*>EPA

United States
Environmental Protection
Agency

Economic Analysis for the Final Lead and Copper Rule

Improvements


-------
Office of Water (4607M)
EPA 810-R-24-005
October 2024


-------
Table of Contents

Executive Summary	ES-1

References	ES-8

1	Introduction	1-1

1.1	Summary of the Final LCRI	1-3

1.2	Document Organization	1-21

1.3	Calculations and Citations	1-21

1.4	References	1-27

2	Need for the Rule	2-1

2.1	Statutory Requirements, Regulatory Actions and National EPA Initiatives Affecting Lead and
Copper in Drinking Water	2-2

2.1.1	Safe Drinking Water Act (SDWA) Requirements and Drinking Water Regulations
Addressing Lead Prior to 1991	2-2

2.1.2	Lead and Copper Rule (1991)	2-3

2.1.3	SDWA Amendments (1996)	2-4

2.1.4	Lead and Copper Rule Minor Revisions (2000)	2-4

2.1.5	2004 National Review of the LCR Leading up to the LCR Short-Term Revisions of 2007 ..2-4

2.1.6	Lead and Copper Rule Short-Term Revisions and Clarifications (2007)	2-5

2.1.7	Lead and Copper Rule Revisions (2021)	2-5

2.1.8	Additional Actions to Reduce Lead in Plumbing Materials (2008-present)	2-6

2.2	Outreach, Consultation, Workgroup Activities, and Other Events Contributing to the Lead and
Copper Rule Revisions	2-6

2.2.1	Stakeholder Meetings	2-7

2.2.2	Input from Small Business Stakeholders	2-7

2.2.3	Input from SAB and NDWAC	2-8

2.2.3.1	SAB Review	2-8

2.2.3.2	NDWAC Meetings	2-8

2.2.4	Consultation with Tribal Governments	2-10

2.2.5	Public Meeting on Environmental Justice	2-11

2.2.6	Consultation with State and Local Government Organizations	2-11

2.2.6.1	November 2011 Federalism Consultation	2-11

2.2.6.2	ASDWA Questionnaire to States on Possible LCRR Requirements	2-11

2.2.6.3	Questionnaire to States on LSL Inventory and Other LSL-Related Information	2-12

2.2.6.4	January 2018 Federalism Consultation	2-12

2.2.6.5	Meetings with ASDWA	2-12

2.2.7	Public Water Systems	2-13

2.2.7.1 Input from PWSs	2-13

2.2.8	EPA Letter to Governors and State Environment and Public Health Commissions and
Tribal Leaders	2-14

2.2.9	Administrator's Meeting with States, PWS, and Non-Government Organizations	2-14

2.2.10	Public Comments on the Proposed LCRR	2-15

2.3	Outreach, Consultation, and Other Engagements Contributing to the Proposed Lead and
Copper Rule Improvements	2-15

Final LCRI Economic Analysis

i

October 2024


-------
2.3.1	LCRR Review	2-15

2.3.2	Consultations and Engagements to Support the Development of the Proposed LCRI....2-17

2.3.2.1	Small Business Stakeholders	2-17

2.3.2.2	Public Meeting on Environmental Justice	2-17

2.3.2.3	Consultation with Tribal Governments	2-18

2.3.2.4	SAB Consultation	2-18

2.3.2.5	NDWAC Consultation	2-19

2.3.2.6	2022 Federalism/Unfunded Mandates Reform Act (UMRA) Consultation	2-19

2.3.2.7	Meetings with ASDWA	2-19

2.3.2.8	HHS Consultation	2-20

2.3.3	Public Water Systems	2-20

2.4	Outreach, Consultation, and Other Engagements Contributing to the Final Lead and Copper
Rule Improvements	2-20

2.4.1	Informational Webinar and Public Hearing	2-20

2.4.1.1	Webinar on Preparing Communities to Engage in the Proposed LCRI Regulatory
Process	2-21

2.4.1.2	Informational Webinar	2-21

2.4.1.3	Public Hearing	2-21

2.4.2	Public Comments on the Proposed LCRI	2-21

2.4.3	Input from NDWAC	2-21

2.4.4	HHS Consultation	2-21

2.5	Statutory Authority for Promulgating the Rule	2-22

2.6	Economic Rationale	2-23

2.7	References	2-25

3 Baseline Drinking Water System Characteristics	3-1

3.1	Introduction	3-1

3.2	Data Sources	3-1

3.2.1	SDWIS/Fed 2020 	3-3

3.2.1.1	Classification of Systems Using SDWIS/Fed Data	3-3

3.2.1.2	Lead and Copper Rule-Specific Data	3-5

3.2.1.3	Treatment Facility Information	3-7

3.2.1.4	Verification of SDWIS/Fed Data	3-7

3.2.2	2006 Community Water System Survey	3-8

3.2.3	Unregulated Contaminant Monitoring Rule 3	3-9

3.2.4	Geometries and Characteristics of Public Water Systems (2000)	3-9

3.2.5	7th Drinking Water Infrastructure Needs Survey and Assessment (DWINSA)	3-9

3.2.6	Six-Year Review Data	3-11

3.2.7	State of Michigan Lead Compliance Monitoring Data	3-12

3.2.8	Data Sources for Schools, Child Care Facilities, Local Health Agencies, and Targeted
Medical Providers	3-13

3.2.8.1	Schools	3-13

3.2.8.2	Child Care Facilities	3-14

3.2.8.3	Local Health Agencies and Targeted Medical Providers	3-14

3.3	Drinking Water System Baseline	3-15

3.3.1	Water System Inventory	3-16

3.3.1.1 Discussion of Data Limitations and Uncertainty	3-17

3.3.2	Population and Households Served	3-18

Final LCRI Economic Analysis

October 2024


-------
3.3.2.1 Discussion of Data Limitations and Uncertainty	3-20

3.3.3	Corrosion Control Treatment (CCT) Status	3-21

3.3.3.1 Discussion of Data Limitations and Uncertainty	3-25

3.3.4	Service Line Material Characterization	3-25

3.3.4.1	Service Line Material Characterization for CWSs	3-25

3.3.4.2	LSL Inventory for NTNCWSs	3-49

3.3.4.3	State Service Line Replacement Regulations	3-53

3.3.5	Lead and Copper Tap Levels	3-54

3.3.5.1	Percent of Systems by Lead 90th Percentile Classification	3-55

3.3.5.2	Likelihood of a System Having Multiple Lead ALEs	3-69

3.3.5.3	Likelihood of an Individual Lead Sample Exceeding the Lead AL	3-71

3.3.5.4	Systems with Copper Only ALEs	3-75

3.3.6	Treatment Plant Characterization	3-79

3.3.6.1	Entry Points per System	3-79

3.3.6.2	Average Daily Flow and Design Flow	3-81

3.3.6.3	Discussion of Data Limitations and Uncertainties	3-81

3.3.7	Lead and Copper Tap Schedules	3-81

3.3.7.1	Estimating Initial Lead and Copper Tap Monitoring Period under the Pre-2021 LCR ..3-
82

3.3.7.2	Estimating Lead and Copper Tap Monitoring Schedules under the 2021 LCRR	3-86

3.3.7.3	Estimating Lead and Copper Tap Monitoring Schedules under the LCRI	3-87

3.3.7.4	Discussion of Data Limitations and Uncertainty	3-91

3.3.8	Water Quality Parameter Monitoring	3-92

3.3.8.1	2021 LCRR	3-92

3.3.8.2	Final LCRI	3-95

3.3.8.3	Discussion of Data Limitations and Uncertainty	3-98

3.3.9	Source and Treatment Changes	3-98

3.3.9.1	Source Change	3-98

3.3.9.2	Primary Source Change	3-100

3.3.9.3	Treatment Change	3-101

3.3.10	Schools, Child Care Facilities, Local Health Departments, and Targeted Medical
Providers3-104

3.3.10.1	Estimated Number of Facilities	3-105

3.3.10.2	Estimated Percentage of Schools and Child Care Facilities that Are Waived from
Monitoring Requirements	3-118

3.3.11	Labor Rates	3-139

3.3.11.1	Public Water System Labor Rates	3-139

3.3.11.2	State Labor Rates	3-142

3.3.11.3	Discussion of Data Limitations and Uncertainty	3-143

3.4	Uncertainties in the Baseline and Compliance Characteristics of Systems	3-144

3.5	References	3-147

4 Economic Impact and Cost Analysis of the Final Lead and Copper Rule Improvements	4-1

4.1	Introduction	4-1

4.1.1	Summary of Rule Costs	4-1

4.1.2	Overview of the Chapter	4-4

4.2	Overview of the SafeWater LCR Model	4-5

4.2.1 Modeling PWS Variability in the SafeWater LCR Model	4-5

Final LCRI Economic Analysis

October 2024


-------
4.2.2	Modeling Uncertainty in the SafeWater LCR Model	4-6

4.2.2.1	Percent of Model PWSs that are Expected to Fall within Five Compliance Tap Sample
90th Percentile Categories	4-8

4.2.2.2	SLR Unit Costs	4-10

4.2.2.3	CCT Unit Costs	4-11

4.2.3	Model PWSs, Very Large Systems, Discounting and Cost of Capital, Compliance
Schedule, and Simulating Compliance Activities	4-12

4.2.3.1	Model Public Water Systems	4-12

4.2.3.2	Very Large Systems	4-13

4.2.3.3	Discounting and Cost of Capital	4-13

4.2.3.4	Schedule	4-14

4.2.3.5	Simulating Compliance Activities	4-14

4.3 Estimating Public Water System Costs	4-15

4.3.1	PWS Implementation and Administrative Costs	4-22

4.3.1.1	PWS One-Time Implementation and Administrative Costs	4-22

4.3.1.2	Estimate of PWS National Implementation and Administrative Costs	4-24

4.3.2	PWS Sampling Costs	4-24

4.3.2.1	PWS Lead Tap Sampling	4-24

4.3.2.2	PWS Lead Water Quality Parameter Monitoring	4-62

4.3.2.3	PWS Copper Water Quality Parameter Monitoring	4-86

4.3.2.4	PWS Source Water Monitoring	4-95

4.3.2.5	CWS School and Child Care Facility Lead Sampling Costs	4-99

4.3.2.6	Estimate of PWS National Sampling Costs	4-141

4.3.3	PWS Corrosion Control Costs	4-141

4.3.3.1	CCT Installation	4-147

4.3.3.2	Re-optimization of Existing Corrosion Control Treatment	4-149

4.3.3.3	DSSA Costs	4-153

4.3.3.4	System Lead CCT Routine Costs	4-167

4.3.3.5	Estimate of PWS National Corrosion Control Treatment Costs	4-172

4.3.4	PWS Service Line Inventory and Replacement Costs	4-174

4.3.4.1	Service Line Inventory	4-174

4.3.4.2	Service Line Replacement Plan	4-196

4.3.4.3	Physical Service Line Replacements	4-202

4.3.4.4	Ancillary Service Line Replacement Activities	4-203

4.3.4.5	Estimate of national service line testing and replacement costs	4-210

4.3.5	PWS POU-Related Costs	4-211

4.3.5.1	POU Device Installation and Maintenance	4-212

4.3.5.2	POU Ancillary Activities	4-215

4.3.5.3	Estimate of PWS National Point-of-Use Device Installation and Maintenance Costs ..4-
225

4.3.6	PWS Lead Public Education, Outreach, and Notification Costs	4-226

4.3.6.1	Consumer Notice	4-226

4.3.6.2	Activities Regardless of Lead 90th Percentile Level	4-227

4.3.6.3	Public Education Activities in Response to Lead ALE	4-251

4.3.6.4	Public Education Activities in Response to Multiple Lead ALEs	4-262

4.3.6.5	Estimate of National Lead Public Education and Outreach Costs	4-270

4.3.7	Summary of PWS Costs	4-270

4.3.7.1 PWS counts and population affected by rule components	4-270

Final LCRI Economic Analysis	iv	October 2024


-------
4.3.7.2	Estimated Cost per Public Water System by System Category	4-272

4.3.7.3	Household Costs by CWS Size and Source Water Type	4-280

4.4	Estimating State (Primacy Agency) Costs	4-285

4.4.1	State Implementation and Administrative Costs	4-289

4.4.1.1	State Start-up Implementation and Administrative Activities	4-289

4.4.1.2	State Annual Implementation and Administrative Activities	4-291

4.4.2	State Sampling Related Costs	4-294

4.4.2.1	State Lead Tap Sampling Costs	4-295

4.4.2.2	State Lead WQP Sampling Costs	4-305

4.4.2.3	State Copper WQP Monitoring Costs	4-306

4.4.2.4	State Source Water Monitoring Costs	4-308

4.4.2.5	State School Sampling Costs	4-310

4.4.3	State CCT Related Costs	4-313

4.4.3.1	State CCT Installation Costs	4-313

4.4.3.2	State CCT Re-optimization Costs	4-315

4.4.3.3	State Distribution System and Site Assessment Costs	4-317

4.4.3.4	State Lead CCT Routine Costs	4-318

4.4.4	State Service Line Inventory and Replacement Related Costs	4-323

4.4.4.1	SL Inventory Costs	4-323

4.4.4.2	SLR Plan Review Costs	4-324

4.4.4.3	SLR Report Review Costs	4-327

4.4.5	State POU Related Costs	4-329

4.4.5.1	One-Time POU Program Costs	4-329

4.4.5.2	Ongoing POU Program Costs	4-331

4.4.6	State Public Education-Related Costs	4-335

4.4.6.1	Consumer Notice	4-335

4.4.6.2	Activities Regardless of the Lead 90th Percentile Level	4-336

4.4.6.3	Public Education Activities in Response to Lead ALE	4-342

4.4.6.4	Public Education Activities in Response to Multiple Lead ALEs	4-343

4.4.7	Summary of Estimated State Costs	4-348

4.5	Costs and Ecological Impacts Associated with Additional Phosphate Usage	4-348

4.5.1	Estimating the Costs of Increased Phosphorus Loadings	4-348

4.5.1.1	Incremental phosphorus loading to wastewater treatment plants	4-348

4.5.1.2	Incremental phosphorus removal costs at wastewater treatment plants	4-350

4.5.2	Ecological Impacts of Phosphorus Loadings	4-353

4.5.2.1	Incremental total phosphorus loadings in water bodies	4-354

4.5.2.2	Ecological impacts of potential increases in phosphate loadings	4-356

4.6	References	4-357

5 Benefits Resulting from the Lead and Copper Rule Improvements	5-1

5.1	Introduction	5-1

5.2	Baseline and Post-Rule Drinking Water Lead Exposures	5-2

5.2.1	Drinking Water Lead Concentration Profile Data	5-4

5.2.1.1	Lead Concentration Profiles	5-6

5.2.1.2	Data Cleaning	5-8

5.2.1.3	Coding	5-11

5.2.2	Drinking Water Lead Concentration Model Fitting and Selection	5-12

5.2.3	Simulated Drinking Water Lead Concentrations Based on Selected Model Fit	5-16

Final LCRI Economic Analysis

v

October 2024


-------
5.2.4 Determination of GRR, and Point-of-Use and Pitcher Filter Water Lead Concentrations...5-

18

5.2.5 Limitations of Baseline and Post-Rule Water Concentration Estimates	5-19

5.3	Assignment of Drinking Water Lead Tap Concentrations to PWS Populations	5-19

5.4	Methods for Estimating Blood Lead Levels	5-23

5.4.1	Methods for Estimating Blood Lead Levels in Children Ages 0-7	5-23

5.4.1.1	SHEDS-Multimedia Modeling	5-24

5.4.1.2	IEUBK Model	5-24

5.4.1.3	Background Lead Exposure Inputs into SHEDS-Pb	5-25

5.4.1.4	Coupling of SHEDS-Multimedia and IEUBK Models for SHEDS-Pb Modeling	5-27

5.4.1.5	Estimates of Pre- and Post-Rule Blood Lead Levels in Young Children	5-29

5.4.2	Methods for Estimating Blood Lead Levels in Older Children and Adults	5-31

5.4.2.1	Overview of the All Ages Lead Model	5-31

5.4.2.2	Estimates of Pre- and Post-Rule Blood Lead Levels in Adults	5-35

5.5	Concentration Response Functions and Valuations used in the Estimation of Benefits to
Children and Adults	5-38

5.5.2	Valuation of Avoided IQ Loss	5-41

5.5.3	Concentration-Response Function for Lead and ADHD	5-43

5.5.4	Valuation of Avoided ADHD	5-46

5.5.5	Concentration-Response Function for Lead and Birth Weight of Infants Born to Women

of Child-Bearing Age	5-48

5.5.6	Valuation of Avoided Reductions in Birth Weight	5-50

5.5.7	Concentration-Response Function for Lead and Cardiovascular Disease Premature
Mortality5-53

5.5.8	Valuation of Avoided Cardiovascular Disease Premature Mortality	5-55

5.6	National Level Benefits Estimates	5-56

5.6.1	Implementation of Benefit Calculations in the SafeWater LCR model	5-56

5.6.2	Monetized National Annual Benefits	5-57

5.7	Uncertainty in the Quantified Benefits	5-64

5.7.1	Uncertainty in Blood Lead Modeling	5-66

5.7.2	General Uncertainty in Concentration-Response Relationships and Population	5-66

5.7.3	General Uncertainty in Valuation	5-67

5.7.4	Uncertainty in IQ	5-68

5.7.5	Uncertainty in ADHD	5-69

5.7.6	Uncertainty in Reductions in Birth Weight	5-70

5.7.7	Uncertainty in Cardiovascular Disease Premature Mortality Benefits	5-70

5.8	Summary of Non-Quantified and Non-Monetized Benefits	5-72

5.9	Disbenefits from Greenhouse Gas Emissions	5-73

5.9.1	Energy Consumption and Unit Greenhouse Gas Emissions	5-74

5.9.1.1	Energy Consumption Estimates	5-74

5.9.1.2	Converting Consumed Energy Estimates into Greenhouse Gas Emissions	5-76

5.9.2	Calculating Annual Total Incremental Emissions in SafeWater LCR	5-78

5.9.3	Valuation of GHG Emissions	5-81

5.10	References	5-87

6 Comparison of Costs to Benefits	6-1

6.1 Summary of the Incremental Costs of the Final LCRI	6-1

6.1.1 Monetized Incremental Costs	6-1

Final LCRI Economic Analysis

vi

October 2024


-------
6.1.2 Non-monetized and Non-quantified Costs	6-3

6.2	Summary of the Incremental Benefits of the Final LCRI	6-4

6.2.1	Monetized Incremental Benefits	6-4

6.2.2	Non-monetized and Non-quantified Benefits	6-6

6.3	Comparison of Incremental Costs to Incremental Benefits	6-7

6.4	References	6-13

7 Statutory and Administrative Requirements	7-1

7.1	Introduction	7-1

7.2	Executive Order 12866: Regulatory Planning and Review and Executive Order 14094:
Modernizing Regulatory Review	7-1

7.3	Paperwork Reduction Act	7-2

7.3.1	State Activities	7-3

7.3.2	System Activities	7-3

7.4	The Regulatory Flexibility Act	7-6

7.4.1	Need for and Objectives of the Rule	7-7

7.4.2	Summary of SBAR Comments and Recommendations	7-7

7.4.3	Number and Description of Small Entities Affected	7-11

7.4.4	Description of the Compliance Requirements of the Rule	7-12

7.4.5	Costs and Benefits of the Final LCRI by Small System Size Category	7-13

7.4.6	Analysis of Alternative Small System Rule Requirements	7-13

7.4.6.1 Alternative Small System Flexibility Option	7-18

7.5	Unfunded Mandates Reform Act	7-18

7.6	Executive Order 13132: Federalism	7-22

7.7	Executive Order 13175: Consultation and Coordination with Indian Tribal Governments ....7-23

7.8	Executive Order 13045: Protection of Children from Environmental Health and Safety Risks ..7-
24

7.9	Executive Order 13211: Actions That Significantly Affect Energy Supply, Distribution, or Use..7-

25

7.9.1	Energy Supply	7-26

7.9.2	Energy Distribution	7-26

7.9.3	Energy Use	7-26

7.10	National Technology Transfer and Advancement Act (NTTAA)	7-26

7.11	Executive Order 12898: Federal Actions to Address Environmental Justice in Minority
Populations and Low-Income Populations and Executive Order 14096 (Revitalizing Our
Nation's Commitment to Environmental Justice for All)	7-27

7.12	Consultations with the Science Advisory Board, National Drinking Water Advisory Council, and
the Secretary of Health and Human Services	7-29

7.12.1	Consultation with Science Advisory Board	7-29

7.12.2	Consultation with National Drinking Water Advisory Council (NDWAC)	7-31

7.12.3	Consultation with Health and Human Services	7-31

7.13	References	7-31

8 Other Options Considered	8-1

8.1	Introduction	8-1

8.2	Alternative Lead Action Levels	8-2

8.3	Alternative Service Line Replacement Rate	8-7

8.4	Alternative Definition of Lead Content Service Lines to Be Replaced	8-9

8.5	Alternative Service Line Replacement Deferral Deadline	8-12

Final LCRI Economic Analysis

vii

October 2024


-------
8.6	Alternative Tap Sampling Requirements	8-15

8.7	Alternative Temporary Filter Programs for Systems with Multiple ALEs	8-16

8.8	Small System Flexibility	8-19

8.9	Summary of Alternative Options Considerations	8-22

8.10	References	8-24

Final LCRI Economic Analysis

viii

October 2024


-------
List of Exhibits

Exhibit ES-1: Estimated National Annualized Monetized Incremental Costs of the Final LCRI at 2 Percent

Discount Rate (millions of 2022 USD)	ES-4

Exhibit ES-2: Estimated National Annualized Monetized Benefits of the Final LCRI at 2 Percent Discount

Rate (millions of 2022 USD)	ES-6

Exhibit ES-3: Comparison of Estimated Monetized National Annualized Incremental Costs to Benefits of

the LCRI - 2 Percent Discount Rate (millions 2022 USD)	ES-8

Exhibit 1-1: Comparison of the 2021 LCRR, Proposed LCRI, and Final LCRI Requirements	1-6

Exhibit 1-2: Supporting Report and Spreadsheet Files	1-22

Exhibit 3-1: Data Sources Used to Develop the Baseline for the Final LCRI	3-2

Exhibit 3-2: Inventory of CWSs	3-16

Exhibit 3-3: Inventory of NTNCWSs	3-17

Exhibit 3-4: Population and Number of Households Served by CWSs	3-19

Exhibit 3-5: Population Served by NTNCWSs	3-20

Exhibit 3-6: Number of CWSs with and without CCT	3-23

Exhibit 3-7: Number of NTNCWS with and without CCT	3-24

Exhibit 3-8: Relationship between Service Line Categories in the Final LCRI Economic Analysis, the 7th

DWINSA, and the DWINSA One-time Update	3-28

Exhibit 3-9: CWS Categorization Based on Service Line Material	3-29

Exhibit 3-10: Percent and Number of CWSs in Service Line Material Categories	3-31

Exhibit 3-11: Characterization of Service Lines from the 7th DWINSA by Material Type and System Size..3-
33

Exhibit 3-12: Total Number of Service Lines in Categories 1 through 4 from DWINSA	3-34

Exhibit 3-13: Allocation of Known Lead Content, Non-Lead, and Unknown Service Lines to Categories 1

through 4	3-35

Exhibit 3-14: Total Number of Service Lines by System Size and CCT Status based on SDWIS/Fed 4th

Quarter 2020 Data	3-38

Exhibit 3-15: Known and Projected Lead Content Service Lines in Category 1 Systems	3-39

Exhibit 3-16: Projected Lead Content Service Lines in Category 3 Systems	3-41

Exhibit 3-17: Projected Lead Content Service Lines in Category 4 Systems	3-42

Exhibit 3-18: Total Number of Known and Projected Lead Content Service Lines by Category	3-44

Exhibit 3-19: Characterization of Lead Content Service Lines in CWSs	3-46

Exhibit 3-20: Similarities and Differences in Service Line Material Data Analysis for the Final LCRI

Economic Analysis and the DWINSA LSL Allocation Model	3-47

Exhibit 3-21: Summary of State Responses Regarding the Percentage of NTNCWSs with LSLs	3-50

Exhibit 3-22: Estimated Number of NTNCWSs With and Without LSLs by CCT Status	3-51

Exhibit 3-23: Number of LSLs in NTNCWSs with CCT by Size Category	3-52

Exhibit 3-24: Number of CWSs with LSL Determination Based on State, Tribal, DWINSA Responses, and

Web Data1	3-57

Exhibit 3-25: Percent of CWSs by Lead 90th Percentile Classification under the 2021 LCRR	3-61

Exhibit 3-26: Percent of CWSs by Lead 90th Percentile Classification under the Final LCRI	3-62

Exhibit 3-27: Comparison Percent of CWSs with Known LSL Status to All CWSs by System Size	3-65

Final LCRI Economic Analysis	ix	October 2024


-------
Exhibit 3-28: Comparison of P90 Data for CWSs with At Least One Reported Value to the Set of CWSs
with Known LSL Status and P90 Data by Two P90 Ranges, System Size, and CCT Status (Percent) Using

the Baseline/High Estimate	3-66

Exhibit 3-29: Number and Percent of CWSs with No ALE, and ALE - Comparison of Results from Five

Geographic Regions with Known LSL Status Using the Baseline/High Estimate	3-68

Exhibit 3-30: Percentage of CWSs and NTNCWSs with At Least One ALE that Have At Least Two Lead

Action Level Exceedances (Above 10 ng/L) in Five Years	3-70

Exhibit 3-31: Percentage of CWSs and NTNCWSs with At Least Two Lead ALEs in Five Years that Have At

Least One Additional Lead ALE (Above 10 ng/L) in Five Years	3-71

Exhibit 3-32: Percent of Individual Lead Sample Results Above 15 ng/L Based on Michigan CWSs with

Known LSL Status for the 2021 LCRR	3-73

Exhibit 3-33: Percent of Individual Lead Sample Result Above 10 ng/L Based on Michigan CWSs with

Known LSL Status for the Final LCRI	3-74

Exhibit 3-34: Average Percent of CWSs that Had Any Copper Only ALE from 2012-2020 	3-77

Exhibit 3-35: Average Percent of NTNCWSs that Had Any Copper Only ALE from 2012-2020	3-78

Exhibit 3-36: Frequency Distribution of Entry Point Inputs for CWSs	3-80

Exhibit 3-37: Frequency Distribution of Entry Point Inputs for NTNCWSs	3-80

Exhibit 3-38: SDWIS/Fed Criteria Used to Estimate Lead Tap Sampling Monitoring Schedules under the

Pre-2021 LCR	3-82

Exhibit 3-39: Estimated Percentage of CWSs with CCT on Various Lead Tap Monitoring Schedules by Size

and Source Type under the Pre-2021 LCR	3-83

Exhibit 3-40: Estimated Percentage of CWSs without CCT on Various Lead Tap Monitoring Schedules by

Size and Source Type under the Pre-2021 LCR	3-84

Exhibit 3-41: Estimated Percentage of NTNCWSs with CCT on Various Lead Monitoring Schedules by Size

and Source Type under the Pre-2021 LCR	3-85

Exhibit 3-42: Estimated Percentage of NTNCWSs without CCT on Various Lead Tap Monitoring Schedules

by Size and Source Type under the Pre-2021 LCR	3-85

Exhibit 3-43: Comparison of the Criteria for Standard and Reduced Tap Sample Monitoring under the

Pre-2021 LCR, 2021 LCRR, and Final LCRI	3-88

Exhibit 3-44: Estimated Number and Percentage of CWSs with Reported Lead ALEs Only under the Pre-

2021 LCR (2012-2020)	3-91

Exhibit 3-45: SDWIS/Fed Data Criteria Used to Determine Reduced WQP Tap Monitoring Schedules for

Systems Serving > 50,000 People With CCT and No Lead TLE or Copper ALE	3-93

Exhibit 3-46: Percentage of Ground Water CWSs Serving > 50,000 People with CCT and No Lead TLE or
Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting in the First Modeled

Compliance Period Given 2021 LCRR Requirements)	3-94

Exhibit 3-47: Percentage of Surface Water CWSs Serving > 50,000 People with CCT and No Lead TLE or
Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting in the First Modeled

Compliance Period Given 2021 LCRR Requirements)	3-94

Exhibit 3-48: Percent of Surface Water NTNCWSs Serving > 50,000 People with CCT and No Lead TLE or
Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting in the First Modeled
Compliance Period Given 2021 LCRR Requirements)	3-94

Final LCRI Economic Analysis

x

October 2024


-------
Exhibit 3-49: Percent of Ground Water CWSs Serving > 10,000 People with CCT and No Lead or Copper
ALE on Various WQP Distribution System Monitoring Schedules (Starting in the First Modeled

Compliance Period Given Final LCRI Requirements)	3-96

Exhibit 3-50: Percentage of Surface Water CWSs Serving > 10,000 People with CCT and No Lead or
Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting in the First Modeled

Compliance Period Given Final LCRI Requirements)	3-96

Exhibit 3-51: Percentage of Ground Water NTNCWSs Serving > 10,000 People with CCT and No Lead or
Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting in the First Modeled

Compliance Period Given Final LCRI Requirements)	3-97

Exhibit 3-52: Percentage of Surface Water NTNCWSs Serving > 10,000 People with CCT and No Lead or
Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting in the First Modeled

Compliance Period Given Final LCRI Requirements)	3-97

Exhibit 3-53: Estimated Percent of CWSs that Will Add a New Source Each Year	3-99

Exhibit 3-54: Estimated Percent of NTNCWS that Will Change Source Each Year	3-100

Exhibit 3-55: Estimated Percent of CWSs that Will Change Treatment Each Year1	3-103

Exhibit 3-56: Estimated Percent of NTNCWSs that Will Change Treatment Each Year	3-103

Exhibit 3-57: Number of Schools and Child Care Facilities by State and United States Territory, Adjusted

to Remove NTNCWS Schools and Child Care Facilities	3-108

Exhibit 3-58: Number of Schools per Person Served by a CWS per State, all Categories, Adjusted to

Remove NTNCWS Schools and Child Care Facilities	3-112

Exhibit 3-59: Estimated Average Number of Local Health Agencies and Targeted Medical Providers per

CWS	3-117

Exhibit 3-60: States with Existing Programs that Satisfy the Waiver Requirements under the 2021 LCRR

for the First Five-Year Cycle1	3-119

Exhibit 3-61: States with Existing Programs that Satisfy the Waiver Requirements under the Final LCRI for

the First Five-Year Cycle1	3-119

Exhibit 3-62: States with Existing Programs that Satisfy the Waiver Requirements Under the 2021 LCRR

for the Second Five-Year Cycle and Subsequent Five-Year Cycles1	3-120

Exhibit 3-63: States with Existing Programs that Satisfy the Waiver Requirements Under the Final LCRI

for the Second Five-Year Cycle and Subsequent Five-Year Cycles1	3-121

Exhibit 3-64: Low and High Estimate for the Number of Taps to be Sampled for Elementary Schools and

Child Care Facilities	3-122

Exhibit 3-65: Estimated Burden per Elementary School Sampling Event	3-123

Exhibit 3-66: Estimated Burden per Child Care Facility Sampling Event	3-123

Exhibit 3-67: Estimated Total Cost per Elementary School Sample Event (2020$)	3-124

Exhibit 3-68: Estimated Total Cost per Child Care Facility Sample Event (2020$)	3-125

Exhibit 3-69: 2021-2023 WIIN Grant Allotment and Projected WIIN Grant Funding for FY 2024 - FY 2026

	3-126

Exhibit 3-70: Estimated Percent of Public Elementary Schools and Child Care Facilities that Could be
Sampled to Comply with the 2021 LCRR Using WIIN Grant Funds from October 16, 2024 - FY 2026 .3-129
Exhibit 3-71: Estimated Percent of Public Elementary Schools and Child Care Facilities that Could be

Sampled to Comply with the Final LCRI Using WIIN Grant Funds from January 1, 2021 - FY 2026	3-131

Exhibit 3-72: Percent of Schools and Child Care Facilities Eligible for Waivers under the 2021 LCRR based
on State Regulations and WIIN Grant Funding	3-133

Final LCRI Economic Analysis

xi

October 2024


-------
Exhibit 3-73: Percent of Schools and Child Care Facilities Eligible for Waivers under the Final LCRI based

on State Regulations and WIIN Grant Funding	3-136

Exhibit 3-74: Comparison of Wage Rate Surveys	3-140

Exhibit 3-75: Hourly Labor Costs Including Wages Plus Benefits (2019$)	3-141

Exhibit 3-76: Weighted Labor Rates for CWSs and NTNCWSs	3-142

Exhibit 3-77: Loaded Labor Rate for State Staff (2020$)	3-143

Exhibit 3-78: Summary of Uncertainties in the Baseline and Compliance Characteristics of Drinking Water

Systems	3-144

Exhibit 4-1: Estimated National Annualized Rule Costs - 2 Percent Discount Rate (millions of 2022 USD) 4-
2

Exhibit 4-2: Summary of Uncertainties in the Estimation of Compliance Actions and Costs	4-7

Exhibit 4-3: Likelihood of Initial Model PWS 90th Percentile Placement under the 2021 LCRR	4-9

Exhibit 4-4: Percent of CWSs by Lead 90th Percentile Classification under the Final LCRI	4-10

Exhibit 4-5: Summary of SLR Costs from DWINSA Survey ($/SLR, 2020$)	4-11

Exhibit 4-6: PWS Cost Components, Subcomponents, and Activities Organized by Section1	4-16

Exhibit 4-7: PWS One-Time Administration Activities and Unit Burden Estimates	4-22

Exhibit 4-8: PWS Administration and Rule Implementation Cost Estimation in SafeWater LCR by Activity

	4-23

Exhibit 4-9: Minimum Number of Lead Tap Sampling Sites for Standard and Reduced Monitoring	4-26

Exhibit 4-10: PWS Lead Tap Sampling Unit Burden and Cost Estimates	4-26

Exhibit 4-11: CWS Burden to Achieve a Sampling Pool with 100 Percent Lead Service Line Sites	4-29

Exhibit 4-12: Non-Labor Costs for CWS without LSLs to Provide Test Kits (per Sample)	4-33

Exhibit 4-13: Non-Labor Costs for CWS with LSLs to Provide Test Kits (per Sample)	4-34

Exhibit 4-14: Travel Burden and Cost for Lead Tap Sample Pickup	4-35

Exhibit 4-15: Burden to Submit Lead Tap Sampling Results and 90th Percentile Level	4-39

Exhibit 4-16: PWS Lead Tap Sampling Cost Estimation in SafeWater LCR by Activity1,2	4-40

Exhibit 4-17: Baseline Percentage of Systems Modifying pH and/or Adding P04	4-63

Exhibit 4-18: Normalized Baseline Percentage of Systems Modifying pH and/or Adding P04	4-64

Exhibit 4-19: Minimum Number of WQP Distribution Samples for Systems on Standard or Reduced

Monitoring	4-65

Exhibit 4-20: PWS Lead WQP Monitoring Unit Burden and Cost Estimates	4-66

Exhibit 4-21: CWS Material Costs Associated with Distribution System Sample Collection	4-68

Exhibit 4-22: NTNCWS Material Costs Associated with Distribution System Sample Collection	4-69

Exhibit 4-23: CWS In-House WQP Analytical Burden for Distribution System Samples (hrs/sample) ....4-70
Exhibit 4-24: NTNCWS In-House WQP Analytical Burden for Distribution System Samples (hrs/sample) .4-
70

Exhibit 4-25: CWS In-House WQP Analytical Cost for Distribution System Samples ($/sample)	4-71

Exhibit 4-26: NTNCWS In-House WQP Analytical Cost for Distribution System Samples ($/sample)	4-71

Exhibit 4-27: CWS Commercial WQP Analytical Cost for Distribution System Samples ($/sample)	4-72

Exhibit 4-28: NTNCWS Commercial WQP Analytical Cost for Distribution System Samples ($/sample) 4-73

Exhibit 4-29: CWS Material Costs Associated with Entry Point Sample Collection	4-74

Exhibit 4-30: NTNCWS Material Costs Associated with Entry Point Sample Collection	4-74

Exhibit 4-31: CWS In-House WQP Analytical Burden for Entry Point Samples (hrs/sample)	4-75

Exhibit 4-32: NTNCWS In-House WQP Analytical Burden for Entry Point Samples (hrs/sample)	4-75

Final LCRI Economic Analysis

xii

October 2024


-------
Exhibit 4-33: CWS In-House WQP Analytical Cost for Entry Point Samples ($/sample)	4-75

Exhibit 4-34: NTNCWS In-House WQP Analytical Cost for Entry Point Samples ($/sample)	4-76

Exhibit 4-35: CWS Commercial WQP Analytical Cost for Entry Point Samples ($/sample)	4-76

Exhibit 4-36: NTNCWS Commercial WQP Analytical Cost for Entry Point Samples ($/sample)	4-77

Exhibit 4-37: PWS Lead WQP Monitoring Cost Estimation in SafeWater LCR by Activity1	4-78

Exhibit 4-38: Estimated Likelihood a CWS Will Have a Copper Only ALE (2012 - 2020)	4-87

Exhibit 4-39: Estimated Likelihood a NTNCWS Will Have a Copper Only ALE (2012 - 2020)	4-88

Exhibit 4-40: PWS Copper WQP Monitoring Unit Burden and Cost Estimates	4-89

Exhibit 4-41: PWS Copper WQP Monitoring Cost Estimation in SafeWater LCR by Activity1	4-90

Exhibit 4-42: PWS Source Monitoring Burden and Cost Estimates	4-96

Exhibit 4-43: PWS Source Water Monitoring Cost Estimation in SafeWater LCR by Activity1	4-98

Exhibit 4-44: CWS School and Child Care Facility Sampling Unit Burden and Cost Estimates for the First

Five-Year Testing Cycle (Years 4-8)	4-101

Exhibit 4-45: CWS School and Child Care Facility First Five-Year Testing Cycle Cost Estimation in

SafeWater LCR by Activity1,2	4-108

Exhibit 4-46: CWS School and Child Care Facility Sampling Unit Burden and Cost Estimates under the

Second Five-Year Testing Cycle On	4-124

Exhibit 4-47: CWS School and Child Care Facility Second Five-Year Testing Cycle Cost Estimation in

SafeWater LCR by Activity1,2	4-128

Exhibit 4-48: Estimated National Annualized Sampling Costs - 2 Percent Discount Rate (millions of 2022

USD)	4-141

Exhibit 4-49: Distribution of Baseline Finished Water pH by Source Water Type and pH Adjustment

Status	4-143

Exhibit 4-50: Distribution of Finished Water pH by Source Water Type for Model-PWSs without pH

Adjustment in Place by CCT Status	4-143

Exhibit 4-51: Distribution of Finished Water pH by Source Water Type for Model-PWSs with pH

Adjustment in Place by CCT Status	4-144

Exhibit 4-52: Derivation of Baseline P04 Dose by System Size and LSL Status	4-146

Exhibit 4-53: Baseline P04 Doses by System Size and LSL Status Used in Cost Modeling	4-146

Exhibit 4-54: PWS CCT Installation-Related Unit Burden and Cost Estimates	4-148

Exhibit 4-55: PWS Ancillary CCT Installation Cost Estimation in SafeWater LCR by Activity1	4-149

Exhibit 4-56: PWS CCT Ancillary Re-optimization Unit Burden and Cost Estimates	4-151

Exhibit 4-57: PWS CCT Ancillary Re-optimization Cost Estimation in SafeWater LCR by Activity1	4-152

Exhibit 4-58: Likelihood of an Individual Lead Sample Result Above 10 ng/L	4-153

Exhibit 4-59: PWS Burden and Cost to Flush as DSSA Response (2020$)	4-154

Exhibit 4-60: PWS Ancillary DSSA Unit Burden and Cost Estimates	4-155

Exhibit 4-61: Burden (hours) for CWSs to Contact Customers and Collect Tap Samples for Locations with

a Lead Tap Sample > 10 ng/L (hrs_samp_above_al_op)	4-157

Exhibit 4-62: Costs for CWSs to Contact Customers and Collect Tap Samples for Locations with a Lead

Tap Sample > 10 ng/L (cost_samp_above_al)	4-158

Exhibit 4-63: Likelihood a CWS Will Add a WQP Sampling Site in Response to the DSSA	4-159

Exhibit 4-64: PWS Burden to Conduct Distribution System Assessment	4-161

Exhibit 4-65: PWS Ancillary DSSA Cost Estimation in SafeWater LCR by Activity12	4-162

Exhibit 4-66: PWS CCT Routine Unit Burden and Cost Estimates	4-167

Final LCR! Economic Analysis

xiii

October 2024


-------
Exhibit 4-67: Estimated PWS Burden to Gather Data and Review CCT-Related Data during Sanitary Survey

to Determine if CCT Is Still Optimized	4-168

Exhibit 4-68: Estimated Percent of Ground Water CWSs Achieving 4-log Virus Inactivation	4-169

Exhibit 4-69: Estimated Hours per System to Report and Consult on Source Water Change	4-170

Exhibit 4-70: Estimated Hours per System to Report and Consult on Treatment Change	4-171

Exhibit 4-71: PWS Lead CCT Routine Cost Estimation in SafeWater LCR by Activity1	4-171

Exhibit 4-72: Estimated National Annualized Corrosion Control Costs - 2 Percent Discount Rate (millions

of 2022 USD)	4-173

Exhibit 4-73: PWS Service Line Inventory Connector Review Unit Burden and Cost Estimates	4-175

Exhibit 4-74: Estimated Unit Burden for CWSs to Review Records for Connector Material	4-179

Exhibit 4-75: Estimated Unit Burden for CWSs and NTNCWSs to Compile and Submit the Connector

Updated LCRR Initial Inventory	4-181

Exhibit 4-76: PWS Service Line Inventory Update Unit Burden and Cost Estimates	4-183

Exhibit 4-77: Average Cost per Service Line Investigated for Four Investigation Methods	4-186

Exhibit 4-78: Least-Cost Decision Tree and Weighted Average Cost for Field Investigations	4-188

Exhibit 4-79: Weighted Average Unit Cost ($/SL) for Identifying service line material of "Unknowns" for

the Inventory Updates	4-189

Exhibit 4-80: PWS Inventory Validation Unit Burden and Cost Estimates	4-190

Exhibit 4-81: Minimum Non-lead Service Line Validation Requirements of the Final LCRI	4-191

Exhibit 4-82: Unit Cost ($/SL) for Validation	4-194

Exhibit 4-83: Lead Service Line Inventory Cost Estimation in SafeWater LCR by Activity1	4-195

Exhibit 4-84: PWS SLR Plan Unit Burden and Cost Estimates	4-196

Exhibit 4-85: Estimated Burden for Systems with Lead, GRR, and/or Unknown Service Lines to Develop

an SLR Plan	4-198

Exhibit 4-86: PWS Burden to Identify Funding Options for SLRs	4-199

Exhibit 4-87: Estimated Additional Burden for the Initial SLR Plan Development for Systems Requesting a

Deferred SLR Rate	4-200

Exhibit 4-88: Estimated Annual Burden for Systems to Update the SLR Plan or Certify No Changes ...4-200

Exhibit 4-89: LSLR Plan Cost Estimation in SafeWater LCR by Activity1	4-201

Exhibit 4-90: PWS LSLR Cost Estimates	4-202

Exhibit 4-91: Lead Service Line Replacement Cost Estimation in SafeWater LCR by Activity1	4-203

Exhibit 4-92: PWS SL Replacement Ancillary Unit Burden and Cost Estimates	4-204

Exhibit 4-93: Estimated Burden Associated with Contacting Customers and Site Visit Prior to LSLR

(hours/replaced SL) (hrs_replaced_lsl_contact_op)	4-205

Exhibit 4-94: Estimated Non-Labor Costs Associated with Contacting Customers and Site Visit Prior to

SLR ($/replaced SL) (cost_replaced_lsl_contact)	4-205

Exhibit 4-95: CWS Unit Burden to Collect Post-SLR Tap Sample	4-206

Exhibit 4-96: CWS Non-labor Unit Cost to Collect Post-SLR Tap Sample	4-206

Exhibit 4-97: Service Line Inventory Ancillary Cost Estimation in SafeWater LCR by Activity1	4-209

Exhibit 4-98: Estimated National Annualized Lead Service Line Replacement Costs - 2 Percent Discount

Rate (millions of 2022 USD)	4-211

Exhibit 4-99: Average Number of Households and POU Devices per CWS	4-212

Exhibit 4-100: Minimum and Maximum Estimated Number of Taps Requiring POU Devices per NTNCWS
	4-213

Final LCRI Economic Analysis

xiv

October 2024


-------
Exhibit 4-101: Point-of-Use Device Installation and Maintenance Cost Estimation in SafeWater LCR by

Activity1	4-214

Exhibit 4-102: PWS Ancillary POU-Related Burden and Cost Estimates1	4-215

Exhibit 4-103: CWS Burden to Develop a POU Plan (hrs/system)	4-217

Exhibit 4-104: NTNCWS Burden to Develop a POU Plan (hours/system)	4-217

Exhibit 4-105: PWS Annual POU Program Report Preparation and Submission Burden	4-221

Exhibit 4-106: PWS Point-of-Use Ancillary Costing Estimation in SafeWater LCR by Activity1,2,3	4-221

Exhibit 4-107: PWS Burden for Consumer Notification of Lead and Copper Tap Sampling Results	4-226

Exhibit 4-108: PWS Burden and Cost for Public Education Activities that Are Independent of Lead 90th

Percentile Levels	4-227

Exhibit 4-109: One-Time Burden (per CWS) to Develop Approach for Improved Access to Lead

Information	4-230

Exhibit 4-110: Likelihood that a Resident Will Request Information about potential lead content SLs4-231

Exhibit 4-111: Households (HHs) with Children under 6 and That Moved	4-232

Exhibit 4-112: Number of Potential Lead Content SL Information Requests from Realtors, Home

Inspectors, and Potential Home Buyers	4-232

Exhibit 4-113: Estimated Number of Health Agencies	4-233

Exhibit 4-114: Annual CWS Burden (per system) to Conduct Outreach to Local and State Health Agencies

	4-234

Exhibit 4-115: CWS Annual Burden (per household) to Distribute General Inventory-related Outreach...4-
238

Exhibit 4-116: Likelihood that the CWS Has a High Proportion of non-English Speaking Customers ...4-240
Exhibit 4-117: Unit Burden for CWSs to Provide Phone Translation by Type of Public Education Material

	4-243

Exhibit 4-118: Unit Cost for CWSs to Provide Written Translation by Type of Public Education Material .4-
245

Exhibit 4-119: PWS Lead Public Education Unit Costing Approach in SafeWater LCR by Activity1	4-248

Exhibit 4-120: PWS Public Education Burden in Response to Lead ALE	4-251

Exhibit 4-121: Number of Local Health Agencies, Schools, Child Care Facilities, and Targeted Medical

Providers Proportionally Distributed by CWS Population Served	4-255

Exhibit 4-122: System Burden for Public Meetings	4-256

Exhibit 4-123: System Burden for Additional Public Education Activities after a Lead ALE	4-257

Exhibit 4-124: Cost for Paid Ads (2021$)	4-258

Exhibit 4-125: System Non-Labor Costs for Additional Public Education Activities after a Lead ALE ...4-260
Exhibit 4-126: PWS Lead ALE Public Education Unit Costing Approach in SafeWater LCR by Activity1.4-261

Exhibit 4-127: PWS Public Education Burden in Response to Multiple Lead ALEs	4-263

Exhibit 4-128: Community Water System Burden for Enhanced Outreach Following a Minimum of 3 Lead

Action Level Exceedances in a 5-Year Period (per system per 6-month period)	4-264

Exhibit 4-129: Estimated Average Annual Burden to Conduct Enhanced Outreach for CWSs with Multiple

Lead ALEs (per system)	4-266

Exhibit 4-130: Community Water System Non-Labor Cost for Enhanced Outreach Following a Minimum

of 3 Lead Action Level Exceedances in a 5-Year Period (per system per 6-month period)	4-267

Exhibit 4-131: Estimated Average Annual Non-Labor Costs to Conduct Enhanced Outreach for CWSs with
Multiple Lead ALEs (per system)	4-268

Final LCR! Economic Analysis

xv

October 2024


-------
Exhibit 4-132: PWS Lead Multiple ALEs Public Education Unit Costing Approach in SafeWater LCR by

Activity1,2	4-269

Exhibit 4-133: Estimated National Annualized Public Education Costs - 2 Percent Discount Rate (millions

of 2022 USD)	4-270

Exhibit 4-134: Estimated System Counts and Population Impacted (Over 35 Year Period of Analysis)4-271

Exhibit 4-135: Estimated Annualized Incremental Cost per CWS - Low Scenario (2022 USD)	4-273

Exhibit 4-136: Estimated Annualized Incremental Cost per CWS - High Scenario (2022 USD)	4-275

Exhibit 4-137: Estimated Annualized Incremental Cost per NTNCWS - Low Scenario (2022 USD)	4-277

Exhibit 4-138: Estimated Annualized Incremental Cost per NTNCWS - High Scenario (2022 USD)	4-278

Exhibit 4-139: Estimated Annualized Incremental Cost per Household - Low Scenario (2022 USD)....4-281
Exhibit 4-140: Estimated Annualized Incremental Cost per Household - High Scenario (2022 USD)...4-283

Exhibit 4-141: State Cost Components, Subcomponents, and Activities Organized by Section1	4-286

Exhibit 4-142: State Administration Activities and Unit Burden Estimates (Occur during Years 1 through

5)	4-290

Exhibit 4-143: Estimated Burden for States to Provide Staff Training during Years 1 through 5	4-291

Exhibit 4-144: State Annual Administration Activities and Unit Burden Estimates	4-291

Exhibit 4-145: State Administration and Rule Implementation Cost Estimation in SafeWater LCR (by

Activity)1	4-293

Exhibit 4-146: State Lead Tap Sampling Burden Estimates	4-295

Exhibit 4-147: Burden to Review Lead Tap Sampling Results and 90th Percentile Level	4-298

Exhibit 4-148: State Lead Tap Sampling Unit Cost Estimation in SafeWater LCR by Activity1,2	4-299

Exhibit 4-149: State Lead WQP Monitoring Burden Estimates	4-305

Exhibit 4-150: State Lead WQP Monitoring Cost Estimation in SafeWater LCR by Activity1	4-306

Exhibit 4-151: State Copper WQP Monitoring Burden Estimates	4-307

Exhibit 4-152: State Copper WQP Monitoring Cost Estimation in SafeWater LCR by Activity1	4-308

Exhibit 4-153: State Source Monitoring Burden Estimates	4-309

Exhibit 4-154: State Source Water Monitoring Cost Estimation in SafeWater LCR by Activity1	4-309

Exhibit 4-155: State School Sampling Burden Estimates	4-310

Exhibit 4-156: State School and Child Care Facility Sampling Cost Estimation in SafeWater LCR by

Activity1-2	4-312

Exhibit 4-157: State CCT Installation Related Burden Estimates	4-313

Exhibit 4-158: Estimated Burden for States to Review Initial CCT Study	4-313

Exhibit 4-159: Estimated Burden for State Review to Set OWQPs	4-314

Exhibit 4-160: State CCT Installation Cost Estimation in SafeWater LCR by Activity1	4-315

Exhibit 4-161: State CCT Re-Optimization-Related Burden Estimates	4-315

Exhibit 4-162: Estimated Burden for States to Review a Revised CCT Study and Determine Needed CCT

Adjustment	4-316

Exhibit 4-163: State CCT Re-optimization Cost Estimation in SafeWater LCR by Activity1	4-316

Exhibit 4-164: State DSSA Burden Estimates	4-317

Exhibit 4-165: State CCT DSSA Cost Estimation in SafeWater LCR by Activity1,2	4-318

Exhibit 4-166: State CCT Installation Related Burden Estimates	4-318

Exhibit 4-167: Estimated State Burden to Review CCT-Related Data during Sanitary Survey	4-319

Exhibit 4-168: Estimated Hours per System for State to Consult on Source Water Change	4-320

Exhibit 4-169: Estimated Hours per System for State to Consult on Treatment Change	4-321

Final LCR! Economic Analysis

xvi

October 2024


-------
Exhibit 4-170: State CCT Re-optimization Cost Estimation in SafeWater LCR by Activity1	4-321

Exhibit 4-171: State SL Inventory Burden Estimates	4-324

Exhibit 4-172: State SLR Plan and Deferred Replacement Deadline Review Burden Estimates	4-325

Exhibit 4-173: Estimated Additional Burden for States to Review the Initial SLR Plan for Systems

Requesting a Deferred Replacement Deadline	4-326

Exhibit 4-174: Estimated Annual Burden for States to Review SLR Plan Updates or Certifications of No

Changes	4-326

Exhibit 4-175: State Report Review Burden Estimates	4-327

Exhibit 4-176: State Burden to Review System's Annual Service Line Replacement Report (hrs per

system)	4-327

Exhibit 4-177: State Service Line Replacement Cost Estimation in SafeWater LCR by Activity1,2	4-328

Exhibit 4-178: State One-Time POU-Related Burden Estimates	4-330

Exhibit 4-179: Estimated Hours for State Review of POU Plan (hrs/system)	4-330

Exhibit 4-180: State Ongoing POU-Related Burden Estimates	4-332

Exhibit 4-181: State Burden to Review Annual POU Program Report (hours/system)	4-333

Exhibit 4-182: State POU Cost Estimation in SafeWater LCR (by Activity)1,2	4-334

Exhibit 4-183: PWS Burden for Consumer Notification	4-335

Exhibit 4-184: State Burden for Public Education Activities that Are Independent of Lead 90th Percentile

Levels	4-336

Exhibit 4-185: Unit Burden per CWS for States to Provide Phone Translation by Type of Public Education

Material	4-340

Exhibit 4-186: Unit Costs per CWS for States to Provide Written Translations by Type of Public Education

Material	4-341

Exhibit 4-187: State Public Education Burden in Response to Lead ALE	4-342

Exhibit 4-188: State Public Education Burden in Response to Multiple Lead ALE	4-344

Exhibit 4-189: State Lead Public Education Cost Estimation in SafeWater LCR (by Activity)1,2	4-345

Exhibit 4-190: Phosphorus Mass Balance Conceptual Model	4-349

Exhibit 4-191: Summary of Assumptions Used in Estimating Phosphorus Loading Increase	4-349

Exhibit 4-192: WWTP Status with Respect to Phosphorus Discharge Permit Limits	4-350

Exhibit 4-193: Summary of Assumptions Used in Estimating Phosphorus Removal Unit Cost	4-352

Exhibit 4-194: Estimated Nationwide Annual Phosphorus Reaching WWTPs after Implementation of the

LCRI under Low Cost Scenario	4-354

Exhibit 4-195: Estimated Nationwide Annual Phosphorus Reaching WWTPs after Implementation of the

LCRI under High Cost Scenario	4-354

Exhibit 4-196: Estimated Nationwide Annual Phosphorus Reaching Waterbodies after Implementation of

the LCRI under Low Cost Scenario	4-355

Exhibit 4-197: Estimated Nationwide Annual Phosphorus Reaching Waterbodies after Implementation of

the LCRI under High Cost Scenario	4-355

Exhibit 5-1: Tap Water Lead Concentration Sample Data: Source Citations, City Water System, LSL and

CCT Status Represented in the Data Source, and Number of Individual Sample Bottles per Source*	5-5

Exhibit 5-2: Diagram Showing Plumbing Where Water Can Become Contaminated with Lead	5-7

Exhibit 5-3: Example of a Complete Consecutive Liter Profile of Lead Concentrations in Tap Water from a
Location with a Lead Service Line	5-7

Final LCRI Economic Analysis

xvii

October 2024


-------
Exhibit 5-4: Summary Statistics for Tap Water Lead Concentrations by LSL and CCT Status Combinations,

Country, and Citation	5-9

Exhibit 5-5: Summary Statistics, Including Geometric Mean, Standard Deviation (SD), Maximum Value,
and Sample Size for Tap Water Lead Concentration Sample Data by LSL and CCT Status Used in

Statistical Modeling	5-11

Exhibit 5-6: Numeric Values Assigned to Two Discrete Contrast Variables Representing LSL Status in the

Estimated Drinking Water Lead Concentration Regression Model	5-12

Exhibit 5-7: Numeric Values Assigned to a Discrete Contrast Variable Representing CCT Status Use in the

Estimated Drinking Water Lead Concentration Regression Model	5-12

Exhibit 5-8: Comparison of Tap Sample Lead Concentration Model Results Based on Maximum

Likelihood Estimators for Goodness of Fit	5-15

Exhibit 5-9: Results from the Reduced Cubic Spline Interaction Model with CCT Interactions: Fixed Effects

and Random Effects for Sampling Event, Site, and City Water System	5-15

Exhibit 5-10: Estimates for the Simulated Data Showing the Relationship between Tap Lead

Concentration and Profile Liter for Each Combination of CCT and LSL Status	5-17

Exhibit 5-11: LSL and CCT Scenarios and Simulated Geometric Mean Tap Water Lead Concentrations and
Standard Deviations for the First Ten Liters Drawn after Stagnation for Each Combination of LSL and CCT

Status	5-17

Exhibit 5-12: Mapping Simulated Drinking Water Lead Tap Concentrations to Benefit Scenarios	5-20

Exhibit 5-13: Summary of Daily Water Consumption Inputs for Drinking Water Consumption in SHEDS-Pb

Coupling (Zartarian et al., 2017)	5-25

Exhibit 5-14: Summary of Daily Inputs for Dietary Lead Intake (ng/day) in SHEDS-Pb (Zartarian et al.

(2017))	5-26

Exhibit 5-15: Summary of Inputs for Soil and Dust Lead Concentration (ng/gram) in SHEDS-Pb Coupling

(USHUD 2011, 2021)	5-26

Exhibit 5-16: Summary of Daily Inputs for Soil/Dust Ingestion (mg/day) in SHEDS-Pb (Ozkaynak et al.,

2022)	5-27

Exhibit 5-17: Default Lead Absorption Fractions across Media Used in SHEDS-Pb Model Runs	5-27

Exhibit 5-18: Age-Specific Polynomial Regressions Equations for Approximating IEUBK (Zartarian et al.,

2017)	5-28

Exhibit 5-19: Modeled SHEDS-Pb Geometric Mean (GM) Blood Lead Levels in Children for Each Possible

Drinking Water Lead Exposure Scenario for Each Year of Life	5-29

Exhibit 5-20: Anticipated Decreases in Blood Lead Levels in Children	5-30

Exhibit 5-21: Constant Variables Entered into the AALM for Both Sexes	5-32

Exhibit 5-22: Estimates of Blood Lead Levels in Adults Associated with Drinking Water Lead Exposures

from LSL/CCT or POU Combinations	5-36

Exhibit 5-23: Estimated Lifetime Average Blood Lead Level Decrease for Adults Experiencing Alternate

LSL/GRR, CCT, pitcher filter and POU Status Combinations	5-37

Exhibit 5-24: Comparison of Adjusted Coefficients from Lanphear et al. Erratum (2019) with Those
Obtained in the Kirrane and Patel (2014), and the Reanalysis and Independent Analysis of Lanphear et al.

(2005) by Crump et al. (2013)	5-40

Exhibit 5-25 Updated Estimates for Lifetime Earnings, Additional Education Costs, and Lost Earnings from

Additional Education (2022 USD), discounted at 2 percent to age 7	5-42

Exhibit 5-26: Present Value of Avoided ADHD Cases 2022 USD, Per Case	5-48

Final LCRI Economic Analysis

xviii

October 2024


-------
Exhibit 5-27: Association between a Change in Blood Lead Concentration and Birth Weight, Upstate New

York, 2003-2005 from Zhu et al. (2010)	5-50

Exhibit 5-28: Simulated Cost Changes (2010 USD) on Annual Medical Expenditures for Inpatient Hospital

Stays, using Birth Weight Spline Specifications (N with Positive Expenditures = 450)	5-51

Exhibit 5-29: Simulated Cost Changes (2010 USD) on Annual Medical Expenditures for Inpatient Hospital
Stays, for Birth Weight Indicator and a Pre-term Indicator Only Model (N with Positive Expenditures

= 450)	5-52

Exhibit 5-30: Distribution of Birth Weights in the United States	5-53

Exhibit 5-31: Inputs to the Health Impact Function Based on Selected Studies	5-55

Exhibit 5-32: Estimated National Annual Children's IQ Benefits, All PWSs, 2 Percent Discount Rate

(millions of 2022 USD)	5-59

Exhibit 5-33: Estimated National Annual Benefits of Avoided ADHD Cases, All PWSs, 2 Percent Discount

Rate (millions of 2022 USD)	5-60

Exhibit 5-34: Estimated National Annual Benefits of Low-Weight Births, All PWSs, 2 Percent Discount

Rate (millions of 2022 USD)	5-61

Exhibit 5-35: Estimated National Annual Benefits of Avoided from Cardiovascular Disease Premature

Mortalities, All PWSs, 2 Percent Discount Rate (millions of 2022 USD)	5-62

Exhibit 5-36: Estimated National Annual Benefits - 2 Percent Discount Rate (millions of 2022 USD) ....5-63

Exhibit 5-37: Uncertainties in the Benefits Analysis	5-64

Exhibit 5-38: Corrosion Control Treatment Total Annual Electricity Consumption by System Size and Type

of Chemical Addition	5-75

Exhibit 5-39: Emissions per MWh Calculated from Post-IRA 2022 IPM Reference Case	5-77

Exhibit 5-40: Greenhouse Gas Emission Values Per Mile or Gallon of Fuel	5-78

Exhibit 5-41: Estimated Emissions per CCT Installation	5-79

Exhibit 5-42: Estimated Emissions Per Service Line Replacement	5-80

Exhibit 5-43: Estimated Total Annual Incremental Greenhouse Gas Emissions for Final LCRI	5-80

Exhibit 5-44: Estimates of the Social Cost of C02, CH4, and N20, 2024-2058 (in 2022 USD)	5-84

Exhibit 5-45: Climate Disbenefits of the Final LCRI Low Scenario (millions of 2022 USD)	5-86

Exhibit 5-46: Climate Disbenefits of the Final LCRI High Scenario (millions of 2022 USD)	5-86

Exhibit 6-1: Estimated National Annualized Monetized Incremental Costs of the Final LCRI at 2 Percent

Discount Rate (millions of 2022 USD)	6-2

Exhibit 6-2: Wastewater Treatment Plants with Phosphorous Limits in 2024	6-4

Exhibit 6-3: Estimated National Annualized Monetized Benefits of the Final LCRI at 2 Percent Discount

Rate (millions of 2022 USD)	6-5

Exhibit 6-4: Comparison of Yearly Monetized National Incremental Costs to Benefits of the LCRI under

Low Scenario (millions 2022 USD)	6-8

Exhibit 6-5: Comparison of Yearly Monetized National Incremental Costs to Benefits of the LCRI under

High Scenario (millions 2022 USD)	6-9

Exhibit 6-6: Comparison of Estimated Monetized National Annualized Incremental Costs to Benefits of

the LCRI - 2 Percent Discount Rate (millions 2022 USD)	6-10

Exhibit 7-1: Estimated Change in Average Annual Net Burden and Costs for the Final LCRI ICR	7-4

Exhibit 7-2: Estimated Total Responses, Burden, and Costs for the Final LCRI ICR for Each Required
Activity	7-5

Final LCRI Economic Analysis

xix

October 2024


-------
Exhibit 7-3: Estimated Incremental Costs and Benefits of the Final LCRI by Small System Size Category - 2

Percent Discount Rate (millions of 2022 USD)	7-13

Exhibit 7-4: Estimated Incremental Costs vs. Revenue for Small CWSs - Low Scenario*	7-16

Exhibit 7-5: Estimated Incremental Costs vs. Revenue for Small CWSs - High Scenario*	7-17

Exhibit 7-6: Summary of Alternative Other Options Considered for the Final LCRI	7-19

Exhibit 7-7: Estimated Total Annualized Incremental Costs and Benefits at 2 Percent Discount Rate

(millions of 2022 Dollars)	7-22

Exhibit 7-8: Estimated Total Annualized Incremental Costs and Benefits for Small PWSs (< 10,000

people) at 2 Percent Discount Rate (millions of 2022 Dollars)	7-22

Exhibit 8-1: Summary of Alternative Other Options Considered for the Final LCRI	8-1

Exhibit 8-2: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and Alternative
Lead Action Level Option (AL < 0.015 mg/L) - High Scenario - 2 Percent Discount Rate (millions of 2022

USD)	8-4

Exhibit 8-3: Estimated National Annual Benefit Comparison Between the Final LCRI and Alternative Lead
Action Level Option (AL < 0.015 mg/L) - High Scenario - 2 Percent Discount Rate (millions of 2022 USD) 8-
5

Exhibit 8-4: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and Alternative
Lead Action Level Option (AL < 0.005 mg/L) - High Scenario - 2 Percent Discount Rate (millions of 2022

USD)	8-6

Exhibit 8-5: Estimated National Annual Benefit Comparison Between the Final LCRI and Alternative Lead
Action Level Option (AL < 0.005 mg/L) - High Scenario - 2 Percent Discount Rate (millions of 2022 USD) 8-
7

Exhibit 8-6: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and Alternative
Service Line Replacement Option (SLR Rate = 7%) - High Scenario - 2 Percent Discount Rate (millions of

2022 USD)	8-8

Exhibit 8-7: Estimated National Annual Benefit Comparison Between the Final LCRI and Alternative
Service Line Replacement Option (SLR Rate = 7%) - High Scenario - 2 Percent Discount Rate (millions of

2022 USD)	8-9

Exhibit 8-8: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and Alternative
Option Including Lead Connectors in Definition of Service Lines to be Replaced - High Scenario - 2

Percent Discount Rate (millions of 2022 USD)	8-10

Exhibit 8-9: Estimated National Annual Benefit Comparison Between the Final LCRI and Alternative
Option Including Lead Connectors in Definition of Service Lines to be Replaced - High Scenario - 2

Percent Discount Rate (millions of 2022 USD)	8-11

Exhibit 8-10: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and
Alternative Deferred Deadline Option (Adding Max Rate of 10,000 SL Per Year) - High Scenario - 2

Percent Discount Rate (millions of 2022 USD)	8-12

Exhibit 8-11: Estimated National Annual Benefit Comparison Between the Final LCRI and Alternative
Deferred Deadline Option (Adding Max Rate of 10,000 SL Per Year) - High Scenario - 2 Percent Discount

Rate (millions of 2022 USD)	8-13

Exhibit 8-12: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and
Alternative Deferred Deadline Option (Adding Max Rate of 8,000 SL Per Year) - High Scenario - 2 Percent
Discount Rate (millions of 2022 USD)	8-14

Final LCRI Economic Analysis

xx

October 2024


-------
Exhibit 8-13: Estimated National Annual Benefit Comparison Between the Final LCRI and Alternative
Deferred Deadline Option (Adding Max Rate of 8,000 SL Per Year) - High Scenario - 2 Percent Discount

Rate (millions of 2022 USD)	8-15

Exhibit 8-14: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and
Alternative Temporary Filters Program for Multiple ALE Systems Option (Filters Made Available to Lead,
GRR, and Unknown Service Line Customers Only) - High Scenario - 2 Percent Discount Rate (millions of

2022 USD)	8-16

Exhibit 8-15: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and
Alternative Temporary Filters Program for Multiple ALE Systems Option (Deliver Filters to All Customers)

- High Scenario - 2 Percent Discount Rate (millions of 2022 USD)	8-18

Exhibit 8-16: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and
Alternative Small System Flexibility Option (Flexibility for CWSs Serving up to 10,000 Persons) - High

Scenario - 2 Percent Discount Rate (millions of 2022 USD)	8-20

Exhibit 8-17: Estimated National Annual Benefit Comparison Between the Final LCRI and Alternative
Small System Flexibility Option (Flexibility for CWSs Serving up to 10,000 Persons) - High Scenario - 2

Percent Discount Rate (millions of 2022 USD)	8-21

Exhibit 8-18: Estimated National Annualized Rule Cost, Benefit, and Net Benefit Comparison Between
the Final LCRI and Alternative Options Considered - High Scenario - 2 Percent Discount Rate (millions of
2022 USD)	8-22

Final LCRI Economic Analysis

xx i

October 2024


-------
List of Acronyms

Hg/dL

micrograms per deciliter

Hg/L

micrograms per liter

AALM

All Ages Lead model

ACS

American Community Survey

ADF

Average daily flow

ADHD

Attention-deficit/hyperactivity disorder

AHHS

American Healthy Homes Survey

Al

American Indian

AIC

Akaike's Information Criterion

AL

Action level

ALE

Action level exceedance

ANSI

American National Standards Institute

ANV

Alaska Native

ASDWA

Association of State Drinking Water Administrators

ATSDR

Agency for Toxic Substances and Disease Registry

AWIA

America's Water Infrastructure Act

AWWA

American Water Works Association

BIC

Bayesian information criterion

BIL

Bipartisan Infrastructure Law

BLL

Blood lead level

BLS

Bureau of Labor Statistics

CCR

Consumer Confidence Report

CCT

Corrosion control treatment

CCTV

Closed circuit television

CDC

Centers for Disease Control and Prevention

CED

Committee for Economic Development

CFR

Code of Federal Regulations

CFSAN

Center for Food Safety and Applied Nutrition

CHAD

Consolidated Human Activity Database

CI

Confidence interval

CND

Canada

costs

Costs of State Transactions Study

cpi

Consumer price index

CVD

Cardiovascular disease

CWS

Community water system

cwss

Community Water System Survey

DF

Design flow

DISC

Diagnostic Interview Schedule for Children

DMR

Discharge monitoring report

DSM

Diagnostic and Statistical Manual of Mental Disorders

DSSA

Distribution System and Site Assessment

DWINSA

Drinking Water Infrastructure Needs Survey and Assessment

DWM

Department of Water Management

DWSRF

Drinking Water State Revolving Fund

EA

Economic analysis

Final LCRI Economic Analysis

xxii

October 2024


-------
ECI	Employment cost index

ECTT	Error code tracking tool

EGLE	Environment, Great Lakes, and Energy

EJ	Environmental justice

EO	Executive order

EP	Entry point

EPA	United States Environmental Protection Agency

FDA	Food and Drug Administration

FIML	Full-information maximum likelihood

FR	Federal Register

FRFA	Final regulatory flexibility analysis

FRN	Federal Register Notice

FTE	Full-time equivalent

FY	Fiscal year

GCWW	Greater Cincinnati Water Works

GHG	Greenhouse gas

Gl	Gastro-intestinal

GIS	Geographic information system

GM	Geometric mean

GRR	Galvanized requiring replacement

Ground water under the direct influence of surface water (GU, shortened from
GUDWI).

GW	Ground water

GWUDI	Ground water under the direct influence of surface water

HAB	Harmful algal blooms

HDPE	High-density polyethylene

HH	Households

HHS	Department of Health and Human Services

HMR	Heavym eta I s registry

HR	Hazard ratio

HRRCA	Health Risk Reduction and Cost Analysis

HUD	U.S. Department of Housing and Urban Development

ICR	Information collection request

ID	Identification

IEUBK	Integrated exposure uptake biokinetic

IPC	Internal plumbing code

IQ	Intelligence quotient

ISA	Integrated Science Assessment for Lead

LBW	Low birth weight

LCCA	Lead Contamination Control Act

LCR	Lead and Copper Rule

LCRI	Lead and Copper Rule Improvements

LCRMR	Lead and Copper Rule Minor Revisions

LCRR	Lead and Copper Rule Revisions

LCRWG	Lead and Copper Rule Working Group

LOD	Limit of detection

LSL	Lead service line

LSLR	Lead service line replacement

Final LCRI Economic Analysis

xxiii

October 2024


-------
LT	Long-Term

MCL	Maximum contaminant level

MCLG	Maximum contaminant level goal

MDL	Method detection limit

mg/L	milligrams per liter

MGD	Million gallons per day

NACCHO	National Association of County and City Health Officials

NAICS	North American Industry Classification System

NCES	National Center for Education Statistics

NDWAC	National Drinking Water Advisory Council

NGO	Non-government organization

NHANES	National Health and Nutrition Examination Survey

NHEXAS	National Human Exposure Assessment Survey

NIH	National Institutes of Health

NPDES	National Pollutant Discharge Elimination System

NPDWR	National Primary Drinking Water Regulation

NPNCWS Not-for-profit non-community water systems

NSF	NSF International

NTNCWS	Non-transient non-community water system

NTP	National Toxicology Program

NTTAA	National Technology Transfer and Advancement Act

OCCT	Optimal corrosion control treatment

OES	Occupational employment survey

OEWS	Occupational Employment and Wage Statistics

OGWDW Office of Ground Water and Drinking Water

OIRA	Office of Information and Regulatory Affairs

O&M	Operation and maintenance

OMB	Office of Management and Budget

OWQP	Optimal water quality parameter

P90	Lead 90th percentile level

Pb	Lead

PE	Public education

PN	Public notice

POE	Point-of-entry

POU	Point-of-Use

PQL	Practical quantitation limit

PR	Puerto Rico

PRA	Paperwork Reduction Act

PSA	Public service announcement

PVC	Polyvinyl chloride

PWD	Philadelphia Water Department

PWS	Public water system

PWSID	Public water system identification number

PWSS	Public water system supervision

OA	Quality assurance

QC	Quality control

REML	Restricted maximum likelihood

RFA	Regulatory Flexibility Act

Final LCRI Economic Analysis

xxiv

October 2024


-------
RLDWA

Reduction of Lead in Drinking Water Act

RTCR

Revised Total Coliform Rule

SAB

Science Advisory Board

SACCHO

State Associations of County and City Health Officials

SBA

Small Business Administration

SBAR

Small Business Advocacy Review

SBREFA

Small Business Regulatory Enforcement Fairness Act

SD

Standard deviation

SDWA

Safe Drinking Water Act

SDWIS

Safe Drinking Water Information System

SDWIS/Fed

Safe Drinking Water Information System/Federal version

SE

Standard error

SER

Small entity representatives

SHEDS

Stochastic Human Exposure and Dose Simulation

SISNOSE

Significant economic impact on a substantial number of small entities

SL

Service line

SLR

Service line replacement

SRF

State revolving fund

SS

Sums of squares

SW

Surface water

SYR3 ICR

Six-Year Review 3 Information Collection Request

TCR

Total Coliform Rule

TDS

Total Diet Study 2007-2013

TL

Trigger level

TLE

Trigger level exceedance

UCMR

Unregulated Contaminant Monitoring Rule

UMRA

Unfunded Mandates Reform Act

USA

United States of America

USD

United States Dollar

USEPA

Unites States Environmental Protection Agency

USGS

United States Geological Survey

USOMB

United States Office of Management and Budget

USPS

United States Postal Service

VLBW

Very low birth weight

VLS

Very large system

WBS

Work breakdown structure

WIC

Women, infants and children

WIFIA

Water Infrastructure Finance and Innovation Act

WIIN

Water Infrastructure Improvements for the Nation

WLL

Water lead levels

WPCA

Water Pollution Control Authority

WQP

Water quality parameter

WWTP

Wastewater treatment plant

Final LCRI Economic Analysis

xxv

October 2024


-------
Executive Summary

The final Lead and Copper Rule Improvements (LCRI) will significantly reduce the risk of exposure to lead
from drinking water. The rule builds on the pre-2021 Lead and Copper Rule (LCR), which was
promulgated in 1991 and last revised in 2007, and the 2021 Lead and Copper Rule Revisions (LCRR). The
EPA conducted a review of the 2021 LCRR in accordance with Executive Order (EO) 13990 and
announced its intention to strengthen the 2021 LCRR with this LCRI rulemaking. The final LCRI addresses
the priorities the EPA identified in the 2021 LCRR review and public comments received on the proposed
LCRI. The final rule includes strengthened requirements in priority areas and provides a fundamental
shift to a more protective lead drinking water rule. In this rule, the agency is finalizing requirements for
drinking water systems to replace lead and certain galvanized service lines. The final rule also removes
the lead trigger level (TL), reduces the lead action level (AL) to 0.010 mg/L, and strengthens tap sampling
procedures to improve public health protection and simplify implementation relative to the 2021 LCRR.
Further, this final rule strengthens corrosion control treatment (CCT), public education and consumer
awareness, requirements for small systems, and sampling in schools and child care facilities.

The final LCRI National Primary Drinking Water Regulation (NPDWR) is a significant regulatory action that
was submitted to the Office of Management and Budget (OMB) for review. An economic analysis (EA) is
required for all significant rules under EO 12866 (Regulatory Planning and Review), as amended by EO
14096. In addition, section 1412(b)(3)(C) of the Safe Drinking Water Act (SDWA) requires the EPA to
prepare a Health Risk Reduction and Cost Analysis (HRRCA). This EA addresses these and other
regulatory reporting requirements, including those that direct the EPA to conduct distributional and
environmental justice analysis. With respect to the SDWA HRRCA requirements, section 1412(b)(3)(C)(i)
lists the analytical elements of the required HRRCA as follows: (1) quantifiable and non-quantifiable
health risk reduction benefits; (2) quantifiable and non-quantifiable health risk reduction benefits from
reductions in co-occurring contaminants; (3) quantifiable and non-quantifiable costs that are likely to
occur solely as a result of compliance; (4) incremental costs and benefits of rule options; (5) effects of
the contaminant on the general population and sensitive subpopulations including infants, children,
pregnant women, the elderly, and individuals with a history of serious illness; (6) any increased health
risks that may occur as a result of compliance, including risks associated with co-occurring contaminants;
and (7) other relevant factors such as uncertainties in the analysis and factors with respect to the degree
and nature of the risk.

The entities potentially affected by the final LCRI are public water systems (PWSs) classified as either
community water systems (CWSs) or non-transient non-community water systems (NTNCWSs) and
primacy agencies (States). In the economic modeling performed, the EPA uses the Federal version of the
Safe Drinking Water Information System (SDWIS/Fed) to derive the number of CWSs and NTNCWSs,
49,529 and 17,418, respectively. The agency also assumed, for modeling purposes, 56 primacy agencies.1
In this EA, the EPA assumes that the final LCRI will be promulgated in 2024. The agency estimated the
year or years in which all costs and benefits accrue over a 35-year period of analysis. The 35-year
window was selected to capture costs associated with rule implementation as well as water systems

1 The 56 primacy agencies include 49 States (excluding Wyoming), Puerto Rico, Guam, United States Virgin Islands,
American Samoa, North Mariana Islands, and Navajo Nation. For cost modeling purposes, the EPA also included
the District of Columbia (D.C.) as a primacy agency when assigning burden and costs of the rule although some of
these costs are incurred by the actual primacy agency, EPA Region 3.

Final LCRI Economic Analysis

ES-1

October 2024


-------
conducting service line replacement (SLR) and installing and operating optimal corrosion control
treatment (OCCT).

The EPA annualized the estimated future streams of costs and benefits that accrue from compliance
activities occurring over this same period of analysis symmetrically. The EPA does not capture the effects
of compliance with the final LCRI after the end of the period of analysis, although, the agency does
account for benefits that continue to accrue in the future from compliance activities that occur during
the 35-year window. Costs and benefits are presented as annualized values in 2022 dollars. The EPA
determined the present value of these costs and benefits using a discount rate of two percent as
prescribed by the OMB Circular A-4 (OMB, 2023).

The EPA used its SafeWater LCR model to analyze the costs and benefits of the final LCRI. For a detailed
description of the model, see Chapter 5 . PWSs will face different compliance scenarios depending on
the size and type of the water system; the presence of lead, galvanized requiring replacement (GRR), and
unknown service lines; water quality; and existing corrosion controls. In addition, PWSs will also face
different unit costs based on water system baseline characteristics including size, type, and number of
entry points (e.g., labor rates, and CCT capital and operation and maintenance unit costs).

One of the strengths of the SafeWater LCR model is that it incorporates a large degree of variability
across water system baseline characteristics that influence compliance and costs. One limitation of the
cost-benefit analysis is that the EPA does not have all of the PWS-specific data needed to fully reflect
baseline and compliance variability across PWSs; therefore, the SafeWater LCR model applies a "model
PWS" approach. The SafeWater LCR model creates model PWSs that represent systems in each, of 72
PWS categories, by combining the PWS-specific data available in SDWIS/Fed with data on baseline and
compliance characteristics available at the PWS category level. When categorical data are point
estimates, every model PWS in a category is assigned the same value. When the EPA has probabilistic
data representing system variability, the SafeWater LCR model assigns each model PWS a value sampled
from the distribution.

Chapter 3 describes in detail the baseline data elements, their derivations, and the inherent sources of
uncertainty in the developed data elements. The EPA estimates the incremental costs and benefits of the
final LCRI relative to a baseline, as described in Chapter 3, that assumes compliance with the 2021 LCRR
and other State regulations requiring lead service line replacement (Illinois, Michigan, New Jersey, and
Rhode Island) and tap sampling in schools and child cares (17 States and the District of Columbia) that go
beyond the 2021 LCRR requirements.

As described in Chapter 4, the EPA determined it does not have enough information to perform a
probabilistic uncertainty analysis as part of the SafeWater LCR model analysis for this rule. Instead, to
capture uncertainty, the EPA estimated compliance costs (and benefits) by running the SafeWater LCR
model under low and high bracketing scenarios. For costs, the bracketing scenarios are defined by the
following three cost drivers:

1.	Likelihood a model PWS will exceed the lead AL and/or TL under the 2021 LCRR and the AL
under the final LCRI.

2.	SLR unit costs.

3.	CCT unit costs.

The low and high benefits bracketing scenarios are defined by the following benefits variables:

Final LCRI Economic Analysis

ES-2

October 2024


-------
1.	Likelihood a model PWS will exceed the AL and/or TL under the 2021 LCRR and the AL under the
final LCRI (also used to define the low and high cost scenarios in the cost analysis).

2.	The concentration-response functions that characterize how reductions in blood lead levels
(caused by changes in lead exposure) translate into avoided intelligence quotient (IQ) reductions,
cases of Attention-deficit/hyperactivity disorder (ADHD), and cardiovascular disease (CVD)
premature mortality.

3.	Two alternative low and high valuations for the ADHD cost of illness.

The EPA expects the significant portion of potential uncertainty is captured by this bracketing approach.
However, some uncharacterized uncertainties still exist which may result in cost and benefit estimates
that fall outside of the range of costs and benefits described in the bracketing model results. All
significant limitations and uncertainties of this economic analysis are described in the following chapters,
particularly sections 3.4, 4.2.2, and 5.7.

National Estimated Costs

In order to estimate the incremental national cost of the final LCRI, the EPA estimated the additional
costs that PWSs, households, and States will incur in response to the final LCRI, above the cost they
would face under the 2021 LCRR if the LCRI was not enacted. The EPA developed estimates of the LCRI
regulatory requirement costs that accrue to PWSs for the following cost components: rule
implementation and administration, sampling, service line inventory and replacement, CCT, point-of-use
program (if a small system selects this compliance option), and public education and outreach. For each
of these six categories of PWS cost the EPA also estimates State oversite costs. In Chapter 4, the EPA
provides the data and algorithms used to calculate the cost of each activity that PWSs and States will
undertake to comply with the final rule.

The EPA estimates that the final LCRI CCT requirements will result in systems adding orthophosphate to
their finished water to creates a protective inner coating on pipes that can inhibit lead leaching.

However, once phosphate is added to a public water distribution system, some of this incremental
loading remains in the water stream as it flows into wastewater treatment plants (WWTPs) downstream.
This generates treatment costs for certain WWTPs. Due to many water systems operating both the
wastewater and drinking water systems, the EPA evaluated the costs of additional phosphate usage for
informational purposes. Because these costs are associated with wastewater treatment to meet Clean
Water Act regulatory requirements, they are not "likely to occur solely as a result of compliance" with
the final LCRI, and, therefore, are not costs considered as part of the HRRCA under SDWA, section
1412(b)(3)(C)(i)(lll).

Exhibit ES provides the estimated incremental monetized costs of the final LCRI, for both the low and
high scenarios, at a 2 percent discount rate, in millions of 2022 dollars.2 Total annualized monetized
incremental costs for the final LCRI range from $1.5 to $2.0 billion, in 2022 dollars discounted at 2
percent.

2 Note that the incremental national costs of the final LCRI when compared to the pre-2021 LCR have also been
computed and are provided in Appendix C.

Final LCRI Economic Analysis

ES-3

October 2024


-------
Exhibit ES-1: Estimated National Annualized Monetized Incremental Costs of the Final LCRI at

2 Percent Discount Rate (millions of 2022 USD)





Low Estimate





High Estimate





Baseline

LCRI Incremental

Baseline

LCRI

ncremental

PWS Annual Costs













Sampling

$134.0

$166.0

$32.0

$143.6

$176.2

$32.6

PWS SLR*

$84.6

$1,259.0

$1,174.4

$124.5

$1,763.9

$1,639.4

Corrosion Control
Technology

Point-of Use Installation
and Maintenance

$552.0
$2.4

$591.1
$5.1

$39.1
$2.7

$647.8
$5.9

$692.9
$9.6

$45.1
$3.7

Public Education and
Outreach

$69.6

$267.3

$197.7

$72.1

$302.2

$230.1

Rule Implementation and
Administration

$0.1

$3.4

$3.3

$0.2

$3.4

$3.2

Total Annual PWS Costs

$842.7

$2,291.9

$1,449.2

$994.1

$2,948.2

$1,954.1

Household SLR Costs**

$8.1

$0.0

-$8.1

$26.4

$0.0

-$26.4

State Rule Implementation
and Administration

$38.4

$66.1

$27.7

$41.8

$67.6

$25.8

Wastewater Treatment
Plant Costs***

$3.0

$3.0

$0.0

$4.8

$5.1

$0.3

Total Annual Rule Costs

$892.2

$2,361.0

$1,468.8

$1,067.1

$3,020.9

$1,953.8

Acronyms: LCRI = Lead and Copper Rule Improvements; SLR = service line replacement; PWS = public water
system; USD = United States dollars.

Notes: Previous baseline costs are projected over the 35-year period of analysis and are affected by the EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

*Service line replacement (SLR) includes full and partial lead service lines and galvanized requiring replacement
service lines.

**The EPA in the 2021 LCRR economic analysis (USEPA, 2020) assumed that the cost of customer-side SLRs made
under the goal-based replacement requirement would be paid for by households. The agency also assumed that
system-side SLRs under the goal-based replacement requirement and all SLRs (both customer-side and systems-
side) would be paid by the PWS under the 3 percent mandatory replacement requirement. The EPA made these
modeling assumptions based on the different levels of regulatory responsibility systems faced operating under a
goal-based replacement requirement versus a mandatory replacement requirement. While systems would not be
subject to a potential violation for not meeting the replacement target under the goal-based replacement
requirement, under the 3 percent mandatory replacement requirement the possibility of a violation could
motivate more systems to meet the replacement target even if they had to adopt customer incentive programs
that would shift the cost of replacing customer-side service lines from customers to the system. To be consistent
with these 2021 LCRR modeling assumptions, under the final LCRI, the EPA assumed that mandatory replacement
costs would fall only on systems. Therefore, the negative incremental values reported for the "Household SLR
Costs" category do not represent a net cost savings to households. They represent an assumed shift of the
estimated SLR costs from households to systems. The EPA has insufficient information to estimate the actual SLR
cost sharing relationship between customers and systems at the national level of analysis.

***Due to many water systems operating both the wastewater and drinking water systems, the EPA is evaluating
the costs of additional phosphate usage for informational purposes. These costs are not "likely to occur solely as a
result of compliance" with the final LCRI, and therefore are not costs considered as part of the Health Risk
Reduction and Cost Analysis (HRRCA) under the Safe Drinking Water Act (SDWA), section 1412(b)(3)(C)(i)(lll).

Final LCRI Economic Analysis

ES-4

October 2024


-------
The final LCRI is expected to result in additional phosphate being added to drinking water to reduce the
amount of lead leaching into the water in the distribution system. Although the downstream ecological
impacts are not costs considered as part of the HRRCA under the SDWA, section 1412(b)(3)(C)(i)(lll), the
EPA, for informational purposes, has quantified incremental phosphorus loadings and outlined potential
downstream ecological impacts. The EPA estimated that, nationwide, the final LCRI may result in post
WWTP total incremental phosphorus loads to receiving waterbodies increasing over the period of
analysis, under the low and high scenarios, by a range of 225,000 to 272,000 pounds fifteen years after
promulgation, and by a range of 216,000 to 260,000 pounds at Year 35. At the national level, under the
high scenario, this additional phosphorus loading to waterbodies is relatively small, less than 0.03
percent of the total phosphorous load deposited annually from all other anthropogenic sources.

However, national average national average receiving waterbody phosphorous load impacts may obscure
significant localized ecological impacts. Impacts, such as eutrophication, may occur in water bodies
without restrictions on additional phosphate loadings, or in locations with existing elevated phosphate
levels.

The EPA also notes that there exist unquantified costs associated with SLR. Costs associated with the
disruption of normal traffic patterns in communities implementing SLR programs are not accounted for
in the monetized cost estimates of the rule. This impact to traffic could be significant in localized areas
where lead, GRR, and unknown service lines are co-located with high traffic roads. During SLR worksite
activities and characteristics have the potential to increase car and pedestrian accidents. Also given the
necessity to shut off water service to buildings and residences during SLR the probability of fire damage
and negative health/sanitation impacts may increase.

National Estimated Benefits

Estimated benefits, in terms of health risk reduction from the final LCRI, result from the activities
performed by water systems, which are expected to reduce risk to the public from exposure to lead and
copper in drinking water at the tap. The EPA quantifies and monetizes some of this health risk reduction
from lead exposure by estimating the decrease in lead exposures accruing to both children and adults
from the installation and re-optimization of OCCT, SLR, the implementation of point-of-use filter devices,
and the provision of pitcher filters in systems with multiple action level exceedances and by quantifying
and monetizing the resulting increases in IQ in children zero to seven years old, ADHD in older children,
reductions in incidents of low birth weight, and adult CVD premature mortality. For a detailed discussion
of the estimation of national incremental annualized benefits see Chapter 5.

Total estimated incremental monetized annualized benefits for these four health endpoints range from
$13.5 to $25.1 billion, in 2022 dollars discounted at a 2 percent discount rate. See Exhibit ES for
additional detail.

Final LCRI Economic Analysis

ES-5

October 2024


-------
Exhibit ES-2: Estimated National Annualized Monetized Benefits of the Final LCRI at 2 Percent

Discount Rate (millions of 2022 USD)

Low Estimate

High Estimate



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual Child













Cognitive
Development

$1,208.5

$6,831.3

$5,622.8

$3,279.0

$10,963.0

$7,684.0

Benefits













Annual Low-Birth
Weight Benefits

$1.0

$5.4

$4.4

$1.8

$5.7

$3.9

Annual ADHD
Benefits

$33.6

$196.3

$162.7

$179.9

$599.5

$419.6

Annual Adult CVD













Premature

$1,750.7

$9,454.3

$7,703.6

$8,174.9

$25,210.0

$17,035.1

Mortality Benefits













Total Annual
Benefits

$2,993.8

$16,487.3

$13,493.5

$11,635.6

$36,778.2

$25,142.6

Acronyms: ADHD = Attention-Deficit/Hyperactivity Disorder; CVD = cardiovascular disease; LCRI = Lead and Copper
Rule Improvements; USD = United States dollar.

While the EPA is not required by SDWA 1412(b)(3)(C)(i)(lII) to consider climate disbenefits under the
HRRCA, the agency has estimated, solely for the purpose of complying with EO 12866, the potential
climate disbenefits caused by increased greenhouse gas (GHG) emissions associated with the operation
of CCT at drinking water treatment facilities and the use of construction and transport vehicles in the
replacement of lead and GRR service lines. The estimated monetized annualized disbenefits range from
$2.1 million under the low scenario to $2.0 million under the high scenario discounted at 2 percent, in
2022 dollars.

In addition to the monetized benefits of the final LCRI, there are several other benefits that are not
quantified. The EPA focused its non-quantified impacts assessment on the endpoints identified using two
comprehensive United States Government documents summarizing the literature on lead exposure
health impacts. These documents are the EPA's Integrated Science Assessment for Lead (ISA) (USEPA,
2024), and the United States Department of Health and Human Services' National Toxicology Program
(NTP) Monograph on Health Effects of Low-Level Lead (NTP, 2012). The risk of adverse health effects due
to lead exposure that are expected to decrease as a result of the final LCRI are summarized in Appendix
D and are expected to affect both children and adults. These endpoints include CVD morbidity effects,
renal effects, reproductive and developmental effects (apart from ADHD and low birth weight initial
hospitalization), immunological effects, neurological effects (apart from children's IQ), and cancer.

There are a number of final LCRI requirements that reduce lead exposure to both children and adults
that the EPA could not quantify. New public education and expanded service line inventory information
requirements will lead to additional averting behavior on the part of the exposed public, resulting in
reductions in the negative impacts of lead.

Final LCRI Economic Analysis

ES-6

October 2024


-------
The EPA did not quantify the CCT benefits of reduced lead exposure from lead-containing plumbing
components (not including from lead and/or GRR service lines) to individuals who reside in both: 1)
homes that have lead and/or GRR service lines but also have other lead-containing plumbing
components, and 2) those that do not have lead and/or GRR service lines but do have lead-containing
plumbing components.3 The EPA has determined that the final LCRI requirements may result in reduced
lead exposure to the occupants of both these types of buildings as a result of improved monitoring and
additional actions to optimize CCT. In the analysis of the LCRI, the number of both homes served by lead
and/or GRR service lines and homes not served by lead and/or GRR service lines potentially affected by
water systems increasing their corrosion control during the 35-year period of analysis is 5.2 million in the
low scenario and 9.1 million in the high scenario. Some of these households may have leaded plumbing
materials apart from lead or GRR service lines, including lead connectors, leaded brass fixtures, and lead
solder. These households could potentially see reductions in tap water lead concentrations. Also,
because of the lack of granularity in the lead tap water concentration data available to the EPA for the
regulatory analysis, the benefits of small improvements in CCT to individuals residing in homes with lead
content service lines, like those modeled under the Distribution System and Site Assessment
requirements, are not quantified.

Non-quantified cobenefits also exist when the corrosion inhibitors used by systems that are required to
install or re-optimize CCT as a result of the final LCRI result in increased useful life of the plumbing
components and appliances (e.g., water heaters), reduced maintenance costs, reduced treated water
loss from the distribution system due to leaks, and reduced potential liability and damages from broken
pipes in buildings that receive treated water from the system (Levin, 2023). The replacement of GRR
service lines may also lead to reduced treated water loss from the distribution system due to leaks
(AwwaRF and DVGW-Technologiezentrum Wasser, 1996).

Additionally, the risk of adverse health effects associated with copper that are expected to be reduced by
the final LCRI are summarized in Appendix E. These risks include acute gastrointestinal symptoms, which
are the most common adverse effect observed among adults and children. In sensitive groups, there may
be reductions in chronic hepatic effects, particularly for those with rare conditions such as Wilson's
disease and children predisposed to genetic cirrhosis syndromes. These diseases disrupt copper
homeostasis, leading to excessive accumulation that can be worsened by excessive copper ingestion
(NRC, 2000).

Comparison of Nation Costs and Benefits

Exhibit ES compares the estimated annualized monetized incremental costs and the estimated
annualized monetized incremental benefits of the final LCRI at a 2 percent discount rate; the monetized
net annualized incremental benefits range from $12.0 billion to $23.2 billion.

3 Although the EPA estimated an average lead concentration for the first 10 liters of drinking water to inform the
water lead concentration estimates used to quantify benefits the EPA could not calculate the CCT benefits
associated with lead containing plumbing components (apart from lead and/or LSL/GRR service lines), because the
EPA used a pooled estimate for all CCT conditions in residences with no lead and/or LSL/GRR service lines in place
(See Chapter 5, section 5.2.3 for additional information).

Final LCRI Economic Analysis

ES-7

October 2024


-------
Exhibit ES-3: Comparison of Estimated Monetized National Annualized Incremental Costs to
Benefits of the LCRI - 2 Percent Discount Rate (millions 2022 USD)



Low Scenario

High Scenario

Annualized Incremental Costs

$1,468.8

$1,953.8

Annualized Incremental Benefits

$13,493.5

$25,142.6

Annual Net Benefits

$12,024.7

$23,188.8

Acronyms: LCRI = Lead and Copper Rule Improvements; USD = United States dollar.

References

AwwaRF and DVGW-Technologiezentrum Wasser. 1996. Internal Corrosion of Water Distribution
Systems. 2nd edition. AwwaRF Order 90508. Project #725. AWWA Research Foundation (now Water
Research Foundation) and AWWA. Denver, CO.

Executive Order 12866. 1993. Regulatory Planning and Review. Federal Register. 58 FR 51735, October
4, 1993. Available at https://www.reginfo.gov/public/isp/Utilities/EO 12866.pdf.

Executive Order 13990. 2021. Executive Order on Protecting Public Health and the Environment and
Restoring Science to Tackle the Climate Crisis. January 20, 2021. https://www.whitehouse.gov/briefing-
room/presidential-actions/2021/01/20/executive-order-protecting-public-health-and-environment-and-
restoring-science-to-tackle-climate-crisis/.

Executive Order 14094. Modernizing Regulatory Review. April 6, 2023. Federal Register. 88 FR 21789.
https://www.govinfo.gov/content/pkg/FR-2023-04-ll/pdf/2023-07760.pdf

Levin, R., and J. Schwartz. 2023. A better cost:benefit analysis yields better and fairer results: EPA's lead
and copper rule revision. Environmental Research 229: 115738.
https://doi.Org/10.1016/j.envres.2023.115738

National Toxicology Program (NTP). 2012. NTP Monograph: Health Effects of Low-Level Lead. U.S.
Department of Health and Human Services. Office of Health Assessment and Translation. Division of the
National Toxicology Program. Available at

https://ntp.niehs.nih.gov/ntp/ohat/lead/final/monographhealtheffectslowlevellead_newissn_508.pdf.

National Research Council (NRC). 2000. Copper in Drinking Water. Washington, D.C.: The National
Academies Press.

OMB. (2023). Circular A-4. November 9, 2023. Retrieved from https://www.whitehouse.gov/wp-
content/uploads/2023/ll/CircularA-4.pdf

USEPA. 2020. Economic Analysis for the Final Lead and Copper Rule Revisions. December 2020. Office of
Water. EPA 816-R-20-008.

USEPA. 2024. Integrated Science Assessment for Lead (Final Report). U.S. Environmental Protection
Agency, Washington, DC. EPA/600/R-23/375.

Final LCRI Economic Analysis

ES-8

October 2024


-------
1 Introduction

Exposure to lead can cause harmful health effects for people of all ages, especially pregnant people,
infants, and young children (CDC, 2022a; CDC, 2022b; CDC, 2023). Lead has acute and chronic impacts
on the body. Lead exposure causes damage to the brain and kidneys and can interfere with the
production of red blood cells that carry oxygen to all parts of the body (ATSDR, 2020). Developing
fetuses, infants, and young children are most susceptible to the harmful health effects of lead (ATSDR,
2020). Exposure to lead is known to present serious health risks to the brain and nervous system of
children (USEPA, 2013; USEPA, 2024a). Young children and infants are particularly vulnerable to the
physical, cognitive, and behavioral effects of lead due to their sensitive developmental stages. There is
no known safe level of exposure to lead. Scientific studies have demonstrated that there is an increased
risk of health effects in children even when their blood lead levels are less than 3.5 micrograms per
deciliter (CDC, 2022c) and in adults even when blood lead levels are less than 10 micrograms per
deciliter (NTP, 2012). Low-level lead exposure is of particular concern for children because their growing
bodies absorb more lead than adults do, and their brains and nervous systems are more sensitive to the
damaging effects of lead (ATSDR, 2020). Sources of lead include, but are not limited to, lead-based paint,
drinking water, and soil contaminated by historical sources.

The U.S. Environmental Protection Agency (EPA) has taken several steps over the past 40 years to
reduce lead exposure through drinking water. To reduce the amount of lead in plumbing materials,
Congress prohibited the use or introduction into commerce of pipes and pipe fittings and fixtures that
contained more than 8 percent lead as well as solder or flux that contained more than 0.2 percent of
lead in 1986. Up until that time, lead was widely used in plumbing materials. Because lead service lines
(LSLs) were typically constructed with a maximum diameter of two inches (LSLR Collaborative, n.d.), it is
highly unlikely that there are lead water mains. Water mains are typically 6 to 16 inches in diameter,
whereas service lines have a smaller diameter. The common water main materials include ductile iron,
polyvinyl chloride (PVC), asbestos cement, high-density polyethylene (HDPE), and concrete steel
(Folkman, 2018).

The EPA estimates there are about 9.0 million LSLs in communities nationwide (USEPA, 2024b) in
addition to potentially millions of older buildings with lead solder and faucets that contain lead. In 2011,
Congress passed the Reduction of Lead in Drinking Water Act (RLDWA) revising the definition of lead
free by lowering the maximum lead content of the wetted surfaces of plumbing products (such as pipes,
pipe fittings, plumbing fittings and fixtures) from 8 percent to a weighted average of 0.25 percent,
establishing a statutory method for the calculation of lead content. On September 1, 2020, the EPA
published the final rule: Use of Lead Free Pipes, Fittings, Fixtures, Solder, and Flux for Drinking Water
that made conforming changes to regulations consistent with the RLDWA and which also requires that
manufacturers or importers certify that their products meet the requirements using a consistent
verification process (USEPA, 2020).

The EPA first promulgated a National Primary Drinking Water Regulation (NPDWR) for Lead and Copper
(LCR) in 1991 (56 FR 26460; USEPA, 1991). The LCR is a treatment technique rule that requires systems
to monitor drinking water at customer taps. If lead concentrations exceed an action level of 0.015
milligrams per liter (mg/L) or copper concentrations exceed an action level of 1.3 mg/L in more than 10
percent of customer taps sampled (90th percentile level), the LCR required systems to undertake

Final LCR! Economic Analysis

1-1

October 2024


-------
corrosion control treatment (CCT) steps. Following a lead action level exceedance, the LCR also required
the system to inform the public about steps they should take to protect their health and replace LSLs
after installing CCT and/or source water treatment. On January 12, 2000, the EPA promulgated Minor
Revisions to the LCR (LCRMR) (65 FR 1950; USEPA, 2000). These Minor Revisions streamlined the LCR,
promoted consistent national implementation, and reduced the reporting burden on affected entities.
The EPA promulgated the Short-Term Revisions in 2007 to improve implementation of the rule (72 FR
57782; USEPA, 2007). For additional information on the EPA's statutory and regulatory actions related to
the LCR, refer to Chapter 2, Section 2.1. The EPA is also committed to assisting schools and child care
facilities with testing for lead in drinking water through the 3Tsfor Reducing Lead in Drinking Water in
Schools and Child Care Facilities: A Training, Testing, and Taking Action Approach (Revised Manual)
(USEPA, 2018).

On January 15, 2021, the EPA published in the Federal Register the "National Primary Drinking Water
Regulation: Lead and Copper Rule Revisions" (86 FR4198; USEPA, 2021a) (2021 LCRR) with an effective
date of March 16, 2021, and a compliance date of January 16, 2024. The 2021 LCRR sought to better
identify areas with the greatest potential for lead contamination, strengthen CCT requirements,
accelerate and strengthen lead service line replacement (LSLR), expand consumer awareness and
improve risk communication, and require systems to offer lead-in-water testing and public education in
schools and child care facilities. On June 16, 2021, the EPA published the agency's decision to delay the
effective and compliance dates of the 2021 LCRR (86 FR 31939; USEPA, 2021b) to allow time for the EPA
to review the rule in accordance with Presidential directives issued on January 20, 2021 (Executive Order
13990) and conduct important consultations with affected parties. Based on this review, the EPA
decided to proceed with developing a rule, known as the Lead and Copper Rule Improvements (LCRI),
that would revise certain key sections of the 2021 LCRR while allowing the rule to take effect. In the
2021 LCRR review, the EPA noted that it does not intend to make any changes to the initial inventory
requirements in the LCRI. Additionally, the review highlighted other nonregulatory actions outside of the
Safe Drinking Water Act (SDWA) framework to reduce exposure to lead in drinking water, including
funding, targeted technical assistance, and risk communication tools.

The December 2021 Biden-Harris Lead Pipe and Paint Action Plan presented a multi-agency effort with a
goal of replacing all lead pipes over the following decade and providing support to local communities for
lead paint removal (The White House, 2021). The development of a final NPDWR, the LCRI, is a key
action of the Lead Pipe and Paint Action Plan. The aim of the plan is to mobilize resources from across
the federal government through funding made available from the Infrastructure Investment and Jobs
Act, also referred to as the Bipartisan Infrastructure Law (BIL), to reduce lead exposure from pipes and
paint containing lead. The BIL invested an unprecedented $50 billion in the nation's water and
wastewater infrastructure, including $15 billion dedicated to LSLR. The plan includes a goal of replacing
all LSLs in the nation and remediating lead paint.

In October 2022, the EPA published the Strategy to Reduce Lead Exposures and Disparities in U.S.
Communities (or "Lead Strategy") to "advance EPA's work to protect all people from lead with an
emphasis on high-risk communities" (USEPA, 2022). This agency-wide Lead Strategy promotes
environmental justice in communities challenged with lead and includes four key goals: (1) reduce
community exposures to lead sources; (2) identify communities with high lead exposures and improve
their health outcomes; (3) communicate more effectively with stakeholders; and (4) support and
conduct critical research to inform efforts to reduce lead exposures and related health risks. The

Final LCRI Economic Analysis

1-2

October 2024


-------
development of the LCRI is a key action within the EPA's Lead Strategy and "reflects EPA's commitment
to fulfilling the Biden-Harris Administration's historic commitment of resources to replace lead pipes and
support lead paint removal under the Lead Pipe and Paint Action Plan" (USEPA, 2022).

On December 6, 2023, the EPA published in the Federal Register the proposed regulation, "National
Primary Drinking Water Regulations for Lead and Copper Rule Improvements" (88 FR 84878; USEPA,
2023a). The proposal included advancements in protecting people from the health effects from
exposures to lead in drinking water. These advancements are based on the science and existing
practices utilized by drinking water systems. Key provisions in the proposal include requiring the vast
majority of all water systems across the country to replace lead and galvanized requiring replacement
(GRR) service lines regardless of lead levels within 10 years, locating legacy lead pipes, improving tap
sampling, lowering the lead action level, and strengthening protections to reduce exposure. The EPA
proposed to retain the 2021 LCRR requirements, and associated October 16, 2024 compliance date, for
the initial service line inventory, notifications to consumers served by a lead, GRR, or lead status
unknown service lines, Tier 1 public notification of a lead action level exceedance, and associated
reporting requirements.

The final LCRI addresses the priorities the EPA identified in the 2021 LCRR review, including the
equitable replacement of all LSLs in the nation, better identification of where LSLs are and action in
communities most at risk of lead exposure, and streamlined and improved implementation of the rule.
This final LCRI is the culmination of numerous meaningful consultations and engagements over several
years, including during the 2021 LCRR review, and in stakeholder outreach conducted to inform the
development of the proposed and final LCRI, along with almost 200,000 public comments submitted to
the docket for the proposed LCRI.

This economic analysis (EA) presents the evaluation of the benefits and costs of the final LCRI. The
analysis is performed in compliance with Executive Order 12866, Regulatory Planning and Review (58 FR
51735, October 4, 1993), as amended by Executive Order 14094 (88 FR 21879, April 6, 2023),
Modernizing Regulatory Review. These executive orders require the EPA to estimate the economic
impact of rules that have an annual effect on the economy of over $200 million, to make that analysis
available to the public for comment prior to publication of the final rule, and to consider ways to reduce
regulatory burden and maintain flexibility for the public. In addition, SDWA requires the EPA
Administrator to "publish and seek public comment on an analysis of the health risk reduction benefits
and costs likely to be experienced as the result of compliance with the treatment technique and
alternative treatment techniques that are being considered . . ." (SDWA section 1412(b)(3)(C)(ii)). The
EPA solicited public comment on all aspects of the data and analysis presented in the proposed EA and
associated Appendices as part of the public commenter period on the proposed LCRI.

This chapter provides a summary of the final LCRI in Section 1.1, outlines the organization of this EA in
Section 1.2, and provides information regarding supporting calculations and citations in Section 1.3.

1.1 Summary of the Final LCRI

The final LCRI will significantly reduce the risk of exposure to lead from drinking water. The rule builds
on the pre-2021 LCR (promulgated in 1991 and last revised in 2007) and the 2021 LCRR. The LCRI
addresses the priorities the EPA identified in the 2021 LCRR review and public comments received on

Final LCRI Economic Analysis

1-3

October 2024


-------
the proposed LCRI. The final rule includes strengthened elements of the rule in priority areas and
provides a fundamental shift to a more protective lead drinking water rule. The LCRI focuses on the
following key areas:

1.	Achieving Lead Pipe Replacement within 10 Years. The final LCRI requires mandatory full
service line replacement of lead and GRR service lines under the control of the system within 10
years unless the State4 sets a shortened deadline or the State approves a deferred deadline for
those systems that are eligible. The final LCRI retains the requirement for systems to replace
lead connectors when encountered and updates the requirements to develop a service line
replacement plan.

2.	Locating Legacy Lead Pipes. Knowing where lead pipes are located is critical to replacing them
efficiently and equitably. Under the final LCRI, all water systems are required to identify the
material of all service lines by the mandatory service line replacement deadline. Water systems
are required to make their service line inventories publicly available and to regularly update
them. In addition, water systems must use a validation process to ensure the service line
inventory is accurate. Water systems are also required to track lead connectors in their
inventories and replace them as they are encountered.

3.	Improving Tap Sampling. The final LCRI makes key changes to the protocol that water systems
must use for tap sampling informed by best practices already being deployed at the local and
State level. Water systems are required to collect first- and fifth-liter samples at sites with LSLs
and use the higher of the two values when determining compliance. This method will better
represent water that has been stagnant both within the LSL and the premise plumbing, helping
water systems better understand the effectiveness of their CCT.

4.	Lowering the Lead Action Level. The final LCRI lowers the lead action level from 0.015 mg/L to
0.010 mg/L. When a water system's lead sampling exceeds the action level, water systems are
required to inform the public and take actions associated with CCT and public education that
will reduce lead exposure, while concurrently working to replace all lead and GRR service lines.
For example, the system may be required to install or adjust CCT to reduce lead that leaches
into drinking water. While lowering the lead action level requires systems to take actions to
reduce lead exposure sooner, the EPA also emphasizes the many final rule requirements will
result in additional public health benefits irrespective of systemwide lead levels, recognizing
there is no safe level of lead in drinking water, including full service line replacement and other
public education provisions.

5.	Strengthening Protections to Reduce Exposure. The final LCRI requires water systems with
continually high lead levels to conduct additional outreach to consumers and make filters
certified to reduce lead available to all consumers. These additional actions can reduce
consumer exposure to higher levels of lead in drinking water while the water system works to

4 State means the agency of the State or Tribal government that has jurisdiction over public water systems. During
any period when a State or Tribal government does not have primary enforcement responsibility pursuant to
Section 1413 of the Public Health Service Act (as amended by the Safe Drinking Water Act, Public Law 93-523), the
term "State" means the Regional Administrator, U.S. Environmental Protection Agency (40 CFR §141.2).

Final LCRI Economic Analysis

1-4

October 2024


-------
reduce systemwide lead levels {e.g., achieving 100 percent lead and GRR service line
replacement, installation or re-optimization of optimal corrosion control treatment (OCCT)),
which may take years to fully implement.

Exhibit 1 compares the major differences among the pre-2021 LCR, the 2021 LCRR, and the final LCRI. In
general, only the changes among the pre-2021 LCR, the 2021 LCRR, and the final LCRI are shown in the
exhibit. Asterisks (*) in the pre-2021 LCR and 2021 LCRR columns denote requirements that are retained
in the final LCRI, and these requirements are, therefore, not repeated in the final LCRI column.

Final LCRI Economic Analysis

1-5

October 2024


-------
Exhibit 1-1: Comparison of the 2021 LCRR, Proposed LCRI, and Final LCRI Requirements

Pre-2021 LCR

2021LCRR

Final LCRI

Service Line Inventory

•	Systems were required to complete a
materials evaluation by the time of initial
sampling.

•	No requirement to regularly update materials
evaluation.

•	All systems must develop an initial lead
service line (LSL) inventory by October 16,
2024, that includes all service lines,
regardless of ownership, categorized as lead,
non-lead, galvanized requiring replacement
(GRR), and unknown.*

•	The inventory must be made publicly
accessible and available online for systems
serving > 50,000 persons.*

•	The publicly available inventory must include
a locational identifier for each lead and GRR
service line.

•	The LSL inventory must be updated based on
the system's tap sampling frequency but no
more than annually.

•	All systems must review specified information
that describes connector materials and
locations.

•	Systems must include each identified
connector in their baseline inventory by the
Lead and Copper Rule Improvements (LCRI)
compliance date.

•	Connector material categories include lead,
non-lead, unknown, and no connector
present.

•	The inventory must include a street address
with each service line and connector, if
available.

•	The inventory must be updated annually.

•	Systems must include in their inventories the
total number of each type of service line, the
number of lead and unknown connectors, the
number of full lead and GRR service line
replacements, and the number of partial lead
and GRR service line replacements.

•	Systems must respond to customer inquiries
on incorrect material categorizations within
60 days.

•	Systems must validate the accuracy of their
methods to categorize non-lead service lines
in their inventory no later than 7 years after
the compliance date by the end of the
calendar year unless on a shortened or
deferred deadline.

o The validation pool includes all non-lead
service lines except for those installed
after the applicable Federal, State, or
local lead ban, visually inspected at a

Final LCRI Economic Analysis

1-6

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI





minimum of two points on the pipe
exterior, or previously replaced,
o Systems may submit previous validation
efforts in lieu of the LCRI requirements if
they are at least as stringent as the
requirements, and States must review
and approve of these previous efforts.

• Systems must identify all unknown service
lines by their mandatory service line
replacement deadline.

Service Line Replacement

Replacement Plan

• No requirement.

Replacement Plan

•	All systems with at least one lead, GRR, or
unknown service line must develop a lead
service line replacement (LSLR) plan by the
compliance date.

•	The plan must include a strategy to prioritize
service line replacement.*

Replacement Plan

•	All systems with at least one lead, GRR, or
unknown service line must develop the
service line replacement plan by the
compliance date. The plan includes the
elements from the Lead and Copper Rule
Revisions (LCRR) as well as two new elements:
(1) a strategy to inform customers and
consumers about the plan and replacement
program and (2) an identification of any legal
requirements or water tariff agreement
provisions that affect a system's ability to gain
access to conduct full service line
replacement.*

•	The service line replacement plan must
include additional plan elements if the system
has at least one lead-lined galvanized service
line or if the system is eligible for a deferred
deadline.

•	Service line replacement plan must be publicly
accessible; and available online for systems
serving > 50,000 persons.

•	The plan must be updated annually to include
any new or updated information and
submitted to the State on an annual basis.

Final LCRI Economic Analysis

1-7

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI





•	By the compliance date, systems eligible for
and planning to use deferred deadlines must
include in the plan information on what the
system identifies as the earliest deadline and
fastest feasible rate to replace lead and GRR
service lines that is no slower than 39 annual
replacements per 1,000 service connections.

•	By the end of the second program year, the
State is required to determine in writing
whether a system with a deferred deadline is
replacing lead and GRR service lines at the
fastest feasible rate, either by approving the
continued use of that deferred deadline or by
setting the fastest feasible rate for the system.
In addition to annual updates, systems with
deferred deadlines must submit their plan
every three years with updated information
about why the replacement rate is still the
fastest feasible. The State must review this
information and determine in writing if the
system with a deferred deadline is still
replacing lead and GRR service lines at the
fastest feasible rate, either by approving the
continued use of that deferred deadline or by
setting the fastest feasible rate.

Lead Service Line Replacement

•	Replacement program requirements are
based on the 90th percentile (P90) lead level,
corrosion control treatment (CCT)
installation, and/or source water treatment.

•	Systems conducting LSLR must annually
replace at least 7 percent of LSLs in their
distribution system.

•	Systems must replace the LSL portion they
own and offer to replace the private portion.
Systems are not required to bear the cost of
replacing the private portion.

Lead Service Line Replacement

• Replacement program requirements are
dependent on P90 lead level for community
water systems (CWSs) serving > 10,000
persons:

o If P90 > 0.015 mg/L: Must fully replace 3
percent of lead and GRR service lines per
year based upon a 2-year rolling average
(mandatory replacement) for at least 4
consecutive 6-month monitoring
periods.

Service Line Replacement

•	Replacement program requirements are
independent of systems' P90 lead levels.

•	All CWSs and NTNCWSs with one or more
lead, GRR, or unknown service line in their
inventory must replace lead and GRR service
lines under their control within 10 years,
unless subject to a shortened or deferred
deadline.

•	Systems must replace service lines at a
cumulative average replacement rate of 10

Final LCRI Economic Analysis

1-8

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI

•	Full LSLR, partial LSLR, and LSLs with lead
sample results < 0.015 mg/L ("test-outs")
count toward the 7 percent replacement
rate.

•	Systems can discontinue LSLR after 2
consecutive 6-month monitoring periods at
or below the lead action level.

•	Requires replacement of LSLs only (i.e., no
GRR service lines).

o If P90> 0.010 mg/L but <0.015 mg/L:
Implement a goal-based LSLR program
and consult the primacy agency (or
State) on replacement goals for 2
consecutive 1-year monitoring periods.

•	CWSs serving < 10,000 persons and all non-
transient, non-community water systems
(NTNCWSs) that select LSLR as their
compliance option must complete LSLR
within 15 years if P90 > 0.015 mg/L. See the
Small System Flexibility section of this
exhibit.

•	Annual LSLR rate is applied to the number of
lead and GRR service lines when the system
first exceeds the trigger or action level plus
the number of unknown service lines at the
beginning of the year.

•	Only full LSLR (replacement of the entire
length of the service line) counts toward
mandatory rate* and goal-based rate.

•	All systems must replace their portion of an
LSL if notified by consumer of private side
replacement within 45 days of notification of
the private replacement. If the system
cannot replace the system's portion within
45 days, it must notify the State and replace
the system's portion within 180 days.*

•	Following each service line replacement,
systems must:

o Provide pitcher filters or point-of-use
devices and 6 months of replacement
cartridges to each customer after
replacement.*
o Provide pitcher filters and cartridges
before the affected portion of the line or
the fully replaced service line is returned
to service.*

percent, unless subject to a shortened or
deferred deadline.

•	Cumulative average replacement rate is
applied to the total number of unknown, lead,
and GRR service lines in the baseline inventory
minus the number of unknown service lines
that have been determined to be non-lead
since the baseline inventory.

•	Systems that would have to annually replace
more than 39 service lines per 1,000 service
connections would be eligible for deferred
deadlines longer than 10 years.

•	States are required to set a shorter deadline
for a system where it determines that a
shorter deadline is feasible.

•	Where property owner consent is required for
a system to access the service line, systems
must make a reasonable effort (at least 4
attempts) to engage property owners about
full service line replacement.

•	Systems conducting partial service line
replacement, if not prohibited by the rule,
must offer to replace the remaining portion of
the service line not under their control (within
45 days if replaced in coordination with an
emergency repair).

Final LCRI Economic Analysis

1-9

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI



o Offer to collect a lead tap sample at
locations served by the replaced line
within 3 to 6 months after replacement.*

•	Requires replacement of lead connectors
when encountered.*

•	Systems must make 2 good faith efforts to
engage customers about LSLR.

•	Systems conducting partial LSLR must offer
to replace the remaining portion of the
service line.



LSL-Related Outreach

• If a system replaces its portion only:
o Provide notification to affected
residences within 45 days prior to
replacement on possible elevated short-
term lead levels and measures to
minimize exposure.*
o Include offer to collect lead tap sample

within 72 hours of replacement,
o Provide test results within 3 business
days after receiving results.

LSL-Related Outreach

•	Notify consumers annually if they are served
by a lead, GRR, or unknown service line.*

•	Deliver notice and educational materials to
consumers during water-related work that
could disturb LSLs.*

•	Systems subject to goal-based program must:
o Conduct targeted outreach that

encourages consumers with LSLs to
participate in the LSLR program,
o Conduct an additional outreach activity if
they fail to meet their goal.

•	Systems required to conduct LSLR must
include information about the LSLR program
in public education (PE) materials that are
provided in response to P90 > action level.*

Service Line-Related Outreach

•	Deliver notice and educational materials
during water-related work that could disturb
lead, GRR, or unknown service lines, including
disturbances due to inventorying efforts, to
consumers within 24 hours or before service
line is returned to service, and to customers
within 30 days.

•	Provide filters to consumers for disturbances
to a lead, GRR, or unknown service line caused
by replacement of an inline water meter,
water meter setter, connector, or water main.

•	If a CWS does not meet the mandatory service
line replacement rate, the CWS must conduct
additional public outreach activities to
encourage customers with lead, GRR, and
unknown service lines to participate in the
service line replacement program.

•	Removes goal-based program outreach
activities.

Action Level and Trigger Level

•	P90 level above lead action level of 0.015
mg/L or copper action level of 1.3 mg/L
requires additional actions.

•	Lead action level exceedance requires 7
percent LSLR (includes partial replacements),

• P90 level above lead action level of 0.015
mg/L or copper action level of 1.3 mg/L
requires more actions than the previous rule.

•	Removes the lead trigger level.

•	P90 level above lead action level of 0.010
mg/L or copper action level of 1.3 mg/L
requires actions including installation or re-

Final LCRI Economic Analysis

1-10

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI

CCT recommendation and possible study and
installation, and PE within 60 days after the
end of the monitoring period.

•	Defines lead trigger level as P90 > 0.010 mg/L
and triggers additional planning, monitoring,
and treatment requirements.

•	Lead action level exceedance requires 3
percent full LSLR, optimal corrosion control
treatment (OCCT) installation or re-
optimization, public education (PE), and
public notification (PN) within 24 hours.

•	Trigger level exceedance requires goal-based
LSLR and steps taken towards CCT
installation or re-optimization.

optimization of CCT, PE, and 24-hour PN (for
lead action level exceedances).

• Mandatory full service line replacement of
lead and GRR service lines is independent of
P90 lead levels.

Lead and Copper Tap Monitoring

Sample Site Selection

•	Prioritizes collection of samples from sites
with sources of lead in contact with drinking
water.

•	Highest priority given to sites served by
copper pipes with lead solder installed after
1982 or containing lead pipes and sites
served by LSLs.

•	Systems must collect 50 percent of samples
from LSLs, if available.

Sample Site Selection

•	Prioritizes collecting samples from sites
served by LSLs. All samples must be collected
from sites served by LSLs, if available.*

•	Equal priority to copper pipes with lead
solder, irrespective of installation date.*

•	Adds 2 tiers to prioritize sampling at lead and
GRR service line sites above sites with copper
with lead solder.*

Sample Site Selection

•	Combines the tap sample site selection tiering
criteria for CWSs and NTNCWSs.

•	Removes galvanized service line or premise
plumbing formerly downstream of a lead
connector from Tier 3 sites.

•	Removes requirement for replacement
sampling sites to be selected within
reasonable proximity.

•	Clarifies that sites are considered no longer
available for sampling after customer refusal
or non-response after 2 outreach attempts.

Sample Collection and Inclusion in 90th

Percentile Calculation

• Requires collection of the first-liter sample
after water has sat stagnant for a minimum of
6 hours.

Sample Collection and Inclusion in 90th
Percentile Calculation

•	Requires collection of the fifth-liter sample in
homes with LSLs after water has sat stagnant
for a minimum of 6 hours.

•	Requires first-liter sample collection in
homes without LSLs.*

•	Requires P90 lead calculation for systems
with insufficient LSLs to meet the minimum
number of samples required to include the
highest samples from lower tiers for a total

Sample Collection and Inclusion in 90th
Percentile Calculation

•	Requires collection of the first- and fifth-liter
samples in structures with LSLs after water
has sat stagnant for a minimum of 6 hours.

•	Requires systems with insufficient Tier 1 and 2
sites to meet the minimum number of
samples required by calculating the P90 from
the highest sample values from the highest
tiers samples, equal to the minimum number
required.

Final LCRI Economic Analysis

1-11

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI



number of samples equal to the minimum
number required.

•	Prohibits inclusion of samples collected
under find-and-fix in the P90 calculation.*

•	Adds requirement that samples must be
collected in wide-mouth bottles.*

•	Prohibits sampling instructions that include
recommendations for aerator
cleaning/removal and pre-stagnation flushing
prior to sample collection.*

•	Requires the higher value of the first- and
fifth-liter lead concentration in structures with
LSLs to be used to calculate the P90 value for
lead.

•	Prohibits inclusion of samples following
service line replacement in the P90
calculation. Prohibits the inclusion of more
than one sample per site in each P90
calculation.

•	Revises the definition of a wide-mouth bottle.

Monitoring Frequency

•	Samples are analyzed for both lead and
copper.

•	Systems must collect standard number of
samples based on population; semi-annually
unless they qualify for reduced monitoring.

•	Systems can qualify for annual or triennial
monitoring at reduced number of sites.
Monitoring schedule based on the number of
consecutive years meeting the following
criteria:

o Serves < 50,000 persons and P90 is at or
below the lead and copper action levels,
o Serves any population size, meets State-
specified optimized water quality
parameters (OWQPs), and P90 < lead
action level.

•	Triennial monitoring also applies to any
system with lead P90 < 0.005 mg/L and
copper P90 < 0.65 mg/L for 2 consecutive 6-
month monitoring periods.

•	Based on rule criteria, systems serving <
3,300 persons can apply for a 9-year
monitoring waiver.*

Monitoring Frequency

•	Samples are analyzed for lead and copper,
only copper, or only lead. This occurs when
lead monitoring is conducted more
frequently or at more sites than copper, and
at LSL sites where a fifth-liter sample is only
analyzed for lead.*

•	Lead monitoring schedule is based on the
P90 level for all systems as follows:

o P90 > 0.015 mg/L: Semi-annually at the

standard number of sites,
o P90 > 0.010 mg/L but < 0.015 mg/L:
Annually at the standard number of
sites.

o P90 < 0.010 mg/L: Annually at the

standard number of sites and triennially
at reduced number of sites using same
criteria as the LCR except copper P90
level is not considered.

•	Initial standard monitoring required for
systems with lead and GRR service lines, and
any system that does not sample under the
requirements of the LCRR by the compliance
date.

•	Systems must resume standard monitoring if
they exceed the action level, have a water

Monitoring Frequency

•	Monitoring schedule is based on both the P90
for lead and copper for all systems. Systems
may retain or qualify for reduced monitoring
based on the number of consecutive tap
monitoring periods:

o P90 < action level for 2 consecutive 6-
month periods: Annual monitoring at
standard number of sites for lead and
reduced number of sites for copper,
o P90 < practical quantitation limit (PQL) for
2 consecutive 6-month periods: Triennial
monitoring at the reduced number of sites
for both lead and copper.

•	Initial standard monitoring schedule required
for most systems with lead and/or GRR
service lines in their inventory on the
compliance.

•	Additional criterion for when systems must
start standard monitoring: Systems with no
lead or GRR service lines in their inventory on
the compliance date must start standard
monitoring if they identify a lead or GRR
service line in the future.

Final LCRI Economic Analysis

1-12

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI



quality parameter (WQP) excursion, and
other criteria.



Corrosion Control Treatment and Water Quality Parameters

CCT

•	Systems serving > 50,000 persons were
required to install treatment by January 1,
1997, with limited exception.

•	Systems serving < 50,000 that exceed lead
and/or copper action level(s) are subject to
CCT requirements [e.g., CCT
recommendation, study if required by the
State, CCT installation). They can discontinue
CCT steps if no longer exceed both action
levels for 2 consecutive 6-month monitoring
periods.

•	Systems must operate CCT to meet any
OWQPs designated by the State that define
optimal CCT.

•	There is no requirement for systems to re-
optimize.

CCT

•	Specifies CCT requirements for systems with
P90 lead level > 0.010 mg/L but < 0.015 mg/L:
o No CCT: Must conduct a CCT study if

required by the State,
o With CCT: Must follow the steps for re-
optimizing CCT, as specified in the rule.

•	Systems with P90 lead level > 0.015 mg/L:

o No CCT: Must complete CCT installation
regardless of subsequent P90 levels if
system has started to install CCT.
o With CCT: Must re-optimize CCT.

•	CWSs serving < 10,000 persons and all
NTNCWSs can select an option other than
CCT to address lead. See the Small System
Flexibility section of this exhibit.

CCT

•	Systems with P90 lead level > 0.010 mg/L:
o No CCT: Must install CCT regardless of

their subsequent P90 levels if they have
started to install CCT.
o With CCT: Must re-optimize OCCT.
o Systems with OCCT meeting OWQPs need
only re-optimize OCCT once, unless
required to do so by the State,
o Systems with OCCT that exceed lead
action level exceedance after complete
removal of all lead and GRR service lines
will need to re-optimize again.

•	CWSs serving < 3,300 persons and all
NTNCWSs can select an option other than CCT
to address lead. See the Small System
Flexibility section of this exhibit.

•	Deferred OCCT or re-optimized OCCT for
systems that can complete removal of 100
percent of lead and GRR service lines within 5
years or less of the date they are triggered
into CCT steps. Systems with CCT must
maintain CCT during the 5-year-or-less service
line replacement program.

CCT Options

Includes alkalinity and pH adjustment, calcium
hardness adjustment, and phosphate or silicate-
based corrosion inhibitor.

CCT Options

Removes calcium hardness as an option and
specifies any phosphate inhibitor must be
orthophosphate.*

CCT Options

No changes from the LCRR.

Final LCRI Economic Analysis

1-13

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI

WQPs

•	No CCT: pH, alkalinity, calcium, conductivity,
temperature, orthophosphate (if phosphate-
based inhibitor is used), silica (if silica-based
inhibitor is used).

•	With CCT: pH, alkalinity, and based on type of
CCT either orthophosphate, silica, or calcium.

WQPs

•	Eliminates WQPs related to calcium hardness
(i.e., calcium, conductivity, and
temperature).*

•	All other parameters are the same as in the
LCR.*

WQPs

No changes from the LCRR.

WQP Monitoring

•	Systems serving > 50,000 persons must
conduct regular WQP monitoring at entry
points and within the distribution system.

•	Systems serving < 50,000 persons conduct
monitoring only in those periods that exceed
the lead or copper action level.

•	Contains provisions to sample at reduced
number of sites in distribution system less
frequency for all systems meeting their
OWQPs.

WQP Monitoring

•	Systems serving > 50,000 persons must
conduct regular WQP monitoring at entry
points and within the distribution system.

•	Systems serving < 50,000 persons must
continue WQP monitoring until they no
longer exceed the lead and/or copper action
level(s) for 2 consecutive 6-month
monitoring periods.

•	To qualify for reduced WQP distribution
monitoring, P90 lead level must be < 0.010
mg/L and the system must meet its OWQPs.*

WQP Monitoring

•	Systems with CCT (unless deemed optimized)
serving > 10,000 persons must conduct regular
WQP monitoring at entry points and within
the distribution system.

•	Systems serving <10,000 persons and systems
without CCT serving < 50,000 persons that
exceed the lead and/or copper action level(s)
must conduct WQP monitoring until they no
longer exceed lead and/or copper action
level(s) for 2 consecutive 6-month monitoring
periods.

•	Systems without CCT serving > 10,000 persons
but < 50,000 persons that exceed the lead
action level that are required to install CCT,
must continue to conduct WQP monitoring.

Sanitary Survey Review

Treatment must be reviewed during sanitary
surveys; no specific requirement to assess CCT or
WQPs.

Sanitary Survey Review

CCT and WQP data must be reviewed during
sanitary surveys against most recent CCT
guidance issued by the EPA.*

Sanitary Survey Review

No changes from the LCRR.

Find-and-Fix

No required follow-up samples or additional
actions if an individual sample exceeds the lead
action level.

Find-and-Fix

If individual tap samples > 0.015 mg/L lead, find-
and-fix steps include:

•	Conducting WQP monitoring at or near the
site > 0.015 mg/L.

•	Collecting tap sample at the same tap sample
site within 30 days.*

Distribution System and Site Assessment

•	Changes the name from "Find-and-Fix" to
"Distribution System and Site Assessment" to
describe this requirement more precisely.

•	Requirements from the LCRR affect systems
with individual tap samples > 0.010 mg/L lead.

Final LCRI Economic Analysis

1-14

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI



o For LSL, collect any liter or sample
volume.*

•	Performing needed corrective action.*

•	Documenting customer refusal or non-
response after 2 attempts.*

•	Providing information to local and State
health officials.*

•	Clarifies that the distribution system sample
location must be within a half-mile radius of
each site with a result > 0.010 mg/L.

•	Water systems without CCT are not required
to collect WQP samples for the DSSA CCT
assessment.

Small System Flexibility

No provisions for systems to elect an alternative
treatment approach but sets specific
requirements for CCT and LSLR.

Allows CWSs serving < 10,000 persons and all
NTNCWSs to implement an alternate compliance
option to address lead with State approval:

•	Systems with lead P90 > 0.010 mg/L
recommend CCT, LSLR, provision and
maintenance of point-of-use (POU) devices,
or replacement of all lead-bearing plumbing
materials.

•	If the system's P90 lead level > 0.015 mg/L,
the system must implement the compliance
option.

Allows CWSs serving < 3,300 persons and all
NTNCWSs with P90 levels > lead action level and <
copper action level to conduct the following
actions in lieu of CCT requirements to address lead
with State approval:

•	Choose a compliance option: (1) provision and
maintenance of POU devices or (2)
replacement of all lead-bearing plumbing
materials.

•	Removes the compliance option to conduct
LSLR in 15 years.

Maintains option for systems following CCT
requirements:

•	With CCT: Collect WQPs and evaluate
compliance options and OCCT.

•	No CCT: Evaluate compliance options and CCT.

Public Education and Outreach

•	Systems with P90 > lead action level must
provide PE to customers about lead sources,
health effects, measures to reduce lead
exposure, and additional information
sources.

•	Systems with P90 > lead action level must
offer lead tap sampling to customers who
request it.

•	Water systems must provide updated lead
health effects language in PN and PE
materials. CWSs must provide updated health
effects language in the Consumer Confidence
Reports (CCR).

•	For water systems serving a large proportion
of consumers with limited English proficiency,
PE materials must contain information in the
appropriate language(s) regarding the

•	Revises the mandatory lead health effects
language to improve completeness and clarity.

•	Water systems must provide the updated
health effects language in PN and all PE
materials. CWSs must provide updated health
effects language in the CCR.

•	For water systems serving a large proportion
of consumers with limited English proficiency,
all PE materials must contain information in

Final LCRI Economic Analysis

1-15

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI

•	Systems must provide lead consumer notice
to individuals served at tested taps within 30
days of learning results.

•	For water systems serving a large proportion
of consumers with limited English proficiency,
PE materials must contain information in the
appropriate language(s) regarding the
importance of the materials or information
on where consumers can get a translated
copy or assistance in other languages.

importance of the materials or information
on where consumers can get a translated
copy or assistance in other languages.

•	If P90 > lead action level:

o LCRR PN and LCR PE requirements apply,
o Water systems must offer to sample the
tap for lead for any customer who
requests it.

•	Water systems must provide the lead
consumer notice to consumers whose
individual tap sample is > 0.015 mg/L lead as
soon as practicable but no later than 3
calendar days.

•	CWSs must provide information to local and
State health agencies.*

Also see the Public Notification, Consumer

Confidence Report, and LSL-Related Outreach

sections of this exhibit.

the appropriate language(s) regarding the
importance of the materials and information
on where consumers can get a translated copy
or assistance in other languages, or the
materials must be in the appropriate
language(s).

•	Water systems must deliver consumer notice
of lead and copper tap sampling results to
consumers whenever their tap is sampled as
soon as practicable but no later than 3
business days after receiving the results,
regardless of the level.

•	If P90 > lead action level:

o LCRR PN requirements apply,
o Water systems must conduct PE no later
than 60 days after the end of the tap
sampling period until the system no
longer exceeds the action level unless the
State approves an extension,
o Water systems must deliver PE materials
to bill paying customers and every service
connection address served.

• Water systems with multiple lead action level
exceedances (at least 3 action level
exceedances in a 5-year period) must
conduct additional public outreach activities
and make filters available. Water systems
must submit a filter distribution plan to the
State within 60 days of the second action
level exceedance, and the State will have 60
days to review. The State has discretion to
allow the system to discontinue outreach
activities and filter provision earlier if it
completes actions to reduce lead levels.

•	Water systems must offer to sample the tap
for lead for any consumer with a lead, GRR, or
unknown service line who requests it.

Final LCRI Economic Analysis

1-16

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI





Also see the Public Notification, Consumer
Confidence Report, and Service Line Related
Outreach sections of this exhibit.

Public Notification

•	If P90 > action level:

o No PN required for P90 > action level.

•	Tier 2 PN required for violations to § 141.80
through § 141.85.

•	Tier 3 PN required for violations to § 141.86
through § 141.89.

Also see the Public Education and Outreach
section of this exhibit.

•	If P90 > lead action level:

o Systems must notify consumers of P90 >
action level within 24 hours (Tier 1 PN).
Systems must comply by October 16,
2024.

•	Tier 2 PN required for violations to § 141.80
(except § 141.80(c)) through § 141.84, §
141.85(a) through (c) and (h), and § 141.93.

•	Tier 3 PN required for violations to § 141.86
through § 141.90.

Also see the Public Education and Outreach
section of this exhibit.

•	If P90 > lead action level of 0.010 mg/L:

o LCRR Tier 1 PN requirements apply, but
for the proposed LCRI action level of
0.010 mg/L.

•	Tier 2 PN required for violations to § 141.80
(except § 141.80(c)) through § 141.84, §
141.85(a) through (c) (except § 141.85(c)(3))
and (h) and (j), and § 141.93.

•	Tier 3 PN required for violations to § 141.86
through § 141.90 and § 141.92.

•	Water systems must provide updated lead
health effects language in PN.

Also see the Public Education and Outreach section
of this exhibit.

Consumer Confidence Report

• All CWSs must provide educational material
in the annual CCR.

•	CWSs must provide updated health effects
language in the CCR.

•	All CWSs are required to include information
on how to access the LSL inventory and how
to access the results of all tap sampling in the
CCR.

•	Revises the mandatory health effects
language to improve accuracy and clarity.

•	Revises the mandatory lead health effects
language and informational statement as well
as includes additional information about risk
of lead exposure in the informational
statement about lead in the CCR to improve
completeness and clarity.

•	CWSs must provide updated health effects
language in the CCR.

•	CWSs must include a statement in the CCR
about the system sampling for lead in schools
and child care facilities and direct the public to
contact their school or child care facility for
further information.

•	CWSs with lead, GRR, or unknown service lines
must include a statement in the CCR about

Final LCRI Economic Analysis

1-17

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI





how to access the service line inventory and
replacement plan.

Also see the Public Education and Outreach section
of this exhibit.

Change in Source of Treatment

Systems on a reduced tap monitoring schedule
must obtain prior State approval before changing
their source or treatment.

Systems on any tap monitoring schedule must
obtain prior State approval before changing their
source or treatment. These systems must also
resume a standard lead and copper tap
monitoring schedule.*

No changes from the LCRR.

Source Water Monitoring and Treatment

Periodic source water monitoring for lead and
copper is required for systems with:

•	Source water treatment; or

•	P90 > action level and no source water
treatment.

States can waive continued source water

monitoring for lead and copper if the:*

•	System has already conducted source water
monitoring for a previous P90 > action level;

•	State has determined that source water
treatment is not required; and

•	System has not added any new water
sources.

Updated cross-reference to requirement for
conducting standard monitoring when there is a
source water addition.

Lead in Drinking Water at Schools and Child Care Facilities

•	Does not include separate testing and
education program for CWSs at schools and
child care facilities.

•	Schools and child care facilities that are
classified as NTNCWSs must sample for lead
and copper.*

•	CWSs must provide annual public education
materials to all schools and licensed child
care facilities they serve.

•	CWSs must conduct sampling at 20 percent
of elementary schools and 20 percent of
licensed child care facilities they serve per
year and conduct sampling at secondary
schools on request for first testing cycle (5
years) and conduct sampling on request of all
schools and child care facilities thereafter.

•	Sample results must be provided to each
sampled school/child care facility, State, and
local or State health department.

Expands on LCRR requirements to include:

•	Waivers for CWSs to sample in schools and
child care facilities during the first 5-year
testing cycle if the facility has been sampled
between January 1, 2021, and the LCRI
compliance date.

•	Requires CWSs to include a statement about
the opportunity for schools and child care
facilities to be sampled in the CCR.

•	Excludes schools and licensed child care
facilities constructed or had full plumbing
replacement on or after January 1, 2014 and

Final LCRI Economic Analysis

1-18

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI



•	Excludes schools and licensed child care
facilities constructed on or after January 1,
2014.

•	Waives schools and child care facilities that
were sampled under a State or other
program after October 16, 2024.

that are also not served by a lead, GRR, or
unknown service line.

• Includes clarifications on the applicability of
the requirements and on the content of public
education material CWSs must provide to
schools and licensed child care facilities.

Primacy Agency (or State) Reporting

States must report information to the EPA that

includes, but is not limited to:

•	All P90 lead levels for systems serving > 3,300
persons, and only levels > 0.015 mg/L for
smaller systems.

•	Only copper P90 levels above the copper
action level for all systems.

•	Systems that are required to initiate LSLR and
the date replacement must begin.

•	Systems for which OCCT has been
designated.

States must keep records on information that

includes, but is not limited to:

•	Records of the currently applicable or most
recent State determinations, including all
supporting information and an explanation of
the technical basis for each decision

State primacy requirements include, but are not

limited to:

•	Designating OCCT

•	Designating source water treatment methods

•	Verifying service line replacement schedules

States must report information to the EPA that

includes, but is not limited to:

•	All lead and copper P90 levels for all system
sizes.*

•	The number of lead, GRR, and unknown
service lines for every water system.*

•	The goal-based or mandatory replacement
rate and the date each system must begin
LSLR.

•	OCCT status of all systems including OWQPs
specified by the State.*

•	For systems triggered into source water
treatment, the State-designated date or
determination for no treatment required.*

States must keep records on information that

includes, but is not limited to:

•	LSLR plans.*

•	Compliance sampling pools.*

•	Determinations related to source water
treatment.*

•	Determinations related to compliance
alternatives for small CWSs and NTNCWSs.*

•	LSL inventories.*

State primacy requirements include, but are not

limited to:

•	Reviewing service line inventory.*

•	Approving LSLR goals.

States must report information to the EPA that

includes, but is not limited to:

•	The current numbers of lead, GRR, unknown,
and non-lead service lines, lead connectors,
and unknown connectors in each system's
inventory.

•	The numbers and types of service lines
replaced and the replacement rate for every
system conducting mandatory service line
replacement.

•	The deadline for the system to complete
replacement of all lead and GRR service lines.

•	The expected date of completion of service
line replacement.

•	The lead P90 levels of systems with an action
level exceedance within 15 days of the end of
the monitoring period or, if earlier, within 24
hours of receiving the notice from the system.

•	The result of the State's determination as to
whether the deferred deadline is the fastest
feasible, the deadline at the fastest feasible
rate, and the reasons for the State's decision.

States must keep records on information that

includes, but is not limited to:

•	Samples that do not meet the six-hour
minimum stagnation time.

•	Determinations concerning systems eligible
for deferred deadlines for service line

Final LCRI Economic Analysis

1-19

October 2024


-------
Pre-2021 LCR

2021LCRR

Final LCRI



•	Determining if a greater LSLR rate is
feasible.*

•	Defining school and child care program and
determining if State or local testing program
is at least as stringent as Federal
requirements.

•	Verifying compliance with "Find-and-Fix"
requirements.*

•	Reviewing any change in source water
treatment.*

replacement.

Adds State primacy requirements to:

•	Identify State laws that pertain to a water
system's access to conduct full service line
replacement.

•	Make determinations about systems eligible
for service line replacement deferred
deadlines.

•	Make determinations about which water
systems serve a large proportion with limited
English proficiency and provide technical
assistance to those systems required to meet
the requirements to provide translated PE or
translation assistance to their consumers.

•	Review and approve inventory validations.

Final LCRI Economic Analysis

1-20

October 2024


-------
1.2	Document Organization

The remainder of this EA is organized into the following chapters:

•	Chapter 2: Need for the Rule summarizes the goal of the final LCRI, why the EPA revised the
prior lead and copper regulations, and the regulatory history. It also explains the statutory
authority for the final LCRI and the economic rationale for the regulatory approach.

•	Chapter 3: Baseline Drinking Water System describes the systems subject to the final LCRI,
including the populations they serve and CCT status, lead and copper tap water concentration
levels, the characterization of service line material in the United States, the proportion of
systems on reduced monitoring, and the rate of historical source water and treatment changes
to characterize the baseline before the EPA models estimated changes that result from
complying with the final LCRI requirements.

•	Chapter 4: Economic Impact and Cost Analysis of the Final Lead and Copper Rule
Improvements describe how the final LCRI regulatory requirements are implemented, and the
unit cost of actions taken by PWSs and the State to comply with the requirements. The chapter
also provides estimates of the total costs of the 2021 LCRR and the final LCRI, as well as the
estimated incremental costs of the final LCRI.

•	Chapter 5: Benefits Resulting from the Lead and Copper Rule provides a description of the
estimated health benefits for the final regulatory changes affecting systems and States and
provides estimates of total benefits for the 2021 LCRR and the final LCRI, as well as the
estimated incremental benefits for the final LCRI requirements.

•	Chapter 6: Comparison of Costs to Benefits provides a summary of costs and benefits
associated with the provisions of the 2021 LCRR and the final LCRI requirements. The chapter
also describes the incremental costs and benefits of the final LCRI.

•	Chapter 7: Statutory and Administrative Requirements discusses distributional analyses
performed to evaluate the effects of the final LCRI options on different segments of the
population in accordance with 13 federal mandates and statutory reviews, including but not
limited to the Regulatory Flexibility Act, Unfunded Mandates Reform Act, and Executive Order
12898 on, Federal Actions to Address Environmental Justice in Minority Populations and Low-
Income Populations.

•	Chapter 8: Other Options Considered presents other alternatives the EPA evaluated when
developing the final LCRI. These include alternative lead ALs and LSLR rates.

1.3	Calculations and Citations

This EA involves numerous detailed and complex analyses, and the following are provided to help the

reader understand how those analyses were conducted and their underlying data and assumptions:

•	Appendices containing supporting spreadsheets and analyses:
o Appendix A: LSLR Unit Costs

Final LCRI Economic Analysis

1-21

October 2024


-------
o Appendix B: Modeling Costs in the SafeWater LCR Model for the Final LCRI, 2021 LCRR, and
the Pre-2021 LCR

o Appendix C: Incremental Costs of the Final LCRI from the Pre-2021 LCR

o Appendix D: Adverse Health Effects Associated with Lead Exposures

o Appendix E: Adverse Health Effects Associated with Copper Exposures

o Appendix F: Sensitivity Analysis for IQ Valuation in Children and Costs and Benefits of the
Final Rule at a 3 Percent and 7 Percent Discount Rate

•	Tabular exhibits, most of which include a row with the formulas used to calculate the contents
of each column and information sources for values that are not calculated in the exhibits.

•	Exhibits that illustrate methodologies of analyses as well as final LCRI requirements.

•	Supporting report and electronic spreadsheet files, as explained in Exhibit 1-2 below.

Exhibit 1-2: Supporting Report and Spreadsheet Files

File Name

Description

Administrative Burden and
Costs_Final.xlsx

Provides one-time and ongoing administrative burden and costs
associated with the pre-2021 LCR, 2021 LCRR, and final LCRI for
water systems and States.

Analysis of School_Child Care Sample
Number_Final.xlsx

Provides an estimate of the number of taps from which a school or
child care facility would collect lead samples, based on the 3Ts
guidance. Used to estimate the number of schools and child care
facilities that can be tested using WIIN grant funding. Serves as a
supporting file for "School_Child Care lnputs_Final.xlsx."

ASDWA CoSTS_2024_Revised

Provides ASDWA's estimated increase burden estimates for States
to oversee the requirements of the LCRI. The EPA used these
estimates and those provided in the file, "Final CoSTS 2-6-20.xlsx"
to develop costs for the final LCRI.

CCT Study and Review Costs_Final.xlsx

Provides the EPA's assumptions regarding which systems will be
required to conduct a CCT study and if applicable, if the study will
be a desktop or demonstration study, and the estimated costs of
these studies under the pre-2021 LCR. Also, provides State CCT
review-related activities for the pre-2021 LCR, 2021 LCRR, and the
final LCRI.

Customer Requested Sample
Percent_Final.xlsx

Provides the percentage of samples requested based on
information provided on five system's websites. Used as a
supporting file for "Lead Analytical Burden and Costs_Final.xlsx."

CWS Inventory Characteristics_Final.xlsx

Provides inventory, milestone, violation, and treatment
information from the SDWIS/Fed 4th quarter 2020 "frozen"
dataset for 49,529 CWSs and how these data are used to provide
baseline system characteristics described in Chapter 3 and
Appendix B.

DWINSA_StateDate_LSL_Status_Final.xlsx

Provides system-specific information for the subset of CWSs with

Final LCRI Economic Analysis

1-22

October 2024


-------
File Name

Description



known LSL status (either presence or absence of LSLs or
unknowns) based on data from (1) the 7th DWINSA and its one-
time update of service line materials; (2) service line inventory
information for 13 States and Region 9 tribal systems; and (3)
additional web searches of systems with prior or ongoing LSLR
programs; (4) and discussions with systems serving more than 1
million people. This file also includes the geographic
representativeness of States with known LSL status.

Estimated Driving Distances_Final.xlsx

Outlines the EPA's approach for estimating the distance a water
system would drive to a customer's home for lead sampling, site
investigation, or other reasons.

Extent of P90 Data_LCR_Final.xlsx

Provides the estimated percentage of CWSs with at least one
reported P90 value during 2012 - 2020 and the percentage with
known LSL status (the presence or absence of LSLs) for three
system size categories to determine if systems serving < 3,300
people were underrepresented.

Failure to Meet LSLR Goals_Final.xlsx

Calculates the burden and costs for CWSs serving > 10,000 people
with a TLE to conduct outreach activities if they fail to meet their
annual LSLR goal under the 2021 LCRR.

Final CoSTS 2-6-20.xlsx

Provides ASDWA's estimated increase burden estimates for States
to oversee the requirements of the final 2021 LCRR. The EPA used
these estimates and those provided in ASDWA
CoSTS_2024_Revised.xlsx" to develop costs for the 2021 LCRR and
final LCRI.

General Cost Model lnputs_Final.xlsx

Provides general costing inputs that include system and State labor
costs, postage, paper, ink, and envelopes.

Initial P90 Categorization_LCR_Final.xlsx

Assigns CWSs to one of five P90 categories using two approaches
under the pre-2021 LCR: A low estimate based on their average
P90 level and high estimate based on their highest P90 level. These
estimates represent the baseline condition before systems
implement the requirements of the final LCRI.

Initial P90 Categorization_LCRI_Final.xlsx

Assigns CWSs to one of five P90 categories using two approaches
under the final LCRI: A low estimate based on their average P90
level and high estimate based on their highest P90 level. Also
includes adjusted lead 90th percentile results from systems with
LSLs using two multipliers to reflect new sampling requirements
under the final LCRI, including an adjustment to simulate the
expected increase on the P90 if a system with LSLs uses the higher
of the paired first- and fifth-liter sample.

Initial P90 Categorization_LCRR_Final.xlsx

Assigns CWSs to one of five P90 categories using two approaches
under the 2021 LCRR: A low estimate based on their average P90
level and high estimate based on their highest P90 level. Also
includes adjusted lead 90th percentile results from systems with
LSLs using two multipliers to reflect new sampling requirements
under the 2021 LCRR.

Initial P90

Categorization_CWS_NTNCWS_LCR

Compares the distribution of NTNCWSs across the five lead 90th
percentile categories as a function of CCT and LSL status to that

Final LCRI Economic Analysis

1-23

October 2024


-------
File Name

Description

Compare_Final.xlsx

developed for CWSs.

Inventory Updates and
Validation_Final.xlsx

Provides the burden and costs for updating the service line
inventory under the 2021 LCRR and final LCRI and performing
validation for a subset of service lines categorized as non-lead
under the final LCRI.

LCRI Updated Initial Inventory with
Connectors_Final.xlsx

Provides the burden and costs to update the initial service line
inventory with connector material information.

Lead Analytical Burden and Costs
_Final.xlsx

Provides estimated burden and costs for lead sample collection,
analysis, and reporting as well as assumptions and data sources for
each estimate.

Lead_WQP_Sample Bottle Costs_Final.xlsx

Provides 250 mL, 500 mL, and 1 liter bottle costs. Used as a
supporting file for estimating lead and WQP analytical burden and
costs and school and child care facility testing.

Likelihood_Sample_Above_AL_LCRI_DSSA
_Final.xlsx

Provides estimates of the likelihood of an individual tap sample
being above the lead AL based on system size, LSL status, and P90
classification under the final LCRI. Also provides estimated burden
and costs associated with the DSSA1.

Likelihood_Sample_Above_AL_LCRR_Find
_Fix_Final.xlsx

Provides estimates of the likelihood of an individual tap sample
being above the lead AL based on system size, LSL status, and P90
classification under the 2021 LCRR. Also provides estimated
burden and costs associated with find-and-fix.

Likelihood_SourceChange_Final.xlsx

Provides the estimated likelihood that a CWS or NTNCWS will add
a new source or change its primary source. Also includes reporting,
review, and State consultation associated with this change.

Likelihood_TreatmentChange_Final.xlsx

Provides the estimated likelihood that a CWS or NTNCWS will add
a new treatment. Also includes reporting, review, and State
consultation associated with this change.

LSLR Ancillary Costs_Final.xlsx

Provides the derivation of costs associated with SLR activities other
than physical replacement.

LSLR Unit Cost.xlsx

Provides the derivation of the SL physical replacement costs based
on the 7th DWINSA.

LSLR_Time_Span_Analysis_CWS_Final.xlsx

Estimates the average length of time a CWS that is triggered into
LSLR replaces LSLs under the pre-2021 LCR. The results of this
analysis are also used for NTNCWSs under the pre-2021 LCR.

MI_LCR_Sample_Database

This workbook contains the Michigan lead tap sampling data from
2019, 2020, and 2021. The data are used in several analyses
presented in this EA.

Multiple Lead ALE_LCRI_5_AL_Final.xlsx

Provides the percentage of systems with at least 2 lead ALEs in 5
years that had at least 3 lead ALEs based on a lead AL of 5 |jg/L as
opposed to the final LCRI AL of 10 ng/L. This information is used to
develop costs for other options the EPA considered, as presented
in Chapter 8 of this EA.

Multiple Lead ALEs_LCRI_10_AL_Final.xlsx

Provides the percentage of systems with at least 2 lead ALEs in 5
years that had at least 3 lead ALEs under the final LCRI lead AL of
10 Mg/L.

Multiple Lead ALE_LCRI_15_AL_Final.xlsx

Provides the percentage of systems with at least 2 lead ALEs in 5

Final LCRI Economic Analysis

1-24

October 2024


-------
File Name

Description



years that had at least 3 lead ALEs based on a lead AL of 15 ng/L as
opposed to the final LCRI AL of 10 ng/L. This information is used to
develop costs for other options the EPA considered, as presented
in Chapter 8 of this EA.

NTNCWS Inventory
Characteristics_Final.xlsx

Provides inventory, milestone, violation, and treatment
information from the SDWIS/Fed 4th quarter 2020 "frozen" dataset
for 17,418 NTNCWSs and how these data are used to provide
baseline system characteristics described in Chapter 3 and
Appendix B.

P90_Unknown LSL vs. LSL Known Status
CWSs_Final xlsx

Compares the P90 data for the subset of systems with known LSLs
status and reported P90 values to the larger set of CWSs with at
least one reported P90 value (but unknown LSL status) in
SDWIS/Fed for 2012 - 2020 to determine the representativeness
of the subset.

Pb Schedules_CWS_Final.xlsx

Estimates baseline lead tap sampling schedules for CWSs using the
SDWIS/Fed 4th quarter 2020 "frozen" dataset starting in Year 4 of
the 35-year analysis period. These schedules are also used for
CWSs without LSLs starting in Year 5 for those with P90 < 15 ng/L
and P90 < 10 ng/L, under the 2021 LCRR and final LCRI,
respectively.

Pb Schedules_NTNCWSs_Final.xlsx

Estimates baseline lead tap sampling schedules for NTNCWSs using
the SDWIS/Fed 4th quarter 2020 "frozen" dataset starting in Year 4
of the 35-year analysis period. These schedules are also used for
NTNCWSs without LSLs starting in Year 5 for those with P90 < 15
Hg/L and P90 < 10 ng/L, under the 2021 LCRR and final LCRI,
respectively.

POU lnputs_Final.xlsx

Provides costing inputs for small CWSs and those NTNCWSs that
select POU devices as their compliance option. Includes the
estimated number of required POU devices, development of a
POU process, annual reporting, and State review.

Public Education lnputs_CWS_Final.xlsx

Provides the derivation of the inputs used to estimate PE burden
and costs under the pre-2021 LCR, 2021 LCRR, and final LCRI for
CWSs.

Public Education
lnputs_NTNCWS_Final.xlsx

Provides the derivation of the inputs used to estimate PE burden
and costs under the pre-2021 LCR, 2021 LCRR, and final LCRI for
NTNCWSs.

Robocall Pricing Estimates_Final.xlsx

Provides the quotes from three companies for robocalling services.
The average of costs from these companies is used to estimate
costs for some outreach activities for CWSs.

Sample Kits and Shipping Costs_Final.xlsx

Provides the cost of a sample kit that includes the container, paper
for instructions and chain-of-custody, plastic bag to prevent the
paper for getting wet, and labels. Bottle costs are included in the
file "Lead_WQP_Sample Bottle Costs," "Lead Analytical Burden
and Costs_Final.xlsx," and "WQP Analytical Burden and
Costs_Final.xlsx."

School_Child Care lnputs_Final.xlsx

Provides the derivation of the inputs used to estimate the burden
and costs for CWSs to conduct a lead in drinking water testing

Final LCRI Economic Analysis

1-25

October 2024


-------
File Name

Description



program at schools and licensed child care facilities.

Service Line Characterization Using
DWINSA_Final.xlsx

Provides the derivation of the estimated percent of systems with
service lines of different material, and the percent of service lines
of different materials in those systems. For the pre-2021 LCR only,
also provides the percent of systems that test their LSLs before
replacing them to determine if any meet the tested-out criteria
and would not need to be replaced, and an estimated percent of
those service lines that meet the tested-out criteria.

Summary of Lab Responses_7
labs_Final.xlsx

Provides commercial lab costs for lead, orthophosphate, alkalinity,
and calcium. These estimates are used for lead and WQP analytical
cost files, school and child care facility testing, and "CCT Study and
Review_Final.xlsx."

Two Lead ALEs_LCRI_5_AL_Final.xlsx

Provides the percentage of systems with at least 2 lead ALEs in 5
years based on a lead AL of 5 ng/L as opposed to the final LCRI
action level of 10 |jg/L. This information is used to develop costs
for other options the EPA considered, as presented in Chapter 8 of
this EA.

Two Lead ALEs_LCRI_10_AL_Final.xlsx

Provides the percentage of systems with at least 2 lead ALEs in 5
years under the final LCRI.

Two Lead ALEs_LCRI_15_AL_Final.xlsx

Provides the percentage of systems with at least 2 lead ALEs in 5
years based on a lead action level of 15 ng/L as opposed to the
final LCRI action level of 10 ng/L. This information is used to
develop costs for other options the EPA considered, as presented
in Chapter 8 of this EA.

VLSEntryPointValues_Final.xlsx

Provides a summary of entry point-level data compiled by the EPA
for LSL and CCT estimates for systems serving more than 1 million
people.

VLSSystemData.xIsx

Provides a summary of system-level data compiled by the EPA for
LSL and CCT estimates for systems serving more than 1 million
people.

WQP Analytical Burden and
Costs_Final.xlsx

Provides the derivation of the inputs used to estimate burden and
costs for system WQP sample collection, analysis, and reporting
and State review.

WQP Schedules_CWS_LCR_Final.xlsx

Estimates the initial WQP distribution system monitoring
schedules under the pre-2021 LCR for CWSs using the SDWIS/Fed
4th quarter 2020 "frozen" dataset.

WQP Schedules_CWS_LCRI_Final.xlsx

Estimates the initial WQP distribution system monitoring
schedules under the final LCRI for CWSs using the SDWIS/Fed 4th
quarter 2020 "frozen" dataset.

WQP Schedules_CWS_LCRR_Final.xlsx

Estimates the initial WQP distribution system monitoring
schedules under the 2021 LCRR for CWSs using the SDWIS/Fed 4th
quarter 2020 "frozen" dataset.

WQP Schedules_NTNCWS_LCR_Final.xlsx

Estimates the initial WQP distribution system monitoring
schedules under the pre-2021 LCR for NTNCWSs using the
SDWIS/Fed 4th quarter 2020 "frozen" dataset.

WQP Schedules_NTNCWS_LCRI_Final.xlsx

Estimates the initial WQP distribution system monitoring

Final LCRI Economic Analysis

1-26

October 2024


-------
File Name

Description



schedules for NTNCWSs under the final LCRI using the SDWIS/Fed
4th quarter 2020 "frozen" dataset.

WQP Schedules_NTNCWS_LCRR_Final.xlsx

Estimates the initial WQP distribution system monitoring
schedules for NTNCWSs under the 2021 LCRR using the SDWIS/Fed
4th quarter 2020 "frozen" dataset.

Acronyms: AL = action level; ALE = action level exceedance; ASDWA = Association of State Drinking Water
Administrators; CCT = corrosion control treatment; CoSTS = Costs of State Transactions Study; CWS = community
water system; DSSA = Distribution System and Site Assessment; DWINSA = Drinking Water Infrastructure and
Needs Assessment; EA = economic analysis; LCR = Lead and Copper Rule; LCRR = Lead and Copper Rule revisions;
LSL = lead service line; LSLR = lead service line replacement; NTNCWS = non-transient non-community water
system; P90 = lead 90th percentile level; POU = point-of use; SDWIS/Fed: Safe Drinking Water Information
System/Federal Version; SL = service line; SLR = service line replacement; TLE = trigger level exceedance; VLS = very
large system; WQP = water quality parameter.

Notes:

General: These documents are available in the docket for the final rule under docket number EPA-HQ-OW-2022-
0801 at https://www.regulations.gov.

1 In the final LCRI, the EPA replaced the term "find-and-fix" with "Distribution System and Site Assessment."
1.4 References

Agency for Toxic Substances and Disease Registry (ATSDR). 2020. Toxicological Profile for Lead. Atlanta,
GA: U.S. Department of Health and Human Services, Public Health Service. August 2020.
https://www.atsdr.cdc.gov/toxprofiles/tpl3.pdf

Centers for Disease Control (CDC). 2022a. Health Effects of Lead Exposure. Retrieved July 19, 2023 from
https://www.cdc.gov/nceh/lead/prevention/health-effects.htm

CDC. 2022b. Breastfeeding and Special Circumstances: Environmental and Chemical Exposures: Lead.
Last reviewed May 18, 2022. Retrieved from https://www.cdc.gov/breastfeeding/breastfeeding-special-
circumstances/environmental-exposures/lead.html.

CDC. 2022c. CDC updates blood lead reference value to 3.5 mg/dL. Last reviewed December 16, 2022.
Retrieved from https://www.cdc.gov/nceh/lead/news/cdc-updates-blood-lead-referencevalue.html.

CDC. 2023. Lead in Drinking Water. Last reviewed February 28, 2023. Retrieved from
https://www.cdc.gov/nceh/lead/prevention/sources/water.htm.

Executive Order 12866. 1993. Regulatory Planning and Review. Federal Register. 58 FR 51735. October
4, 1993. Available at https://www.reginfo.gov/public/isp/Utilities/EO 12866.pdf.

Executive Order 13990. 2021. Executive Order on Protecting Public Health and the Environment and
Restoring Science to Tackle the Climate Crisis. January 20, 2021. https://www.whitehouse.gov/briefing-
room/presidential-actions/2021/01/20/executive-order-protecting-public-health-and-environment-and-
restoring-science-to-tackle-climate-crisis/.

Executive Order 14094. Modernizing Regulatory Review. April 6, 2023. Federal Register. 88 FR 21789.
https://www.govinfo.gov/content/pkg/FR-2023-04-ll/pdf/2023-07760.pdf

Final LCRI Economic Analysis

1-27

October 2024


-------
Folkman, S. 2018. Water Main Break Rates In the USA and Canada: A Comprehensive Study. Utah State
University, Buried Structures Laboratory.

LSLR Collaborative, n.d. Getting Started on an LSL Inventory. Retrieved November 11, 2023, from
https://www.lslr-collaborative.org/preparing-an-

inventorv.html#:~:text=Lead%20pipe%20was%20tvpicallv%20installed.as%203%20inches%20in%20dia
meter.

National Toxicology Program (NTP). 2012. NTP Monograph: Health Effects of Low-Level Lead. U.S.
Department of Health and Human Services. Office of Health Assessment and Translation. Division of the
National Toxicology Program.

https://ntp.niehs.nih.gov/ntp/ohat/lead/final/monographhealtheffectslowlevellead newissn 508.pdf.

The White House. 2021. Fact Sheet: The Biden-Harris Lead Pipe and Paint Action Plan. December 16,
2021. https://www.whitehouse.gov/briefing-room/statements-releases/2021/12/16/fact-sheet-the-
biden-harris-lead-pipe-and-paint-action-plan/

United States Environmental Protection Agency (USEPA). 1991. Drinking Water Regulations; Maximum
Contaminant Level Goals and National Primary Drinking Water Regulations for Lead and Copper; Final
Rule. Federal Register. 56 FR 26460. June 7, 1991. Washington, D.C.: Government Printing Office.

USEPA. 2000. National Primary Drinking Water Regulations for Lead and Copper. Federal Register. 65 FR
1950. January 12, 2000. https://www.govinfo.gov/content/pkg/FR-2000-01-12/pdf/0Q-3.pdf.

USEPA. 2007. National Primary Drinking Water Regulations for Lead and Copper: Short-Term Regulatory
Revisions and Clarifications; Final Rule. Federal Register 72 FR 57782. October 10, 2007. Washington,
D.C.: Government Printing Office.

USEPA. 2013. Integrated Science Assessment for Lead. (EPA/600/R-10/075F). Office of Research and
Development. (EPA/600/R-10/075F). Research Triangle Park, NC.

USEPA. 2018. 3Tsfor Reducing Lead in Drinking Water in Schools and Child Care Facilities: A Training,
Testing, and Taking Action Approach (Revised Manual). October 2018. Office of Water. EPA 815-B-18-
007. https://www.epa.gov/ground-water-and-drinking-water/3ts-reducing-lead-drinking-water-toolkit.

USEPA. 2020. Use of Lead Free Pipes, Fittings, Fixtures, Solder, and Flux for Drinking Water; Final Rule.
Federal Register 85 FR 54235. September 1, 2020. https://www.govinfo.gov/content/pkg/FR-2020-Q9-
01/pdf/2020-16869.pdfhttps://www.govinfo.gov/content/pkg/FR-2020-09-01/pdf/2020-
16869. pdfhttps://www.govinfo.gov/content/pkg/FR-2020-09-01/pdf/2020-16869.pdf.

USEPA. 2021a. National Primary Drinking Water Regulations: Lead and Copper Rule Revisions; Final Rule.
Federal Register. 86 FR 4198. January 15, 2021. Washington, D.C.: Government Printing Office.

USEPA. 2021b. National Primary Drinking Water Regulations: Lead and Copper Rule Revisions; Delay of
Effective and Compliance Dates. Final Rule. Federal Register. 86 FR 31939. June 16, 2021. Washington,
D.C.: Government Printing Office.

USEPA. 2022. Strategy to Reduce Lead Exposures and Disparities in U.S. Communities. EPA 540-R-22-
006. October 2022. Retrieved from https://www.epa.gov/lead/final-strategy-reduce-lead-exposures-
and-disparities-us-communities.

Final LCRI Economic Analysis

1-28

October 2024


-------
USEPA. 2023. National Primary Drinking Water Regulations: Lead and Copper Rule Improvements.
Proposed Rule. Federal Register. 88 FR 84878. December 6, 2023. Washington, D.C.: Government
Printing Office.

USEPA. 2024a. Integrated Science Assessment for Lead (Final Report). EPA 600-R-23-375. Office of
Research and Development. January 2024.

USEPA. 2024b. Updated 7th Drinking Water Infrastructure Needs Survey and Assessment. Fact Sheet.
May 2024. Retrieved from https://www.epa.gov/svstem/files/documents/2024-05/fact-sheet one-
time-update 2024.04.30 508 compliant l.pdf

Final LCRI Economic Analysis

1-29

October 2024


-------
2 Need for the Rule

Lead and copper enter drinking water primarily through the corrosion of distribution system and
household plumbing materials that contain these metals. The goal of the Lead and Copper Rule (LCR) is
to protect public health by reducing exposure to lead and copper in drinking water and the associated
health risks from this exposure. The LCR accomplishes this primarily by controlling water corrosivity,
thereby minimizing the leaching of these metals from household plumbing and drinking water
distribution system components.

The Lead and Copper Rule Revisions (LCRR) was published in the Federal Register on January 15, 2021.
As previously discussed in Chapter 1, the Environmental Protection Agency (EPA) published the agency's
decision to delay the effective and compliance dates of the 2021 LCRR (86 FR 71574; USEPA, 2021a) on
June 16, 2021, to allow time for the EPA to review the rule in accordance with Presidential directives
issued on January 20, 2021 (Executive Order 13990), that directed Federal agencies to review certain
regulations and conduct important consultations with affected parties. The agency reviewed the 2021
LCRR to further evaluate if the rule protects families and communities, particularly those that have been
disproportionately impacted by lead in drinking water. The agency concluded that there are significant
opportunities to improve the 2021 LCRR. The EPA identified priority improvements for the Lead and
Copper Rule Improvements (LCRI): proactive and equitable lead service line replacement (LSLR),
strengthening compliance tap sampling to better identify communities most at risk of lead in drinking
water and to compel lead reduction actions, and reducing the complexity of the regulation through
improvement of the action and trigger level construct. Based on this review (also referred to as the LCRR
Review), the EPA decided to proceed with the development of the proposed LCRI that would revise
certain key sections of the 2021 LCRR while allowing the rule to take effect, and highlighted other
nonregulatory actions that the EPA and other Federal agencies could take to reduce exposure to lead in
drinking water. The purpose of this economic analysis (EA) is to provide additional technical information
on the final LCRI.5

A number of activities and sources of information and input have contributed to the development of the
final LCRI, including but not limited to LCR stakeholder meetings held by the EPA; input from the Science
Advisory Board (SAB); recommendations made by the National Drinking Water Advisory Council
(NDWAC) and its Lead and Copper Rule Working Group (LCRWG); comments received in response to
consultations with State, local, and tribal governments and intergovernmental organizations in 2018 and
in prior years; and comments received from the public in response to the November 13, 2019 proposed
LCRR. More currently, these include engagements conducted as part of the EPA's LCRR Review and
development of the proposed LCRI, consultations and engagements conducted with key stakeholders on
aspects of the 2021 LCRR, the EPA is considering for revision under the LCRI. These activities and sources
of input are described further in Section 2.2 and collectively contributed to the development of the final
LCRI as summarized in Chapter 1.

5 The EPA is required to adhere to the Administrative Procedure Act during the process of developing and issuing
regulations: https://www.archives.gov/federal-register/laws/administrative-procedure.

Final LCRI Economic Analysis

2-1

October 2024


-------
The remainder of this chapter is organized as follows:

•	Section 2.1 provides the statutory requirements, a chronology of the regulatory actions, and
initiatives affecting lead and copper in drinking water prior to the publication of the final LCRI.

•	Section 2.2 provides a description of the activities following the LCR Short-Term Revisions that
have informed development of the revised LCRI.

•	Section 2.3 includes a description of the EPA's mandated review of the 2021 LCRR and
subsequent consultations that the EPA considered in the development of the final LCRI.

•	Section 2.5 discusses regulatory authority for the regulation.

•	Section 2.6 discusses the economic rationale for the regulation.

2.1 Statutory Requirements, Regulatory Actions and National EPA Initiatives Affecting Lead
and Copper in Drinking Water

This section provides a chronology of regulatory actions and initiatives affecting lead and copper in
drinking water prior to the publication of the final LCRI.

2.1.1 Safe Drinking Water Act (SDWA) Requirements and Drinking Water Regulations Addressing
Lead Prior to 1991

SDWA (Public Law 93-523), passed in 1974, authorized the EPA to establish National Primary Drinking
Water Regulations (NPDWRs) for public water systems (PWSs). The EPA published national interim
primary drinking water regulations on December 24, 1975. Included among those regulations was a
maximum contaminant level (MCL) of 0.05 mg/L (or 50 ng/L) for lead. The monitoring requirements for
lead under these interim regulations focused on limiting the lead levels of drinking water entering the
distribution system. The supporting materials for these interim regulations (USEPA, 1976) recognized to
some degree that elevated lead levels were due to corrosion problems in the distribution system and
household plumbing; however, the regulation did not address this source of contamination.

Amendments to SDWA in 1986 (Public Law 99-339) required the use of "lead free" materials in the
installation or repair of pipes, fixtures, solders, and fluxes in any facility that provides water for human
consumption. As defined in SDWA Section 1417(d), "lead free" solders and fluxes may not contain more
than 0.2 percent lead, and "lead free" pipes, pipe fittings, and well pumps could not at the time contain
more than 8.0 percent lead. All States were required to implement the "lead ban" by August 6, 1988 (52
FR 20674; USEPA, 1987).

To limit children's exposure to lead, one of the most sensitive populations, Congress passed the Lead
Contamination Control Act (LCCA) (Public Law 100-572) in 1988 that further amended the SDWA. The
LCCA is aimed at the identification and reduction of lead in drinking water at schools and child care
centers, including the recall of drinking water coolers with lead lined tanks and the publication of a list
of drinking water coolers that were not "lead free." It required the EPA to provide guidance to States
and localities to test for and remedy lead contamination in drinking water at schools and child care

Final LCRI Economic Analysis

2-2

October 2024


-------
centers.6 In addition, the LCCA required testing, recall, repair, and/or replacement of water coolers with
lead-lined storage tanks or with other parts containing lead. One section of the LCCA that required
States to establish program to conduct testing and remedial actions has since been repealed as part of
the Water Infrastructure Improvements for the Nation Act (WIIN Act) (Public Law 114-322, Dec. 16,
2016; United States, 2016). Prior to the WIIN Act repeal of that section of the LCCA, in 1996, the U.S.
Court of Appeals for the 5th Circuit held that this now-repealed provision requiring States to establish
programs for testing and remediating lead was unconstitutional under the Tenth Amendment because it
directly compelled States to enact and enforce a federal regulatory program and provided no options for
the States to decline.7 Since that time, the EPA developed and revised its voluntary program for States,
schools, and child care facilities to address lead in drinking water (USEPA, 2018). In 2016, the WIIN Act
replaced the repealed version of Section 1464(d) of the SDWA with a new provision establishing a
voluntary school and child care lead testing grant program. 42 US.C. § 300j-24(d). Many States have also
enacted their own testing programs.8

2.1.2 Lead and Copper Rule (1991)

The 1986 SDWA amendments directed the EPA to revise the regulations for lead and copper in drinking
water. In response to this directive, the agency proposed revisions in 1988. On June 7, 1991, the EPA
promulgated the LCR (56 FR 26460; USEPA, 1991), and established a maximum contaminant level goal
(MCLG) of 0 for lead and 1.3 mg/L for copper. The LCR established treatment technique requirements
instead of an MCL. Section 1412(b)(7)(A) of SDWA authorizes the EPA to "promulgate a national primary
drinking water regulation that requires the use of a treatment technique in lieu of establishing an MCL,
if the Administrator makes a finding that it is not economically or technologically feasible to ascertain
the level of the contaminant." The EPA's decision to promulgate a treatment technique rule for lead
instead of an MCL in 1991 has been upheld by the United States Court of Appeals for the District of
Columbia Circuit. American Water Works Association (AWWA) v. EPA, 40 F.3d 1266, 1270-71 (D.C Cir.
1994).

In establishing treatment technique requirements, the Administrator is required to identify those
treatment techniques "which in the Administrator's judgment, would prevent known or anticipated
adverse effects on the health of persons to the extent feasible." 42 U.S.C. § 300g-l(b)(7)(A). "Feasible" is
defined in Section 1412(b)(4)(D) of SDWA as "feasible with the use of the best technology, treatment
techniques and other means which the Administrator finds after examination for efficacy under field

6	In response to the LCCA, the EPA developed the guidance document, "Lead in Drinking Water in Schools and Non-
residential Buildings in April 1994." Some states have initiated their own testing efforts, which may be dictated by
state-specific regulations. See Chapter 3, Section 3.3.10.2 for more information on States with lead testing
programs in schools and child care facilities.

7	No. 94-30714 (81 F.3d 1387) (5th Cir. April 22,1996). For more information about this case, see:
https://caselaw.findlaw.com/us-5th-circuit/1340297.html.

8	The EPA assessed existing State-level requirements for lead in drinking water testing in schools and child care
facilities. Currently 17 States have mandatory testing requirements for schools of which 16 have comparable
programs to the proposed rule. Eleven States have mandatory requirements for testing at child care facilities of
which nine have requirements that are comparable to the final LCRI mandatory testing criteria. All States have
received WIIN grant funding to conduct testing. See Chapter 3, Section 3.3.10.2 for additional detail on the EPA's
assessment of current State school and child care facility testing requirements, how they compare with the final
LCRI requirements, and the assumed application of WIIN grant funding for testing.

Final LCRI Economic Analysis

2-3

October 2024


-------
conditions and not solely under laboratory conditions, are available (taking cost into consideration)."
The 1991 LCR established requirements for PWSs to conduct tap sampling at households with plumbing
materials containing lead and copper. The 1991 LCR set an action level (AL) of 0.015 mg/L (or 15 ng/L)
for lead and 1.3 mg/L (or 1,300 ng/L) for copper. The AL is exceeded if the concentration in more than
10 percent of water samples (i.e., the 90th percentile level) collected at interior taps during any
monitoring period is greater than 0.015 mg/L for lead or 1.3 mg/L for copper. Water systems that
exceed the AL are not in violation of the LCR, but these systems are required to take actions to reduce
drinking water lead and copper exposure including corrosion control treatment (CCT),9 public education
(PE), and LSLR.

2.1.3	SDWA Amendments (1996)

The 1996 Amendments to SDWA added that "lead free" plumbing fittings and fixtures must meet
standards established under Section 1417(e) (42 U.S.C. 300g-6(e)). Section 1417(e) of SDWA required
the EPA to accept a voluntary standard within a year or issue a regulation within two years.

Furthermore, for the voluntary standard to be accepted, the EPA Administrator must provide technical
assistance to a qualified third-party in the development of the voluntary standard and associated testing
protocols for examining lead leaching from new plumbing fittings and fixtures.

In 1996, the National Sanitation Foundation (NSF) developed National Sanitation Foundation/American
National Standards Institute (NSF/ANSI) Standard 61, Section 9, which limits the amount of lead that can
be leached from endpoint devices for water intended for human consumption (NSF, 2019). The EPA
published, in the Federal Register (FR) (FR 62 44607; USEPA, 1997), its view that NSF 61, Section 9
satisfied the requirement of SDWA Section 1417(e). Specifically, the EPA found that NSF 61, Section 9 is
an established voluntary standard. Therefore, the obligation to issue a new regulation was not triggered.
As a result, from August 1997 to January 2014, only those plumbing fixtures and fittings that had a
maximum lead content of eight percent and were NSF/ANSI Standard 61, Section 9 certified could be
defined as "lead free" per SDWA.

2.1.4	Lead and Copper Rule Minor Revisions (2000)

On January 12, 2000, the EPA published the final Lead and Copper Rule Minor Revisions (65 FR 1950;
USEPA, 2000). The goals of the revisions were to streamline requirements, promote consistent national
implementation, and in many cases, reduce the burden for community water systems (CWSs) and non-
transient non-community water systems (NTNCWSs). The changes affected the following rule
requirements: demonstration of optimal corrosion control treatment (OCCT), LSLR, PE, monitoring,
analytical methods, reporting and recordkeeping, and special primacy considerations.

2.1.5	2004 National Review of the LCR Leading up to the LCR Short-Term Revisions of 2007

In early 2004, the EPA began a wide-ranging review of the implementation of the LCR in response to
high profile action level exceedances (ALEs) experienced by the District of Columbia Water and Sewer

9 The LCR required PWSs serving more than 50,000 people including those at or below their ALs to install CCT
unless they: 1) had completed treatment steps that are equivalent to those described in the 1991 LCR prior to
December 7,1992 or 2) could demonstrate they had very low levels of lead and copper in the distribution system
(i.e., qualified as a "b3" system).

Final LCR! Economic Analysis

2-4

October 2024


-------
Authority (DC Water, formerly known as DC Water and Sewer Authority). For a detailed discussion of the
elements of the review and consultations with the NDWAC and other key stakeholders, refer to Section

2.1.5	in Chapter 2 of the EPA's December 2020 Economic Analysis for the Final Lead and Copper Rule
Revisions (hereafter referred to as the "Final 2021 LCRR EA") (USEPA, 2020a).

2.1.6	Lead and Copper Rule Short-Term Revisions and Clarifications (2007)

The LCR Short-Term Revisions were published in the Federal Register on October 10, 2007 (72 FR 57782;
USEPA, 2007). This rulemaking contained additional requirements to improve the implementation of the
pre-2021 LCR. For additional information, refer Section 2.1.6 in Chapter 2 of the Final 2021 LCRR EA
(USEPA, 2020a).

2.1.7	Lead and Copper Rule Revisions (2021)

The 2021 LCRR was published in the Federal Register on January 15, 2021 (86 FR 4198; USEPA, 2021b)
with an effective date of March 16, 2021, and a compliance date of January 16, 2024. The 2021 LCRR
includes a suite of actions to address lead contamination in drinking water to improve the LCR and
further reduce lead exposure in comparison with the pre-2021 LCR. The 2021 LCRR created new
requirements for:

•	Water systems to develop an inventory of their service lines to better understand the number
and location of lead service lines (LSLs) in a community.

•	Tap sampling to require water systems to sample lead from sites served by LSLs and to collect a
fifth-liter sample in lieu of a first-liter sample at LSL sites.

•	A lead trigger level (TL) of 0.010 mg/L in addition to the lead AL. Systems that exceed the lead TL
must conduct actions sooner that may include a CCT study, adjustment to existing CCT, or
initiation of a goal-based LSLR program.

•	Improved CCT that requires systems to continue the CCT installation process even if they no
longer exceed the lead AL. Also requires systems to implement a find-and-fix approach to
evaluate individual sites with lead tap samples above 0.015 mg/L.

•	Enhanced LSLR requirements that include a two-year rolling average replacement rate of three
percent (that includes lead, galvanized requiring replacement, and unknown service lines) for
systems with a lead ALE, a goal-based LSLR program for systems that serve greater than 10,000
people that exceed the lead TL, and the elimination of the test-out provision that allowed LSLs
to remain in place if LSL samples do not exceed 0.015 mg/L.

•	Small system flexibility for CWSs serving 10,000 or fewer people and all NTNCWSs to select the
compliance option that best suits their system including CCT, LSLR, point-of-use (POU)
treatment, and replacement of lead-bearing plumbing materials.

•	Improved risk communication to require water systems to notify consumers within 24 hours if
the system exceeds the AL, to notify consumers whose individual tap sample exceeds 0.015
mg/L within 3 days, and to deliver PE materials to impacted consumers during water-related
work that may disturb LSLs. Also includes revisions to the CCT requirements to provide clear

Final LCRI Economic Analysis

2-5

October 2024


-------
health effects language, a statement on the availability of the service line inventory, and the
range of tap sample levels and public access to results.

•	CWSs to conduct PE and lead in drinking water testing at elementary schools and child care
facilities once over a five-year period and thereafter, conduct monitoring only at these facilities
that request testing. Conduct monitoring at secondary schools on request only.

On January 20, 2021, President Biden issued the "Executive Order on Protecting Public Health and the
Environment and Restoring Science to Tackle the Climate Crisis" (Executive Order 13990). In response to
Executive Order 13990, the EPA reviewed the 2021 LCRR to further evaluate if the rule protects families
and communities, particularly those that have been disproportionately impacted by lead in drinking
water. The EPA concluded that there are significant opportunities to improve the LCRR. For more details
on Executive Order 13990, refer to Section 2.3.1.

2.1.8 Additional Actions to Reduce Lead in Plumbing Materials (2008-present)

An annex to the NSF/ANSI Standard 61 was developed in 2008 that established a standard to determine
product compliance with the lead content requirements of California's Health and Safety Code Section
116875 (commonly known as California Assembly Bill 1953 [AB 1953]), which specifies a maximum
weighted average lead content of 0.25 percent calculated across the wetted surface of most plumbing
pipe, fittings, and fixtures. Further, more stringent requirements under NSF/ANSI Standard 61 leaching
standard (effective July 2012) include lowering the leaching standard from 11 ng/L to 3 ng/L under
Section 9 for supply stops, flexible plumbing, connectors, and miscellaneous components, and from 11
Hg/L to 5 ng/L for all other Section 9 devices (NSF, 2019).

Congress enacted the Reduction of Lead in Drinking Water Act (RLDWA) (Public Law 111-380) on January
4, 2011, to amend Section 1417 of SDWA to revise the definition of "lead free" in solder, flux, pipe, and
fixtures. The law reduced the level of permissible lead in drinking water plumbing fixtures from a
maximum of 8 percent to 0.2 percent lead in solder and flux and specifies a maximum weighted average
of 0.25 for wetted surfaces of most pipes, fittings, and fixtures. The RLDWA became effective on January
4, 2014. The Community Fire Safety Act of 2013 (Public Law 113-64) further amended Section 1417 to
exempt fire hydrants from having to meet the "lead free" requirements under the RLDWA. The EPA
announced the final rule titled "Use of Lead Free Pipes, Fittings, Fixtures, Solder, and Flux for Drinking
Water "on July 29, 2020 (85 FR 54235; USEPA, 2020b). This rule codified the requirements of the RLDWA
and established certification requirements for demonstrating compliance.

2.2 Outreach, Consultation, Workgroup Activities, and Other Events Contributing to the
Lead and Copper Rule Revisions

On January 15, 2021, the EPA published in the Federal Register the "National Primary Drinking Water
Regulation: Lead and Copper Rule Revisions" (86 FR 4198; USEPA, 2021b). The goal for the 2021 LCRR is
to improve public health protection provided by the LCR by making substantive changes to the rule
based on issues identified through the EPA's 2004 National Review and as described in the March 2005
Drinking Water Lead Reduction Plan (USEPA, 2005). To help the EPA better define these changes, the
agency:

•	Held various stakeholder meetings and consultations.

Final LCR! Economic Analysis

2-6

October 2024


-------
•	Charged the SAB to evaluate the effectiveness of partial LSLRs.

•	Solicited input from small business stakeholders.

•	Continued to consult with NDWAC, whose LCRWG was convened in 2014 and met during 2014 -
2015.

•	Consulted with tribal governments.

•	Held a public meeting on environmental justice.

•	Consulted with State, local government organizations, and PWSs.

•	Convened a meeting of high-level staff from the EPA, State, PWS, and non-government
organizations (NGOs).

Outreach activities and other events that impacted the 2021 LCRR are discussed in more detail in
Sections 2.2.1 through 2.2.10, and summaries and presentation materials, or other documents from
meetings and consultations discussed in these sections are available in the docket under EPA-HQ-OW-
2022-0801 at https://www.regulations.gov.

2.2.1	Stakeholder Meetings

In October 2008, the EPA held a two-day stakeholder meeting at the Carnegie Institution for Science in
Washington, D.C. The purpose of this meeting was to gather stakeholder input on areas to consider in
the revisions to the LCR. Stakeholders present at the meeting included State drinking water regulators,
members of city level water departments, regional water companies, State health departments, and
smaller water testing groups. Discussion topics included changes to the tiering criteria for lead and
copper sample site selection LSLR requirements, particulate lead in tap water samples, optimal water
quality parameters (OWQPs), tap sampling issues, and CCT technologies. The EPA presented summaries
of the scientific data that the agency had compiled on these issues. The EPA also requested stakeholder
input and feedback on these and other issues the EPA could consider for potential future action on the
LCR.

In November 2010, the EPA held a one-day stakeholder meeting in Philadelphia, PA. Expert participants
from utilities, academia, State governments, and other stakeholder groups met to discuss three areas
that the EPA considered for revision: tiering criteria for lead and copper sample site selection, LSLR
requirements, and potential requirements for testing of lead in drinking water at schools.

2.2.2	Input from Small Business Stakeholders

In July 2012, the EPA solicited input from the Small Business Administration, the Office of Management
and Budget (OMB), and nine potentially affected small entity representatives (SERs) on the LCRR,
pursuant to the Regulatory Flexibility Act (see Section 7.4). On August 14, 2012, the EPA convened a
Small Business Advocacy Review (SBAR) Panel and provided the Panel with input from the SERs. The
SBAR Panel submitted its report to the EPA in October 2012, which incorporated additional input from
the SERs. The report provided the number and type of small entities that may be affected by the
proposed rule; a recommendation to consider CCT techniques other than orthophosphate due to

Final LCR! Economic Analysis

2-7

October 2024


-------
possible conflicts with National Pollutant Discharge Elimination System permit limits for phosphorus;
and alternatives that would minimize any significant economic impact of the proposed rule on small
entities. Specifically, the Panel submitted recommendations regarding the sample site selection criteria,
PE for copper, the process for re-evaluating and revising CCT, copper monitoring waivers for systems
that can demonstrate their water is non-aggressive toward copper; POU treatment units in lieu of CCT
for NTNCWSs serving 10,000 or fewer people; the sampling protocol at sites served by LSLs; and
mandatory LSLR requirements.

2.2.3 Input from SAB and NDWAC

Throughout the LCRR rulemaking process, the EPA consulted with the SAB and the NDWAC. Sections
2.2.3.1 and 2.2.3.2 provide a summary of the EPA's consultations with the SAB and with the NDWAC,
respectively.

2.2.3.1	SAB Review

The SAB provides scientific advice to the EPA Administrator including reviewing the quality and
relevance of the scientific and technical information being used by the EPA or proposed as the basis for
agency regulations. This section describes consultations with the SAB during 2011 and 2020 on the LCRR.

2.2.3.1.1	2011 SAB Consultation

The EPA formally charged the SAB to review and provide advice regarding studies examining the
effectiveness of partial LSLRs. The SAB held a public meeting on this review on March 30 and 31, 2011 in
Washington, D.C. with a follow up conference call on May 16, 2011. SAB's final report, entitled "SAB
Evaluation of the Effectiveness of Partial Lead Service Line Replacements" was transmitted along with a
memorandum to the EPA Administrator on September 28, 2011 (USEPA, 2011a).

2.2.3.1.2	2020 SAB Review of the Proposed LCRR

Following the LCRR proposal, the SAB elected to review the scientific and technical basis of the proposed
rule, on March 30, 2020. The drinking water sub workgroup took the lead in the SAB deliberations on
this topic at a public teleconference held on May 11, 2020. The SAB provided advice and comments in its
June 12, 2020 report (USEPA, 2020c). SAB comments were similar to those raised by public commenters.
A copy of the report is included in the docket for the rule.

2.2.3.2	NDWAC Meetings

The NDWAC is a Federal Advisory Committee that supports the EPA in performing its duties and
responsibilities related to the national drinking water program and was created through a provision in
the SDWA in 1974. In accordance with Section 1412(d) and (e) of the SDWA, the EPA consulted with the
NDWAC on efforts to develop revisions to the LCR. These consultations are further described in this
section.

2.2.3.2.1 2011 NDWAC Consultation

On November 18, 2011, the EPA held a public teleconference with NDWAC to discuss a study completed
by the Centers for Disease Control and Prevention as well as to address the SAB evaluations regarding
partial LSLR. In December 2011, the NDWAC held a 2-day public meeting to address various issues

Final LCR! Economic Analysis

2-8

October 2024


-------
associated with drinking water protection including actions to assist small water systems. The NDWAC
provided the EPA with recommendations on the potential LCR regulatory revisions, which are outlined in
a letter dated December 23, 2011 (NDWAC, 2011).

2.2.3.2.2	2013 NDWAC Consultation

In December 2013, the EPA met with the NDWAC in Washington, D.C. to provide a national drinking
water program update (NDWAC, 2013). The EPA provided background on the LCRR and highlighted for
the Council five areas where the EPA was considering a range of regulatory revisions and seeking
detailed stakeholder input. The five areas were: 1) sample site selection criteria for tap monitoring, 2)
lead sampling protocol, 3) copper PE, 4) measure to ensure OCCT, and 5) LSLR. The public also had an
opportunity to provide information to the NDWAC on issues with which they were concerned and
wanted to be considered in the rule revisions. During this meeting, the EPA formally requested that the
NDWAC form a working group to support the EPA in the development of the LCRR. The NDWAC
unanimously voted on forming this working group. A summary of these LCRWG meetings is provided in
the next section.

2.2.3.2.3	2014-2015 NDWAC LCRWG Meetings

The NDWAC formed the LCRWG to provide additional advice to the EPA on potential options for the
LCRR. The 15-member LCRWG consisted of representatives from States, water systems, health agencies,
and public interest groups. The LCRWG held seven in-person meetings from March 2014 through June
2015, participated in multiple conference calls, and spent time outside these meetings to provide input
to the NDWAC on the five key issues that the EPA identified during the December 2013 NDWAC
meeting. The LCRWG also provided additional recommendations on other areas such as expanded lead
education and outreach and the need to engage other stakeholders that include the health community.

The LCRWG provided their final report, including recommendations, to the larger NDWAC committee in
August 2015 (NDWAC, 2015a) and presented their recommendations to the NDWAC in November 2015.
The NDWAC accepted the LCRWG recommendations and submitted their recommendation via letter to
the EPA on December 15, 2015 (NDWAC, 2015b).

In the report, the NDWAC acknowledged that reducing lead exposure is a shared responsibility among
consumers, PWSs, building owners, public health officials, and others. In addition, they recognized that
creative financing is necessary to reach the LSL removal goals, especially for disparate and vulnerable
communities. The NDWAC advised the EPA to maintain the LCR as a treatment technique rule but with
enhanced improvements. The NDWAC qualitatively considered costs before finalizing its
recommendations, emphasizing that PWSs and States should focus efforts where the greatest public
health protection can be achieved and incorporating their anticipated costs in their capital improvement
program or the requests for Drinking Water State Revolving Funds (DWSRF). The LCRWG outlined an
extensive list of recommendations for the LCRR including establishing a goal-based LSLR program,
strengthening CCT requirements, and tailoring water quality parameters (WQPs) to the specific CCT plan
for each water system.

The report the NDWAC provided for the EPA also included recommendations for renewed collaborative
commitments between all levels of government and the public while recognizing the EPA's leadership
role in this area. These complementary actions as well as a detailed description of the provisions for
NDWAC's recommendations for the proposed rule can be found in the "Report of the Lead and Copper

Final LCR! Economic Analysis

2-9

October 2024


-------
Rule Working Group to the National Drinking Water Advisory Council" (NDWAC, 2015a). One member of
the NDWAC working group provided a dissenting opinion (Parents for Nontoxic Alternatives, 2015). The
EPA took into consideration the NDWAC's recommendations and the dissenting opinion when
developing the final revisions to the LCR.

2.2.3.2.4 2019 NDWAC Consultation

On December 4-5, 2019, the EPA held a NDWAC meeting in Washington, D.C. where the EPA presented
the proposed LCRR. In the presentation, the major LCR revisions were highlighted (e.g., the LSL
inventory, the new TL of 10 ng/L, and new sampling protocols). The presentation focused on six key
areas: identifying areas most impacted, strengthening treatment requirements, replacing LSLs,
increasing sampling reliability, improving risk communication, and protecting children in schools. The
EPA reiterated the LCRR was developed with extensive consultation from State, local, and tribal partners
to identify opportunities that would reduce elevated levels of lead in drinking water. The EPA reaffirmed
its commitment to transparency and improved communication to the public.

2.2.4 Consultation with Tribal Governments

Consistent with the EPA Policy on Consultation and Coordination with Indian Tribes (May 4, 2011), the
EPA consulted with tribal officials during the development of the LCRR to gain an understanding of tribal
views of potential revisions to key areas of the LCR (USEPA, 2011b). The EPA coordinated and consulted
with federally-recognized Indian tribes on the LCR proposed regulatory revisions, pursuant to Executive
Order 13175, Consultation and Coordination With Indian Tribal Governments (65 FR 67249, November 9,
2000) (see Chapter 7, Section 7.7). Any revisions to the LCR will impact a tribal government that
operates a PWS or that has primary enforcement authority for PWSs on tribal lands. The EPA requested
input from tribal governments on how the agency should revise the LCR while maintaining or improving
public health protection. The EPA held tribal consultations, beginning with a national tribal consultation
teleconference on December 1, 2011 to obtain input from tribal governments on the proposed LCRR and
to determine which revisions would assist tribal governments in implementing and complying with the
rule while maintaining or improving public health.

From January 16 to March 16, 2018, the EPA held a consultation with federally-recognized Indian tribes.
The EPA sent a consultation invitation letter to all 567 federally-recognized tribes along with a
consultation and coordination plan, a link to written technical background information, and an invitation
to two national webinars for tribes. The first national webinar was held January 31, 2018, and a second
national webinar was held February 15, 2018. A total of 48 tribal representatives participated in the two
webinars. Updates on the consultation process were provided to the National Tribal Water Council,
upon request, at regularly scheduled monthly meetings during the consultation process. Also, upon
request, informational webinars were provided to the National Tribal Toxics Council's Lead
Subcommittee on January 30, 2018, and the EPA Region 9's Regional Tribal Operations Committee on
February 8, 2018. The information presented included key challenges to the previous LCR and potential
revisions regarding LSLR, CCT, tap sampling, PE and transparency, and copper requirements.

Five tribes or tribal organizations (Navajo Tribal Utility Authority, National Tribal Water Council, United
South and Eastern Tribes Sovereignty Protection Fund, Yukon River Inter-Tribal Watershed Council, and
Indian Health Service - Sanitation Facilities Construction, Seattle Office) submitted written consultation

Final LCR! Economic Analysis

2-10

October 2024


-------
comments to the EPA.10 A summary report of the views expressed during tribal consultations is available
in the docket (EPA-HQ-OW-2022-0801) at www.regulations.gov.

2.2.5	Public Meeting on Environmental Justice

During March 2011, the EPA held a public meeting to discuss and solicit input on environmental justice
considerations related to several upcoming regulatory efforts that included the LCRR. The meeting was
attended in-person and remotely by a diverse group including advocacy groups, water systems, State
agencies and trade associations, and private corporations. LSLR was a main area of discussion during this
meeting. The EPA provided information on the LCR and rule revisions that the agency was considering to
alleviate disproportionate impacts. The EPA also solicited input from the public regarding ways in which
the agency could further consider environmental justice concerns in the LCR revision process.

2.2.6	Consultation with State and Local Government Organizations

This section provides information on the EPA's 2011 and 2018 federal consultations and interactions
with the Association of State Drinking Water Administrators (ASDWA) on development of the LCRR.

2.2.6.1	November 2011 Federalism Consultation

On November 15, 2011, the EPA held a Federalism consultation with representatives from State and
local government organizations to solicit feedback on potential regulatory revisions to the LCR, pursuant
to Executive Order 13132, Federalism (64 FR 43255, August 10, 1999) (see Chapter 7, Section 7.6).

In its capacity as an advisory committee to the EPA, the Local Government Advisory Committee
periodically makes recommendations and comments to the agency on issues impacting local
governments. The EPA received comments that addressed sample site collection criteria and lead
sampling protocol at LSL sites.

2.2.6.2	ASDWA Questionnaire to States on Possible LCRR Requirements

In 2016, ASDWA developed a State questionnaire regarding potential LCRR requirements. The purpose
of the questionnaire was to obtain labor and cost estimates associated with some of the pre-2021 LCR
and potential requirements under the proposed LCRR to include in the Proposed LCRR EA (USEPA, 2019).
States were questioned about pre-2021 LCR oversight activities and additional implementation (i.e.,
sampling invalidation, WQP monitoring, CCT re-assessment, changes in source or treatment, and LSLR).
In terms of possible LCRR oversight activities, States were asked about burden and costs associated with
lead sampling instructions, updating the materials inventory, annual review of lead information,
discussion of sampling data during sanitary surveys, water aggressiveness to copper determinations,
drinking water treatment process control charting, periodic review of updated CCT guidance, and how
systems could demonstrate they had no LSLs. Two States (Indiana and North Carolina) responded to the
questionnaire.

10 More information on LCR-specific tribal consultation is available at the EPA's LCR website:
http://water.epa.gov/lawsregs/rulesregs/sdwa/lcr/index.cfm.

Final LCR! Economic Analysis

2-11

October 2024


-------
2.2.6.3	Questionnaire to States on LSL Inventory and Other LSL-Related Information

In 2017, the EPA disseminated a questionnaire to nine States regarding the burden and cost associated
with NDWAC's recommendation to require all systems to develop a comprehensive LSL inventory and to
expand the definition of an LSL to include lead connectors even if the service line is not made of lead.
The questionnaire asked States how they would manage the LSL inventory requirement and their
estimates for costs associated with reviewing PWS inventory documentation. The nine States were
selected based on geographic diversity, high incidence of LSLs, and knowledge of existing LSLR programs.
Seven States (Illinois, Michigan, New Jersey, Ohio, Rhode Island, Washington, and Wisconsin) out of the
nine States responded to the questionnaire.

2.2.6.4	January 2018 Federalism Consultation

Pursuant to Executive Order 13132, Federalism, the EPA held an initial Federalism meeting on January 8,
2018 in Washington, D.C. with 17 intergovernmental associations and several associations representing
State and local governments.11 EPA provided the associations' membership an opportunity to provide
input during follow-up meetings. The EPA also held five follow-up briefings between January 8 and
March 8, 2018. A total of 82 State and local governments and related associations provided input during
the meetings and within 60 days after the initial meeting. The EPA received comments from 24
municipal water utilities, 21 local government agencies, 20 intergovernmental associations, 15 State
agencies, and two Members of the United States House of Representatives. Common issues discussed
included LSLR, CCT, transparency and PE, tap sampling, and copper.

A summary report of the views expressed during Federalism consultations is available in the docket
(EPA-HQ-OW-2022-0801) at www.regulations.gov.

2.2.6.5	Meetings with ASDWA

This section describes the EPA's meetings with ASDWA during August 2018 to further discuss their
Federalism comments and March 2020 on projected State costs to implement the possible revisions to
the LCR.

2.2.6.5.1 AuRust 2018 MeetinR

The EPA met with ASDWA in August 2018 to further discuss ASDWA's comments provided during the
Federalism consultation period discussed above. The EPA gave an abbreviated version of the Federalism
presentation for the ASDWA members, highlighting the major topics the EPA was contemplating for
revision for the LCR. ASDWA presented preliminary estimates of State costs for CCT-related activities,
including State review of CCT and find-and-fix activities. ASDWA noted that they planned to continue to
refine their estimates and analysis and to eventually conduct a survey of their members. The EPA and
ASDWA also discussed LSLR, CCT, transparency and PE, tap sampling, and copper.

11 For more information regarding the LCR Federalism Consultation, refer to:
https://www.epa.gov/dwstandardsregulations/lcr-federalism-consultation.

Final LCRI Economic Analysis	2-12	October 2024


-------
2.2.6.5.2 March 2020 Meeting

The EPA met with ASDWA during March 2020 to discuss revisions to their 2018 Costs of State
Transactions Study (CoSTS) (ASDWA, 2018). The model projected the increase in the States' workload
from the anticipated revisions to the LCR. ASDWA submitted the 2018 version of the model during the
Federalism consultation and submitted a revised version to the EPA during the public comment period
for the proposed rule. The EPA revised several of its costing inputs used for the proposed rule to reflect
information provided in ASDWA's 2020 version of CoSTS (ASDWA, 2020). The file, "Final CoSTS 2-6-
20.xlsx" is available in the LCRR docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

2.2.7 Public Water Systems

The lead in drinking water crisis in Flint, Ml12 brought increased attention to lead in drinking water and
to the need to improve the pre-2021 LCR. It underscored significant challenges in the implementation of
the pre-2021 LCR, including a rule structure that for many systems only compels protective actions after
public health threats have been identified (USEPA, 2016a). The EPA took into account the experience in
Flint, Ml in developing the LCRR. In addition, the EPA solicited input from other PWSs across the country
regarding burden and costs of potential revisions to consider in the development of the LCRR. A
summary of the input from PWSs (through the dissemination of surveys and questionnaires) is discussed
in Section 2.2.7.1.

2.2.7.1 Input from PWSs

The EPA sought input from PWSs regarding the cost and burden of potential provisions in the LCRR.
Specifically, the EPA issued questionnaires to the New York City Department of Environmental
Protection and the Chicago Department of Water Management about their free lead in drinking water
testing program and to nine systems regarding their LSL inventories. The EPA also met with systems by
phone to obtain information. The EPA met with Greater Cincinnati Water Works (GCWW) about their
school testing program for lead in drinking water and with the Philadelphia Water Department (PWD)
regarding their protocol to address high lead levels at individual households. Each of these is discussed
in more detail in the following sections.

2.2.7.1.1 New York City Department of Environmental Protection and Chicago Department of
Water Management

The EPA sent a questionnaire in 2016 to the New York City Department of Environmental Protection and
the Chicago Department of Water Management regarding their free testing programs for lead in
drinking water. The purpose of this questionnaire was to give the EPA a sense of the burden and cost
associated with implementing such a program. In particular, the questionnaire asked about when these
programs were started, methods of advertising and communication, how many customers requested
sampling per year, percentage of sample results that exceeded the lead AL, public accessibility of the
lead results, and other types of testing and analyses offered to customers.

12 See, https://www.epa.gov/flint for additional information on the EPA's Flint drinking water response along with
website links to additional information. Also see, https://www.cdc.gov/nceh/lead/programs/flint-registrv.htm, the
Centers for Decease Control and Preventions' voluntary Flint lead exposure registry website.

Final LCR! Economic Analysis

2-13

October 2024


-------
2.2.7.1.2	LSL Inventory and LSLR Questionnaire

The EPA sent a questionnaire to nine PWSs with active LSLR programs. The questionnaire was designed
to obtain information about the activities and costs needed to develop a comprehensive LSL inventory,
how systems have achieved successful LSLR programs, and the cost associated with LSLR. Fort Worth
was the only PWS to respond to the questionnaire.

2.2.7.1.3	Greater Cincinnati Water Works

On May 25, 2018, the EPA met with GCWW to discuss their proactive school testing program for lead in
drinking water. Representatives from GCWW provided an overview of the program and discussed the
services offered to schools, the roles of other agencies in the program, and the integration of child cares
into the program. GCWW also provided the EPA with an Excel spreadsheet that outlined the steps taken
to sample at a school, average time it takes to complete each step, and the average cost per school.

2.2.7.1.4	Philadelphia Water Department

The EPA met with PWD on November 2, 2018 to discuss how the system addresses high lead levels at
individual residences. PWD served on the NDWAC LCRWG and indicated that PWD conducts find-and-fix
steps when LCR compliance sampling yields high lead results. During this meeting, PWD discussed its
free lead tap sampling program for customers who request testing. PWD also provided the EPA with
some of its lead PE materials.

2.2.8	EPA Letter to Governors and State Environment and Public Health Commissions and Tribal
Leaders

In 2016, the EPA sent letters to Governors, State Environment and Public Health Commissioners, and
Tribal Leaders regarding the LCR.13 The intent of the letters was to ensure that the LCR was being
properly implemented. In the letter, the EPA explained their immediate effort to oversee State
implementation of the LCR and to work with States to identify ways to strengthen implementation and
ultimately improve public health protection. The letter also asked these parties to take action to
improve public transparency and accountability in the implementation of the rule.

2.2.9	Administrator's Meeting with States, PWS, and Non-Government Organizations

In May and June of 2016, the Administrator and other high-ranking EPA officials conducted meetings
with State officials, water system officials, and NGOs. Sixteen State officials and 16 PWS officials met
with the EPA on May 26 and June 1, 2016, respectively. The EPA met with 15 NGOs on June 2, 2016.
During each meeting, the EPA and stakeholder officials discussed critical needs and key opportunities for
addressing drinking water challenges and four priority issues including the LCR with the goal of
strengthening implementation of the pre-2021 LCR and improving public health protection through
updates to the rule. The results of these meetings informed the EPA's Drinking Water Action Plan,
published in November 2016 (USEPA, 2016b).

13 For templates of these letters and stakeholder responses, refer to: https://www.epa.gov/dwreginfo/epa-letter-
governors-and-state-environment-and-public-health-commissioners.

Final LCR! Economic Analysis

2-14

October 2024


-------
2.2.10 Public Comments on the Proposed LCRR

Following publication of the proposed LCRR, the EPA accepted public comments for 90 days. The EPA
received comments from over 79,000 individuals and organizations representing a wide range of
stakeholders, including PWSs, States, tribes, other organizations, and private citizens. Each unique
comment was read and considered in determining the final rule requirements. A record of the
comments received on the proposal, as well as the EPA's responses to these comments, can be found in
the "Public Comment and Response Document for the Final Lead and Copper Rule Revisions" (USEPA,
2020d). Copies of unique individual comments are also available as part of the public record and can be
accessed through the EPA's docket (EPA-HQ-OW-2017-0300 at www.regulations.gov).

2.3 Outreach, Consultation, and Other Engagements Contributing to the Proposed Lead and
Copper Rule Improvements

This section provides a summary of the EPA's engagements that occurred as part of the LCRR Review
(Section 2.3.1) and engagements and consultations held to support the development of the proposed
LCRI (Section 2.3.2). The EPA's summaries and presentation materials, or other documents from
meetings and consultations discussed in these sections are available in the docket for the proposed rule
under EPA-HQ-OW-2022-0801 at https://www.regulations.gov.

2.3.1 LCRR Review

On January 15, 2021, the EPA published in the Federal Register the "National Primary Drinking Water
Regulation: Lead and Copper Rule Revisions" (86 FR 4198; USEPA, 2021b) with an effective date of
March 16, 2021, and a compliance date of January 16, 2024. On January 20, 2021, President Biden
issued the "Executive Order on Protecting Public Health and the Environment and Restoring Science to
Tackle the Climate Crisis" (Executive Order 13990).

Section 1 of Executive Order 13990 states that it is "the policy of the Administration to listen to the
science, to improve public health and protect our environment, to ensure access to clean air and water.
. ., and to prioritize both environmental justice and the creation of the well-paying union jobs necessary
to deliver on these goals." Executive Order 13990 directs the heads of all Federal agencies to
immediately review regulations that may be inconsistent with, or present obstacles to, the policy it
establishes. On March 12, 2021, the EPA published the National Primary Drinking Water Regulations:
Lead and Copper Rule Revisions; Delay of Effective Date (86 FR 14003; USEPA, 2021c), which delayed the
effective date of the LCRR from March 16, 2021, to June 17, 2021. On the same day, the EPA published
the National Primary Drinking Water Regulations: Lead and Copper Rule Revisions; Delay of Effective and
Compliance Dates (86 FR 14063; USEPA, 2021d), which proposed further delaying the effective date of
LCRR to December 16, 2021 to allow the EPA to "conduct a review of the LCRR and consult with
stakeholders, including those who have been historically underserved by, or subject to discrimination in,
Federal policies and programs prior to the LCRR going into effect" (86 FR 14063; USEPA, 2021d). On June
16, 2021, the EPA published a final rule, the National Primary Drinking Water Regulations: Lead and
Copper Rule Revisions; Delay of Effective and Compliance Dates (86 FR 31939; USEPA, 2021e), which
delayed the LCRR effective date until December 16, 2021, and the compliance date until October 16,
2024. While the LCRR was delayed, the EPA engaged with stakeholders to better understand their
thoughts and concerns about the LCRR.

Final LCRI Economic Analysis

2-15

October 2024


-------
The EPA hosted a series of virtual engagements from April to August 2021 to obtain public input on the
review of the LCRR. The EPA also opened a docket, from April 5, 2021, to July 30, 2021, to accept written
comments, suggestions, and data from the public. Summaries of these engagements, including
summaries of the meetings and written comments, can be found in the docket, the EPA-HQ-OW-2021-
0255 at https://www.regulations.gov/. Recordings of the public listening sessions and community, tribal,
and national stakeholder association roundtables can also be found in the docket. The virtual
engagement meetings included two public listening sessions, 10 community roundtables, a tribal
roundtable, a national stakeholder association roundtable, a national co-regulator meeting, and a
meeting with organizations representing elected officials.

The EPA specifically sought engagement with communities that have been disproportionately impacted
by lead in drinking water, especially lower-income people and communities of color that have been
underrepresented in past rule-making efforts. The EPA hosted roundtables with individuals and
organizations from Pittsburgh, Pennsylvania; Newark, New Jersey; Maiden, Massachusetts; Washington,
D.C.; Newburgh, New York; Benton Harbor and Highland Park, Michigan; Flint and Detroit, Michigan;
Memphis, Tennessee; Chicago, Illinois; and Milwaukee, Wisconsin. These geographically-focused
roundtables included a range of participants including local government entities, community
organizations, environmental groups, local public water utilities, and public officials. The EPA worked
with community representatives to develop meeting agendas that reflected community priorities. Each
community roundtable included a presentation by local community members. The EPA held a separate
roundtable with representatives from tribes and tribal communities. Participants in all roundtables were
invited to share diverse perspectives with the agency through verbal discussion and a chat feature. The
EPA obtained detailed, valuable feedback from these engagements, which often focused on the lived
experiences of people impacted by lead in drinking water.

On December 17, 2021, the EPA published its findings from the review that included specific areas on
which commenters provided feedback and the agency's intention to develop a new rule to revise the
LCRR (86 FR 71574, USEPA, 2021a). Specific areas on which commenters provided feedback included:

•	Concern that the LCRR would not provide equitable public health protections, may create
confusion about drinking water safety, and would be difficult to implement.

•	The topic of 100 percent LSLR including the replacement timeframe and obstacles to achieving
it.

•	The lead AL and TL including whether to lower the AL and/or remove the TL.

•	Tap sampling requirements for systems with LSLs.

•	Requiring accessible PE materials and outreach to residents about lead risk and suggestions for
achieving this goal.

•	Water testing at schools and child care facilities.

•	Suggestions for revising the CCT, WQP and find-and-fix requirements.

•	Limiting the small system flexibility provisions.

Final LCRI Economic Analysis

2-16

October 2024


-------
2.3.2 Consultations and Engagements to Support the Development of the Proposed LCRI

The EPA held consultations and engagements September 2022 through August 2023 to obtain additional
feedback on areas the EPA identified for improvement during the LCRR Review. These consultations and
engagements are provided in more detail in Sections 2.3.2.1 through 2.3.2.7.

2.3.2.1	Small Business Stakeholders

On September 12, 2022, the EPA conducted a SBAR pre-panel outreach meeting to solicit input from 11
SERs on the potential small systems implications of the forthcoming proposed LCRI.14 EPA received
verbal comments from seven small entities and written comments from four small entities. SER
comments included challenges in achieving 100 percent LSLR in small systems including acquiring
contractor support, engaging customers and local governments, financial and administrative burden,
incorporating equity, and service line ownership issues. SERs also provided comments on complying with
a revised TL and AL construct, complying with revised tap sampling protocol, the need for national
training and technical assistance, concerns with school and child care facility sampling, small system
flexibility, issues with public notification requirements, opportunities to reduce burden through
clarifications, and simultaneous compliance with other rules.

On November 29, 2022, the EPA convened a second SBAR outreach panel to solicit further input from
SERs. A total of eight SERs attended the meeting, with six providing verbal comments and six providing
written comments following the meeting. The comments included incorporating equity into LSLR,
importance of funding and support, and challenges of service line ownership. SERs also provided
comments on challenges in collecting first- and fifth-liter samples, relating tap monitoring results to risk
communication and CCT, the effect of the AL on tier 1 public notice, and simplifying the rule. Additional
comments included concerns for schools, risk reduction through filters and bottled water, the rule
implementation timeframe, and additional regulatory flexibility including compliance options for water
corrosivity and LSLR and compliance options for low- or no-lead systems. The SBAR panel submitted its
report to the EPA on May 31, 2023.

2.3.2.2	Public Meeting on Environmental Justice

The EPA held two public meetings related to environmental justice (EJ) and the development of the
proposed LCRI on October 25, 2022, and November 1, 2022. These sessions provided opportunities for
the EPA to share information about the upcoming LCRI rulemaking and for individuals to offer input on
EJ considerations related to the rule. During the meeting, the EPA presented a brief overview of lead
health effects, lead occurrence in drinking water, and the SDWA process for developing a drinking water
regulation, in particular highlighting the EJ-related components. The EPA received public input through
verbal and written public comments, as well as interactive polling responses. The EPA received a total of
30 public comments during the 60-day post meeting comment period. Public comments included
incorporating equity into 100 percent LSLR replacement goals, methods of identifying and prioritizing
disadvantaged communities who are disproportionately impacted by lead in in drinking water for LSLR,
and methods of overcoming customers' financial and access barriers to full LSLR. A summary report of

14 For more information about the SBAR panel, visit https://www.epa.gov/reg-flex/potential-sbar-panel-national-
primarv-drinking-water-regulation-lead-and-copper-rule

Final LCRI Economic Analysis

2-17

October 2024


-------
the views expressed during both EJ consultations is available in the docket (EPA-HQ-OW-2022-0801) at
www.regulations.gov.

2.3.2.3	Consultation with Tribal Governments

The EPA initiated consultations and coordination with federally recognized Indian tribes to obtain input
on the agency's proposed LCRI, pursuant to Executive Order 13175, Consultation and Coordination With
Indian Tribal Governments (Executive Order 13175). The EPA signed a tribal consultation notification
letter inviting tribal officials to participate in consultation and coordination events and provide
comments to the EPA, and emailed this letter to all 574 federally-recognized tribal leaders at that time.
In addition to the consultation invitation letter, the EPA provided a consultation and coordination plan
background information, and an invitation to two national informational webinars for tribal
governments. All tribal consultation materials were made available via the EPA's Tribal Consultation
Opportunities Tracking System (https://tcots.epa.gov).

The national informational webinars were held on October 27, 2022, and November 9, 2022. Consistent
with the EPA Policy on Consultation and Coordination with Indian Tribes (May 4, 2011), the EPA
consulted with tribal officials to gain an understanding of tribal views of key areas of the proposed LCRI.
As part of the meeting, the EPA representatives presented background information on the pre-2021 LCR
and LCRR regulations regarding lead and copper content in drinking water. The EPA also presented on
the rule considerations for the proposed LCRI. During the consultation process, the EPA requested input
from tribal governments on considerations to inform the development of the proposed LCRI, including
elements related to potential regulatory requirements and suggestions that would assist tribal
governments in implementing and complying with the rule. Four specific areas of the proposed rule on
which the EPA requested input included achieving 100 percent LSLR, tap sampling and compliance,
reducing rule complexity, and small system flexibility.

A total of 11 tribal representatives participated in the two webinars. Webinar participants provided
verbal comments, but the EPA did not receive any written consultation comments from tribal
organizations during the comment period that followed the webinars.

2.3.2.4	SAB Consultation

The EPA consulted with the 37 members of the SAB on the key areas being considered for the proposed
LCRI and tools, indicators, and measures for use in future analyses to determine EJ impacts of LSL
presence and replacement in drinking water systems. Prior to the meeting, which was held on
November 3, 2022, the EPA provided the charge to the SAB and shared the agency's preliminary
analyses and draft results on case studies for three cities to help inform the agency's EJ analysis for the
proposed LCRI (USEPA, 2022). The EPA charged the SAB with the following three questions:

•	Are there potential EJ concerns associated with environmental stressors affected by the
regulatory action for population groups of concern in the baseline?

•	Are there potential EJ concerns associated with environmental stressors affected by the
regulatory action for population groups of concern for each regulatory option under
consideration?

Final LCRI Economic Analysis

2-18

October 2024


-------
• For each regulatory option under consideration, are potential EJ concerns created or mitigated
compared to the baseline?

The SAB provided initial verbal advice and comments on the proposed rule and case studies, as well as
written comments through November 21, 2022. The SAB provided its final report to the EPA
Administrator on December 20, 2022, regarding the agency's EJ analysis for LCRI (USEPA SAB, 2022).

2.3.2.5	NDWAC Consultation

On December 1, 2022, the EPA held a public teleconference with NDWAC during which the EPA
presented the proposed LCRI and solicited input from the NDWAC. The EPA provided background on
lead in drinking water and the LCR, an overview of the LCRR published in January 2021, annualized cost
estimates from the LCRR EA, and a summary of the outcome of the EPA's review of the LCRR. The
NDWAC provided key input on four key areas: achieving 100 percent LSLR, tap sampling and compliance,
reducing rule complexity, and small system flexibility. The public was also given an opportunity to
provide their comments to the NDWAC.

2.3.2.6	2022 Federalism/Unfunded Mandates Reform Act (UMRA) Consultation

The Federalism Consultation began on October 13, 2022, and ended on December 13, 2022. On
September 29, 2022, the Director of the Office of Ground Water and Drinking Water (OGWDW), Jennifer
McLain, signed a Federalism consultation notification letter inviting State and local government officials
as well as their representative associations to participate in a meeting and consultation and provide
comments to the EPA during the consultation process. The EPA sent this letter to a number of State and
local agencies as well as several water and utility professional organizations that may have State and
local government members.

The EPA held the Federalism and UMRA meeting on October 13, 2022. During the meeting, the EPA
presented background information and questions for feedback on key areas of the proposed rule. The
EPA specifically requested input on the following key rule areas: achieving 100 percent LSLR, tap
sampling and compliance, reducing rule complexity, and small system flexibility. Fifteen organizations,
as well as several associations with expertise in drinking water, were represented at the
Federalism/UMRA consultation meeting. Although this virtual briefing was for intergovernmental
association staff only, participants were able to schedule follow-up briefings for their memberships and
were encouraged to forward the briefing information and materials to their members. The EPA provided
a 60-day public comment period following the October 13, 2022 meeting.

2.3.2.7	Meetings with ASDWA

The EPA met with ASDWA on October 5, 2022, to solicit feedback from State co-regulators on the
development of LCRI. A total of 21 State co-regulators from 16 States participated in this early
engagement meeting, in addition to 5 representatives from ASDWA and 10 representatives from the
EPA OGWDW. The EPA representatives presented background regarding the pre-2021 LCR, an overview
of LCRR, and cost information for actions to reduce drinking water lead levels. ASDWA and State co-
regulators discussed how quickly systems can achieve 100 percent LSLR, factors that impact a system's
rate of LSLR, barriers to engaging customers for full LSLR, and how systems can ensure equity in
replacements.

Final LCRI Economic Analysis

2-19

October 2024


-------
The EPA held a second meeting with ASDWA on November 2, 2022, at which a total of 18 co-regulators
from 15 States participated, in addition to 5 representatives from ASDWA and 6 representatives from
the EPA OGWDW. The EPA representatives asked ASDWA and State co-regulators to provide feedback
on tap sampling and compliance and opportunities to reduce complexity mainly around the AL and TL
construct. In addition to these topics, ASDWA and State co-regulators discussed CCT, WQPs, find-and-fix
provisions, school and childcare sampling, PE, and Safe Drinking Water Information System (SDWIS)
capabilities to track data.

2.3.2.8 HHS Consultation

On August 18, 2023, the EPA conducted a virtual consultation meeting with the Department of Health
and Human Services (HHS) on the proposed LCRI. The purpose of the meeting was to provide an
overview of the proposed rule and to allow participants to ask clarifying questions. HHS participants
sought clarifications on full LSLR, justification for changing the lead AL, factors influencing water
systems' ability to meet the lead AL, regulatory authority over schools and child care centers, lead tap
sampling in schools and child care centers, language accessibility of PE materials, small system
flexibilities, use of LSL inventory data, materials used to replace LSLs, and resources for protecting
workers during LSLR. The EPA considered HHS input as part of the interagency review process.

2.3.3 Public Water Systems

On December 7, 2023, the EPA sent a questionnaire to nine water systems regarding the burden and
cost to develop and maintain a service line inventory under the LCRR. The EPA requested feedback by
February 28, 2023, and received responses from three water systems, Grand Rapids, Michigan;
Pittsburgh, Pennsylvania; and Cincinnati, Ohio (USEPA, 2023a). As explained in Chapter 5, Section
5.3.4.1, the EPA used the information from these three water systems among other sources to help
develop burden and cost related to service line inventory updates and validation.

2.4 Outreach, Consultation, and Other Engagements Contributing to the Final Lead and
Copper Rule Improvements

This section provides a summary of the EPA's engagements, consultations, and opportunity for public
comment that occurred to support finalizing the LCRI. The EPA's summaries and presentation materials
from meetings and consultations discussed in these sections are available at the EPA's website
https://www.epa.gov/ground-water-and-drinking-water/proposed-lead-and-copper-rule-improvements
and are available in the docket for the proposed rule under EPA-HQ-OW-2022-0801 at
https://www.regulations.gov.

2.4.1 Informational Webinar and Public Hearing

This section provides a summary of two webinars and a public hearing on the proposed LCRI that are
discussed in more detail in Section 2.4.1.1 through Section 2.4.1.3. The EPA's summaries and
presentation materials from these meetings are available at the EPA's website
https://www.epa.gov/ground-water-and-drinking-water/proposed-lead-and-copper-rule-
improvements.

Final LCRI Economic Analysis

2-20

October 2024


-------
2.4.1.1	Webinar on Preparing Communities to Engage in the Proposed LCRI Regulatory Process

On October 17, 2023, the EPA held a webinar to provide information to the public on how to participate
in the rulemaking process and how to offer the EPA input on the proposed LCRI.

2.4.1.2	Informational Webinar

On December 6, 2023, the EPA held an informational webinar on the proposed LCRI and to provide
information on the public comment period and how the public can submit their comments to the
docket.

2.4.1.3	Public Hearing

On January 16, 2024, the EPA held a virtual public hearing to provide an opportunity for the public to
share their input on the proposed LCRI. Members of the public were notified of the public hearing on
December 6, 2023, through the Proposed LCRI Federal Register Notice and on the EPA's LCRI website.
Eighty-two individuals or organizations provided testimony. Their comments were considered equally to
the written public comments received through the docket (see Section 2.4.2). Their comments can be
accessed through the docket at https://www.regulations.gov/comment/EPA-HQ-OW-2022-0801-2268.

2.4.2	Public Comments on the Proposed LCRI

Following publication of the proposed LCRI on December 6, 2023 (USEPA, 2023b), the EPA accepted
public comments for 60 days. The EPA received nearly 200,000 comments from individuals or
organizations representing a wide range of stakeholders, including PWSs, States, tribes, other
organizations, and private citizens. Each unique comment including those from the January 16, 2024
public hearing were read and considered in determining the final rule requirements. A record of the
comments received on the proposal, as well as the EPA's responses to these comments, can be found in
the Public Comment and Response Document for the Final Lead and Copper Rule Improvements (USEPA,
2024a). Copies of unique individual comments are also available as part of the public record and can be
accessed through the EPA's docket (EPA-HQ-OW-2022-0801 at www.regulations.gov).

2.4.3	Input from NDWAC

On January 31, 2024, the EPA held a public teleconference to consult with the NDWAC on five key areas
for the final rule: (1) achieving 100 percent lead pipe replacement within 10 years, (2) locating legacy
pipe, (3) improving tap sampling and compliance, (4) lowering the lead AL, and (5) strengthening
protection to reduce exposure. The public was also given an opportunity to provide their comments to
the NDWAC. A summary of the NDWAC meeting, the public comments to the NDWAC, and the EPA's
presentation are available in the NDWAC Summary Report (NDWAC, 2024) and is also available in the
docket.

2.4.4	HHS Consultation

On July 15, 2024, the EPA consulted with HHS for the final LCRI. The purpose of the meeting was to
provide an overview of the final rule and to allow participants to ask clarifying questions. HHS
participants sought clarifications on developing equitable service line replacement plans, public

Final LCRI Economic Analysis

2-21

October 2024


-------
availability of the service line inventory, communication with consumers about identification of lead
status unknown service lines, communication with health care providers about lead and steps to protect
health, lead tap sampling results in schools and child care facilities, requirements for water systems with
multiple lead action level exceedances, EPA communication with National Institutes of Health (NIH) on
occurrence of lead action level exceedances, and communications materials around promulgation of the
final rule. The EPA considered HHS input as part of the interagency review process in accordance with
Executive Order 12866: Regulatory Planning and Review.

2.5 Statutory Authority for Promulgating the Rule

The EPA derives its statutory authority to regulate contaminants in drinking water through the SDWA.
The SDWA requires the EPA to establish MCLGs and NPDWRs for contaminants that may have an
adverse effect on the health of persons and may occur in systems at a frequency and level of public
concern and for which, in the sole judgment of the Administrator, regulation of the contaminant would
present a meaningful opportunity for health risk reduction for persons served by PWSs (SDWA Section
1412(b)(1)(A)). The 1986 amendments to the SDWA established a list of 83 contaminants for which the
EPA is to develop MCLGs and NPDWRs, which included lead and copper. The 1991 NPDWR for lead and
copper (56 FR 26460; USEPA, 1991) fulfilled the requirements of the 1986 SDWA amendments with
respect to lead and copper.

The EPA is finalizing revisions to the lead and copper regulations under the authority of the following
sections of the SDWA: 1412, 1413, 1414, 1417, 1445, and 1450 (42 U.S.C. §§ 300f et seq.).

Section 1412(b)(7)(A) of the SDWA authorizes the EPA to promulgate a treatment technique "which in
the Administrator's judgement, would prevent known or anticipated adverse effects on the health of
persons to the extent feasible." (42 U.S.C. § 300g-l(b)(7)(A)). Section 1412(b)(9) provides that "[T]he
Administrator shall, not less often than every 6 years, review and revise, as appropriate, each national
primary drinking water regulation promulgated under this subchapter. Any revision of a national
primary drinking water regulation shall be promulgated in accordance with this section, except that each
revision shall maintain, or provide for greater, protection of the health of persons." (42 U.S.C. § 300g-
1(b)(9)). In finalizing a revised NPDWR, the EPA follows the applicable procedures and requirements
described in Section 1412, including those related to 1) the use of the best available, peer-reviewed
science and supporting studies; 2) presentation of information on public health effects; and 3) a health
risk reduction and cost analysis of the rule (42 U.S.C. § 300g-l(b)(3)(A)-(C)).

Section 1413(a)(1) of the SDWA allows the EPA to grant a State primary enforcement responsibility
("primacy") for NPDWRs when the EPA has determined that the State has adopted regulations that are
no less stringent than the EPA's regulations (42 U.S.C. § 300g-2(a)(l)). To obtain primacy for this rule,
States must adopt comparable regulations within two years of the EPA's promulgation of the final rule,
unless the EPA grants the State a two-year extension. State primacy requires, among other things,
adequate enforcement (including monitoring and inspections) and reporting. The EPA must approve or
deny State primacy applications within 90 days of submission to the EPA (42 U.S.C. § 300g-2(b)(2)). In
some cases, a State submitting revisions to adopt an NPDWR has primary enforcement authority for the
new regulation while the EPA's decision on the revision is pending (42 U.S.C. § 300g-2(c)).

Final LCRI Economic Analysis

2-22

October 2024


-------
Section 1414(c) of the SDWA, as amended by the WIIN Act, requires PWSs to provide notice to the
public if the water system exceeds the lead AL (42 U.S.C. § 300g-3(c)). Section 1414(c)(2) provides that
the Administrator "shall, by regulation ... prescribe the manner, frequency, form, and content for giving
notice" (42 U.S.C. § 300g-3(c)(2)). Section 1414(c)(2)(C) specifies additional requirements for those
regulations related to public notification of a lead ALE "that has the potential to have serious adverse
effects on human health as a result of short-term exposure," including requirements for providing
notification to the EPA.

Section 1417(a)(2) of the SDWA provides that PWSs "shall identify and provide notice to persons that
may be affected by lead contamination of their drinking water where such contamination results from
the lead content of the construction materials of the public water distribution system and/or corrosivity
of the water supply sufficient to cause leaching of lead" (42 U.S.C. § 300g-6(a)(2)(A)(i) and (ii)). The
notice "shall be provided notwithstanding the absence of a violation of any national drinking water
standard" (42 U.S.C. § 300g-6(a)(2)(A)).

Section 1445(a) of the SDWA authorizes the Administrator to establish monitoring, recordkeeping, and
reporting regulations, to assist the Administrator in establishing regulations under the SDWA,
determining compliance with the SDWA, and in advising the public of the risks of unregulated
contaminants (42 U.S.C. § 300j-4(a)). In requiring a PWS to monitor under Section 1445(a), the
Administrator may take into consideration the water system size and the contaminants likely to be
found in the system's drinking water (42 U.S.C. § 300j-4(a)). Section 1445(a)(1)(C) of the SDWA provides
that "every person who is subject to a national primary drinking water regulation" must provide such
information as the Administrator may reasonably require to assist the Administrator in establishing
regulations under Section 1412 (42 U.S.C § 300j-4(a)(l)(C)).

Section 1450 of the SDWA authorizes the Administrator to prescribe such regulations as are
necessary or appropriate to carry out his or her functions under the Act (42 U.S.C § 300j-9).

2.6 Economic Rationale

This section addresses the economic rationale, as described in Executive Order 12866, Regulatory
Planning and Review (58 FR 51735, October 4, 1993), for choosing a regulatory approach to regulate
lead and copper levels in drinking water supplies rather than nonregulatory alternatives. Executive
Order 12866 states the following:

[E]ach agency shall identify the problem that it intends to address (including, where applicable,
the failures of the private markets or public institutions that warrant new agency action) as well
as assess the significance of that problem (Section 1, b(l)).

In addition, OMB Circular A-4, dated September 17, 2003, states that:

"... [the analyst] should try to explain whether the action is intended to address a significant
market failure or to meet some other compelling public need such as improving governmental
processes or promoting intangible values such as distributional fairness or privacy" (USOMB,
2003).

In the case of the final LCRI, several properties of public water suppliers do not satisfy the conditions for
a perfectly competitive market and thus lead to market failures that require regulation. In a perfectly

Final LCRI Economic Analysis

2-23

October 2024


-------
competitive market, prices and quantities are determined solely by the aggregated decisions of buyers
and sellers. Such a market occurs when many producers of a product are selling to many buyers, and
where both producers and consumers have perfect information on the characteristics and prices of each
firm's products. Barriers to entry in the industry cannot exist, and individual buyers and sellers must be
"price takers" (i.e., their individual decisions cannot affect the price).

Many water systems are natural monopolies. A natural monopoly exists when the most efficient
number of firms is one due to high fixed costs, economies of scale and barriers to entry. For PWSs, there
are high fixed costs associated with reservoirs and wells, transmission and distribution systems,
treatment plants, and other facilities. For other potential suppliers to enter the market, they would need
to provide the same extensive infrastructure to realize similar economies of scale and be competitive. A
splitting of the market with increased fixed costs (e.g., two supplier networks in a single market) usually
makes this situation unprofitable. The result is a market suitable for a single supplier and hostile to
additional suppliers. In such natural monopolies, the monopoly will charge a price that exceeds the
marginal cost, earning monopoly profits. The monopolistic firm faces fewer incentives to provide quality
services than if operating in a competitive market. In the case of the drinking water market a problem
with asymmetric/incomplete information also exists where drinking water systems are not incentivized
to provide high quality water to consumers because the customers know very little about the quality of
the water they are purchasing. In these situations, governments often intervene to help protect the
public interest by setting rates and ensuring the quality of the good or service. Consumers may purchase
bottled water, but this option can be much more expensive per unit than tap water of similar quality.
Consumers may also install and operate home treatment systems, but this can also be considerably
more expensive because they do not have the economies of scale of large, centralized water systems
and home treatment systems potentially can lead to increased health risks when not regularly
maintained by the consumer.

The public may not understand the health and safety issues associated with poor drinking water quality,
resulting in the existence of inadequate or asymmetric information. Understanding the health risks
posed by trace quantities of drinking water contaminants involves analysis and synthesis of complex
toxicological and health sciences data. Therefore, the public may not be aware of the risks it faces and
therefore may not advocated for improved water quality. Monopolistic drinking water systems and
oversite bodies may be slower to react to water quality issues without an informed public advocating for
improvements. The EPA has implemented a Consumer Confidence Report (CCR) Rule (63 FR 44512;
USEPA, 1998) that makes water quality information more easily available to consumers. In addition, the
EPA promulgated revisions to this regulation on May 24, 2024 (89 FR 45980; USEPA, 2024b) in
accordance with America's Water Infrastructure Act (AWIA) of 2018 (United States, 2018) and to require
reporting of compliance monitoring data to the EPA. The revisions to the CCR improve the readability,
clarity, and understandability of CCRs as well as the accuracy of the information presented, improve risk
communication in CCRs, incorporate electronic delivery options, and provide supplemental information
regarding lead levels and control efforts. This rule requires CWSs that serve 10,000 or more people to
provide CCRs to customers biannually (twice per year) as opposed to annually. Consumers, however, still
need to analyze this information for its health risk implications. Furthermore, even if informed
consumers can engage systems in a dialogue about health issues, the transaction costs of such
interaction (measured in personal time and monetary outlays) present another significant impediment
to consumer expression of risk reduction preferences.

Final LCRI Economic Analysis

2-24

October 2024


-------
Several of the rule changes under the final LCRI specifically compensate for inadequate or asymmetric
information. For example, the final LCRI greatly expands the PE and outreach requirements to provide
consumer notice to individuals who participated in the compliance sampling pool with their lead and
copper test results as soon as practicable but no later than three business days, and educational
materials to those served by service lines with known or possible lead content, and those potentially
impacted by disturbances to a known or potential service line containing lead. The requirements also
extend beyond the customer base to provide State and local health agencies with PE materials and
require greater public accessibility to information on lead-related information, such as LSL locations and
lead tap sample results. The more robust PE will provide consumers will more timely and useful
information to make more informed decisions and subsequently reduce their exposure to lead.

Overall, the SDWA regulations are intended to provide health protection from exposure to drinking
water contaminants. The regulations set minimum safety standards to protect consumers from
exposure to contaminants in drinking water supplies. The SDWA regulations are not intended to
restructure market mechanisms or establish competition in supply; rather, they establish the level of
service to be provided that best reflects public preference for safety. Federal regulations reduce the high
information and transaction costs by acting on behalf of consumers in balancing risk reduction and the
social costs of achieving this risk reduction.

2.7 References

Association of State Drinking Water Administrators (ASDWA). 2018 Costs of States' Transactions Study
(CoSTS) for Potential Long-Term Revisions to the Lead and Copper Rule (LT-LCR). April 2018.

ASDWA. 2020. Costs of States Transactions Study (CoSTS) for EPA's Proposed LCRR. February 6, 2020.

American Water Works Association v. EPA, 40 F.3d 1266, 1270-71 (D.C. Cir. 1994).

Community Fire Safety Act of 2013. Public Law 113-64. 113th Congress.
https://www.congress.gov/113/plaws/publ64/PLAW-113publ64.pdf.

Executive Order 13990. Executive Order on Protecting Public Health and the Environment and Restoring
Science to Tackle the Climate Crisis. January 20, 2021. https://www.whitehouse.gov/briefing-
room/presidential-actions/2021/01/20/executive-order-protecting-public-health-and-environment-and-
restoring-science-to-tackle-climate-crisis/.

Executive Order 13175. 2000. Consultation and Coordination with Indian Tribal Governments. Federal
Register. 65 FR 67249, November 9, 2000. Available at https://www.gpo.gov/fdsys/pkg/FR-2000-ll-
09/pdf/00-29003.pdf

Executive Order 13132. 1999. Federalism. Federal Register. 64 FR 43255, August 10, 1999. Available at
https://www.gpo.gov/fdsvs/pkg/FR-1999-08-10/pdf/99-20729.pdf.

Executive Order 12866. 1993. Regulatory Planning and Review. Federal Register 58 FR 51735, October 4,
1993. Available at https://www.reginfo.gov/public/isp/Utilities/EO 12866.pdf.

Lead Contamination Control Act of 1988. Public Law 100-572. 100th Congress.
https://www.govinfo.gov/content/pkg/STATUTE-102/pdf/STATUTE-102-Pg2884.pdf.

Final LCRI Economic Analysis

2-25

October 2024


-------
National Drinking Water Advisory Council (NDWAC). 2011. December 23, 2011: NDWAC Letter to EPA.
https://www.epa.gov/sites/procluction/files/2015-10/clocuments/2nclwaclettertoepaclec2011 O.pdf.

NDWAC. 2013. National Drinking Water Advisory Council Meeting Summary, December 11-12, 2013.
Prepared for EPA Office of Ground Water and Drinking Water. Available at

https://www.epa.gov/sites/production/files/2015-10/documents/2ndwacmeetingsummdecl22013.pdf.

NDWAC. 2015a. Report of the Lead and Copper Rule Working Group to the National Drinking Water
Advisory Council. August 24, 2015. https://www.epa.gov/sites/production/files/2016-
01/documents/ndwaclcrwgfinalreportaug2015.pdf.

NDWAC. 2015b. Recommendations to the Administrator for the Long Term Revisions to the Lead and
Copper Rule (LCR). December 15, 2015. https://www.epa.gov/sites/production/files/2016-
01/documents/ndwacrecommtoadminl21515.pdf.

NDWAC. 2024. National Drinking Water Advisory Council Meeting Summary, January 31, 2024. Retrieved
from https://www.epa.gov/svstem/files/documents/2024-06/ndwac-meeting-summarv-ianuarv-2Q24-
508 l.pdf

National Sanitation Foundation (NSF). 2019. NSF/ANSI 61-2019: Drinking Water System Components -
Health Effects. Ann Arbor, Michigan: NSF International, https://www.techstreet.com/nsf/standards/nsf-
61-2019?product id=2086734.

Parents for Nontoxic Alternatives. 2015. Memorandum from Yanna Lambrinidou, President., to the EPA
National Drinking Water Advisory Council (NDWAC). Long-term revisions for the Lead and Copper Rule
(LCR). October 28, 2015.

Reduction of Lead in Drinking Water Act. Public Law 111-380. IIIth Congress.
https://www.congress.gov/lll/plaws/publ380/PLAW-lllpubl380.pdf.

United States. 2016. Water Infrastructure Improvements for the Nation Act. 2016. Public Law 114-322,
130 Stat. 1628 (Dec. 16, 2016).

United States. 2018. America's Water Infrastructure Act. Public Law 115-270, 132 Stat. 3765.

United States Environmental Protection Agency (USEPA). 1976. National Interim Primary Drinking Water
Regulations. EPA-570/9-76-003.

USEPA. 1987. Amendments to the Safe Drinking Water Act. Federal Register. 52 FR 20674. June 2, 1987.
Washington, D.C.: Government Printing Office.

USEPA. 1991. Drinking Water Regulations; Maximum Contaminant Level Goals and National Primary
Drinking Water Regulations for Lead and Copper; Final Rule. Federal Register. 56 FR 26460. June 7, 1991.
Washington, D.C.: Government Printing Office.

USEPA. 1997. Interpretation of New Drinking Water Requirements Relating to Lead Free Plumbing
Fittings and Fixtures. Federal Register. 62 FR 44607. August 22, 1997. Washington, D.C.: Government
Printing Office, https://www.ftc.gov/sites/default/files/documents/federal register notices/extension-
time-guides-watch-industry-16-cfr-part-245/970822watchindustry.pdf.

Final LCR! Economic Analysis

2-26

October 2024


-------
USEPA. 1998. National Primary Drinking Water Regulations: Consumer Confidence Reports; Final Rule.
Federal Register. 63 FR 44512. August 19, 1998. Washington, D.C.: Government Printing Office.

USEPA. 2000. National Primary Drinking Water Regulations for Lead and Copper. Federal Register. 65 FR
1950. January 12, 2000. https://www.govinfo.gov/content/pkg/FR-2000-01-12/pdf/0Q-3.pdf.

USEPA. 2005. Drinking Water Lead Reduction Plan - EPA Activities to Improve Implementation of the
Lead and Copper Rule. March 2005. EPA 810-F-05-001.
https://nepis.epa.gov/Exe/ZvPDF.cgi?Dockev=P10051WL.txt.

USEPA. 2007. National Primary Drinking Water Regulations for Lead and Copper: Short-Term Regulatory
Revisions and Clarifications; Final Rule. Federal Register. 72 FR 57782. October 10, 2007. Washington,
D.C.: Government Printing Office.

USEPA. 2011a. Science Advisory Board (SAB) Evaluation of the Effectiveness of Partial Lead Service Line
Replacements. September 2011. Science Advisory Board. EPA-SAB-11-015.

https://www.epa.gov/sdwa/science-advisorv-board-evaluation-effectiveness-partial-lead-service-line-
replacements.

USEPA. 2011b. EPA Policy on Consultation and Coordination with Indian Tribes. May 2011.

https://www.epa.gov/sites/production/files/2013-Q8/documents/cons-and-coord-with-indian-tribes-

policy.pdf.

USEPA. 2016a. Lead and Copper Rule Revisions White Paper. October 2016. Office of Water.

https://www.epa.gov/sites/production/files/2016-

10/documents/508 Icr revisions white paper final 10.26.16.pdf.

USEPA. 2016b. Drinking Water Action Plan. November 2016. Office of Water.
https://19january2017snapshot.epa.gov/sites/production/files/2016-
ll/documents/508.final_. usepa_.drinking.water_.action.plan_11.30.16.v0.pdf.

USEPA. 2018. 3Tsfor Reducing Lead in Drinking Water in Schools and Child Care Facilities: A Training,
Testing, and Taking Action Approach (Revised Manual). October 2018. Office of Water. EPA815-B-18-
007. https://www.epa.gov/ground-water-and-drinking-water/3ts-reducing-lead-drinking-water-toolkit.

USEPA. 2019. Economic Analysis for the Proposed Lead and Copper Rule Revisions. October 2019. Office
of Water.

USEPA. 2020a. Economic Analysis for the Final Lead and Copper Rule Revisions. December 2020. Office of
Water. EPA 816-R-20-008.

USEPA. 2020b. Use of Lead Free Pipes, Fittings, Fixtures, Solder, and Flux for Drinking Water; Final Rule.
Federal Register. 85 FR 54235. September 1, 2020. https://www.govinfo.gov/content/pkg/FR-2020-Q9-
01/pdf/2020-16869.pdf.

USEPA. 2020c. Science Advisory Board (SAB) Consideration of the Scientific and Technical Basis of EPA's
Proposed Rule Titled National Primary Drinking Water Regulations: Proposed Lead and Copper Rule
Revisions. June 2020. Science Advisory Board. EPA-SAB-20-007.

USEPA. 2020d. Public Comment and Response Document for the Final Lead and Copper Rule Revisions.

Final LCRI Economic Analysis

2-27

October 2024


-------
USEPA. 2021a. Review of the National Primary Drinking Water Regulation: Lead and Copper Rule
Revisions (LCRR). Federal Register. 86 FR71574. December 17, 2021.

https://www.federalregister.gov/documents/2021/12/17/2021-27457/review-of-the-national-primary-
drinking-water-regulation-lead-and-copper-rule-revisions-lcrr.

USEPA. 2021b. National Primary Drinking Water Regulations: Lead and Copper Rule Revisions. Final Rule.
Federal Register. 86 FR4198. January 15, 2021. Washington, D.C.: Government Printing Office.

USEPA. 2021c. National Primary Drinking Water Regulations: Lead and Copper Rule Revisions; Delay of
Effective Date. Federal Register. 86 FR 14003. March 12, 2021. Washington, D.C.: Government Printing
Office.

USEPA. 2021d. National Primary Drinking Water Regulations: Lead and Copper Rule Revisions; Delay of
Effective and Compliance Dates. Federal Register. 86 FR 14063. March 12, 2021. Washington, D.C.:
Government Printing Office.

USEPA. 2021e. National Primary Drinking Water Regulations: Lead and Copper Rule Revisions; Delay of
Effective and Compliance Dates. Final Rule. Federal Register. 86 FR 31939. June 16, 2021. Washington,
D.C.: Government Printing Office.

USEPA. 2022. Draft Case Studies to Inform EPA's Environmental Justice Analysis for the Proposed Lead
and Copper Rule Improvements. Office of Water. EPA 816-D-22-001. October 2022.
https://sab.epa.gov/ords/sab/f?p=114:19:7221208161736:::RP.19:P19 ID:978#draft.

USEPA. 2023a. System responses to questionnaire on time needed to develop and maintain a service
line inventory under the LCRR. Responses received from Grand Rapids, Ml; Pittsburgh, PA; and
Cincinnati, OH.

USEPA. 2023b. National Primary Drinking Water Regulations: Lead and Copper Rule Improvements.
Proposed Rule. Federal Register 88(233): 84878. December 6, 2023. Washington, D.C.: Government
Printing Office.

USEPA. 2024a. Response to Public Comments on the Lead and Copper Rule Improvements. EPA 815-R-24-
029. October 2024

USEPA. 2024b. National Primary Drinking Water Regulations: Consumer Confidence Report. Final Rule.
Federal Register. 89 FR 45980. May 24, 2024. Washington, D.C.: Government Printing Office.

USEPA Science Advisory Board. 2022. Consultation on Environmental Justice Analysis for EPA's Lead and
Copper Rule Improvements. From Alison CCullen, Sc. D Chair to EPA Administrator Michael S. Regan.
EPA-SAB-23_003. December 20, 2022.

https://sab.epa.gOv/ords/sab/r/sab apex/sab/advisorvactivitvdetail?pl8 id=2628&clear=RP.18&sessio
n=11133043673738#reporthttps://sab.epa.gov/ords/sab/r/sab apex/sab/advisorvactivitydetail?pl8 id
=2628&clear=RP.18&session=11133043673738#report

United States Office of Management and Budget (USOMB). 2003. Circular A-4: Regulatory Analysis.
Circular. Washington, D.C.: Government Printing Office.

Final LCRI Economic Analysis

2-28

October 2024


-------
3 Baseline Drinking Water System Characteristics

3.1	Introduction

In its Guidelines for Preparing Economic Analyses, the United States Environmental Protection Agency
(EPA, or USEPA) characterizes the "baseline" as a reference point that reflects the world without the
final regulation (USEPA, 2014). It is the starting point for estimating the potential benefits and costs. This
chapter presents a characterization of public water systems (PWSs) and their current operations (i.e.,
the baseline) before changes are made to meet the regulatory requirements for the 2021 Lead and
Copper Rule Revisions (LCRR) or the final Lead and Copper Rule Improvements (LCRI). Section 3.2
identifies each major source used to develop the baseline. Section 3.3 explains the derivation of each
baseline characteristic and presents results in detailed tables. Section 3.4 summarizes limitations of the
major data sources and uncertainties in the baseline characterization (both quantified and unquantified)
in tabular format.

Note that the EPA uses the SafeWater Lead and Copper Rule (LCR) model to estimate national costs of
the 2021 LCRR and the final LCRI. The estimated national cost of the 2021 LCRR are then subtracted
from the final LCRI cost estimates to compute the incremental estimated cost of the final LCRI. See
Chapter 4, section 4.2 for an in-depth discussion of the SafeWater LCR model. See Appendix B, Sections
B.5 and B.6 for a discussion of the data variables and the estimated burden and costs associated with
the implementation of the 2021 LCRR. Also, review the Chapter 4, Sections 4.3 and 4.4 data variable
descriptions and estimated burden and costs for the final LCRI.

3.2	Data Sources

The EPA used a variety of data sources to develop the baseline. Additional background on each of these
data sources is provided in the following subsections:

•	Section 3.2.1 explains the relevant information provided in the federal version of the Safe
Drinking Water Information System (SDWIS/Fed) and measures the EPA has taken to verify the
data.

•	Section 3.2.2 explains the purpose of the 2006 Community Water System Survey (CWSS) and
the representativeness of the data.

•	Section 3.2.3 explains the relevant information that was used from the third Unregulated
Contaminant Monitoring Rule (UCMR 3).

•	Section 3.2.4 describes a key information source used to characterize corrosion control
treatment (CCT) costs.

•	Section 3.2.5 describes the 7th Drinking Water Infrastructure Needs Survey and Assessment
(DWINSA) data that were used to develop the EPA's characterization of service line material,
identify individual systems with lead content service lines, and estimate service line replacement
costs.

Final LCRI Economic Analysis

3-1

October 2024


-------
•	Section 3.2.6 provides an overview of the system compliance monitoring data voluntarily
submitted to the EPA by States from 2006 to 2011, data cleaning steps, and data
representativeness.

•	Section 3.2.7 describes the State of Michigan lead tap monitoring dataset that included first -
and fifth-liter compliance monitoring samples collected in 2019, 2020, and 2021, as well as data
cleaning steps.

•	Section 3.2.8 describes other information sources used to characterize a subset of the
population served by CWSs that provide services to sensitive subpopulations (i.e., infants,
children, and pregnant women). Note that the EPA used several studies to characterize sensitive
subpopulations affected by the rule. These studies are discussed in Chapter 5.

Exhibit 3-1 identifies each major data source detailed in Sections 3.2.1 through 3.2.7 and the baseline
data element(s) derived from them. Data sources used for pre-2021 LCR and 2021 LCRR are provided in
Appendix B.

Exhibit 3-1: Data Sources Used to Develop the Baseline for the Final LCRI

Data Source

Baseline Data Derived from the Source



• PWS inventory, including population served, number of service connections,



source water type, and water system type. Also used to identify NTNCWSs that



are schools and child care facilities.



• Status of CCT, including identification of water systems with CCT and the



proportion of water systems serving < 50,000 people that installed CCT in



response to the pre-2021 LCR.



• Analysis of lead 90th percentile concentrations to identify water systems

SDWIS/Fed fourth quarter
2020 "frozen" dataset1

below, at, or above the lead and/or copper ALs at the start of rule

implementation by LSL status, i.e., presence or absence of LSLs for the pre-

2021 LCR, 2021 LCRR, and final LCRI. Used in concert with data from Michigan
described below for the final LCRI.2

•	The proportion of water systems that are on various reduced monitoring
schedules for lead tap and WQP monitoring.

•	The frequency of source and treatment changes and those source changes that
can result in additional source water monitoring.

•	Number of distribution system entry points per drinking water system for
systems that were not included in the UCMR 3 dataset.

2006 CWSS (USEPA, 2009)

• PWS labor rates.

UCMR 3 (2013-2015)

• Number of distribution system entry points per drinking water system.

7th DWINSA and

• Service line material characterization.

Supplemental One-time

• Service line replacement costs.

Update



State service line

• Service line material characterization.

information



Geometries and

• Design and average daily flow per system.

Characteristics of Public



Water Systems (USEPA,



2000)



Final LCRI Economic Analysis

3-2

October 2024


-------
Data Source

Baseline Data Derived from the Source

Six-Year Review 3 ICR
Occurrence Dataset (2006-
2011)

•	Baseline distribution of pH for various CCT conditions.

•	Baseline orthophosphate dose for CCT.

State of Michigan Lead and
Copper Compliance
Monitoring Data (Michigan
EGLE, 2019-2021)

•	Analysis of the ratio of fifth- to first-liter lead tap samples to estimate the
increase in lead 90th percentile levels for LSL systems based on the use of the
higher of the first- or fifth-liter sample result. Ratios are applied to SDWIS/Fed
lead 90th percentile data to identify systems below, at, or above the AL under
the final LCRI by LSL status.

•	Percent of individual samples exceeding 0.010 mg/L for the final LCRI.

Acronyms: AL = action level; CCT = corrosion control treatment; CWSS = Community Water System Survey;

DWINSA = Drinking Water Infrastructure Needs Survey and Assessment; ICR = Information Collection Request; LCR
= Lead and Copper Rule; LCRR = Lead and Copper Rule Revisions; LCRI = Lead and Copper Rule Improvements; LSL =
lead service line; Michigan EGLE = Michigan Department of Environment, Great Lakes, and Energy; NTNCWS = non-
transient non-community water system; public water system; SDWIS/Fed = Safe Drinking Water Information
System/Federal version; UCMR 3 = Third Unregulated Contaminant Monitoring Rule; USEPA = United States
Environmental Protection Agency; WQP = water quality parameter.

Note:

1	Contains information reported through December 31, 2020.

2	A system's lead 90th percentile level is a key factor in determining a system's requirements under the pre-2021
LCR, 2021 LCRR, and final LCRI.

3.2.1 SDWIS/Fed 2020

SDWIS/Fed is the EPA's national regulatory compliance database for the drinking water program. It
contains water system inventory, 90th percentile lead and copper levels, treatment facility information,
violation, and enforcement information for PWSs as reported by States, the EPA Regions, and the EPA
Headquarters personnel. States report data quarterly to the EPA. The information presented in the
economic analysis (EA) is based on the fourth quarter 2020 "frozen" dataset that contains information
reported through December 31, 2020.15

SDWIS/Fed contains information to characterize the United States inventory of PWSs, namely: system
name and location; retail population served; source water type (i.e., ground water (GW), surface water
(SW), or ground water under the direct influence of surface water (GWUDI)); and PWS type, as
described in Section 3.2.1.1. SDWIS/Fed also includes 90th percentile lead and copper levels, milestones,
violations, and enforcement actions, as detailed in Section 3.2.1.2. A description of the treatment facility
information in SDWIS/Fed is in Section 3.2.1.3. Section 3.2.1.4 summarizes steps by the EPA to verify
SDWIS/Fed information.

3.2.1.1 Classification of Systems Using SDWIS/Fed Data

This section describes how the EPA classified systems by type (Section 3.2.1.1.1), population served
(Section 3.2.1.1.2, and source water (Section 3.2.1.1.3) using data from SDWIS/Fed.

3.2.1.1.1 System Type

The Safe Drinking Water Act (SDWA) defines a system as one that provides water for human
consumption through pipes or other constructed conveyances to at least 15 service connections or

15 This dataset represented the most current full year of data at the time the EPA started the proposed LCRI EA.

Final LCRI Economic Analysis

3-3

October 2024


-------
regularly serves an average of at least 25 individuals per day for at least 60 days per year. Systems are
categorized as follows:

•	Community water systems (CWSs) are systems that supply water to the same population year-
round.

•	Non-community water systems (NCWSs) are systems that supply water to a varying population
or one that is served less than year-round. They are sub-categorized as follows:

o Non-transient non-community water systems (NTNCWSs) are systems that are not CWSs
and that regularly supply water to at least 25 of the same people at least six months per
year, for example, schools.

o Transient non-community water systems (TNCWSs) are NCWSs that provide water in
places such as gas stations or seasonal campgrounds where people do not remain for
long periods of time.

The final LCRI would not apply to TNCWSs. Therefore, system inventories in this EA are classified into
two categories, CWSs and NTNCWSs.

3.2.1.1.2 Population Served

Systems are also categorized by the number of people they serve.16 The following nine categories of
populations served by systems are used throughout this EA:

< 100

101-500

501-1,000

1,001-3,300

3,301-10,000

10,001-50,000

50,001-100,000

100,001-1,000,000 (1M)

> 1M

The EPA has developed these system size categories based on distinctions in the way systems operate as
the amount of water supplied and number of service connections increases. Systems within each size
category can be expected to face similar implementation and cost challenges when complying with the
regulatory requirements for the final LCRI.

16 SDWIS/Fed classifies systems according to "retail" population that does not include the population served by
other systems that purchase water from them.

Final LCRI Economic Analysis

3-4

October 2024


-------
3.2.1.1.3 Source Water Type
SDWIS/Fed classifies systems by source water using the following six categories:

1.	GW

2.	GW purchased

3.	GWUDI17

4.	GWUDI purchased

5.	SW

6.	SW purchased

For this final LCRI analysis, the EPA broadly categorized systems as SW if any of their sources were SW,
SW purchased, GWUDI, or purchased GWUDI. Source water type is important in estimating treatment
costs for the rule due to the fact that GW systems typically have more entry points and thus, more
treatment and water quality parameter (WQP) monitoring locations, than SW systems. Systems were
classified as GW if they exclusively used GW or purchased GW. See Section 3.3.1 for the EPA's approach
for assigning a source type to the small number of CWSs and NTNCWSs without a reported source water
type to develop the system inventory for this EA.

3.2.1.2 Lead and Copper Rule-Specific Data

This section describes specific data that States must report to the EPA using SDWIS/Fed under the pre-
2021 LCR and reflects the requirements prior to the implementation of the final LCRI. It is organized into
the following subsections:

•	3.2.1.2.1: 90th Percentile Levels

•	3.2.1.2.2: Violations/Compliance Achieved

•	3.2.1.2.3: Milestones.

3.2.1.2.1 90th Percentile Levels

Under the pre-2021 LCR, systems are required to report all lead and copper tap sample results used to
calculate their lead and copper 90th percentile levels to their State. States are required to report to
SDWIS/Fed all lead 90th percentile values in mg/L for systems serving more than 3,300 people18 and 90th
percentile values in mg/L above the action level (AL) of 0.015 mg/L19 for systems serving 3,300 or fewer.

17	40 CFR section 141.2 defines GWUDI as "any water beneath the surface of the ground with significant
occurrence of insects or other macro-organisms, algae, or large-diameter pathogens such as Giardia lamblia or
Cryptosporidium, or significant and relatively rapid shifts in water characteristics such as turbidity, temperature,
conductivity, or pH which closely correlate to climatological or surface water conditions."

18	Prior to 2002, States were not required to report lead 90th percentile levels that were at or below the lead AL for
systems serving 3,300 or fewer people.

19	As discussed throughout the economic analysis, the AL under the final LCRI has been lowered to 0.010 mg/L.

Final LCRI Economic Analysis

3-5

October 2024


-------
For all systems, States are also required to report to SDWIS/Fed copper 90th percentile levels above the
AL of 1.3 mg/L.

Under the pre-2021 LCR, a system has an action level exceedance (ALE) if more than 10 percent of tap
water samples collected during any monitoring period are found to be greater than 0.015 mg/L for lead
or 1.3 mg/L for copper (i.e., if the 90th percentile level is greater than the AL). An ALE is not a violation
but triggers additional actions. These actions include CCT steps, WQP monitoring, source water
monitoring and source water treatment, if needed. A lead ALE also triggers lead service line replacement
(LSLR) for systems with treatment in place for lead and public education for all systems with a lead ALE.

3.2.1.2.2	Violations/Compliance Achieved

Systems are in violation of the pre-2021 LCR if they do not meet the treatment technique requirements
related to LSLR, CCT, source water treatment, public education (PE), or monitoring and reporting
requirements. States are required to report to SDWIS/Fed, systems that are in violation of these
requirements using specific codes that identify the type of violation and the action taken by the system
or State to address these violations. As explained in Section 3.3.3, the EPA used the following subset of
violations to estimate the number of systems with CCT20:

•	Violation code 58 denotes systems that failed to meet their CCT requirements. This includes
failure to properly install or operate State-approved CCT, submit a certification that CCT is being
properly installed and operated, or to demonstrate that optimal corrosion control treatment
(OCCT) already exists in accordance with 40 CFR 141.81(b)(l)-(3) and 141.90(c)(1).21

•	Violation code 59 denotes systems that fail to meet the optimal water quality parameter
(OWQP) values set by the State. OWQPs are set by the State after a system has collected WQP
samples during two consecutive, six-month monitoring periods, following the installation of CCT.
OWQPs are measured to determine whether a system is operating its CCT at a level that most
effectively minimizes the lead and copper concentrations at users' taps.

States are also required to report enforcement actions taken by the State or the EPA in response to a
violation, and to report when a system has achieved compliance. As discussed in Section 3.3.3, the EPA
enforcement action code for compliance achieved is "SOX" or "EOX." Systems that have returned to
compliance with a type 58 violation are likely to have installed CCT.

See "Safe Drinking Water Information System Federal (SDWIS Fed) Data Reporting Requirements" for
additional information on SDWIS/Fed reporting requirements (USEPA, 2016a).

3.2.1.2.3	Milestones

States report milestone information to indicate the initiation or completion of key requirements under
the pre-2021 LCR. The EPA used the following milestones data to characterize the baseline. Specifically,
the EPA used the "Deem" and "Done" milestones to help estimate the number of systems with CCT (see

20	Each violation has a specific code. A violation is reported in SDWIS/Fed using the specific violation code (e.g., 58
or 59) vs. Y or N. States do not report if the system has no violation, only if there is a violation.

21	Code 58 is also used to identify water systems that are in violation of the source water treatment installation
requirements. However, very few water systems have high lead and/or copper source water levels and are
required to install source water treatment.

Final LCR! Economic Analysis

3-6

October 2024


-------
Section 3.3.3) and LSLR milestone to estimate the average number of years a system is required to
replace lead service lines (LSLs) under the pre-2021 LCR.

•	"Deem" represents the basis for the State's determination that a system is "deemed" to be
optimized under the LCR. Systems with a reason code of "WQP" have installed CCT.

•	"Done" indicates when a water system has completed all required steps to reduce lead and/or
copper levels. Systems with a reason code of "WQP" have installed CCT.

•	"LSLR" indicates water systems that are required to initiate LSLR; States are also required to
report when this replacement is scheduled to begin.

3.2.1.3	Treatment Facility Information

States report treatment information to SDWIS/Fed for each system's drinking water treatment facilities.
Specifically, for each treatment plant, States report 1) the treatment objective codes from a list of 13
available options and 2) treatment process codes from a list of 71 available options. For example, the
treatment objective code of "C" denotes corrosion control and the treatment process code of 445
indicates orthophosphate inhibitor. States can report multiple treatment objective codes for each plant,
and multiple treatment process codes for each plant or for each objective code.

The EPA uses treatment code information to help determine the percent of systems with CCT and which
type of CCT they have in place (pH adjustment, orthophosphate, or both) as described in Section 3.3.3
and Section 4.3.3 in Chapter 4, respectively. The EPA also uses treatment information from SDWIS/Fed
to evaluate changes in treatment over time to predict the percent of systems that would change
treatment each year, as described in Section 3.3.9.

3.2.1.4	Verification of SDWIS/Fed Data

The EPA routinely conducts Program Reviews to verify whether information in the States' databases and
files, such as inventory, 90th percentile data, and violations for all regulations are correctly represented
in SDWIS/Fed. Between 2006 and 2016, the EPA recorded the findings from these reviews in the
national Error Code Tracking Tool (ECTT) (USEPA, 2007).22 The ECTT contains, as individual records, all
actions assessed during each Program Review. The EPA identifies records as confirmed actions (correct
compliance determinations and correct reporting to SDWIS/Fed), compliance determination
discrepancies (incorrect compliance determinations), or data flow discrepancies (correct compliance
determination but incorrect reporting). This section presents data from the ECTT from Program Reviews
conducted from 2006 to 2016 related to water system inventory (Section 3.2.1.4.1) and LCR compliance
data (Section 3.2.1.4.2).

It is important to note that treatment data (objective codes and process codes for plants in SDWIS/Fed)
are not evaluated during Program Reviews and therefore have more uncertainty associated with the
data as compared to water system inventory and compliance data.

22 More recent data were not available for use in this analysis as the EPA no longer used the ECTT after 2016.

Final LCR! Economic Analysis

3-7

October 2024


-------
3.2.1.4.1	Water System Inventory

From 2006 to 2016, the EPA evaluated water system inventory data for a total of 2,180 systems. Prior to
August 2007, the Program Reviews evaluated eight water system inventory fields: system type, system
status, activity status, source type, population, service connection, administrative contact, and
administrative address. Afterwards, the reviews did not include administrative contact or address. In
addition, in August 2007, the review policy changed so that discrepancies for water system inventory
were only identified if they affected monitoring requirements (e.g., change in population that would
increase or decrease the minimum number of required samples).

Of the water system inventory fields evaluated from 2006 to 2016, only 82 (<1 percent) inventory
discrepancies were identified. Some of these discrepancies could be for things that do not impact the
PWS baseline characterization such as administrative contact and address. The water system inventory
data in ECTT indicate a high degree of completeness and accuracy in SDWIS/Fed.

3.2.1.4.2	LCR Compliance MonitorinR Data

To assess the completeness and accuracy of the SDWIS/Fed LCR compliance monitoring data, which is
reported to the State as the 90th percentile of tap monitoring results, the EPA determined whether
States had reported the following:

•	The correct 90th percentile levels to SDWIS/Fed by comparing it to the computed 90th percentile
levels from the individual monitoring results submitted by systems.23

•	All required 90th percentile levels.

File reviews conducted between 2006 and 2016 evaluated 2,180 systems for two rounds of lead
sampling and evaluated 4,360 rounds of lead samples for 53 primacy agencies. Of these data, the 90th
percentile level sample values were properly calculated and reported to SDWIS/State for 4,212 (87
percent) of the sample rounds. The file review also evaluated whether the samples were properly
collected, including a sufficient number of samples, correct sampling procedure, collection during the
correct monitoring period. The review determined that systems complied with these additional
requirements for 87 percent of the sample rounds. The file reviews also determined that systems failed
to take the required steps after a lead ALE, including PE, CCT study (when required), WQP sampling, or
follow-up monitoring after installation of CCT in some instances.

3.2.2 2006 Community Water System Survey

The EPA periodically conducts the CWSS to obtain data to support the agency's development and
evaluation of drinking water regulations. The 2006 CWSS is the most recent survey (USEPA, 2009). For
this EA, the EPA used the 2006 CWSS to develop hourly labor rates by system size (see Section 3.3.11.1).
These rates are multiplied by the burden estimates in the SafeWater LCR cost model to develop labor
cost for water systems to comply with the requirements of the final LCRI. See Chapter 4 for additional
detail pertaining to the final LCRI and Appendix B for information related to the pre-2021 LCR and 2021
LCRR.

23 This evaluation also assessed whether the 90th percentile level was reported for the correct monitoring period.

Final LCRI Economic Analysis

3-8

October 2024


-------
3.2.3	Unregulated Contaminant Monitoring Rule 3

The EPA uses the UCMR to collect nationally representative data for contaminants that are suspected to
be present in drinking water and do not have health-based standards set under SDWA. Monitoring
under the third round of UCMR (UCMR 3) was conducted from 2013 through 2015.24 Similar in design to
the first two rounds of UCMR sampling (UCMR 1 and UCMR 2), UCMR 3 required SW systems to monitor
quarterly and GW systems to monitor semi-annually to capture seasonal variability. For UCMR 3, all
large and very large PWSs (serving between 10,001 and 100,000 people and serving more than 100,000
people, respectively), plus a statistically representative national sample of 800 small PWSs (serving
10,000 people or fewer), were required to conduct Assessment Monitoring during a 12-month period
between January 2013 and December 2015. For all UCMR 3 contaminants, systems were required to
gather samples at the entry point to the distribution system. As described in Section 3.3.6.1, the EPA
developed estimates of entry points per system using unique sampling point data from UCMR 3.

3.2.4	Geometries and Characteristics of Public Water Systems (2000)

An important factor in determining costs of CCT is average daily flow and design flow, in gallons per day
or million gallons per day, at a treatment plant. The EPA estimated the average daily flow and design
flow for each entry point in the system based on the relationship between retail population and flow as
derived in the document, Geometries and Characteristics of Public Water Systems (USEPA, 2000).25

Utilizing data from the 1995 CWSS, the EPA conducted an extensive data cleaning process26 to develop a
dataset consisting of 1,734 records with paired responses for population and total average daily flow.
These data were then weighted to account for non-responses to individual questions from the CWSS.
This dataset was used to develop regression equations that predict average daily flow based on retail
population served (for both publicly-owned and privately-owned systems). The data show a very good
correlation as indicated by a high R value of 0.90. Additional information and background data are
provided in Chapter 4 of the Geometries and Characteristics of Public Water Systems (USEPA, 2000) and
in Section Bl.4.2 of the Drinking Water Baseline Handbook, Fourth Edition (USEPA, 2003). Note that
household water use has generally declined over the period since this analysis was completed and
therefore the EPA's estimated national costs for CCT are likely overestimated. For additional information
see Section 3.4.

3.2.5	7th Drinking Water Infrastructure Needs Survey and Assessment (DWINSA)

Every four years, the EPA works with States and CWSs to conduct the Drinking Water Infrastructure
Needs Survey and Assessment (DWINSA) to estimate the Drinking Water State Revolving Fund (DWSRF)-
eligible needs of systems by State. Through this survey, systems submit DWSRF-eligible infrastructure
projects that are necessary over the next 20 years to continue to provide safe drinking water to the
public. These projects include infrastructure needs that are eligible for, but not necessarily financed by,

24	See USEPA (2012a) and USEPA (2019) for more information on the UCMR 3 study design and data analysis,
including a complete list of analytes.

25	The analysis was republished in the Drinking Water Baseline Handbook, Fourth Edition (USEPA, 2003).

26	EPA adjusted the dataset to remove non-zero values; adjusted flow if needed to represent retail flow only
removing wholesale water flow; and adjusted for reporting discrepancies in population, flow, or service
connections.

Final LCRI Economic Analysis

3-9

October 2024


-------
the DWSRF, including the installation of new drinking water infrastructure and the rehabilitation,
expansion, or replacement of existing infrastructure.

The EPA's 7th DWINSA consisted of a survey of all systems in the country serving more than 100,000
people, a per-State sample of systems serving between 3,301 and 100,000 people, and a national
sample of systems serving 3,300 or fewer people (USEPA, 2023a).27 The surveyed systems for the 7th
DWINSA included CWSs and not-for-profit non-community water systems (NPNCWS) in States, U.S.
Territories, and American Indian (Al) and Alaska Native Village (ANV) water systems. The assessment
selected a stratified random sample of systems, dividing systems into mutually exclusive categories
based on the systems' water source and the number of people served.28 The EPA administered the 7th
DWINSA to a total of 3,629 water systems and received 3,526 responses. This large number of
participants and 97 percent response rate provides a high degree of confidence in the statistical
precision of the assessment's findings.

As part of the 7th DWINSA, the EPA collected service line material information for the first time in 2021.
The same 3,629 water systems participating in the primary DWINSA were surveyed using the 7th
DWINSA service line questionnaire, which collected information on the number of service lines by
material type. The service line questionnaire was optional; however, approximately 80 percent of water
systems provided complete responses about their service lines.29 This dataset included CWSs from all
States plus D.C., Puerto Rico, Guam, American Samoa, the Northern Mariana Islands, and the U.S. Virgin
Islands. In this questionnaire form, systems were also asked to group their service line inventory into
eight categories: lead pipe, lead connectors, galvanized service lines downstream from a lead pipe,
galvanized service lines downstream from a lead connector, galvanized service lines downstream from
an unknown lead source, standalone galvanized service lines, or non-lead/non-galvanized service lines.30
The inventories of systems that did not respond to the written survey were classified as "unreported."

27	The 7th DWINSA used the list of systems from the second quarter of 2019 freeze of SDWIS/Fed. It used the retail
and consecutive population to determine each system's population. This population was reviewed by the states
and revised when necessary. The final population of each system in the survey is based on the systems' survey
responses.

28	The 7th DWINSA used a sample of water systems to estimate total infrastructure needs and the number of lead
service lines in each state and in the nation. Within each State, water systems were divided into several categories
based on each system's water source and the size of the population served. The 7th DWINSA included all systems
serving more than 100,000 people and a random sample of systems serving between 3,301-100,000 people from
each category of systems in each State, as well as a random sample of systems serving 3,000 of fewer people from
each category of systems nationally. The number of systems in each State and each system's population are based
on information in SDWIS/Fed as of the second quarter of 2019, as reviewed and revised by the States. To estimate
state totals for medium and large systems using the sample, each system was assigned a weight equal to the
number of systems in the category divided by the number of systems sampled from that category. For example, if
the survey included a sample of three systems from a category that consists of 12 systems, each of the three
systems from that category would receive a weight of 4 (12 -f 3 = 4). The final sampling weights are adjusted to
account for non-response.

29	A modified version of the survey was provided to American Indian and Alaska Native Village CWSs, but the
responses were not included in the totals.

30	Note survey information does not provide specific detail on service line lengths being replaced. EPA cost
estimates assume the distribution of DWINSA data line lengths is equal to the national distribution. Note although
length of service line being replaced is positively correlated with cost, the larger cost drivers in SLR are associated with

Final LCRI Economic Analysis

3-10

October 2024


-------
In August 2023, the EPA initiated a one-time effort to update the LSL counts from the 7th DWINSA
(USEPA, 2024a; 2024b). This update allowed previously surveyed water systems and States to revise
their original response based on new service line inventory information or to provide responses if they
had not participated earlier. Participation in this update, which was limited to the service line material
questionnaire, was voluntary.

Through this effort, the EPA surveyed 2,888 medium and large systems 50 States, the District of
Columbia, Puerto Rico and four territories, receiving 2,089 responses (a 72% response rate): 596
reporting updates and 1,493 indicating no changes. Additionally, the EPA surveyed 695 small systems,
receiving 132 responses (a 19% response rate): 64 reporting updates and 68 indicating no changes. Note
the low response rate by small systems to the one-time update is related to the use of trained site
visitors to assist in the collection of the small system data during the original survey. The EPA gave small
systems the opportunity to participate in the one-time update but did not expect many responses
because the site visitors provided the original data. Some systems took advantage of the opportunity
and provided updates, but most of the original data were accurate because site visitors assisted in the
collection of the data. The EPA then combined the data from the initial survey and the August 2023
update by both adding new respondent data from the update survey to the 7th DWINSA dataset and by
replacing the original 7th DWINSA information with new data when respondents updated information as
part of the August 2023 effort. Note that the sample weights were not recalculated, as the sample of
water systems for both the one-time update and the 7th DWINSA were identical.

The EPA used the results of the 7th DWINSA as cost model inputs to the LCRI EA as follows:

•	To characterize service line material. The EPA used the combined results of the original 7th
DWINSA and the one-time update to develop the service line characterization. For details on the
methodology and results, see Section 3.3.4.

•	To identify systems with known LSL status. For the purpose of analyzing 90th percentile lead
levels for systems with known LSLs and all non-lead service lines as described in Section 3.3.5.

•	To estimate unit costs of LSLR. The EPA reviewed LSL project costs submitted and accepted by
the 7th DWINSA and prepared a distribution of unit costs for full and partial replacements.
Appendix A provides a discussion of the methods the EPA used to select LSLR projects and the
final estimated replacement cost results.

3.2.6 Six-Year Review Data

The EPA used information from the third Six-Year Review Information Collection Request (ICR) Dataset
(hereafter referred to as the "SYR3 ICR dataset") to characterize the pH of finished water and the
distribution of orthophosphate dose. The SYR3 ICR dataset contains more than 47 million records of
water system compliance monitoring data for chemical, microbial, disinfection byproduct, and
radionuclides collected from 2006 through 2011. The SYR3 ICR dataset and general quality
assurance/quality control (QA/QC) procedures are further described in USEPA (2016b) and USEPA
(2016c).

the mobilization of crews, the method of replacement, depth of pipe, and the amount of restoration (concrete and
road repair) work required.

Final LCRI Economic Analysis

3-11

October 2024


-------
Forty-four States, D.C., American Samoa, and five EPA Regions submitted individual compliance
monitoring sample result for lead as part of the SYR3 ICR data set.31 A number of QA steps were applied
to the SYR3 ICR dataset to identify water quality data on system pH and orthophosphate concentration
records suitable for analyses. Data were excluded via the following QA steps:

•	Records from non-public water systems.

•	Records marked as not being for compliance.

•	Records marked with a sample type code equal to something other than "RT" (routine) or "CO"
(confirmation). For example, "RP" for "repeat" or "SP" for "special."

•	Records from outside of the SYR3 date range of 2006 - 2011.

•	Records from systems that were missing water system inventory information such as the
system's population served or source water type.

3.2.7 State of Michigan Lead Compliance Monitoring Data

The EPA evaluated lead and copper compliance monitoring data provided by the State of Michigan,
Department of Environment, Great Lakes, and Energy (EGLE) for January 1, 2019, through December 31,
2021. Michigan's State-level lead and copper regulations require systems to collect a first- and fifth-liter
sample at sites with an LSL (any portion of the service line containing lead) and to collect a first liter
sample only for all other sites (service lines made of galvanized, copper, or plastic pipe). Based on
Michigan's requirements, the EPA identified systems collecting both a first- and fifth-liter sample as a
system with LSLs and those collecting only a first-liter as a system without LSLs. SDWIS/Fed does not
indicate the LSL status of water systems (i.e., the presence or absence of LSLs). Thus, the EPA used the
subset of CWSs in Michigan with LSLs to adjust the 90th percentile value reported to SDWIS/Fed for non-
Michigan systems that the EPA identified as having LSLs (see 3.3.5.1.1 for EPA's approach for identifying
systems with LSLs). The EPA was then able to categorize these systems by LSL status into one of five lead
90th percentile classifications under the pre-2021 LCR and the 2021 LCRR (see Appendix B), and the final
LCRI (see Section 3.3.5.1).32 Thus, the EPA used the subset of CWSs in Michigan with known LSL status
and 90th percentile data reported to estimate the national percentage of systems by LSL status that
would be categorized into one of five lead 90th percentile classifications under the pre-2021 LCR, the
2021 LCRR, and the final LCRI (see Section 3.3.5.1). The EPA also used the Michigan date to estimate the
likelihood a single sample would exceed 0.015 mg/L under the 2021 LCRR (see Appendix B) and 0.010
mg/L under the final LCRI (see Section 3.3.5.3). The EPA recognizes the uncertainty introduced in using
data from a single State that may not represent the values on a national level. The Michigan data on
first- and fifth-liter sampling was the best available data to inform this analysis at the time the analysis
was conducted.

31	With the exception of the Navajo Nation, the EPA Regions are the primacy agencies for Tribal water systems.

32	The five lead 90th percentile classifications are: lead 90th percentile (P90) < 5 ng/L; 5 ng/L < P90 < 10 ng/L; 10
Hg/L < P90 < 12 ng/L; 12 ng/L < P90 < 15 ng/L; and P90 > 15 ng/L.

Final LCRI Economic Analysis

3-12

October 2024


-------
A total of 93,882 lead and copper sample results were submitted by the 1,373 CWSs in Michigan that
sampled during January 2019 through December 2021. The following QA steps were applied to the data
to identify records suitable for analyses:

•	Excluded records collected for copper.

•	Substituted 1 ng/Lfor all concentrations <1 ng/L in the original dataset.

•	Excluded data from 10 systems that were not included in the SDWIS/Fed 2020 fourth quarter
frozen dataset. Thus, the EPA did not have needed information for the EA, such as CCT status,
population served, and 90th percentile data.

•	Classified systems as non-lead service line systems if they collected only first-liter samples based
on 2020 and 2021 compliance samples and were listed in Michigan's preliminary distribution
system materials inventory (Michigan EGLE, 2020) with 0 values for "known lead" service line
materials, "unknown - likely lead" service line materials, and "unknown - No information"
service line materials.

•	Assumed systems that collected paired first- and fifth-liter samples had LSLs or if the system was
listed as having "known lead" based on Michigan's preliminary distribution system materials
inventory (Michigan EGLE, 2020).

3.2.8 Data Sources for Schools, Child Care Facilities, Local Health Agencies, and Targeted Medical
Providers

The number of schools, child care facilities, local health agencies, and targeted medical providers are
inputs in calculating the costs and benefits of the final LCRI given the school and child care facility
sampling and public education requirements of the final rule. Sections 3.2.8.1 through 3.2.8.3 describe
the data sources used to estimate the number of these facilities.

3.2.8.1 Schools

The EPA primarily used information from the United States Department of Education's National Center
for Education Statistics (NCES) to estimate the number of elementary and secondary schools, both
public and private, for each State (including Washington, D.C.), United States territories, and on tribal
lands operated by the Bureau of Indian Education (BIE). For public schools, the EPA used 2018 -2019
data from "Table 216.70. Public elementary and secondary schools, by level, type, and State or
jurisdiction: 1990-91, 2000-01, 2010-11, and 2018-19" (NCES, 2020a). For private schools, the EPA used
"Table 15: Number of private schools, students, full-time equivalent (FTE) teachers, and 2018-2019 high
school graduates, by State: United States, 2019-2020 from the NCES Private School Universe Survey"
(NCES, 2020b). The EPA supplemented the NCES data with other sources33 to estimate the number of
public and private schools in the Navajo Nation and the number of private schools in United States

33 Table 1: List of Tribal Public Schools Managed by the Bureau of Indian Education was obtained from this website
on April 30, 2020: https://www.bie.edu/Schools/index.htm. Other sources were identified through an internet
search on schools in the Navajo Nation and in U.S. Territories conducted in March of 2020. For detailed findings,
see the derivation file, "Schools_Child Care lnputs_Final.xlsx", worksheets "NN Pub Priv & CC" and "Private and CC
for Territories".

Final LCRI Economic Analysis

3-13

October 2024


-------
territories (American Samoa, Guam, Northern Marianas, Puerto Rico, and the United States Virgin
Islands). None of the data sources differentiate elementary vs. secondary schools, so the EPA used the
proportion of elementary to secondary public schools per State, United States territory, and BIE-
operated schools from NCES (NCES, 2020a) to estimate the proportion of elementary to secondary
private schools by State, United States territories, and the Navajo Nation. The EPA supplemented the
NCES data with other sources to estimate the number of public and private schools in the Navajo Nation
and the number of private schools in United States territories (American Samoa, Guam, Northern
Marianas, Puerto Rico, and the United States Virgin Islands). The estimated total number of schools
(public and private, elementary and secondary in all States and territories) inclusive of NTNCWSs is
131,264. See the file "School_Child Care lnputs_Final.xlsx" for details.

The estimated number of schools was adjusted to remove the 3,406 public schools and 1,951 private
schools reported in SDWIS/Fed as NTNCWSs, as of December 31, 2020. The adjusted number of public
and private schools is 96,691 and 29,221, respectively.

3.2.8.2	Child Care Facilities

The EPA used data from the 2019 Committee for Economic Development (CED) report analyzing the role
of child care facilities in the economy (CED, 2019). The data for this report was collected in 2017. The
EPA specifically used the information from "Figure 24: Comparative Cost of Child Care (2017)" of the
CED report. The EPA supplemented CED data with additional web-based information on the number of
child care facilities in the Navajo Nation and in United States territories. See the file "School_Child Care
lnputs_Final.xlsx" for details.

The EPA adjusted the estimated number of United States child care facilities (674,794) to remove the
1,252 child care facilities reported in SDWIS/Fed as NTNCWSs, as of December 31, 2020. The adjusted
number of child care facilities is 673,542.

3.2.8.3	Local Health Agencies and Targeted Medical Providers

The EPA used the following sources to estimate the number of local health agencies and medical
providers that are obstetricians/gynecologists (ob/gyn) and pediatricians in the United States:

•	National Association of County and City Health Officials (NACCHO) 2019 National Profile of Local
Health Departments (NACCHO, 2019): This source estimated the number of local health
agencies at 2,459.34

•	The number of ob-gyns (20,700), pediatricians (30,200), and family medicine physicians
(107,700) is from the U.S. Bureau of Labor Statistics' "Occupational Outlook Handbook" (U.S.
Bureau of Labor Statistics, 2021). The EPA downloaded the section that is specific to Physicians

34 A 2020 report was not available. NACCHO uses a database of local health departments based on previous profile
studies and consults with state health agencies and State Associations of Local Health Officials (SACCHOs) to
identify local health departments for inclusion in the study population. For the 2019 Profile study, a total of 2,459
local health departments were included in the study population. Rhode Island was excluded from the study
because the state health agency operates on behalf of local public health and has no sub-state units. For the first
time, Hawaii was included.

Final LCRI Economic Analysis

3-14

October 2024


-------
and Surgeons on September 8, 2021, and used the information presented in the "Work
Environment" section for these three categories of physicians.

Using the sources listed above, the EPA estimated that there are 161,059 local health agencies, ob/gyns,
pediatricians, and family medicine physicians in the United States.

3.3 Drinking Water System Baseline

This section presents the following baseline characterizations for the purposes of estimating costs and
benefits for the final LCRI:

•	Section 3.3.1 provides a characterization of the inventory of CWSs and NTNCWSs that are
subject to the lead and copper regulations.

•	Section 3.3.2 includes the population served by CWSs and NTNCWSs and the number of
households served by CWSs.

•	Section 3.3.3 includes the derivation of the number of CWSs and NTNCWSs with existing CCT
from SDWIS/Fed data, current through December 2020.

•	Section 3.3.4 provides the characterization of service line material for CWSs and NTNCWSs.

•	Section 3.3.5 details how lead and copper 90th percentile data and individual lead sampling data
were used to characterize water systems.

•	Section 3.3.6 provides treatment plant characteristics used to determine treatment costs.

•	Section 3.3.7 provides the derivation of initial lead and copper tap sampling based on
SDWIS/Fed data, current through December 2020.

•	Section 3.3.8 provides the derivation of initial WQP monitoring schedules based on SDWIS/Fed
data, current through December 2020.

•	Section 3.3.9 provides the derivation of the percent of systems that annually add a new source
or treatment from SDWIS/Fed data, current through December 2020.

•	Section 3.3.10 details the derivation of the number of schools, child care facilities, and targeted
medical providers as well as the estimated percent of schools and child care facilities for which a
CWS would receive a waiver from the testing requirements under the final LCRI.

•	Section 3.3.11 describes the derivation of PWS and State labor rates.

Each section includes a characterization of the baseline for CWSs, followed by NTNCWSs, if applicable,
and a characterization of data limitations and uncertainty.

With respect to CCT and LSL status, the EPA contacted 21 of the CWSs serving more than one million
people in 2020 for information. Whenever possible, the EPA used this system-specific information
instead of the estimated values presented in this section for systems serving greater than one million
people in the cost and benefits analysis. See Chapter 4, Section 4.2.3 and Appendix B, Section B.2.3 for
additional information on the data collected for systems serving greater than one million people.

Final LCRI Economic Analysis

3-15

October 2024


-------
3.3.1 Water System Inventory

A key component of the baseline is the inventory of systems subject to the pre-2021 LCR. As shown in
Exhibit 3-2, 40,113 of 49,529 (about 81 percent) of all CWSs serve 3,300 or fewer people, and the 26,816
CWSs serving 500 or fewer account for about 54 percent of all CWSs. The 8,400 CWSs serving 3,301 -
50,000 people comprise about 17 percent of all CWSs, and the 1,016 CWSs serving 50,000 or more
people account for only about 2 percent. Most CWSs (37,904 or about 77 percent) use GW as their
primary source. However, most of the 4,390 CWSs serving above 10,000 people are classified as SW
systems (2,788 CWSs or about 64 percent).

Exhibit 3-2: Inventory of CWSs

System Size
(Population Served)

CWSs

Ground Water

Surface Water

Total

A

B

C = A + B

<100

10,809

923

11,732

101-500

13,028

2,056

15,084

501-1,000

4,168

1,162

5,330

1,001-3,300

5,502

2,465

7,967

3,301-10,000

2,795

2,231

5,026

10,001-50,000

1,365

2,009

3,374

50,001-100,000

161

410

571

100,001-1M

74

347

421

> 1M

2

22

24

TOTAL

37,904

11,625

49,529

Sources: SDWIS/Fed fourth quarter 2020 "frozen" dataset that contains information reported through December
31, 2020. Includes all active CWSs. See Section 3.2.1.1 for detail on system classification (system type, source water
type, and population served using SDWIS). Additional information can be found in "CWS Inventory
Characteristics_Final.xlsx" available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Notes:

A, B: Includes 19 CWSs serving 10,000 or fewer people for which no primary source water type was reported to
SDWIS/Fed. These systems were assigned to the source type of GW or SW based on the ratio of systems with
known GW to SW source type for each size category. Based on this ratio, 16 systems were assigned to the source
type of GW and three to SW.

As shown in Exhibit 3-3, 17,217 of 17,418 (about 99 percent) of all NTNCWSs serve 3,300 or fewer
people. The 199 NTNCWSs serving 3,301 - 50,000 people account for approximately one percent of all
NTNCWSs. Only two NTNCWSs (0.01 percent) serve more than 50,000 people and none serve more than
1 million people. Most NTNCWSs (16,633 or about 95 percent) use GW as their primary source. Eighteen
(46 percent) of those serving 10,001 to 100,000 people use GW versus SW and the one system serving
100,001 to 1 million people is classified as a SW system.

Final LCR! Economic Analysis

3-16

October 2024


-------
Exhibit 3-3: Inventory of NTNCWSs

System Size
(Population Served)

NTNCWSs

Ground Water

Surface Water

Total

A

B

C=A+B

<100

8,138

250

8,388

101-500

6,133

247

6,380

501-1,000

1,489

89

1,578

1,001-3,300

752

119

871

3,301-10,000

103

59

162

10,001-50,000

18

19

37

50,001-100,000

0

1

1

100,001-1M

0

1

1

> 1M

0

0

0

TOTAL

16,633

785

17,418

Sources: SDWIS/Fed fourth quarter 2020 "frozen" dataset that contains information reported through December
31, 2020. Includes all active NTNCWSs. See Section 3.2.1.1 for detail on system classification (system type, source
water type, and population served using SDWIS). Additional information can be found in "NTNCWS Inventory
Characteristics_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Notes:

A: Includes 8 NTNCWSs serving 3,300 or fewer people for which no primary source type was reported to
SDWIS/Fed. These systems were assigned to the source water type of GW or SW based on the ratio of systems
with known GW to SW source type for each size category. The majority of small NTNCWSs are GW systems and
based on these ratios, all 8 systems were assigned to the source type of GW.

3.3.1.1 Discussion of Data Limitations and Uncertainty

As described in Section 3.2.1.4.1, the EPA periodically performed program reviews to verify inventory
information in SDWIS/Fed. From 2006 to 2016, the EPA identified only 82 individual discrepancies (<1
percent), although some discrepancies in the reviews conducted prior to August 200735 could be
unrelated to the population, source type, or system type, such as contact information or address, based
on a detailed review of 2,180 systems, indicating a high level of completeness and accuracy. Although
the EPA has information that shows a low discrepancy rate from program reviews conducted during
2006 - 2016, the agency does not have current national information on discrepancy rates for 2017 -
2020, which are the last four years of the frozen 2020 SDWIS/Fed dataset used for the EA. Thus, the EPA
cannot state with certainty if the discrepancy rate for 2017 - 2020 is similar to that found for 2006 -
2016. However, the EPA continues to evaluate compliance with the pre-2021 LCR through file reviews
with the goal of helping States improve their programs.

There is uncertainty in the approach used to assign source water type to PWSs where no primary source
type was reported to SDWIS/Fed. The EPA assumed that the systems with an unknown source would

35 As previously discussed in Section 3.2.1.4.1, the review policy changed in August 2007 to no longer include the
administrative contact or address and to only identify water system inventory discrepancies that impacted
monitoring requirements, such as a change in population.

Final LCR! Economic Analysis

3-17

October 2024


-------
have the same proportion of GW to SW source types as the overall population of PWSs. This could result
in an under or overestimate of costs in those instances where the cost model inputs vary by source type,
e.g., number of entry points per system. However, the EPA expects the impact to be low because
systems with no source type in SDWIS/Fed represent a small proportion of systems subject to the rule.
Specifically, they comprise 19 or 0.04 percent of the total 49,529 CWSs and 8 or 0.05 percent of the total
17,418 NTNCWSs or 0.04 percent of all systems subject to the rule and all serve 10,000 or fewer people.

3.3.2 Population and Households Served

An accurate characterization of the populations served by water systems is necessary when assessing
the potential benefits of the final LCRI. Population served is also used to estimate volume of water
treated and associated CCT costs.

SDWIS/Fed tracks "retail" population served, meaning that it counts only the population that purchase
water directly from the water system and does not include the population of a water system that
purchase water from another system. Consecutive water systems are recorded in SDWIS/Fed as a
separate system with a unique public water system identification (PWSID) number.

Exhibit 3-4 and Exhibit 3-5 show the total population served and average population served per system
by size category for both CWSs and NTNCWSs, respectively. Each exhibit is organized by source water
type (SW or GW) and is based on SDWIS/Fed fourth quarter 2020 "frozen" dataset that contains
information reported by States through December 31, 2020.

Because systems often pass their costs onto customers in the form of rate increases, the final LCRI cost
analysis also includes analyses to assess the impact of the requirements on a household level. The
number of households served by CWSs expected to be subject to the final LCRI requirements is
estimated by dividing the population for each system size category by the average number of people per
household. For CWSs, the EPA assumed an average of 2.53 persons per household based on 2020 United
States Census data (United States Census Bureau, 2020). This information is also included in Exhibit 3-4
by system size and source type. NTNCWSs do not serve households and thus, this information is not
included in Exhibit 3-5.

As shown in Exhibit 3-4, although CWSs serving 3,300 or fewer account about 81 percent of all CWSs,
they serve fewer than eight percent of the population and households that receive their water from a
CWS. On the other hand, although CWSs serving more than 50,000 people account for only two percent
of all CWSs, they serve more than half (59 percent) of the population and households that receive their
water from a CWS.

Final LCRI Economic Analysis

3-18

October 2024


-------
Exhibit 3-4: Population and Number of Households Served by CWSs

System Size
(Population
Served)

Ground Water

Surface Water

TOTAL

(Includes 19 CWSs with unspecified
primary source)

Population
Served

Average
Population
Per System

Number of
Households
Served

Population
Served

Average
Population
Per System

Number of
Households
Served

Population
Served

Average
Population
Per System

Number of
Households
Served

A

B

C = A/2.53

D

E

F=D/2.53

G=A+D

H

l=G/2.53

<100

658,125

61

260,128

49,688

54

19,640

708,236

60

279,935

101-500

3,249,684

249

1,284,460

578,788

282

228,770

3,830,126

254

1,513,884

501-1,000

3,058,307

734

1,208,817

870,975

750

344,259

3,931,488

738

1,553,948

1,001-3,300

10,267,678

1,866

4,058,371

4,950,969

2,009

1,956,905

15,218,647

1,910

6,015,275

3,301-10,000

15,898,651

5,688

6,284,052

13,660,859

6,123

5,399,549

29,565,710

5,883

11,686,051

10,001-50,000

28,316,279

20,745

11,192,205

45,846,395

22,821

18,121,105

74,162,674

21,981

29,313,310

50,001-100,000

10,785,606

66,991

4,263,085

28,843,811

70,351

11,400,716

39,629,417

69,404

15,663,801

100,001-1M

14,963,849

202,214

5,914,565

84,395,513

243,215

33,357,910

99,359,362

236,008

39,272,475

> 1M

3,400,000

1,700,000

1,343,874

43,238,891

1,965,404

17,090,471

46,638,891

1,943,287

18,434,344

TOTAL

90,598,179

2,390

35,809,557

222,435,889

19,134

87,919,324

313,044,551

6,320

123,733,024

Sources: A, B, D, and E: SDWIS/Fed fourth quarter 2020 "frozen" dataset that contains information reported through December 31, 2020. See file "CWS
Inventory Characteristics_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov. Note that for CWSs in the size category of
serving <100 people in which the reported population was < 24 people, the EPA increased the population to 25. This resulted in an increase in total population
change from 701,258 to 708,236 for this size category.

Notes:

B,	E, and H: Derived by dividing the population served by the number of systems presented in Exhibit 3-2.

C,	F, and I: The average of 2.53 persons per household is from 2020 Census data (Table AVG1. Average Number of People per Household, by Race and Hispanic
Origin, Marital Status, Age, and Education of Householder: 2020).

G-l: CWSs with unreported primary source were not summarized individually, however they were included in the 'TOTAL" columns. Thus, the 'TOTAL" column
reflects an additional 19 CWSs with unreported primary source type.

Final LCRI Economic Analysis

3-19

October 2024


-------
As previously discussed, NTNCWSs serving 3,300 or fewer account for approximately 99 percent of all
NTNCWSs. As shown in Exhibit 3-5, these systems serve approximately 71 percent of the population that
receives their water from a NTNCWS. Those serving 3,301 to 50,000 people and more than 50,000
people serve approximately 25 percent and four percent of the population that receives water from a
NTNCWS, respectively.

Exhibit 3-5: Population Served by NTNCWSs

System Size
(Population
Served)

Ground Water

Surface Water

TOTAL

Population
Served

Average
Population
Per System

Population
Served

Average
Population
Per System

Population
Served

Average
Population
Per System

A

B

D

E

F

G

<100

454,125

56

12,498

50

466,808

56

101-500

1,522,528

248

66,010

267

1,588,708

249

501-1,000

1,060,097

712

66,567

748

1,126,664

714

1,001-3,300

1,254,365

1,675

232,569

1,954

1,493,446

1,715

3,301-10,000

542,409

5,266

350,086

5,934

892,495

5,509

10,001-50,000

330,457

18,359

410,046

21,581

740,503

20,014

50,001-100,000

0

0

71,963

71,963

71,963

71,963

100,001-1M

0

0

203,375

203,375

203,375

203,375

> 1M

0

0

0

0

0

0

TOTAL

5,163,981

310

1,413,114

1,800

6,583,962

378

Sources: SDWIS/Fed fourth quarter 2020 "frozen" dataset that contains information reported through December
31, 2020. See file "NTNCWS Inventory Characteristics_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801
at www.regulations.gov.

Notes:

B, E, and G: Derived by dividing the population served by the number of systems presented in Exhibit 3-3.

F and G: NTNCWSs with unreported primary source were not summarized individually, however they were
included in the "TOTAL" columns. Thus, the 'TOTAL" column reflects an additional eight systems with unspecified
primary source.

3.3.2.1 Discussion of Data Limitations and Uncertainty

As described in Section 3.2.1.4.1, the EPA periodically performs Program Reviews to verify key
parameters in SDWIS/Fed including, but not limited to, population served, system type, and source type
(USEPA, 2007). From 2006 to 2016, the EPA identified only 82 individual water system inventory
discrepancies (<1 percent) based on a detailed review of 2,180 systems, although some discrepancies
could be unrelated to the population, source type, or system type, such as contact information or
address. The results of the Program Review indicate a high level of completeness and accuracy in the
SDWIS/Fed population data (USEPA, 2007). Also as noted in Section 3.3.1.1, the EPA does not have
current national information on discrepancy rates for 2017 - 2020, which are the last four years of the
frozen 2020 SDWIS/Fed dataset used for this EA. Thus, the EPA cannot state with certainty if the
discrepancy rate for 2017 - 2020 is similar to that found for 2006 - 2016. However, the EPA continues to

Final LCRI Economic Analysis

3-20

October 2024


-------
evaluate compliance with the pre-2021 LCR through file reviews with the goal of helping States improve
their programs.

As noted previously, the EPA consistently classifies systems in SDWIS/Fed according to the retail
population served by the system and does not include the population served by wholesale customers.
Wholesale customers that purchase water from another system and meet the PWS definition have their
own unique PWSID, retail population, and associated regulatory requirements under the SDWA. As
described in Chapter 4, Section 4.3.3, the EPA uses retail population to estimate design and average
daily flow parameters, which are then used to estimate CCT costs associated with the rule. Use of retail
population may overestimate costs by assuming that each PWSID will have an individual treatment plant
instead of the more common scenario of the seller having one large plant and selling treated water to its
wholesale customers.

3.3.3 Corrosion Control Treatment (CCT) Status

Under the pre-2021 LCR, the 2021 LCRR, and the final LCRI, systems with CCT in place have different
requirements than those without this treatment. This section includes the EPA's derivation of the
number of CWSs and NTNCWSs with CCT. As noted in the introduction to Section 3.3, the EPA used
system specific CCT information for systems serving greater than one million people where available.

To estimate the percent of CWSs and NTNCWSs with CCT, the EPA used one approach for systems
serving 50,000 or fewer people and a different approach for those systems serving more than 50,000
people. Both approaches rely on information reported to SDWIS/Fed but use different data fields and
assumptions. Systems serving 50,000 or fewer are required under the pre-2021 LCR to install CCT if they
have a lead and/or copper ALE.36 As a first step, the EPA identified CWSs and NTNCWSs for which the
State reported a treatment objective of "C" to identify those with CCT. As noted in Section 3.2.1.4,
treatment code data in SDWIS/Fed is not part of the program review; thus, there is more uncertainty
associated with these data as compared to SDWIS/Fed population and violation data. Therefore, to
supplement the treatment code analysis, the EPA reviewed milestone and violation data to identify
additional CWSs that were required to install CCT as follows:

•	The State reported a "DONE" or "DEEM" milestone with a reason code of "WQP." This indicates
systems for which the State has set OWQPs, and thus would have CCT.37

•	The system was in violation for failure to install CCT {i.e., was assigned violation code 58) and
subsequently addressed this violation. Systems with an addressed code 58 violation were
identified by the enforcement code of "SOX" or "EOX" that denotes compliance achieved.

36	A system serving 50,000 or fewer is triggered into CCT steps that can include a study prior to CCT installation.
However, these systems can discontinue CCT steps if they have two consecutive six-month monitoring periods at
or below both the lead and copper ALs. If they have a subsequent ALE, they must recommence CCT steps but can
discontinue the steps if they again have no ALEs for two consecutive six-month monitoring periods.

37	Following the installation of CCT, the State will set OWQPs that represent the conditions under which systems
must operate their CCT to most effectively minimize the lead and copper concentrations at their users' taps while
not violating any National Primary Drinking Water Regulation.

Final LCRI Economic Analysis

3-21

October 2024


-------
•	The system has an OWQP (59) violation code. As noted above, OWQPs are set for systems with
CCT.

•	The system purchased water from another system that the EPA has identified as having CCT.

CWSs and NTNCWSs serving more than 50,000 people were required under the pre-2021 LCR to install
CCT unless they: 1) had completed treatment steps that are equivalent to those described in the 1991
LCR prior to December 7, 1992 {i.e., meet the criteria of 40 CFR 141.81(b)) or 2) could demonstrate they
have very low levels of lead and copper in the distribution system (i.e., qualify as a "b3" system).38
Therefore, the EPA classified all systems as having CCT except those identified as a b3 system. The EPA
used the following criteria to identify b3 systems:

•	Had a reported "b3" milestone,

•	Did not have CCT using the criteria described above for systems serving < 50,000 people, and

•	Did not have a lead or copper ALE from 1992-2020 and all reported lead 90th percentile levels
are < 5 ng/L or non-detect.

Only 16 CWSs were found to be b3 systems.

As shown in Exhibit 3-6, the EPA estimated that overall, approximately 31 percent of all CWSs have CCT.
The percentage of CWSs with CCT is higher in the larger size categories. Specifically, about 24 percent of
CWSs serving 3,300 or fewer have CCT. Whereas, approximately 59 percent of those serving 3,301 to
50,000 people and approximately 98 percent of those serving more than 50,000 people have CCT.

As shown in Exhibit 3-7, the EPA estimated that overall, approximately 12 percent of all NTNCWSs have
CCT. Approximately 12 percent of those serving 3,300 or fewer and 20 percent of those serving 3,301 to
50,000 people have CCT. No NTNCWS met the b3 criteria; thus, the EPA assumed all NTNCWSs serving
more than 50,000 people had CCT.

38 "b3 systems" is an abbreviated term for those systems that meet the criteria in 40 CFR 141.81(b)(3). Specifically,
under the pre-2021 LCR, for two consecutive six-month monitoring periods, the system's: 1) 90th percentile lead
level minus the highest source water level is < 0.005 mg/L (i.e., 5 ng/L); or 2) source water lead levels are below the
method detection limit (MDL) and the 90th percentile lead level is <0.005 mg/L. As stated above, the EPA applied
more stringent criteria in its analysis by limiting the b3 criteria to system serving more than 50,000 people for
which all reported lead 90th percentile levels were ^0.005 mg/L.

Final LCR! Economic Analysis

3-22

October 2024


-------
Exhibit 3-6: Number of CWSs with and without CCT



Number of CWSs with CCT

Number of CWSs without CCT



System Size
(Population Served)

Ground Water

Surface Water

Total

Ground Water

Surface Water

Total

TOTAL



A

B

C=A+B

D

E

F=D+E

G=C+F

<100

985

405

1,390

9,824

518

10,342

11,732

101-500

2,120

926

3,046

10,908

1,130

12,038

15,084

501-1,000

1,077

584

1,661

3,091

578

3,669

5,330

1,001-3,300

1,851

1,525

3,376

3,651

940

4,591

7,967

3,301-10,000

1,171

1,566

2,737

1,624

665

2,289

5,026

10,001-50,000

693

1,550

2,243

672

459

1,131

3,374

50,001-100,000

158

402

560

3

8

11

571

100,001-1M

72

344

416

2

3

5

421

>1 M

2

22

24

0

0

0

24

TOTAL

8,129

7,324

15,453

29,775

4,301

34,076

49,529

Source: SDWIS/Fed fourth quarter 2020 "frozen" dataset that contains information reported through December 31, 2020. See file, "CWS Inventory

Characteristics_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Notes:

D & E: Includes 19 CWSs serving 10,000 or fewer people with no CCT for which no primary source type was reported to SDWIS/Fed. These systems were
assigned to the source type of GW or SW based on the ratio of systems with known GW to SW source type for each size category. Based on this ratio, 16
systems were assigned to the source type of GW and three to SW. All CWSs identified as having CCT had a reported source type.

Final LCRI Economic Analysis

3-23

October 2024


-------
Exhibit 3-7: Number of NTNCWS with and without CCT

System Size
(Population Served)

Number of NTNCWSs with CCT

Number of NTNCWSs without CCT

TOTAL

Ground Water

Surface Water

Total

Ground Water

Surface Water

Total

A

B

C=A+B

D

E

F=D+E

G=C+F

<100

704

25

729

7,434

225

7,659

8,388

101-500

823

46

869

5,310

201

5,511

6,380

501-1,000

255

22

277

1,234

67

1,301

1,578

1,001-3,300

160

28

188

592

91

683

871

3,301-10,000

25

9

34

78

50

128

162

10,001-50,000

3

2

5

15

17

32

37

50,001-100,000

0

1

1

0

0

0

1

100,001-1M

0

1

1

0

0

0

1

>1 M

0

0

0

0

0

0

0

TOTAL

1,970

134

2,104

14,663

651

15,314

17,418

Source: SDWIS/Fed fourth quarter 2020 "frozen" dataset that contains information reported through December 31, 2020. See file, "NTNCWS Inventory

Characteristics_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Notes:

D: Includes eight NTNCWSs serving 3,300 or fewer people with no CCT for which no primary source type was reported to SDWIS/Fed. These systems were
assigned to the source type of GW or SW based on the ratio of systems with known GW to SW source type for each size category. The majority of small
NTNCWSs are GW systems and based on these ratios, all eight systems were assigned to the source type of GW. All NTNCWSs identified as having CCT had a
reported source type.

Final LCRI Economic Analysis

3-24

October 2024


-------
3.3.3.1 Discussion of Data Limitations and Uncertainty

There is uncertainty in the estimated percent of CWSs and NTNCWSs with CCT. For systems serving
more than 50,000 people, the assumptions are based on the pre-2021 LCR requirements that all systems
must install CCT unless they had installed it previously (i.e., as required in the 1991 rule) or have very
low lead and copper levels, which signifies that they have naturally non-corrosive water (i.e., they are b3
systems). Therefore, the uncertainty in these estimates is not expected to have a significant impact on
benefits and costs of the final LCRI. For systems serving 50,000 or fewer people, the EPA recognizes
greater uncertainty in using treatment objective code data from SDWIS/Fed to identify systems with
CCT. Thus, the EPA supplemented these data with milestone and violation information to identify those
systems that would have been required to install CCT under the pre-2021 LCR. The EPA recognizes that it
is unlikely that SDWIS/Fed contains complete and current treatment and milestone data for all water
systems, especially for smaller water systems. The uncertainty in the percent of systems with CCT may
result in an under or overestimate of costs and benefits of the final LCRI.

3.3.4 Service Line Material Characterization

The characterization of service line material and the characterization of systems based on their mix of
service line materials are key inputs to calculating the costs and benefits of the final LCRI. Sections
3.3.4.1 and 3.3.4.2 provide the detailed characterization of service line material for CWSs and NTNCWSs,
respectively. Section 3.3.4.3 follows with a discussion of current State regulations related to service line
replacement that impact the estimated cost of service line replacement under the LCRI requirements.

3.3.4.1 Service Line Material Characterization for CWSs

This introductory section presents:

•	The EPA's rationale for using the 7th DWINSA results (including results from the one-time
update) as the primary data source for service line material characterization.

•	An explanation of the differences in classification of service line material between this EA and
the analysis conducted for the 7th DWINSA LSL Drinking Water State Revolving Fund (DWSRF)
allocation.

•	How this analysis classifies systems based on their mix of service line material.

Sections 3.3.4.1.1 and 3.3.4.1.2 follow with the characterization of CWSs based on service line material
and the characterization of service line materials within those CWSs, respectively. Estimates for
NTNCWSs are in Section 3.3.4.2.

The EPA reviewed available national scope data sources for characterizing service line material types.
The Economic Analysis for the Final Lead and Copper Rule Revisions (hereafter referred to as the "Final
2021 LCRR EA") (USEPA, 2020a) used two datasets to characterize LSLs based on surveys done by the
American Water Works Association (AWWA).39 Since the 2021 LCRR was finalized, LSL survey data from

39 The sources were: (1) the Lead Information Survey conducted by AWWA in 1989, as published in the USEPA.
1991. Final Regulatory Impact Analysis of National Primary Drinking Water Regulations for Lead and Copper. April

Final LCRI Economic Analysis

3-25

October 2024


-------
the 7th DWINSA, collected primarily from February 2021 - December 2021, have become available. As
described in Section 3.2.5, the 7th DWINSA surveyed: all systems in the country serving more than
100,000 people, a per-State sample of systems serving between 3,301 and 100,000 people, and a
national sample of systems serving 3,300 or fewer people. In addition to information on the presence of
LSL, the 7th DWINSA is the first time that comprehensive national data has been collected on galvanized
service lines requiring replacement (GRR) and lead connectors. The 7th DWINSA is the largest and
broadest scope data collection effort since the survey's inception in 1995. The dataset contains a wide
range of responses from small, medium, and large systems, and from urban and rural systems. The
response rate for the survey overall was 97 percent, with the response rate for the supplemental LSL
questionnaire being lower but still high at 80 percent. Furthermore, systems were assigned sample
weights to improve the degree to which surveyed systems were representative of systems within their
assigned strata.40 Due to the extensiveness and representativeness of the dataset and the detailed
information gathered on service line material, the EPA used the 7th DWINSA results to characterize
service line material for this EA in place of the two previous AWWA surveys.

The 7th DWINSA asked systems for detailed information on service line material including the number of
service lines for the following material categories:

Lead Pipe - Service lines that contain any lead pipe.

Lead Connectors - Service lines that do not contain any lead pipe but have lead connectors such
as goosenecks or pigtails.

Galvanized/Lead Pipe - Service lines that contain galvanized pipe and were previously
downstream from a lead pipe that was removed from the service line.

Galvanized/Lead Connector - Service lines that contain galvanized pipe and were previously
downstream from a lead connector that was removed from the service line.

Galvanized/Unknown Lead - Service lines that contain galvanized pipe and were previously
downstream of an unknown source of lead that was removed from the service line.

Galvanized/Standalone - Service lines that contain galvanized pipe that have never been
downstream from any lead pipe or lead connector in the service line.

No Lead or Galvanized - Service lines that do not contain any lead pipe or galvanized pipe and do
not have lead connectors.

Unknown - Service lines for which the material makeup of the service line and of the connector
are not known.

Unreported - Services lines for which the system did not provide any information on their
material.

1991. Office of Water, and (2) AWWA surveys conducted in 2011 and 2013, results published in Cornwell, D.A, R.A.
Brown, and S.H Via. 2016. National Survey of Lead Service Line Occurrence. Journal AWWA. 108(4):E182-E191.

40 See Section 3.2.5 for an explanation of the DWINSA system sampling weights.

Final LCRI Economic Analysis

3-26

October 2024


-------
As described in Section 3.2.5, the EPA provided a one-time opportunity for previously surveyed water
systems to update their 7th DWINSA service line material questionnaire responses. This update,
conducted between September - November 2023, allowed these water systems and States to revise
their original response based on new service line inventory information or to provide a response if they
had not participated previously. Participation in this update was voluntary and limited to the service line
material questionnaire. For the one-time update, the EPA simplified the questionnaire, merging the
categories of galvanized previously downstream of lead pipe, galvanized previously downstream of
unknown pipe, and galvanized previously downstream of lead connectors.

The EPA combined the new data from this update with the original 7th DWINSA, replacing the system
level data where new information was available. This combined dataset is referred to as the "DWINSA
LSL Allocation Model" throughout this document. The EPA used this model to generate ratios of lead,
non-lead, unknown, and unreported service lines for each State.41 These ratios were applied to the total
number of service lines in the State to generate the counts in each category. For small systems and
States lacking sufficient data for State-specific models, the EPA used ratios based on national data.42 The
results are used by the EPA to allocate DWSRF grants, including funding from the Bipartisan
Infrastructure Law, to States. This allocation model was also used by the EPA to characterize service line
material for this EA.

To be consistent with the final LCRI regulatory definitions, the EPA combined some of the service
material categories for this EA. See Exhibit 3-8 for a comparison of LCRI and DWINSA categories. For the
purposes of these analyses, the EPA uses the term lead content service lines to indicate the broader
group of service lines that are LSLs, GRR service lines, lead connectors, and galvanized previously
downstream of lead connectors. The EPA also tracks unknown service lines because the EPA estimates
that systems will incur burden and costs as they investigate their unknowns, prepare their service line
inventory updates, and replace unknowns that are found to be lead or GRR service lines under the final
LCRI. For the purposes of the final LCRI, the EPA grouped the unreported responses to the survey with
the unknown responses. This may result in an overestimate of the unknown service lines, as some
systems may have elected not to complete the service line questions as opposed to not knowing their
service line material. As will be discussed later in this section, the EPA estimates that a portion of the
unknown service lines will be found to have lead content.

41 Note the DWINSA LSL Allocation Model calculates service line characteristic ratios at the state level using sample
weights which reflect the probability of a system being sampled. The DWINSA LSL Allocation Model does not
estimate service line characteristic ratios by system size. The number of service lines per system varies within each
state, which adds uncertainty to the estimate of the total number of service lines. The additional steps taken to
estimate SL ratios and counts by system size are presented in the remainder of this section.

42For small systems state level SL counts the use of national ratios will introduce additional uncertainty into the
state specific estimates. However, because the small system survey sample design and weights are designed to be
nationally representative, using the national ratios in place of state values (when there is insufficient state level
data) does not result in additional uncertainty in the final national estimates used in the LCRI analysis.

Final LCRI Economic Analysis

3-27

October 2024


-------
Exhibit 3-8: Relationship between Service Line Categories in the Final LCRI Economic Analysis,
the 7th DWINSA, and the DWINSA One-time Update

LCRI Service Line Category1

7th DWINSA Service Line
Category

DWINSA One-time
Update Service Line
Category

Lead Content
Service Lines

Lead Service Line (LSL)

Lead Pipe

Lead Pipe



Galvanized Requiring
Replacement (GRR) Service
Line

Galvanized/Lead Pipe
Galvanized/Unknown

Galvanized/Lead Pipe



Lead Connectors

Lead Connectors

Lead Connectors



Galvanized Previously
Downstream of Lead
Connectors

Galvanized/Lead Connectors

Galvanized/Lead Pipe

Unknown



Unknown

Unknown





Unreported2

Unreported2

Non-Lead



No Lead or Galvanized
Galvanized/Standalone

No Lead or Galvanized
Galvanized/Standalone

Notes:

1.	The final LCRI service line categories match the regulatory definitions. For the purposes of this EA, the first 4
rows represent "lead content service lines."

2.	While the DWINSA LSL Allocation Model maintained the distinction between unknown and unreported service
lines but applied the same method to estimate the proportion of these respective categories of service lines that
are lead, this analysis groups unreported service lines with unknowns. The results of this approach are the same in
this EA and the DWINSA LSL Allocation Model.

Solely for the purposes of modeling LCRI costs and benefits using the SafeWater LCR model, the EPA also
categorized systems based on their mix of service line material as follows:

•	Category 1: Systems with any known lead content service lines.

•	Category 2: Systems with all non-lead service lines.

•	Category 3: Systems with all unknown content (i.e., unknown) service lines.

•	Category 4: Systems with a mix of non-lead and unknown service lines.

These system categories are shown in Exhibit 3-9. Note that Category 1 systems can contain known lead
content, non-lead, and unknown service lines. Category 3 systems reported all unknown service lines.
Category 4 systems have a mix of non-lead and unknown service lines. Note that for Categories 1, 3, and
4, some proportion of the unknown service lines were projected to be lead content service lines.

Final LCRI Economic Analysis

3-28

October 2024


-------
The EPA used the DWINSA LSL Allocation Model to estimate the percent of systems in each of the four
categories (see the fourth column for the SafeWater LCR model 43 variable name). The EPA also used the
model and decision rules to characterize the service lines for each of the four system categories (see the
fifth column of Exhibit 3-9 for relevant SafeWater LCR model variable names for characterization of
service lines). The detailed approach and results are described in the next two subsections.

Exhibit 3-9: CWS Categorization Based on Service Line Material

CWS
Category

CWS Category
Description

Types of
Service Lines
Within Each
CWS Category

SafeWater LCR Model

Variable Name for
percent of SYSTEMS in
the CWS category

SafeWater LCR Model Variable Names
for classification of SERVICE LINES in the
CWS Category

1

Systems with
any known lead-
content service
lines

•	Lead
Content

•	Non-Lead

•	Unknown

pjsl

•	Percent of service lines with known
material (perc_lsl_known)

•	Percent of known service lines that are
lead (perc_lsl_known_lead)

•	Percent of unknown service lines that
are found to be lead

(perc_ unkn ownjead)

2

Systems with all
non-lead service
lines

• Non-lead

See note 1

• None (all are non-lead)

3

Systems with all
unknown service
lines

• Unknown

p_lsl_unknown

• Percent of unknown service lines that
are found to be lead
(perc_ unkn ownjead)

4

Systems with
both non-lead
and unknown
service lines

•	Non-Lead

•	Unknown

p_lsl_nolead_unknown

•	Percent of service lines that are
unknown

[percJsl_nolead_unknown_unknown]

•	Percent of unknown service lines that
are found to be lead

(perc_ unkn ownjead)

Notes:

1. The percent of systems with all non-lead service lines is equal to 1 minus pjsl, p_lsl_unknown, and
p_lsl_nolead_unknown).

3.3.4.1.1 CWSs with Known or Potential Lead

This section presents the percent and number of CWSs in each of the four service line material
categories: 1) any known lead content, 2) all non-lead, 3) all unknown, and 4) both non-lead and
unknown. As described previously, these percentages were derived using the DWINSA LSL Allocation
Model for the nine system size categories used in this EA and also by CCT status.

Exhibit 3-10 shows the percent of CWSs in service line material Categories 1 through 4 for each CCT and
system size category using the DWINSA LSL Allocation Model.44 Column G shows the estimated total

43	See Chapter 4, Section 4.2 for an overview of the SafeWater LCR cost model.

44	All systems serving fewer than 3,301 people were analyzed as a single bin due to the limited number of
observations for systems of this size, even though they are presented separately in this analysis.

Final LCR! Economic Analysis

3-29

October 2024


-------
number of systems with lead content and/or unknowns, calculated by multiplying the percentages by
the SDWIS/Fed 4th quarter 2020 systems data (note that SDWIS/Fed 4th quarter 2020 data is used
throughout this EA for consistency). For modeling purposes, the EPA assumed that lead content service
lines can be found in any system reporting unknowns. Thus, the total number of systems in Column G
represents the total number of CWSs with potential lead content service lines.

The EPA recognizes the uncertainty in this assumption; some systems reporting all or some unknowns
are likely to discover that they have no lead content service lines. With the current data, it is not
possible to estimate the proportion of systems that will find all their unknowns to be non-lead. This
assumption may overestimate the percentage of systems with potential lead content. This could result
in an overestimate of systems that exceed the action level and subsequently install or change treatment,
given that the EPA assumes higher lead concentrations in lead content systems vs. non-lead systems
(See Section 3.3.5 for the analysis). However, this assumption does not affect the total projected
number of lead content service lines that will be presented later in this section. See Section 3.3.4.1.3 for
an additional discussion of uncertainty regarding the EPA's service line material characterization.

Final LCRI Economic Analysis

3-30

October 2024


-------
Exhibit 3-10: Percent and Number of CWSs in Service Line Material Categories

System Size
(Population
Served)

Number of
CWSs

CCT Status

Percent of CWSs by Service Line Category

Number of CWSs with Lead
Content and/or Unknowns
[i.e., Potential Lead Content)

Any Lead
Content (1)

All Non-
Lead (2)

All Unknowns (3)

Non-Lead and Unknown
(4)

p_lsl

p_lsl_unknown

p_lsl_nolead_unknown

A

B

C

D

E

F

G = (C+E+F)*A

<100

10,342

No

1.8%

53.9%

38.0%

6.3%

4,766

<100

1,390

Yes

8.8%

47.7%

33.5%

10.0%

727

101-500

12,038

No

1.8%

53.9%

38.0%

6.3%

5,547

101-500

3,046

Yes

8.8%

47.7%

33.5%

10.0%

1,593

501-1,000

3,669

No

1.8%

53.9%

38.0%

6.3%

1,691

501-1,000

1,661

Yes

8.8%

47.7%

33.5%

10.0%

868

1,001-3,300

4,591

No

1.8%

53.9%

38.0%

6.3%

2,115

1,001-3,300

3,376

Yes

8.8%

47.7%

33.5%

10.0%

1,765

3,301-10,000

2,289

No

17.7%

39.2%

33.9%

9.2%

1,391

3,301-10,000

2,737

Yes

25.2%

35.2%

25.1%

14.5%

1,772

10,001-50,000

1,131

No

25.7%

35.1%

26.1%

13.1%

734

10,001-50,000

2,243

Yes

32.2%

25.6%

26.2%

16.0%

1,669

50,001-100,000

11

No

0.0%

46.6%

35.8%

17.6%

6

50,001-100,000

560

Yes

37.7%

23.2%

24.2%

14.9%

430

100,001-1,000,000

5

No

0.0%

28.9%

64.8%

6.3%

4

100,001-1,000,000

416

Yes

37.5%

23.2%

24.9%

14.4%

319

>1,000,000

0

No

0.0%

0.0%

0.0%

0.0%

0

> 1,000,000

24

Yes

45.8%

18.8%

20.8%

14.6%

20

Total

49,529











25,416

Notes: Category 1: systems have any lead known content service lines; Category 2: systems have all non-lead service lines; Category 3: systems have all
unknown service lines; Category 4: systems have a mix of non-lead and unknown service lines. Note that Category 1 are systems with known lead content
service lines, and Categories 3 and 4 are systems without known lead content lines but have potential lead content service lines.

Sources:

A: SDWIS/Fed 4th Quarter 2020 freeze. See Section 3.3.3 for the EPA's approach for assigning CCT status for each CWS.

C - F: The DWINSA LSL Allocation Model. See the file "Service Line Characterization using DWINSA_Final.xlsx", worksheet "CWS Lead Service Line Status."
G: Totals may not add due to rounding.

Final LCRI Economic Analysis

3-31

October 2024


-------
3.3.4.1.2 Characterization of Known and Potential Lead Content Service Lines by System Category

The previous section estimated the percentage and number of water systems (specifically CWSs) with
potential lead content service lines. The next step is to estimate the number of service lines with
potential lead content within those systems. In this EA, these potential lead content lines are treated as
if they are in fact lead content lines. For the purposes of SafeWater LCR modeling, the EPA developed
separate service line characterizations for each of the four system categories described in the previous
section.

The primary purpose of the DWINSA is to allocate DWSRF funding to each State, so the DWINSA LSL
Allocation Model generated ratios to calculate the number of lead, non-lead, unknown, and unreported
service lines by State, not by the four system material categories or the nine system size categories
needed for the final LCRI analysis. Though the DWINSA sample is large, it is not big enough to reliably
estimate ratios by system size for the four system categories by state. Instead, the model is used to
estimate the ratios in the aggregate. Using these results, the EPA developed the following seven-step
approach to characterize service line materials for each of the four system categories as presented in
the referenced exhibits:

•	Step 1: Estimate the number of service lines by material type (known lead content, unknown,
and non-lead) in the nation using the DWINSA LSL Allocation Model. See Exhibit 3-11.

•	Step 2: Estimate the number of service lines in system Categories 1 through 4 using the DWINSA
LSL Allocation Model. See Exhibit 3-12.

•	Step 3: Use the results of Step 2 and decision rules to allocate the known lead content,
unknown, and non-lead service lines from Step 1 to system Categories 1 through 4. See Exhibit
3-13, which includes a table for each of the four system categories.

•	Step 4: Divide the results from Step 3 by the total service lines in Step 2 to calculate the
percentages of each service line type per category for the SafeWater LCR model.

•	Step 5: Estimate the percent of unknown service lines that are projected to be found to have
lead content.

•	Step 6: Use the percentages from Step 4 and 5 and the adjusted number of service lines from
SDWIS/Fed 4th quarter 2020 to calculate the total number of service lines with potential lead
content in each system category. See:

o	Exhibit 3-14, which presents total service lines in CWSs

o	Exhibit 3-15 for Category 1 Systems

o	Exhibit 3-16 for Category 3 Systems

o	Exhibit 3-17 for Category 4 Systems

o	Exhibit 3-18, which shows the total lead content service lines for all categories.

Note that there is no exhibit for Category 2 systems because these systems have no potential
lead content service lines.

Final LCRI Economic Analysis

3-32

October 2024


-------
• Step 7: Characterize the breakouts of lead content service lines, specifically what portion are
LSLs, GRR, lead connectors, and galvanized lines previously downstream of lead connectors. See
Exhibit 3-18.

The details of each step and results are presented below.

Step 1: Use the DWINSA LSL allocation model to estimate the total number of service lines nationally
and the proportion that are known lead content, unknown, and non-lead. As noted earlier, the
unreported were grouped with the unknown. Results are shown in Exhibit 3-11.

Exhibit 3-11: Characterization of Service Lines from the 7th DWINSA by Material Type and

System Size

System size
(population served)

Number of Service Lines Nationally from
DWINSA by Material Type

Total Number
of Service
Lines

Estimated Percent of
DWINSA Reported
Unknowns that Are
Projected to be Lead
Content Service Lines

Lead Content

Non-Lead

Unknown

A

B

C

D = A+B+C

E



3,143

223,019

121,141

347,303

1.4%

101-500

13,824

981,072

532,907

1,527,803

1.4%

501-1,000

13,525

959,841

521,375

1,494,741

1.4%

1,001-3,300

47,831

3,394,511

1,843,860

5,286,202

1.4%

3,301-10,000

566,386

5,026,111

4,154,122

9,746,619

11.2%

10,001-50,000

1,513,830

12,327,293

9,766,986

23,608,109

12.6%

50,001-100,000

679,129

6,656,380

4,711,647

12,047,156

12.5%

100,001-1,000,000

1,591,955

16,952,488

11,504,703

30,049,146

11.5%

>1,000,000

683,352

7,965,044

5,123,021

13,771,417

10.8%

Total

5,112,975

54,485,759

38,279,762

97,878,496



Source: DWINSA LSL Allocation Model with system weights applied to the service lines as an approximation.

Notes: The Unknown service lines in Column C include the unreported survey responses. Due to the additional size
categories added for the final LCRI economic analyses, some service lines were dropped due to rounding. The EPA
estimated the number of lead, non-lead, and unknown content service lines for each state. Each of these service
lines was then assigned to a size category, determined by the proportion of service lines falling within each size
category in the state. After completing these calculations for each state, the EPA aggregated the totals for lead,
non-lead, and unknown content lines for each size category across all states. The EPA then estimated the number
of unknown service lines projected to be lead by applying the ratio of known lead to the sum of known lead and
non-lead service lines in each state (value in column A divided by columns A plus B but at the state level). The EPA
then allocated the number of unknown service lines projected to be lead to their respective size categories,
following the same proportional method used in estimating columns A-C. After completing these calculations for
each state, the EPA aggregated the totals for unknown service line projected to be lead by each size category
across all states. Finally, to determine the Estimated Percent of Unknowns that are projected to be lead, the EPA
divided the number of unknowns projected to be lead nationally in each size category by the total number of
unknown service lines nationally in that size category. Note that because EPA analyzed data by State rather than by
system size, which is consistent with the 7th DWINSA-based allocation calculations, column E cannot be computed
using the numbers in this table.

Final LCRI Economic Analysis

3-33

October 2024


-------
The total number of service lines shown in Exhibit 3-11 is 97.9 million, which is slightly less than the
reported 100 million service lines in the 7th DWINSA Fact Sheet.45 This difference occurs because this EA
used additional categories for lead content service lines, unlike the DWINSA LSL Allocation Model where
these lines were combined. The additional categories resulted in smaller ratios, causing some service
lines to be dropped due to rounding. Note that the data in Exhibit 3-11 are used to estimate the percent
of service line material types, which are then applied in the SafeWater LCR model to adjusted
SDWIS/Fed 4th Quarter 2020 connection data. Therefore, differences between the total service lines in
Exhibit 3-11 and the DWINSA Fact Sheet are not expected to significantly affect the accuracy of the final
LCRI cost and benefit analysis.

Step 2: Use the 7th DWINSA to estimate the total number of service lines in Categories 1 through 4. The
EPA used the system weights from the 7th DWINSA as an approximation for service line weights. Results
are shown in Exhibit 3-12.

Exhibit 3-12: Total Number of Service Lines in Categories 1 through 4 from DWINSA

System Size
(population
Served

LCRI System Categories

Total Service
Lines

Any Lead
Content (1)

All Non-Lead
(2)

All Unknown
(3)

Non-lead and
Unknown (4)

<100

18,759

180,285

103,421

44,838

347,303

101-500

82,523

793,085

454,953

197,243

1,527,803

501-1,000

80,737

775,922

445,108

192,974

1,494,741

1,001-3,300

285,528

2,744,074

1,574,139

682,460

5,286,202

3,301-10,000

2,426,840

3,382,879

2,676,109

1,260,792

9,746,619

10,001-50,000

7,950,684

6,542,558

5,499,984

3,614,884

23,608,109

50,001-100,000

4,263,166

3,094,325

2,997,895

1,691,771

12,047,156

100,001-1,000,000

14,012,813

4,706,307

6,563,039

4,766,986

30,049,146

>1,000,000

7,646,197

1,476,881

1,865,054

2,783,284

13,771,417

Total

36,767,246

23,696,316

22,179,702

15,235,232

97,878,496

Notes: Category 1: systems have any lead known content service lines; Category 2: systems have all non-lead
service lines; Category 3: systems have all unknown service lines; Category 4: systems have a mix of non-lead and
unknown service lines. Unreported service lines are grouped with unknown service lines.

Source:

Results of the DWINSA LSL Allocation Model system weights used as an approximation for service line weights. See
derivation file "Service Line Characterization using DWINSA_Final.xlsx", worksheet "SL by Category."

Step 3: Use the service line characterizations from Exhibit 3-11 and Exhibit 3-12 and simplifying
assumptions to estimate the distribution of known lead content, non-lead, and unknown service lines in
Categories 1 through 4. The EPA's approach is described below.

45 Fact Sheet: 7th Drinking Water Infrastructure Needs Survey and Assessment, April 2023. Available online at
https://www.epa.gov/svstem/files/documents/2023-04/Final DWINSA%20Public%20Factsheet%204.4.23.pdf

Final LCRI Economic Analysis

3-34

October 2024


-------
Exhibit 3-13 includes a set of three tables that provide the number of service lines for each of the nine
size categories for Categories 1 through 4 with the number of lead content in the first table; the number
of non-lead service lines in the second table; and the number of unknowns in the third table. Note that
the total number of known lead content (5.1 million), non-lead (54.5 million), and unknown (38.3
million) service lines in the last column, last row of each table in Exhibit 3-13 matches the total for each
service line type in Exhibit 3-11.

The EPA used the following assumptions to determine the number of each type of service line by
category:

1.	Assign all known lead content service lines (5.1 million) to Category 1 based on system
category definitions.

2.	Of the total 54.5 million known non-lead service lines (as presented in Column B in Exhibit
3-11), assign 23.7 million to Category 2 since all service lines in Category 2 are non-lead (see
the total non-lead service lines for Category 2 in Exhibit 3-12).

3.	Of the total service lines in Category 4 systems (15.2 million), make a simplifying assumption
that two-thirds are non-lead. The EPA chose this ratio based on the distribution of unknown
and non-lead service lines in the original DWINSA survey and the one-time update in the
absence of detailed estimates for each size category. The EPA recognizes this is a simplifying
assumption that may introduce uncertainty; however, the EPA does not calculate national
costs independently for Categories 1 through 4. These categories were used for modeling
purposes only. The total number of non-lead service lines assigned to category 4 using this
approach is approximately 10.2 million.

4.	Assign the remaining non-lead service lines (54.5- 23.7 - 10.2 = 20.6 million) to Category 1.

5.	Of the total 38.3 million unknown service lines (see Column C in Exhibit 3-11), assign 22.2
million to Category 3 since all 22.2 million service lines in Category 3 are unknown by
definition (see Exhibit 3-12).

6.	Assume that the other third of the service lines in Category 4 systems are unknown. This
represents the remaining service lines in Category 4, which is equal to the total of 15.2
million service lines (see Category (4) Exhibit 3-12) minus 10.2 million assigned non-lead
service lines from Step 3 above, for a total of 15.2 - 10.2 = 5.0 million unknown service lines
in Category 4.

7.	Assign the remaining unknown service lines (38.3 - 22.2- 5.0 = 11.1 million) to Category 1.

Exhibit 3-13: Allocation of Known Lead Content, Non-Lead, and Unknown Service Lines to

Categories 1 through 4

Table 1 Known Lead Content Service Lines

Final LCRI Economic Analysis

3-35

October 2024


-------
System Size
(Population Served)

Number of Lead Content Service Lines per System Category

Total Lead Content
Service Lines

Any Lead
Content (1)

All Non-Lead
(2)

All Unknown
(3)

Non-lead and
Unknown (4)

<100

3,143

-

-

-

3,143

101-500

13,824

-

-

-

13,824

501-1,000

13,525

-

-

-

13,525

1,001-3,300

47,831

-

-

-

47,831

3,301-10,000

566,386

-

-

-

566,386

10,001-50,000

1,513,830

-

-

-

1,513,830

50,001-100,000

679,129

-

-

-

679,129

100,001-1,000,000

1,591,955

-

-

-

1,591,955

>1,000,000

683,352

-

-

-

683,352

Total

5,112,975

-

-

-

5,112,975

Table 2 Non Lead Service Lines

System Size
(Population Served

Number of Non-Lead Service Lines per System Category

Total Non-Lead
Service Lines

Any Lead
Content (1)

All Non-Lead
(2)

All Unknown
(3)

Non-lead and
Unknown (4)

<100

12,692

180,285

-

30,041

223,019

101-500

55,835

793,085

-

132,153

981,072

501-1,000

54,626

775,922

-

129,293

959,841

1,001-3,300

193,188

2,744,074

-

457,248

3,394,511

3,301-10,000

798,502

3,382,879

-

844,731

5,026,111

10,001-50,000

3,362,763

6,542,558

-

2,421,972

12,327,293

50,001-100,000

2,428,569

3,094,325

-

1,133,486

6,656,380

100,001-1,000,000

9,052,300

4,706,307

-

3,193,881

16,952,488

>1,000,000

4,623,362

1,476,881

-

1,864,801

7,965,044

Total

20,581,838

23,696,316

-

10,207,605

54,485,759

Table 3 Unknown Service Lines

System Size
(Population Served

Number of Unknown Service Lines per System Category

Total Unknown
Service Lines

Any Lead
Content (1)

All Non-Lead
(2)

All Unknown
(3)

Non-lead and
Unknown (4)

<100

2,924

-

103,421

14,796

121,141

101-500

12,864

-

454,953

65,090

532,907

501-1,000

12,585

-

445,108

63,682

521,375

1,001-3,300

44,509

-

1,574,139

225,212

1,843,860

3,301-10,000

1,061,952

-

2,676,109

416,061

4,154,122

10,001-50,000

3,074,090

-

5,499,984

1,192,912

9,766,986

50,001-100,000

1,155,468

-

2,997,895

558,284

4,711,647

Final LCRI Economic Analysis

3-36

October 2024


-------
System Size
(Population Served

Number of Unknown Service Lines per System Category

Total Unknown
Service Lines

Any Lead
Content (1)

All Non-Lead
(2)

All Unknown
(3)

Non-lead and
Unknown (4)

100,001-1,000,000

3,368,558

-

6,563,039

1,573,105

11,504,703

>1,000,000

2,339,483

-

1,865,054

918,484

5,123,021

Total

11,072,433

-

22,179,702

5,027,626

38,279,762

Source: DWINSA LSL Allocation Model. See the derivation file, "Service Line Characterization using
DWINSA_Final.xlsx", worksheet "SL by Category."

Step 4: Calculate the percentages needed to characterize service line material type for the SafeWater
LCR model by dividing the known lead content, non-lead, and unknown service lines counts per system
type and size category in Exhibit 3-13 by the total service lines by system type and size category in
Exhibit 3-12. For example, for systems serving 1,001 - 3,300 in Category 1, the percent of lines that are
lead content is calculated as 47,831 (from Exhibit 3-13, Table 1) divided by 285,528 (from Exhibit 3-12)
equals 16.8 percent. The results of this step for all SafeWater LCR variables are combined with the
results of Steps 5 and 6 below.

Step 5: Estimate the proportion of unknown service lines that contain lead. Many survey respondents
reported that they either did not know the material of their service lines or left the section blank (i.e.,
the "unreported" responses). For the DWSRF allocation, the EPA explored different methods to estimate
the proportion of these unknown and unreported service lines that might contain lead.

In the absence of more detailed information, the EPA assumed that for each State, the proportion of
unknown and unreported service lines with potential lead content is equal to the percent of all known
service lines in that State (lead + non-lead) that contain lead. The EPA estimated the number of lead
content service lines in each State and calculated the ratio of lead content service lines to the total
number known service lines. This ratio was then applied to the number of unknown service lines in each
State to project the number lead content service lines. The EPA then distributed these estimated lead
content service lines across different system size categories based on the percent of service lines in each
size category within each State. The counts were aggregated to get the projected number lead content
service lines in each size category nationally. This national number was divided by the total number of
unknown service lines in each size category to find the projected percentage of lead content service
lines. Results are shown in Exhibit 3-11, Column E. The EPA recognizes the uncertainty in this
assumption and that the actual number of lead content service lines among the unknowns and
unreported may be higher or lower.

Step 6: Use the results from Steps 4 and 5 (in the form of percents) and the number of service
connections from the SDWIS/Fed 4th quarter 2020 data to calculate the total number of potential lead
content service lines per system category and size strata.

Although the DWINSA LSL Allocation Model is based on the second quarter of 2019 SDWIS/Fed
population data, adjusted based on survey responses, this EA uses SDWIS 2020 4th quarter freeze data
consistently across all analyses. For the purposes of modeling systems in the SafeWater LCR model, the
EPA evaluated the 2020 connection data compared to the minimum population for PWSs and the

Final LCR! Economic Analysis

3-37

October 2024


-------
relationship between connections and populations {i.e., population per number of connections). The
EPA adjusted the SDWIS/Fed 4th quarter 2020 connection data for systems serving 100 or fewer people
as follows.

•	Systems with listed populations less than 25 had their populations revised to 25 to conform to
the definition of a PWS under SDWA.

•	If the number of people per connection (population/number of connections) was less than one
or greater than five, then the number of connections for that system was adjusted to be the
population of that system divided by the mean number of connections per person for its size
and source water category. This data cleaning step is consistent with the approach used under
the 2021 LCRR analysis.

The total number of connections per size category and CCT status derived from the SDWIS/Fed 4th
quarter 2020 data are shown in Exhibit 3-14.

Exhibit 3-14: Total Number of Service Lines by System Size and CCT Status based on

SDWIS/Fed 4th Quarter 2020 Data

System Size
(Population Served)

Total Number of Connections

Number of
CWSs

Average Number of
Connections per CWS

with CCT

without CCT

Total

A

B

C = A+B

D

Q

u"

II

LU

<100

37,942

277,850

315,792

11,732

27

101-500

358,968

1,247,941

1,606,909

15,084

107

501-1,000

519,552

1,098,346

1,617,898

5,330

304

1,001-3,300

2,660,788

3,382,584

6,043,372

7,967

759

3,301-10,000

6,198,589

4,984,027

11,182,616

5,026

2,225

10,001-50,000

18,022,965

8,562,958

26,585,923

3,374

7,880

50,001-100,000

13,030,588

298,917

13,329,505

571

23,344

100,001-1,000,000

32,908,364

245,317

33,153,681

421

78,750

>1,000,000

16,084,234

0

16,084,234

24

670,176

Source: File "Service Line Characterization using DWINSA_Final.xlsx," worksheet "Systems and Connections."

Note: Based on connection data from SDWIS 4th quarter 2020 frozen dataset, current through December 31, 2020.
Adjusted for systems serving < 100 people if the reported population was less than 25 or if the number of people
per connection was less than 1 or greater than five.

Exhibit 3-15 through Exhibit 3-17 summarize the results of Steps 4 through 6, showing the known and
projected lead content service lines (percent and number) for Categories 1, 3, and 4 and the
corresponding SafeWater LCR variable names. Exhibit 3-18 shows the total known and projected lead
content service lines per category and the national total estimate of 9.8 million potential lead content
lines.

Final LCR! Economic Analysis

3-38

October 2024


-------
Exhibit 3-15: Known and Projected Lead Content Service Lines in Category 1 Systems

System Size
(population
served)

SDWIS 2020,
Adjusted
Connection
Data

Percent
of CWSs
in

Category
1

Lead Content SL in Category 1
Systems

Projected Lead Content SL in
Category 1 Systems





Total
Number of
CWSs

Average
No. of SL
/ System



Percent SL
that are
Known
(i.e., lead
or non-
lead)

Percent of
Known SL
that are

Lead
Content

Number of SL
that are Lead
Content in
Category 1
Systems

Percent SL
that are
Unknown

Percent of
Unknown
SL

Projected
to be
Lead
Content

Number of
Projected

Lead
Content SL
in Category
1 Systems

Total Known
and
Projected
Lead Content
SL in
Category 1
Systems







pjsl

perc_lsl_kn
own

perc_lsl_kno
wn lead





perc_unkn
own lead







A

B

C

D

E

F =

A*B*C*D*E

0

II

l->

1

o

H

1 =

A*B*C*G*
H

J = F+l

<100

11,732

27

1.9%

84.4%

19.8%

992

15.6%

1.4%

13

1,005

101-500

15,084

107

1.9%

84.4%

19.8%

5,046

15.6%

1.4%

65

5,112

501-1,000

5,330

304

1.9%

84.4%

19.8%

5,081

15.6%

1.4%

66

5,147

1,001-3,300

7,967

759

1.9%

84.4%

19.8%

18,979

15.6%

1.4%

245

19,224

3,301-10,000

5,026

2,225

20.6%

56.2%

41.5%

537,272

43.8%

11.2%

113,057

650,329

10,001-50,000

3,374

7,880

29.8%

61.3%

31.0%

1,508,873

38.7%

12.6%

386,474

1,895,347

50,001-100,000

571

23,344

35.7%

72.9%

21.9%

757,931

27.1%

12.5%

160,810

918,741

100,001-
1,000,000

421

78,750

34.3%

76.0%

15.0%

1,293,766

24.0%

11.5%

313,603

1,607,369

>1M

24

670,176

43.1%

69.4%

12.9%

620,086

30.6%

10.8%

228,891

848,977

Total

49,529









4,748,026





1,203,224

5,951,250

Source: 7th DWINSA, Derivation file "Service Line Characterization using DWINSA_Final.xlsx", worksheet "Category 1 Characterization."

Notes:

Category 1 includes systems with any known lead content. SL = service line; CWS = Community Water System.

A, B: Exhibit 3-14, Columns D and E, respectively.

C: Exhibit 3-10, Column C, combined for systems with and without CCT. Note that this is for Category 1 systems only and represents the percentage of systems
rather than the percentage of service lines.

Final LCRI Economic Analysis

3-39

October 2024


-------
D: Calculated by taking the sum of lead content and non-lead service lines in Category 1 from Exhibit 3-13, Tables 1 and 2, and dividing it by the total service
lines in Category 1 from Exhibit 3-12. Calculated separately for each size category. For instance, for systems serving 101-500, total known service lines = 13,824
+ 55,835 = 69,659 from Exhibit 3-13, Tables 1 and 2. The total number of service lines in Category 1 for systems serving 101 - 500 from Exhibit 3-12 is 82,523.
Divide 69,659 by 82,523 to estimate the percent of all service lines that are known (i.e., lead content or non-lead) = 84 percent.

E: Calculated by taking the number of lead content service lines in Category 1 from Exhibit 3-13, Table 1 and dividing it by the sum of lead content and non-lead
content service lines in Category 1 from Exhibit 3-13, Tables 1 and 2. For instance, for systems serving 101 - 500, the number of lead content service lines is
13,824. The total number of known lines is lead plus non-lead, or 13,824 + 55,835 = 69,659. The percent of known lines that are lead content equals 13,824
divided by 69,659 = 20 percent.

H: Exhibit 3-11, Column E.

F, I, and J: Estimated values by row may differ from those calculated by using exhibit inputs-because of independent rounding.

A, F, I, and J: Totals may not add due to independent rounding.

Final LCRI Economic Analysis

3-40

October 2024


-------
Exhibit 3-16: Projected Lead Content Service Lines in Category 3 Systems

System Size
(population
served)

SDWIS 2020, Adjusted
Connection Data

Percent of
Systems in
Category 3

Percent of Unknown SL
Projected to be Lead
Content

Number of
Projected Lead
Content SLs in
Category 3
Systems

Total
Number of
CWSs

Average No.
of SL/
System

p_lsl_unknown

perc_unknown_lead

A

B

C

D

e=a*b*c*d

<100

11,732

27

41.4%

1.4%

1,817

101-500

15,084

107

41.4%

1.4%

9,244

501-1,000

5,330

304

41.4%

1.4%

9,307

1,001-3300

7,967

759

41.4%

1.4%

34,764

3,301-10,000

5,026

2,225

32.9%

11.2%

413,360

10,001-50,000

3,374

7,880

26.5%

12.6%

890,095

50,000-100,000

571

23,344

25.1%

12.5%

417,435

100,000-1,000,000

421

78,750

28.3%

11.5%

1,075,833

>1,000,000

24

670,176

25.5%

10.8%

442,055

Total

49,529







3,293,910

Acronyms: CWS = community water system; SL = service line.

Notes:

General: Category 3 is systems with only all unknown service lines.

A, B: Exhibit 3-14, Columns D and E, respectively.

C: Exhibit 3-10, Column E, combined for with and without CCT. Note that this is for Category 3 systems only and represents the percentage of
systems rather than the percentage of service lines.

D: Exhibit 3-11, Column E.

E: Estimated values by row may differ from those calculated by using exhibit inputs-because of independent rounding.

A and E: Totals may not add due to independent rounding.

Final LCRI Economic Analysis

3-41

October 2024


-------
Exhibit 3-17: Projected Lead Content Service Lines in Category 4 Systems

System Size
(Population
Served)

SDWIS 2020, Adjusted
Connection Data

Percent of
Systems in
Category 4

Percent of SL that
are Unknown in
Category 4

Estimated
Number of
Unknown Service
Lines in Category
4 Systems

Percent of Unknown
Service Lines that
are Projected to be
Lead Content

Number of
Projected Lead
Content SL in
Category 4 Systems

Total
Number of
CWSs

Average
No. of SL/
System

p_lsl_nolea
d_un known

perc_lsl_nolead_ un
known_unknown

perc_unknown_lead

A

B

C

D

E = A*B*C*D

F

G = F*E

<100

11,732

27

4.9%

33%

5,128

1.4%

71

101-500

15,084

107

4.9%

33%

26,093

1.4%

363

501-1,000

5,330

304

4.9%

33%

26,271

1.4%

365

1,001-3,300

7,967

759

4.9%

33%

98,131

1.4%

1,364

3,301-10,000

5,026

2,225

11.4%

33%

421,262

11.2%

47,278

10,001-50,000

3,374

7,880

14.9%

33%

1,305,766

12.6%

164,700

50,001-100,000

571

23,344

15.0%

33%

660,065

12.5%

82,312

100,001-1,000,000

421

78,750

13.7%

33%

1,494,393

11.5%

171,189

>1M

24

670,176

13.7%

33%

728,521

10.8%

78,550

Total

49,529







4,765,630



546,192

Acronyms: CWS = community water system; SL = service line.

Notes:

General: Category 4 is systems with a mix of non-lead and unknown service lines.

A, B: Exhibit 3-14, Columns D and E, respectively.

C: Exhibit 3-10, Column F, combined for with and without CCT. Note that this is for Category 4 systems only and represents the percentage of systems rather
than the percentage of service lines.

D: Calculated by taking the number of unknown service lines in Category 4 from Exhibit 3-13, Table 3 and dividing it by the total number of service lines in
Category 4 from Exhibit 3-12. Calculated separately for each system size category. For instance, for systems serving 501 - 1,000, the unknown service lines are
63,682. The total number of service lines for Category 4 systems serving 501 - 1000 from Exhibit 3-12 is 192,974. Divide 63,682 by 192,974 to estimate the
percent of service lines that are unknown = 33%. Note that in Step 3 of the analysis, the EPA made a simplifying assumption for the percent of service lines that
are unknown for this category. The remaining unknown service lines were allocated to Category 1.

F: Exhibit 3-11, Column E.

Final LCRI Economic Analysis

3-42

October 2024


-------
E and G: Estimated values by row may differ from those calculated by using exhibit inputs-because of independent rounding.
A, E, and G: Totals may not add due to independent rounding.

Final LCRI Economic Analysis

3-43

October 2024


-------
Exhibit 3-18: Total Number of Known and Projected Lead Content Service Lines by Category

System Size (Population
Served)

Category 1

Category 2

Category 3

Category 4

Total

Number of
Known Lead
Content SLs

Number of
Projected Lead
Content SLs

Number of Known or
Projected Lead
Content SL

Number of
Projected Lead
Content SLs

Number of
Projected Lead
Content SLs

A

B

C

D

E

F = SumA:E

<100

992

13

0

1,817

71

2,893

101-500

5,046

65

0

9,244

363

14,718

501-1,000

5,081

66

0

9,307

365

14,819

1,001-3,300

18,979

245

0

34,764

1,364

55,352

3,301-10,000

537,272

113,057

0

413,360

47,278

1,110,967

10,001-50,000

1,508,873

386,474

0

890,095

164,700

2,950,142

50,001-100,000

757,931

160,810

0

417,435

82,312

1,418,488

100,001-1,000,000

1,293,766

313,603

0

1,075,833

171,189

2,854,391

>1M

620,086

228,891

0

442,055

78,550

1,369,582

Total

4,748,026

1,203,224

0

3,293,910

546,192

9,791,351

Acronyms: CWS = community water system; SL = service line.

Notes:

General: Category 1 are systems with any lead content service lines; Category 2 are systems with all non-lead service lines; Category 3 are systems with all
unknown service lines, and Category 4 are systems with a mix of non-lead and unknown service lines.

A: Exhibit 3-15, Column F.

B: Exhibit 3-15, Column I.

C: Zero because these systems have all non-lead service lines.

D: Exhibit 3-16, Column E.

E: Exhibit 3-17, Column G.

F: Estimated values by row may differ from those calculated by using exhibit inputs-because of independent rounding.

A, B, D, E, and F: Totals may not add due to independent rounding.

Final LCRI Economic Analysis

3-44

October 2024


-------
Step 7: Characterize the percent of lead content service lines that are full LSLs, partial LSLs, GRR service
lines, lead connectors, and galvanized previously downstream of lead connectors.

As described earlier in this section, the 7th DWINSA collected detailed information on lead content
service lines, asking respondents to classify lead content lines as LSLs, lead connectors, GRR service lines,
and galvanized previous downstream from lead connectors. The agency used the 7th DWINSA survey
results to estimate, by system size strata, the percent of total lead content service lines (known +
projected), from Exhibit 3-18 that fall into each of these lead content categories.46 The EPA made one
additional distinction for LSLs: For the purposes of this analysis, the EPA estimated that 80 percent of
LSLs are full service lines and 20 percent are partial service lines, meaning that part of the service line
(e.g., either the service line on the system-side property or the customer side property) had been
replaced. This estimate is consistent with the approach used in the Final 2021 LCRR EA (USEPA, 2020a)
as summarized below. The percentages including the estimated 80/20 split between full and partial LSLs
are presented in Exhibit 3-19.

The EPA recognized that many systems have been replacing LSLs as a component of infrastructure
improvement programs The EPA assumed a one percent replacement rate from the promulgation of the
original LCR in 1991 to 2020 when the LCRR became final or 29 years. Based on a survey by Black &
Veatch (2004), the EPA assumed that approximately 72 percent of those replacements were partial
LSLRs. Thus, the EPA estimated that approximately 20 percent (29 years * 1 percent LSLR per year * 72
percent of replacements are partials) of LSLs have already been partially replaced on the utility side prior
to the 2021 LCRR (USEPA, 2020a).

For the final LCRI, the compliance date is projected to be 2027 which may result in additional partial
replacements between 2020 through 2027. However, the EPA has been strongly encouraging full
replacements, and many funding sources have become available for customer side replacements (for a
list of funding sources, see EPA's website: https://www.epa.gov/ground-water-and-drinking-
water/funding-lead-service-line-replacement). In the absence of additional data on partial vs. full
replacements, the EPA continues to assume an estimated 20 percent of LSLs are partial LSLs for this EA.

The EPA applied these percentages to the adjusted SDWIS/Fed 4th quarter 2020 data estimate of lead
content service lines for each system size category, both known and projected, to determine the
number of each type of lead content service line in each size category. The results of this
characterization are shown in Exhibit 3-19 below. Note that the counts of service lines include known
lead content service lines from Category 1, and projected lead content service lines in Categories 1, 3,
and 4.

46 As previously noted in this section, the DWINSA one-time update used combined categories for galvanized lines.
To estimate the percent of galvanized previously downstream of lead connectors separately from GRR, the EPA
used the proportions from the original 7th DWINSA dataset.

Final LCRI Economic Analysis

3-45

October 2024


-------
Exhibit 3-19: Characterization of Lead Content Service Lines in CWSs

System Size
(Population
Served)

Total
Known and
Projected
Lead
Service
Lines

Full LSLs

Partial LSLs

Lead Connectors

GRR

Galvanized
Previously
Downstream of
Connectors

Percent

Number

Percent

Number

Percent

Number

Percent

Number

Percent

Number

ppjsljull

pp_lsl_partial

pp_lsl_connector

pp_lsl_grr

pp_lsl_gpdlc

A

B

C = A*B

D

E = A*D

F

G = A*F

H

1 = A*H

J

K = A*J

<100

2,893

65.3%

1,888

16.3%

472

10.5%

304

0.6%

17

7.3%

211

101-500

14,718

65.3%

9,607

16.3%

2,402

10.5%

1,549

0.6%

89

7.3%

1,071

501-1,000

14,819

65.3%

9,673

16.3%

2,418

10.5%

1,559

0.6%

90

7.3%

1,079

1,001-3,300

55,352

65.3%

36,131

16.3%

9,033

10.5%

5,824

0.6%

335

7.3%

4,030

3,301-
10,000

1,110,967

28.4%

315,450

7.1%

78,863

49.5%

549,574

5.7%

63,457

9.3%

103,623

10,001-
50,000

2,950,142

43.3%

1,276,476

10.8%

319,119

25.4%

750,192

8.4%

247,567

12.1%

356,789

50,001-
100,000

1,418,488

47.6%

675,478

11.9%

168,869

26.6%

376,687

6.3%

89,568

7.6%

107,886

100,001-
1,000,000

2,854,391

62.5%

1,782,775

15.6%

445,694

12.2%

348,571

4.8%

137,612

4.9%

139,740

>1M

1,369,582

69.1%

946,236

17.3%

236,559

11.7%

159,983

0.8%

11,227

1.1%

15,577

Total

9,791,351



5,053,714



1,263,429



2,194,242



549,962



730,005

Notes: See Exhibit 3-8 for details on which categories reported in the 7th DWINSA were included in each service material category shown here. Totals may not

add due to independent rounding.

Source:

B,	D, F, H, and J: Generated using the DWINSA LSL Allocation Model. Assumes that 20 percent of all LSLs are partial based on infrastructure replacement rate of
1 percent per year between 1991 when the LCR became finalized and 2020. The relationship between GRR and galvanized previously downstream of a lead
connector was determined using the original 7th DWINSA survey dataset because the categories were combined in the one-time update.

C,	E, G, I, and K: Estimated values by row may differ from those calculated by using exhibit inputs-because of independent rounding.

A, C, E, G, I, and K: Totals may not add due to independent rounding.

Final LCR! Economic Analysis

3-46

October 2024


-------
For purposes of this EA, the EPA estimates that there are approximately 9.8 million lead content service
lines (i.e., LSLs, GRR service line, lead connectors, and galvanized previously downstream of lead
connectors). As previously explained, the LCRI focuses on mandatory replacement of lead and GRR
service lines. This rule requires the replacement of lead connectors when encountered during normal
operations like main replacement and does not require replacement of galvanized lines downstream of
lead connectors. Therefore, using the values in Exhibit 3-19 the EPA estimates that the total number of
lead and GRR service lines that will require replacement under the LCRI is approximately 6.9 million (5.1
million full LSLs, 1.3 million partial LSLs, and 0.5 million GRR service lines).

The EPA recognizes that the approach for estimating the total number of lead content service lines for
this EA is slightly different than the approach used to estimate the total number of lead content service
lines for the DWINSA Allocation Model, and that these differences produce a different estimated
number of total lead content service lines. Exhibit 3-20 compares the total number of known and
potential lead content service lines from this Final LCRI EA and the 7th DWINSA Fact Sheet. The totals are
higher for this EA at 9.8 million lead content service lines compared to the DWINSA Fact Sheet at 9.0
million lead content service lines (USEPA, 2024a). The exhibit also describes assumptions for different
elements of the analysis, noting where they are the same or different. The largest difference in the two
datasets is the total connection data as reported in SDWIS/Fed, which is multiplied by percentages of
known and potential lead content to produce total service line counts. This EA used values from
SDWIS/Fed 4th quarter 2020 frozen dataset, adjusted upward using decision rules as described in Step 6
above. The 7th DWINSA Fact Sheet value is based on 2nd quarter 2019 SDWIS/Fed data without similar
adjustments but updated based on survey responses. Note that the percent of systems with lead
content service lines and percent of connections that are lead content is similar between this EA and the
DWINSA Allocation Model.

Exhibit 3-20: Similarities and Differences in Service Line Material Data Analysis for the Final
LCRI Economic Analysis and the DWINSA LSL Allocation Model

Element of the Service Line
Material Analysis

Final LCRI Economic Analysis

DWINSA LSL Allocation Model1

Total Number of Known and
Potential Lead Content
Service Lines

9.8 million

9.0 million

Source Data

Uses responses from both the 7th
DWINSA and one-time update.

Same.

Types of Systems Included

Includes CWSs from all 50 States,
DC, PR, and the territories. Service
line material data from ANV and
Al systems were not included.

Same.

Use of Sampling Weights

Uses system sampling weights to
estimate national totals.

Uses system sampling weights
to estimate State totals and

Final LCRI Economic Analysis

3-47

October 2024


-------
Element of the Service Line
Material Analysis

Final LCRI Economic Analysis

DWINSA LSL Allocation Model1





adds them together for
national totals.

Unreported vs Unknown

Groups unknown and unreported
together.

Maintains unknown and
unreported as separate
categories.

Methodology for Estimating
the Number of Service Lines
by Material Type

Used State ratios to estimate the
proportion of service lines that are
known lead content, non-lead and
unknown (which includes
unreported) by 9 system size
categories.

Used decision rules to estimate
the proportion of lead content,
non-lead, and unknown in 4
system categories.

Uses State ratios to estimate
the proportion of service lines
that are known lead content,
non-lead, unknown, and
unreported for all system size
categories.

Projections

Projections of unknown and
unreported service lines that are
found to be lead content are
based on the percent of known
service lines (both lead and non-
lead) that are known lead-content.

Same.

Classification of Lead
Content Service Line
Material

Uses five categories of lead
content service lines:

•	Full LSL

•	Partial LSL

•	GRR Service Lines

•	Lead Connectors

•	Galvanized previously downstream
of lead connector

Groups all lead content service
lines.

Use of SDWIS/Fed data

Uses 4th quarter 2020 SDWIS/Fed
data with adjustments for service
connections for systems serving <
100 people.

Uses 2nd quarter 2019 data
from SDWIS/Fed, updated
where appropriate with system
responses to the 7th DWINSA
survey

Acronyms: CWS = community water system; DC = District of Columbia, PR = Puerto Rico, ANV = Alaska Native
Village, Al = American Indian, LSL = lead service line, SDWIS = Safe Drinking Water Information System.
Note:1 See USEPA 2023d for a description of the model used to estimate the number of lead service lines.

Final LCRI Economic Analysis

3-48

October 2024


-------
3.3.4.1.3 Discussion of Data Limitations and Uncertainty

There are analytical uncertainties associated with the use of the 7th DWINSAto characterize systems
containing lead content service lines and the proportions of service lines in those systems that are lead.
While efforts were made to ensure that the sample of systems surveyed was representative of the
population of all systems, such as the assignment of system weights and their incorporation into the
analysis of survey results, the representativeness of these results may have been diminished due to
inaccurate and non-responses to the survey. For example, systems may not know the material make-up
of the service lines in their service area. Furthermore, the presence or lack of lead lines may affect a
systems' willingness to respond to the survey. The EPA mitigated the potential error associated with
these survey responses by letting systems include lines of unknown material, comparing responses to
the total connections reported in SDWIS/Fed, and providing the states with the opportunity to review
the responses.

To estimate national costs and benefits of the final LCRI, the EPA made several assumptions regarding
the unknown and unreported service line materials in the 7th DWINSA results:

•	The proportion of unknown service lines that are lead is the same as the proportion of known
service lines that are lead. This estimate was based on the proportion of systems with at least
one known lead content service line - derived from the results of the survey - and the number
of lead content service lines nationally - projected from the results of the survey using State
ratios and connection data from SDWIS/Fed.

•	The percent of unknown service lines that are found to be lead is the same regardless of system
category.

•	All systems with unknown content service lines are likely to have at least one LSL.

These assumptions and the following additional assumptions could underestimate or overestimate the
costs and benefits of the final LCRI:

•	The EPA made simplifying assumptions about the proportion of unknown and non-lead service
lines in Categories 1 (systems with any lead content) and 4 (systems with non-lead and
unknowns) and the percent of lead service lines that are partial as opposed to full.

•	There is uncertainty associated with using connection data from the SDWIS/Fed 4th quarter data
freeze to apply to DWINSA-derived percentages to estimate the total lead content service lines
(known + potential).

3.3.4.2 LSL Inventory for NTNCWSs

Information comparable to the DWINSA LSL Allocation Model for CWSs described in Section 3.3.4.1 has
not been collected on the occurrence of lead content service lines in NTNCWSs47. Therefore, the EPA
used the approach from the Final 2021 LCRR EA (USEPA, 2020a) to estimate the occurrence of LSLs in

47 The DWINSA LSL Allocation Model included a limited number of responses for NTNCWSs, but the data were
insufficient to draw conclusions regarding the extent of lead service lines for NTNCWSs.

Final LCRI Economic Analysis

3-49

October 2024


-------
NTNCWSs. Note that the EPA assumes that the NTNCWSs have no stand-alone lead connectors and no
GRR service lines since NTNCWSs are unlikely to replace only the upstream portion of an LSL and leave a
galvanized service line in place, because the full service line is likely on the property of the system, and
galvanized lines have a high failure rate given age and disturbance like that produced by removing a
section of LSL.48

This section outlines the EPA's approach for estimating the number of NTNCWSs with LSLs (Section
3.3.4.2.1) and number of LSLs (Section 3.3.4.2.2). A discussion of data limitations and uncertainties is
provided in Section 3.3.4.2.3.

3.3.4.2.1 Number of NTNCWSs with LSLs

In August 2017, the EPA disseminated a questionnaire to nine States regarding the burden and cost
associated with the National Drinking Water Advisory Council's (NDWAC's) recommendation to require
all systems to develop a comprehensive LSL inventory and to expand the definition of an LSL to include
lead connectors even if the service line is not made of lead.49 The questionnaire included questions on
the estimated number or percentage of NTNCWSs with one or more LSLs and the total number of LSLs in
NTNCWSs. States were selected for geographical diversity, known occurrence of LSLs in CWSs, and
active LSLR projects. The EPA received responses from seven States. Four States did not provide any
estimates. The remaining three States provided estimates ranging from zero to five percent. Exhibit 3-21
below provides a summary of the seven States' responses.

Exhibit 3-21: Summary of State Responses Regarding the Percentage of NTNCWSs with LSLs

Response

Number of States with this Response

Unknown/Information not readily available

3

Unknown but expected to be very low

1

Unknown but expected to be 0

1

Estimated to be 0 - 5 percent

1

Estimated to be <5 percent (gross estimate)

1

Estimated to be >5 percent

0

Total

7

Source: A copy of the questionnaire and each State's response is available in the docket at EPA-HQ-OW-
2022-0801 at www.regulations.gov.

Due to the uncertainty of the responses and the respondents being in States with known LSLs in CWSs,
the EPA used the midpoint of the range reported by States to estimate the number of NTNCWSs that
had LSLs. Specifically, the EPA assumed 2.5 percent (corresponds to SafeWater LCR model data variable,
p_lsl) or 435 NTNCWSs have LSLs. The EPA further assumed that systems without CCT are less likely to
have LSLs because they would have installed CCT if they had sustained lead ALEs. Thus, the EPA assumed
all 435 NTNCWSs with LSLs are those with CCT. Exhibit 3-22 indicates the estimated number of
NTNCWSs with and without LSLs for systems with and without CCT.

48	Florida Department of State (2010). Rule: 25-30.140. Florida Administrative Code & Administrative Register.
Retrieved July 24, 2023, from https://www.flrules.org/gatewav/ruleNo.asp?id=25-30.140

49	A copy of the questionnaire is available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Final LCR! Economic Analysis

3-50

October 2024


-------
Exhibit 3-22: Estimated Number of NTNCWSs With and Without LSLs by CCT Status

System Size

Number of NTNCWSs
by CCT Status

Estimated Number
with LSLs
by CCT Status

Estimated Number
without LSLs by CCT
Status



With CCT

No CCT

Total

With CCT

No CCT

With CCT

No CCT









D = C*0.025









A

B

C = A+B

pjsl

o

II

LU

F = A-D

LU

i

CO
II

o

<100

729

7,659

8,388

210

0

519

7,659

101-500

869

5,511

6,380

160

0

710

5,511

501-1,000

277

1,301

1,578

39

0

238

1,301

1,001-3,300

188

683

871

22

0

166

683

3,301-10,000

34

128

162

4

0

30

128

10,001-50,000

5

32

37

1

0

4

32

50,001-100,000

1

0

1

0

0

1

0

100,001-1M

1

0

1

0

0

1

0

> 1M

0

0

0

0

0





Total

2,104

15,314

17,418

435

0

1,669

15,314

Notes:

General:

1.	Only systems with CCT are assumed to have LSLs. No NTNCWSs serve more than 1 million people, and the EPA estimates that the two NTNCWSs
serving 50,001 - 1 million which are airports do not have LSLs

2.	Values by size category may not equal total values do to independent rounding.

A, B: Exhibit 3-7.

D, E: Estimate of 2.5 percent based on information from three States regarding the percentage of NTNCWSs in their State with any LSLs (see Exhibit
3-21). As a simplifying assumption, the EPA assumed that all 2.5 percent of NTNCWSs with LSLs are those with existing CCT. Also, the EPA assumed that
the two NTNCWSs serving > 50,000 which are large airports and are unlikely to have LSLs and assigned them to the no LSL category.

Final LCRI Economic Analysis

3-51

October 2024


-------
3.3.4.2.2 Number of LSLs in NTNCWSs

Two States provided an estimate of the number of LSLs in NTNCWSs in response to the August 2017
questionnaire. One estimated the number to be between zero and five. The second estimated between
zero and 50 LSLs and noted that the majority of their NTNCWSs (67 percent) have five or fewer
connections. Due to the uncertainty of the responses and representativeness of the data, the EPA did
not use the responses to the State questionnaire to develop the national number of LSLs in NTNCWSs.
Instead, consistent with the 2021 LCRR, the EPA used the following approach.

1.	Determine the median number of service connections from SDWIS/Fed for each of the NTNCWS
size categories serving one million or fewer people.

2.	For systems with LSLs:

a.	Assume 100 percent of service connections are lead when the median number of service
connections is 10 or fewer.

b.	Assume NTNCWSs with more than 10 service connections have experienced expansion over
time resulting in service lines of different materials. For these systems, the EPA developed a
range, with a minimum of 50 percent and a maximum of 100 percent of service lines
assumed to be lead.

Exhibit 3-23 provides the estimated total number of LSLs for each system size category. The
corresponding SafeWater LCR model data variable used to estimate costs in Chapter 4 is provided in red
italics. Note that the minimum is the same as the maximum except for the three size categories that
serve populations of 10,001 - 1 million where the median number of service connections is above 10. As
previously stated, the EPA assumes that LSLs in NTNCWSs are limited to the subset of NTNCWSs with
CCT.

Exhibit 3-23: Number of LSLs in NTNCWSs with CCT by Size Category

System Size
Category

Number of
Systems with
LSLs (assumes
2.5%)

Median
Number of

Service
Connections

Estimated Percent

of Service
Connections that
Are LSLs

Total Estimated
Number of LSLs per
Size Category

perc_lsl_known

Minimum

Maximum

Minimum

Maximum

A

B

C

D

e=a*b*c

F=A*B*D

<100

210

1

100%

100%

210

210

101-500

160

1

100%

100%

160

160

501-1,000

39

1

100%

100%

39

39

1,001-3,300

22

2

100%

100%

44

44

3,301-10,000

4

7

100%

100%

28

28

10,001-50,000

1

13

50%

100%

6

12

50,001-100,000













100,001-1,000,000













>1,000,000













Final LCR! Economic Analysis

3-52

October 2024


-------
System Size
Category

Number of
Systems with
LSLs (assumes
2.5%)

Median
Number of

Service
Connections

Estimated Percent

of Service
Connections that
Are LSLs

Total Estimated
Number of LSLs per
Size Category

perc_lsl_known

Minimum

Maximum

Minimum

Maximum

A

B

C

D

e=a*b*c

F=A*B*D

Total

435







492

504

Source: "Service Line Characteristics Using DWINSA_Final.xlsx," worksheet, "NTNCWS Lead Service Line Status."

Notes:

General:

1.	Only systems with CCT are assumed to have LSLs. No NTNCWSs serve more than 1 million people, and the EPA
estimates that the two NTNCWSs serving 50,001 - 1 million which are airports do not have LSLs

2.	Values by size category may not equal total values do to independent rounding.

Sources:

A: Exhibit 3-22 Column D.

B: Based on SDWIS/Fed 4th Quarter 2020 connection data.

C, D: For systems with LSLs, the EPA assumed 100 percent of service connections are lead when the median
number of service connections was < 10. For NTNCWSs with > 10 service connections, the EPA assumed that
service connections have been laid over a period of time and may be composed of different materials. Thus, for
these systems, the EPA assumed a minimum of 50 percent and maximum of 100 percent of service lines are lead.

As noted previously, the EPA assumes that all lead content service lines in NTNCWSs are lead pipe. The
EPA further assumes that all LSLs are full LSLs because NTNCWSs would not do partial replacement since
the entire service line is likely fully on the system's property. The EPA assumes that NTNCWSs have no
stand-alone lead connectors and no GRR service lines because they would not have replaced a portion
of the service line (the upstream portion) and left the galvanized portion in place.

3.3.4.2.3 Discussion of Data Limitations and Uncertainty

There is a high degree of uncertainty in using the 2.5 midpoint of the range as the estimated percentage
of NTNCWSs with LSLs is based on survey results from three States. This uncertainty could result in an
under- or overestimate of national costs and benefits of the final LCRI. The EPA assumed that all service
lines would be lead in those NTNCWSs with LSLs serving 10,000 or fewer based on the reported median
number of service connections in SDWIS/Fed for each size category. This may result in an overestimate
of costs and benefits based on the accuracy of the service connection information. However, the impact
of these uncertainties is expected to be small due to the low estimated number of NTNCWSs with LSLs.

3.3.4.3 State Service Line Replacement Regulations

In order to estimate future LSLR in the baseline over the period of regulatory analysis (35-years) driven
by factors other than potential service line replacement requirements under the final LCRI, the EPA
evaluated the extent of State regulations related to the replacement of LSLs. Below is a summary of four
existing State regulations:

• Illinois: Requires replacement rates for CWSs based on the number of LSLs, which includes lead
connectors, in the system's final inventory and replacement plan. The service line replacement

Final LCRI Economic Analysis

3-53

October 2024


-------
rate ranges from 2-7 percent annually, with timelines ranging from 15 years to 50 years for
completion. The final replacement plan is due April 15, 2027.50

•	Michigan: Beginning a year after the preliminary inventory was completed, which was due on
January 1, 2021, CWSs and NTNCWSs must replace 5 percent of LSLs that include lead
connectors and affected galvanized service lines annually, not to exceed 20 years (by 2041). If
the system exceeds the lead AL, it must replace 7 percent of LSLs annually.

•	New Jersey: CWSs must replace all LSLs, which include galvanized SLs and lead connectors,
within 10 years or no later than July 22, 2031 (i.e., annual replacement of at least 10 percent of
all known LSLs).

•	Rhode Island: CWSs and NTNCWSs must replace all public and private LSLs within 10 years of the
June 24, 2023 effective date of the law or by June 24, 2033. This corresponds to an annual
replacement of 10 percent of LSLs.

The EPA estimates that these four States have approximately 1.8 million LSLs, which is equivalent to
about one-fifth of LSLs in the country. However, only the New Jersey and Rhode Island laws require full
replacement of all LSLs by 2034. Therefore, the EPA does not assume that all 1.8 million service lines in
these four States would be replaced in the baseline without the 2021 LCRR or the final LCRI. Rather, the
EPA estimates that 451,000 service lines would be replaced over the 35-year analysis period because of
these State laws.

In addition to these four States, some States or municipalities have voluntary or goal-based programs to
replace all LSLs within the next 10 or more years. For example, the preamble section V.B.8 mentions
programs in Minnesota to replace all LSLs within 10 years and in Washington to replace all LSLs within
15 years. Because these are not legal requirements, the EPA does not include them in its estimate of the
number of LSLs that would be replaced in the baseline.

See Chapter 4, Section 4.3.4.3 for a discussion of how these regulations were considered when
estimating service lines replacement costs for the final LCRI EA.

3.3.5 Lead and Copper Tap Levels

The analyses described in this section draw from multiple sources to characterize baseline water quality,
including lead and copper levels at customers' taps. Lead 90th percentile data were obtained from
SDWIS/Fed, along with information on systems' CCT status. The EPA also used information from 13
States, Region 9 Tribes, and a web search of individual system LSLR programs to identify systems with
LSLs. As previously discussed, SDWIS/Fed does not identify which systems have LSLs, only those that are
required to initiate LSLR.

The remainder of this section is organized as follows:

50 For the required replacement rate based on number of lead service lines, see Public Act 0613 102ND GENERAL
ASSEMBLY (ilga.gov)

Final LCRI Economic Analysis

3-54

October 2024


-------
•	Section 3.3.5.1 explains the derivation of the percentage of systems that fall into one of five
classifications based on their lead 90th percentile level as a function of LSL and CCT status during
the first year of implementation of both the 2021 LCRR and the final LCRI.

•	Section 3.3.5.2 describes the EPA's approach for determining the likelihood a system with an
ALE will have two lead ALEs in a five year period, and the likelihood that those systems (with two
lead ALEs in a five-year period) will have at least one additional lead ALE within a five-year
period (i.e., has multiple ALEs) under the final LCRI.

•	Section 3.3.5.3 provides the likelihood of an individual lead sample being greater than the lead
AL of 10 ng/L under the final LCRI.

•	Section 3.3.5.4 provides the likelihood that a system exceeds the copper AL of 1.3 mg/L but not
the lead AL.

3.3.5.1 Percent of Systems by Lead 90th Percentile Classification

A system's lead 90th percentile level is an important factor in determining a system's requirements. For
the purposes of estimating the incremental costs of the final LCRI relative to the 2021 LCRR (see Chapter
4) and the potential alternative lead AL options (see Chapter 8), the EPA first estimated the proportion
of systems that would be placed into the following five lead 90th percentile level classifications under the
baseline (2021 LCRR) and the final LCRI:

•	Lead 90th percentile (P90) < 5 ng/L

•	5 ng/L < P90 < 10 ng/L

•	10 ng/L < P90 < 12 ng/L

•	12 ng/L < P90 < 15 ng/L

•	P90 > 15 ng/L

Sections 3.3.5.1.1 and 3.3.5.1.2 detail the EPA's approach for the 2021 LCRR and final LCRI,
respectively.51 Section 3.3.5.1.3 provides a discussion of the data limitations and uncertainties
associated with these estimations.

3.3.5.1.1 Baseline (2021 LCRR)

Under the 2021 LCRR, which is the baseline scenario for the incremental costs analysis of the final LCRI,
the EPA modified the sampling protocol for systems with LSLs to require all samples to be collected from

51 For the 2021 LCRR economic analysis, the EPA used a similar approach to place systems into one of three lead
classifications that corresponded to no lead ALE or trigger level exceedance (TLE) (P90 < 10 ng/L), TLE (10 ng/L <
P90 < 15 ng/L, and lead ALE (P90> 15 ng/L). However, the EPA used 2016 SDWIS/Fed data for the 2021 LCRR
economic analysis. See Chapter 4, Section 4.3.5.1 of the 2021 LCRR economic analysis for a complete description of
the EPA's approach.

Final LCRI Economic Analysis

3-55

October 2024


-------
sites served by LSLs, if available and to require systems to collect a fifth-liter sample at these sites in lieu
of a first-liter sample.

The estimated percent of systems in each P90 category is based on SDWIS/Fed historical 90th percentile
lead tap sample data from 2012 to 2020. The EPA recognizes that there are uncertainties in predicting
the future 90th percentile ranges from historical SDWIS/Fed data. Also, the agency recognizes that these
uncertainties could have a significant impact on estimated costs and benefits of the 2021 LCRR. To
provide a range of costs and benefits that reflects this uncertainty, the EPA generated both a "low" and
"high" estimate for the baseline conditions as detailed in the following four steps:

Step 1 - Identified "Low" and "High" 90th percentile level based on historical data: The EPA reviewed
all lead 90th percentile data from 38,348 CWSs with P90 results reported to SDWIS/Fed during 2012 and
2020 and excluded those results that were: 1) negative sample values (Maryland only) and 2) values >
1,500 ng/L, which is lOOx higher than the AL of 15 ng/L under the pre-2021 LCR and the 2021 LCRR.

From the remaining 38,339 CWSs, the EPA selected the average lead 90th percentile level between 2012
and 2020 for each system as the "low" estimate and the maximum lead 90th percentile for the "high"
estimate.

Step 2 - Designated systems by LSL status: Data were grouped according to LSL status for analysis. LSL
status for individual systems is not available in SDWIS/Fed. Therefore, the EPA compiled data from
numerous State surveys or databases, general web searches conducted during 2018 - 2021 of systems
with prior or ongoing LSLR programs, and discussions with some systems serving greater than one
million people. The EPA also used responses to the 7th DWINSA (including the one-time update). In the
case of conflicting information, the DWINSA LSL determination was prioritized. Exhibit 3-24 summarizes
the information from 8,339 systems that were assigned a "yes" / "no" LSL status based on this effort.
For additional detail on systems' LSL determination for individual States, see file
"DWINSA_StateData_LSL_Status_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

Exhibit 3-24 provides the total number of CWSs, the subset with LSLs, the subset with no LSLs, and the
percentage of CWSs with known LSL status, for each State and Region 9 tribal systems for which the EPA
has some information on the LSL status of its CWSs. In all, the number of systems with known LSL status
(8,339) represents approximately 17 percent of the total CWS inventory of 49,529, with information
collected from more than 50 percent of States.52 Column D indicates the percentage of all CWSs for
which the EPA has known LSL status information. The fact that the EPA is constrained by the available
data that represents 17 percent of total systems results in a high degree of uncertainty around these
estimated percentages for systems falling into the ALE or no ALE category. For systems serving > 1M, the
EPA obtained system-level LSL estimates from available sources (see the data summary table provided
as Exhibit B-2 in Appendix B).

52 The web searches included some LSL data from 21 States plus Washington, D.C.

Final LCR! Economic Analysis

3-56

October 2024


-------
Exhibit 3-24: Number of CWSs with LSL Determination Based on State, Tribal, DWINSA

Responses, and Web Data1



Total Number of
CWSs in
Represented
States/R9 Tribes

Number of CWSs

Number of CWSs

Percent of All CWSs

State

with "YES" LSL

with "NO" LSL

in Dataset with



determination

determination

Known LSL Status



A

B

C

D = (B+CJ/A

EPA Region 9

213

0

213

100%

Alaska

406

1

3

1%

Alabama

509

0

39

8%

Arkansas

682

5

22

4%

American Samoa

18

0

3

17%

Arizona

745

1

18

3%

California

2,878

2

1,44

5%

Colorado

909

10

32

5%

Connecticut

486

11

1

2%

District of Columbia

4

1

0

25%

Delaware

206

3

8

5%

Florida

1,616

57

25

5%

Georgia

1,731

29

32

4%

Hawaii

118

0

118

100%

Iowa

1,082

31

20

5%

Idaho

743

3

20

3%

Illinois

1,760

320

901

69%

Indiana

775

121

205

42%

Kansas

870

13

5

2%

Kentucky

381

8

11

5%

Louisiana

973

10

49

6%

Massachusetts

531

41

27

13%

Maryland

465

26

171

42%

Maine

383

2

5

2%

Michigan

1,380

201

983

86%

Minnesota

965

25

13

4%

Missouri

1,431

43

47

6%

Northern Mariana Islands

36

0

1

3%

Mississippi

1,028

4

25

3%

Montana

757

5

8

2%

North Carolina

2,001

173

1,646

91%

North Dakota

315

6

7

4%

Nebraska

598

10

9

3%

New Hampshire

708

4

13

2%

New Jersey

571

202

90

51%

Final LCRI Economic Analysis

3-57

October 2024


-------


Total Number of
CWSs in
Represented
States/R9 Tribes

Number of CWSs

Number of CWSs

Percent of All CWSs

State

with "YES" LSL

with "NO" LSL

in Dataset with



determination

determination

Known LSL Status

New Mexico

571

2

5

1%

Nevada

198

0

198

100%

New York

2,294

31

11

2%

Ohio

1,169

183

688

75%

Oklahoma

899

15

28

5%

Oregon

900

0

15

2%

Pennsylvania

1,916

50

16

3%

Puerto Rico

407

0

6

1%

Rhode Island

91

9

1

11%

South Carolina

574

4

13

3%

South Dakota

463

9

13

5%

Tennessee

456

4

19

5%

Texas

4,648

5

61

1%

Utah

502

1

10

2%

Virginia

1,089

4

8

1%

US Virgin Islands

77

0

2

3%

Vermont

411

6

11

4%

Washington

2,291

60

397

20%

Wisconsin

1,040

137

2

13%

West Virginia

433

5

23

6%

Wyoming

315

1

4

2%

TOTAL



1,894

6,445



Notes:

1	The data presented in this exhibit were compiled from numerous State surveys or data bases, web searches of
systems with prior or ongoing lead service line replacement programs (LSLR) programs, responses to the 7th
DWINSA (including the one-time update), and discussions with some systems serving greater than 1 million people.
Determinations of whether CWSs had LSLs within each State were dependent on a hierarchy of these data sources.
If the LSL status conflicts between DWINSA and information from the State inventory/web searches, the DWINSA
status was used. (For additional detail on systems' LSL determinations, see
"DWINSA_StateData_LSL_Status_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.) Note that this exhibit does not show counts of CWSs that reported "unknown" LSL status or
for which the data were insufficient for the EPA to determine the system's LSL status. Only those systems whose
LSL status was known are described here.

2	The LSL information for the State of Hawaii was extracted from an April 9, 2016 Associated Press article53 that
stated no drinking water systems in Hawaii have lead pipes.

3	The LSL information for the State of Nevada was submitted as a public comment on the proposed 2021 LCRR, by
the Nevada Division of Environmental Protection, stating that no drinking water systems in Nevada had
documented LSLs. Refer to Attachment A that is available in the docket for the 2021 LCRR at EPA-HQ-OW-2022-
0801 at www.regulations.gov.

53 Available at https://www.staradvertiser.com/2016/04/Q9/breaking-news/hawaii-tap-water-safer-from-lead-
than-other-states/.

Final LCRI Economic Analysis

3-58

October 2024


-------
4	The web search identified LSL status information for systems from 21 States plus Washington, D.C. For more
details, see file "DWINSA_StateData_LSLStatus_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

5	Note that there was overlap with some of the systems with known LSL status in individual States, systems
identified through web searches, and those that serve more than 1 million people.

Step 3 - Identified systems with reported lead 90th percentile results and known LSL status: The EPA

identified which systems had at least one reported lead 90th percentile value in SDWIS/Fed between
2012 and 2020 and known LSL status. This subset of 6,551 systems54 was used for the remainder of the
analysis described in Step 4. Of the 6,551 systems, 27 percent (1,758 systems) were identified as having
LSLs and 73 percent (4,793 systems) were identified as having no LSLs.

Step 4 - Adjust lead 90th percentile results from LSL systems: The EPA adjusted lead 90th percentile
results from systems with known LSL status using two multipliers to reflect new sampling requirements
under the 2021 LCRR.

1) The first multiplier was used to reflect the requirement for LSL systems to collect all samples
from LSL sites where possible, as opposed to the pre-2021 LCR minimum of 50 percent of
samples being collected from LSL sites. A lower multiplier (1.20) was used to adjust the "Low"
90th percentile results, and a higher multiplier (1.35) was used to adjust the "High" 90th
percentile results.

The EPA used Slabaugh et al. (2015) to derive these two multipliers. Slabaugh et al. (2015)
evaluated LCR compliance data from 17 systems over 72 tap sampling periods, comparing the
lead 90th percentile concentrations based on samples collected from all LSLs (either Tier 1
sites—single family structures, or Tier 2 sites—multiple-family residences) to lead 90th
percentiles based on samples collected from both LSL and non-lead service line sites. For the
lower multiplier, the EPA used the lead 90th percentile median value of 2.5 ng/L obtained for the
1,758 CWSs with LSLs from the set of 6,551 systems with known LSL status. This value, which is
based on all reported samples for LSL systems for 2012-2020, is assumed to be representative of
LSL systems in general, corresponding to approximately the 9th percentile of the 72 monitoring
periods based on the Slabaugh data for samples taken from All Sites in Figure 2.55 The value for
samples taken from Tier 1 and Tier 2 LSL sites only at the 9th percentile value was 3 ng/L. The
ratio of these values, 3/2.5= 1.20, was used as the lower multiplier.

For the higher value, the EPA compared the median 90th percentile for LSL only sites of 8.95 to a
median 90th percentile of 6.63 from all monitoring sites.56The ratio of the median 90th
percentiles for LSL sites compared to all sites was 8.95/6.63 = 1.35. The EPA selected this value
(1.35) as a "high" multiplier, meaning that 1.35 was applied to all 90th percentile lead values for
LSL systems to reflect the potential impacts of sampling from only LSL sites to predict the initial
lead categorization under the 2021 LCRR. This adjustment value may be biased high because the

54	With the addition of information from the DWINSA one-time supplement, the number of systems with known
LSL status increased in the final LCRI EA from the proposed LCRI EA from 6,529 to 6,551.

55	Percentiles and values taken from Figure 2 of Slabaugh et al. (2015) were read from the graph by
WebPlotDigitizer (available at: https://automeris.io/WebPlotDigitizer/).

56	The median 90th percentile values are based on data presented in Figure 2 of Slabaugh et al. (2015) and were
read from the graph by WebPlotDigitizer (available at: https://automeris.io/WebPlotDigitizer/).

Final LCRI Economic Analysis

3-59

October 2024


-------
median 90th percentile value for the 1,758 systems with LSLs (from among the 6,551 systems
used in the EPA's analysis) was only 2.5 ng/L. This 90th percentile value of 2.5 ng/L corresponded
with the 9th percentile for the 72 tap sampling periods from the Slabaugh et al. (2015) dataset
using all sampling sites. Ideally, this analysis would have been done at the system level, but the
EPA did not have access to the dataset of 72 tap sampling periods from the 17 systems. Thus,
the EPA could not confirm if the data from the tap sampling periods varied within and among
the 17 systems and if system(s) above the AL were overrepresented as nine percent of the
sampling periods exceeded the AL.

2) The EPA developed a second multiplier to simulate the expected increase in lead 90th percentile
levels resulting from the 2021 LCRR requirement for LSL systems to use the fifth-liter sample
results as opposed to the first-liter sample results from each LSL site when calculating lead 90th
percentiles. To develop this multiplier, the EPA used paired fifth- and first-liter data57 from 181
systems in Michigan at LSL locations that were collected in 2019, 2020, and 2021. Only the most
recent monitoring period of sampling data was used for systems that had multiple monitoring
periods of sampling. The EPA calculated the ratio of the fifth-liter lead 90th percentile
concentration to the first-liter lead 90th percentile concentration for each of these 181 systems.
Note that there was insufficient data to allow for calculation of separate fifth- to first-liter ratios
with respect to CCT status. Overall, the ratios ranged from 0.18 to 25.64 with a mean of 1.41.
(Note that reported concentrations below 1 ng/L and non-detects were changed to 1 ng/L).

For this analysis, the EPA applied the mean ratio of 1.41 to the low and high 90th percentile
values reported to SDWIS/Fed for 2012 - 2020 for systems having LSLs to account for LSL
systems using the fifth-liter sample results rather than first-liter sample results. The mean ratio
was not applied to the non-lead service line systems.

Step 5 - Estimated the Percentage of CWSs in Each Category: Based on Steps 1 through 4, the EPA

assigned each CWS one of five lead 90th percentile classifications by the system's LSL status.

Exhibit 3-25 presents the "low" and "high" estimates of the percentage of systems in each lead 90th
percentile category by LSL status. The "low estimate" is based on the average 90th percentile lead value
reported to SDWIS/Fed from 2012 to 2020; the "high estimate" is based on the highest 90th percentile
lead value reported to SDWIS/Fed from 2012 to 2020. Based on the "low estimate," the percentage of
systems with LSLs having an ALE, under the 2021 LCRR AL of 15 ng/L, was 9.6 percent as opposed to 2.3
percent for systems without LSLs. For the "high estimate," a much higher percentage of systems with
LSLs were classified as having an ALE than those without LSLs, 24.1 percent compared to 4.8 percent. It
is important to note that systems without LSLs can have lead sources that can contribute lead to
drinking water such as plumbing with lead solder and brass or chrome-plated brass faucets.

Minimal data were available on the LSL status for NTNCWSs. Thus, the above analysis could not be
conducted for NTNCWSs. However, an analysis was conducted to evaluate the likelihood of the
NTNCWSs' 90th percentile values falling into one of the five lead classifications without the consideration
of LSL status. The likelihoods for the NTNCWSs were very similar to those calculated for CWSs. Based on
this comparison and the lack of LSL information for NTNCWSs, the EPA assumed the same estimated

57 Paired data are first- and fifth-liter samples that are collected from the same sampling location during the same
sampling event.

Final LCRI Economic Analysis

3-60

October 2024


-------
percentages for NTNCWSs as those presented in Exhibit 3-25 for CWSs. For additional detail, see file
"Initial P90 Categorization_CWS_NTNCWS_LCR_Compare_Final.xlsx," available in the docket at EPA-HQ-
OW-2022-0801 at www.regulations.gov.

Exhibit 3-25: Percent of CWSs by Lead 90th Percentile Classification under the 2021 LCRR

Category

No LSLs

Has LSLs

Low Estimate

^ 5 ng/L

88.5%

55.5%

>5 and <10 ng/L

7.1%

24.5%

10 ng/L < P90 < 12 ng/L

1.0%

5.3%

12 ng/L < P90 < 15 ng/L

1.0%

5.2%

P90 > 15 ng/L

2.3%

9.6%

High Estimate

^ 5 ng/L

79.6%

37.8%

>5 and <10 ng/L

11.7%

25.4%

10 ng/L < P90 < 12 ng/L

2.0%

6.8%

12 ng/L < P90 < 15 ng/L

1.9%

6.0%

P90 > 15 ng/L

4.8%

24.1%

Acronyms: CWS = community water system; LCRR= Lead and Copper Rule Revisions; LSL = lead service line; P90 =
lead 90th percentile level.

Source: "Initial P90 Categorization_LCRR_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

Notes:

1	Includes CWSs with known LSL status that also reported at least one 90th percentile value to SDWIS between
2012 and 2020.

2	Totals may not add due to independent rounding.

3	Percentages have changed slightly from those presented in the Economic Analysis for the Proposed Lead and
Copper Rule Improvements (hereafter referred to as the "Proposed LCRI EA") (USEPA, 2023b) because the final
rule calculations are based on additional systems with known LSL status from the DWINSA one-time update.

3.3.5.1.2 Final LCRI

The final LCRI reduces the lead AL from 15 ng/L to 10 ng/L. For the purposes of estimating the
incremental costs of the final LCRI and potential alterative lead AL options that are presented in Chapter
8, the EPA first estimated the proportion of systems that would be placed into the five lead 90th
percentile classifications, which were previously discussed, at the start of the implementation of the
final LCRI. The protocol used to generate the "low" and "high" estimate for the final LCRI baseline
conditions is identical to the process described in Steps 1-5 of Section 3.3.5.1.1, with one difference in
the approach used to adjust lead 90th percentile results from LSL systems. That difference is described
below.

As was true for the 2021 LCRR, the EPA adjusted lead 90th percentile results using two multipliers to
reflect new sampling requirements under the final LCRI. The first multiplier, used to reflect the
requirement for LSL systems to collect all samples from LSL sites where possible, is the same as in

Final LCRI Economic Analysis

3-61

October 2024


-------
Section 3.3.5.1.1. The lower multiplier used to generate the "Low" estimate was 1.20, and the higher
multiplier used to generate the "High" estimate was 1.35.

However, the EPA modified the approach for the estimating the second multiplier to simulate the
expected increase on the lead 90th percentile levels if a system with LSLs uses the higher of the paired
first- and fifth-liter sample, as required under the final LCRI. To estimate the likelihood that a system
having LSLs being placed into one of five lead 90th percentile classifications, the EPA used the same
paired fifth- and first-liter data from 181 systems in Michigan at LSL locations that were collected in
2019, 2020, and 2021. Only the most recent monitoring period of sampling data was used for systems
that had multiple monitoring periods of sampling. However, for the final LCRI, the EPA used the ratio of
the higher of the fifth- and first-liter to the first-liter lead concentrations for each of these 181 systems.
Overall, the ratios ranged from 1 to 4.99 with a mean ratio of 1.48. Note that reported concentrations
below 1 ng/L and non-detects were changed to 1 ng/L.)

For this analysis, the average ratio of 1.48 was applied to the low and high 90th percentile values for
systems having LSLs to account for LSL systems using the higher of the fifth- and first-liter samples rather
than first-liter samples. The average ratio was not applied to the non-lead service line systems.

Exhibit 3-26 presents the "low" and "high" estimates of the percentage of systems in each lead 90th
percentile category by LSL status. The "low estimate" is based on the average 90th percentile lead value
reported to SDWIS/Fed from 2012 to 2020; the "high estimate" is based on the highest 90th percentile
lead value reported to SDWIS/Fed from 2012 to 2020. Based on the "low estimate," the percentage of
systems with LSLs having an ALE, under the final AL of 10 ng/L, was about 21.0 percent as opposed to
4.4 percent. For the "high estimate," a much higher percentage of systems with LSLs were classified as
having an ALE than those without LSLs, 38.9 percent compared to 8.7 percent.58

Exhibit 3-26: Percent of CWSs by Lead 90th Percentile Classification under the Final LCRI

Category

No LSLs

Has LSLs

Low Estimate

No ALE (P90 <10 ng/L)

95.6%

79.0%

<5/xg/L

88.5%

54.4%

>5 and <10 ng/L

7.1%

24.6%

ALE (>10 ng/L)

4.4%

21.0%

10 ng/L < P90 < 12 ng/L

1.0%

5.2%

12 ng/L < P90 < 15 ng/L

1.0%

5.6%

P90 > 15 ng/L

2.3%

10.3%

High Estimate

58 Note that under the final LCRI water systems must use the highest sample values in their 90th percentile
calculation. This could include samples from non-lead service line sites. The second adjustment does not explicitly
model the impact of this rule requirement. However, the EPA uses a low and high adjustment to reflect the
uncertainty of the new rule requirements on 90th percentile tap results.

Final LCRI Economic Analysis

3-62

October 2024


-------
Category

No LSLs

Has LSLs

No ALE (P90 <10 ng/L)

91.3%

61.1%

<5/xg/L

79.6%

37.3%

>5 and <10 ng/L

11.7%

23.8%

ALE (>10 ng/L)

8.7%

38.9%

10 ng/L < P90 < 12 ng/L

2.0%

7.8%

12 ng/L < P90 < 15 ng/L

1.9%

6.0%

P90 > 15 ng/L

4.8%

25.0%

Acronyms: ALE = action level exceedance; CWS = community water system; LCRI = Lead and Copper Rule
Improvements; LSL = lead service line; P90 = lead 90th percentile level.

Source: "Initial P90 Categorization_LCRI_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

Notes:

1.	Gray shaded rows indicate the final LCRI AL of 10 ng/L. Other AL values that the EPA considered are described
in greater detail in Chapter 8.

2.	Includes CWSs with known LSL status that also reported at least one 90th percentile value to SDWIS between
2012 and 2020.

3.	Totals may not add due to independent rounding.

4.	Percentages have changed slightly from those presented in the Proposed LCRI EA because the final rule
calculations are based on additional systems with known LSL status from the DWINSA one-time update.

3.3.5.1.3 Discussion of Data Limitations and Uncertainty

There are several factors that introduce uncertainty into the initial lead 90th percentile classification as
follows:

•	Use of historical SDWIS/Fed data to predict future 90th percentile levels.

•	Uncertainty in predicting the effects of sampling from 100 percent LSLs.

•	Reliance on an incomplete universe of systems with known LSL status.

•	Representativeness of first- and fifth-liter sample results from a single State (Michigan).

•	Variability of the ratio of first- and fifth-liter sample results from a single State.

Each of these limitations are described in more detail below.

1. Use of Historical SDWIS/Fed Data

As described previously in this section, the EPA recognizes the uncertainty in using historical SDWIS/Fed
data to predict future 90th percentile values by developing "low" and "high" end estimates of the
percent of CWSs in each P90 category.

2 Uncertainty in Predicting the Effects of Sampling from 100 Percent Sites Served by LSLs

For the 2021 LCRR and final LCRI, there is additional uncertainty in the effect of LSL systems being
required to take all samples from LSL sites instead of the 50 percent minimum as required under the
pre-2021 LCR. The EPA addressed this uncertainty by having a low and high estimate based on data

Final LCRI Economic Analysis

3-63

October 2024


-------
provided in Slabaugh al. (2015) paper, as was done for the Final 2021 LCRR EA59. However, as discussed
in Section 3.3.5.1.1, the EPA also noted that the Slabaugh et al. (2015) was based on 72 monitoring
periods from only 17 systems and the EPA did not have access to the data needed to conduct the
analysis at a system level.

3 Reliance of Incomplete Universe of Systems with Known LSL Status

An important factor in the analyses to determine a system's initial lead 90th percentile categorization is
the distinction between systems with LSLs and systems without LSLs. Limited data are available that
indicate a system's LSL status; thus, the EPA conducted a series of analyses to evaluate the
representativeness of the subset of 6,551 CWSs with known LSL status and at least one reported lead
90th percentile level during 2012-2020 as follows:

•	Compared the subset of 6,551 CWSs with known LSL status and at least one reported lead 90th
percentile level during 2012-2020 to all 49,529 CWSs in the SDWIS/Fed inventory.

•	Compared the subset of 6,551 CWSs with known LSL status and at least one reported lead 90th
percentile level during 2012-2020 to all 31,788 CWSs with at least one reported lead 90th
percentile level during 2012 - 2020 and unknown LSL status.

•	Determined the geographic representation of the 6,551 CWSs.

Each of these analyses are described in more detail below.

Comparison of Known LSL Status Subset to All CWSs

To help characterize the uncertainty of the subset of 6,551 with known LSL status and lead 90th
percentile data used to determine a system's initial lead 90th percentile classification, the EPA compared
this subset to the 49,529 active CWSs in SDWIS/Fed. As shown in Exhibit 3-27, although most of the
6,551 CWSs were those serving 3,300 or fewer people, they only represented 9 percent of all small
CWSs. The subset of the 6,551 CWSs serving 3,301 to 50,000 people and serving more than 50,000
people comprised 26 percent and 52 percent of all CWSs that serve these size categories, respectively.
The dataset of known LSL status systems is therefore less robust in its representation of water systems
serving fewer than 3,300 people. The dataset is consistent in the degree of representation across the
larger size categories representing water systems serving between 3,301 and 50,000 people and those
serving greater than 50,000 people.

59 The multiplier for the low estimate has changed since the 2021 LCRR economic analyses, due to more recent
SDWIS/Fed data, as well as an updated list of systems with LSLs.

Final LCRI Economic Analysis

October 2024

3-64


-------
Exhibit 3-27: Comparison Percent of CWSs with Known LSL Status to All CWSs by System Size

System Size
Population Served

Number of Active
CWSs

CWSs with Known LSL Status and Lead 90th Percentile
Data

Number of Systems

Percent of All CWSs by
Size

A

B

C= B/A

< 3,300

40,113

3,798

9%

3,301 to 50,000

8,400

2,220

26%

> 50,000

1,016

533

52%

Total

49,529

6,551



Source: SDWIS/Fed, 4th quarter 2020 frozen dataset. Also see file, "Extent of P90 Data_LCR_Final.xlsx" for

additional information.

Notes:

General: Refer to Section 3.3.5.1.1 and Exhibit 3-24 for more details on the development of the universe of
systems with known LSL status.

A: Includes all active CWSs in SDWIS/Fed based on 4th quarter 2020 frozen dataset.

B: Includes systems with known LSL status (either presence or absence of LSLs) and at least one reported lead 90th
percentile value to SDWIS/Fed during the 9-year analysis period of 2012 - 2020. Lead 90th percentile values for this
subset of systems were used to determine a system's initial lead 90th percentile classification. See file
"DWINSA_StateData_LSL_Status_Final.xlsx" for more information on how LSL status was determined.

Comparison of Known LSL Status Subset of All CWSs with Reported Lead 90th Percentile Data

The EPA compared the subset of systems with known LSL status and reported lead 90th percentile values
to the larger set of CWSs with at least one reported lead 90th percentile value (but unknown LSL status)
in SDWIS/Fed for 2012 - 2020. The first step was to generate the percentage of CWSs placed into each of
the five lead 90th percentile categories (based on the maximum or "high estimate" lead 90th percentile
value for 2012 - 2020) by four system size categories and CCT status using the larger dataset of 31,788
CWSs (i.e., 38,339 minus 6,551 CWSs) that includes CWSs with at least one reported lead 90th percentile
level during 2012 - 2020 and LSL status is unknown and the 6,551 CWSs with known LSL status. These
values represent the data reported to SDWIS/Fed before any adjustments were made to simulate the
requirements of the final LCRI. The results of this lead 90th percentile assessment are shown in Exhibit
3-28. Next, the EPA used a z-test to statistically evaluate the proportions for systems in each lead 90th
percentile category for the two sets of systems. There were 16 categories of system size, CCT status, and
lead 90th percentile category. However, since there were only two lead 90th percentile categories used
(above and below the final AL of 10 ng/L), there were only eight independent z-tests. Of the eight z-
tests, two returned a z-value falling within the critical range, indicating that differences in proportions
observed between the two sets were not statistically significant. See file "P90_Unknown LSL vs LSL
Known Status CWSs Final.xlsx" for additional information.

Final LCRI Economic Analysis

3-65

October 2024


-------
Exhibit 3-28: Comparison of P90 Data for CWSs with At Least One Reported Value to the Set of CWSs with Known LSL Status and
P90 Data by Two P90 Ranges, System Size, and CCT Status (Percent) Using the Baseline/High Estimate

System Size

CCT

Percent

P90 < 10 ng/L

P90 >10 ng/L

w/ Reported P90

Reported P90 &
Known LSL
Status

Difference

w/
Reported
P90

Reported P90
& Known LSL
Status

Difference

A

B

C = A-B

D

E

F = D-E

< 3,300

No

88.7%

89.5%

-0.9%

11.3%

10.5%

0.9%

< 3,300

Yes

85.4%

87.8%

-2.3%

14.6%

12.2%

2.3%

3,301-10K

No

93.2%

91.5%

1.6%

6.8%

8.5%

-1.6%

3,301-10K

Yes

93.4%

88.9%

4.5%

6.6%

11.1%

-4.5%

10,001-50K

No

96.2%

92.1%

4.1%

3.8%

7.9%

-4.1%

10,001-50K

Yes

94.4%

88.4%

6.0%

5.6%

11.6%

-6.0%

> 50K

No

No Data

100.0%

N/A

No Data

0.0%

N/A

> 50K

Yes

95.2%

87.9%

7.4%

4.8%

12.1%

-7.4%

Acronyms: LSL = lead service line; P90 = lead 90th percentile.

Source: SDWIS/Fed 4th Quarter Frozen Dataset, current through December 31, 2020. Also see file, "P90_Unknown LSL vs LSL Known Status CWSs_Final.xlsx" for

additional detail.

Notes:

The summaries in this table represent data reported to SDWIS/Fed before any adjustments were made to simulate the requirements of the final LCRI.

A,	D: Includes all 31,788 CWSs that equals all 38,339 CWSs with at least one reported P90 value minus the 6,551 CWSs with both a reported P90 value and
known LSL status (i.e., presence or absence of LSLs).

B,	E: Includes the subset of 6,551 CWSs with both a reported P90 value and known LSL status.

Final LCRI Economic Analysis

3-66

October 2024


-------
Geographic Representativeness of Known LSL Status Subset

The EPA recognizes that using the subset of systems with known LSL status for the analysis may under or
over represent 90th percentile results from specific geographic regions. To evaluate the potential
impacts of this uncertainty, the EPA grouped known LSL status systems into five geographic regions:

1.	East (Connecticut, Delaware, Maine, Massachusetts, Maryland, New Hampshire, New Jersey,
New York, North Carolina, Pennsylvania, Rhode Island, Vermont, Virginia, Washington, D.C.,
West Virginia),

2.	Midwest (Iowa, Illinois, Indiana, Kansas, Kentucky, Michigan, Minnesota, Missouri, Nebraska,
Ohio, Oklahoma, and Wisconsin),

3.	West (Arizona, California, Colorado, Idaho, Montana, New Mexico, North Dakota, Nevada,
Oregon, South Dakota, Texas, Utah, Washington, and Wyoming),

4.	South (Alabama, Arkansas, Florida, Georgia, Louisiana, Mississippi, South Carolina, and
Tennessee),

5.	Other (Alaska, American Samoa, Hawaii, Northern Mariana Islands, and the EPA Region 9 Tribal
Systems).

The 90th percentile values for these groups and for the full set of known LSL status systems (6,551) are
shown in Exhibit 3-29. Note that in the 7th DWINSA Allocation Memorandum (USEPA, 2023a), a high
percent of the allocation for lead service line replacement occurs for several midwest and northeast
States, including Illinois, Indiana, Michigan, New York, Ohio, and Pennsylvania; however, the EPA
recognizes uncertainty in not representing other geographic regions.

Final LCRI Economic Analysis

3-67

October 2024


-------
Exhibit 3-29: Number and Percent of CWSs with No ALE, and ALE - Comparison of Results from Five Geographic Regions with

Known LSL Status Using the Baseline/High Estimate

Category

No LSLs

Has LSLs

Number of CWSs

Percent

Number of CWSs

Percent

No CCT

Yes CCT

No CCT

Yes CCT

No CCT

Yes CCT

No CCT

Yes CCT

East (Connecticut, Delaware, Maine, Massachusetts, Maryland, New Hampshire, New Jersey, New York, North Carolina,
Pennsylvania, Rhode Island, Vermont, Virginia, Washington, D.C., West Virginia)

No ALE (P90 <10 ng/L)

629

776

91%

92%

73

386

94%

83%

ALE (P90 > 10 ng/L)

60

69

9%

8%

5

80

6%

17%

Midwest (Iowa, Illinois, Indiana, Kansas, Kentucky, Michigan, Minnesota, Missouri, Nebraska, Ohio, Oklahoma, and Wisconsin)

No ALE (P90 <10 ng/L)

1,017

706

89%

91%

288

513

84%

78%

ALE (P90 > 10 ng/L)

121

70

11%

9%

54

145

16%

22%

West (Arizona, California, Colorado, Idaho, Montana, New Mexico, North Dakota, Nevada, Oregon, South Dakota, Texas,

Utah, Washington, and Wyoming)

No ALE (P90 <10 ng/L)

498

258

92%

95%

37

56

86%

97%

ALE (P90 > 10 ng/L)

43

14

8%

5%

6

2

14%

3%

South (Alabama, Arkansas, Florida, Georgia, Louisiana, Mississippi, South Carolina, and Tennessee)

No ALE (P90 <10 ng/L)

74

127

93%

95%

17

90

100%

95%

ALE (P90 > 10 ng/L)

6

6

8%

5%

0

5

0%

5%

Other (Alaska, American Samoa, Hawaii, Northern Mariana Islands, and the EPA Region 9 Tribal Systems)

No ALE (P90 <10 ng/L)

260

32

92%

89%

1

100

100%

100%

ALE (P90 > 10 ng/L)

23

4

8%

11%

0

0

0%

0%

All Systems with Known LSL Status

No ALE (P90 <10 ng/L)

2,478

1,899

91%

92%

416

1,145

86%

83%

ALE (P90 > 10 ng/L)

253

163

9%

8%

65

232

14%

17%

Acronyms: ALE = action level exceedance; CCT = corrosion control treatment; CWS = community water system; LSL = lead service line; P90 = lead 90th
percentile level

Source: DWINSA_StateData_LSL_Status_Final.xlsx.

Notes: Includes only systems with known LSLs status and at least one reported lead 90th percentile (P90) to SDWIS/Fed for 2012 - 2020. CWSs were assigned to
a P90 category of "No ALE" or "ALE" based on their highest reported lead P90 value reported to SDWIS/Fed for 2012-2020.

Final LCRI Economic Analysis

3-68

October 2024


-------
Representativeness of First- and Fifth-Liter Data from a Single State

The EPA used data from the State of Michigan to estimate the impact on the lead 90th percentile levels
to simulate the expected increase in P90 values if they were based on the higher of the first- and fifth-
liter sample under the final LCRI for LSL systems. As described earlier, an average ratio of the maximum
of the first- and fifth-liter 90th percentile values to the first liter 90th percentile values from 181 systems
in Michigan was applied to 90th percentile values from the subset of 6,551 systems with known LSL
status that have a status of "Has LSLs" (1,758 systems) to determine the P90 category for systems under
the final LCRI. The EPA recognizes the uncertainty introduced in using data from a single State that may
not represent the values on a national level. However, the Michigan data represent actual compliance
monitoring data collected recently from all systems within the State.

3.3.5.2 Likelihood of a System Having Multiple Lead ALEs

The EPA's final LCRI requires water systems that have two lead ALEs in five years to prepare and submit
a temporary filter plan to the State within 60 days of their second lead ALE. In addition, systems with at
least three lead ALEs in a five-year period (i.e., multiple lead ALEs) must provide enhanced community
outreach and make pitcher filters available to the people that they serve (see Chapter 4, Section 4.3.6.4
for additional detail). The remainder of this section first provides the EPA's approach for determining
the percentage of systems with at least two lead ALEs, followed by those that subsequently have three
or more lead ALE in five years.

3.3.5.2.1 Likelihood of a System HavinR Two Lead ALEs in Five Years

The EPA determined the percentages of CWSs, with at least one lead ALE, that within a period of five
years had a second lead ALE, based on data reported to SDWIS/Fed during 2012 - 2020 in the fourth
quarter 2020 frozen dataset. The analysis is restricted to the 6,551 CWSs with known LSL status and lead
P90 data from SDWIS/Fed. Both a high and low estimate were calculated, using a similar method as
discussed in Section 3.3.5.1.1 to adjust 90th percentile values to simulate the final rule changes to the
lead tap sampling requirements. Note that this adjustment to the values from the 1,758 systems with
LSLs is the only difference between the high and low estimates. The EPA determined that a CWS could
only have two P90 results in any given year due to monitoring requirements. If a CWS reported more
than two P90 values in a given year, the two highest reported values were used. These percentages are
provided in Exhibit 3-30. Note that an ALE in this analysis refers to a system's 90th percentile value being
above 10 ng/L for lead, which is consistent with the final rule. Systems with LSLs were more likely than
systems without LSLs to have two lead ALEs in a five-year period for every size category and CCT status.
Note that the high estimate may result in a lower percentage than the low estimate. This is not
unexpected, as using the high estimate will produce more systems that report an ALE, expanding the
total pool of systems from which the percentages are calculated. The EPA assumed that NTNCWSs
would have the same percentage of systems having at least two lead ALEs in a five-year period as CWSs.

Final LCRI Economic Analysis

3-69

October 2024


-------
Exhibit 3-30: Percentage of CWSs and NTNCWSs with At Least One ALE that Have At Least
Two Lead Action Level Exceedances (Above 10 |ig/L) in Five Years

System Size
(Population
Served)

Has LSLs

No LSLs

Has CCT

No CCT

Has CCT

No CCT

Low Estimate

<3,300

41.5%

27.1%

19.4%

10.5%

3,301-10,000

45.0%

56.8%

33.3%

15.8%

10,001-50,000

54.2%

29.5%

42.9%

16.7%

>50,000

66.1%

No Data

60.0%

No Data

High Estimate

<3,300

39.1%

22.9%

19.4%

10.5%

3,301-10,000

44.0%

59.2%

33.3%

15.8%

10,001-50,000

56.7%

30.0%

42.9%

16.7%

>50,000

66.9%

No Data

60.0%

No Data

Acronyms: CCT = corrosion control treatment; CWSs = community water system; LSLs = Lead Service Lines;
NTNCWS = non-transient non-community water system.

Source: "Two Lead ALE_LCRI_10_AL_Final.xlsx."

Notes: The EPA assumed that the same percentage of NTNCWSs would have at least two lead ALEs in a five-year
period as CWSs and that these percentages are representative of water systems with the same characteristics (i.e.,
size, CCT, and LSL status).

3,3.5.2.2 Likelihood of a System with Two Lead ALEs Having Three or More Lead ALE in Five Years

The EPA also determined the percentages, by size and LSL and CCT status, of CWSs with at least two lead
ALEs in a five-year period that have at least one additional lead ALE in five years, based on data reported
to SDWIS/Fed during 2012 - 2020 in the fourth quarter 2020 frozen dataset.60 The percentages were
derived by calculating the proportion of systems that had at least three lead ALEs within a five-year
timeframe, relative to the number of systems that had two ALEs in a period of five years (represented in
Exhibit 3-30) between 2012 and 2020. The results of the analysis are detailed in Exhibit 3-31 below. Note
that the exhibit shows a higher percentage of systems without LSLs having a third ALE in a five-year
period than systems with LSLs in most categories. This is due to the small number of systems in the
sample pool; note that when the percentages in Exhibit 3-31 are multiplied by the percentages of
systems with two ALEs in Exhibit 3-30, and multiplied again by the percent of all systems with at least
one ALE in Exhibit 3-26, the overall percent of systems with three lead ALEs in five years is higher for
systems with LSLs compared to systems without LSLs for all categories except systems serving 10,001 to
50,000 with no CCT. There were no systems that have more than three lead ALEs in five years in this

60 For the final LCRI EA, the EPA modified its approach for estimating multiple lead ALEs. For the proposed LCRI EA,
the EPA determined the percentage of CWSs with at least three lead ALEs in a five-year period. For the final LCRI
EA, the EPA determined the percentage of CWSs that have at least two lead ALEs in a five-year period that have at
least one additional lead ALE. The EPA needed to change its approach to model the final rule requirement that
systems with two lead ALEs would prepare a filter plan, but would not be required to conduct public education
activities and filter distribution until the third lead ALE in the same five-year period.

Final LCRI Economic Analysis

3-70

October 2024


-------
category. The EPA assumed that NTNCWSs would have the same percentage of multiple lead ALEs as
CWSs.

Exhibit 3-31: Percentage of CWSs and NTNCWSs with At Least Two Lead ALEs in Five Years
that Have At Least One Additional Lead ALE (Above 10 |ig/L) in Five Years

System Size
(Population
Served)

Has LSLs

No LSLs

Has CCT

No CCT

Has CCT

No CCT

Low Estimate

<3,300

37.0%

25.0%

50.0%

29.2%

3,301-10,000

42.2%

32.0%

55.6%

33.3%

10,001-50,000

35.0%

38.5%

0.0%

100.0%

>50,000

50.0%

No Data

33.3%

No Data

High Estimate

<3,300

48.1%

25.0%

50.0%

29.2%

3,301-10,000

39.6%

31.0%

55.6%

33.3%

10,001-50,000

33.6%

40.0%

0.0%

100.0%

>50,000

49.4%

No Data

33.3%

No Data

Acronyms: CCT = corrosion control treatment; CWSs = community water system; LSLs = Lead Service Lines;
NTNCWS = non-transient non-community water system.

Source: "Multiple Lead ALE_LCRI_10_AL_Final.xlsx."

Notes: The EPA assumed that the same percentage of NTNCWSs would have multiple lead ALEs as CWSs and that
these percentages are representative of water systems with the same characteristics (i.e., size, CCT, and LSL
status).

3.3.5.3 Likelihood of an Individual Lead Sample Exceeding the Lead AL

3.3.5.3.1 Baseline (2021 LCRR)

Under the 2021 LCRR, the EPA requires all systems to take specific actions in response to any single lead
tap sample that is above 15 ng/L. Individual sample results are not available in SDWIS/Fed. Therefore,
the EPA used available compliance monitoring data from the State of Michigan, which is the only State
to date to require a first- and fifth-liter sample for LSL systems, to calculate the likelihood of an
individual sample being greater than 15 ng/L based on system size, LSL status, and the five lead 90th
percentile categories presented in Section 3.3.5.1. The analysis for the 2021 LCRR uses fifth-liter samples
for systems with LSLs and first-liter samples for systems without LSLs from the Michigan dataset using
the following steps:

Step 1 - Categorized Michigan systems as with or without LSLs based on compliance monitoring data
as follows:

Final LCRI Economic Analysis

3-71

October 2024


-------
•	The EPA assumed 19361 CWSs have LSLs because the system provided first- and fifth-liter lead
samples as required by Michigan's State-level regulation and were identified as having LSLs in
their online service line inventory information (Michigan EGLE, 2020).

•	The EPA assumed 975 CWSs have no LSLs because the system submitted only first-liter data and
did not report any LSLs in the online service line inventory information.

Step 2 - Calculated lead 90th percentile levels: The EPA calculated lead 90th percentile (P90) values for
CWSs as described below. The purpose of this step is to categorize each dataset by five P90 categories
so that the likelihood of a sample being above 15 ng/L can be calculated for each 90th percentile
category separately.

•	For all systems with LSLs. to approximate the lead 90th percentile value if all samples were
collected from LSL sites and all samples are fifth-liter samples as is required under the 2021
LCRR, the EPA calculated a P90 value using the fifth-liter concentrations. If an insufficient
number of fifth-liter samples were available to meet minimum sampling requirements, the EPA
used the highest first-liter sample results to meet the minimum requirements. The EPA used this
P90 value to categorize the dataset by the five lead 90th percentile categories.

•	For all systems with no LSLs. the EPA calculated the 90th percentile using all first-liter
concentration data for each dataset and used this information to categorize the dataset by P90
level.62

Step 3 - Calculated the likelihood of a sample > 15 ng/L: The EPA calculated the proportion of samples
above 15 ng/L for each 90th percentile category. Results are shown in Exhibit 3-32. Note that if first-liter
samples were used to calculate the 90th percentile value for systems with LSLs having insufficient fifth-
liter sampling data, those first-liter samples were not considered when calculating the proportion of
samples above 15 ng/L.

Individual samples from systems with LSLs had a higher likelihood of being above 15 ng/L than individual
samples from systems without LSLs when the system's calculated 90th percentile was less than or equal
to 10 ng/L. Individual samples from systems without LSLs had a higher likelihood of being above 15 ng/L
than individual samples from systems with LSLs when the system's calculated 90th percentile was greater
than 10 ng/L. Note that although the percentage of individual lead samples above 15 ng/L from systems
in the lead 90th percentile categories above the lead AL of 10 ng/L is higher for systems without LSLs
than with LSLs, the overall percentage of individual lead samples above 15 ng/L from systems with LSLs
is higher than from systems without LSLs (2.7 percent from systems with LSLs as opposed to 0.8 percent
from systems without LSLs).63 For additional detail on the number and percent of samples in the
Michigan dataset that were greater than 15 ng/L, see file

61	This total of 193 CWSs is different than the 181 Michigan CWSs used to adjust the lead 90th percentile data for
systems with LSLs. The EPA excluded 12 systems that only reported non-detects for their fifth-liter samples, and
had a determination of "No LSLs" based solely on Michigan's online service line inventory information.

62	Note that the EPA discusses other potential sources of lead in premise plumbing, apart from lead and GRR
service lines, in the final LCRI Federal Register notice in sections IV.A Regulatory Approach and IV.E Tap Sampling for
the Lead and Copper.

63	Overall, systems with LSLs have 101 samples > 15 ng/L out of a total of 3,685 samples (2.7%) while systems
without LSLs have 79 samples > 15 ng/L out of a total of 9,613 samples (0.8%).

Final LCRI Economic Analysis

3-72

October 2024


-------
"Likelihood_Sample_Above_AL_LCRR_Find_Fix_Final.xlsx," available in the docket at EPA-HQ-OW-2022-
0801 at www.regulations.gov.

Note that the Michigan dataset does not include first- and fifth-liter data for NTNCWSs. Therefore, the
EPA assumed the same likelihood for NTNCWSs as those presented in Exhibit 3-32 for CWSs. The text in
red is the data input name used in the SafeWater LCR model.

Exhibit 3-32: Percent of Individual Lead Sample Results Above 15 |ig/L Based on Michigan
CWSs with Known LSL Status for the 2021 LCRR

LSL Status

P90 >15 ng/L

12 ng/L < P90 <
15 (J-g/L

10 ng/L < P90 <
12 (J-g/L

5 |ig/L < P90 <
10 ng/L

P90 < 5 ng/L

pp90above
all5 1

pp90above
all5 2

pp90above
all5 3

pp90above
all5 4

pp90above
a 115 5

Has LSLs

16.9%

9.3%

5.3%

3.1%

0.7%

No LSLs

22.2%

10.0%

7.9%

3.0%

0.4%

Acronyms: P90 = lead 90th percentile level.

Notes:

1.	Although the percentage of individual lead samples above 15 ng/L from systems in the lead 90th percentile
categories above 10 ng/L is higher for systems without LSLs than with LSLs, the overall percentage of individual
lead samples from systems with LSLs is higher than from systems without LSLs (2.7 percent from systems with LSLs
as opposed to 0.8 percent from systems without LSLs).

2.	For additional detail, see file "Likelihood_Sample_Above_AL_LCRR_Find_Fix_Final.xlsx," available in the
docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

3.3.5.3.2 Final LCRI

Under the final LCRI, the EPA's is maintaining the requirement for systems to take specific actions in
response to any single lead tap sample that is above the lead action level, which is final at 10 ng/L. As
was true for the 2021 LCRR, the EPA used available compliance monitoring data from the State of
Michigan to calculate the likelihood of an individual sample being greater than 10 ng/L based on system
size, LSL status, and the five lead 90th percentile categories presented in Section 3.3.5.1. The analysis for
the final LCRI uses the maximum of the first- and fifth-liter samples for systems with LSLs and first liter
samples for systems without LSLs from the Michigan dataset using the following steps:

Step 1 - Categorized Michigan systems as with or without LSLs based on compliance monitoring data.

The EPA used the same assumptions to categorize water systems as described under Step 1 for the 2021
LCRR analysis (see Section 3.3.5.3.1).

Step 2 - Calculated lead 90th percentile levels: The EPA calculated lead 90th percentile values for CWSs
with and without LSLs so that the likelihood of a sample being above 10 ng/L can be calculated for each
of the five 90th percentile categories separately.

• For all systems with LSLs. to approximate the lead 90th percentile value if all samples were
collected from LSL sites and all samples are the maximum of the first- and fifth-liter samples
(new requirements in the final LCRI), the EPA calculated a lead 90th percentile value using the
maximum of the first- and fifth-liter concentrations. The EPA used this 90th percentile to
categorize the dataset by the five P90 categories.

Final LCRI Economic Analysis

3-73

October 2024


-------
• For all systems with no LSLs, the EPA used the same approach as the 2021 LCRR analysis by
calculating the 90th percentile using all first-liter concentration data for each dataset and used
this information to categorize the dataset by P90 level.

Step 3 - Calculated the likelihood of a sample > 10 pg/L: The EPA calculated the proportion of samples
above 10 ng/Lfor each 90th percentile category. Results are shown in Exhibit 3-33.

Individual samples from systems with LSLs had a higher likelihood of being above 10 ng/L than individual
samples from systems without LSLs when the system's calculated 90th percentile was greater than 15
Hg/L or less than or equal to 5 ng/L. Individual samples from systems without LSLs had a higher
likelihood of being above 10 ng/L than individual samples from systems with LSLs when the system's
calculated 90th percentile was greater than 10 ng/L but less than 15 ng/L. When the system's calculated
90th percentile was between 5 and 10 ng/L, individual samples from systems with LSLs had an
approximately equal chance of being above 10 ng/L as those from systems without LSLs. Note that the
percentage of individual lead samples above 10 ng/L from systems in the two lead 90th percentile
categories above the lead AL of 10 ng/L but below 15 ng/L is higher for systems without LSLs than with
LSLs. The reason is the number of individual samples collected from systems with LSLs and a 90th
percentile level above 10 ng/L is much higher than the number of individual samples from systems
without LSLs and a 90th percentile above 10 ng/L. For example, of the systems with LSLs that are
classified in the lead 90th percentile category of 12 ng/L < P90 < 15 ng/L, 83 of the 495 individual lead
samples from those systems are above 10 ng/L (16.8%) compared to 30 of 130 samples (23.1%) for non-
lead service line systems. Note that the overall percentage of individual lead samples above 15 ng/L
from systems with LSLs is higher than from systems without LSLs (2.7 percent from systems with LSLs as
opposed to 0.8 percent from systems without LSLs).64 For additional detail on the number and percent
of samples in the Michigan dataset that were greater than 10 ng/L, see file

"Likelihood_Sample_Above_AL_LCRI_DSSA_Final.xlsx/' available in the docket at EPA-HQ-OW-2022-
0801 at www.regulations.gov.

Also note that the Michigan dataset does not include first- and fifth-liter data for NTNCWSs. Therefore,
the EPA assumed the same likelihood for NTNCWSs as those presented in Exhibit 3-33 for CWSs. The
text in red is the data input name used in the SafeWater LCR model.

Exhibit 3-33: Percent of Individual Lead Sample Result Above 10 pg/L Based on Michigan
CWSs with Known LSL Status for the Final LCRI

LSL Status

P90 >15 ng/L

12 ng/L < P90 <
15 (J-g/L

10 ng/L < P90 <
12 (J-g/L

5 |ig/L < P90 <
10 ng/L

P90 < 5 ng/L

pp90above
alio 1

pp90above
alio 2

pp90above
alio 3

pp90above
alio 4

pp90above
alio 5

Has LSLs

25.2%

16.8%

13.8%

6.5%

1.8%

No LSLs

22.2%

23.1%

21.1%

6.5%

0.5%

Acronyms: P90 = lead 90th percentile level.
Notes:

64 Overall, systems with LSLs have 284 samples > 10 ng/L out of a total of 4,575 samples (6.2%) while systems
without LSLs have 130 samples > 10 ng/L out of a total of 9,613 samples (1.4%).

Final LCRI Economic Analysis

3-74

October 2024


-------
1.	Although the percentage of individual lead samples above 15 ng/Lfrom systems in the lead 90th percentile
categories above the lead AL of 10 ng/L but below 15 ng/L is higher for systems without LSLs than with LSLs,
the overall percentage of individual lead samples from systems with LSLs is higher than from systems without
LSLs.

2.	For additional detail, see file "Likelihood_Sample_Above_AL_LCRI_DSSA_Final.xlsx," available in the docket at
EPA-HQ-OW-2022-0801 at www.regulations.gov.

3.3.5.3.3 Discussion of Data Limitations and Uncertainty

Recent data from the State of Michigan were used to estimates the likelihood of a single sample result
above 10 ng/L for systems with LSLs as compared to systems without LSLs. While there is uncertainty in
the national representativeness of the data (i.e., does lead tap sample data from Michigan represent
lead tap data from other States), there are advantages to the use of these data. The Michigan data
contains more than 14,100 individual lead sample results from systems with and without LSLs, with both
first- and fifth-liter sampling results. To date, no other State requires both a first- and fifth-liter sample
from sites served by LSLs. However, even with the large sample size, the relatively smaller sample size
within each data category can introduce uncertainty.

3.3.5.4 Systems with Copper Only ALEs

The pre-2021 LCR set an AL concentration of 1.3 mg/L for copper. If a system exceeds the AL in more
than 10 percent of tap water samples collected during any monitoring period (i.e., if the 90th percentile
level is greater than the AL), the system has not violated the rule but must conduct additional actions
such as CCT steps, WQP monitoring, and source water monitoring. Requirements for systems that
exceed both the lead and copper AL are considered in the analysis of lead ALE systems; thus, this section
presents the estimates of systems with only copper ALEs.

The EPA reviewed SDWIS/Fed 90th percentile copper data from 2012 through 2020 to identify systems
that had exceeded the copper AL but not the lead AL under the baseline conditions (i.e., prior to the
implementation of the final LCRI).65 Thus, the analysis uses the ALE of 15 ng/L from the pre-2021 LCR as
opposed to the final LCRI lead AL of 10 ng/L.66

The average annual percentage of all CWSs exceeding the copper AL without a lead exceedance during
this time period is shown in Exhibit 3-34 by size, CCT status, and source type and was extremely low,
indicating a low number of systems exceeded the copper AL only compared to the total number of
systems in each category. For CWSs with CCT, the percentages ranged from zero percent for CWSs

65	The EPA expanded the analysis period from the Proposed LCRI EA (USEPA, 2023b) from 2017 - 2020 to 2012 -
2020 to be more consistent with other EA analyses that use a nine-year analysis period. For CWSs with CCT, the
overall percentage with a copper only ALE increased by about 0.2% using the expanded data set and was
essentially the same for those without CCT. For NTNCWSs with CCT, the overall percentage with a copper only ALE
increased by about 0.4% for those using GW and 0.9% for SW systems, based on the expanded analysis period. On
the other hand, the percentages dropped to 0.1 to 0.2% for NTNCWSs without CCT using GW and SW, respectively,
based on the expanded analysis period.

66	Note this approach will overestimate the percentage of systems exceeding the copper AL under the 2021 LCRR
and LCRI because more systems are expected to initially exceed the lead AL due to changes in the sampling
protocol and 90th percentile calculations for systems with LSLs. In addition, the lower AL of 10 ng/L will also
contribute to an expected increase in the number of systems initially exceeding the lead AL under the final LCRI.

Final LCRI Economic Analysis

3-75

October 2024


-------
serving greater than 1,000,000 people to 2.4 percent for GW systems serving 501 - 1,000 people. The
overall percentage of CWSs with CCT and a copper ALE was approximately one percent. For those
without CCT, 16 systems serving more than 50,000 people were b3 systems and had no copper ALEs. No
CWS size or source without CCT category had a copper ALE percentage above 1.0 percent and, overall,
all CWSs without CCT had an estimated copper exceedance percentage of around 0.4 percent. For a
detailed information and for the number of systems exceeding the copper AL only in each category, see
"CWS Inventory Characteristics_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

Similar information is shown in Exhibit 3-35 for NTNCWSs. The overall percentages of NTNCWSs with a
copper ALE were 3.2 percent and less than one percent for NTNCWSs with and without CCT,
respectively. For additional detail, see "NTNCWS Inventory Characteristics_Final.xlsx," available in the
docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Note that for the cost estimates presented in Chapter 4 for the final LCRI and Appendix B for the pre-
2021 LCR and the 2021 LCRR, the EPA made a simplifying assumption that no system with CCT would
have a copper ALE, because approximately 1 percent of CWSs and 3.0 percent of NTNCWSs with CCT
were estimated to have a copper ALE. See Chapter 4, Section 4.3.2.3.1 and Appendix B for additional
detail.

Final LCRI Economic Analysis

3-76

October 2024


-------
Exhibit 3-34: Average Percent of CWSs that Had Any Copper Only ALE from 2012-2020

System Size
(Population
Served)

Average (2012 -2020)

with CCT

without CCT

Ground Water

Surface Water

All Sources

Ground Water

Surface Water

All Sources

<100

1.9%

0.7%

1.6%

0.3%

1.0%

0.4%

101-500

1.7%

0.9%

1.4%

0.4%

0.7%

0.4%

501-1,000

2.4%

0.6%

1.7%

0.4%

0.6%

0.5%

1,001-3,300

2.1%

0.4%

1.4%

0.5%

0.3%

0.4%

3,301-10,000

1.2%

0.2%

0.6%

0.3%

0.2%

0.3%

10,001-50,000

1.0%

0.2%

0.4%

0.1%

0.1%

0.1%

50,001-100,000

0.1%

0.1%

0.1%

0.0%

0.0%

0.0%

100,001-1M

0.2%

0.0%

0.0%

0.0%

0.0%

0.0%

>1M

0.0%

0.5%

0.5%







All Sizes

1.7%

0.4%

1.1%

0.4%

0.5%

0.4%

Source: "CWS Inventory Characteristics_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Notes:

1. The EPA estimated that 16 CWSs are b3 systems, serve 50,001 - 1 million people, and have no CCT. No b3 systems serve more than 1 million people. Refer

to Section 3.3.3 for the EPA's approach for estimating the number of b3 systems based on SDWIS/Fed fourth quarter 2020 frozen dataset.
2. The gray shaded cells denote that there were no CWSs serving >1M people without CCT in the CWS inventory.

Final LCRI Economic Analysis

3-77

October 2024


-------
Exhibit 3-35: Average Percent of NTNCWSs that Had Any Copper Only ALE from 2012-2020

System Size

Average (2012-2020)

with CCT

without CCT

Ground Water

Surface Water

All Sources

Ground Water

Surface Water

All Sources

<100

3.8%

4.4%

3.8%

0.6%

0.3%

0.6%

101-500

2.8%

1.7%

2.7%

0.6%

1.2%

0.7%

501-1,000

3.5%

1.0%

3.3%

0.6%

1.2%

0.6%

1,001-3,300

3.7%

1.2%

3.3%

0.8%

1.7%

0.9%

3,301-10,000

1.3%

0.0%

1.0%

0.1%

0.7%

0.3%

10,001-50,000

0.0%

0.0%

0.0%

1.5%

0.7%

1.0%

50,001-100,000



0.0%

0.0%







100,001-1M



0.0%

0.0%







>1M













All Sizes

3.3%

1.8%

3.2%

0.6%

0.9%

0.6%

Source: "NTNCWS Inventory Characteristics_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Notes:

1.	The gray shaded cells denote that for NTNCWSs with CCT, no GW NTNCWSs serve more than 50,000 people and no SW NTNCWSs serve more than > 1M
people.

2.	For NTNCWSs without CCT, none serve more than 50,000 people, regardless of their water source.

Final LCRI Economic Analysis

3-78

October 2024


-------
3.3.6	Treatment Plant Characterization

This section explains the baseline inputs for the following treatment-related PWS characteristics:

•	Entry points per system

•	Average daily flow

•	Design flow

•	pH of finished water

•	Orthophosphate (P04) dose

For additional detail and values used in this EA, see the file, "Baseline CCT Characteristics.xlsx/' available
in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

3.3.6.1 Entry Points per System

Entry points are the locations where source water is treated (in the case of the LCR where CCT occurs)
and enters the distribution system. Systems can have multiple entry points. The EPA developed
estimates of entry points per system using unique sampling point data from UCMR 3 (USEPA, 2017),
along with SDWIS/Fed facility data, and a modeled frequency distribution.

The UCMR 3 data record a unique identifying number for the entry point sample location(s) for each
system. Given the information provided, the EPA assumed that the number of unique sample point IDs
per system approximates the total number of entry points per system.

For systems without UCMR 3 occurrence data, the EPA developed estimates based on SDWIS/Fed facility
data. The SDWIS/Fed data include unique identification numbers for system facilities, as well as facility
type and activity status. This analysis relies on active facilities identified as treatment plants. Using the
assumption that treatment plants are associated with one entry point, the SDWIS/Fed facility data
provide an approximation for the number of entry points per system when a system does not have
UCMR 3 occurrence data. The EPA considers the UCMR 3 sampling point data to be of higher quality
than the SDWIS/Fed treatment facility data. If the SDWIS/Fed treatment facility data value for a system
exceeded the maximum number found for the equivalent system size and source water combination in
the UCMR 3 data, the EPA limited the system entry point value to the UCMR 3 maximum number of
entry points.

For systems without UCMR 3 occurrence data or SDWIS/Fed facility data, the EPA relies on an estimate
of the number of entry points. The estimated value for each system with missing entry point count data
was imputed from known entry point counts for stratified SDWIS/Fed data. Within each stratum,
defined by a combination of system size and source water, the EPA sampled from systems with known
entry point counts. Sampling was done with replacement after truncating the entry point counts to the
maximum recorded in UCMR 3. For reproducibility, the EPA performed this sample-based imputation in
R using the 'base::sample' function (R Core Team, 2021).

Following this process, the EPA relied on sample point values recorded in UCMR 3 for 5,419 systems,
SDWIS/Fed facility data for 43,563 systems, and imputed entry point values for 17,523 systems. All

Final LCR! Economic Analysis

3-79

October 2024


-------
systems have at least one entry point. Among CWSs, the maximum number of entry points is 202, and
the mean is 1.80. Among NTNCWSs, the maximum number of entry points is 22, and the mean is 1.31.

Exhibit 3-36 summarizes the final frequency distribution of entry point input ranges for each CWS
stratum of size and source water combination. Exhibit 3-37 summarizes the final frequency distribution
of entry point input ranges for each NTNCWS stratum of size and source water combination. These
distributions are used to proportionally assign numbers of entry points to systems in each system size
and type category.67

Exhibit 3-36: Frequency Distribution of Entry Point Inputs for CWSs

System Size
(Population Served

Ground Water

Surface Water

1 EP

2-5
EP

6-
10
EP

il-
ls

EP

16-

20

EP

21-
100
EP

> 100
EP

1 EP

2-5
EP

6-
10
EP

il-
ls

EP

16-
20 EP

21-
100
EP

>

100
EP

< 100

90%

10%

0.1%

0

0

0

0

87%

13%

0

0

0

0

0

101-500

76%

24%

0

0

0

0

0

84%

16%

0

0

0

0

0

501-1,000

62%

38%

0.5%

0

0

0

0

76%

23%

0.8%

0

0

0

0

1,001-3,300

48%

50%

1%

0

0

0

0

70%

30%

0.7%

0

0

0

0

3,301-10,000

32%

59%

8%

0.9%

0.1%

0

0

54%

43%

3%

0.5%

0.04%

0

0

10,001-50,000

3%

58%

28%

7%

3%

1%

0.07%

3%

82%

10%

2%

1%

0.6%

0

50,001-100,000

0

51%

25%

8%

8%

9%

0

0.2%

74%

13%

6%

2%

4%

0

100,001-1M

0

34%

22%

11%

8%

24%

1%

0.3%

67%

13%

4%

9%

6%

0.3%

Acronyms: CWS - community water systems; EP - entry point.

Exhibit 3-37: Frequency Distribution of Entry Point Inputs for NTNCWSs



Ground Water

Surface Water

System Size

1 EP

2-5

6-10

11-20

>20

1 EP

2-5

6-10

11-20

>20

(Population Served



EP

EP

EP

EP



EP

EP

EP

EP

< 100

84%

16%

0.4%

0

0

82%

18%

0

0

0

101-500

81%

19%

0

0

0

74%

26%

0

0

0

501-1,000

0

0

0

0

0

0

0

0

0

0

1,001-3,300

68%

30%

2%

0

0

61%

31%

8%

0

0

3,301-10,000

53%

44%

2%

1%

0

35%

44%

14%

6%

0

10,001-50,000

10%

80%

0

10%

0

30%

40%

5%

20%

5%

50,001-100,000

0

0

0

0

0

0

100%

0

0

0

100,001-1M

0

0

0

0

0

0

100%

0

0

0

Acronyms: NTNCWS" - non-transient non-community water systems; EP - entry point.

67 The SDWIS/Fed data provide information on the PWS characteristics that typically define PWS categories, or
strata, for which the EPA develops costs in rulemakings. These characteristics include system type (CWS, NTNCWS),
number of people served by the PWS, PWS's primary raw water source (GW or SW), PWS's ownership type (public
or private), and PWS state. For more information on the use of baseline and compliance characteristics to define
model systems in the EPA's cost analysis, please see Section 3.2.

Final LCRI Economic Analysis

3-80

October 2024


-------
3.3.6.2	Average Daily Flow and Design Flow

Average daily production flow and design flow per system are based on regression equations from the
EPA Report, Geometries and Characteristics of Public Water Supplies (USEPA, 2000). The average daily
flow and design flow are functions of the population served, with different equations for source water
type (surface or ground water), ownership (public or private) and for purchased and non-purchased
systems. The flow was then divided by the number of entry points to calculate the flow per treatment
plant for the system (assuming each entry point has one treatment plant). As a conservative estimate,
the flow-population regression equations for CWSs were also used for NTNCWSs.

The EPA evaluated historical SDWIS/Fed data to determine the proportion of systems with CCT that use
pH adjustment, orthophosphate (P04) treatment, or both. This analysis is detailed in Chapter 4, Section
4.3.2.2.1.

Baseline pH levels and P04 dosages are also important inputs in calculating the incremental costs of the
final LCRI. The EPA used the SYR3 ICR dataset to characterize the distribution of finished water pH for
those systems that have CCT installed and those that do not under baseline conditions. The EPA also
estimated the distribution of P04 dosages for large, medium, and small systems with and without LSLs.
See the file, "Baseline CCT Characterization.xlsx" for additional detail and for final baseline pH and P04
input values used to develop costs and benefits for this EA.

3.3.6.3	Discussion of Data Limitations and Uncertainties

The EPA recognizes that there is uncertainty in assuming a system's total flow is divided equally among
each entry point because a single system may have a mix of large and small plants to support their
population. There is also uncertainty in using the equations from the 2000 Geometries Document
(USEPA, 2000) to predict future average daily and design flow based on a system's retail population.
Water use efficiency has increased substantially since the 1980's, with a major improvement between
2005 and 2010 (Rockaway et al., 2011). A 2016 Water Research Foundation study reported a 22 percent
decline in indoor water use between 1999 and 2016 (WaterRF, 2016). The trend of lower residential
water use could result in lower flow per population and lower treatment costs as compared to predicted
values in this EA.

3.3.7 Lead and Copper Tap Schedules

This section describes the EPA's approach for estimating water systems' initial lead and copper tap
monitoring schedules under the 2021 LCRR and final LCRI. As a starting point, the EPA estimated the
likelihood a system would be on a standard six-month monitoring or one of the reduced lead and copper
tap monitoring schedules under the pre-2021 LCR. This approach is detailed in Section 3.3.7.1.

Section 3.3.7.2 describes how the EPA adapted the pre-2021 schedules to determine the initial lead and
copper tap monitoring period under the 2021 LCRR and final LCRI. Note that these schedules do not
apply to systems implementing a point-of-use (POU) program under the small system flexibility option.
Those systems must sample one-third of POU sites annually to assess performance. Section 3.3.7.4
provides a discussion of the data limitations and uncertainty.

Final LCRI Economic Analysis

3-81

October 2024


-------
3.3.7.1 Estimating Initial Lead and Copper Tap Monitoring Period under the Pre-2021 LCR

Under the pre-2021 LCR, systems on routine (semi-annual) lead and copper tap sampling could qualify
for reduced sampling by meeting specific criteria. These criteria varied for the three broad LCR system
size categories.68 Reduced monitoring allows a system to collect lead and copper tap samples from a
reduced number of sites on an annual, triennial, or 9-year tap sampling monitoring period.

Exhibit 3-38 provides a summary of the criteria used to estimate the various lead and copper tap
monitoring schedules under the pre-2021 LCR based on information reported to SDWIS/Fed in the
fourth quarter frozen dataset, current through December 31, 2020.

Exhibit 3-38: SDWIS/Fed Criteria Used to Estimate Lead Tap Sampling Monitoring Schedules

under the Pre-2021 LCR

Monitoring Frequency

Description



• System serves 50,000 or fewer people and its latest lead or copper action level



exceedance (ALE) of 15 ng/L or 1.3 mg/L, respectively, occurred after 12/31/2019.

6-Month at

Thus, the system did not have two consecutive 6-month monitoring periods without a

standard number of

lead and/or copper ALE.

sites

• System serves more than 50,000 people, has CCT, and a lead ALE, or a 59 violation that
indicates non-compliance with State-specified optimal water quality parameters
(OWQPs), any of which occurred after 12/31/2019.



• System serves 50,000 or fewer people and its latest lead or copper ALE occurred

Annual at

between 1/1/2018 and 12/31/2019. Thus, the system had two consecutive 6-month

reduced number of

monitoring periods without a lead or copper ALE.

sites

• System serves more than 50,000 people, has CCT, and its latest lead ALE a 59 violation)
occurred between 1/1/2018 and 12/31/2019.



• System serves 50,000 or fewer people and meets one of the following criteria:



a. Any lead or copper ALE occurred before 1/1/2018; or



b. Does not meet the criteria for 9-year monitoring listed below.

Triennial at

• System serves >50,000 people and meets one of the following criteria:

reduced number of

a. Satisfies the b3 criteria1; or

sites

b.	Has CCT, no lead ALE, and no 59 violation for the most recent 3 or more
consecutive years; or

c.	Has CCT, no lead ALE, and a 59 violation for which the system has achieved
compliance for at least the 3 most consecutive years.



• System serves a population of < 1,0002 and meet all of the following conditions in a.



through e.:

Every 9 years at

a.	Is a mobile home park (CWS only)2, and

b.	Has no CCT2, and

c.	Had no lead or copper exceedances during 1992 - 20203, and

reduced number of
sites

d.	The sampling period is for both lead and copper sampling is > 9 years3 or no 90th
percentile data were reported to SDWIS/Fed4.

e.	Has a first reported date in SDWIS/Fed on or after January 1,1989.5

Source: For additional information see "Pb Schedules_CWS_Final.xlsx" and "Pb Schedules_NTNCWS_Final.xlsx,"
both available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

68 The pre-2021 LCR defines three broad size categories: Systems serving more than 50,000 people, systems
serving 3,301 to 50,000 people, and systems serving 3,300 or fewer people. Some of the requirements of the rule
varied across these size categories.

Final LCR! Economic Analysis

3-82

October 2024


-------
Notes:

1	For purposes of this analysis, the EPA identified a systems as a b3 system if it met all of the following criteria: 1)
served more than 50,000 people; 2) had a reported"B3" milestone; 3) did not have CCT (refer back to Section
3.3.3); and 4) did not have a lead or copper ALE and all reported lead 90th percentile values are < 5 ng/L or non-
detect during 1992 -2020.

2	SDWIS/Fed does not have a milestone or other required reporting that identifies systems on 9-year monitoring.
Although the rule allows systems serving < 3,300 people to qualify for 9-year monitoring, the EPA assumed only a
subset of systems serving < 1,000 people met this requirement. The EPA further assumed only water systems that
became active after January 1,1989 (based on the first reported date) would qualify for 9-year monitoring. The
EPA selected this date because it is well after when systems stopped using LSLs and when all States had to adopt
the lead provisions (i.e., by August 6,1988) that limited the amount of lead in plumbing materials.

3The length of the tap sampling period was determined by the difference between the sampling period begin and
end dates. The EPA assumed if the difference was greater than one year, but the system did not meet the 9-year
monitoring criteria, it was on triennial monitoring.

4The pre-2021 LCR only requires States to report 90th percentile levels to SDWIS/Fed that are above the lead AL for
systems serving < 3,300 people and above the copper AL for any size system.

5 The first reported date may indicate when the system became operational. The 1986 SDWA Amendments banned
the use of lead pipe and required the use of "lead-free" solders, fluxes, pipes and pipe fittings in the installation or
repair of public water systems. States were required to implement this ban by August 6,1988. The EPA assumed
these systems that came on-line after 1988 and the system and customers they serve would be more likely to use
lead-free plumbing materials that would allow the system to meet the requirements for a 9-year monitoring
waiver.

Based on the criteria in Exhibit 3-38 the majority of CWSs are on triennial monitoring under the pre-
2021 LCR. See Exhibit 3-39 and Exhibit 3-40, for additional detail on CWSs with CCT and without CCT,
respectively. Note that the text in red font and italics are variable names of the costing inputs for the
SafeWater LCR model.

Exhibit 3-39: Estimated Percentage of CWSs with CCT on Various Lead Tap Monitoring
Schedules by Size and Source Type under the Pre-2021 LCR

System Size
(Population
Served)

CWS with CCT: Surface Water

CWS with CCT: Ground Water

6 Month
(Standard)

Annual
(Reduced)

Triennial
(Reduced)

9 Year
(Reduced)

6 Month
(Standard)

Annual
(Reduced)

Triennial
(Reduced)

9 Year
(Reduced)

A

B

C

D

E

F

G

H

1-(B+C+D)

p_tap_annual

p_tap_triennial

p_tap_nine

l-( F+G+H)

p_tap_annual

p_tap_triennial

p_tap_nine

<100

1.2%

2.2%

96.5%

0%

2.8%

5.5%

91.7%

0%

101-500

1.7%

2.5%

95.8%

0%

2.5%

3.7%

93.9%

0%

501-1,000

0.5%

1.5%

97.9%

0%

2.6%

3.8%

93.6%

0%

1,001-3,300

0.7%

2.3%

97.0%

0%

2.6%

3.4%

94.0%

0%

3,301-10,000

0.6%

1.4%

98.0%

0%

1.5%

2.6%

95.8%

0%

10,001-50,000

0.5%

1.5%

98.0%

0%

1.2%

1.6%

97.3%

0%

50,001-
100,000

2.5%

2.5%

95.0%

0%

0.6%

1.3%

98.1%

0%

100,001-1M

2.9%

1.7%

95.3%

0%

1.4%

0.0%

98.6%

0%

>1M

0.0%

0.0%

100.0%

0%

0.0%

0.0%

100.0%

0%

Final LCR! Economic Analysis

3-83

October 2024


-------
Source: For additional information, see "Pb Schedules_CWS_Final.xlsx," available in the docket at EPA-HQ-OW-

2022-0801 at www.regulations.gov.

Notes:

1.	Refer to Exhibit 3-38 for the criteria the EPA applied to determine systems' lead and copper tap monitoring
schedules and Section 3.3.3 for the criteria the EPA used to identify systems with and without CCT.

2.	Systems on annual, triennial, or 9-year monitoring collect samples at the reduced number of sites specified in
the rule (see 40 CFR 141.86(c)). As will be discussed in Section 3.3.7.2, under the 2021 LCRR and final LCRI,
systems monitoring annually must collect from the standard number of sites.

Exhibit 3-40: Estimated Percentage of CWSs without CCT on Various Lead Tap Monitoring
Schedules by Size and Source Type under the Pre-2021 LCR

System Size
(Population
Served)

CWS without CCT: Surface Water

CWS without CCT: Ground Water

6 Month
(Standard)

Annual
(Reduced)

Triennial
(Reduced)

9 Year
(Reduced)

6 Month
(Standard)

Annual
(Reduced)

Triennial
(Reduced)

9 Year
(Reduced)

A

B

C

D

E

F

G

H

1-(B+C+D)

p_tap_annual

p_tap_triennial

p_tap_nine

1-(F+G+H)

p_tap_annual

p_tap_triennial

p_tap_nine

<100

1.7%

3.9%

92.5%

1.9%

1.1%

2.2%

95.4%

1.4%

101-500

1.3%

3.2%

94.3%

1.2%

1.0%

2.3%

95.9%

0.8%

501-1,000

0.9%

2.9%

95.8%

0.3%

1.0%

1.5%

97.2%

0.4%

1,001-3,300

0.6%

1.9%

97.4%

0.0%

0.8%

1.4%

97.9%

0.0%

3,301-10,000

1.2%

1.2%

97.6%

0.0%

0.2%

0.8%

99.0%

0.0%

10,001-50,000

0.4%

0.4%

99.1%

0.0%

0.6%

0.7%

98.7%

0.0%

50,001-100,000

0.0%

0.0%

100.0%

0.0%

0.0%

0.0%

100.0%

0.0%

100,001-1M

0.0%

0.0%

100.0%

0.0%

0.0%

0.0%

100.0%

0.0%

>1M

















Source: For additional information, see "Pb Schedules_CWS_Final.xlsx," available in the docket at EPA-HQ-OW-

2022-0801 at www.regulations.gov.

Notes:

1.	Refer to Exhibit 3-38 for the criteria the EPA applied to determine systems' lead and copper tap monitoring
schedules and Section 3.3.3 for the criteria the EPA used to identify systems with and without CCT.

2.	Systems on annual, triennial, or 9-year monitoring collect samples at the reduced number of sites specified in
the rule (see 40 CFR 141.86(c)). As will be discussed in Section 3.3.7.2, under the 2021 LCRR and final LCRI,
systems monitoring annually must collect from the standard number of sites.

3.	The gray shaded cells denote that there were no CWSs serving >1M people without CCT in the CWS inventory.

Exhibit 3-41 and Exhibit 3-42 provide similar information for NTNCWSs with CCT and without CCT,

respectively under the pre-2021 LCR.

Final LCRI Economic Analysis

3-84

October 2024


-------
Exhibit 3-41: Estimated Percentage of NTNCWSs with CCT on Various Lead Monitoring
Schedules by Size and Source Type under the Pre-2021 LCR

System Size
(Population
Served)

NTNCWS with CCT: Surface Water

NTNCWS with CCT: Ground Water

6 Month
(Standard)

Annual
(Reduced)

Triennial
(Reduced)

9 Year
(Reduced)

6 Month
(Standard)

Annual
(Reduced)

Triennial
(Reduced)

9 Year
(Reduced)

A

B

C

D

E

F

G

H

1-(B+C+D)

p_tap_annual

p_tap_triennial

p_tap_nine

1-(F+G+H)

p_tap_annual

p_tap_triennial

p_tap_nine

<100

0.0%

8.0%

92.0%

0.0%

4.7%

10.7%

84.7%

0.0%

101-500

4.3%

4.3%

91.3%

0.0%

4.5%

6.1%

89.4%

0.0%

501-1,000

4.5%

0.0%

95.5%

0.0%

4.7%

8.6%

86.7%

0.0%

1,001-3,300

0.0%

10.7%

89.3%

0.0%

4.4%

5.6%

90.0%

0.0%

3,301-10,000

0.0%

11.1%

88.9%

0.0%

4.0%

12.0%

84.0%

0.0%

10,001-50,000

0.0%

0.0%

100.0%

0.0%

0.0%

33.3%

66.7%

0.0%

50,001-100,000

100.0%

0.0%

0.0%

0.0%









100,001-1M

0.0%

0.0%

100.0%

0.0%









>1M

















Source: For additional information, see "Pb Schedules_NTNCWS_Final.xlsx," available in the docket at EPA-HQ-

OW-2022-0801 at www.regulations.gov.

Notes:

1.	Refer to Exhibit 3-38 for the criteria the EPA applied to determine systems' lead and copper tap monitoring
schedules and Section 3.3.3 for the criteria the EPA used to identify systems with and without CCT.

2.	Systems on annual, triennial, or 9-year monitoring collect samples at the reduced number of sites specified in
the rule (see 40 CFR 141.86(c)). As will be discussed in Section 3.3.7.2, under the 2021 LCRR and final LCRI,
systems monitoring annually must collect from the standard number of sites.

3.	The gray shaded cells denote that for NTNCWSs with CCT, no SW NTNCWSs serve more than 1 M people and
no GW NTNCWSs serve more than 50,000 people i.

Exhibit 3-42: Estimated Percentage of NTNCWSs without CCT on Various Lead Tap Monitoring
Schedules by Size and Source Type under the Pre-2021 LCR

System Size
(Population
Served)

NTNCWS without CCT: Surface
Water

NTNCWS without CCT: Ground
Water

6 Month
(Standard)

Annual
(Reduced)

Triennial
(Reduced)

9 Year
(Reduced)

6 Month
(Standard)

Annual
(Reduced)

Triennial
(Reduced)

9 Year
(Reduced)

A

B

C

D

E

F

G

H

1-(B+C+D)

p_tap_annual

p_tap_triennial

p_tap_nine

l-( F+G+H)

p_tap_annual

p_tap_triennial

p_tap_nine

<100

1.3%

2.2%

64.0%

32.4%

2.1%

3.7%

73.9%

20.3%

101-500

2.5%

5.0%

78.1%

14.4%

2.1%

3.7%

82.7%

11.6%

501-1,000

1.5%

7.5%

73.1%

17.9%

2.0%

2.5%

85.3%

10.2%

1,001-3,300

4.4%

5.5%

90.1%

0.0%

2.5%

3.9%

93.6%

0.0%

3,301-10,000

2.0%

0.0%

98.0%

0.0%

1.3%

1.3%

97.4%

0.0%

10,001-50,000

0.0%

0.0%

100.0%

0.0%

0.0%

6.7%

93.3%

0.0%

Final LCRI Economic Analysis

3-85

October 2024


-------
System Size
(Population
Served)

NTNCWS without CCT: Surface
Water

NTNCWS without CCT: Ground
Water

6 Month
(Standard)

Annual
(Reduced)

Triennial
(Reduced)

9 Year
(Reduced)

6 Month
(Standard)

Annual
(Reduced)

Triennial
(Reduced)

9 Year
(Reduced)

A

B

C

D

E

F

G

H

1-(B+C+D)

p_tap_annual

p_tap_triennial

p_tap_nine

l-( F+G+H)

p_tap_annual

p_tap_triennial

p_tap_nine

50,001-100,000

















100,001-1M

















>1M

















Source: For additional information, see "Pb Schedules_NTNCWS_Final.xlsx," available in the docket at EPA-HQ-
OW-2022-0801 at www.regulations.gov.

Notes:

1	Refer to Exhibit 3-38 for the criteria the EPA applied to determine systems' lead and copper tap monitoring
schedules and Section 3.3.3 for the criteria the EPA used to identify systems with and without CCT.

2	Systems on annual, triennial, or 9-year monitoring collect samples at the reduced number of sites specified in
the rule (see 40 CFR 141.86(c)). As will be discussed in Section 3.3.7.2, under the 2021 LCRR and final LCRI,
systems monitoring annually must collect from the standard number of sites.

3	The gray shaded cells denote no NTNCWSs without CCT serve more than 50,000 people, regardless of their
water source.

3.3.7.2 Estimating Lead and Copper Tap Monitoring Schedules under the 2021 LCRR

To determine the initial monitoring requirements under the 2021 LCRR, the EPA assumed all systems
with lead content or unknowns would monitor semi-annually for the first year of the analysis period
(Year 4) with the exception of systems in Michigan because they would have already monitored
according to the new sampling protocol required under the 2021 LCRR prior to the rule's compliance
date. As a simplifying approach, the EPA modeled all water systems in Michigan as having all non-lead
service lines.69

For systems with all non-lead service lines that do not exceed the lead AL of 15 ng/L under the 2021
LCRR, the EPA assumed systems will retain their monitoring schedule from the pre-2021 LCR. Thus, they
would have the same likelihood of being on one of the four monitoring schedules presented in Exhibit
3-39 through Exhibit 3-42, except that those qualifying for annual monitoring must collect the standard
number of samples under the 2021 LCRR.

Systems with all non-lead service lines that have:

•	A lead ALE under the 2021 LCRR must monitor semi-annually at the standard number of sites
until they qualify for reduced monitoring.

•	A lead TLE under the 2021 LCRR but no lead or copper ALE must monitor annually at the
standard number of sites.

69 There is uncertainty in using this approach because Michigan did not require first- and fifth-liter samples for
systems with GRR service lines but no LSLs. For these systems, the burden and cost for lead tap monitoring may be
underestimated.

Final LCRI Economic Analysis

3-86

October 2024


-------
As a simplifying assumption, the EPA assumed all systems will begin their monitoring cycle in Year 4 of
the analysis period resulting in an overestimation of sampling costs associated with the 2021 LCRR.

3.3.7.3 Estimating Lead and Copper Tap Monitoring Schedules under the LCRI

To determine the initial monitoring requirements under the LCRI, the EPA assumed all systems with lead
content would monitor semi-annually for the first year (Year 4) with the exception of systems in
Michigan because they would have monitored according to the new sampling protocol required under
the LCRI prior to the rule's compliance date.

Systems with no lead content service lines or unknowns that do not exceed the lead AL of 10 ng/L under
the LCRI will retain their monitoring schedule from the pre-2021 LCR. Thus, they would have the same
likelihood of being on one of the four monitoring schedules presented in Exhibit 3-39 through Exhibit
3-42, except that those qualifying for annual monitoring must collect the standard number of samples
under the LCRI. Systems with no lead content service lines that have a lead ALE or OWQP violation must
monitor semi-annually at the standard number of sites until they qualify for reduced monitoring.70 As a
simplifying assumption, the EPA assumed all systems will begin their monitoring cycles in Year 4 of the
analysis period resulting in an overestimation of sampling costs associated with the LCRI.

Exhibit 3-43 provides a comparison of the criteria for increased and reduced tap sample monitoring
under the pre-2021 LCR, 2021 LCRR, and final LCRI.

70 As a simplifying assumption, the tap monitoring schedules do not take into account copper ALEs, which are
handled separately as described in Section 3.3.5.4.

Final LCRI Economic Analysis

October 2024

3-87


-------
Exhibit 3-43: Comparison of the Criteria for Standard and Reduced Tap Sample Monitoring under the Pre-2021 LCR, 2021 LCRR,

and Final LCRI

Frequency
and # of
Samples

Pre-2021 LCR Criteria for
Lead and Copper

2021LCRR
Criteria for Lead

2021LCRR
Criteria for Copper

Final LCRI
Criteria for Lead

Final LCRI
Criteria for Copper

Semi-

Annually at
Routine
Number of
Sites

•	Lead and/or copper ALE1
during any tap sampling
monitoring period;
and/or

•	Has an OWQP excursion2
for more than 9 days in a
6-month period.

•	Has a lead ALE1 during
any tap monitoring
period;

•	After State sets OWQPs
following CCT installation
or re-optimization;

•	Lead ALE or has an
OWQP excursion2 for
more than 9 days in a 6-
month period;

•	New water systems that
begin operation after
effective date; or

•	Initial monitoring:
Systems with LSLs
including b3 systems3
unless have prior
monitoring data.4

•	Cu90 is > 1.3 mg/L during
any tap monitoring
period; and/or

•	Has an OWQP excursion2
for more than 9 days in a
6-month period; or

•	New water systems that
begin operation after
effective date.

•	Has a lead and/or
copper ALE5 during
any tap monitoring
period; and/or

•	Has an OWQP
excursion2 for more
than 9 days in a 6-
month period, or

•	New water systems
that begin operation
after effective date.

•	Initial monitoring:
Systems with lead
and GRR service
lines, including b3
systems3, unless have
prior monitoring
data.4

• Same criteria as lead.

Annually at
Standard
Number of
Sites

N/A

• No lead or copper ALE, &
meets OWQP
specifications (if
applicable) for 2
consecutive 6-month tap
monitoring periods.

• Has a lead TLE6, no lead
or copper ALE, & meets
OWQP specifications (if
applicable) for 2
consecutive 6-month tap
monitoring periods.

• No lead and/or
copper ALE5 & meets
OWQP specifications
(if applicable) for 2
consecutive 6
months.

• N/A. Systems that
qualify for annual
monitoring collect
copper samples at
the reduced number
of sites.

Final LCRI Economic Analysis

3-88

October 2024


-------
Frequency
and # of
Samples

Pre-2021 LCR Criteria for
Lead and Copper

2021LCRR
Criteria for Lead

2021LCRR
Criteria for Copper

Final LCRI
Criteria for Lead

Final LCRI
Criteria for Copper

Annually at
Reduced
Number of
Sites

•	Serves < 50,000 people:
No lead or copper ALE for
2 consecutive 6-month
tap monitoring periods.

•	All sizes: No lead ALE &
meets OWQP.
specifications (if
applicable) for 2
consecutive 6-month tap
monitoring periods.

N//A. Systems cannot
monitor annually at the
reduced number of sites for
lead.

• No lead TLE6, no copper
ALE, & meets OWQP
specifications (if
applicable) for 2
consecutive 6-month tap
monitoring periods.

N/A

• No lead or copper
ALE & meets OWQP
specifications (if
applicable) for 2
consecutive 6-month
tap monitoring
periods.

Triennially
at Reduced
Number of
Sites

•	Serves < 50,000 people:
No lead or copper ALE for
3 consecutive years.

•	All sizes:

o No lead ALE and met
OWQP specifications
for 3 consecutive
years; or
o Meet 40 CFR

141.81(b)(3);2 or
o Meets accelerated
reduced criteria for
/lead and copper.6

•	Serves < 50,000 people:
No lead TLE5 or ALE, no
copper ALE, & meets
OWQP specifications (if
applicable) for > 3
consecutive years.

•	All sizes: Meets
accelerated reduced
criteria for lead.6

•	Serves < 50,000 people:
Cu90 is < 1.3 mg/L and
meets OWQP
specifications for 3
consecutive years.

•	All sizes:

o Meets 40 CFR

141.81(b)(3) criteria3
and OWQP
specifications (if
applicable); or
o Meets accelerated
criteria for copper.6

•	No lead or copper
ALE5 & meets OWQP
specifications (if
applicable) for 3
consecutive years
(with State
approval); or

•	Meets 40 CFR
141.81(b)(3) criteria3
and OWQP
specifications (if
applicable); or

•	Meets accelerated
reduced criteria for
lead and copper.7

• Same criteria as lead.

Every Nine
Years at
Reduced
Number of
Sites

Serves < 3,300 people: Lead and copper 90th percentile levels are < 5 ng/L and < 0.65 mg/L, respectively, and all plumbing
materials are free of lead- and copper-containing materials including those in buildings and residences served by the system.

Final LCRI Economic Analysis

3-89

October 2024


-------
1 Under the pre-2021 LCR and 2021 LCRR, a system has a lead ALE if its 90th percentile lead level and 90th percentile copper level were above 15 ng/L and 1.3
mg/L, respectively.

2OWQPs are measured to determine whether a system is operating its CCT at a level that most effectively minimizes the lead and copper concentrations at
users' taps. An excursion occurs when the daily value of a WQP is below the minimum value or outside the OWQP range set by the State. This definition is the
same for the pre-2021 LCR, 2021 LCRR, and LCRI.

3 Under the pre-2021 LCR, a system met the criteria in 40 CFR 141.81(b)(3) if for two consecutive 6-month monitoring periods, the system's lead 90th percentile
level minus its highest source water level was < 5 ng/L or its source water lead was less than the lead method detection limit and its P90 was < 5 ng/L. The
2021 LCRR modified these criteria to specify they are met if for two consecutive six-month tap sampling monitoring periods, the system's lead 90th percentile
level is < the practical quantitation limit of 0.005 mg/L. The LCRI further expands the "b3" criteria in the 2021 LCRR to specify that the water system cannot
have State-designated OWQPs. Under the 2021 LCRR and LCRI, the initial monitoring period refer to the first monitoring period under 2021 LCRR or LCRI.
4Systems that have conducted monitoring that meets the site location and sampling protocol between the date the rule was published in the Federal Register
and three year can use that data to determine their sampling schedule in lieu of conducting initial monitoring.

5	Under the LCRI, a system has a lead ALE if its 90th percentile level is above the new AL of 10 ng/L. The EPA has not modified the definition of a copper ALE
(see note 1).

6	Under the 2021 LCRR, a system has a lead TLE if its 90th percentile is above 10 ng/L but not above 15 ng/L.

7	Systems with a lead 90th percentile level of < 0.005 mg/L and copper 90th percentile level of < 0.65 mg/L for 2 consecutive 6-month tap sampling monitoring
periods can qualify for triennial monitoring at the reduced number of sites. Under the 2021 LCRR, lead and copper are evaluated separately, such that a system
could qualify for to monitor for lead only at a triennial schedule but not copper or vice versa. In addition, under the 2021 LCRR and LCRI systems with CCT must
also be in compliance with their OWQPs.

Final LCRI Economic Analysis

3-90

October 2024


-------
3.3.7.4 Discussion of Data Limitations and Uncertainty

As previously discussed, for systems serving 3,300 or fewer people, the pre-2021 LCR required States to
only report those lead 90th percentile values that exceed the lead AL of 15 ng/L, but to report all lead
90th percentile values for larger water systems. To determine if systems in this smallest size category
were underrepresented, the EPA estimated the percentage of systems with any reported lead 90th
percentile data during 2012 - 2020 within this size category, as well as for systems serving 3,301 to
50,000 people and greater than 50,000 people. As shown in Column C of Exhibit 3-44 below, lead 90th
percentile data were reported for about 72 percent of all CWSs in this smallest size category compared
to more than 98 percent in the larger two categories. The EPA also estimated the percentage of CWSs in
which only lead exceedance data were reported to try to assess any bias in reporting for the smallest
size category. As shown in Column F of Exhibit 3-44, in general, both exceedances and non-exceedances
were reported for approximately 98 percent of systems that reported any lead 90th percentile data in
the smallest size category, and essentially all of those in the highest two categories. This indicates that
most States report exceedance and non-exceedance data for even the smallest size category.

Exhibit 3-44: Estimated Number and Percentage of CWSs with Reported Lead ALEs Only under

the Pre-2021 LCR (2012-2020)

System Size
(Population
Served)

All CWSs

All CWSs w/ any Reported
Lead 90th Percentile Data

All CWSs w/ Reported P90 Data
Above the Lead Action Level Only of
15 M-g/L

Number

Number

Percent of All
CWSs

Number

Percent of All
CWSs that
Only Report
Lead ALEs

Percent of
CWSs w/
Reported P90
Data that Only
Report Lead
ALEs

A

B

C= B/A

D

<
Q*

II
LU

CO

Q*

II

U-

< 3,300

40,113

29,046

72.41%

533

1.33%

1.84%

3,301 to 50,000

8,400

8,296

98.76%

1

0.01%

0.01%

> 50,000

1,016

997

98.13%

1

0.10%

0.10%

Total

49,529

38,339



535

1.08%

1.40%

Source: SDWIS/Fed'4th quarter 2020 freeze. Also see, "Extent of P90 Data_LCR_Final.xlsx."

Notes:

General:

A: Includes all active CWSs in SDWIS/Fed based on fourth quarter 2020 freeze, current through December 31,
2020.

B: Includes CWSs with one or more P90 value reported to SDWIS/Fed during 2012- 2020.

C: Note, for the Proposed LCRI EA, the EPA used the most recent sample start date (i.e., the date that denotes the
start of the monitoring period). The EPA revised its approach for the Final LCRI EA to instead use the lowest value
reported when a system reported more than one result in a single year that had the same sample start date. By
using the minimum value versus the most recent, the EPA more accurately captured systems that only reported
lead levels above the lead AL

D: Includes the subset of CWSs for which all reported P90 values are above the pre-2021 LCR AL of 15 ng/L.

Final LCRI Economic Analysis

3-91

October 2024


-------
3.3.8 Water Quality Parameter Monitoring

Under the pre-2021 LCR, 2021 LCRR, and final LCRI, water systems can reduce the frequency, and in
some cases the number of samples for WQP monitoring in the distribution system (also referred to as
WQP tap monitoring) based on their 90th percentile lead levels and compliance with their OWQPs. Note
that under all three versions of the LCR rule, systems cannot qualify for reduced WQP monitoring that
occurs at entry points to the distribution system.

The EPA determined the initial WQP tap monitoring schedules under the pre-2021 LCR, the 2021 LCRR,
and the final LCRI separately to reflect differences in the WQP monitoring requirements and 90th
percentile calculations. The monitoring schedules presented in these sections are the estimated
schedules that systems start with during the first period of cost modeling. Sections 3.3.8.1 and 3.3.8.2
describe the development of the WQP tap sampling schedules for the 2021 LCRR and the final LCRI,
respectively. Also see Appendix B for a discussion of the derivation of the WQP tap monitoring schedules
under the pre-2021 LCR. A discussion of data limitations and uncertainty associated with the analysis are
provided in Section 3.3.8.3.

3.3.8.1 2021 LCRR

Under the 2021 LCRR, systems cannot conduct reduced WQP tap monitoring on a triennial schedule, as
was allowed under the LCR. The 2021 LCRR also maintains the requirement for systems to conduct
monitoring at entry points to the distribution system with no allowance for reduced entry point
monitoring at a frequency less than every two weeks.

For modeling purposes, the EPA assumed the following to estimate a system's lead WQP tap monitoring
schedule that starts in Year 4 of the 35-year period of analysis:

1.	Systems serving 50,000 or fewer people are only required to conduct WQP monitoring under
the following circumstances:

•	Six-month monitoring when they have a lead TLE and;

•	For two consecutive six-month monitoring periods after installation or re-optimization of
CCT. These systems are not required to conduct long-term WQP monitoring to comply with
OWQPs under the 2021 LCRR unless they are required by the State (not modeled) or have
an ongoing lead TLE. Thus, the EPA assumed no systems serving 50,000 or fewer people
would conduct WQP tap monitoring at a reduced rate.

2.	Systems serving more than 50,000 people with CCT71 and

•	A lead TLE would be on six-month routine WQP monitoring for as long as they exceed the
TLE.

•	No lead TLE would be eligible for reduced WQP distribution monitoring. Eligibility is based
on the water system meeting optimal water quality parameters (OWQPs).

71 All systems serving more than 50,000 people were required to install CCT, except b3 systems. See Section 3.3.3
for an explanation of how the EPA derived the number of b3 systems.

Final LCRI Economic Analysis

3-92

October 2024


-------
3.	The EPA used the lead 90th percentile classification, as described in Section 3.3.5.1.1, to
determine if a system of any size has a TLE beginning in Year 4 of the rule analysis period.

4.	The EPA modeled systems with a copper ALE separately.72

Exhibit 3-45 provides a summary of how the EPA used SDWIS/Fed data (current through December 31,
2020) to determine if a system serving more than 50,000 people without a lead TLE or copper ALE would
qualify for reduced WQP tap monitoring based on compliance with OWQPs. Systems that do not meet
the reduced monitoring criteria are assumed to monitor semi-annually at the standard number of sites
(see Exhibit 4-19 in Chapter 4 for the number of standard and reduced WQP tap monitoring sites).

Exhibit 3-45: SDWIS/Fed Data Criteria Used to Determine Reduced WQP Tap Monitoring
Schedules for Systems Serving > 50,000 People With CCT and No Lead TLE or Copper ALE

Monitoring Frequency

Criteria

6-Month (Reduced number
of sites)1

System with CCT serves > 50,000 people and for > 1 but < 3 years, it is in
compliance with its OWQPs (i.e., no 59 violation or a 59 violation for which the
system has achieved compliance).2

Annual (Reduced number of
sites)1

System with CCT serves > 50,000 people and for a minimum of 3 consecutive years,
is in compliance with its OWQPs (i.e., no 59 violation or a 59 violation for which the
system has achieved compliance).2

Acronyms: CCT = corrosion control treatment; OWQP = optimal water quality parameters.

Notes:

1See Exhibit 4-19 in Chapter 4 for the number of distribution system sites required for reduced monitoring.

2 Based on analysis of SDWIS/Fed data from 2012 through 2020. To meet the reduced monitoring criteria, systems
with an OWQP violation must have achieved compliance, denoted by the SDWIS code of SOX or EOX by December
31, 2019.

Exhibit 3-46 and Exhibit 3-47 provide the percentage of CWSs serving more than 50,000 people with CCT
and no lead TLE or copper ALE on each of the three possible WQP tap monitoring schedules under the
2021 LCRR by source water type based on analysis of SDWIS/Fed data for 2012 through 2020. For
NTNCWSs, this information is provided in Exhibit 3-48 for SW systems. Note that no GW NTNCWS serves
more than 50,000 people. Also, these exhibits exclude systems without CCT because WQP monitoring to
comply with OWQPs is not required for these systems. The exhibits show that:

•	All CWS GW systems and the majority of CWS SW systems (97.8 to 100%) serving more than
50,000 people met the criteria for annual reduced WQP tap monitoring.

•	Of the two SW NTNCWSs that serve more than 50,000 people, one met the criteria for annual
reduced WQP tap monitoring and the other is on six-month standard monitoring.

72 As discussed in Chapter 4, Section 4.3.2.3.1, the SafeWater LCR models copper WQP monitoring separately from
the lead WQP monitoring. To avoid double counting the cost of WQP monitoring for systems experiencing both a
copper ALE and a lead ALE simultaneously, the SafeWater LCR models the costs of copper and lead WQP
monitoring separately and restricts copper WQP monitoring to systems with a copper ALE only (SafeWater input:
p_copper_ale) and lead 90th percentile not greater than the lead AL.

Final LCR! Economic Analysis

3-93

October 2024


-------
Exhibit 3-46: Percentage of Ground Water CWSs Serving > 50,000 People with CCT and No
Lead TLE or Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting
in the First Modeled Compliance Period Given 2021 LCRR Requirements)

System Size
(Population Served)

6 Month (Standard)

6 Month (Reduced)

Annual(Reduced)

A = 1- (B+C)

p_wqp_six_red

p_wqp_annual

B

C

50,001-100,000

0.0%

0.0%

100.0%

100,001-1,000,000

0.0%

0.0%

100.0%

>1 M

0.0%

0.0%

100.0%

Source: For additional information, see "WQP Schedules_CWS_LCRR_Final.xlsx," available in the docket at EPA-HQ-
OW-2022-0801 at www.regulations.gov.

Note: Percentages are based on OWQP violation and compliance data reported to SDWIS/Fed for 2012 - 2020 in
the fourth quarter frozen 2020 dataset, current through December 31, 2020.

Exhibit 3-47: Percentage of Surface Water CWSs Serving > 50,000 People with CCT and No
Lead TLE or Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting
in the First Modeled Compliance Period Given 2021 LCRR Requirements)

System Size

6 Month (Standard)

6 Month (Reduced)

Annual(Reduced)

A = 1- (B+C)

p_wqp_six_red

p_wqp_annual

(Population Served)

B

C

50,001-100,000

1.1%

1.1%

97.8%

100,001-1,000,000

1.6%

0.0%

98.4%

>1 M

0.0%

0.0%

100.0%

Source: For additional information, see "WQP Schedules_CWS_LCRR_Final.xlsx," available in the docket at EPA-HQ-
OW-2022-0801 at www.regulations.gov.

Note: Percentages are based on OWQP violation and compliance data reported to SDWIS/Fed for 2012 - 2020 in
the fourth quarter frozen 2020 dataset, current through December 31, 2020.

Exhibit 3-48: Percent of Surface Water NTNCWSs Serving > 50,000 People with CCT and No
Lead TLE or Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting
in the First Modeled Compliance Period Given 2021 LCRR Requirements)

System Size
(Population Served)

6 Month (Standard)

6 Month (Reduced)

Annual(Reduced)

A = 1- (B+C)

p_wqp_six_red

p_wqp_annual

B

C

50,001-100,000

100.0%

0.0%

0.0%

100,001-1,000,000

0.0%

0.0%

100.0%

>1 M







Source: For additional information, see "WQP Schedules_NTNCWS_LCRR_Final.xlsx," available in the docket at

Final LCRI Economic Analysis

3-94

October 2024


-------
EPA-HQ-OW-2022-0801 at www.regulations.gov.

Notes:

1.	Percentages are based on OWQP violation and compliance data reported to SDWIS/Fed for 2012 - 2020 in the
fourth quarter frozen 2020 dataset, current through December 31, 2020.

2.	The gray shaded cells denote that no SW NTNCWS serves more than 1 million people.

3.3.8.2 Final LCRI

Under the final LCRI, all systems serving more than 10,000 people with CCT, systems without CCT that
serve 10,001 to 50,000 people that have a lead or copper ALE, and systems serving 10,000 or fewer
people that have a lead or copper ALE must conduct WQP sampling at entry points to the distribution
system and within the distribution system.73 Systems serving more than 10,000 people with CCT can
qualify for reduced WQP monitoring in the distribution system if for at least two consecutive six-month
monitoring periods they are in compliance with their State-set OWQP ranges or minimums and their
lead and copper 90th percentile levels are at or below the action levels of 10 ng/L and 1.3 mg/L,
respectively. The number of consecutive monitoring periods in which a system meets its OWQPs
determines if a system qualifies for reduced semi-annual or annual monitoring. Under the final LCRI, as
maintained from the 2021 LCRR, systems cannot conduct reduced WQP tap monitoring on a triennial
schedule. The final rule also maintains the requirement for systems to conduct monitoring at entry
points to the distribution system with no allowance for reduced entry point monitoring at a frequency
less than every two weeks.

For modeling purposes, the EPA applied the same approach for the LCRI as was used for the 2021 LCRR,
as previously described in Section 3.3.8.1, with one exception. Systems serving 10,001 to 50,000 people
with CCT must continue monitoring under the final LCRI irrespective of their lead 90th percentile level,
comply with State-set OWQP ranges or minimums, and are eligible for reduced monitoring. Because
historical SDWIS/Fed data on OWQP compliance is not available for these systems, the EPA assumed the
percent of these systems that qualify for reduced monitoring would be the same as for systems serving
50,001 to 100,000 people.

Exhibit 3-49 and Exhibit 3-50 provide the percentage of CWSs with CCT and no lead or copper ALE on
each of the three possible distribution system monitoring schedules under the final LCRI by source water
type based on analysis of SDWIS/Fed data from 2012 through 2020. For NTNCWSs, this information is
provided in Exhibit 3-51 and Exhibit 3-52. The exhibits show that:

•	All GW CWSs and the majority of SW systems serving more than 10,000 people (97.8 to 100
percent) met the criteria for annual reduced WQP tap monitoring.

•	All SW NTNCWSs serving 10,001 to 100,000 people will be on standard six-month monitoring
and those serving more than 1 million people met the criteria for annual reduced WQP
monitoring.

73 The EPA set more stringent requirements under the LCRI for systems serving 10,001 to 50,000 people with CCT.
These systems must continue WQP monitoring irrespective of their lead or copper 90th percentile level. Under the
pre-2021 LCR and 2021 LCRR, these systems were only required to conduct WQP monitoring when they had a lead
or copper ALE, unless required by the State.

Final LCRI Economic Analysis

3-95

October 2024


-------
Exhibit 3-49: Percent of Ground Water CWSs Serving > 10,000 People with CCT and No Lead or
Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting in the First
Modeled Compliance Period Given Final LCRI Requirements)

System Size

6 Month (Standard)

6 Month (Reduced)

Annual(Reduced)

A = 1- (B+C)

p_wqp_six_red

p_wqp_annual

(Population Served)

B

C

10,001-50,000

0.0%

0.0%

100.0%

50,001-100,000

0.0%

0.0%

100.0%

100,001-1,000,000

0.0%

0.0%

100.0%

>1 M

0.0%

0.0%

100.0%

Source: For additional information, see "WQP Schedules_CWS_LCRI_Final.xlsx," available in the docket at EPA-HQ-

OW-2022-0801 at www.regulations.gov.

Notes:

1.	Percentages are based on OWQP violation data reported to SDWIS/Fed for 2012 - 2020 in the fourth quarter
frozen 2020 dataset, current through December 31, 2020.

2.	Under the final LCRI, systems serving 10,001 to 50,000 people with CCT must conduct WQP monitoring
irrespective of their lead 90th percentile levels. The EPA assumed that the same percent of these systems
would qualify for reduced monitoring as systems serving 50,001 to 100,000 people.

Exhibit 3-50: Percentage of Surface Water CWSs Serving > 10,000 People with CCT and No
Lead or Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting in
the First Modeled Compliance Period Given Final LCRI Requirements)

System Size

6 Month (Standard)

6 Month (Reduced)

Annual(Reduced)

A = 1- (B+C)

p_wqp_six_red

p_wqp_annual

(Population Served)

B

C

10,001-50,000

1.1%

1.1%

97.8%

50,001-100,000

1.1%

1.1%

97.8%

100,001-1,000,000

1.6%

0.0%

98.4%

>1 M

0.0%

0.0%

100.0%

Source: For additional information, see "WQP Schedules_CWS_LCRI_Final.xlsx," available in the docket at EPA-HQ-

OW-2022-0801 at www.regulations.gov.

Notes:

1.	Percentages are based on OWQP violation data reported to SDWIS/Fed for 2012 - 2020 in the fourth quarter
frozen 2020 dataset, current through December 31, 2020.

2.	Under the final LCRI, systems serving 10,001 to 50,000 people with CCT must conduct WQP monitoring
irrespective of their lead 90th percentile levels. The EPA assumed that the same percent of these systems
would qualify for reduced monitoring as systems serving 50,001 to 100,000 people.

Final LCRI Economic Analysis

3-96

October 2024


-------
Exhibit 3-51: Percentage of Ground Water NTNCWSs Serving > 10,000 People with CCT and No
Lead or Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting in
the First Modeled Compliance Period Given Final LCRI Requirements)

System Size

6 Month (Standard)

6 Month (Reduced)

Annual(Reduced)

A = 1- (B+C)

p_wqp_six_red

p_wqp_annual

(Population Served)

B

C

10,001-50,000

100.0%

0.0%

0.0%

50,001-100,000







100,001-1,000,000







>1 M







Source: For additional information, see "WQP Schedules_NTNCWS_LCRI_Final.xlsx," available in the docket at EPA-

HQ-OW-2022-0801 at www.regulations.gov.

Notes:

1.	This table includes the monitoring schedules for SW NTNCWSs serving more than 10,000 people with CCT that
did not have a lead or copper ALE based on data reported to SDWIS/Fed for 2012 - 2020 in the fourth quarter
frozen 2020 dataset, current through December 31, 2020.

2.	Under the final LCRI, systems serving 10,001 to 50,000 people with CCT must conduct WQP monitoring
irrespective of their lead 90th percentile levels. There was insufficient data to estimate monitoring schedules
based on SDWIS/Fed for this size category and no data for NTNCWS serving more than 50,000 people (see
note 3) on which to base the WQP tap monitoring schedule. Thus, the EPA conservatively assumed these
systems would be on semi-annual standard monitoring.

3.	The gray shaded cells denote that no GW NTNCWS serve more than 50,000 people.

Exhibit 3-52: Percentage of Surface Water NTNCWSs Serving > 10,000 People with CCT and No
Lead or Copper ALE on Various WQP Distribution System Monitoring Schedules (Starting in
the First Modeled Compliance Period Given Final LCRI Requirements)

System Size

6 Month (Standard)

6 Month (Reduced)

Annual(Reduced)

A = 1- (B+C)

p_wqp_six_red

p_wqp_annual

(Population Served)

B

C

10,001-50,000

100.0%

0.0%

0.0%

50,001-100,000

100.0%

0.0%

0.0%

100,001-1,000,000

0.0%

0.0%

100.0%

>1 M







Source: For additional information, see "WQP Schedules_NTNCWS_LCRI_Final.xlsx," available in the docket at EPA-

HQ-OW-2022-0801 at www.regulations.gov.

Notes:

1.	This table includes the monitoring schedules for SW CWSs serving more than 10,000 people with CCT that did
not have a lead or copper ALE based on data reported to SDWIS/Fed for 2012 - 2020 in the fourth quarter
frozen 2020 dataset, current through December 31, 2020.

2.	Under the final LCRI, systems serving 10,001 to 50,000 people with CCT must conduct WQP monitoring
irrespective of their lead 90th percentile levels. There was insufficient data to estimate monitoring schedules
based on SDWIS/Fed for this size category so the EPA assumed these systems would be on the same schedule
as systems serving 50,001 to 100,000 people.

Final LCRI Economic Analysis

3-97

October 2024


-------
3. The gray shaded cells denote that no SW NTNCWS serve more than 1M people.

3.3.8.3 Discussion of Data Limitations and Uncertainty

To estimate the WQP monitoring schedules for the 2021 LCRR, starting in Year 4, the EPA assumed
systems serving 50,000 or fewer people with CCT would discontinue WQP monitoring when they no
longer exceeded the lead TL. Similarly, the EPA assumed for the final LCRI, systems serving 10,000 or
fewer with CCT would discontinue WQP monitoring when they no longer exceeded the lead AL. Thus,
these systems would never conduct reduced WQP tap monitoring. There may be uncertainty in these
assumptions because some States may have required these smaller systems to continue to conduct
long-term WQP monitoring and comply with OWQPs regardless of whether they exceeded the lead TL or
AL, to ensure CCT is operating properly.

For the final LCRI, the EPA assumed that systems serving 10,001 to 50,000 with CCT and with no ALE
would have the same likelihood of achieving reduced monitoring status as systems serving 50,001 to
100,000 based on analysis of historical SDWIS/Fed data from 2012 to 2020. If there exists some
systematic difference between systems' ability to achieve OWQP set by States, in the 10,001 to 50,000
and 50,001 to 100,000 size categories, this assumption could result in an under- or overestimate of WQP
monitoring costs.

3.3.9 Source and Treatment Changes

This section presents the EPA's methodology for estimating the annual likelihood that a system will add
a new source or change treatment. Under the pre-2021 LCR, systems that conduct lead and copper tap
sampling less frequently than semi-annually had to report plans to add a source or make a long-term
treatment change to the State and obtain approval prior to making this change. The State could require
systems to conduct additional monitoring or take other actions it deemed appropriate in response to
this change. Under the 2021 LCRR and final LCRI, these requirements would apply to any system
regardless of its monitoring schedule.

3.3.9.1 Source Change

The EPA used historical data from SDWIS/Fed reported through December 31, 2020 to estimate the
likelihood that systems would have a source change in any given year. SDWIS/Fed assigns a unique
facility ID for each source in a system. A change in source was defined as the addition of a facility ID for a
system that was not present in the year before. The EPA evaluated source changes between 2013 and
2020.74 Note that the addition of multiple facilities was considered a single change in source if they all
occurred in the same calendar year. Only systems that had facility IDs listed in all years of the analysis
(2013-2020) were included in the analysis.

The percentage of CWSs that had a change in source was calculated for each year interval from 2013 to
2020 (i.e., 2013-2014, 2014-2015, 2015-2016, 2016-2017, 2017-2018, 2018-2019, and 2019-2020). The
values were then averaged across the seven individual year sequences. These estimates are shown in

74 The EPA expanded the analysis period from proposed LCRI EA (USEPA, 2023b) from 2017 - 2020 to 2013 - 2020
to be more consistent with other analyses (data for 2012 was not available). Based on the expanded analysis, the
estimated percentage of CWSs that will add a new source annually increased from 3.43 to 3.88 percent and for
NTNCWSs, from 1.58 to 2.78 percent.

Final LCRI Economic Analysis

3-98

October 2024


-------
Exhibit 3-53 and Exhibit 3-54 for CWSs and NTNCWSs, respectively. Although most results are similar
across CCT status and source water type, results for larger sized GW systems are high, likely due to the
small total number of systems in those size categories.

To produce estimates that can be used to predict future changes over a 30+ year time period, the EPA
combined the size categories. Specifically, the EPA calculated a weighted average using the number of
CWSs in each stratum multiplied by their result to get an overall percentage for all systems of a given
source water type and CCT status. In general, the estimates for CWSs were similar regardless of CCT
status or source water type with the exception of those serving more than 50,000 people that ranged
from about 7 to 48 percent and are based on a smaller number of systems in these size categories. The
weighted averages for systems with and without CCT were 4.75 percent and 3.48 percent, respectively.
Because of these similarities the EPA used one estimate of 3.88 percent across all CWSs in its cost
estimates, which corresponds to the SafeWater LCR model data variable, p_source_chng.

Exhibit 3-53: Estimated Percent of CWSs that Will Add a New Source Each Year

Size Category
(Population Served)

Estimated Percent of CWSs that Will Add a New Source
(Based on 2013 - 2020 SDWIS/Fed Data) Source

With CCT

Without CCT

Ground Water

Surface Water

Ground Water

Surface Water

<100

2.18%

3.31%

2.39%

3.35%

101-500

2.80%

2.96%

3.10%

2.91%

501-1,000

3.12%

3.07%

3.56%

3.01%

1,001-3,300

4.19%

3.44%

4.32%

3.08%

3,301-10,000

6.81%

4.71%

6.82%

4.77%

10,001-50,000

8.34%

5.62%

9.91%

5.96%

50,001-100,000

15.83%

8.78%

47.62%

7.14%

100,001-1M

23.34%

12.83%

14.29%

23.81%

>1M

42.86%

21.43%





WEIGHTED AVERAGE

4.60%

4.91%

3.45%

3.67%

4.75%

3.48%

3.88% (p_source_chng)

Notes:

1. For additional information, see file "Likelihood_SourceChange_Final.xlsx," available in the docket at EPA-HQ-

OW-2022-0801 at www.regulations.gov.

2. The gray shaded cells denote that no CWSs without CCT serve more than 1 million people.

In general, the estimates for NTNCWSs were similar regardless of CCT status or source water type with
two exceptions. The EPA estimated that 10.71 percent of NTNCWSs with CCT using SW and serving
3,001-10,000 and 12.09 of NTNCWSs without CCT using GW that serve 10,001 to 50,000 people would
change their source each year. These high likelihoods are based on a small number of systems in each
of these categories {i.e., 8 and 13 systems, respectively). Thus, the EPA combined size categories for
NTNCWSs and estimated a weighted average for each CCT and source stratum, for those with and

Final LCR! Economic Analysis

3-99

October 2024


-------
without CCT and all NTNCWSs. The combined weighted averages by CCT status and for all NTNCWSs
yield an estimate of 2.78 percent (p_source_chng).

Exhibit 3-54: Estimated Percent of NTNCWS that Will Change Source Each Year

Size Category (Population
Served)

Estimated Percent of NTNCWSs that Will Add a New
Source (Based on 2013 - 2020 SDWIS Data)

With CCT

Without CCT

Ground Water

Surface Water

Ground Water

Surface Water

<100

2.90%

3.73%

2.51%

5.34%

101-500

3.16%

2.04%

2.71%

3.63%

501-1,000

2.74%

1.43%

2.54%

3.65%

1,001-3,300

5.35%

4.95%

3.45%

2.42%

3,301-10,000

3.73%

10.71%

4.55%

1.59%

10,001-50,000

4.76%

0.00%

12.09%

4.76%

50,001-100,000



0.00%





100,001-1M



0.00%





>1M









WEIGHTED AVERAGE

3.21%

3.40%

2.66%

3.97%

3.22%

2.71%

2.78% (p_source_chng)

Notes:

1.	For additional information, see file "Likelihood_SourceChange_Final.xlsx," available in the docket at EPA-HQ-
OW-2022-0801 at www.regulations.gov.

2.	The gray shaded cells denote that for NTNCWSs with CCT, no GW NTNCWSs serve more than 50,000 people
and no SW NTNCWSs serve more than > 1 million people. For those without CCT, none serve more than
50,000 people, regardless of their water source.

3.3.9.1.1 Discussion of Data Limitations and Uncertainty

Although SDWIS/Fed provides the most comprehensive dataset of available system information, the
reporting of source information to SDWIS/Fed has associated uncertainties. See Section 3.2.1 for a
discussion of SDWIS/Fed. The EPA worked to minimize the impacts of these uncertainties by counting
only non-emergency sources, net increases in the number of sources, averaging results over three two-
year periods, and combining size categories to minimize over-representation of small numbers of large
systems in a single size category in order to develop a more representative prediction of changes
throughout the rule analysis period for all water systems in the United States. The EPA recognizes that
using SDWIS/Fed may underestimate the percent of systems changing sources because it does not
include systems that add and subtract the same type of source in a given calendar year.

3.3.9.2 Primary Source Change

The EPA assumed States at a minimum would require systems that change their primary source to take
additional actions such as source water monitoring. The EPA defined a change in primary source as a
year-to-year change in purchasing status and/or source type. Changes in primary source were evaluated

Final LCRI Economic Analysis

3-100

October 2024


-------
at the facility level, so that each system/facility combination was counted as a distinct change. New
systems that were not in SDWIS/Fed during the entire evaluation period of 2013 -2020 were excluded
from the analysis.75 Specifically, a change in primary source includes the following options:

•	Change from a system that receives or purchases some or all of its finished water from a
wholesale system to one that uses its own source or vice versa but continues to use GW, SW, or
ground water under the direct influence of surface water (GU, shortened from GUDWI). For
example, changes from GW purchased to GW, SW to SW purchased.

•	Change from GW to SW or GU; from SW to GW or GU; or GU to GW or SW.

•	Change in source type and purchased status, e.g., from GW purchased to SW, GU to SW
purchased.

The counts of systems meeting these criteria based on information in SDWIS/Fed for 2013 to 2020 were
small for both CWSs (4,660 system/facility combinations) and NTNCWSs (145 system/facility
combinations). This is expected since changing source water type would change the overall water
chemistry significantly and affect numerous regulatory requirements. The EPA estimated that 0.42
percent and 0.10 percent of CWSs and NTNCWSs, respectively, would change primary source water each
year. For additional information, see "Likelihood_SourceChange_Final.xlsx," available in the docket at
EPA-HQ-OW-2022-0801 at www.regulations.gov. However, for reasons cited in the next section, the EPA
assumed that one percent of CWSs and NTNCWSs would change their primary source (p_source_sig).

3.3.9.2.1 Discussion of Data Limitations and Uncertainty

The EPA estimate of one percent of systems changing their primary source water type in a given year
over the rule analysis period is uncertain given that the agency is using historical data to predict future
rates of change. The EPA found that very few systems, less than one percent, changed primary source
water designation as reported to SDWIS/Fed between 2013 and 2020. However, the EPA believes that a
baseline rate of change of one percent is a reasonable predictor of future changes allowing for the
potential increases in source water type changes due to population movement, GW quality changes,
drought, and other climate-related factors.

3.3.9.3 Treatment Change

The EPA used historical data from SDWIS/Fed to estimate the percent of systems that would change
treatment in a given year. For this analysis, the EPA identified a treatment change as a system adding a
treatment that was not used in the previous year.

The analysis was limited to:

•	Treatment changes that were associated with non-emergency sources.

•	Treatment code entries with a reported treatment code as opposed to a blank field or with a
dash.

75 As previously discussed in Section 3.3.9.1, the EPA modified the analysis period used in the proposed LCRI EA
(USEPA, 2023b) from 2017 - 2020 to 2013 - 2020, but retained the assumption that 1 percent of systems would
have a change in their primacy source each year.

Final LCRI Economic Analysis

3-101

October 2024


-------
•	Systems that had at least one valid treatment code in all years of the analysis.

•	Systems that did not also have a source change in a given year, to avoid double counting.

•	Systems that were in SDWIS/Fed during the entire evaluation period of 2013 -2020.76 Changes in
treatment were identified for each 2-year sequence from 2013-2020. This yields a total of seven,
2-year sequences. These 2-year sequences include 2013-2014, 2014-2015, 2015-2016, 2016-
2017, 2017-2018, 2018-2019, and 2019-2020.

The estimated percent of systems that will change treatment each year is shown in Exhibit 3-55 and
Exhibit 3-56 for CWSs and NTNCWSs, respectively. The percentages are the average of the annual
percent of systems with a treatment change from 2013-2020. Similar to the approach taken for
estimating source water changes, the EPA estimated the weighted average for CWSs by CCT status,
source water type, and for all CWSs. The weighted average percentages are between three and seven
percent considering only CCT status and not source type, with an overall weighted average of 4.2
percent, which corresponds to the value of the SafeWater model LCR data variable, p_treat_change.

The estimated percent of NTNCWSs that will change treatment each year is low across all size
categories. The EPA estimated the weighted average for NTNCWSs by CCT status, source water type,
and for all NTNCWSs. The weighted average percentages were between two and six percent considering
only CCT status and not source type. The overall weighted average for NTNCWSs was 3.2 percent
(p_treat_change).

76 Similar to the analysis for the change in source water, the EPA also extended the analysis period from 2017 -
2020 in the proposed LCRI EA (USEPA, 2023b) to 2013 - 2020 to estimate the percentage of systems that would
have a long-term change in treatment each year. Based on the 2013 - 2020 expanded analysis period, the
percentage of systems with a treatment change decreased slightly from 4.6 to 4.2 percent for CWSs and from 3.3
to 3.2 percent for NTNCWSs.

Final LCRI Economic Analysis

3-102

October 2024


-------
Exhibit 3-55: Estimated Percent of CWSs that Will Change Treatment Each Year1

Size Category

Estimated Percent of CWSs that will Change Treatment
(Based on 2013 - 2020 SDWIS Data)

With CCT

Without CCT

Ground Water

Surface Water

Ground Water

Surface Water

<100

4.5%

7.7%

2.2%

5.6%

101-500

5.0%

5.3%

2.5%

3.9%

501-1,000

5.9%

4.6%

3.2%

6.2%

1,001-3,300

6.5%

5.5%

4.0%

5.3%

3,301-10,000

7.2%

7.6%

4.8%

6.6%

10,001-50,000

9.1%

7.8%

6.2%

7.1%

50,001-100,000

9.1%

9.6%

0.0%

0.0%

100,001-1,000,000

8.6%

11.8%

0.0%

14.3%

>1M

28.6%2

7.9%

0.0%

0.0%

WEIGHTED AVERAGE

6.1%

7.2%

2.9%

5.6%

6.5%

3.2%

4.2% (p_treat_change)

Note:

1	For additional information, see file "Likelihood_TreatmentChange_Final.xlsx," available in the docket at EPA-HQ-
OW-2022-0801 at www.regulations.gov.

2	This percentage is based on a single system that had a treatment change in some but not all years.

Exhibit 3-56: Estimated Percent of NTNCWSs that Will Change Treatment Each Year

Size Category

Estimated Percent of NTNCWSs that Will Change Treatment (Based
on 2013 - 2020 SDWIS Data)

With CCT

Without CCT

Ground Water

Surface Water

Ground Water

Surface Water

<100

5.1%

12.2%

2.6%

3.2%

101-500

5.1%

4.8%

2.8%

4.2%

501-1,000

5.3%

8.6%

2.9%

4.3%

1,001-3,300

5.8%

6.7%

3.1%

4.6%

3,301-10,000

5.2%

8.6%

1.3%

7.7%

10,001-50,000

0.0%



4.8%

14.3%

50,001-100,000



0.0%





100,001-1,000,000



0.0%





>1M









WEIGHTED AVERAGE

5.2%

7.4%

2.7%

4.3%

5.3%

2.8%

3.2% (p_treat_change)

Notes:

1. For additional information, see file "Likelihood_TreatmentChange_Final.xlsx," available in the docket at EPA-

HQ-OW-2022-0801 at www.regulations.gov.

Final LCRI Economic Analysis

3-103

October 2024


-------
2. The gray shaded cells denote that for NTNCWSs with CCT, no GW NTNCWSs serve more than 50,000 people
and no SW NTNCWSs serve more than > 1M people. For those without CCT, none serve more than 50,000
people, regardless of their water source.

3.3.9.3.1 Discussion of Data Limitations and Uncertainty

There is uncertainty in using counts of treatment codes from SDWIS/Fed to predict the future likelihood
of treatment changes. In addition, there is uncertainty in how consistently States reported this
information to SDWIS/Fed. Using an average rate from 2013 to 2020 may over or underrepresent costs
across the final LCRI implementation period depending on future drinking water regulations and trends
in source water quality. The EPA worked to minimize the impacts of these uncertainties by averaging
results over three two-year periods and combining size categories to minimize over-representation of
small numbers of large systems in a single size category in order to develop a more representative
prediction of changes throughout the rule analysis period for all water systems in the United States.

3.3.10 Schools, Child Care Facilities, Local Health Departments, and Targeted Medical Providers

The pre-2021 LCR and 2021 LCRR require CWSs that exceed the lead AL to provide lead PE materials to
facilities that include, but are not limited to schools, child care facilities, community-based
organizations, and medical providers that offer services to pregnant women, children, and infants to
better reach these at-risk populations and their caregivers. CWSs must also contact local health
departments by phone or in person to request the health agency's support in disseminating information
on lead in drinking water and the steps that vulnerable populations can take to reduce their exposure.
These requirements are maintained under the final LCRI. Section 3.3.10.1 explains how the EPA derived
the average number of each of these facility types per system.

The 2021 LCRR established requirements for CWSs to conduct PE and lead in drinking water testing in
schools and licensed child care facilities in their service area. Consistent with the 2021 LCRR, the final
LCRI requires CWSs to conduct drinking water monitoring in schools and child care facilities as follows:

•	Sample for lead in schools and licensed child care facilities served by the CWS unless they were
constructed or had full plumbing replacement on or after January 1, 2014 or the date the State
adopted standards that meet the definition of lead free in accordance with Section 1417 of the
SDWA, as amended by the Reduction of Lead in Drinking Water Act and are not served by a lead,
GRR, or unknown service line. This requirement does not apply to a school or child care facility
that is regulated as a public water system.

•	During the first five years after the final LCRI compliance date (first five-year cycle), conduct
monitoring at a minimum of 20 percent of the elementary schools and 20 percent of the
licensed child care facilities they serve per year. CWSs are required to schedule sampling with
elementary schools and licensed child care facilities and may count schools or licensed child care
facilities that decline sampling or are non-responsive towards the minimum 20 percent.
Secondary schools are sampled when requested. After the first five-year cycle, conduct
monitoring only at schools and licensed child care facilities that request testing.

•	Collect five samples for lead per school and two samples per child care facility. Samples must be
250-mL in volume with a stagnation time of 8 -18 hours.

Final LCRI Economic Analysis

3-104

October 2024


-------
•	Provide sampling results to tested facilities, States, and local and State health departments.

•	Provide a more in-depth annual report to the State.

The 2021 LCRR and final LCRI allows States to waive school and/or licensed child care facility sampling
requirements for individual CWSs under the following conditions:

•	The State or locality has an existing program or the facility or district has a policy that meets all
of the requirement in the LCRR/LCRI;

•	The State or locality has an existing program or the facility or district has a policy that meets all
of the requirement in the LCRR/LCRI except its program uses a different sample volume for
testing or stagnation time but requires remediation actions in response to a high lead level (e.g.,
disconnecting or replacing affected fixtures and installation of POU devices);

•	The State or locality has an existing program or the facility or district has a policy that meets all
the requirements in the LCRR/LCRI except its program samples less frequently than once every
five years but requires remediation actions in response to a high lead level; or

•	The sampling was conducted under the Water Infrastructure Improvements for the Nation Act
(WIIN Act) Grant Program for Lead Testing in School and Child Care Program Drinking Water and
therefore was consistent with the grant requirements.

New under the final LCRI, States can also waive CWSs from sampling a school or child care facility that
installs and maintains POU devices that are certified by an ANSI-accredited certifier to reduce lead on all
outlets used to provide water for human consumption.

The final LCRI expands the eligibility of waivers to allow States to waive requirements for the first five-
year sampling cycle after the final LCRI compliance date in schools and licensed child care facilities that
were sampled between January 1, 2021 and the final LCRI compliance date. The 2021 LCRR does not
allow States to waive requirements based on sampling conducted prior to the LCRR compliance date of
October 16, 2024.

3.3.10.1 Estimated Number of Facilities

This section is organized into four subsections as follows:

•	3.3.10.1.1: Schools

•	3.3.10.1.2: Child Care Facilities

•	3.3.10.1.3: Local Health Agencies and Targeted Medical Providers

•	3.3.10.1.4: Discussion of Data Limitations and Uncertainty

Schools and child care facilities that are NTNCWSs are not served by CWSs and have separate PE
requirements and, under the final LCRI, would not be included in the CWS's lead in drinking water
monitoring required at schools and licensed child care facilities. Thus, as shown below they are excluded
from the estimated number of schools and child care facilities provided in Sections 3.3.10.1.1 and
3.3.10.1.2, respectively. Also, for additional detail, see file, "School_Child Care lnputs_Final.xlsx,"
available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Final LCRI Economic Analysis

3-105

October 2024


-------
3.3.10.1.1	Schools

The EPA used the following approach to estimate the total number of public and private elementary and
secondary schools per State and United States territory and for the Navajo Nation:

1.	Obtained the most current estimate of public elementary and secondary schools per State and
United States territory from the United States Department of Education, NCES (NCES, 2020a).
Categorized combined elementary/secondary schools and schools that did not report a grade
span as elementary schools. Obtained the most current estimate of private schools per State
from the NCES Private School Universe Survey (NCES, 2020b) and used the ratio of the numbers
of elementary and secondary public schools to estimate the proportion of private schools that
are elementary vs. secondary. Supplemented NCES data with data from other sources to
estimate the number of public and private schools in the Navajo Nation and the number of
private schools in United States territories.

2.	Determined the number of NTNCWSs that are schools in each State and United States territory
based on the system's reported service area type code for schools of "SC" in SDWIS/Fed fourth
quarter 2020 frozen dataset. Used the owner type information to determine how many schools
were public vs. private. Used the ratio of elementary and secondary schools for all public schools
from NCES (NCES, 2020a) to estimate the proportion of NTNCWSs that are elementary vs.
secondary.

3.	Subtracted the number of public and private NTNCWSs schools per State and United States
territory calculated in Step 2 from the national number of public and private elementary and
secondary schools per State and United States territories estimated in Step 1 to produce the
adjusted number of schools served by CWSs.

Exhibit 3-57 (presented following the description of the estimated child care facilities) shows the

results of these steps in columns A through F.

3.3.10.1.2	Child Care Facilities

The EPA used a similar approach to the one used for schools to estimate the average number of child
care facilities per CWS:

1.	Obtained the national number of "organized" child care facilities77 per State from Figure 24, U.S.
Child Care Industry Statistics (CED, 2019). Note that the CED study utilized data collected in
2017; therefore, the estimated total number of child care facilities in 2017 was 674,332. The EPA
supplemented CED data with additional web-based information on the number of child care
facilities in the Navajo Nation and in United States territories. See the file "School_Child Care
lnputs_Final.xlsx" for details.

2.	Determined the number of NTNCWSs that are child care facilities in each State and United
States territories based on the system's reported service area type code for Daycare Center of
"DC" in SDWIS/Fed fourth quarter 2020 frozen dataset.

77 Organized child care providers are those who typically offer care on a paid basis.

Final LCRI Economic Analysis

3-106

October 2024


-------
3. Subtracted the number of NTNCWS child care facilities per State and United States territory
calculated in Step 2 from the national number of child care facilities per State and United States
territory, Step 1, to produce the adjusted number of child care facilities served by CWSs.

Exhibit 3-57 shows the results of these steps in Column G.

Final LCRI Economic Analysis

3-107

October 2024


-------
Exhibit 3-57: Number of Schools and Child Care Facilities by State and United States Territory, Adjusted to Remove NTNCWS Schools

and Child Care Facilities



Public Schools (adjusted to remove
NTNCWSs)

Private Schools (adjusted to remove
NTNCWSs)

















Number of Child

States/
Territories/Tribes

Total

Secondary

Elementary

Total

Secondary

Elementary

Care Facilities
(adjusted to remove
NTNCWSs)



A

B

C

D

E

F

G

United States (incl D.C.)

95,469

23,098

72,371

28,564

6,635

21,929

673,084

United States (incl
D.C.)/Territories/ Tribes

96,691

23,350

73,341

29,221

6,765

22,456

673,542



Alabama

1,528

399

1,129

403

105

298

7,163

Alaska

438

67

371

31

5

26

1,532

Arizona

2,286

748

1,538

394

129

265

11,432

Arkansas

1,078

370

708

170

59

111

5,186

California

10,095

2,409

7,686

3,153

752

2,401

95,126

Colorado

1,875

374

1,501

344

69

275

9,017

Connecticut

926

199

727

290

62

228

7,775

Delaware

222

38

184

127

22

105

1,384

Dist. of Columbia

228

38

190

72

12

60

1,299

Florida

4,111

640

3,471

2,451

382

2,069

34,510

Georgia

2,274

450

1,824

837

166

671

22,967

Hawaii

292

53

239

164

30

134

1,209

Idaho

708

183

525

131

34

97

2,769

Illinois

4,245

998

3,247

1,211

285

926

40,943

Indiana

1,772

419

1,353

828

196

632

12,514

Iowa

1,300

338

962

210

55

155

11,584

Final LCRI Economic Analysis

3-108

October 2024


-------
Id

ove

,751

,430

,855

,585

,189

,436

,121

,787

,654

,004

,084

,671

,513

,647

,322

,679

,992

,552

,890

,429

,019

,781

Public Schools (adjusted to remove
NTNCWSs)

Private Schools (adjusted to remove
NTNCWSs)

Secondary

338
482
282
109
217
360
946
825
334
632
281
326
130

	82_

497
230
1,080
530
180
951
558
247

Elementary

966

1,049
1,090
339
1,068
1,416
2,784
1,651
719
1,732
449
728
565
288
1,914
630
3,583
2,027
354
2,575
1,228
883

Total

210
408
403
79
631
633
348
503
181
621
98
189
128
165
977
158
1,542
723

	57_

1,130
173
345

Secondary

54
128
83
19
107
128
88
168

57
166

38

58
24
37

201
42
357
150
19
305
54
76

Elementary

156
280
320
60
524
505
260
335
124
455
60
131
104
128
776
116
1,185
573
38
825
119
269

3-109


-------


Public Schools (adjusted to remove
NTNCWSs)

Private Schools (adjusted to remove
NTNCWSs)

















Number of Child

States/
Territories/Tribes

Total

Secondary

Elementary

Total

Secondary

Elementary

Care Facilities
(adjusted to remove
NTNCWSs)



A

B

C

D

E

F

G

Pennsylvania

2,650

685

1,965

2,398

619

1,779

16,881

Rhode Island

298

66

232

105

23

82

1,672

South Carolina

1,253

283

970

409

92

317

8,018

South Dakota

712

231

481

78

25

53

2,825

Tennessee

1,856

368

1,488

564

112

452

13,185

Texas

8,899

2,017

6,882

1,701

385

1,316

56,358

Utah

1,058

281

777

167

44

123

4,970

Vermont

202

41

161

80

16

64

1,699

Virginia

1,948

401

1,547

958

197

761

15,847

Washington

2,408

625

1,783

660

171

489

9,763

West Virginia

701

135

566

124

24

100

2,307

Wisconsin

2,096

526

1,570

762

191

571

10,399

Wyoming

355

96

259

40

11

29

1,359



Puerto Rico

846

159

687

565

106

459

228

Guam

44

7

37

22

4

19

41

United States Virgin
Islands1

6

2

4

0

0

0

137

American Samoa

29

6

23

15

3

12

22

North Mariana Islands

20

11

9

15

8

7

13

Navajo Nation

277

67

210

40

10

30

17

Source: "School_Child Care lnputs_Final.xlsx," worksheet "Adjusted Sch & CC by State", Table 1.

Final LCRI Economic Analysis

3-110

October 2024


-------
Exhibit 3-58, is a continuation of Exhibit 3-57, and includes the number of schools and child care facilities
per person served by a CWS in each State. The SafeWater LCR model applies the number of schools per
person served by a CWS per State, to estimate the number of:

•	Public elementary schools per system that corresponds to SafeWater LCR model data variable,
numb_elem_schools pub (see Column K);

•	Private elementary schools per system that corresponds to SafeWater LCR model data variable,
numb_elem_schools priv (see Column N);

•	Public secondary schools per system that corresponds to SafeWater LCR model data variable,
numb_second_schools pub (see Column J);

•	Private secondary schools per system that corresponds to SafeWater LCR model data variable,
numb_second_schools priv (see Column M); and

•	Child care facilities per system that corresponds to SafeWater LCR model data variable,
numb_daycares (see Column O).

For example, assume a model CWS in Virginia serves 15,000 people. To determine the number of public
secondary schools for this water system, the CWS population of 15,000 is multiplied by the number of
public secondary schools per person served by a CWS in Virginia from Column J of Exhibit 3-58
(0.000056)78, which equals 0.84 public secondary schools.

78 The number of secondary schools per person in Virginia, 0.000056, is derived by dividing the number of public
secondary schools in Virginia from Column B of Exhibit 3-57 (401) by the total Virginia population served by all
CWSs in the State from Column H of Exhibit 3-58 (7,114,191).

Final LCR! Economic Analysis

October 2024

3-111


-------
Exhibit 3-58: Number of Schools per Person Served by a CWS per State, all Categories, Adjusted to Remove NTNCWS Schools and

Child Care Facilities

States/Territories/
Tribes

Total
Population
Served by
CWSs

Estimated Public Schools per
Person Served by a CWS

Estimated Private Schools per Person
Served by a CWS

Estimated Child
Care Facilities
per Person
Served by a
CWS

Public
Schools
TOTAL

Public Schools
SECONDARY

Public Schools
ELEMENTARY

Private
Schools
TOTAL

Private
Schools
SECONDARY

Private Schools
ELEMENTARY

H

1 = A / H

J = B/ H

K = C/H

L= D/ H

M = E/ H

N = F / H

O = G / H

United States (incl
D.C.)

309,061,248

0.000309

0.000075

0.000234

0.000092

0.000021

0.000071

0.002178

United States (incl

D.C.)/Territories/

Tribes

313,044,551

0.000309

0.000075

0.000234

0.000093

0.000022

0.000072

0.002152



Alabama

5,949,334

0.000257

0.000067

0.000190

0.000068

0.000018

0.000050

0.001204

Alaska

689,487

0.000635

0.000098

0.000538

0.000045

0.000007

0.000038

0.002222

Arizona

6,727,375

0.000340

0.000111

0.000229

0.000059

0.000019

0.000039

0.001699

Arkansas

2,938,783

0.000367

0.000126

0.000241

0.000058

0.000020

0.000038

0.001765

California

39,960,569

0.000253

0.000060

0.000192

0.000079

0.000019

0.000060

0.002380

Colorado

6,533,948

0.000287

0.000057

0.000230

0.000053

0.000011

0.000042

0.001380

Connecticut

2,776,268

0.000334

0.000072

0.000262

0.000104

0.000022

0.000082

0.002801

Delaware

937,477

0.000237

0.000041

0.000196

0.000135

0.000023

0.000112

0.001476

Dist. of Columbia

664,597

0.000343

0.000057

0.000286

0.000108

0.000018

0.000090

0.001955

Florida

20,533,551

0.000200

0.000031

0.000169

0.000119

0.000019

0.000101

0.001681

Georgia

9,549,632

0.000238

0.000047

0.000191

0.000088

0.000017

0.000070

0.002405

Final LCRI Economic Analysis

3-112

October 2024


-------
States/Territories/
Tribes

Total
Population
Served by
CWSs

Estimated Public Schools per
Person Served by a CWS

Estimated Private Schools per Person
Served by a CWS

Estimated Child
Care Facilities
per Person
Served by a
CWS

Public
Schools
TOTAL

Public Schools
SECONDARY

Public Schools
ELEMENTARY

Private
Schools
TOTAL

Private
Schools
SECONDARY

Private Schools
ELEMENTARY

H

1 = A / H

J = B/ H

K = C/H

L= D/ H

M = E/ H

N = F / H

O = G / H

Hawaii

1,507,465

0.000194

0.000035

0.000159

0.000109

0.000020

0.000089

0.000802

Idaho

1,365,170

0.000519

0.000134

0.000384

0.000096

0.000025

0.000071

0.002028

Illinois

12,028,786

0.000353

0.000083

0.000270

0.000101

0.000024

0.000077

0.003404

Indiana

4,980,984

0.000356

0.000084

0.000272

0.000166

0.000039

0.000127

0.002512

Iowa

2,849,783

0.000456

0.000119

0.000338

0.000074

0.000019

0.000055

0.004065

Kansas

2,821,989

0.000462

0.000120

0.000342

0.000074

0.000019

0.000055

0.002747

Kentucky

4,497,262

0.000340

0.000107

0.000233

0.000091

0.000029

0.000062

0.001430

Louisiana

5,004,321

0.000274

0.000056

0.000218

0.000081

0.000017

0.000064

0.001969

Maine

680,244

0.000659

0.000161

0.000498

0.000116

0.000028

0.000088

0.003800

Maryland

5,371,635

0.000239

0.000040

0.000199

0.000117

0.000020

0.000098

0.002641

Massachusetts

9,725,252

0.000183

0.000037

0.000146

0.000065

0.000013

0.000052

0.001073

Michigan

7,410,236

0.000503

0.000128

0.000376

0.000047

0.000012

0.000035

0.002580

Minnesota

4,524,951

0.000547

0.000182

0.000365

0.000111

0.000037

0.000074

0.003489

Mississippi

3,079,305

0.000342

0.000108

0.000233

0.000059

0.000019

0.000040

0.002810

Missouri

5,418,783

0.000436

0.000117

0.000320

0.000115

0.000031

0.000084

0.002400

Montana

770,369

0.000948

0.000365

0.000583

0.000127

0.000049

0.000078

0.002705

Nebraska

1,588,421

0.000664

0.000205

0.000459

0.000119

0.000037

0.000082

0.004200

Nevada

2,847,531

0.000244

0.000046

0.000198

0.000045

0.000008

0.000037

0.001936

New Hampshire

895,785

0.000413

0.000092

0.000321

0.000184

0.000041

0.000143

0.001839

Final LCRI Economic Analysis

3-113

October 2024


-------
States/Territories/
Tribes

Total
Population
Served by
CWSs

Estimated Public Schools per
Person Served by a CWS

Estimated Private Schools per Person
Served by a CWS

Estimated Child
Care Facilities
per Person
Served by a
CWS

Public
Schools
TOTAL

Public Schools
SECONDARY

Public Schools
ELEMENTARY

Private
Schools
TOTAL

Private
Schools
SECONDARY

Private Schools
ELEMENTARY

H

1 = A / H

J = B/ H

K = C/H

L= D/ H

M = E/ H

N = F / H

O = G / H

New Jersey

8,845,156

0.000273

0.000056

0.000216

0.000110

0.000023

0.000088

0.001845

New Mexico

2,077,412

0.000414

0.000111

0.000303

0.000076

0.000020

0.000056

0.001290

New York

18,251,232

0.000255

0.000059

0.000196

0.000084

0.000020

0.000065

0.003506

North Carolina

8,820,387

0.000290

0.000060

0.000230

0.000082

0.000017

0.000065

0.001763

North Dakota

738,289

0.000723

0.000243

0.000480

0.000077

0.000026

0.000051

0.003914

Ohio

10,486,511

0.000336

0.000091

0.000246

0.000108

0.000029

0.000079

0.002043

Oklahoma

3,703,121

0.000482

0.000151

0.000332

0.000047

0.000015

0.000032

0.001625

Oregon

3,517,136

0.000321

0.000070

0.000251

0.000098

0.000021

0.000077

0.002497

Pennsylvania

11,425,462

0.000232

0.000060

0.000172

0.000210

0.000054

0.000156

0.001477

Rhode Island

1,035,889

0.000288

0.000064

0.000224

0.000101

0.000022

0.000079

0.001614

South Carolina

4,080,458

0.000307

0.000069

0.000238

0.000100

0.000023

0.000078

0.001965

South Dakota

853,073

0.000835

0.000271

0.000564

0.000091

0.000030

0.000062

0.003312

Tennessee

7,193,174

0.000258

0.000051

0.000207

0.000078

0.000016

0.000063

0.001833

Texas

28,670,617

0.000310

0.000070

0.000240

0.000059

0.000013

0.000046

0.001966

Utah

3,280,153

0.000323

0.000086

0.000237

0.000051

0.000014

0.000037

0.001515

Vermont

449,956

0.000449

0.000092

0.000357

0.000178

0.000036

0.000141

0.003776

Virginia

7,114,191

0.000274

0.000056

0.000218

0.000135

0.000028

0.000107

0.002228

Washington

7,743,099

0.000311

0.000081

0.000230

0.000085

0.000022

0.000063

0.001261

West Virginia

1,527,381

0.000459

0.000088

0.000370

0.000081

0.000016

0.000066

0.001510

Final LCRI Economic Analysis

3-114

October 2024


-------
States/Territories/
Tribes

Total
Population
Served by
CWSs

Estimated Public Schools per
Person Served by a CWS

Estimated Private Schools per Person
Served by a CWS

Estimated Child
Care Facilities
per Person
Served by a
CWS

Public
Schools
TOTAL

Public Schools
SECONDARY

Public Schools
ELEMENTARY

Private
Schools
TOTAL

Private
Schools
SECONDARY

Private Schools
ELEMENTARY

H

1 = A / H

J = B/ H

K = C/H

L= D/ H

M = E/ H

N = F / H

O = G / H

Wisconsin

4,119,398

0.000509

0.000128

0.000381

0.000185

0.000046

0.000139

0.002524

Wyoming

499,860

0.000710

0.000192

0.000519

0.000080

0.000022

0.000058

0.002719



Puerto Rico

3,415,890

0.000248

0.000046

0.000201

0.000165

0.000031

0.000134

0.000067

Guam

191,786

0.000229

0.000036

0.000193

0.000115

0.000018

0.000096

0.000214

U.S. Virgin Islands

81,072

0.000074

0.000026

0.000048

0.000000

0.000000

0.000000

0.001690

American Samoa

56,728

0.000511

0.000106

0.000405

0.000264

0.000055

0.000210

0.000388

North Mariana
Islands

65,949

0.000303

0.000167

0.000136

0.000227

0.000125

0.000102

0.000197

Navajo Nation

171,878

0.001612

0.000390

0.001222

0.000233

0.000056

0.000176

0.000099

Source: "School_Child Care lnputs_Final.xlsx," worksheet "Adjusted Sch & CC by State", Table 2.

Final LCRI Economic Analysis

3-115

October 2024


-------
3.3.10.1.3 Local Health ARencies and TarReted Medical Providers

The EPA used the following approach to estimate the average number of local health agencies and
targeted medical providers per CWS:

1.	Determined the total number of local health agencies and targeted medical providers by
summing the numbers of local health agencies, obstetrician/gynecologists (ob/gyns),
pediatricians, and family medicine physicians, which were obtained from various data sources:

a.	The number of local health agencies was obtained from data collected by the NACCHO in
the 2019 National Profile of Local Health Departments (NACCHO, 2019). The estimated
number of local health agencies in 2019 was 2,459.79

b.	The number of ob-gyns (20,700), pediatricians (30,200), and family medicine physicians
(107,700) is from the U.S. Bureau of Labor Statistics' "Occupational Outlook Handbook"
(U.S. Bureau of Labor Statistics, 2021), as previously discussed in Section 3.2.8.

2.	Assumed the number of local health agencies and targeted medical providers were
proportionally distributed across the size categories. For example, as previously discussed, the
percentage of the people served by smallest size category of CWSs is approximately 0.23
percent of the total population served by CWSs (i.e., 708,236/313,044,551). The 0.23 percent
was multiplied by the number of health agencies and targeted medical providers to yield an
estimated number of health agencies and targeted medical providers served by all systems in
this size category (0.23%*161,059 = 364).

3.	Divided the number of health agencies and targeted medical providers in each of the nine
system size categories by the number of systems in the category.

4.	Rounded up values to the nearest whole number. The EPA assumed all CWSs would contact at
least one agency because the pre-2021 LCR requires CWSs to contact local public health
agencies even if they are outside their service area. Exhibit 3-59 provides the average number of
health agencies and targeted medical providers per CWSs.

79 A 2020 report was not available. For the 2019 profile study, NACCHO used a database of local health
departments (LHDs) based on previous profile studies and consulted with State health agencies and the State
Associations of Local Health Officials (SACCHOs) to identify additional LHDs for inclusion in the study population.
For the 2019 study, a total of 2,459 LHDs were included in the study population. Rhode Island was excluded from
the study because the State health agency operates on behalf of local public health and has no sub-state units. For
the first time, Hawaii was included.

Final LCR! Economic Analysis

3-116

October 2024


-------
Exhibit 3-59: Estimated Average Number of Local Health Agencies and Targeted Medical

Providers per CWS

System Size

# of Systems

Population
Served

Number of
Agencies
Proportionally
Distributed

Number of
Agencies per
System

Number of
Agencies per
System (Rounded

Up to Nearest
Whole Number)

A

B

C

<

u"

II

Q

E

< 100

11,732

708,236

364

0.0

1.0

101-500

15,084

3,830,126

1,971

0.1

1.0

501-1000

5,330

3,931,488

2,023

0.4

1.0

1001-3300

7,967

15,218,647

7,830

1.0

1.0

3301-10000

5,026

29,565,710

15,211

3.0

3.0

10001-50000

3,374

74,162,674

38,156

11.3

12.0

50001-100000

571

39,629,417

20,389

35.7

36.0

100001-1M

421

99,359,362

51,120

121.4

122.0

>1M

24

46,638,891

23,995

999.8

1,000.0

Total

49,529

313,044,551

161,059





Sources:

A: Exhibit 3-2.

B: Exhibit 3-4.

C: Calculated from SDWIS/Fed data 4th quarter 2020 frozen data set, 2010 NACCHO data (NACCHO, 2019) for the
number of local health department, and the U.S. Bureau of Labor Statistics "Occupational Outlook Handbook" (U.S.
Bureau of Labor Statistics, 2021) for the number of ob-gyns, pediatricians, and family medicine physicians (see
Steps 1 through 4 above).

3,3.10.1.4 Discussion of Data Limitations and Uncertainty

The number of entities that will receive PE under the pre-2021 LCR, 2021 LCRR, and final LCRI in
response to a lead ALE may be an underestimation because the number of entities may continue to
increase each year to meet the needs of growing populations. In addition, the estimated number of
facilities focused on schools, child care facilities, pediatricians, ob/gyns, and family medical providers
does not include other groups that are required to receive PE, i.e., Women, Infants, and Children (WIC);
Head Start; public and private hospitals and clinics; family planning centers; and local welfare agencies.
From a national perspective, the EPA does not anticipate these limitations to have an impact on the
incremental costs to deliver PE in response to a lead ALE under the final LCRI because the requirements
pertaining to delivery to these groups remain unchanged from the pre-2021 LCR and 2021 LCRR.

The uncertainty in the estimated number of schools that exist when the final LCRI goes into effect may
result in an underestimate of costs for CWSs to conduct lead sampling at these facilities. In addition, the
number of child care facilities may be overestimated because the source, CED (2019), may include non-
licensed facilities, which are not subject to LCRI requirements and would result in an overestimate of
costs. The resulting impact of all these factors may be an under or overestimate of national costs. The

Final LCRI Economic Analysis

3-117

October 2024


-------
EPA does not expect this uncertainty to have a significant impact on national cost and benefit estimates
in this EA.

3.3.10.2 Estimated Percentage of Schools and Child Care Facilities that Are Waived from
Monitoring Requirements

As noted previously, the 2021 LCRR and final LCRI allow States to waive school and/or licensed child care
facility sampling requirements for individual CWSs under certain circumstances when States have
existing programs or when sampling has been completed using WIIN grant funds. For detailed
information on eligibility under the 2021 LCRR and final LCRI, see Section 3.3.10. The EPA expanded the
time period for waivers under the final LCRI to allow the State to waive a CWS from testing a facility that
was sampled from January 1, 2021 through the compliance date of the LCRI (assumed to be October
2027). The 2021 LCRR does not allow States to waive requirements based on sampling conducted prior
to the LCRR compliance date of October 16, 2024.

The EPA used two methods to identify instances where schools and child care facilities served by CWSs
would be waived from monitoring requirements under the 2021 LCRR and final LCRI:

1.	Review State regulations for required programs, and

2.	Review funds allocated to States for lead testing through the Water Infrastructure Improvement
for the Nation (WIIN) grant.

Systems that would meet the minimum requirements of the 2021 LCRR and final LCRI would be
considered waived from the school sampling requirements and would not incur any burden or cost for
these activities. This section provides the evaluation of State regulations first, followed by the analysis of
school and child care facility sampling performed using WIIN grant funding.

3.3.10.2.1 Analysis of State Regulations

Some States have developed their own requirements for lead testing of drinking water at schools and
child care facilities. The purpose of this section is to describe the EPA's approach for identifying States
with programs that are at least as stringent as the 2021 LCRR and final LCRI for public and/or private
elementary, public and/or private secondary schools, and licensed child care facilities. The EPA assumed
CWSs in these States would not incur burden or costs to meet these requirements under the final LCRI
because States will elect to waive these requirements.

During 2022 and 2023, the EPA collected data and conducted internet searches to identify States with
regulations that are at least as stringent as those required under the 2021 LCRR and final LCRI for public
elementary schools, private elementary schools, public secondary schools, private secondary schools,
and/or child care facilities. Exhibit 3-60 and Exhibit 3-61 summarize the results of this review for the first
five-year cycle under 2021 the LCRR and final LCRI, respectively. Note that because of the expanded
waiver eligibility to include sampling conducted prior to the rule compliance date under the final LCRI
compared to the 2021 LCRR, additional State programs qualify for waivers for elementary schools,
secondary schools, and licensed child care facilities.

Final LCRI Economic Analysis

3-118

October 2024


-------
Exhibit 3-60: States with Existing Programs that Satisfy the Waiver Requirements under the

2021 LCRR for the First Five-Year Cycle1

States with Equivalent
Program

For Public
Elementary
Schools

For Private
Elementary
Schools

For Public
Secondary
Schools

For Private
Secondary
Schools

For Child Care
Facilities

District of Columbia

X



X



X

Indiana

X



X





Maryland

X

X

X

X



Minnesota

X



X





Missouri

X

X

X

X



Montana

X

X

X

X



New Hampshire









X

New Jersey

X



X



X

New York

X



X





North Carolina

X



X



X

Oregon

X



X



X

Pennsylvania

X



X





Utah

X

X

X

X



Vermont

X

X

X

X

X

Washington

X



X



X

Total Number of States

14

5

14

5

7

Note:

1 CWSs are assumed to incur no burden for the lead in drinking water testing at facilities in States marked with an
"X" for the first five-year cycle. The table only includes those States that include all of a particular subset (e.g., all
public elementary schools). States that have requirements for a smaller subset of schools, such as school
constructed before 1998, are not included in this table.

Exhibit 3-61: States with Existing Programs that Satisfy the Waiver Requirements under the

Final LCRI for the First Five-Year Cycle1

States with Equivalent
Program

For Public
Elementary
Schools

For Private
Elementary
Schools

For Public
Secondary
Schools

For Private
Secondary
Schools

For Child Care
Facilities

California









X2

Colorado

X







X

District of Columbia

X



X



X

Indiana

X



X





Maine

X

X

X

X



Maryland

X

X

X

X



Minnesota

X



X





Missouri

X

X

X

X



Montana

X

X

X

X



New Hampshire









X

New Jersey

X



X



X

New York

X



X





North Carolina

X



X



X

Final LCRI Economic Analysis

3-119

October 2024


-------
States with Equivalent
Program

For Public
Elementary
Schools

For Private
Elementary
Schools

For Public
Secondary
Schools

For Private
Secondary
Schools

For Child Care
Facilities

Oregon

X



X



X

Pennsylvania

X



X





Utah

X

X

X

X

X

Vermont

X

X

X

X

X

Washington

X



X



X

Total Number of States

16

6

15

6

10

Note:

1	CWSs are assumed to incur no burden for the lead in drinking water testing at facilities in States marked with an
"X" for the first five-year cycle. The table only includes those States that include all of a particular subset (e.g., all
public elementary schools). States that have requirements for a smaller subset of schools, such as school
constructed before 1998, are not included in this table.

2	In California's Assembly Bill 2370 required testing of all licensed child care facilities on private property
constructed before Jan 1, 2010 (which is the effective date of California lead ban), requires one-time testing
between January 1, 2020 and January 1, 2023. Only data from 2021 - 2023 can be used to satisfy the requirements
for the final LCRI. Thus, the EPA assumed that 75 percent of child care facilities would qualify for a waiver under
the final LCRI.

Exhibit 3-62 and Exhibit 3-63 summarizes the results for the subsequent five-year cycles under the 2021
LCRR and final LCRI, respectively. Note that one additional State program for licensed child care facility
testing (Utah) would be qualified to issue waivers under the final LCRI compared to the 2021 LCRR.

Exhibit 3-62: States with Existing Programs that Satisfy the Waiver Requirements Under the
2021 LCRR for the Second Five-Year Cycle and Subsequent Five-Year Cycles1

States with Equivalent
Program

For Public
Elementary
Schools

For Private
Elementary
Schools

For Public
Secondary
Schools

For Private
Secondary
Schools

For Child Care
Facilities

District of Columbia

X



X



X

Maryland

X

X

X

X



Minnesota

X



X





Missouri

X

X

X

X



Montana

X

X

X

X



New Hampshire









X

New Jersey

X



X



X

New York

X



X





North Carolina









X

Oregon

X



X



X

Pennsylvania

X



X





Vermont

X

X

X

X

X

Washington

X



X



X

Total Number of States

11

4

11

4

7

Note:

1 CWSs are assumed to incur no burden for the lead in drinking water testing at facilities in States marked with an
"X" for the second and subsequent five-year cycles.

Final LCRI Economic Analysis

3-120

October 2024


-------
Exhibit 3-63: States with Existing Programs that Satisfy the Waiver Requirements Under the
Final LCRI for the Second Five-Year Cycle and Subsequent Five-Year Cycles1

States with Equivalent
Program

For Public
Elementary
Schools

For Private
Elementary
Schools

For Public
Secondary
Schools

For Private
Secondary
Schools

For Child Care
Facilities

District of Columbia

X



X



X

Maryland

X

X

X

X



Minnesota

X



X





Missouri

X

X

X

X



Montana

X

X

X

X



New Hampshire









X

New Jersey

X



X



X

New York

X



X





North Carolina









X

Oregon

X



X



X

Pennsylvania

X



X





Utah









X

Vermont

X

X

X

X

X

Washington

X



X



X

Total Number of States

11

4

11

4

8

Note:

1 CWSs are assumed to incur no burden for the lead in drinking water testing at facilities in States marked with an
"X" for the second and subsequent five-year cycles.

3.3.10.2.2 Analysis of WIIN Rrand fundinR

Section 2107 of the WIIN Act of 2016 established the Lead Testing in School and Child Care Program
Drinking Water grant to award funding to States and Tribes to test for lead in public elementary schools
and child care facilities. Testing must be in accordance with the EPA's 3T guidance (USEPA, 2018) or
applicable State regulations, and results must be made publicly available. The 3T's recommends that
schools sample for lead at every tap, which would meet the minimum requirements of the 2021 LCRR
sampling that is maintained in the final LCRI (which is required at a minimum of five taps in schools).

To estimate the percent of public elementary schools80 and child care facilities that could be tested using
WIIN grant funds in each State, the EPA used the following three-step approach. Note that each step is
discussed in detail following the bullets.

•	Step 1: Estimate the average burden and costs for conducting lead tap sampling at a single
school or child care facility.

•	Step 2: Estimate WIIN grant funds allocated to each State from January 1, 2021 through federal
fiscal year (FY) 2026.

80 The WIIN grant funding cannot be used for private schools. WIIN grant funding can be used for public
elementary and secondary schools and public and private child care facilities. The EPA recommends States
prioritize elementary schools and child care facilities that serve children ages 6 and younger. For purposes of this
analysis, the EPA assumed the funding would be used to only test elementary schools and child care facilities.

Final LCRI Economic Analysis

3-121

October 2024


-------
• Step 3: Use the results from Steps 1 and 2 to estimate the proportion of public elementary
schools and child care facilities that can be tested between October 16, 2024 through FY 2026
under for 2021 LCRR and between January 1, 2021 and FY 2026 under the final LCRI using WIIN
grant funds. These schools and child care facilities would be waived from testing requirements
during the first five-year cycle under the 2021 LCRR and final LCRI, respectively.

The results from this analysis for elementary schools and child care facilities are combined with the
waivers based on State testing requirements to estimate the total percent of schools eligible for a
waiver under the 2021 LCRR and final LCRI for first five-year cycle.

Step 1: Estimate the average burden and costs for conducting lead tap sampling at a single school or
child care facility.

The EPA estimated non-labor and labor costs associated with a lead in drinking water sampling event for
an elementary school and a child care facility. Due to the wide range in school size and plumbing
configurations, the EPA prepared a low and high unit cost estimate. The major components of the cost
estimate are described below. For additional details and assumptions, see the derivation file
"School_Child Care lnputs_Final.xlsx", worksheet "Unit_Burden Costs."

As a starting point, the EPA estimated the number of taps that are sampled per elementary school and
child care facility. Because 3Ts recommends sampling at all taps used for cooking and drinking, the EPA
assumed that any elementary school or child care facility utilizing WIIN grant funds would sample at all
their taps. To represent the possible range of school configurations, the EPA used two data sources: the
minimum number of required plumbing fixtures from Table 403.1 of the 2021 International Plumbing
Code (IPC) as a low estimate, and sampling results from five States as a high estimate. To estimate the
number of taps sampled per child care facility, the EPA used the required number of taps from the 2021
LCRR (and maintained in the final LCRI) as a low estimate and data from two States (New York and
Nebraska) as the high estimate. The results are shown in Exhibit 3-64.

Exhibit 3-64: Low and High Estimate for the Number of Taps to be Sampled for Elementary

Schools and Child Care Facilities

Schools

Child Care Facilities

Average Number (Low)

Average Number (High)

Minimum Required (Low)

Average Number (High)

A

B

C

D

9

36

2

3

Source: Derivation file "School_Child Care lnputs_Final.xlsx", worksheet "Unit_Burden Costs."

Notes:

A: Based on weighted average number of taps for elementary schools based on minimum number of required
plumbing fixture from Table 403.1 of the 2021 IPC. See file "Analysis of School_Child Care Sample
Number_Final.xlsx", worksheet "School Sample # Based on IPC."

B: Based on the average of five States with sampling data in Table B-2 from the file, "Analysis of School_Child Care
Sample Number_Final.xlsx", worksheet "School Summary Statistics." Note that this average does not include
sample values that were less than the 5th percentile or greater than the 95th percentile.

C: Equals the minimum number of samples specified in the 2021 LCRR, which has not been revised under the final
LCRI).

Final LCRI Economic Analysis

3-122

October 2024


-------
D: Based on average number of taps sampled by child care facilities in New York and Nebraska. See file " Analysis
of School_Child Care Sample Number_Final.xlsx worksheet "Child Care Summary Statistics."

To calculate non-labor costs, the EPA used an estimated commercial laboratory cost per lead sample
analyzed of $23.50, in 2020 dollars. This value is based on quotes from seven laboratories (see the
Derivation file "Lead Analytical Burden and Cost_Final.xlsx", worksheet "Commercial Analytical_$" for
details). The EPA also added shipping costs based on sample weight estimates and United States Postal
Service (USPS) shipping rates.

To estimate labor costs associated with lead in drinking water sampling at elementary schools and child
care facilities, the EPA first estimated the burden associated with sampling activities. See Exhibit 3-65
and Exhibit 3-66 for estimates of burden per sampling event for elementary schools and child care
facilities, respectively.

Exhibit 3-65: Estimated Burden per Elementary School Sampling Event

Activity

Hours/Sampling Event

Low End

High End

Average

A: Develop a Sampling Plan

8.0

8.0

8.0

B: Collect Samples

1.5

6.0

3.8

C: Have Samples Analyzed

0.0

0.0

0.0

D: Provide Results to State/Parents

2.0

2.0

2.0

Total hours per sampling event

11.50

16.0

13.8

Source: Derivation file "School_Child Care lnputs_Final.xlsx," worksheet "Unit_Burden Costs."

Notes:

General: The school will follow the procedures outlined in the 3Ts guidance.

A: The EPA assumes schools will require 8 hours to read information provided by the State on the sampling
program and to develop a sampling plan.

B: The EPA assumed schools will require 10 minutes per sample because they will have prepared a sampling plan
and will be familiar with the location of taps to be samples. The per sample burden is multiplied by the number of
samples in Column A in Exhibit 3-64 for the low end and Column B in Exhibit 3-64 for the high end.

C: All samples are assumed to be analyzed by a commercial lab.

D: The EPA assumes schools will require 2 hours to prepare the notice, email, and post the results.

Exhibit 3-66: Estimated Burden per Child Care Facility Sampling Event

Activity

Hours/Sampling Event

Low End

High End

Average

A: Develop a Sampling Plan

4.0

4.0

4.00

B: Collect Samples

0.3

0.5

0.42

C: Have Samples Analyzed

0.0

0.0

0.00

D: Provide Results to State/Parents

2.0

2.0

2.00

Total hours per sampling event

6.3

6.5

6.42

Source: Derivation file "School_Child Care lnputs_Final.xlsx," worksheet "Unit_Burden Costs."
Notes:

Final LCRI Economic Analysis

3-123

October 2024


-------
General: The child care facility will follow the procedures outlined in the 3Ts guidance.

A: The EPA assumes child care facilities will require 4 hours to read information provided by the State on the
sampling program and to develop a sampling plan. The EPA assumed fewer hours are required for child care
facilities than schools to develop the sampling plan because the child care facility is more likely to be smaller than a
school and will have fewer taps to sample.

B: The EPA assumed child care facilities will require 10 minutes per sample because they will have prepared a
sampling plan and will be familiar with the location of taps to be samples. The per sample burden is multiplied by
the number of samples in Column A in Exhibit 3-64 for the low end and Column B in Exhibit 3-64 for the high end.
C: All samples are assumed to be analyzed by a commercial lab.

D: The EPA assumes child care facilities will require 2 hours to prepare the notice, email, and post the results.

To convert labor burden to cost, the EPA estimated labor rates for school maintenance workers and for
child care facility workers using national averages from the U. S. Bureau of Labor Statistics. The per hour
labor rates used for this analysis, in 2020 dollars, are $21.05 for a school maintenance worker and
$12.88 for a child care facility worker. For additional details and citations, see the derivation file
"School_Child Care lnputs_Final.xlsx," worksheet "Labor Rates."

The EPA combined non-labor costs for sample analysis and shipping with the labor costs associated with
the activities in Exhibit 3-65 and Exhibit 3-66 to develop a low, high, and average cost estimate per
sampling event per elementary school and child care facility. Results are shown in Exhibit 3-67 and
Exhibit 3-68, respectively. In summary, the EPA estimated that the total cost of a sample event for an
elementary school costs on average $840.41, and the total cost of a sample event for a child care facility
costs on average $149.27.

Exhibit 3-67: Estimated Total Cost per Elementary School Sample Event (2020$)

Activity

Cost/Sampling Event

Low End

High End

Average

A: Develop a Sampling Plan

$168.40

$168.40

$168.40

B: Collect Samples

$31.58

$126.30

$78.94

C: Have Samples Analyzed

$223.45

$878.50

$550.98

D: Provide Results to State/Parents

$42.10

$42.10

$42.10

Total non-labor & labor cost per
sampling event

$465.53

$1,215.30

$840.41

Source: Derivation file "School_Child Care lnputs_Final.xlsx," worksheet "Unit_Burden Costs."

Notes:

A, B, D: Labor costs estimated by multiplying the burden in Exhibit 3-66 by the hourly labor rate for school
maintenance worker of $21.05, in 2020$.

C: Non labor costs estimated by multiplying the number of samples in Exhibit 3-75 by the per sample commercial
laboratory cost of $23.50 plus shipping costs, in 2020$. For detailed assumptions for shipping costs, see the
derivation file "School_Child Care lnputs_Final.xlsx," worksheet "Unit_Burden Costs."

Final LCRI Economic Analysis

3-124

October 2024


-------
Exhibit 3-68: Estimated Total Cost per Child Care Facility Sample Event (2020$)

Activity

Cost/Sampling Event

Low End

High End

Average

A: Develop a Sampling Plan

$51.52

$51.52

$51.52

B: Collect Samples

$4.29

$6.44

$5.37

C: Have Samples Analyzed

$55.25

$78.00

$66.63

D: Provide Results to State/Parents

$25.76

$25.76

$25.76

Total non-labor & labor cost per
sampling event

$136.82

$161.72

$149.27

Source: Derivation file "School_Child Care lnputs_Final.xlsx", worksheet "Unit_Burden Costs."

Notes:

A, B, D: Labor costs estimated by multiplying the burden in Exhibit 3-66 by the hourly labor rate for child care
facility worker of $12.88, in 2020$.

C: Non labor costs estimated by multiplying the number of samples in Exhibit 3-75 by the per sample commercial
laboratory cost of $23.50 plus shipping costs, in 2020$. For detailed assumptions for shipping costs, see the
derivation file "School_Child Care lnputs_Final.xlsx", worksheet "Unit_Burden Costs."

Step 2: Estimate the WIIN grant funds per State from 2021 through 2026.

This analysis started with final allotments to each State for federal FY 2021 through 2023.81 Estimated
WIIN grant funding for FY 2024 is based on an authorized total appropriation of $40 million, with total
appropriations increasing by $5 million each year thereafter through FY 2026.82 The EPA reduced the
allotments by four percent to account for maximum allowable State costs to administer the grant
program (USEPA, 2020b). The EPA assumed that the percent of funds allotted to each State would be
the same in FY 2024 through 2026 as it was in FY 2023. Exhibit 3-69 provides the results of the analysis

81	Updated allotments for FY 2021 are available online here https://www.epa.gov/svstem/files/documents/2022-
02/fv2021 lead testing allotments wiin 2107.pdf, released June 2021. 2022 and 2023 allotments are available
online at https://www.epa.gov/svstem/files/documents/2023-

07/School%20Lead%20Testing%20and%20Reduction%20AllotmentMemo-Julv%202023 Final O.pdf, issued July
21, 2023. Note that the Federal Fiscal Year is from October 1 through September 30. For example, Fiscal Year 2021
(FY 2021) started on October 1, 2020, and ended on September 30, 2021.

82	Infrastructure Investment and Jobs Act, Section 50110, Lead Contamination in School Drinking water. Amendment
to Section 1464 of the SDWA (42 U.S.C. 300j-24). Authorization of appropriations. Available online at
https://www.conaress.aov/117/plaws/publ58/PLAW-117publ58.pdf.

Final LCRI Economic Analysis

3-125

October 2024


-------
Exhibit 3-69: 2021-2023 WIIN Grant Allotment and Projected WIIN Grant Funding for FY 2024 - FY 2026

State

FY 2021
Allotment

FY 2022
Allotment

FY 2023
Allotment

Percent
Allotment
per State for
FY 2023

Projected
FY 2024

Projected
FY 2025

Projected
FY 2026

A

B

C

D

E=38.4M*D

F=43.2M*D

G=48M*D

Totals

$23,534,400

$24,422,400

$27,086,400

100%

$38,400,000

$43,200,000

$48,000,000

Alabama

$312,960

$330,240

$366,720

1.4%

$519,894

$584,880

$649,867

Alaska

$65,280

$67,200

$74,880

0.3%

$106,156

$119,426

$132,695

Arizona

$380,160

$392,640

$434,880

1.6%

$616,523

$693,589

$770,654

Arkansas

$238,080

$245,760

$272,640

1.0%

$386,518

$434,833

$483,147

California

$2,210,880

$2,445,120

$2,712,000

10.0%

$3,844,763

$4,325,359

$4,805,954

Colorado

$366,720

$380,160

$421,440

1.6%

$597,469

$672,153

$746,837

Connecticut

$345,600

$386,880

$429,120

1.6%

$608,357

$684,402

$760,447

Delaware

$86,400

$89,280

$98,880

0.4%

$140,181

$157,703

$175,226

District of Columbia

$82,560

$84,480

$93,120

0.3%

$132,015

$148,517

$165,019

Florida

$1,005,120

$1,033,920

$1,146,240

4.2%

$1,625,008

$1,828,134

$2,031,260

Georgia

$823,680

$1,087,680

$1,206,720

4.5%

$1,710,750

$1,924,593

$2,138,437

Hawaii

$85,440

$83,520

$93,120

0.3%

$132,015

$148,517

$165,019

Idaho

$221,760

$151,680

$168,000

0.6%

$238,171

$267,943

$297,714

Illinois

$989,760

$1,032,000

$1,144,320

4.2%

$1,622,286

$1,825,072

$2,027,858

Indiana

$406,080

$557,760

$618,240

2.3%

$876,470

$986,029

$1,095,587

Iowa

$283,200

$289,920

$321,600

1.2%

$455,928

$512,919

$569,910

Kansas

$278,400

$290,880

$322,560

1.2%

$457,289

$514,450

$571,611

Kentucky

$334,080

$318,720

$353,280

1.3%

$500,840

$563,445

$626,050

Louisiana

$559,680

$368,640

$408,960

1.5%

$579,777

$652,249

$724,721

Maine

$323,520

$169,920

$189,120

0.7%

$268,113

$301,627

$335,141

Maryland

$318,720

$324,480

$360,000

1.3%

$510,367

$574,163

$637,959

Massachusetts

$313,920

$841,920

$934,080

3.4%

$1,324,232

$1,489,761

$1,655,290

Michigan

$666,240

$878,400

$974,400

3.6%

$1,381,393

$1,554,067

$1,726,741

Final LCRI Economic Analysis

3-126

October 2024


-------
State

FY 2021
Allotment

FY 2022
Allotment

FY 2023
Allotment

Percent
Allotment
per State for
FY 2023

Projected
FY 2024

Projected
FY 2025

Projected
FY 2026

A

B

C

D

E=38.4M*D

F=43.2M*D

G=48M*D

Minnesota

$412,800

$420,480

$466,560

1.7%

$661,435

$744,115

$826,794

Mississippi

$288,960

$302,400

$335,040

1.2%

$474,981

$534,354

$593,727

Missouri

$493,440

$446,400

$495,360

1.8%

$702,265

$790,048

$877,831

Montana

$259,200

$257,280

$285,120

1.1%

$404,211

$454,737

$505,263

Nebraska

$342,720

$351,360

$389,760

1.4%

$552,557

$621,627

$690,696

Nevada

$154,560

$251,520

$278,400

1.0%

$394,684

$444,019

$493,355

New Hampshire

$851,520

$220,800

$245,760

0.9%

$348,410

$391,962

$435,513

New Jersey

$465,600

$480,960

$533,760

2.0%

$756,704

$851,292

$945,880

New Mexico

$167,040

$174,720

$192,960

0.7%

$273,557

$307,751

$341,946

New York

$1,131,840

$1,128,000

$1,250,880

4.6%

$1,773,355

$1,995,024

$2,216,693

North Carolina

$544,320

$584,640

$648,000

2.4%

$918,660

$1,033,493

$1,148,325

North Dakota

$82,560

$73,920

$81,600

0.3%

$115,683

$130,144

$144,604

Ohio

$720,960

$669,120

$742,080

2.7%

$1,052,036

$1,183,541

$1,315,045

Oklahoma

$352,320

$674,880

$747,840

2.8%

$1,060,202

$1,192,727

$1,325,253

Oregon

$624,960

$524,160

$580,800

2.1%

$823,392

$926,316

$1,029,240

Pennsylvania

$943,680

$904,320

$1,003,200

3.7%

$1,422,222

$1,600,000

$1,777,778

Rhode Island

$382,080

$280,320

$310,080

1.1%

$439,596

$494,545

$549,495

South Carolina

$296,640

$299,520

$333,120

1.2%

$472,259

$531,292

$590,324

South Dakota

$157,440

$193,920

$215,040

0.8%

$304,859

$342,967

$381,074

Tennessee

$325,440

$370,560

$411,840

1.5%

$583,860

$656,842

$729,825

Texas

$1,924,800

$1,960,320

$2,174,400

8.0%

$3,082,616

$3,467,943

$3,853,270

Utah

$267,840

$272,640

$302,400

1.1%

$428,708

$482,297

$535,885

Vermont

$112,320

$116,160

$129,600

0.5%

$183,732

$206,699

$229,665

Virginia

$412,800

$423,360

$469,440

1.7%

$665,518

$748,708

$831,898

Washington

$459,840

$589,440

$653,760

2.4%

$926,826

$1,042,679

$1,158,533

West Virginia

$147,840

$148,800

$165,120

0.6%

$234,088

$263,349

$292,610

Final LCRI Economic Analysis

3-127

October 2024


-------
State

FY 2021
Allotment

FY 2022
Allotment

FY 2023
Allotment

Percent
Allotment
per State for
FY 2023

Projected
FY 2024

Projected
FY 2025

Projected
FY 2026

A

B

C

D

E=38.4M*D

F=43.2M*D

G=48M*D

Wisconsin

$449,280

$385,920

$428,160

1.6%

$606,996

$682,871

$758,745

Wyoming

$63,360

$65,280

$72,000

0.3%

$102,073

$114,833

$127,592

Source: "School_Child Care lnputs_Final.xlsx," worksheet, "Project WIIN Funding Available."

Notes:

A: The EPA allotment memo for FY 2021 (USEPA, 2020b), issued June 2021. Does not include American Samoa, Puerto Rico, and the Virgin Islands. Reduced by
4% to account for maximum allowable percent of grant funds that can be used by States to pay for the administrative costs of carrying out the program, per
USEPA (2020c).

B and C: The EPA allotment memo for FY 2022 and 2023 (USEPA, 2023c), issued July 21, 2023. Does not include American Samoa, Puerto Rico, and the Virgin
Islands. Reduced by 4% to account for maximum allowable percent of grant funds that can be used by States to pay for the administrative costs of carrying out
the program, per USEPA (2020c).

D: Percent of WIIN grant funds allotted to each State in FY 2023.

F-J: Total allotment from the Infrastructure Investment and Jobs Act, Section 50110, Lead Contamination in School Drinking water. Amendment to Section 1464
of the Safe Drinking Water Act (42 U.S.C. 300j-24). Authorization of appropriations. Available online at https://www.congress.gov/117/plaws/publ58/PLAW-
117publ58.pdf. Assume that the proportion of total funds allocated to each State is consistent with FY 23 allocations. Reduced by 4 percent to account for the
maximum allowable percent of grant funds that can be used by States to pay for the administrative costs of carrying out the program, per USEPA (2020c). Thus
for each State, their funding is equal to the total funding for that fiscal year multiplied by the percent in Column D.

Final LCRI Economic Analysis

3-128

October 2024


-------
Step 3: Use the results from Steps 1 and 2 to estimate the proportion of elementary schools and child
care facilities that can be tested using WIIN grant funds under the 2021 LCRR and final LCRI. Under the
2021 LCRR, the EPA considered funds available beginning October 2024 through FY 2026. For the final
LCRI, the EPA considered the funds available starting January 2021 through FY 2026 to reflect the
change in waiver eligibility. These calculations are in the form of large tables that can be found in the
Derivation file, "School_Child Care lnputs_Final.xlsx", worksheets "Proj_Tested_WIIN Grant_LCRR" and
"Proj_Tested_WIIN Grant_LCRI." This step involves two main sub-steps:

•	The average of the high and low unit costs per sampling event from Step 1 was used to estimate
what it would cost to sample all public elementary schools and child care facilities in each State.
See Sections 3.3.10.1.1 and 3.3.10.1.2 for the estimated number of elementary schools and child
care facilities, respectively, in each State. The States with existing programs that satisfy the
waiver requirements for the first five-year cycle for public elementary schools and child care
facilities, as shown in Exhibit 3-70 for the 2021 LCRR and in Exhibit 3-71 for the final LCRI, were
also excluded from the analysis.

•	The total costs for sampling public elementary schools and child care facilities was compared to
the WIIN grant allotments starting on October 16, 2024 for the 2021 LCRR and January 1, 2021
for the LCRI. The EPA assumed that States would take full advantage of the available funding to
sample elementary schools and child care facilities. Therefore, the EPA subtracted available
WIIN grant funds from the total cost for sampling all public elementary schools and child care
facilities to estimate the remaining funds needed to sample all public elementary schools and
child care facilities. The EPA did this for each year starting in FY 2025 and FY 2026 under the
2021 LCRR, and January 2021 through FY 2026 under the LCRI. Based on these data, nearly all
States should be able to use the WIIN grant funds to sample 100 percent of their public
elementary schools and child care facilities during the first five-year cycle under the final LCRI.
See Exhibit 3-70 and Exhibit 3-71 for results per State for the 2021 LCRR and final LCRI,
respectively.

Note that the EPA assumed CWSs in States in which the rows have gray shading would be
waived from conducting lead in school and/or child care facilities testing, due to State
regulations. Therefore, the EPA did not consider the schools and child care facilities in those
States as part of this step.

Exhibit 3-70: Estimated Percent of Public Elementary Schools and Child Care Facilities that
Could be Sampled to Comply with the 2021 LCRR Using WIIN Grant Funds from October 16,

2024 - FY 2026

State

Percent of Child Care Facilities Sampled
Using WIIN Grant Funds from October 16,
2024-FY 2026

Percent of Public Elementary Schools
Sampled Using WIIN Grant Funds from
October 16, 2024 - FY 2026

Alabama

61.2%

61.2%

Alaska

46.7%

46.7%

Arizona

48.8%

48.8%

Arkansas

67.0%

67.0%

California

44.2%

44.2%

Final LCRI Economic Analysis

3-129

October 2024


-------
State

Percent of Child Care Facilities Sampled
Using WIIN Grant Funds from October 16,
2024-FY 2026

Percent of Public Elementary Schools
Sampled Using WIIN Grant Funds from
October 16, 2024 - FY 2026

Colorado

54.4%

54.4%

Connecticut

81.6%

81.6%

Delaware

92.3%

92.3%

District of Columbia





Florida

47.8%

47.8%

Georgia

81.9%

81.9%

Hawaii

82.2%

82.2%

Idaho

66.2%

66.2%

Illinois

43.6%

43.6%

Indiana

100.0%



Iowa

42.7%

42.7%

Kansas

55.2%

55.2%

Kentucky

64.6%

64.6%

Louisiana

57.7%

57.7%

Maine

95.0%

95.0%

Maryland

57.2%



Massachusetts

100.0%

100.0%

Michigan

63.2%

63.2%

Minnesota

66.7%



Mississippi

59.5%

59.5%

Missouri

85.9%



Montana

100.0%



Nebraska

81.6%

81.6%

Nevada

72.2%

72.2%

New Hampshire



100.0%

New Jersey





New Mexico

69.9%

69.9%

New York

44.1%



North Carolina





North Dakota

37.7%

37.7%

Ohio

46.6%

46.6%

Oklahoma

100.0%

100.0%

Oregon





Pennsylvania

100.0%



Rhode Island

100.0%

100.0%

South Carolina

55.8%

55.8%

South Dakota

87.7%

87.7%

Tennessee

43.1%

43.1%

Texas

51.6%

51.6%

Utah

100.0%



Final LCRI Economic Analysis

3-130

October 2024


-------


Percent of Child Care Facilities Sampled

Percent of Public Elementary Schools

State

Using WIIN Grant Funds from October 16,

Sampled Using WIIN Grant Funds from



2024-FY 2026

October 16, 2024 - FY 2026

Vermont





Virginia

43.1%

43.1%

Washington





West Virginia

67.8%

67.8%

Wisconsin

50.2%

50.2%

Wyoming

57.6%

57.6%

Source: "School_Child Care lnputs_Final.xlsx," worksheet, "Proj_Tested_WIIN Grant_LCRR."

Note: The EPA assumed CWSs in States in which the rows have gray shading would be waived from conducting

lead in school and/or child care facility testing due to State regulations.

Exhibit 3-71: Estimated Percent of Public Elementary Schools and Child Care Facilities that
Could be Sampled to Comply with the Final LCRI Using WIIN Grant Funds from January 1, 2021

- FY 2026

State

Percent of Child Care Facilities Sampled
Using WIIN Grant Funds from January 1,
2021 - FY 2026

Percent of Public Elementary Schools
Sampled Using WIIN Grant Funds from
January 1, 2021 - FY 2026

Alabama

100.0%

100.0%

Alaska

100.0%

100.0%

Arizona

100.0%

100.0%

Arkansas

100.0%

100.0%

California



95.8%

Colorado





Connecticut

100.0%

100.0%

Delaware

100.0%

100.0%

District of Columbia





Florida

100.0%

100.0%

Georgia

100.0%

100.0%

Hawaii

100.0%

100.0%

Idaho

100.0%

100.0%

Illinois

94.9%

94.9%

Indiana

100.0%



Iowa

93.1%

93.1%

Kansas

100.0%

100.0%

Kentucky

100.0%

100.0%

Louisiana

100.0%

100.0%

Maine

100.0%



Maryland

100.0%



Massachusetts

100.0%

100.0%

Michigan

100.0%

100.0%

Minnesota

100.0%



Mississippi

100.0%

98.3%

Final LCRI Economic Analysis

3-131

October 2024


-------
State

Percent of Child Care Facilities Sampled
Using WIIN Grant Funds from January 1,
2021 - FY 2026

Percent of Public Elementary Schools
Sampled Using WIIN Grant Funds from
January 1, 2021 - FY 2026

Missouri

100.0%



Montana

100.0%



Nebraska

100.0%

100.0%

Nevada

100.0%

100.0%

New Hampshire



100.0%

New Jersey





New Mexico

100.0%

100.0%

New York

96.4%



North Carolina





North Dakota

83.4%

83.4%

Ohio

100.0%

100.0%

Oklahoma

100.0%

100.0%

Oregon





Pennsylvania

100.0%



Rhode Island

100.0%

100.0%

South Carolina

100.0%

100.0%

South Dakota

100.0%

100.0%

Tennessee

93.1%

93.1%

Texas

100.0%

100.0%

Utah





Vermont





Virginia

94.1%

94.1%

Washington





West Virginia

100.0%

100.0%

Wisconsin

100.0%

100.0%

Wyoming

100.0%

100.0%

Source: "School_Child Care lnputs_Final.xlsx," worksheet, "Proj_Tested_WIIN Grant_LCRI."

Note: The EPA assumed CWSs in States in which the rows have gray shading would be waived from conducting

lead in school and/or child care facility testing due to State regulations.

The EPA combined the results of the State regulatory analysis and the WIIN grant fund analysis to
predict the percent of each child care facility and type of school (public and private, elementary and
secondary) that would meet the criteria for a waiver under the 2021 LCRR and the final LCRI for the first
five-year and second five-year cycles. The final results of the analysis are provided in Exhibit 3-72 for the
2021 LCRR and Exhibit 3-73 for the final LCRI. Note that results for child care facilities and public
elementary schools for the first five-year cycle reflect the analysis of available WIIN grant funds in
addition to State regulations. The results for child care facilities and public elementary schools for the
second five-year cycle, and all results for private elementary, public secondary, and public secondary
schools reflect the analysis of State regulations only. The text in red font are variable names of the
costing inputs for the SafeWater LCR model.

Final LCRI Economic Analysis

3-132

October 2024


-------
Exhibit 3-72: Percent of Schools and Child Care Facilities Eligible for Waivers under the 2021 LCRR based on State Regulations and

WIIN Grant Funding

State

Percent of Child Care
Facilities Eligible for
Waiver for:

Percent of Public
Elementary Schools
Eligible for Waiver for:

Percent of Private
Elementary Schools
Eligible for Waiver
for:

Percent of Public

Secondary
Schools Eligible
for Waiver for:

Percent of

Private
Secondary
Schools Eligible
for Waiver for:

1st 5-yr
cycle

2nd 5-yr cycles
on

1st 5-yr cycle

2nd 5-yr cycles
on

1st 5-yr cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

pp_childcar
e_mand_w
aiver

pp_childcare_o
nreq_waiver

pp_pub_elem_
mand_waiver

pp_pub_elem_
onreq_waiver

pp_priv_ele
m_mand_w
aiver

pp_priv_ele
m_onreq_w
aiver

pp_pub_se
cond_onre
ql_waiver

pp_pub_se
cond_onre
q2on_waiv
er

pp_priv_se
cond_onre
ql_waiver

pp_priv_s
econd_on
req2on_w
aiver

Alabama

61.2%

0%

61.2%

0%

0%

0%

0%

0%

0%

0%

Alaska

46.7%

0%

46.7%

0%

0%

0%

0%

0%

0%

0%

Arizona

48.8%

0%

48.8%

0%

0%

0%

0%

0%

0%

0%

Arkansas

67.0%

0%

67.0%

0%

0%

0%

0%

0%

0%

0%

California

44.2%

0%

44.2%

0%

0%

0%

0%

0%

0%

0%

Colorado

54.4%

0%

54.4%

0%

0%

0%

0%

0%

0%

0%

Connecticut

81.6%

0%

81.6%

0%

0%

0%

0%

0%

0%

0%

Delaware

92.3%

0%

92.3%

0%

0%

0%

0%

0%

0%

0%

District of
Columbia

100.0%

100%

100.0%

100%

0%

0%

100%

100%

0%

0%

Florida

47.8%

0%

47.8%

0%

0%

0%

0%

0%

0%

0%

Georgia

81.9%

0%

81.9%

0%

0%

0%

0%

0%

0%

0%

Hawaii

82.2%

0%

82.2%

0%

0%

0%

0%

0%

0%

0%

Idaho

66.2%

0%

66.2%

0%

0%

0%

0%

0%

0%

0%

Illinois

43.6%

0%

43.6%

0%

0%

0%

0%

0%

0%

0%

Indiana

100.0%

0%

100.0%

0%

0%

0%

100%

0%

0%

0%

Iowa

42.7%

0%

42.7%

0%

0%

0%

0%

0%

0%

0%

Kansas

55.2%

0%

55.2%

0%

0%

0%

0%

0%

0%

0%

Kentucky

64.6%

0%

64.6%

0%

0%

0%

0%

0%

0%

0%

Final LCRI Economic Analysis

3-133

October 2024


-------
State

Percent of Child Care
Facilities Eligible for
Waiver for:

Percent of Public
Elementary Schools
Eligible for Waiver for:

Percent of Private
Elementary Schools
Eligible for Waiver
for:

Percent of Public

Secondary
Schools Eligible
for Waiver for:

Percent of

Private
Secondary
Schools Eligible
for Waiver for:

1st 5-yr
cycle

2nd 5-yr cycles
on

1st 5-yr cycle

2nd 5-yr cycles
on

1st 5-yr cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

pp_childcar
e_mand_w
aiver

pp_childcare_o
nreq_waiver

pp_pub_elem_
mand_waiver

pp_pub_elem_
onreq_waiver

pp_priv_ele
m_mand_w
aiver

pp_priv_ele
m_onreq_w
aiver

pp_pub_se
cond_onre
ql_waiver

pp_pub_se
cond_onre
q2on_waiv
er

pp_priv_se
cond_onre
ql_waiver

pp_priv_s
econd_on
req2on_w
aiver

Louisiana

57.7%

0%

57.7%

0%

0%

0%

0%

0%

0%

0%

Maine

95.0%

0%

95.0%

0%

0%

0%

0%

0%

0%

0%

Maryland

57.2%

0%

100.0%

100%

100%

100%

100%

100%

100%

100%

Massachuse
tts

100.0%

0%

100.0%

0%

0%

0%

0%

0%

0%

0%

Michigan

63.2%

0%

63.2%

0%

0%

0%

0%

0%

0%

0%

Minnesota

66.7%

0%

100.0%

100%

0%

0%

100%

100%

0%

0%

Mississippi

59.5%

0%

59.5%

0%

0%

0%

0%

0%

0%

0%

Missouri

85.9%

0%

100.0%

100%

100%

100%

100%

100%

100%

100%

Montana

100.0%

0%

100.0%

100%

100%

100%

100%

100%

100%

100%

Nebraska

81.6%

0%

81.6%

0%

0%

0%

0%

0%

0%

0%

Nevada

72.2%

0%

72.2%

0%

0%

0%

0%

0%

0%

0%

New

Hampshire

100.0%

100%

100.0%

0%

0%

0%

0%

0%

0%

0%

New Jersey

100.0%

100%

100.0%

100%

0%

0%

100%

100%

0%

0%

New Mexico

69.9%

0%

69.9%

0%

0%

0%

0%

0%

0%

0%

New York

44.1%

0%

100.0%

100%

0%

0%

100%

100%

0%

0%

North
Carolina

100.0%

100%

100.0%

0%

0%

0%

100%

0%

0%

0%

North
Dakota

37.7%

0%

37.7%

0%

0%

0%

0%

0%

0%

0%

Ohio

46.6%

0%

46.6%

0%

0%

0%

0%

0%

0%

0%

Oklahoma

100.0%

0%

100.0%

0%

0%

0%

0%

0%

0%

0%

Final LCRI Economic Analysis

3-134

October 2024


-------
State

Percent of Child Care
Facilities Eligible for
Waiver for:

Percent of Public
Elementary Schools
Eligible for Waiver for:

Percent of Private
Elementary Schools
Eligible for Waiver
for:

Percent of Public

Secondary
Schools Eligible
for Waiver for:

Percent of

Private
Secondary
Schools Eligible
for Waiver for:

1st 5-yr
cycle

2nd 5-yr cycles
on

1st 5-yr cycle

2nd 5-yr cycles
on

1st 5-yr cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

pp_childcar
e_mand_w
aiver

pp_childcare_o
nreq_waiver

pp_pub_elem_
mand_waiver

pp_pub_elem_
onreq_waiver

pp_priv_ele
m_mand_w
aiver

pp_priv_ele
m_onreq_w
aiver

pp_pub_se
cond_onre
ql_waiver

pp_pub_se
cond_onre
q2on_waiv
er

pp_priv_se
cond_onre
ql_waiver

pp_priv_s
econd_on
req2on_w
aiver

Oregon

100.0%

100%

100.0%

100%

0%

0%

100%

100%

0%

0%

Pennsylvani
a

100.0%

0%

100.0%

100%

0%

0%

100%

100%

0%

0%

Rhode
Island

100.0%

0%

100.0%

0%

0%

0%

0%

0%

0%

0%

South
Carolina

55.8%

0%

55.8%

0%

0%

0%

0%

0%

0%

0%

South
Dakota

87.7%

0%

87.7%

0%

0%

0%

0%

0%

0%

0%

Tennessee

43.1%

0%

43.1%

0%

0%

0%

0%

0%

0%

0%

Texas

51.6%

0%

51.6%

0%

0%

0%

0%

0%

0%

0%

Utah

100.0%

0%

100.0%

0%

100%

0%

100%

0%

100%

0%

Vermont

100.0%

100%

100.0%

100%

100%

100%

100%

100%

100%

100%

Virginia

43.1%

0%

43.1%

0%

0%

0%

0%

0%

0%

0%

Washington

100.0%

100%

100.0%

100%

0%

0%

100%

100%

0%

0%

West
Virginia

67.8%

0%

67.8%

0%

0%

0%

0%

0%

0%

0%

Wisconsin

50.2%

0%

50.2%

0%

0%

0%

0%

0%

0%

0%

Wyoming

57.6%

0%

57.6%

0%

0%

0%

0%

0%

0%

0%

Source: "School_Child Care lnputs_Final.xlsx," worksheet, "Waiver EIigibiIity_LCRR."

Final LCRI Economic Analysis

3-135

October 2024


-------
Exhibit 3-73: Percent of Schools and Child Care Facilities Eligible for Waivers under the Final LCRI based on State Regulations and

WIIN Grant Funding

State

Percent of Child
Care Facilities
Eligible for
Waiver for:

Percent of Public

Elementary
Schools Eligible
for Waiver for:

Percent of Private
Elementary Schools
Eligible for Waiver
for:

Percent of Public

Secondary
Schools Eligible
for Waiver for:

Percent of

Private
Secondary
Schools Eligible
for Waiver for:

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

pp_childc
are_mand
_waiver

pp_childcar
e_onreq_w
aiver

pp_pub_el
em_mand
_waiver

pp_pub_ele
m_onreq_
waiver

pp_priv_ele
m_mand_
waiver

pp_priv_elem
_onreq_waiv
er

pp_childc
are_mand
_w aiver

pp_childcare
_onreq_wai
ver

pp_pub_el
em_mand
_w aiver

pp_pub_ele
m_onreq_
waiver

Alabama

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Alaska

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Arizona

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Arkansas

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

California

100%

0%

96%

0%

0%

0%

0%

0%

0%

0%

Colorado

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Connecticut

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Delaware

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

District of
Columbia

100%

100%

100%

100%

0%

0%

100%

100%

0%

0%

Florida

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Georgia

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Hawaii

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Idaho

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Illinois

95%

0%

95%

0%

0%

0%

0%

0%

0%

0%

Indiana

100%

0%

100%

0%

0%

0%

100%

0%

0%

0%

Iowa

93%

0%

93%

0%

0%

0%

0%

0%

0%

0%

Kansas

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Kentucky

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Final LCRI Economic Analysis

3-136

October 2024


-------
State

Percent of Child
Care Facilities
Eligible for
Waiver for:

Percent of Public

Elementary
Schools Eligible
for Waiver for:

Percent of Private
Elementary Schools
Eligible for Waiver
for:

Percent of Public

Secondary
Schools Eligible
for Waiver for:

Percent of

Private
Secondary
Schools Eligible
for Waiver for:

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

pp_childc
are_mand
_waiver

pp_childcar
e_onreq_w
aiver

pp_pub_el
em_mand
_waiver

pp_pub_ele
m_onreq_
waiver

pp_priv_ele
m_mand_
waiver

pp_priv_elem
_onreq_waiv
er

pp_childc
are_mand
_w aiver

pp_childcare
_onreq_wai
ver

pp_pub_el
em_mand
_w aiver

pp_pub_ele
m_onreq_
waiver

Louisiana

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Maine

100%

0%

100%

0%

100%

0%

100%

0%

100%

0%

Maryland

100%

0%

100%

100%

100%

100%

100%

100%

100%

100%

Massachusett
s

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Michigan

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Minnesota

100%

0%

100%

100%

0%

0%

100%

100%

0%

0%

Mississippi

100%

0%

98%

0%

0%

0%

0%

0%

0%

0%

Missouri

100%

0%

100%

100%

100%

100%

100%

100%

100%

100%

Montana

100%

0%

100%

100%

100%

100%

100%

100%

100%

100%

Nebraska

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Nevada

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

New

Hampshire

100%

100%

100%

0%

0%

0%

0%

0%

0%

0%

New Jersey

100%

100%

100%

100%

0%

0%

100%

100%

0%

0%

New Mexico

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

New York

96%

0%

100%

100%

0%

0%

100%

100%

0%

0%

North
Carolina

100%

100%

100%

0%

0%

0%

100%

0%

0%

0%

North Dakota

83%

0%

83%

0%

0%

0%

0%

0%

0%

0%

Ohio

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Oklahoma

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Oregon

100%

100%

100%

100%

0%

0%

100%

100%

0%

0%

Final LCRI Economic Analysis

3-137

October 2024


-------
State

Percent of Child
Care Facilities
Eligible for
Waiver for:

Percent of Public

Elementary
Schools Eligible
for Waiver for:

Percent of Private
Elementary Schools
Eligible for Waiver
for:

Percent of Public

Secondary
Schools Eligible
for Waiver for:

Percent of

Private
Secondary
Schools Eligible
for Waiver for:

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

1st 5-yr
cycle

2nd 5-yr
cycles on

pp_childc
are_mand
_waiver

pp_childcar
e_onreq_w
aiver

pp_pub_el
em_mand
_waiver

pp_pub_ele
m_onreq_
waiver

pp_priv_ele
m_mand_
waiver

pp_priv_elem
_onreq_waiv
er

pp_childc
are_mand
_w aiver

pp_childcare
_onreq_wai
ver

pp_pub_el
em_mand
_w aiver

pp_pub_ele
m_onreq_
waiver

Pennsylvania

100%

0%

100%

100%

0%

0%

100%

100%

0%

0%

Rhode Island

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

South
Carolina

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

South Dakota

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Tennessee

93%

0%

93%

0%

0%

0%

0%

0%

0%

0%

Texas

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Utah

100%

100%

100%

0%

100%

0%

100%

0%

100%

0%

Vermont

100%

100%

100%

100%

100%

100%

100%

100%

100%

100%

Virginia

94%

0%

94%

0%

0%

0%

0%

0%

0%

0%

Washington

100%

100%

100%

100%

0%

0%

100%

100%

0%

0%

West Virginia

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Wisconsin

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Wyoming

100%

0%

100%

0%

0%

0%

0%

0%

0%

0%

Source: "School_Child Care lnputs_Final.xlsx," worksheet, "Waiver Eligibility_LCRI."

Final LCRI Economic Analysis

3-138

October 2024


-------
3.3.10.2.3 Discussion of Data Limitations and Uncertainty

There is uncertainty in the use of WIIN grant allocations for estimating the percentage of public
elementary schools and child care facilities that will be tested in the first five-year cycle. The amount of
available funding is based on the amounts appropriated in the BIL, recognizing that the full appropriated
amount may not be allocated. The EPA projected the amount of WIIN grant funding that would be
available for FY 2024 through FY 2026 based on the allocation for FY 2023. To try to reduce the
uncertainty in the number of facilities that could be tested using this WIIN grant funding, the EPA
developed low and high estimates per sampling event and used the average. The EPA assumed in this
analysis that States will use funding in the year it is awarded, but there is uncertainty in when a State
will use the grant funding. There is also uncertainty in how a State decides to spend its funding. There is
further uncertainty in the number of samples per school and in the use of data that include secondary
schools. There is also uncertainty in the voluntary participation of a school or child care facility in the
testing program.

3.3.11 Labor Rates

This section is divided into three subsections:

•	Section 3.3.11.1: presents PWS labor rates,

•	Section 3.3.11.2: presents State labor rates, and

•	Section 3.3.11.3: provides a discussion of data limitations and uncertainty associated with the
labor rates.

3.3.11.1 Public Water System Labor Rates

The EPA recognizes that there may be variation in labor rates across all systems. However, for purposes
of this EA, the EPA used national-level estimates from Labor Costs for National Drinking Water Rules
(USEPA, 2020c) with a few modifications, as described below.

The 2020 document evaluated three data sources for labor rates:

•	The 2019 Occupational Employment Survey (OES), a semi-annual U.S. Bureau of Labor Statistics
(BLS) survey that provides hourly wage estimates by occupation and industry (U.S. Bureau of
Labor Statistics, 2020).

•	The 2019 Water Utility Compensation Survey, an annual AWWA survey that provides hourly
wage estimates for the water and wastewater industry by occupation. Data are in 2008 dollars
(AWWA, 2019a, 2019b, 2019c).

•	The 2006 CWSS, a periodic EPA survey that obtains employment information from a sample of
CWSs. Wage rates are escalated from 2007 to 2019 dollars using an employment cost index
(USEPA, 2009).

Final LCRI Economic Analysis

3-139

October 2024


-------
In 2020, the EPA evaluated these data sources against suitability criteria (see Exhibit 3-74) (USEPA,
2020c).83 The EPA determined that the 2006 CWSS was the most suitable source for labor rates
associated with national drinking water rules particularly because: the data are specific to drinking
water; the survey responses can be extrapolated to national estimates since the survey has a known
sampling framework; and the data can be organized by system size, source, and ownership (USEPA,
2020c).

Exhibit 3-74: Comparison of Wage Rate Surveys

Suitability
Criteria

OES (BLS)(2019)

AWWA (2019)

2006 CWSS

National
average wage
rates

Yes

Annual updates available

No

Sample of systems serving
<10,000 people may not be
representative of all small

systems
Annual updates available

Yes

Updates are periodic

Data quality

High

Statistically precise wage
estimates

Unknown
Sampling procedures
unknown; no information on
statistical precision of wage
estimates

Moderate
Low item response rates
among small systems lead to
large confidence intervals

Drinking water
industry data

No1

Yes

Yes

Management,
Technical, and
Administrative
occupations

Yes

No administrative occupation
to match WBS needs for
medium or large systems

Administrative occupation may
differ from WBS needs

System size
differentiation

No

Yes

Yes

Source water
differentiation

No

No

Yes

Estimates are not statistically
significantly different across
source waters

Ownership
differentiation

No

No

Yes

Estimates are not statistically
significantly different across
private and public ownership

Acronyms: AWWA = American Water Works Association; CWSS = Community Water System Survey; BLS = Bureau
of Labor Statistics; OES = Occupational Employment Survey; WBS = work breakdown structure.

Source: From Exhibit ES-1 (USEPA, 2020c).

Notes:

1 OES data are available for two North American Industry Classification System (NAICS) categories that are likely to
contain drinking water systems: 221300 (Water, Sewage, and Other Systems) for private water systems and
999300 (Local Government) for PWSs.

83 Note that, based on publicly available information, the EPA did not find significant survey reporting updates in
more recent OES or AWWA surveys; therefore, the comparisons between these 2019 surveys and the CWSS are
still accurate.

Final LCRI Economic Analysis

3-140

October 2024


-------
Exhibit 3-75 presents the labor rate estimates used in USEPA (2020b) in 2019 dollars. Labor rates were
calculated for three occupation categories: manager, treatment plant operator, and administrative
personnel. The rates include benefits. The EPA considered benefit multipliers from BLS and the 2006
CWSS. Benefit multipliers from BLS ranged from 1.3 to 1.5, and benefit multipliers from the 2006 CWSS
ranged from 1.2 to 1.4. The BLS estimates are more precise than the 2006 CWSS estimates, but
information was not available at the industry level. The CWSS estimates are related to the drinking
water industry but have large confidence intervals and low precision. Ranges from both sources
overlapped at 1.4; however, the EPA used a benefit multiplier of 1.45 because updated BLS data for
industries and occupations that include water treatment utilities showed that State and Local
Government benefits were 0.11 - 0.13 points higher in 2019 compared to 2006. A benefit multiplier of
1.45 is the current multiplier for all civilians working in service-producing industries (USEPA, 2020c).

The EPA used the employment cost index (ECI) to escalate wage rates and convert dollar values to 2019
dollars. To adjust the managerial wage rates from 2007 to 2019 dollars, the EPA used the ECI escalation
rate of 149.6/104.5, or 43%. The EPA did not use the 43% ECI escalation rate for technical wage rates
because it appeared to overstate growth in technical rates for operator and maintenance activities
compared to the OES and AWWA data, particularly among larger systems (USEPA, 2020c). It also
overstated growth in administrative rates.84 The EPA accounted for this variation by escalating the CWSS
value for technical and administrative wage rates to 2019 dollars using the OES change in mean wage
rate from 2007 to 2019. The EPA used a revised escalation value of 32.1% for the technical rate and an
escalation value of 33% for the administrative rate.

Exhibit 3-75: Hourly Labor Costs Including Wages Plus Benefits (2019$)

System Size (Population
Served)

Hourly Labor Cost by Occupation

Manager

Treatment plant
operator

Administrative
personnel

<500

$48.20

$32.51

$31.31

501-3,300

$48.20

$32.51

$31.31

3,301-10,000

$55.13

$34.67

$31.31

10,001-50,000

$61.41

$36.60

$40.43

50,001-100,000

$71.66

$38.21

$40.43

>100,000

$76.56

$44.66

$40.43

Source: Labor Costs for National Drinking Water Rules (USEPA, 2020c), Exhibit 3-5, which is based on 2006 CWSS
(USEPA, 2009).

To account for the general composition of staff at systems of smaller sizes, {i.e., those serving 3,300 or
fewer people), the EPA used only the technical rate (i.e., treatment plant operator rate). For systems
serving more than 3,300 people, the EPA used proportions of 80 percent technical labor and 20 percent
managerial labor (i.e., manager rate) to arrive at a labor cost, or weighted labor rate. The actual
proportions between technical and managerial rates employed may vary by PWS and among the
different compliance activities under the final LCRI. However, for simplicity, the EPA used the 80/20

84 The EPA determined that the CWSS wage rates for technical workers (USEPA, 2009) escalated using the ECI were
4 to 14 percent higher than the wage rates in the OES survey (United States Bureau of Labor Statistics, 2018) and
the AWWA surveys (AWWA, 2019a; 2019b; 2019c). The analysis is described in Chapter 3 of Labor Costs for
National Drinking Water Rules (USEPA, 2020c).

Final LCRI Economic Analysis

3-141

October 2024


-------
proportions as a general assumption to develop system labor costs for this EA. This approach for
developing system labor rates is consistent with that used for other economic analyses, such as the
Revised Total Coliform Rule (USEPA, 2012b). The EPA further adjusted the labor rates from 2019 dollars
to 2020 dollars using an ECI escalation rate of 140.7/137.0 for all labor categories. The final labor rates
used in this EA are in column D in Exhibit 3-76 below and are used for both CWSs and NTNCWSs. The
final labor rates used in this EA are in column D in Exhibit 3-76 below.

Exhibit 3-76: Weighted Labor Rates for CWSs and NTNCWSs

System Size
(Population Served)

Technical Labor
Rate
(2019$/hour)

Managerial Labor
Rate (2019$/hour)

Weighted System Labor
(2019$/hour)

Weighted System
Labor Adjusted to
2020$ (2020$/hour)

A

B

C = A for PWSs < 3,300);
C = (0.8*A) + (0.2*B) for
PWSs > 3,300

D = C*( 140.7/137.0)

<3,300

$32.51

$48.20

$32.51

$33.39

3,301-10,000

$34.67

$55.13

$38.76

$39.81

10,001-50,000

$36.60

$61.41

$41.56

$42.68

50,001-100,000

$38.21

$71.66

$44.90

$46.11

>100,000

$44.66

$76.56

$51.04

$52.42

Sources: A, B: Labor Costs for National Drinking Water Rules (USEPA, 2020c), Exhibit 3-5. Hourly labor costs include

wages and benefits. Technical labor wage rates are based on wage rates for treatment plant operators.

Notes:

General: Labor rates for each size category are assumed to be the same regardless of system type (CWSs or
NTNCWSs). In general, information in this chapter is presented by the nine size categories used in the SafeWater
LCR model. In this exhibit, the EPA merged size categories with the same hourly rate.

C: The EPA estimates that systems serving 3,300 people or fewer use 100 percent technical labor, whereas systems
serving more than 3,300 use 80 percent technical (operator) labor and 20 percent managerial (engineer) labor.
D: The weighted system hourly rate was adjusted to 2020 dollars using the general employment cost index (ECI)
seasonally adjusted June for 2019 (137.0) and June 2020 (140.7), as shown in the file, "General Cost Model
lnputs_Final," worksheet, "ECI Table 1."

3.3.11.2 State Labor Rates

The EPA used the hourly mean State employee labor rate from the BLS May 2020 Occupational and
Employment Wages (OEWS) table. Specifically, the EPA used the hourly labor rate from the category
"19-2041 Environmental Scientists and Specialists, Including Health" as an approximation for the State
labor rate. Within that category, the EPA used the hourly mean wage for State Government, excluding
schools and hospitals. This approximation is a reasonable estimate because the majority of primacy
agencies are States. The base hourly State labor rate is $33.91 in 2020 dollars. The EPA further adjusted
the labor rate using a 1.62 loading factor that reflects additional employee benefits from the BLS
Employer Costs for Employee Compensation report, "Table 1. Employer Costs for Employee
Compensation by Ownership (June 2020)." (See worksheet "BLS Table 1" in the file, "General Cost
Model lnput_Final.xlsx" for additional information.) The final "loaded" hourly rate of $54.78 is used for
the State labor rate in the SafeWater LCR model and is designated with the data variable name of

Final LCR! Economic Analysis

3-142

October 2024


-------
rateJs. Calculations and the loaded labor rate are shown in the Exhibit 3-77 (also see "General Cost
Model lnputs_Final/' worksheet, "State Labor Rates").85

Exhibit 3-77: Loaded Labor Rate for State Staff (2020$)

Base Hourly Labor Cost

Loading Factor

Loaded Hourly Labor Rate (2020$)

rateJs

A

B

C = A*B

$33.91

1.62

$54.78

Sources:

A: State employee wage rates from National Occupational Employment and Wage Estimates, United States, BLS
SOC Code 19-2041, "State Government - Environmental Scientists and Specialists, Including Health," hourly mean
wage rate for "State Government, excluding schools and hospitals (OEWS Designation)." May 2020 data:
https://www.bls.gov/oes/current/oesl92041.htm. See worksheet OES in the file, "General Cost Model
lnput_Final.xlsx," worksheet "State Labor Rates."

B: Wages are loaded using a factor from the BLS Employer Costs for Employee Compensation by Ownership, Table
1, June 2020. State and local government workers. Percent of compensation.

https://www.bls.gov/news.release/archives/ecec 09172020.htm. See "General Cost Model lnput_Final.xlsx,"
worksheet, "BLS Table 1."

3.3.11.3 Discussion of Data Limitations and Uncertainty

There is uncertainty in the derivation of water system labor rates that could result in an over or
underestimate of national costs of the final LCRI. The wage rates are based on the 2006 CWSS data and
escalated to a particular dollar year using an ECI. The labor rate mix may have changed since the time of
the survey. Moreover, the labor rate used is a national average and does not capture differences across
regions or between urban and rural areas. The EPA accounted for general changes in the cost of labor
over time by adjusting 2007 values to 2020 using the ECI.

Additionally, the wage rates based on the 2006 CWSS data values were found to overstate labor costs
for technical and administrative labor when compared to the OES and AWWA surveys. The EPA revised
the escalation rate from the ECl-based value using the OES change in mean wage rate from 2007 to 2019
for technical and administrative wage rates to account for variation. For managerial hours, the wages
did not clearly over- or understate wages compared to OES data, but were consistently lower than
AWWA wage estimates.

There is also uncertainty in assuming a 1.45 benefits multiplier and that a labor mix of 80/20
technical/managerial staff will apply to activities conducted by CWSs and NTNCWSs serving more than
3,300 people. There may be situations where an activity is performed just by technical staff, e.g., sample

85 Note that although the EPA used more current BLS information for this economic analysis, the State labor rate is
lower than the one used to estimate costs for the final 2021 LCRR of $57.24 (2016$). This is because for the final
2021 LCRR economic analysis (USEPA, 2020a), the EPA used the hourly mean wage for all employees in the
category of "Environmental Scientist and Specialists, Including Health from the May 2016 Occupational and
Employment Wage" information from the BLS, which yielded a base labor hourly labor rate of $36.23 (2016$). For
the current economic analysis, the EPA revised this estimate to use the subcategory that most closely
approximates State primacy agencies labor costs.

Final LCRI Economic Analysis

3-143

October 2024


-------
analysis, or just by managerial staff, e.g., reporting. This may cause an under or overestimation of cost of
the final LCRI.

Similarly, there is uncertainty in the derivation of the State labor rate that could result in an over or
underestimation of national cost. The EPA tried to reduce the uncertainty by using the base hourly labor
rate of the subcategory "State Government, excluding schools and hospitals," within the larger labor
category of an "Environmental Scientists and Specialists, Including Health", as opposed to using the
average from the larger category, as was done for the Final 2021 LCRR EA. Some of the activities
undertaken by the State may include support staff, more technical staff, or management staff that have
a lower or higher base rate. There is also uncertainty in assuming an average State employee hourly
labor rate includes a loading factor of 1.62. This factor, provided by BLS, is an average across all State
and local governments and job categories. This assumption could result in an under or overestimation of
the State labor costs estimated for the final LCRI.

3.4 Uncertainties in the Baseline and Compliance Characteristics of Systems

Uncertainties in the baseline and compliance characteristics of PWSs, which can apply to systems under
the pre-2021 LCR, 2021 LCRR, and final LCRI analysis, are due to the limits of available information. The
largest sources of uncertainty include the following three variables: 1) the number of PWSs that will
exceed the AL under the revised tap sampling requirements, 2) the cost of service line replacement, and
3) the cost of CCT treatment. The EPA is using low and high scenarios defined by the assignment of low
and high values listed above to assess the potential impact of these uncertain variables on the costs and
benefit of the final LCRI. Detailed descriptions of the uncertain variables and the derivation of their
values can be found in Chapter 4, Section 4.2.2.

In addition to the uncertainty, which is represented in the cost range, the EPA acknowledges that there
are other uncertainties associated with the inputs to the cost-benefit model. The EPA has described
these uncertainties related to system characteristics throughout the text in this chapter and Exhibit 3-78
provides a summary of these uncertainties.

Exhibit 3-78: Summary of Uncertainties in the Baseline and Compliance Characteristics of

Drinking Water Systems

Uncertainty Description

Effect on Costs

Effect on
Benefits

Relevant
Section(s)

System and CCT Characterization

For PWSs with unknown source type, uncertainty
in assigning source type based on the ratio of
systems with known primary GW or SW sources.

+/-

+/-

3.3.1

Uncertainty associated with changing population
demographics including fertility and immigration
rates, and within country migration affecting the
number and size of PWSs

+/-

+/-

3.3.1

3.3.2

Uncertainty in using 2020 census data on average
persons per household to estimate number of
households served by CWSs.

+/-

+/-

3.3.2

Final LCRI Economic Analysis

3-144

October 2024


-------
Uncertainty Description

Effect on Costs

Effect on
Benefits

Relevant
Section(s)

Uncertainty in using retail population to predict
average daily flow per PWS in situations where the
PWS is a wholesale system that sells water to a
consecutive PWS.

+

None

3.3.2

Uncertainty in assuming that all PWSs serving
50,000 or more except "b3" systems1 have CCT.
For those serving 50,000 or fewer, uncertainty in
using SDWIS/Fed treatment data to estimate
percent with CCT.

+/-

+/-

3.3.3

Uncertainty in assuming that flow is proportioned
equally among all entry points in a given system.

+/-

+/-

3.3.6

Uncertainty in using historical CWSS data as
analyzed in the Geometries document (USEPA,
2000) to predict population/flow relationships
given water efficiency trends over the last 20
years.

+

None

3.2.4
3.3.6

Uncertainty in estimated percent of PWSs with pH
adjustment only, orthophosphate only, or both
based on SDWIS/Fed historical data. Uncertainty
in using SYR3 ICR dataset to estimate baseline pH
and orthophosphate concentration.

+/-

+/-

3.3.6

Uncertainty in average daily flow used to estimate
CCT costs due to the fact that household water
use has generally declined over the period since
this analysis was completed.

+

None

3.2.4

Lead Service Line Characterization

For CWSs, uncertainty in the extent of unknowns
in the service line material characterization that
are lead or non-lead.

+/-

+/-

3.2.5
3.3.4.1

For NTNCWS, uncertainty in using data from two
States to estimate percent of systems with LSLs.
Uncertainty in assumptions related to percent of
connections that are lead within NTNCWSs that
are known to have LSLs.

+/-

+/-

3.3.4.2

Uncertainty in using historical SDWIS/Fed 90th
percentile data from 2012-2020 to predict future
90th percentile lead levels and percent of systems
with no ALE and an ALE under the final LCRI
conditions. Includes adjustments for the final LCRI
requirement for LSL systems to collect all samples
from locations served by an LSL and to use the
higher of the 1st and 5th liter sample in the 90th
percentile calculation.

+/-

+/-

3.2.1
3.3.5.1

Uncertainty in using a subset of CWSs with known
LSL status to predict future 90th percentile values.

+/-

+/-

3.3.5.1

Final LCRI Economic Analysis

3-145

October 2024


-------
Uncertainty Description

Effect on Costs

Effect on
Benefits

Relevant
Section(s)

Uncertainty in using data from a single State
(Michigan) to estimate the impact on the lead 90th
percentile levels that will be based on fifth-liter vs.
first-liter samples under the final LCRI for LSI-
systems.

+/-

+/-

3.2.7
3.3.5.1

Uncertainty in using historical SDWIS/Fed 90th
percentile data from 2012-2020 to predict future
90th percentile lead levels and percent of systems
that will have at least two or three lead ALEs in a
five-year period.

+/-

+/-

3.2.1
3.3.5.2

Uncertainty in basing the likelihood of a sample
exceeding 10 ng/L based on data from a single
State (Michigan).

+/-

+/-

3.2.7
3.3.5.3

Uncertainty in basing the likelihood a system has a
copper ALE based on historical copper 90th data
reported for 2012 - 2020.

+/-

+/-

3.3.5.4

Lead and WQP Monitoring Schedules

Uncertainty in using schedule based on historical
P90, Cu90, milestone, treatment, and violation
data from SDWIS/Fed to estimate initial WQP
schedules under the final LCRI.

+/-

+/-

3.3.8.2

Change in Source or Treatment

Uncertainty in using historical information on
source water type from SDWIS/Fed to estimate
the percent of systems that will change source
each year. May underestimate costs by not
counting when the same type of source was
added and removed at a system in a given year.

-

+/-

3.2.1
3.3.9.1

Uncertainty in using historical treatment code
data from SDWIS/Fed to estimate the percent of
systems making a treatment change each year.

+/-

+/-

3.2.1
3.3.9.3

Schools, Child Care Facilities, Local Health Departments, and Targeted Medical Providers

Uncertainty in number of schools based on NCES
data for 2018-2020 (NCES, 2020a, 2020b) and in
the number of child care facilities based on the
2019 update to a report by the Committee for
Economic Development (CED, 2019 due to
population growth.

-

None

3.2.8.1
3.3.10.1.1

Uncertainty in number of child care facilities based
on 2019 industry statistics (CED, 2019) due to
potential inclusion of unlicensed at-home
facilities.

+

None

3.2.8.2
3.3.10.1.2

Final LCRI Economic Analysis

3-146

October 2024


-------
Uncertainty Description

Effect on Costs

Effect on
Benefits

Relevant
Section(s)

Uncertainty in identification of local health
agencies number based on 2016 statistics
(NACCHO, 2017).

No effect on
incremental
costs.

Requirement to
deliver to this
group is the same
under the pre-
2021 LCR, 2021
LCRR, and final
LCRI.

None

3.2.8.3
3.3.10.1.3

Possible overestimation of States that may grant
waivers to CWSs for sampling in schools and child
care facilities based on the assumed use of WIIN
grant funding.

-

None

3.3.10.2.2

Labor Rates

Uncertainty in using 2006 CWSS data for PWS
labor rates, 1.45 benefits multiplier, and 80/20
technical/managerial staffing mix for PWSs serving
more than 3,300 people.

+/-

None

3.3.11.1

Uncertainty in basing the State labor rate on the
wage rate category for Environmental Scientists
and Specialists (activities may be done by staff
with higher or lower rates) and uncertainty in
using a single benefits loading factor of 1.62 for
employee compensation.

+/-

None

3.3.11.2

Acronyms: ALE = action level exceedance; CED = Committee for Economic Development; CCT = corrosion control
treatment; CWS = community water system; CWSS = Community Water System Survey; LCRR = Lead and Copper
Rule Revisions; LSL = lead service line; NACCHO = National Association of County and City Health Officials; NCES =
National Center for Education Statistics; NTNCWS = non-transient non-community water system; P90 = lead 90th
percentile value; PWS = public water system; SDWIS/Fed = Safe Drinking Water Information System - Federal
version; SYR3 ICR = Six-Year Review 3 Information Collection Request.

Notes:

General: This exhibit indicates whether each uncertainty factor contributes to understating (-), overstating (+), or
either understating or overstating (+/-) the overall economic impact results.

1 Excluded 16 CWSs serving 50,000 that were assumed to meet the b3 criteria, i.e., have naturally non-corrosive
water, and under the pre-2021 LCR are not required to install CCT. See Section 3.3.3 for additional information.

3.5 References

American Water Works Association (AWWA). 2019a. 2019 AWWA Compensation Survey: Large Water
and Wastewater Utilities serving populations above 100,000. Denver, CO: AWWA.

AWWA. 2019b. 2019 AWWA Compensation Survey: Medium Water and Wastewater Utilities serving
populations between 10,000 and 99,999. Denver, CO: AWWA.

Final LCR! Economic Analysis

3-147

October 2024


-------
AWWA. 2019c. 2019 AWWA Compensation Survey: Small Water and Wastewater Utilities serving
populations under 10,000. Denver, CO: AWWA.

Black & Veatch. 2004. Notes from the EPA Lead Service Line Replacement Workshop. Conducted for
American Water Works Association. December 10, 2004.

Committee for Economic Development (CED). 2019. Child Care in State Economies, 2019 Update.
https://www.ced.org/assets/reports/childcareimpact/181104%20CCSE%20Report%20Jan30.pdf.

Cornwell, D.A, R.A. Brown, and S.H Via. 2016. National Survey of Lead Service Line Occurrence. Journal
AWWA. 108(4):E182-E191.

Michigan EGLE. 2020. Preliminary Distribution System Material Inventory. Available at
https://www.michigan.gov/documents/egle/egle-dwehd-PDSMISummaryData 682673 7.pdf.

National Association of County and City Health Officials (NACCHO). 2019. 2019 National Profile of Local
Health Departments. Available at https://www.naccho.org/uploads/downloadable-
resources/Programs/Public-Health-lnfrastructure/NACCHO 2019 Profile final.pdf.

National Center for Education Statistics (NCES). 2020a. Table 105.50. Number of educational institutions,
by level and control of institution: Selected years, 1980-81 through 2018-19. Available at
https://nces.ed.gov/programs/digest/d20/tables/dt20 105.50.asp.

NCES. 2020b. Private School Universe Survey, 2019 - 2020. Table 15: Number of private schools,
students, full-time equivalent (FTE) teachers, and 2018-2019 high school graduates, by state: United
States, 2019-2020. Available at https://nces.ed.gov/survevs/pss/tables/TABLE15fll920.asp.

Rockaway, T.D., P.A. Coomes, J. Rivard, and B. Kornstein. 2011. Residential water use trends in North
America. Journal AWWA 103(2):76-89.

Slabaugh, R.M., R.B. Arnold, S. Chaparro, and C.P. Hill. 2015. National cost implications of potential long-
term LCR requirements. Journal A WWA. 107(8):E389-E400.

United States Bureau of Labor Statistics. 2018. Occupational Employment Statistics: 2017 Occupation
Profiles. Retrieved from https://www.bls.gOv/oes/2017/may/oes_stru.htm#43-0000.

United States Bureau of Labor Statistics. 2020. May 2019 National Industry-Specific Occupational
Employment and Wage Estimates: NAICS 221300 - Water, Sewage and Other Systems. Retrieved from
https://www.bls.gov/oes/current/naics4 221300.htm.

United States Bureau of Labor Statistics. 2021. Occupational Outlook Handbook, Physicians and
Surgeons. Retrieved from http://www.bls.gov/ooh/healthcare/phvsician-and-surgeons.htm. Last
modified September, 2021. Accessed December 21, 2021.

United States Census Bureau. 2020. Table AVG1. Average Number of People Per Household, By Race
And Hispanic Origin, Marital Status, Age, And Education Of Householder: 2020.
https://www2.census.gov/programs-surveys/demo/tables/families/2020/cps-2020/tabavgl.xls.

United States Environmental Protection Agency (USEPA). 2000. Geometries and Characteristics of Public
Water Systems. December 2000. EPA 815-R-00-24.

Final LCR! Economic Analysis

3-148

October 2024


-------
USEPA. 2003. Drinking Water Baseline Handbook, Fourth Edition.

USEPA. 2007. Office Ground Water and Drinking Water's Error Code Tracking Tool [for SDWIS/Fed].
Developed 2007.

USEPA. 2009. 2006 Community Water System Survey Volume II: Detailed Tables and Survey
Methodology. May 2009. Office of Water. EPA 815-R-09-002. Available at
https://nepis.epa.gov/Exe/ZyPDF.cgi/P1009USA.PDF?Dockev=P1009USA.PDF.

USEPA. 2012a. Revisions to the Unregulated Contaminant Monitoring Regulation (UCMR 3) for Public
Water Systems. Federal Register. 11 FR26072. May 2, 2012. Available at
https://www.govinfo.gov/content/pkg/FR-2012-05-02/pdf/2012-9978.pdf.

USEPA. 2012b. Economic Analysis for the Final Revised Total Coliform Rule. September 2012. Office of
Water. EPA 815-R-12-004. Available at

https://nepis.epa.gov/Exe/ZvPDF.cgi/P100PIVQ. PDF?Dockev=P100PIVQ. PDF.

USEPA. 2014. Guidelines for Preparing and Economic Analyses. Office of Policy. December 17, 2010
(updated May 2014). Available at https://www.epa.gov/sites/default/files/2017-08/documents/ee-
0568-50.pdf.

USEPA. 2016a. Safe Drinking Water Information System Federal Version (SDWIS/Fed) Data Reporting
Requirements, vl.2. March 2016. Office of Ground Water and Drinking Water.

USEPA. 2016b. The Analysis of Regulated Contaminant Occurrence Data from Public Water Systems in
Support of the Third Six-Year Review of National Primary Drinking Water Regulations: Chemical Phase
Rules and Radionuclides Rules. December 2016. Office of Water. EPA 810-R-16-014. Available at
https://www.epa.gov/sites/default/files/2016-12/documents/810rl6014.pdf.

USEPA. 2016c. The Data Management and Quality Assurance/Quality Control Process for the Third Six-
Year Review Information Collection Rule Dataset. December 2016. Office of Water. EPA 810-R-16-015.
Available at https://www.epa.gov/sites/default/files/2016-12/documents/810rl6015 O.pdf.

USEPA. 2017. UCMR 3 (2013 - 2015) Occurrence Data. January 2017. Available at
https://www.epa.gOv/dwucmr/occurrence-data-unregulated-contaminant-monitoring-rule#3.

USEPA. 2018. 3Tsfor Reducing Lead in Drinking Water in Schools and Child Care Facilities: A Training,
Testing, and Taking Action Approach (Revised Manual). October 2018. Office of Water. EPA815-B-18-
007. Available at https://www.epa.gov/ground-water-and-drinking-water/3ts-reducing-lead-drinking-
water-toolkit.

USEPA. 2019. Occurrence Data from the Third Unregulated Contaminant Monitoring Rule (UCMR 3). EPA
815-R-19-007.

USEPA. 2020a. Economic Analysis for the Final Lead and Copper Rule Revisions. December 2020. Office of
Water. EPA 816-R-20-008.

USEPA. 2020b. State Lead in School and Child Care Program Drinking Water Grant Implementation
Document. March 2020. Office of Water. Available at

Final LCRI Economic Analysis

3-149

October 2024


-------
https://www.epa.gov/svstem/files/documents/2021-

ll/fy2020 implementation document for wiin 2107 testing in schools updated.pdf.

USEPA. 2020c. Labor Costs for National Drinking Water Rules. Submitted to Rajiv Khera, Office of Ground
Water and Drinking Water, USEPA. October 2020. EPA Contract No. EP-B16C-0001.

USEPA. 2023a. Fact Sheet: 7th Drinking Water Infrastructure Needs Survey and Assessment, April 2023.
Available at https://www.epa.gov/system/files/documents/2023-
04/Final DWINSA%20Public%20Factsheet%204.4.23.pdf.

USEPA. 2023b. Economic Analysis for the Proposed Lead and Copper Rule Improvements. November

2023.	Office of Water. EPA 815-R-23-005.

USEPA. 2023c. Voluntary School and Child Care Lead Testing & Reduction Grant Program
Implementation Document for States and Territories. July 2023. Office of Water. EPA 815-B-23-009.
https://www.epa.gov/system/files/documents/2023-

07/Final FY23 ImplementationDoc VoluntarvSchoolandChildCareLeadTestingReductionGrantProgram
508.pdf

USEPA. 2023d. Drinking Water Infrastructure Needs Survey and Assessment: 7th Report to Congress.
September 2023. Office of Water .EPA 810-R-23-001.

https://www.epa.gov/system/files/documents/2023-9/Seventh%20DWINSA September2023 Final.pdf

USEPA. 2024a. Fact Sheet: Updated 7th Drinking Water Infrastructure Needs Survey & Assessment, May

2024.	Available at https://www.epa.gov/svstem/files/documents/2024-05/fact-sheet one-time-
update 2024.04.30 508 compliant l.pdf

USEPA. 2024b. Memorandum: Fiscal Year 2024 Lead Service Line Allotments for the Drinking Water
State Revolving Fund Provisions of the Bipartisan Infrastructure Law Funding. Available at
https://www.epa.gov/svstem/files/documents/2024-05/fy24-bil-lslr-allotments-memorandum may-
2024.pdf

Water Research Foundation (WaterRF). 2016. Residential End Uses of Water, Version 2, Executive
Report. Available at https://www.circleofblue.org/wp-content/uploads/2016/04/WRF REU2016.pdf.

Final LCRI Economic Analysis

3-150

October 2024


-------
4 Economic Impact and Cost Analysis of the Final Lead and Copper

Rule Improvements

4.1 Introduction

The final Lead and Copper Rule Improvements (LCRI) accelerates the removal of lead and certain
galvanized service lines and reduces the lead action level to 0.010 mg/L. The final LCRI also strengthens
tap sampling procedures, corrosion control treatment (CCT), public education and consumer awareness,
requirements for small systems, and sampling in schools and child care facilities. In this chapter, the EPA
presents its estimates of, and approach to estimate, the national incremental cost of the final LCRI. To
determine the incremental national cost of the final LCRI, the EPA estimated the additional costs that
public water systems (PWSs), households, and primacy agencies (note: this document uses "States" to
refer generally to primacy agencies) will incur in response to the final LCRI, above the cost they would
face under the 2021 Lead and Copper Rule Revisions (LCRR) if the LCRI was not enacted. To determine
the incremental cost of the final LCRI, the agency first calculated the costs that would be incurred in
continuing to comply with the 2021 LCRR. Next, the agency estimated the cost that PWSs, households,
and States would incur in response to the final LCRI if no 2021 LCRR requirements were currently in
place. Under both the 2021 LCRR and the LCRI, the EPA removed those lead and galvanized requiring
replacement (GRR) service line replacements that would occur in the baseline as a result of State service
line replacement (SLR) requirements.86 The incremental national cost of the final LCRI is the difference
between the cost of compliance with the final LCRI and the cost of compliance with the 2021 LCRR.

Note that the incremental national costs of the final LCRI when compared to the pre-2021 Lead and
Copper Rule (LCR) have also been computed and are provided in Appendix C. Appendix B also explains
how the EPA developed the cost values for the pre-2021 LCR, which were subtracted from the final LCRI
costs to produce the incremental cost of moving from the pre-2021 LCR to the final LCRI requirements.

4.1.1 Summary of Rule Costs

The annualized costs, discounted at 2 percent, that PWSs, households, and States will incur in complying
with the 2021 LCRR and the final LCRI are summarized in Exhibit 4-1. The EPA used the 2 percent
discount rate as prescribed by the Office of Management and Budget's updated Circular A-4 (OMB
Circular A-4, 2023).87 See Section 4.2.3 below for additional information on discounting. The EPA
estimated costs of the final LCRI under both low and high scenarios to reflect uncertainty in the cost
estimates. The low scenario and high scenario differ in their assumptions made about: 1) the number of

86	Four States (Illinois, Michigan, New Jersey, and Rhode Island) have passed state laws that require lead service
line replacement (LSLR) at various rates from 2-10 percent. The EPA has included replacements associated with
these programs in the baseline; therefore, the costs of replacing these LSLs do not appear in the estimated SLR
costs under either the 2021 LCRR or LCRI. There are other statewide and municipal voluntary or goal-based
programs to replace all LSLs within the next 10 or more years. However, because these are not legal requirements,
the EPA does not include them in its estimate of the number of LSLs that would be replaced in the baseline. See
Chapter 3, Section 3.3.4.3 for more information.

87	Because the EPA provided cost estimates discounted at 3 and 7 percent for the proposed LCRI based on OMB
guidance which was in effect at the time of the proposed rule analysis (OMB Circular A-4, 2003), the agency has
also calculated the cost impacts at both the 3 and 7 percent discount rates. See Appendix F for results.

Final LCRI Economic Analysis

4-1

October 2024


-------
PWSs above the action level (AL) for the 2021 LCRR and final LCRI and trigger level (TL) for the 2021
LCRR, and final LCRI monitoring requirements; 2) the cost of installing and optimizing corrosion control
treatment (CCT); and 3) the cost of SLR. The EPA discusses these assumptions in more detail below and
in Section 4.2.2.

The monetized incremental annualized cost of the final LCRI ranges from $1.47 billion to $1.95 billion at
a 2 percent discount rate in 2022 dollars. The exhibits also detail the proportion of the annualized costs
attributable to each rule component.

Exhibit 4-1: Estimated National Annualized Rule Costs - 2 Percent Discount Rate (millions of

2022 USD)





Low Estimate





High Estimate





Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS Annual Costs













Sampling

$134.0

$166.0

$32.0

$143.6

$176.2

$32.6

PWS SLR*

$84.6

$1,259.0

$1,174.4

$124.5

$1,763.9

$1,639.4

Corrosion Control
Technology

$552.0

$591.1

$39.1

$647.8

$692.9

$45.1

Point-of Use













Installation and

$2.4

$5.1

$2.7

$5.9

$9.6

$3.7

Maintenance













Public Education
and Outreach

$69.6

$267.3

$197.7

$72.1

$302.2

$230.1

Rule













Implementation
and

$0.1

$3.4

$3.3

$0.2

$3.4

$3.2

Administration













Total Annual
PWS Costs

$842.7

$2,291.9

$1,449.2

$994.1

$2,948.2

$1,954.1

Household SLR
Costs**

$8.1

$0.0

-$8.1

$26.4

$0.0

-$26.4

State Rule













Implementation
and

$38.4

$66.1

$27.7

$41.8

$67.6

$25.8

Administration













Wastewater













Treatment Plant

$3.0

$3.0

$0.0

$4.8

$5.1

$0.3

Costs***













Total Annual
Rule Costs

$892.2

$2,361.0

$1,468.8

$1,067.1

$3,020.9

$1,953.8

Acronyms: LCRI = Lead and Copper Rule Improvements; SLR = service line replacement; PWS = public water
system; USD = United States dollar.

Notes: Previous Baseline costs are projected over the 35-year period of analysis and are affected by the EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

*SLR includes full and partial LSLs and GRR service lines.

Final LCRI Economic Analysis

4-2

October 2024


-------
**ln the Economic Analysis for the Final Lead and Copper Rule Revisions (USEPA, 2020), the EPA assumed that the
cost of customer-side SLRs made under the goal-based replacement requirement would be paid for by households.
The agency also assumed that system-side SLRs under the goal-based replacement requirement and all SLRs (both
customer-side and systems-side) would be paid by the PWS under the 3 percent mandatory replacement
requirement. The EPA made these modeling assumptions based on the different levels of regulatory responsibility
systems faced operating under a goal-based replacement requirement versus a mandatory replacement
requirement. While systems would not be subject to a potential violation for not meeting the replacement target
under the goal-based replacement requirement, under the 3 percent mandatory replacement requirement the
possibility of a violation could motivate more systems to meet the replacement target even if they had to adopt
customer incentive programs that would shift the cost of replacing customer-side service lines from customers to
the system. To be consistent with these LCRR modeling assumptions, under the final LCRI, the EPA assumed that
mandatory replacement costs would fall only on systems. Therefore, the negative incremental values reported for
the "Household SLR Costs" category do not represent a net cost savings to households. They represent an assumed
transfer of the estimated SLR costs from households to systems. The EPA has insufficient information to estimate
the actual SLR cost sharing relationship between customers and systems at the national level of analysis.

***Due to many water systems operating both the wastewater and drinking water systems, the EPA is evaluating
the costs of additional phosphate usage for informational purposes. These costs are not "likely to occur solely as a
result of compliance" with the final LCRI, and therefore are not costs considered as part of the HRRCA under
SDWA, section 1412(b)(3)(C)(i)(lll).

OMB Circular A-4 (OMB, 2023) defines a "transfer" as: "... a shift in money (or other item of value) from
one party to another. More generally, when a regulation generates a gain for one group and an equal-
dollar-value loss for another group, the regulation is said to cause a transfer from the latter group to the
former." The final LCRI has both known transfers and potential transfers associated with the
implementation of the rule's requirements. The transfers discussed here do not affect the estimated
total monetized annualized social costs of the final LCRI. Tracking these transfers helps the agency
understand the distributional impact of costs across affected groups.

Implementation of the final LCRI will result in inter-community transfers, which is defined as the shift in
cost burden associated primarily with inventory development, and lead and GRR service line
replacement from the PWS and the community it serves, including customers whose private side lead or
GRR service lines were replaced, to an outside entity or group.

Congress enacted the Infrastructure Investment and Jobs Act (Pub. L. 117-58), also referred to as the
Bipartisan Infrastructure Law (BIL), which included $15 billion specifically appropriated for lead service
line replacement (LSLR) projects and associated activities directly connected to the identification and
replacement of LSLs. The BIL also included over $11.7 billion for the Drinking Water State Revolving Fund
(DWSRF) General Supplemental, which can be used for LSLR as well as other drinking water projects. The
$15 billion in specified LSLR BIL funding, when used by PWSs to pay for service line replacement (SLR),
represents a transfer of the payment burden from the community or individuals in the community to
the federal taxpayer at large. Also, to the extent systems utilize the other $11.7 billion in BIL funds or
other DWSRF base appropriation funds to conduct both system and private side SLR, the payment
burden of these LCRI activities will transfer from the implementing community to the federal taxpayer.
The use of funds from other federal programs like the EPA's Water Infrastructure Improvements for the
Nation Act of 2016 (WIIN Act) grant programs, the American Rescue Plan, Community Development
Block Grant programs through the U.S. Department of Housing and Urban Development, Rural
Development through the U.S. Department of Agriculture, and the Public Works Program through the
U.S. Department of Commerce Economic Development Administration for LCRI implementation will

Final LCRI Economic Analysis

4-3

October 2024


-------
result in the same type of transfers. But apart from information on the American Rescue Plan's
budgeted $519 million to remediate lead in drinking water, as of April 2024 (USDT, 2024), the EPA does
not have estimates on the amount of funds that will be used under these programs. Inter-community
transfers my also occur at the State or regional level. States may adopt programs to support lead and
GRR service line replacement (e.g., the State of Minnesota has approved $240 million for replacing LSLs,
mapping and inventory activities, and informing residents about the benefits of LSLR, and New York
State in the past has provided New York City with $30 million in LSLR funding) which would result in a
transfer from the State taxpayer base to LCRI implementing communities.

The implementation of the final LCRI also has the potential to result in intra-community transfers, which
is a shift in cost burden associated primarily with lead and GRR service line replacement from the private
side owner of the service lines to other community members, in this case the PWS which will likely seek
to recoup the cost by raising water rates on some or all of its customers. Note, that although the EPA
strongly encourages water systems to offer full-service line replacement at no cost to the customer;
SDWA does not provide authority for the agency to direct how a water system covers the costs of
compliance with a National Primary Drinking Water Regulation (NPDWR) and the EPA has not used its
section 1412 authority under SDWA to do so. This is a matter of State and local law, as the State and
local governments regulate how water systems provide and charge for services to their customers. At
the time of rule publication, the majority of State and local authorities have not made decisions on
customer/PWS cost sharing for service line replacement therefore the EPA has insufficient information
to estimate the actual SLR cost sharing relationship between customers and systems at the national
level of analysis. The potential size of the intra-community transfers are dependent on a number of
system specific SLR program criteria including the amount or fraction of the private side SLR the system
will pay for and any other income restrictions or other qualifiers associated with private side SLR
participants, the rate structure in the individual system and the degree to which the cost of private side
replacement will be passed through to the customers (e.g., low household income customers may not
receive a rate increase in favor of increasing rates on higher household income customers), and the
degree to which inter-community transfer funds are used to pay for private side replacements. After
accounting for BIL funding of $15 billion, and assuming no other Federal, State, or regional funding of
SLRs, the maximum incremental intra-community transfer under the LCRI associated with private side
SLR would be between $7.1 billion (low scenario) and $11.4 billion (high scenario), in 2022 dollars over
the 35-year period of analysis. This assumes 100 percent of private side costs are paid for by the PWS.

4.1.2 Overview of the Chapter

In Section 4.2, the EPA provides an overview of its approach to estimate the cost of the final LCRI. In
Section 4.3, the EPA provides the data and algorithms used to calculate the cost of each activity that
PWSs will undertake to comply with the final rule. In addition, Section 4.3 provides the EPA's estimates
of these costs. In Section 4.4, the EPA provides the data and algorithms used to calculate the cost of
each activity States will undertake to implement and administer the final LCRI, as well as the EPA's
estimates of these costs. While this chapter includes the EPA's national cost estimates for both the 2021
LCRR and the final LCRI, only details on the approach, data, and algorithms used to calculate the costs of
the final LCRI are provided in this chapter. The details on the approach, data, and algorithms used to
calculate the costs of the 2021 LCRR are provided in Appendix B.

Final LCRI Economic Analysis

4-4

October 2024


-------
An important compliance option for PWSs is to add additional phosphate for corrosion control. Some of
this phosphate may eventually enter wastewater treatment plants (WWTPs). In Section 4.5, the EPA
estimates the costs and impacts associated with increased phosphorous loadings.

4.2 Overview of the SafeWater LCR Model

In order to estimate the compliance costs (and benefits) of the LCRR in 2021, the EPA developed a new
version of its existing SafeWater CBX model.88 This new version, called the SafeWater LCR model, was
designed to estimate the costs and benefits of a treatment technique rule, and focus on water
contamination in the distribution system. The agency has updated the SafeWater LCR model to estimate
the compliance costs (and benefits) of the final LCRI.

4.2.1 Modeling PWS Variability in the SafeWater LCR Model

The SafeWater LCR model incorporates a large degree of variability across water system baseline
characteristics that influence compliance and costs. For example, under the final LCRI, PWSs will face
different compliance scenarios and costs depending on their size, primary source water type, number of
entry points to the distribution system, number of lead service lines (LSLs) and GRR service lines in their
distribution system, and existing corrosion controls in place. The SafeWater LCR model also includes
variability in compliance characteristics like different labor rates and number of tap and water quality
parameter (WQP) samples required by system size.

To reflect variability across PWS categories in modeling the final LCRI, the SafeWater LCR model applies
a "model PWS" approach. From a set of system baseline characteristic data including system type,
system size, and primary water source, the EPA defined 72 PWS categories, or strata, in the SafeWater
LCR model.

The 72 PWS categories consist of each combination of PWS type (2), PWS population size category (9)
PWS primary source water (2), and PWS ownership (2):

•	PWS Type:

o Community Water System
o Non-Transient Non-Community Water System

•	PWS Size Category (Population Served)

o

25 -100

o

101-500

o

501-1,000

o

1,000-3,300

o

3,301-10,000

o

10,001-50,000

o

50,001-100,000

o

100,001-1,000,000

o

Over 1,000,000

88 Information of the development of the SafeWater CBX model and its peer review can be found in Chapter 5,
Section 5.2.3 of the Economic Analysis for the Final Lead and Copper Rule Revisions (USEPA, 2020).

Final LCRI Economic Analysis

4-5

October 2024


-------
•	PWS Primary Source Water
o Surface Water

o Groundwater

•	PWS Ownership

o Publicly Owned
o Privately Owned

The SafeWater LCR model creates model PWSs representing systems in each category by combining the
PWS-specific data available in the Safe Drinking Water Information System/Federal version (SDWIS/Fed)
with data on baseline and compliance characteristics available at the PWS category level. When
categorical data are point estimates, every model PWS in a category is assigned the same value. When
the EPA has probabilistic data representing system variability, SafeWater LCR model assigns each model
PWS a value sampled from the distribution. Examples of the distributional data inputs that characterize
variability in the SafeWater LCR model include the burden for PWS and State staff to conduct tasks like
sampling and compliance documentation and review. For additional detail on the development of
model-PWSs in the SafeWater LCR model, see Appendix B, Section B.2.1. Because of this model PWS
approach, SafeWater LCR does not output any results at the PWS level, but rather, outputs cost (and
benefit) estimates at the PWS category, or strata. Each PWS category is defined by a set of system
characteristics including: the system type (community water system (CWS) and non-transient non-
community water system (NTNCWS)), primary water source (ground or surface), and size category (nine
categories). For each PWS category, the model calculates summary statistics that describe the costs (and
benefits) associated with final LCRI compliance. These summary statistics include total costs and
benefits, total costs per final regulatory requirement, total benefits per final regulatory requirement, the
variability in PWS-level costs (i.e., 10th, 25th, 75th, and 90th percentile system costs), and the variability in
household-level costs. For additional information on the data sources used in the estimation of costs see
Chapter 3 and Chapter 4, Sections 4.2.2, 4.3, 4.4, and 4.5. Also see Chapter 1, Exhibit 1-2 for the names
and descriptions of the reports and spreadsheet files that support the estimation of costs which are
available in the rulemaking docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

4.2.2 Modeling Uncertainty in the SafeWater LCR Model

This treatment technique rulemaking, and therefore the SafeWater LCR model, is complex,
incorporating multiple compliance triggers (e.g., AL exceedance (ALE), single sample exceedance,
multiple ALEs) that require multiple and varying compliance actions (CCT installation or re-optimization,
distribution system and site assessment, public education, temporary filter distribution) requiring a large
number of inputs for the estimation of total compliance costs. Many of these inputs, which are specific
to the assessment of the cost impacts of the final LCRI, are uncertain.

The EPA described in Chapter 3, Exhibit 3-78, the uncertainties in the baseline and compliance
characteristics of public water systems that impact the estimation of both costs and/or benefits of the
final LCRI. In addition to these baseline and system characteristics, there are additional uncertainties
associated with estimating the costs of the final rule. These are listed in Exhibit 4-2.

Final LCRI Economic Analysis

4-6

October 2024


-------
Exhibit 4-2: Summary of Uncertainties in the Estimation of Compliance Actions and Costs

Source of Uncertainty

Section
Discussed

Included in
High/Low
Scenarios

Potential
Direction of Bias

SLR unit costs

4.3.4

Yes

Unclear1

Baseline pH levels at PWSs with or without
existing CCT

4.3.3

No

Unclear2

Baseline Orthophosphate levels at PWSs with
existing CCT installed

4.3.3

No

Unclear2

CCT capital and O&M unit costs

4.3.3

Yes

Unclear2

POU unit costs

4.3.5

No

Unclear2

PWS administrative activity unit costs

4.3.1

No

Unclear3

Sampling unit costs

4.3.2

No

Unclear3

Public education unit costs

4.3.6

No

Unclear2

State administrative unit costs

4.4

No

Unclear3

Wastewater treatment plant phosphorus
treatment costs

4.5

No

Unclear2

Acronyms: CCT = corrosion control treatment; O&M = operations and maintenance; POU = point-of-use; PWS =
public water system; SLR = service line replacement.

Note:1 The EPA received public comments on the proposed rule unit cost for SLR indicating that the EPA's
estimated unit costs for SLR where to low biasing modeled total cost downward, however the data provided by
commenters was insufficient to allow EPA to re-estimate SLR unit cost or evaluate the directional bias claims. For
additional information on the EPA's estimated SLR unit costs and comparisons to commenter provided data see
Appendix A.

2The EPA did not receive sufficient specific data on the estimated unit cost or baseline characteristic that would
allow the EPA to either re-estimates the cost or characteristic for the final rule or discern a potential direction of
bias that may exist in the EPA values.

3The EPA received unit cost information through the proposed rule public comment process from the Association
of State Drinking Water Administrators that allowed the EPA to re-estimates administrative costs for States and
PWSs upward. The new cost information can be characterized as being more certain, but some degree of
uncertainty still exists, and the direction of bias is unclear. See Chapter 4, Sections 4.3 and 4.4for additional
information of the adjusted unit cost values.

The EPA determined it does not have enough information to perform a probabilistic uncertainty analysis
as part of the SafeWater LCR model analysis for this rule. Instead, to capture uncertainty, the EPA
estimated compliance costs (and benefits) using the SafeWater LCR model under low and high
bracketing scenarios that capture the three most significant cost drivers (the first is a PWS baseline
characteristic and the other two are compliance activity unit cost):

1. Likelihood a model PWS will exceed the AL and/or TL under the 2021 LCRR and the AL under the
final LCRI89

89 Exceedance of the AL and/or TL under the 2021 LCRR and the AL under the proposed LCRI will result in systems
making changes to CCT, implementing public education, and potentially using point-of-use (POU) filters. This drives
both costs and benefits in a consistent manner.

Final LCRI Economic Analysis

4-7

October 2024


-------
2.	SLR unit costs

3.	CCT capital and operations & maintenance (O&M) unit costs

Descriptions of these uncertain cost variables and the derivation of their values under the low and high
scenarios follows in Sections 4.2.2.1, 4.2.2.2, and 4.2.2.3.

The low and high benefits bracketing scenarios are further defined by the following benefits variables:

1.	Likelihood a model PWS will exceed the AL and/or TL under the 2021 LCRR and the AL under
the final LCRI (the same variable used to define the low and high scenarios in the cost
analysis).90

2.	The concentration-response functions that characterize how reductions in blood lead levels
(caused by changes in lead exposure) translate into avoided IQ reductions, cases of ADHD, and
cardiovascular disease premature mortality.

See Chapter 5 for additional information on the selection of the concentration-response functions for
the low and high scenarios in the benefits analysis.

4.2.2.1 Percent of Model PWSs that are Expected to Fall within Five Compliance Tap Sample 90th
Percentile Categories

As described in Chapter 3, Section 3.3.5.1, the likelihood that a model PWS would have an initial lead
90th percentile value (P90) greater than or equal to the AL is based on SDWIS/Fed historical 90th
percentile lead data from 2012 to 2020. The EPA recognizes that there are uncertainties in predicting
the future 90th percentile values from historical SDWIS data. Also, the agency recognizes that these
uncertainties could have a significant impact on estimated costs and benefits of the final LCRI.

Therefore, the EPA developed two sets of expected percentages for placement of model PWSs into one
of the five possible 90th percentile ranges (note that the subcategories of 90th percentile levels are used
to estimate costs associated with options presented in Chapter 8). Because the implementation of a
number of final rule requirements is driven by ALEs, the greater the estimated percent of systems above
those levels, the greater the total final rule costs. Therefore, the low cost scenario uses data derived
from the average 90th percentile value each PWS reported in SDWIS/Fed between 2012 to 2020. The
data used in the high cost scenario is derived by using the highest 90th percentile value each PWS
reported in SDWIS/Fed between 2012 to 2020. Exhibit 4-3 provides the likelihood that a model PWS,
with or without LSLs, would be assigned a 90th percentile in each of the 90th percentile-ranges by the
SafeWater LCR model under the 2021 LCRR in the low- and high-cost scenarios.

90 Note, the estimated P90 values used to determine AL and/or TL exceedances are not directly used in the model
to estimate lead exposure. See Chapter 5, Section 5.2 for detail on the tap water lead concentration sample data
used in the estimation of exposure changes.

Final LCRI Economic Analysis

4-8

October 2024


-------
Exhibit 4-3: Likelihood of Initial Model PWS 90th Percentile Placement under the 2021 LCRR

Category

No LSLs

Has LSLs

Low Estimate

^ 5 ng/L

88.5%

55.5%

>5 and <10 ng/L

7.1%

24.5%

10 ng/L < P90 < 12 ng/L

1.0%

5.3%

12 ng/L < P90 < 15 ng/L

1.0%

5.2%

P90 > 15 ng/L

2.3%

9.6%

High Estimate

^ 5 ng/L

79.6%

37.8%

>5 and <10 ng/L

11.7%

25.4%

10 ng/L < P90 < 12 ng/L

2.0%

6.8%

12 ng/L < P90 < 15 ng/L

1.9%

6.0%

P90 > 15 ng/L

4.8%

24.1%

Acronyms: LCRR = Lead and Copper Rule Revisions; LSL = lead service line; P90 = lead 90th percentile level; PWS =
public water system.

Note: Totals may not add due to independent rounding.

As discussed in Chapter 3, Section 3.3.5.1, under the final LCRI, the EPA estimated the percent of CWSs
that would be assigned to one of five bins using historical SDWIS/Fed 90th percentile tap sample and
applied the same following adjustments as was done for the 2021 LCRR. However, under the final LCRI,
the AL has been lowered to 10 ng/L as opposed to 15 ng/L. Additional updates include:

1.	An adjustment to reflect the new requirement for LSL systems to collect all samples from LSL
sites where possible, as opposed to the previous LCR minimum of 50 percent of samples being
collected from LSL sites.

2.	An adjustment to reflect new requirements for LSL systems to collect both first- and fifth-liter
samples from LSL sites instead of just first-liter samples (as required under the pre-2021 LCR) or
just fifth-liter samples (as required under the 2021 LCRR) and to use the higher of the first- and
fifth-liter sample in the 90th percentile calculation.91

Exhibit 4-4 provides the likelihood a model PWS, with or without LSLs, would be assigned a 90th
percentile in each of the 90th percentile ranges by the SafeWater LCR model under the final rule in the
low and high cost scenarios. Shaded (summary) rows are associated with the final rule requirement.

91 In addition to this requirement, under the final LCRI, water systems must use the highest sample values in their
90th percentile calculation. These high samples could be from non-lead service line sites. The second adjustment
does not explicitly model the impact of this rule requirement. The EPA uses a low and high adjustment to reflect
the majority of the uncertainty of the new rule requirements on 90th percentile tap results. However, the
possibility of using high non-lead service line samples in the calculation of the system's 90th percentile which is not
captured in EPA's adjustment to historical 90th percentile information may result in an underestimate of ALEs in
the analysis of the final rule.

Final LCRI Economic Analysis

4-9

October 2024


-------
Exhibit 4-4: Percent of CWSs by Lead 90th Percentile Classification under the Final LCRI

Category

No LSLs

Has LSLs

Low Estimate

No ALE (P90 <10 ng/L)

95.6%

79.0%

<5/xg/L

88.5%

54.4%

>5 and <10 ng/L

7.1%

24.6%

ALE (P90 >10 ng/L)

4.4%

21.0%

10 ng/L < P90 < 12 ng/L

1.0%

5.2%

12 ng/L < P90 < 15 ng/L

1.0%

5.6%

P90 > 15 ng/L

2.3%

10.3%

High Estimate

No ALE (P90 <10 ng/L)

91.3%

61.1%

<5/xg/L

79.6%

37.3%

>5 and <10 ng/L

11.7%

23.8%

ALE (P90 >10 ng/L)

8.7%

38.9%

10 ng/L < P90 < 12 ng/L

2.0%

7.8%

12 ng/L < P90 < 15 ng/L

1.9%

6.0%

P90 > 15 ng/L

4.8%

25.0%

Acronyms: ALE = action level exceedance; CWS = community water system; LCRI = Lead and Copper Rule
Improvements; LSL = lead service line; P90 = lead 90th percentile level.

Source: "Initial P90 Categorization_5 bins_LCRI_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

Note: Systems without LSLs can have lead sources that can contribute lead to drinking water such as premise
plumbing with lead solder and brass or chrome-plated brass faucets.

4.2.2.2 SLR Unit Costs

SLR cost estimates are based on the EPA's review of submitted and completed LSLR projects reported in
the 7th Drinking Water Infrastructure Needs Survey and Assessment (DWINSA) (USEPA, 2023a). For the
LCRI, the EPA reviewed SLR projects with independent documentation of their estimated costs and a
reported number of replaced service lines. Projects with unusually low-cost estimates (less than $700
per line), projects that included other non-lead SLR activities, and projects in which it was unclear
whether the replacement was partial or full were excluded from the dataset. The final dataset included
33 LSLR projects across 31 water systems in 13 States with populations serving from 3,000 to over
2,000,000. Low and high SLR cost estimates are based on the 25th and 75th percentile data from 33
DWINSA reported projects. The EPA recognizes uncertainty in SLR unit costs by having a low- and high-
cost estimate that are used in the low and high costing scenario, respectively. The detailed methodology
for identifying qualified projects and estimating the LSLR costs is provided in Appendix A, Section A.2.
Section A.3 of the appendix compares the SLR data from DWINSA to other data sources, including new
data sources provided since the proposed LCRI, and provides a discussion of geographic
representativeness.

Final LCRI Economic Analysis

4-10

October 2024


-------
The low and high estimates used in this economic analysis (EA) are presented in Exhibit 4-5. They
provide a range of national costs for the final LCRI that reflects the degree of uncertainty in the average
SLR unit costs. The EPA did not use the minimum and maximum values, from the 33 DWINSA reported
projects, for this bounding exercise given that the characteristics of the projects associated with the
minimum and maximum values are not representative of the majority of SLR projects nationally. The
EPA selected the interquartile range (25th to 75th percentile range) to represent uncertainty in the mean
SLR cost value because it is less sensitive to extreme values.92 Using minimum and maximum values
would have produced a national estimate range greater than what is warranted given the uncertainty in
the distribution of SLR unit costs.

Exhibit 4-5: Summary of SLR Costs from DWINSA Survey ($/SLR, 2020$)

Statistic

SLR Unit Costs

Full

Partial

Number of Cost Estimates

23

10

Min

$1,180

$1,677

25th percentile value

$6,507

$1,920

Median

$7,232

$3,273

Mean

$6,930

$3,803

75th percentile value

$8,519

$5,400

Max

$14,966

$8,099

Acronyms: DWINSA = Drinking Water Infrastructure Needs Survey and Assessment; SLR =
service line replacement.

Source: "LSLR Unit Costs_Final.xlsx," worksheet "DWINSA Data Analysis."

Notes: Data in this exhibit replicated data provided in Appendix A, Exhibit A-l.

The unit cost estimates used in the SafeWater LCR model do not include certain indirect and non-market
costs which occur during service line replacement such as traffic congestion costs, inconvenience to
homeowners and neighbors at SLR sites, potential short-term impact to the aesthetic appeal of the
property, and additional impacts to landscaping and cost of replacement beyond lawn repair, which is
included in the unit cost estimates above. See Chapter 6, Section 6.1.2for discussion of how EPA
considered these non-quantified costs in its decision making.

4.2.2.3 CCT Unit Costs

The EPA developed the cost estimates for CCT scenarios using outputs from the caustic feed and
phosphate feed Work Breakdown Structure (WBS) models (see Technologies and Costs for Corrosion

92 Note commenters on the proposed LCRI and industry experts support the use of values other than the maximum
for estimating national costs. CDM Smith (2022) stated that "The minimum and maximum costs for each item are
the extremes and should not be used for estimating cost except under special circumstances with specific criteria
for replacements." In their public comments on the proposed LCRI, Betanzo and Speight (2024) concluded that
"The findings of this analysis show that very high FLSLR costs are real but outliers occur in very limited
circumstances. The majority of FLSLR (full lead service line replacement) costs are substantially lower than the
maximum and reliably below $10,000."

Final LCRI Economic Analysis

4-11

October 2024


-------
Control to Reduce Lead in Drinking Water (USEPA, 2023b)). Outputs from these models are point
estimates of total capital and operation and maintenance (O&M) costs that correspond to a given set of
inputs that include treatment plant design flow (DF) and average flow in million gallons per day (MGD).
To estimate costs for CCT, the EPA fit cost curves to the WBS outputs for up to 49 different flow rates.
Specifically, for each scenario modeled and separately for total capital and for O&M costs, the EPA fit
three curves: one covering small systems (less than 1 MGD DF), one covering medium systems (1 MGD
to less than 10 MGD DF), and one covering large systems (10 MGD DF and greater).

For each CCT scenario modeled, the EPA also estimated separate equations for low, mid, and high costs
(see Technologies and Costs for Corrosion Control to Reduce Lead in Drinking Water (USEPA, 2023b)).
The EPA developed the low, mid, and high cost equations by varying the component level input in the
WBS models. This input drives the selection of materials for items of equipment that can be constructed
of different materials. For example, a low cost system might include fiberglass storage tanks and
polyvinyl chloride (PVC) piping. A high cost system might include stainless steel storage tanks and
stainless steel piping. The component level input also drives other WBS model assumptions that can
affect the total cost, such as building quality and heating and cooling. Because of uncertainty in the
component level materials selection a PWSs would choose for real world installation or re-optimization
of CCT technology, the EPA chose to use the low CCT cost equations in the SafeWater LCR model for the
low cost scenario and, for the high cost scenario, the SafeWater LCR model uses the high CCT cost
equations.

4.2.3 Model PWSs, Very Large Systems, Discounting and Cost of Capital, Compliance Schedule, and
Simulating Compliance Activities

4.2.3.1 Model Public Water Systems

As discussed above in Sections 4.2.1 and 4.2.2, under the regulatory provisions of the final LCRI, PWSs
will face different compliance scenarios depending on the size, the type of water system, the presence
of LSLs, and existing corrosion controls. In addition, PWSs will also face different unit costs based on
water system size, type, and number of entry points (e.g., labor rates and CCT capital, and O&M unit
costs). PWSs have a great deal of inherent variability across the water system characteristics that dictate
both compliance activities and cost.

Because of this variability, to accurately estimate the national level compliance costs (and benefits) of
the final LCRI, as well as describe how compliance costs are expected to vary across types of PWSs, the
SafeWater LCR model creates a sample of representative "model PWSs" by combining the PWS-specific
data available in SDWIS/Fed with data on baseline and compliance characteristics available at the PWS
category level. The SafeWater LCR model follows each model PWS in the sample through each year of
analysis - determining how the PWS will comply with each requirement of the final rule, estimating the
yearly compliance cost, and tracking the impact of the compliance actions on drinking water lead
concentrations. It also tracks how other events, such as changing a water source or treatment, affect the
water system's compliance requirements for the next year.

In constructing the initial model PWS sample for the cost-benefit analysis, the EPA began with the
49,529 CWSs and 17,418 NTNCWSs in SDWIS/Fed. Also, from SDWIS/Fed, the EPA knows each water
system's type (CWS or NTNCWS); primary water source (surface water or groundwater); population

Final LCRI Economic Analysis

4-12

October 2024


-------
served; CCT status (yes/no); ownership (public or private); and number of connections. Because some
PWS baseline characteristics are being assigned from distributional source data to capture the variability
across PWS characteristics, the EPA needed to ensure that its sample size was large enough that the
results of the cost-benefit model were stable for each of the 72 PWS categories. To ensure stability in
modeled results, the EPA oversampled the SDWIS/Fed inventory to increase the number of water
systems in each PWS category. For every PWS category, the EPA set the target minimum number of
model-PWSs to 5,000. To calculate the total estimated costs for each PWS category, the SafeWater LCR
model weights the estimated per water system costs so that, when summed, the total cost is
appropriate for the actual number of water systems known to be in the category. See Appendix B
Section B.2 for more detail.

With the exception of the three uncertain variables, which define the difference between the low and
high cost scenarios, the remaining baseline water system and compliance characteristics are assigned to
model PWSs, as described in Section 4.2.2 above and Appendix B, Section B.2, and remain constant
across the scenarios. This allows the EPA to define the uncertainty characterized in the cost range
provided by the low and high scenarios and maintain consistency between the estimation of costs for
the 2021 LCRR and final LCRI (e.g., number of systems with lead content service lines and percent of
connections that are lead content service lines).

4.2.3.2	Very Large Systems

The exception to the assignment of water system characteristics discussed in Sections 4.2.1, and
Appendix B.2.3 are the 24 very large water systems serving more than one million people. Because of
the small number of water systems in this size category, the uniqueness of their system characteristics,
and the potential large cost for these systems to comply with the final regulatory requirements, using
the methods described above to assign system attributes could result in substantial error in the
estimation of the national costs. Therefore, the EPA attempted to collect information on very large
water systems' CCT practices and chemical doses, pH measurements and pH adjustment practices,
number of lead and GRR service lines, service populations, and average annual flow rates for each entry
point to the distribution system. The EPA gathered this information from publicly available data such as
SDWIS/Fed facility-level data, Consumer Confidence Reports (CCR), and water system websites.93 In
addition, the American Water Works Association (AWWA) provided additional data from member water
systems to fill in gaps.94 When facility-specific data was available, the EPA used it to estimate compliance
costs for the very large water systems. If data were not available, the EPA assigned baseline
characteristics using the same process as previously described. See Appendix B, Section B.2.3 for a
summary of the data the EPA collected on these very large systems.

4.2.3.3	Discounting and Cost of Capital

The SafeWater LCR model estimates the incremental cost of the final LCRI over a 35-year period. In
accordance with the EPA's policy, and based on the current guidance from OMB, when calculating social
costs and benefits, the EPA discounted future costs (and benefits) at a 2 percent discount rate.

93	See "VLSEntryPointValues.xIsx" and "VLSSystemData.xIsx" for the information gathered on VLSs and used in
SafeWater LCR.

94	AWWA, personal communication, December 31, 2017, and March 5, 2018.

Final LCRI Economic Analysis

4-13

October 2024


-------
When evaluating the economic impacts on PWSs and households, the EPA uses the estimated PWS cost
of capital to discount future costs, as this best represents the actual costs of compliance that water
systems would incur over time. The EPA used data from the 2006 Community Water System Survey
(CWSS) to estimate the PWS cost of capital (USEPA, 2009). The EPA calculated the overall weighted
average cost of capital (across all funding sources and loan periods) for each size/ownership category,
weighted by the percentage of funding from each source. The cost of capital for each CWS size category
and ownership type is shown in Exhibit B-3 in Appendix B.2.4. Since similar cost of capital information is
not available for NTNCWSs, the EPA used the CWS cost of capital when calculating the annualized cost
per NTNCWS. Total estimated cost of capital may be greater than actual costs water systems bear when
complying with future regulatory revisions because financing support for lead reduction efforts may be
available from State and local governments, the EPA programs (e.g., the Bipartisan Infrastructure Act
and other federal funding administered through the DWSRF, the Water Infrastructure Finance and
Innovations Act (WIFIA) Program, and the WIIN Act grant programs), and other federal agencies (e.g.,
the United States Department of Housing and Urban Development's (HUD's) Community Development
Block Grants). Also see Section IV.G of the final LCRI Federal Register Notice (FRN) for a list of potential
funding sources. The availability of funds from government sources, while potentially reducing the cost
to individual PWSs, does not reduce the social cost of capital to society.

4.2.3.4	Schedule

The EPA projects that rule implementation activities will begin immediately after rule promulgation.
These activities will include one-time PWS and State costs for staff to read the rule, become familiar
with its provisions, and develop training materials and train employees on the new rule. States will also
incur burden hours associated with adopting the rule into State requirements, updating their LCR
program policies and practices, modifying data record keeping systems, conferring with systems on
initial planning for SLRs, reviewing inventory updates, and assisting and reviewing public education
material associated with service lines with lead or unknown content. PWSs will incur costs to comply
with the service line inventory requirements; develop an initial SLR plan if the system has one or more
known lead, GRR, or unknown service lines, and develop and distribute public education material
associated with service lines that are classified in the inventory as lead, GRR, or unknown material in
Years 1 through 3 of the analysis.95 The EPA expects that water systems will begin complying with all
other final rule requirements three years after promulgation, or in Year 4 of the analysis.

4.2.3.5	Simulating Compliance Activities

Some requirements of the final rule must be implemented by water systems regardless of their water
quality and tap sampling results (e.g., service line material inventory updates, mandatory SLR, and CWS
school and child care facility sampling programs). However, other activities are a function of a water
system's 90th percentile lead tap sample value.96 Because a water system's lead 90th percentile value is

95	For additional information on unit cost by system size for activities associated with developing and updating the
service line inventory, developing the initial SLR plan, and developing and distributing inventory-related outreach
material, see Sections 4.3.4.1, 4.3.4.2, and 4.3.6.2, respectively.

96	Distribution system and site assessment adjustments to CCT are required for a single lead tap sample
exceedances of the AL. This requirement was previously referred to as "find-and-fix" under the 2021 LCRR. The
provision of temporary pitcher filters are triggered by multiple ALE violations. Both of these compliance
requirements are also positively associated with system level 90th percentile tap sample values.

Final LCRI Economic Analysis

4-14

October 2024


-------
so important to determining regulatory requirements and cost under the final LCRI, the SafeWater LCR
model tracks each model PWS's 90th percentile value over each annual time step in the model. The 90th
percentile value, and if it exceeds the lead AL, dictates:

•	the tap water sampling and WQP monitoring schedules,

•	the installation/re-optimization of CCT,

•	the installation of point-of-use (POU) filters at water systems selecting this treatment option as
part of the small water system flexibilities of the final LCRI, and

•	public education and public notification requirements.

Under the final LCRI, the SafeWater LCR model assumes a PWS's 90th percentile tap sample values will
drop below the ALE once they (1) install or re-optimize CCT; (2) install POU or (3) removes all SLs with
lead content.97 When the PWS no longer has a 90th percentile tap sample value above the AL, it incurs
lower sampling, public education, and notification costs.

The SafeWater LCR model allows for future increases in 90th percentile values because of changes in
source water or treatment. The likelihood of these events occurring has been derived from SDWIS/Fed
data (see Chapter 3). When a change in source or treatment occurs in a modeled year, a new 90th
percentile value is assigned to the water system. This value may be higher or lower than the current
value thus potentially triggering new corrective actions. In the SafeWater LCR model, if a water system
already has "optimized" CCT or POU in place, it is assumed that no additional action is needed and that
the current treatment is adequate; therefore, the 90th percentile will not change.

4.3 Estimating Public Water System Costs

This section details how the EPA estimated the cost of water system compliance for each major rule
component of the final LCRI, including:

•	4.3.1: PWS Implementation and Administrative Costs

•	4.3.2: PWS Sampling Costs

•	4.3.3: PWS Corrosion Control Costs

•	4.3.4: PWS Service Line Inventory and Replacement Costs

•	4.3.5: PWS POU-Related Costs

•	4.3.6: PWS Lead Public Education, Outreach, and Notification Costs

Section 4.3.7 provides a summary of PWS costs including PWSs counts and population affected by each
major requirement, as well as costs by system and source water type and size category for low and high
cost scenarios using a 2 percent discount rate. In addition, the cost per household is also presented.

97 In Chapter 8, the EPA has analyzed the costs and benefits of an alternative option that includes an AL of 5 ng/L.
Under this AL, the EPA assumes that 10 percent of PWSs with service lines with lead content, and 3 percent of
PWSs with no SLs with lead content, will not be able to achieve the AL.

Final LCRI Economic Analysis

4-15

October 2024


-------
For most activities, water systems will incur costs in the form of burden (i.e., hours). The burden is
multiplied by the labor rate ($/hr), as presented in Chapter 3, Section 3.3.11.1, to estimate labor unit
costs. Systems will also incur capital and O&M costs for some activities. Exhibit 4-6 provides an overview
of the rule components, subcomponents, and activities for which the EPA estimates water system costs
for the final LCRI. The derivation of unit burden and/or cost is provided in each referenced subsection.

At the end of each subsection, the EPA provides a summary exhibit showing the SafeWater LCR
modeling approach for each water system activity (e.g., Exhibit 4-8, Exhibit 4-16). The exhibits are
organized as follows:

•	The first and second columns show how unit burden and labor rate information is combined to
estimate a CWS and NTNCWS cost per activity, respectively.

•	The third and fourth columns indicate the conditions under which the water system activity
occurs. The columns indicate if the system activity is dependent on:

o The system's 90th percentile range. See Appendix B, Section B.2 for a detailed discussion
of how the SafeWater LCR model tracks a water system's 90th percentile level and
accounts for changes in the 90th percentile level over the 35-year analysis period.

o Other characteristics of the system such as presence or absence of LSL/GRR service lines
and/or CCT, and frequency of monitoring.

•	The fifth column indicates the frequency of the activity (e.g., one-time, annually, every 3 years).

The SafeWater LCR model uses the information from these exhibits to calculate total annualized water
system cost for each activity. See Section 4.2 for detail on the cost modeling methodology.

As noted in Section 4.1, costs for model water systems presented in this section are LCRI costs if no
previous rule was in place. The national costs of the LCRI, or incremental costs, are the difference
between the cost of compliance with the LCRI and the cost of compliance with the 2021 LCRR. These
incremental national costs are presented in Section 4.1.98

For the purpose of the SafeWater LCR modeling, all cost model inputs are assigned a unique data
variable name, usually in the form of abbreviations, or shorthand, separated by underlined spaces (e.g.,
rate_op, hrs_read_rule_op). The SafeWater LCR model uses these data variables to model LCRI scenarios
for different system sizes and types.

Exhibit 4-6: PWS Cost Components, Subcomponents, and Activities Organized by Section1

Component

Subcomponents

Activities2

4.3.1: PWS
Implementation and
Administrative
Costs

4.3.1.1: PWS One-Time
Implementation and
Administrative Costs

a)	Read and understand the rule.

b)	Assign personnel and resources for rule
implementation.

c)	Participate in training and technical assistance provided
by the State during rule implementation.

98 Incremental national costs for the final LCRI using the pre-2021 LCR as the baseline for comparison are available
in Appendix C, Section C.2.

Final LCRI Economic Analysis

4-16

October 2024


-------
Component

Subcomponents

Activities2





d)

Provide small system flexibility option recommendation







to the State.



4.3.2.1: PWS Lead Tap

a)

Update sampling instructions for lead tap sampling and



Sampling



submit to the State.





b)

Contact homes to establish new 100 percent LSL tap







sampling pool.





c)

Update and submit tap sampling plan to the State.





d)

Report any changes in sampling locations to the State.





e)

Confer with the State on initial lead sampling data and







status under the LCRI.





f)

Obtain households for each round of lead tap sampling.





g)

Offer incentives to households to encourage







participation in lead tap sampling program.





h)

Ship tap sampling material and instructions to







participating households.





i)

Collect lead tap samples.





j)

Determine if a sample should be rejected and not







analyzed.





k)

Analyze lead tap samples in-house or commercially.

4.3.2: PWS Sampling



1)

Prepare and submit sample invalidation request to the

Costs





State.





m)

Inform consumers of tap sample results.





n)

Certify to the State that results were reported to







consumers.





o)

Submit request to renew 9-year monitoring waiver to







the State.





P)

Submit sampling results and 90th percentile calculation







to the State.





q)

Oversee the customer-initiated lead sampling program.





r)

Ship tap sampling material and instructions to







participating households for customer-initiated lead







sampling program.





s)

Collect lead tap samples for customer-initiated lead







sampling program.





t)

Analyze lead tap samples in-house or commercially for







customer-initiated lead sampling program.





u)

Inform customers of lead tap sample results for







customer-initiated lead sampling program.



4.3.2.2: PWS Lead Water

v)

Collect lead WQP samples from the distribution system.



Quality Parameter

w)

Analyze lead WQP samples from the distribution



Monitoring



system.





X)

Collect lead WQP samples from entry points.





y)

Analyze lead WQP samples from entry points.





z)

Report lead WQP sampling data and compliance with







OWQPs to the State.



4.3.2.3: PWS Copper

aa)

Collect copper WQP samples from the distribution

4.3.2: PWS Sampling

Water Quality Parameter



system.

Costs (Continued)

Monitoring

bb) Analyze copper WQP samples from the distribution







system.





cc)

Collect copper WQP samples from entry points.

Final LCRI Economic Analysis

4-17

October 2024


-------
Component

Subcomponents

Activities2





dd) Analyze copper WQP samples from entry points,
ee) Report copper WQP sampling data and compliance with
OWQPs to the State.

4.3.2.4: PWS Source
Water Monitoring

ff) Collect source water samples.

gg) Analyze source water samples.

hh) Report source water monitoring results to the State.

4.3.2.5.1: CWS School and
Child Care Facility Lead
Sampling Costs - First
Five-Year Cycle

ii) Create a list of schools and child care facilities served

by CWS and submit to State,
jj) Develop lead outreach materials for schools and child
care facilities.

kk) Prepare and distribute initial letters explaining the

sampling program and the EPA's 3Ts Toolkit.
II) Contact elementary school or child care facility to
determine and finalize its sampling schedule (one-
time) or contact secondary school to offer sampling
(annual).

mm) Contact school or child care facility to coordinate

sample collection logistics,
nn) Conduct walkthrough at school or child care facility

before the start of sampling,
oo) Travel to collect samples,
pp) Collect samples,
qq) Analyze samples.

rr) Provide sampling results to tested facilities,
ss) Discuss sampling results with the school or child care
facility.

tt) Conduct detailed discussion of high sampling results

with schools and child care facilities,
uu) Report school and child care facility sampling results
to the State.

vv) Prepare and provide annual report on school and
child care facility sampling program to the State.

4.3.2.5.2: CWS School and
Child Care Facility Lead
Sampling Costs - Second
Five-Year Cycle On

ww) Update the list of schools and child care facilities and

submit to the State,
xx) Contact schools and child care facilities to offer
sampling.

yy) Contact the school or child care facility to coordinate

sample collection logistics,
zz) Conduct walkthrough at school or child care facility

before the start of sampling,
aaa) Travel to collect samples,
bbb) Collect samples,
ccc) Analyze samples.

ddd) Provide sampling results to tested facilities,
eee) Discuss sampling results with the school and child
care facility.

Final LCRI Economic Analysis

4-18

October 2024


-------
Component

Subcomponents

Activities2





fff) Conduct detailed discussion of high sampling results

with schools and child care facilities,
ggg) Report school and child care facility sampling results
to the State.

hhh) Prepare and provide annual report on school and
child care facility sampling program to the State.

4.3.3: PWS
Corrosion Control
Costs

4.3.3.1: CCT Installation

a)	Conduct a CCT study.

b)	Install CCT (PO4, PO4 with post treatment, pH
adjustment, or modify pH).

4.3.3.2: Re-optimization
of Existing Corrosion
Control Treatment

c)	Revise CCT study.

d)	Re-optimize existing CCT.

4.3.3.3: DSSA Costs

e)	Contact customers and collect follow-up tap sample.

f)	Analyze follow-up lead tap sample.

g)	Collect distribution system WQP sample.

h)	Analyze distribution system WQP sample.

i)	Review incidents of systemwide events and other
system conditions.

j) Consult with the State prior to making CCT changes,
k) Report follow-up sample results and overall DSSA
responses to the State.

4.3.3.4: System Lead CCT
Routine Costs

1) Review CCT guidance.

m) Provide WQP data to the State and discuss during
sanitary survey.

n) Notify and consult with the State on required actions
in response to source water change.

0) Notify and consult with the State on required actions
in response to treatment change.

4.3.4: PWS Service
Line Inventory and
Replacement Costs

4.3.4.1: Service Line
Inventory

a)	Conduct records review for connector materials.

b)	Compile and submit connector updated LCRR initial
inventory information (baseline inventory) to the
State.

c)	Identify material for unknown service lines.

d)	Report annual inventory updates to the State.

e)	Conduct field investigations for inventory validation.

f)	Report validation results to the State.

4.3.4.2: Service Line
Replacement Plan

g)	Develop initial SLR plan and submit to the State for
review.

h)	Identify funding options for full SLRs.

i)	Include information on deferred deadline and
associated replacement rate in the SLR plan.

j) Update SLR plan annually or certify no changes.

k) Provide an updated recommendation of the deferred
deadline and associated replacement rate.

4.3.4.3: Physical Service
Line Replacements

1) Systems replace lead and GRR service lines.

4.3.4.4: Ancillary Service
Line Replacement
Activities

m) Contact customers and conduct site visits prior to
service line replacement.

Final LCRI Economic Analysis

4-19

October 2024


-------
Component

Subcomponents

Activities2





n) Deliver filters and 6 months of replacement cartridges

at time of service line replacement,
o) Collect tap sample post-service line replacement,
p) Analyze post-service line replacement tap sample,
q) Inform customers of tap sample result,
r) Submit annual report on service line replacement
program to the State.

4.3.5: PWS POU-
Related Costs (Small
System Compliance
Option)

4.3.5.1: POU Device
Installation and
Maintenance

a) Provide, monitor, and maintain POU devices.

4.3.5.2: POU Ancillary
Activities

b)	Develop POU plan and submit to the State.

c)	Develop public education materials and submit to the
State.

d)	Print POU education materials.

e)	Obtain households for POU monitoring .

f)	Deliver POU monitoring materials and instructions to
participating households.

g)	Collect tap samples after POU installation.

h)	Determine if sample should be rejected and not
analyzed.

i)	Analyze POU tap samples.

j) Prepare and submit sample invalidation request to the
State.

k) Inform customers of POU tap sample results.

1) Certify to the State that POU tap results were reported
to customers.

m) Prepare and submit annual report on POU program to
the State.

4.3.6: PWS Lead
Public Education,
Outreach, and
Notification Costs

4.3.6.1: Consumer Notice

a)	Develop lead consumer notice materials and submit to
the State for review.

b)	Provide a copy of the consumer notice and
certification to the State.

4.3.6.2: Activities
Regardless of Lead 90th
Percentile Level

c)	Update CCR language.

d)	Develop new customer outreach plan.

e)	Develop approach for improved public access to lead
health-related information and tap sample results.

f)	Establish a process for public access to information on
known or potential lead content SL locations and tap
sample results.

g)	Maintain a process for public access to lead health
information, known or potential lead content SL
locations, and tap sample results.

h)	Respond to customer request for known or potential
lead content SL information.

i)	Respond to requests from realtors, home inspectors,
and potential home buyers for known or potential lead
content SL information.

j) Develop a list of local and State health agencies.

k) Develop lead outreach materials for local and State
health agencies and submit to the State for review.

Final LCRI Economic Analysis

4-20

October 2024


-------
Component

Subcomponents

Activities2





1) Deliver lead outreach materials for local and State

health agencies,
m) Develop public education materials for known or

potential lead content SL disturbances and submit to
the State.

n) Deliver public education for SL disturbances.
0) Deliver filters and 6 months of replacement cartridges

during disturbances of service lines,
p) Develop inventory-related outreach materials and

submit to the State for review,
q) Distribute inventory-related outreach materials,
r) Provide translation services for public education
materials.

s) Certify to the State that lead outreach was
completed.3

4.3.6.3: Public Education
Activities in Response to
Lead ALE

t) Update mandatory language for lead ALE public

education and submit to the State for review,
u) Deliver lead ALE public education materials to all

customers,
v) Post notice to website,
w) Prepare press release.

x) Contact public health agencies to obtain additional

organizations and update recipient list,
y) Notify public health agencies and other organizations,
z) Consult with State on other public education activities,
aa) Implement other public education activities.

4.3.6.4: Public Education
Activities in Response to
Multiple Lead ALEs

bb) Develop plan for making filters available and submit to

the State for review,
cc) Develop outreach materials for systems with multiple

lead ALEs and submit to the State for review,
dd) Conduct enhanced public education for systems with

multiple lead ALEs.
ee) Consult with State on filter program for systems with

multiple lead ALEs.
ff) Administer filter program for systems with multiple
lead ALEs.

gg) Make filters available due to multiple lead ALEs.

Acronyms: ALE = action level exceedance; CCR = consumer confidence report; CCT = corrosion control treatment;
CWS = community water system; DSSA = Distribution System and Site Assessment; GRR = galvanized requiring
replacement; LSL = lead service line; LSLR = lead service line replacement; OCCT = optimal corrosion control
treatment; OWQPs = optimal water quality parameters; PO4 = orthophosphate; POU = point-of-use; PWS = public
water system; SL = service line; SLR = service line replacement; WQP = water quality parameter.

Notes:

1 Systems will also incur burden for recordkeeping activities under the LCRI, such as retaining records of decisions,
supporting documentation, technical basis for decisions, and documentation submitted by the system. The EPA
has included burden for recordkeeping with each activity when applicable and opposed to providing separate
burden estimates.

2The EPA assigned a unique letter identification (ID) for each activity under a given rule component. Activities are
generally organized with upfront, one-time activities first followed by ongoing activities.

3 This certification is inclusive of outreach activities in Sections 4.3.6.1 through 4.3.6.3.

Final LCRI Economic Analysis

4-21

October 2024


-------
4.3.1 PWS Implementation and Administrative Costs

PWSs will incur a one-time burden to implement the new requirements. These activities and associated
SafeWater LCR model cost inputs are described in Section 4.3.1.1. Section 4.3.1.2 provides the estimated
incremental annualized national PWS implementation and administrative costs for the LCRI at a 2
percent discount rate.

4.3.1.1 PWS One-Time Implementation and Administrative Costs

The EPA estimated that systems will incur a one-time burden to begin rule implementation. The EPA has
identified and developed costs for four activities as shown in Exhibit 4-7. The exhibit provides the unit
burden and/or cost estimate for each activity. The last column provides the data variable used in the
SafeWater LCR cost model. The assumptions used in the estimation of each activity follow the exhibit.
The EPA recognizes that systems would also incur administrative burden related to specific
requirements under the final LCRI. In these cases, the system burden is estimated under that particular
rule requirement.

Exhibit 4-7: PWS One-Time Administration Activities and Unit Burden Estimates

Activity

Unit Burden and/or
Cost
(hours/system)

SafeWater LCR Data Variable

a) Read and understand Rule

16 per PWS

hrs_read_rule_op

b) Assign personnel and resources for rule
implementation

8 per PWS

hrs_ assign_staff_imp_ op

c) Participate in training and technical
assistance provided by the State during
rule implementation

8 per PWS

hrs_ in iti al_ ta_ op

d) Provide small system flexibility option
recommendation to the State

12 hrs/CWSs serving
<3,300 and all
NTNCWSs

hrs_smJex_option_op

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; PWS = public

water system

Sources:

a), b): Based on implementation burden estimated for USEPA's 2012, Economic Analysis for the Final Revised Total
Coliform Rule (USEPA, 2012a). Available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

c):	Based on the EPA's 2015 Public Water System Supervision Program Information Collection Request (Renewal)
(USEPA, 2015a). Available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

d):	Association of State Drinking Water Administrators (ASDWA) 2024 Costs of States Transactions Study (CoSTS)
model, section "Small System Flexibility" (ASDWA, 2024).

Note: These data variables are also provided in "Administrative Burden and Costs_Final.xlsx."

a) Read and understand the rule (hrs_read_rule_op). Based on previous experience with rule

implementation, the EPA used the burden estimate of 4 hours from in the Economic Analysis for the
Final Revised Total Coliform Rule (USEPA, 2012a) as a starting point and revised the value upward to
account for the complexity of the regulatory requirements under the final LCRI. The EPA estimated
that systems would require a total of 16 hours to read and understand the rule revisions.

Final LCRI Economic Analysis

4-22

October 2024


-------
b)	Assign personnel and resources for rule implementation (hrs_assign_staff_imp_op). The EPA

assumed systems would require an additional 8 hours to assign appropriate personnel and
resources to carry out the new requirements under the final LCRI. This estimate is also consistent
with estimates used in the Economic Analysis for the Final Revised Total Coliform Rule (USEPA,
2012a).

c)	Participate in training and technical assistance provided by the State during rule implementation
(hrs_initial_ta_op). The EPA assumed systems would require an additional 8 hours to attend
training and receive other technical assistance from the State. This estimate is based on the data
from the EPA's 2015 Public Water System Supervision Program Information Collection Request (ICR)
(Renewal) (USEPA, 2015a).

d)	Provide small System flexibility option to the State (hrs_sm_flex_option_op). CWSs serving 3,300
or fewer people and all NTNCWSs that exceed the revised AL of 10 ng/L must submit a
recommended compliance option to their State to address lead. The EPA estimates each system will
require 12 hours to develop and submit this recommendation, which is twice the burden estimated
by the Association of State Drinking Water Administrators (ASDWA) in their 2024 Costs of States
Transactions Study (CoSTS) model, hereafter referred to as the ASDWA 2024 CoSTS model (ASDWA,
2024) for States to review this plan (data variable, hrs_sm_flex_option_js).99 See Section 4.4.1.1,
activity e) for a discussion of the corresponding State input.

Exhibit 4-8 provides the SafeWater LCR model cost estimation approach for system one-time PWS
administrative and rule implementation activities including additional cost inputs required to calculate
these costs.

Exhibit 4-8: PWS Administration and Rule Implementation Cost Estimation in SafeWater LCR

by Activity





Conditions for Cost to







Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

a) Read and understand the rule

The total hours per system multiplied
by the system labor rate.

Cost applies as
written to

All

All model PWSs

One time

(hrs_read_rule_op*rate_op)

NTNCWSs.







b) Assign personnel and resources for rule implementation

99 For the proposed LCRI EA, the EPA assumed a burden of 10 hours for systems to develop and submit a small
system flexibility option that was twice the burden needed for the States' review, based on ASDWA's 2020 CoSTS
model (ASDWA, 2020b). The 2020 model estimated the increase in costs to States to implement the final 2021
LCRR requirements and was provided to the agency as part of the public comment process on the 2021 LCRR
proposed rulemaking. The EPA subsequently revised its burden estimate for the final rule based on ASDWA's 2024
CoSTS model. ASDWA originally submitted the 2024 model as Appendix C to its public comments on the proposed
LCRI and made slight modifications to a version submitted to the EPA on April 18, 2024. The ASDWA 2020 and
2024 CoSTS models are available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Final LCRI Economic Analysis

4-23

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

The total hours per system multiplied
by the system labor rate.

(hrs_assign_staff_imp_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

All model PWSs

One time

c) Participate in training and technical assistance provided by the State during rule
implementation

The total hours per system multiplied
by the system labor rate.

(hrs_initial_ta_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

All model PWSs

One time

d) Provide small system flexibility lead compliance option to State

The total hours per system multiplied
by the system labor rate.

(hrs_sm_flex_option_op*rate_op)

Cost applies as
written to
NTNCWSs.

Above AL

CWSs serving <
3,300 people and
NTNCWSs

One time

Acronyms: AL = action level; CWS = community water system; NTNCWS = non-transient non-community water
system; PWS = public water system.

Note: The data variables in the exhibit are defined previously in Section 4.3.1.1 with the exception of:

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

4.3.1.2 Estimate of PWS National Implementation and Administrative Costs

As shown in Exhibit 4-1, the estimated monetized incremental annualized national PWS implementation
and administrative costs for the final LCRI range from $3.3 million, under the low scenario, to $3.2
million, under the high scenario, at a 2 percent discount rate in 2022 dollars.

4.3.2 PWS Sampling Costs

This section provides system unit burden and cost for lead tap sampling, lead WQP monitoring, copper
WQP monitoring, source water monitoring, and CWS sampling in schools and child care facilities in
Sections 4.3.2.1 through 4.3.2.5, respectively. Incremental national annualized sampling costs are
presented at a 2 percent discount rate in Section 4.3.2.6 in Exhibit 4-48.

4.3.2.1 PWS Lead Tap Sampling

The discussion of lead tap sampling costs for water systems is presented in three subsections as follows:

•	4.3.2.1.1: Lead Tap Sampling Schedules and Required Number of Samples

•	4.3.2.1.2: Lead Tap Sampling Activities

•	4.3.2.1.3: Lead Tap Sampling PWS Unit Cost Estimation Example

Final LCRI Economic Analysis

4-24

October 2024


-------
Exhibit 4-16 at the end of Section 4.3.2.1 is a summary exhibit that indicates how the cost inputs are
modeled by the SafeWater LCR model. Note that the SafeWater LCR model does not include the costs of
copper tap sampling. Because the final LCRI does not change the current regulatory requirements
associated with copper tap sampling the incremental cost associated with these provisions under the
final LCRI are equal to zero.

Activities and costs for tap monitoring associated with the POU program are not included in this section
but are provided in Section 4.3.5.

4.3.2.1.1 Lead Tap Sampling Schedules and Required Number of Samples

All CWSs and NTNCWSs are subject to lead tap sampling requirements. The frequency and required
number of samples depend on the systems' lead 90th percentile level. All systems with lead and GRR
service lines are assumed to conduct one year of semi-annual monitoring at the start of LCRI compliance
(assumed to be Year 4) with the exception of LSL systems in Michigan because they would have
monitored according to the LCRI sampling protocol (i.e., collect both a first- and fifth-liter lead sample)
prior to the rule's compliance date. As a simplifying approach, the EPA modeled all water systems in
Michigan as having all non-lead service lines.100 Only systems with a 90th percentile level at or below the
AL of 10 ng/L can qualify to conduct lead tap sampling annually at the standard number of sites or
triennially at the reduced number of sites. Some systems may be granted waivers by their State to
sample every 9 years, consistent with the LCR and 2021 LCRR. (Refer to Chapter 3, Section 3.3.7 for
additional detail regarding reduced monitoring schedules and criteria). Those systems with lead ALEs
must conduct lead tap sampling every six months at the standard number of sample sites (i.e., standard
semi-annual monitoring). In addition, systems must sample for a minimum of two, six-month tap
sampling monitoring periods following a change in source water or significant or long-term change in
treatment.

Because the number of required sampling sites and sampling schedules can vary, costs are estimated
separately for systems on the different lead tap sampling monitoring schedules. All systems with lead
and GRR service lines are assumed to conduct semi-annual monitoring in Year 4 to determine their 90th
percentile lead level. After Year 4, the EPA estimated the percentages of systems with a 90th percentile
level at or below 10 ng/L that would be on semi-annual monitoring,101 and on a reduced annual
(p_tap_annual), triennial (p_tap_triennial), or 9-year (p_tap_nine) monitoring schedule based on
historical SDWIS/Fed data. Chapter 3, Section 3.3.7 provides a detailed discussion of how these
percentages were derived. Exhibit 3-39 and Exhibit 3-40 provide the percentage of CWSs with 90th
percentile levels of < 10 ng/L with CCT and without CCT, respectively, on semi-annual monitoring and
annual monitoring at the standard number of sites and triennially or every nine years at the reduced
number of sites. Exhibit 3-41 and Exhibit 3-42 provide similar information for NTNCWSs with and
without CCT, respectively.

100	There is uncertainty in using this approach because Michigan did not require first- and fifth-liter samples for
systems with GRR service lines and having no LSLs. For these systems, the burden and cost for lead tap monitoring
may be underestimated.

101	The likelihood that a system without lead or GRR service lines and with a 90th percentile value at or below 10
Hg/L being on a semi-annual monitoring schedule is 1 minus (p_tap_annual + p_tap_triennial + p_tap_nine).

Final LCRI Economic Analysis

4-25

October 2024


-------
Exhibit 4-9 provides the minimum number of tap samples for CWSs and NTNCWSs on standard
monitoring and reduced monitoring schedules. These requirements have not been modified under the
final LCRI.

Exhibit 4-9: Minimum Number of Lead Tap Sampling Sites for Standard and Reduced

Monitoring



Standard Monitoring

Reduced Monitoring

System Size
(Population Served)

Minimum Number of Tap Samples



numb_samp_customer

numb_reduced_tap



A

B

<100

5

5

101-500

10

5

501-3,300

20

10

3,301-10,000

40

20

10,001-100,000

60

30

>100,000

100

50

Source: Lead and Copper Rule, 40 CFR 141.86(c).

Notes: The final LCRI did not modify the minimum required number of lead tap samples.

A: The required number of sites for CWSs and NTNCWSs on standard monitoring schedules.

B: The required number of sites for CWSs and NTNCWSs on reduced monitoring schedules. Under the final

LCRI, only systems with lead 90th percentile levels at or below 10 ng/L can qualify for reduced monitoring.

4.3.2.1.2 Lead Tap SamplinR Activities

The EPA has developed costs for system activities associated with lead tap sampling as shown in Exhibit
4-10. The exhibit provides the unit burden and/or cost for each activity. The assumptions used in the
estimation of each activity follows the exhibit. The last column provides the corresponding SafeWater
LCR model data variable in red/italic font. In a few instances, some of these activities are conducted by
the State instead of the water system. These activities are identified in the exhibit and further explained
in the exhibit notes. This section does not pertain to CWSs serving 3,300 or fewer people or NTNCWSs
that are using the POU provision and maintenance program as their lead compliance option. These
systems have some different lead tap sampling requirements that are discussed in Section 4.3.5.

Exhibit 4-10: PWS Lead Tap Sampling Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost1

SafeWater LCR Data Variable

a) Update sampling instruction for
lead tap sampling and submit to
the State (one-time)

2 hrs/CWS and NTNCWS

hrs_devel_samp_op2

b) Contact homes to establish new
100 percent LSLtap sampling
pool (one-time)

5 to 100 hrs/CWS with LSLs

hrs_add_lsl_samp_op

c) Update and submit tap sampling
plan to the State (one-time)

No LSLs: 2 to 6 hours per PWS
With LSLs: 8 to 20 hours per PWS

hrs_samp_plan_ op

Final LCRI Economic Analysis

4-26

October 2024


-------
Activity

Unit Burden and/or Cost1

SafeWaterLCR Data Variable

d) Report any changes in sampling
locations to the State

3 hrs/CWS

hrs_chng_tap_op

e) Confer with the State on initial
lead sampling data and status
under the LCRI (one-time)

2 hrs/PWS

hrs_initial_tap_confer_op

f) Obtain households for each
round of lead tap sampling

Burden per sample (CWSs onlv)
No LSLs: 0.5 hrs
With LSLs: 1 hrs

hrs_samp_ volun t_ op

g) Offer incentives to households
to encourage participation in
lead tap sampling program

$10 to $100/sample per CWS

costjncentive

h) Ship tap sampling material and
instructions to participating
households

Burden per sample (CWSs onlv)
0.25 hrs

Cost per sample (CWSs onlv)
No LSLs: $8.57 to $11.33
With LSLs: $8.96 to $23.21

Burden

hrs_discuss_samp_op
Cost

cost_5_lt_samp3

i) Collect lead tap samples

Burden per sample
0.40 to 0.71 hrs per CWS;
0.5 hrs per NTNCWS

Cost per sample
$5.75 to $10.24 per CWS

Burden

hrs_pickup_samp_ op
Cost

cost_pickup_samp

j) Determine if a sample should be
rejected and not analyzed

0.25 hrs/rejected sample for CWSs

hrs_samp_reject_op

k) Analyze lead tap samples in-
house or commercially

In-house Analysis (CWSs > 100K onlv)

In-house Analysis

Burden: 0.44 hrs/sample without
LSLs; 0.89 hrs/sample with LSLs

Cost: $3.92/sample without LSLs;
$7.84/sample with LSLs

Commercial Analysis (CWSs <100K
and all NTNCWSs)

$32.20/ sample without LSLs
$57.20/sample with LSLs

hrs_analyze_samp_op3

cost_lab_lt_samp3

Commercial Analysis
cost_ 5_ commerci al_ lab3

1) Prepare and submit sample
invalidation request to State

2 hrs per sample per CWS and
NTNCWS

hrs_samp_in valid_ op

m) Inform consumers of tap sample
results

CWS per sample
Burden: 0.05 to 0.11 hrs
Cost: $0.72

NTNCWS per sample
Burden: 1 hr
Cost: $0,079

CWS

hrs_inform_samp_ op
cost_cust_lt

NTNCWS

hrs_n tncws_inform_samp_ op
cost_ntncws_cust_lt

n) Certify to the State that results
were reported to consumers

0.66 to 1 hr per CWS or NTNCWS

hrs_cert_cust_lt_op

Final LCRI Economic Analysis

4-27

October 2024


-------
Activity

Unit Burden and/or Cost1

SafeWaterLCR Data Variable

o) Submit request to renew 9-year
monitoring waiver to the State

1 hr/9 years per qualifying CWS or
NTNCWS

hrs_renew_nine_op

p) Submit sampling results and 90th
calculation to the State

No LSLs: 2 to 3 hrs per CWS and
NTNCWS

With LSLs: 2.5 to 3.75 hrs per CWS
and NTNCWS

hrs_annual_lt_op3

q) Oversee the customer-initiated
lead sampling program

1 hr/sample per CWS

hrs_cust_request_oversee_op

r) Ship tap sampling material and
instructions to participating
households for customer-
initiated lead sampling program

Burden per sample (CWSs onlv)
0.25 hrs

Cost per sample (CWSs onlv)
No LSLs: $8.57 to $11.33
With LSLs: $8.96 to $23.21

Burden

hrs_discuss_samp_op
Cost

cost_5_lt_samp3

s) Collect lead tap samples for

customer-initiated lead sampling
program

Burden per sample (CWSs onlv)
0.40 to 0.71 hrs per CWS;

Cost per sample (CWSs onlv)
$5.75 to $10.24 per CWS

Burden

hrs_pickup_samp_ op
Cost

cost_pickup_samp

t) Analyze lead tap samples in-
house or commercially for
customer-initiated lead sampling
program

In-house Analysis (CWSs > 100K onlv)

In-house Analysis

Burden: 0.44 hrs/sample without
LSLs; 0.89 hrs/sample with LSLs

Cost: $3.92/sample without LSLs;
$7.84/sample with LSLs

Commercial Analysis (CWSs <100K
onlv)

$32.20/ sample without LSLs
$57.20/sample with LSLs

hrs_analyze_samp_op3

cost_lab_lt_samp3

Commercial Analysis
cost_ 5_ commerci al_ lab3

u) Inform customers of lead tap
sample results for customer-
initiated lead sampling program

CWS per sample
Burden: 0.05 to 0.11 hrs
Cost: $0.72

CWS

hrs_inform_samp_ op
cost_cust_lt

Acronyms: CWS = community water system; LSL = lead service line; NTNCWS = non-transient non-community
water system; PWS = public water system.

Source: "Lead Analytical Burden and Costs_Final.xlsx." See Section 4.3.2.1 for a summary of how the unit burden is

derived for each activity.

Notes:

1	All activities other than one-time activities are per monitoring period. In addition, many of the activities listed
above do not apply to NTNCWSs because unlike CWSs they collect their own samples from sampling locations
under their control and thus, are unlikely to change sampling sites or reject samples for analysis. They also do not
need to solicit sampling participation for customers or travel to their residences to pick up samples.

2	In Arkansas, Louisiana, Mississippi, Missouri, North Dakota, and South Carolina the State sends sampling
instructions to the water systems and thus are assumed to incur the burden to update the sampling instruction in
lieu of the system (ASDWA, 2020a).

3	In Arkansas, Louisiana, Mississippi, Missouri, and South Carolina the State pays for the cost of bottles, shipping,
analysis, and providing sample results to the system. Thus, the State will incur the burden and cost for these
activities in lieu of the system (ASDWA, 2020a).

Final LCRI Economic Analysis

4-28

October 2024


-------
a)	Update sampling instruction for lead tap sampling and submit to the State
(hrs_devel_samp_op). All CWSs and NTNCWSs will incur a one-time burden to update their
sampling instructions to be consistent with the revised tap sampling procedures related to the
prohibition of aerator removal and pre-stagnation flushing, and the requirement to use wide-
mouth bottles for sample collection. Systems are assumed to use an EPA template provided by
the State as the basis for updating their sampling instructions and would require 2 hours per
system. The EPA also assumed systems would submit their revised instructions electronically
and would not incur non-labor costs.

b)	Contact homes to establish new 100 percent LSL tap sampling pool (hrs_add_lsl_samp_op).

Under the LCRI, CWSs with LSLs incur a one-time burden to contact additional residents to have
enough volunteers to collect all samples from sites served by LSLs meeting their minimum
required number of tap samples. The estimated burden associated with this activity
(hrs_add_LSL_samp_op) is provided in Exhibit 4-11 below. The burden would only apply to those
systems with LSLs. See Chapter 3, Section 3.3.4.1 for the percentage of CWSs with LSLs (p_lsl).
The EPA assumed that CWS without LSLs will not need to update their initial sampling pool
because they are subject to less restrictive sampling criteria regarding the age of the copper and
lead solder sites under the LCRI. Specifically, these systems no longer need to prioritize sampling
at sites with copper pipes and lead solder installed after 1982. In addition, NTNCWSs generally
have control over their entire distribution system and are not expected to incur this additional
burden.

Exhibit 4-11: CWS Burden to Achieve a Sampling Pool with 100 Percent Lead Service Line Sites

System Size
(Population
Served)

Required number of
samples for standard
monitoring

Number of
new sites
needed for
systems with
LSLs

Total hours to
recruit one new
LSL sample
location

Total hours per system to

contact residences and
obtain required additional
LSL sample locations



n umb_samp_ customer





hrs_ add_ LSL_samp_ op



A

B = A*50%

C

D = B*C

<100

5

2.5

2

5

101-500

10

5

2

10

501-3,300

20

10

2

20

3,301-10,000

40

20

2

40

10,001-100,000

60

30

2

60

>100,000

100

50

2

100

Acronyms: CWS = community water system; LSL = lead service line.

Notes:

A: Exhibit 4-9, column A.

C: Based on a November 2, 2018 meeting with Philadelphia Water Department (PWD) regarding steps PWD takes
in response to a high lead level at an individual residence (available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov). Of the 263 people contacted at residences with potential LSLs sites, only 71 had LSLs. This is
approximately 25 percent of contacted customers. Based on this information, the EPA assumed that a water

Final LCRI Economic Analysis

4-29

October 2024


-------
system would need to contact four residences to obtain one new LSL site for their sampling program and would
require 0.5 hours per resident. This may be an overestimate because LSL systems will be updating their LSL
inventory to identify residences with LSLs. Thus, they may need to contact fewer residences to find those with LSLs
that are interested in participating in the sampling program.

c)	Update and submit tap sampling plan to the State (hrs_samp_plan_op). Systems must submit
tap sampling plans to the State prior to the initial monitoring period under the final rule. This is
a one-time burden. The EPA estimated systems with no LSLs will require 2 hours, 4 hours, and 6
hours for systems serving 3,300 or fewer people, systems serving 3,301-100,000 people, and
systems serving greater than 100,000 people, respectively. The EPA assumed systems with LSLs
will require more time to prepare their plans for submission to the State. The EPA estimated
systems with LSLs will require 8 hours, 16 hours, and 20 hours for systems serving 3,300 or
fewer, systems serving 3,301-100,000, and systems serving greater than 100,000, respectively.
These estimates are twice the State burden for reviewing sampling plans
(hrs_rev_samp_plan_js). As discussed in Section 4.4.2.1, activity b), these estimates are based
on the ASDWA 2020 CoSTS model, section "Tap Sampling" (ASDWA, 2020b). The EPA did not use
estimates from the ASDWA 2024 CoSTS model because estimates in the 2020 model were more
conservative.

d)	Report any changes in sampling locations to the State (hrs_chng_tap_op). Systems must report
any changes in their tap sampling locations from the prior monitoring period and the reason for
the change. Water systems must include the number of customers that are non-responsive after
two attempts or refused to participate. The EPA estimates CWSs will require 3 hours per
monitoring period to submit this documentation based on the 2022 Disinfectants/Disinfection
Byproducts, Chemicaland Radionuclides Rules ICR (Renewal), Exhibit 35 (Move Tap Sampling
Location) (USEPA, 2022a). The EPA assumed CWSs would have changes in monitoring locations
every monitoring period, starting in Year 5, due to customers dropping out of the testing
program. NTNCWSs are not assumed to incur this burden because in general they have control
over their entire distribution system and, unlike CWSs, should have access to all sampling
locations. Thus, the EPA assumed NTNCWS would be unlikely to change tap sampling locations.
Note that this assumption would underestimate burden in those instances in which a NTNCWS
had to change sampling sites (e.g., the site no longer meets the tiering criteria because the LSL
was removed). However, the EPA anticipates that once all LSLs are removed, a NTNCWS'
sampling plan would remain static.

e)	Confer with the State on initial lead sampling data and status under the LCRI
(hrs_initial_tap_confer_op). The EPA assumed systems will incur one-time burden in Year 4 to
discuss their requirements with the State based on their most recent two six-month monitoring
periods. The EPA assumes each system will incur a burden of 2 hours for this consultation. The
EPA assumed a 2 hour consultation burden is consistent with other types of consultations and is
based on the estimated burden for systems to consult with their State on public education
activities from pg. 60 of the Economic and Supporting Analyses: Short-Term Regulatory Changes
to the Lead and Copper Rule (USEPA, 2007). This estimate was increased from 1 hour per system
from the Economic Analysis for the Proposed Lead and Copper Rule Improvements (hereafter

Final LCRI Economic Analysis

4-30

October 2024


-------
referred to as the "Proposed LCRI EA" (USEPA, 2023c). The EPA changed this estimate to be
consistent with estimates for other types of consultation in the final rule.

f) Obtain households for each round of lead tap sampling (hrs_samp_volunt_op). For each
monitoring period, CWSs will contact customers from the tap sampling pool (see b above) to
obtain volunteers to participate in the lead tap sampling program. The EPA assumed:

•	CWSs will contact customers by phone.

•	CWSs will spend 20 minutes with those that agree to participate to explain the program, or
50 percent of customers, and 5 minutes with those that do not, for an average of 15
minutes or 0.25 hours per sample.

•	CWSs without LSLs will contact two customers for every one sample, resulting in an average
burden of 0.5 hours per sample.

•	CWSs with LSLs must contact additional customers because they must collect all samples
from LSL sites and previous sites will become ineligible if LSLs are replaced. The EPA
assumed these systems will contact four customers for every one sample, resulting in an
average burden of 1 hour.

An important input for this activity is the number of customers that are contacted each monitoring
period. The EPA started with the required number of samples (numb_samp_customer or
numb_reduced_tap from Section 4.3.2.1.1) and increased it to recognize that systems commonly
start with a larger sampling pool to account for situations where customers do not actually take the
sample, the sample is rejected for improper sampling protocol methods, or invalidated after it is
analyzed. For modeling purposes, the EPA inflated the starting number of customers in the sampling
pool using the following percentages:

•	1 - pp_hh_return_samp: The EPA assumed that 90 percent of volunteer customers would
collect their lead sample each monitoring period, with 1-90 percent, or 10 percent not
returning their sample bottles to be picked up by the water system. This likelihood is based
on New York City's Department of Environmental Protection response to an EPA 2016
questionnaire102 about their voluntary lead testing program in which they indicated that
customers returned the test kits 50 percent of the time. The EPA assumed a higher return
rate of 90 percent because CWSs will have contact with their customers prior to sample
collection as opposed to customer-initiated sampling that may be done via a website. This
likelihood does not apply to NTNCWSs.

•	pp_samp_reject: The EPA assumed CWSs would reject 5 percent of samples prior to sample
analysis based on the sample rejection rate provided by the City of Chicago Department of
Water Management (DWM) regarding their free customer-request testing program. The

102 The EPA sent a questionnaire in 2016 to the New York City Department of Environmental Protection and the
Chicago Department of Water Management (DWM) regarding their free testing programs for lead in drinking
water. The purpose of this questionnaire was to give the EPA a sense of the burden and cost associated with
implementing such a program. The questionnaire and responses are available in the docket (EPA-HQ-OW-2022-
0801 at www.regulations.gov).

Final LCRI Economic Analysis

4-31

October 2024


-------
DWM indicated they reject approximately 26 percent of test kits for improper sampling
procedures. The EPA assumed a lower rejection rate because customers will collect one
sample as opposed to the three sets of samples that are part of Chicago's sampling
protocol.103 In addition, some customers participate in multiple sampling events and should
be familiar with the sample collection protocol. Also refer to activity j) for the burden to
CWSs to determine if a sample should be rejected (hrs_samp_reject_op). The EPA assumed
NTNCWSs would not reject any samples because they collect their own samples and should
be familiar with the sampling protocol.

•	pp_samp_invalid: The EPA estimated that a small percentage (0.6 percent) of samples will
be invalidated by the State after the sample is analyzed. This estimate is based on the
average of Indiana and North Carolina's response to a 2016 ASDWA survey regarding the
number of invalidation requests per year. Indiana indicated they receive about 15
invalidations per year or 1.1 percent of their 1,375 CWSs and NTNCWSs. North Carolina
responded they have 1 to 2 requests per year. This translates to 0.08 percent using the
higher number of requests of 2 per the 2,375 CWSs and NTNCWSs in North Carolina. The
EPA used the average of the two percentages, approximately 0.6 percent. The EPA assumed
the same invalidation percentage for CWSs and NTNCWSs across all system sizes. Refer to
activity I) for the burden to systems to prepare and submit a sample invalidation request to
the State (hrs_samp_invalid_op). A copy of the questionnaire and each State's responses
are available in the docket under EPA-HQ-OW-2022-0801 at www.regulations.gov.

g) Offer incentives to households to encourage participation in lead tap sampling program
(cost_incentive). Some CWSs offer monetary incentives to their customers to encourage their
participation in their lead tap sampling program. Other systems elect not to or are prohibited
from providing financial incentives. The EPA considered the following information provided by
the Greater Cincinnati Water Works (GCWW) for 12 water systems in developing the likelihood
that a system would offer an incentive and the amount of that incentive:

•	Three systems (25 percent) offered no incentives.

•	Nine systems offered incentives ranging from $10 to $100. Most (four) offered $25. Two
systems offered $10, one system each offered $20, $50, and $100.

Based on this information, the EPA:

•	Assumed 75 percent of systems would offer incentives during each monitoring period in
order to obtain customer participation (p_incentive).

•	Set a minimum and maximum value by size category due to the variability across the 12
systems (cost_incentive). The EPA assumed systems serving 3,300 or fewer people would
not have the financial resources to offer large incentives and thus, set a minimum and
maximum of $10 and $20, respectively. The EPA assumed systems serving more than 3,300
would offer a minimum and maximum of $25 and $100, respectively.

103 Chicago DWM's free testing program includes three bottles and instructions for an initial first-draw, a 3 minute
flush, and a 5 minute flush.

Final LCRI Economic Analysis

4-32

October 2024


-------
The EPA assumed that incentives are only provided to customers that collect a sample that is not
later rejected or invalidated.

h) Ship tap sampling materials and instructions (hrs_discuss_samp_op, cost_5_lt_samp). The rule
allows customers to collect tap samples after receiving proper instructions from the water
system. The EPA assumed each CWS will spend an average of 0.25 hours to discuss sampling
instructions with customers (hrs_discuss_samp_op). This estimate is based on information
provided by Chicago DWM regarding its water testing program. DWM responded that on
average staff required 0.25 hours to send out test kits. The EPA assumed this burden included
time to discuss sampling instructions with volunteers and prepare the sampling kit for shipment.

The EPA assumed CWSs will ship sampling materials to customers. Thus, CWSs will also incur non-
labor costs for a CWS to provide a test kit (including bottles and instructions) and ship the kits to
customers (cost_5_lt_samp). The inputs and assumptions for this cost are provided in Exhibit 4-12
for systems without LSLs and in Exhibit 4-13 for systems with LSLs.

Exhibit 4-12: Non-Labor Costs for CWS without LSLs to Provide Test Kits (per Sample)

System Size
(Population
Served)

Test Kit Cost

Shipping Cost to
customers

Total Non-Labor Costs to
Provide Test Kits

cost_5_lt_samp

A

B

C = A+B

<3,300

$1.27

$7.50

$8.77

3,301-100,000

$1.07

$7.50

$8.57

>100,000

$3.83

$7.50

$11.33

Notes:

A: The sample test kit includes a shipping container, bottle label, resealable plastic bag, directions and chain of
custody form, and an empty bottle. The cost for CWSs serving 3,301 to 100,000 people is lower because these
systems are assumed to buy a larger quantity of shipping containers and incur a lower per container cost.
CWSs serving more than 3,300 people are assumed to buy the shipping container in bulk. CWSs serving
100,000 or fewer are assumed to use a commercial laboratory and the cost of the bottle is included as part of
the analytical fee. See file, "Lead Analytical Burden and Costs_Final.xlsx," worksheet
"Tap_Collect_Analyze_CWS_LCRI."

B: The EPA estimated the sample kit to weigh 0.23 pounds. The 2020 USPS retail ground shipping costs for
Zones 1 or 2 for a package of one pound or less is $7.50. Postage costs are available at
https://pe.usps.eom/Archive/NHTML/DMMArchive20201018/Noticel23.htm#_c037%20 (Accessed
6.27.2022).

C: The cost of a test kit is $0.02 lower than the Proposed LCRI EA (USEPA, 2023c) cost estimate due to an
adjustment in the cost of ink.

Final LCRI Economic Analysis

4-33

October 2024


-------
Exhibit 4-13: Non-Labor Costs for CWS with LSLs to Provide Test Kits (per Sample)

System Size
(Population Served)

Test Kit Cost

Shipping Cost to
customers

Total Non-Labor
Costs to Provide Test
Kits

cost_5_lt_samp

A

B

C = A+B

<3,300

$1.73

$7.50

$9.23

3,301-100,000

$1.46

$7.50

$8.96

>100,000

$15.71

$7.50

$23.21

Notes:

A: The sample test kit includes a shipping container, bottle label, resealable plastic bag, directions and chain of
custody form, and five empty bottles. The cost for CWSs serving 3,301 to 100,000 people is lower because
these systems are assumed to buy a larger quantity of shipping containers and incur a lower per container
cost. CWSs serving more than 3,300 people are assumed to buy the shipping container in bulk. CWSs serving
100,000 or fewer are assumed to use a commercial laboratory and the cost of the bottle is included as part of
the analytical fee. The sample kit cost is higher for systems with LSLs because the EPA assumed CWSs would
need a larger shipping container to accommodate five versus one sample bottle. See file, "Lead Analytical
Burden and Costs_Final.xlsx," worksheet "Tap_Collect_Analyze_CWS_LCRI." The test kit cost for CWSs with
LSLs that serve more than 100,000 people includes the cost of four additional 1-liter bottles. For CWSs serving
100,000 or fewer with LSLs, the additional bottle costs are reflected in a higher commercial laboratory cost.
See activity k) below.

B: The EPA estimated the sample kit to weigh 0.23 pounds. The 2020 USPS retail ground shipping costs for
Zones 1 or 2 for a package of one pound or less is $7.50. Postage costs are available at
https://pe.usps.eom/Archive/NHTML/DMMArchive20201018/Noticel23.htm#_c037%20 (Accessed
6.27.2022).

C: The cost of a test kit is $0.01 lower than the Proposed LCRI EA (USEPA, 2023c) cost estimate due to a
formula adjustment.

These unit costs are combined with the total number of tap sampling locations to produce the total
cost in the SafeWater LCR model. To estimate the number of tap sampling locations, the EPA
inflated the number of required samples for CWSs (numb_samp_customer or numb_reduced tap
from Section 4.3.2.1.1) by the likelihood a customer would not collect the sample of 10 percent (1 -
pp_hh_return_samp), the likelihood that the sample would be rejected of 5 percent
(pp_samp_reject), and the likelihood that a sample would be invalidated of 0.6 percent
(pp_samp_invalid). See activity f) for a more detailed discussion of these likelihoods.

i) Collect lead tap samples (hrs_pickup_samp_op, cost_pickup_samp). The EPA assumed CWSs
will pick up filled sample bottles versus having the customer ship them back. The agency has
heard from a number of systems that picking up the sample bottles ensures a demonstrable
chain of custody for the sample and ensures no damage to the sample before being analyzed by
the laboratory. The system will incur burden and O&M costs to travel round-trip to pick-up a
sample from each customer who participated in the sampling event. The EPA calculated the
average driving distance for each of the nine system size categories used in the SafeWater LCR
model. For CWSs serving 100,000 or fewer people, the EPA calculated the total service area for
each active CWS in SDWIS/Fed with available zip code information from the 2006 CWSS and Zip

Final LCRI Economic Analysis

4-34

October 2024


-------
Code Tabulation Areas from United States Census Bureau's Geography program TIGER GIS data
(2020 release of 2010 decennial geographies). For CWSs serving more than 100,000 people, the
EPA determined the service area from the county information reported to SDWIS/Fed or the
city's area. The latter was used for those CWSs that have system names identifying the city
served (e.g., the CWS "San Diego, City Of").

The EPA summed the total service area for all systems in each of the nine system size categories.
The EPA assumed each service area could be approximated by a circular shape and estimated the
average driving distance as 2/3 the radius of the service area. Due to the limited availability of
service area zip code information available in the 2006 CWSS,104 the EPA used a weighted average
for all systems serving 100,000 or fewer people based on the representativeness of the sample of
systems with zip code information and the total number of systems within each size category. A
summary of this analysis is presented in Exhibit 4-14 and additional data can be found in the file,
"Estimated Driving Distances_Final.xlsx."

The EPA also assumed systems would travel an average speed of 25 miles per hour to pick up a lead
sample from participating customers that equals a burden of 0.4 to 0.71 (hrs_pickup_samp_op),
depending on the system size as shown in Column C of Exhibit 4-14 below. In addition, the EPA used
the Federal mileage reimbursement rate of $0,575 per mile to calculate an average cost of $5.75 to
$10.24 (cost_pickup_samp) per trip based on system size as shown in Column E of Exhibit 4-14.

Similar to previous activities, an important input is the number of locations at which systems collect
lead tap samples. The EPA started with the required number of samples for CWSs
(numb_samp_customer or numb_reduced tap in Section 4.3.2.1.1) and increased it by the
likelihoods a customer would not collect the sample of 10 percent (1 - pp_hh_return_samp), the
sample would be rejected of 5 percent (pp_samp_reject), and the sample would be invalidated of
0.6 percent (pp_samp_invalid).

Exhibit 4-14: Travel Burden and Cost for Lead Tap Sample Pickup

System Size
(Population Served)

Miles one
way

Time one
way (hrs)

Time Roundtrip (hrs)

2020 Mileage
Rate

2020 Vehicle Cost

hrs_pickup_samp_ op

cost_pickup_samp

A

B=A/25

C= B*2

D

E = A*2*D

<100,000

5.0

0.20

0.40

$0,575

$5.75

100,001 -1,000,000

6.4

0.26

0.51

$0,575

$7.36

>1,000,000

8.9

0.36

0.71

$0,575

$10.24

Source: See file "Estimated Driving Distances_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at

www.regulations.gov.

Notes:

A & B: Geographic extent of water systems from the 2006 Community Water Systems Survey, and Census Data.
See file "Estimated Driving Distances_Final.xlsx" for derivation of mileage. Assumed travel speed of 25 mph.

104 Between 1 and 13 percent of CWSs serving 100,000 or fewer were currently active systems with zip code
information from the 2006 CWSS. For systems serving 100,000 to 1 million people, 6 percent and 100 percent of
CWSs were included in the average driving distance estimates, respectively.

Final LCRI Economic Analysis

4-35

October 2024


-------
D: Vehicle O&M based on Federal reimbursement rate of $0,575 (2020 mileage rate).

NTNCWSs collect their own samples and are assumed to require 0.5 hours to collect a sample
(hrs_pickup_samp_op). This burden is based on the estimated source water sample collection
burden from the 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR
(Renewal) (Exhibit 15) (USEPA, 2022a). The EPA inflated the number of required samples for
NTNCWS by the likelihood a sample would be invalidated of 0.6 percent (pp_samp_invalid) to
account for the additional burden for a NTNCWS to collect another sample. See activity f) for a more
detailed discussion of this likelihood.

j) Determine if a sample should be rejected and not analyzed (hrs_samp_reject_op). CWSs will
determine if samples collected by customers meet the required sampling protocol and if any
should be rejected prior to analysis. For example, the sample volume may not be one-liter or a
review of the chain-of-custody information could indicate the customer did not follow other
proper sampling procedures. The EPA assumed systems would spend an average of 15 minutes
or 0.25 hours per rejected sample (hrs_samp_reject_op).

The unit burden is multiplied by the number of samples that the system receives from customers,
which is estimated as the required number of rejected samples (numb_samp_customer or
numb_reduced tap from Section 4.3.2.1.1) multiplied by the 5 percent likelihood that the sample
would be rejected (pp_samp_reject).

As discussed under activity f), the EPA assumed all NTNCWSs collect their own samples and should
be familiar with the sampling protocol and thus would not incur burden to determine if a sample
should be rejected.

k) Analyze lead tap samples in-house or commercially (hrs_analyze_samp_op, cost_lab_lt_samp,
cost_5_commercial_lab). Based on input from seven laboratories, the EPA assumed only CWSs
serving more than 100,000 people will have in-house capabilities to analyze lead. All NTNCWSs
and all other CWSs are assumed to use a commercial laboratory for lead analysis. Thus, the
likelihood that a model PWS will conduct lead analyses in-house (pp_lab_samp) is 1 for CWSs
serving more than 100,000 people and 0 for all other systems. Conversely, the assigned
likelihood that a system will use a commercial lab for lead, or pp_commercial_samp, is 0 for
CWSs serving more than 100,000 people and 1 for all other systems.

Based on estimates provided by three laboratories, the EPA assumed that CWSs serving 100,000
would incur an in-house lead analytical burden of 0.44 hours per sample (hrs_analyze_samp_op).
This burden includes sample preparation, sample analysis, quality control checks and data entry.
Refer to "Lead Analytical Burden and Costs_Final.xlsx," worksheet "In-House Burden_hrs" for
additional information. For samples collected by CWSs serving more than 100,000 people from a site
served by an LSL, both a first- and fifth-liter sample must be analyzed and the analytical burden
would be double or 0.89 hours per system. CWSs conducting in-house analyses would also incur
non-labor costs for analytical materials such as preservatives, calibration standards, and quality
assurance (QA) standards of $3.92 per sample from a non-lead service line site and $7.84 for a first -
and fifth-liter sample from an LSL site (cost_lab_lt_samp) based on quotes from three vendors. See

Final LCRI Economic Analysis

4-36

October 2024


-------
worksheet "ln_House_Consumables_Summary_$/' in the file, "Lead Analytical Burden and
Costs_Final.xlsx."

The EPA assumed CWSs serving 100,000 or fewer people and all NTNCWSs that collected lead
samples from sites served by non-lead service lines would incur an average cost of $23.50 for a lead
sample analysis conducted by a commercial laboratory, based on estimates from seven laboratories,
and a cost of $8.70 to ship the sample to the laboratory for a total per sample cost of $32.20
(cost_5_commercial_lab). See worksheet "Commercial Analytical_$" in the file "Lead Analytical
Burden and Costs_Final.xlsx" for additional information. The EPA increased this estimate for systems
with LSLs to account for the analysis and shipping of a first- and fifth-liter sample of $23.50*2 or
$47.00 plus a cost to ship two bottles to the laboratory at $10.20 for a total cost of $57.20 per
sample.

The unit costs are multiplied by the number of samples analyzed each monitoring period to produce
total costs. The EPA began with the required number of samples (numb_samp_customer or
numb_reduced tap from Section 4.3.2.1.1) and increased it by the 0.6 percent likelihood the sample
would be invalidated (pp_samp_invalid) to estimate the number of samples analyzed in-house or
commercially. Note that the number of samples analyzed does not include those rejected by the
water system because they are not analyzed.

I) Prepare and submit a sample invalidation request to State (hrs_samp_invalid_op). Some CWSs
and NTNCWSs will request that the State invalidate a lead tap sample. The EPA assumed that
systems will not require extensive time to prepare and submit their sample invalidation requests
because the rule provides the allowable criteria for sample invalidation. The EPA assumed
systems will incur a burden of 2 hours per request (hrs_samp_invalid_op) based on Indiana's
and North Carolina's responses to a questionnaire. A copy of the questionnaire and each State's
responses are available in the docket under EPA-HQ-OW-2022-0801 at www.regulations.gov.

The EPA estimated that 0.6 percent of samples will be invalidated for CWSs and NTNCWSs
(pp_samp_invalid), as previously discussed in activity f). As a simplifying assumption, the EPA
assumed the State will grant all sample invalidation requests. Thus, the likelihood a system will
request sample invalidation is equal to the likelihood that a sample will be invalidated.

m) Inform consumers of tap sample results (hrs_inform_samp_op, cost_cust_lt,

hrs_ntncws_inform_samp_op, cost_ntncws_cust_lt). CWSs must report individual lead and
copper sample results to consumers who participated in the tap monitoring program as well as
consumers who request sampling under the LCRI final (see activity q) for additional detail). The
EPA estimates that systems serving 3,300 or fewer people will require an average of 0.05 hours
per consumer (hrs_inform_samp_op), while systems serving greater than 3,300 people will
require an average of 0.11 hours per consumer. This estimate is based on the public education
burden for systems to notify occupants of monitoring results estimated as part of the 2022
Disinfectants/Disinfection Byproducts, Chemicaland Radionuclides Rules ICR (Renewal) (Exhibit
39) (USEPA, 2022a). This ICR assumed a burden of 1 hour per 20 letters for all systems serving
3,300 or fewer people and a burden of 1 hour per 9 letters for systems serving greater than
3,300 people. Systems are also assumed to mail these results at a cost in 2020 dollars of $0.72
(cost_cust_lt) that includes postage ($0.55), and paper, ink, and envelope costs based on three
vendors of $0,019, $0,092, and $0.06, respectively (see "General Cost Model lnputs_Final.xlsx").

Final LCRI Economic Analysis

4-37

October 2024


-------
NTNCWSs are also required to provide sampling results to the people they serve. For NTNCWSs, the
EPA assumed the systems will deliver materials via email to all customers and post in a public
location at a burden of 1 hour for all system sizes (hrs_ntncws_inform_samp_op). This estimate
includes 0.5 hours to develop and send the e-mails and an additional 0.5 hours to post public
education materials publicly. The EPA assumed NTNCWSs will incur paper and ink costs of $0,019
and $0.06, respectively, (cost_ntncws_cust_lt) to post the flyer.

n) Certify to State that results were reported to consumers (hrs_cert_cust_lt_op). The EPA

assumed CWSs and NTNCWSs serving 50,000 or fewer people will incur a burden of 0.66 hours
per monitoring period to prepare and submit a certification that consumers who participated in
the compliance sampling or requested samples, were notified of their sampling results. Those
serving more than 50,000 people will incur a burden of 1 hour for this activity. The burden
estimates of 0.33 hours and 0.5 hours are based on North Carolina and Indiana's response,
respectively, to a 2016 ASDWA questionnaire regarding the estimated burden to review these
certifications. The EPA assumed systems would require twice the burden to prepare these
certifications than would be required for the State to review them. The EPA used the higher
estimated burden from Indiana for systems serving more than 50,000 people because these
systems collect a larger number of samples than smaller systems and thus, would be certifying
that they reported results to more consumers. The EPA assumed systems will submit this
certification electronically and thus incur no paper or mailing costs.

o) Submit request to renew 9-year monitoring waiver to the State (hrs_renew_nine_op). CWSs
and NTNCWSs on 9-year monitoring waivers must submit documentation to the State every 9
years that demonstrates their system and their customers continue to have no lead- or copper-
containing plumbing materials. As discussed in Section 3.3.7.1, the EPA assumed only a subset of
systems serving 1,000 or fewer people would qualify for this waiver. The EPA assumed systems
will incur a burden of 1 hour for this request based on the 2022 Disinfectants/Disinfection
Byproducts, Chemicaland Radionuclides Rules ICR (Renewal), Exhibit 35 (Monitoring Waiver
Application) (USEPA, 2022a). See file, "Pb Schedules_CWS_Final.xlsx" for additional information
on how the EPA estimated the number of systems with 9-year monitoring waivers.

p) Submit sampling results and 90th percentile calculations to the State (hrs_annual_lt_op). The

EPA estimated the burden for CWSs and NTNCWSs to submit tap sampling results and their 90th
percentile calculations, and the number of customers that were non-responsive after two
attempts or refused to participate in the sampling program. The burden is provided in Exhibit
4-15 for systems with and without LSLs with more detailed assumptions provided in the exhibit
notes. These estimates were doubled from the proposed rule to mirror changes to State burden
(hrs_annual_ltJs) in Section 4.4.2.1 that are based on the ASDWA 2024 CoSTS model, section
"Tap Sampling" (ASDWA, 2024).

Final LCRI Economic Analysis

4-38

October 2024


-------
Exhibit 4-15: Burden to Submit Lead Tap Sampling Results and 90th Percentile Level

System Size
(Population Served)

Provide Lead Tap Sampling Results and 90th percentile
Calculation (hrs/system/monitoring period)

hrs_annual_lt_op

A

B=A*1.25

No LSL

LSL

<10,000

2

2. 5

10,001-100,000

2. 5

3.13

> 100,000

3

3.75

Acronyms: LSL = lead service line.

Notes:

A: Burden based on the 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR
(Renewal), Exhibit 35 (Tap Sample Calcs) (USEPA, 2022a) and ASDWA's 2024 CoSTS model, section 'Tap Sampling"
(ASDWA, 2024).

B: LSL systems must also provide documentation if they have an insufficient number of sites served by LSLs that
are needed to meet minimum sampling requirements. Thus, the EPA assumed an additional 25 percent burden for
LSL systems.

q) Oversee the customer-initiated lead sampling program (hrs_cust_request_oversee_op). Under
the final LCRI, CWSs that exceed the final lead AL of 10 ng/L must make the offer to sample the
tap water of any customer who requests it more prominent in their public education materials.
This offer must also be included as part of the targeted outreach to customers with lead, GRR, or
unknown service lines. The final LCRI does not require CWSs to bear the sampling costs but for
modeling purposes, the EPA assumed CWSs would pay for the sample collection and analysis.
The EPA assumed the likelihood of a customer requesting a lead tap sample
(p_customer_request) to be 1% based on the testing program of five water systems. See file,
"Customer Requested Sampling Percent_Final.xlsx" for additional detail. The EPA also assumed
systems would require 1 hour per sample administrative burden to ensure customer's requests
were properly fulfilled (hrs_cust_request_oversee_op). In addition, systems would incur the
same burden and costs for shipping sampling materials, sample collection, analysis, and
informing customer of results associated with a first liter sample from non-lead service line sites
and a first- and fifth-liter sample from LSL sites previously described under activities:

•	h) Ship tap sampling material and instructions to participating households
(hrs_discuss_samp_op, cost_5_lt_samp);

•	i) Collect lead tap samples (hrs_pickup_samp_op, cost_pickup_samp);

•	k) Analyze lead tap samples in-house or commercially (hrs_analyze_samp_op, cost_lab_lt_samp,
cost_5_commercial_lab); and

•	m) Inform consumers of lead tap sample results (hrs_inform_samp_op, cost_cust_lt,
cost_ n tn cws_ cust_ It).

Final LCRI Economic Analysis

4-39

October 2024


-------
Exhibit 4-16 shows the SafeWater LCR model cost estimation approach for system lead tap sampling
activities. As shown in the exhibit, the SafeWater LCR model relies upon additional inputs, such as
number of samples for lead tap sampling and the likelihood a system is below an AL, to compute the
cost per activity. For example, unit costs for activity I) to prepare and submit sample invalidation
requests to State is the product of the required number of samples, the likelihood of sample
invalidation, the burden to prepare and submit the sample invalidation request, and the PWS hourly
rate. A description of the data variables and section where they are described in more detail are
provided in footnote 1 to the exhibit.

r) Ship tap sampling materials and instructions for customer-initiated lead sampling program
(hrs_discuss_samp_op, cost_5_lt_samp). See activity h) for discussion of burden estimates.

s) Collect lead tap samples for customer-initiated lead sampling program (hrs_pickup_samp_op,
cost_pickup_samp). See activity i) for discussion of burden estimates.

t) Analyze lead tap samples in-house or commercially for customer-initiated lead sampling
program (hrs_analyze_samp_op, cost_lab_lt_samp, cost_5_commercial_lab). See activity k)
for discussion of burden estimates.

u) Inform customers of lead tap sample results for customer-initiated lead sampling program
(hrs_inform_samp_op, cost_cust_lt, hrs_ntncws_inform_samp_op, cost_ntncws_cust_lt). See

activity m) for discussion of burden estimates.

Exhibit 4-16: PWS Lead Tap Sampling Cost Estimation in SafeWater LCR by Activity1,2





Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions2

Frequency
of Activity

a) Update sampling instructions for lead tap sampling and submit to the State3

Total hours per system multiplied by
the system labor rate.

(hrs_devel_samp_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

All model PWSs

One time

b) Contact homes to establish new 100 percent LSL tap sampling pool

Total hours per system multiplied by
the system labor rate.

(hrs_add_lsl_samp_op*rate_op)

Cost does not
apply to
NTNCWSs.

All

Model PWSs with
service lines of lead
or unknown content

One time

c) Update and submit tap sampling plan to the State

Total hours per system multiplied by
the system labor rate.

(hrs_samp_plan_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

All model PWSs

One time

d) Report any changes in sampling locations to the State

4



Final LCR! Economic Analysis

4-40

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions2

Frequency
of Activity







Model PWS not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
p_tap_triennial +
p_tap_nine)

Twice per
year

Total system hours per monitoring
period multiplied by the system labor
rate.

(hrs_chng_tap_op*rate_op)

Cost does not
apply to
NTNCWSs.

At or
below AL

Model PWS on
annual tap sampling
and not doing POU
sampling

p_tap_annual

Once a year







Model PWS on
triennial reduced tap
sampling and not
doing POU sampling

Every 3
years







p_tap_triennial









Model PWS is on
nine-year reduced
tap sampling and not
doing POU sampling

Every 9
years







p_tap_nine







Above AL

All model PWSs not
doing POU sampling

Twice a
year

e) Confer with the State on initial lead monitoring data and status under the LCRI

Total system hours multiplied by the
system labor rate.

(hrs_initial_tap_confer_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

All model PWSs

One Time

f) Obtain households for each round of lead tap sampling





At or
below AL

Model PWS not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
p_tap_triennial +
p tap nine)

Twice per
year

The number of required samples per
system multiplied by the hours per
sample and the system labor rate. The
number of required samples is inflated
to include those unreturned,
invalidated, and rejected to ensure that

Cost does not
apply to
NTNCWSs.



Model PWS on
annual tap sampling
and not doing POU
sampling

p tap annual

Once a year

Final LCRI Economic Analysis

4-41

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions2

Frequency
of Activity

the cost reflects the additional burden
that must occur to meet the sampling
requirement.









(numb_samp_customer+(numb_samp_
customer*(1-

pp_hh_return_samp))+(numb_samp_c
ustomer*pp_samp_invalid)+(numb_sa
mp_customer*pp_samp_reject))*(hrs_s
amp_volunt_op*rate_op)













Above AL

All model PWSs not
doing POU sampling

Twice per
year

The number of required samples per
system multiplied by the hours per
sample and the system labor rate. The
number of required samples is inflated
to include those unreturned,
invalidated, and rejected to ensure that
the cost reflects the additional burden
that must occur to meet the sampling
requirement.





Model PWS on
triennial reduced tap
sampling and not
doing POU sampling

Every 3
years

(numb_reduced_tap+(numb_reduced_t
ap*(1-

pp_hh_return_samp))+(numb_reduced
_tap*pp_samp_invalid)+(numb_reduce
d_tap*pp_samp_reject))*(hrs_samp_vol
unt_op*rate_op)

Cost does not
apply to
NTNCWSs.

At or
below AL

p_tap_triennial









Model PWS is on
nine-year reduced
tap sampling and not
doing POU sampling

Every 9
years







p_tap_nine



Final LCRI Economic Analysis

4-42

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions2

Frequency
of Activity

g) Offer incentives to households to encourage participation in lead tap sampling program







Model PWS not on
reduced tap sampling
and not doing POU
sampling that offers
an incentive

1 - (p_tap_annuai +
pjtapjtrienniai +
p_tap_nine)] *
p incentive

Twice per
year

The number of required samples per
system multiplied by the cost of the
incentive. This number is not inflated by
the number of samples deemed invalid
or rejected because it is assumed that if
a sample is invalid or rejected the
system will return to the same customer
to resample. The EPA also assumes
that unreturned samples would not be
eligible for an incentive.

numb_samp_customer*cost_incentive

Cost does not
apply to
NTNCWSs.

At or
below AL

Model PWS on
annual tap sampling
and not doing POU
sampling that offers
an incentive

p_tap_annuai *
pjncentive

Once a year





Above AL

Model PWS not doing
POU sampling that
offers an incentive

pjncentive

Twice per
year

The number of required samples per
system multiplied by the cost of the
incentive. This number is not inflated by
the number of samples deemed invalid
or rejected, because it is assumed that
if a sample is invalid or rejected the
system will return to the same customer
to resample. The EPA also assumes
that unreturned samples would not be
eligible for an incentive.

Cost does not
apply to
NTNCWSs.

At or
below AL

Model PWS on
triennial reduced tap
sampling and not
doing POU sampling
that offers an
incentive
pjtapjtrienniai *
pjncentive

Every 3
years

numb reduced tap*cost incentive















Model PWS on nine-
year reduced tap
sampling and not
doing POU sampling
that offers an
incentive

Every 9
years







p_tap_nine *
pjncentive



Final LCRI Economic Analysis

4-43

October 2024


-------
CWS Cost Per Activity

NTNCWS Cost
Per Activity

Conditions for Cost to
Apply to a Model PWS

Lead 90th
- Range

Other Conditions2

Frequency
of Activity

h) Ship tap sample monitoring materials and instructions to participating households5

Number of required samples multiplied
by the total of the hours per sample to
provide instructions times the system
labor rate, plus the cost of materials per
sample. The number of required
samples is inflated to include those
unreturned, invalidated, and rejected, to
ensure that the cost reflects the
additional burden that must occur to
meet the sampling requirement.

(numb_samp_customer+(numb_samp_
customer*(1-

pp_hh_return_samp))+(numb_samp_c

ustomer*pp_samp_invalid)+(numb_sa

mp_customer*pp_samp_reject))*((hrs_

discuss_samp_op*rate_op)+cost_5_lt_

samp)

To calculate the
sampling material
costs for
NTNCWSs this
equation is still
used. Number of
required samples
multiplied by the
cost of materials
per sample. The
number of
required samples
is inflated to
include those
invalidated to
ensure that the
cost reflects the
additional burden
that must occur to
meet the sampling
requirement.

((numb_samp_cu
stomer+(numb_sa
mp_customer*pp_
sampjn valid))*co
st_5_lt_samp)

At or
below AL

Model PWS not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
p_tap_triennial +
p_tap_nine)

Twice per
year

Model PWS on
annual tap sampling
and not doing POU
sampling

p_tap_annual

Once a year

Above AL

All model PWSs not
doing POU sampling

Twice per
year

Number of required samples multiplied
by the total of the hours per sample to
provide instructions times the system
labor rate, plus the cost of materials per
sample. The number of required
samples is inflated to include those
unreturned, invalidated, and rejected, to
ensure that the cost reflects the
additional burden that must occur to
meet the sampling requirement.

(numb reduced tap+(numb reduced t
ap*(1-

To calculate the
sampling material
costs for
NTNCWSs this
equation is still
used. Number of
required samples
multiplied by the
cost of materials
per sample. The
number of
required samples
is inflated to

At or
below AL

Model PWS on
triennial reduced tap
sampling and not
doing POU sampling

p_tap_triennial

Every 3
years

Final LCRI Economic Analysis

4-44

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions2

Frequency
of Activity

pp_hh_return_samp))+(numb_reduced
_tap*pp_samp_invalid)+(numb_reduce
d_tap*pp_samp_reject))*((hrs_discuss_
samp_op*rate_op)+cost_5_lt_samp)

include those
invalidated to
ensure that the
cost reflects the
additional burden
that must occur to
meet the sampling
requirement.

((numb_reduced_t
apr+(numb_reduc
ed_tap*pp_samp_
in vaii d))*cost_5_lt
samp)













Model PWS on nine-
year reduced tap
sampling and not
doing POU sampling

Every 9
years







p_tap_nine



i) Collect lead tap samples







Model PWS not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
p_tap_triennial +
p tap nine)

Twice per
year

The number of required samples per
system multiplied by the hours per
sample and the system labor rate. The
number of required samples is inflated
to include those invalidated and
rejected to ensure that the cost reflects
the additional burden that must occur to
meet the sampling requirement.

(numb_samp_customer+(numb_samp_
customer*pp_samp_invalid)+(numb_sa
mp_customer*pp_samp_reject)+
(numb_samp_customer*(1-
pp_hh_return_samp))*( (hrs_pickup_sa
mp_op*rate_op)+cost_pickup_samp)

Cost applies as
written to
NTNCWSs.

At or
below AL

Model PWS on
annual tap sampling
and not doing POU
sampling

p_tap_annuai

Once a year





Above AL

All model PWSs not
doing POU sampling

Twice per
year

Final LCRI Economic Analysis

4-45

October 2024


-------
CWS Cost Per Activity

NTNCWS Cost
Per Activity

Conditions for Cost to
Apply to a Model PWS

Lead 90th
- Range

Other Conditions2

The number of required samples
multiplied by the total of the hours per
sample to provide instructions times the
system labor rate, plus the cost of
materials per sample. The number of
required samples is inflated to include
those unreturned, invalidated, and
rejected, to ensure that the cost reflects
the additional burden that must occur to
meet the sampling requirement.

(numb_reduced_tap+(numb_reduced_t

apr*pp_samp_invalid)+(numb_reduced

_tap*pp_samp_reject)+

(numb_reduced_tap*(1-

pp_hh_return_samp))*((hrs_pickup_sa

mp_op*rate_op)+cost_pickup_samp)

Cost applies as
written to
NTNCWSs.

At or
below AL

Model PWS on



triennial reduced tap



sampling and not

Every 3

doing POU sampling

years

p_tap_triennial



Model PWS on nine-



year reduced tap



sampling and not

Every 9

doing POU sampling

years

p_tap_nine



j) Determine if a sample should be rejected and not analyzed







Model PWS not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
p_tap_triennial +
p tap nine)

Twice per
year

The number of samples expected to be
rejected (calculated by multiplying the
total number of required samples by the
likelihood of rejection) multiplied by the
hours per sample and the system labor
rate.

(numb_samp_customer*pp_samp_reje
ct)*(hrs_samp_reject_op*rate_op)

Cost does not
apply to
NTNCWSs.

At or
below AL

Model PWS on
annual tap sampling
and not doing POU
sampling

p_tap_annual

Once a year





Above AL

All model PWSs not
doing POU sampling

Twice per
year

Final LCRI Economic Analysis

4-46

October 2024


-------
CWS Cost Per Activity

NTNCWS Cost
Per Activity

The number of samples expected to be
rejected (calculated by multiplying the
total number of required samples by the
likelihood of rejection) multiplied by the
hours per sample and the system labor
rate.

(numb_reduced_tap*pp_samp_reject)*(
hrs_samp_reject_op*rate_op)

Cost does not
apply to
NTNCWSs.

Conditions for Cost to
Apply to a Model PWS



Lead 90th
- Range

Other Conditions2

Frequency
of Activity



Model PWS on
triennial reduced tap
sampling and not
doing POU sampling

Every 3
years

At or
below AL

p_tap_triennial





Model PWS on nine-
year reduced tap
sampling and not
doing POU sampling

Every 9
years



p tap nine



k) Analyze lead tap samples in-house or commercially5

The number of samples multiplied by
the probabilities for a sample analyzed
in house and a sample analyzed in a
commercial lab times the different labor
and material cost burdens for each type
of analysis.

The number of samples is inflated to
include those invalidated, to ensure that
the cost reflects the additional burden
that must occur to meet the sampling
requirement.

(((numb_samp_customer+(numb_samp

_customer*pp_samp_invalid))*pp_lab_

samp)*((hrs_analyze_samp_op*rate_o

p)+cost_lab_lt_samp))+(((numb_samp_

customer+(numb_samp_customer*pp_

samp_invalid))*pp_commercial_samp)*

((hrs_ar>alyze_samp_op*rate_op)+cost

_5_commercial_lab))

Cost applies as
written to
NTNCWSs.

At or
below AL

Model PWS not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
p_tap_triennial +
p tap nine)	

Model PWS on
annual tap sampling
and not doing POU
sampling

p_tap_annual

Twice per
year

Once a year

Final LCRI Economic Analysis

4-47

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions2

Frequency
of Activity





Above AL

All model PWSs not
doing POU sampling

Twice per
year

The number of samples multiplied by
the probabilities for a sample analyzed
in house and a sample analyzed in a
commercial lab times the different labor
and material cost burdens for each type
of analysis.

The number of samples is inflated to
include those invalidated, to ensure that
the cost reflects the additional burden
that must occur to meet the sampling
requirement.

Cost applies as
written to
NTNCWSs.

At or
below AL

Model PWS on
triennial reduced tap
sampling and not
doing POU sampling

p_tap_triennial

Every 3
years

(((numb_reduced_tap+(numb_reduced

_tap*pp_samp_invalid))*pp_lab_samp)*

((hrs_analyze_samp_op*rate_op)+cost

_lab_lt_samp))+(((numb_reduced_tap+

(numb_reduced_tap*pp_samp_invalid))

*pp_commercial_samp)*((hrs_analyze_

samp_op*rate_op)+cost_commercial_l

ab))















Model PWS on nine-
year reduced tap
sampling and not
doing POU sampling

Every 9
years







p_tap_nine



Final LCRI Economic Analysis

4-48

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions2

Frequency
of Activity

1) Prepare and submit sample invalidation request to the State





At or
below AL

Model PWS not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
p_tap_triennial +
p_tap_nine)

Twice per
year

The number of samples expected to be
invalid (calculated by multiplying the
total number of required samples by the
likelihood of invalidation) multiplied by
the hours per sample and the system
labor rate.

(numb_samp_customer*pp_samp_inval
id)*(hrs_samp_invalid_op*rate_op

Cost applies as
written to
NTNCWSs.



Model PWS on
annual tap sampling
and not doing POU
sampling

p_tap_annual

Once a year





Above AL

All model PWSs not
doing POU sampling

Twice per
year

The number of samples expected to be
invalid (calculated by multiplying the
total number of required samples by the
likelihood of invalidation) multiplied by
the hours per sample and the system
labor rate.

(numb_reduced_tap*pp_samp_invalid)*
(hrs_samp_invalid_op*rate_op)

Cost applies as
written to
NTNCWSs.

At or
below AL

Model PWS on
triennial reduced tap
sampling and not
doing POU sampling

p_tap_triennial

Every 3
years







Model PWS on nine-
year reduced tap
sampling and not
doing POU sampling

Every 9
years







p tap nine



Final LCRI Economic Analysis

4-49

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions2

Frequency
of Activity

m) Inform consumers of tap sample results

The number of required of samples per
system multiplied by the total of the
hours per sample times the system
labor rate plus the material cost per
sample.

numb_samp_customer*((hrs_inform_sa
mp_op*rate_op)+cost_cust_lt)

Hours per
sampling event
multiplied by the
system labor rate,
plus the material
cost per sampling
event.

((hrs_ntncws_infor
m_samp_op*rate_
op)+cost_ntncws_
custjt)

At or
below AL

Model PWS not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
p_tap_triennial +
p_tap_nine)

Twice per
year







Model PWS on
annual tap sampling
and not doing POU
sampling

Once a year







p tap annual







Above AL

All model PWSs not
doing POU sampling

Twice per
year

The number of required samples per
system multiplied by the total of the
hours per sample times the system
labor rate plus the material cost per
sample.

numb_reduced_tap*((hrs_inform_samp
_op*rate_op)+cost_cust_lt)

Hours per
sampling event
multiplied by the
system labor rate,
plus the material
cost per sampling
event.

((hrs_ntncws_infor
m_samp_op*rate_
op)+cost_ntncws_
custjt)

At or
below AL

Model PWS on
triennial reduced tap
sampling and not
doing POU sampling

p_tap_triennial

Every 3
years







Model PWS on nine-
year reduced tap
sampling and not
doing POU sampling

Every 9
years







p_tap_nine



n) Certify to State that results were reported to consumers







Model PWS not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
p_tap_triennial +
p_tap_nine)

Twice per
year

Final LCRI Economic Analysis

4-50

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions2

Frequency
of Activity

Total hours per sampling event
multiplied by the system labor rate.

(hrs_cert_cust_lt_op*rate_op)

Cost applies as
written to
NTNCWSs.

At or
below AL

Model PWS on
annual tap sampling
and not doing POU
sampling

p_tap_annual

Once a year







Model PWS on
triennial reduced tap
sampling and not
doing POU sampling

Every 3
years







p_tap_triennial









Model PWS on nine-
year reduced tap
sampling and not
doing POU sampling

Every 9
years







p_tap_nine







Above AL

All model PWSs not
doing POU sampling

Twice per
year

o) Submit request to renew 9-year monitoring waiver to the State6

Total hours per sampling event
multiplied by the system labor rate.

(hrs_renew_nine_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

Model PWS on nine-
year reduced tap
sampling and not
doing POU sampling

p_tap_nine

Every 9
years

p) Submit monitoring results and 90th percentile calculations to the State5







Model PWS not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
p_tap_triennial +
p_tap_nine)

Twice per
year

Total hours per sampling event
multiplied by the system labor rate.

(hrs_annual_lt_op*rate_op)

Cost applies as
written to
NTNCWSs.

At or
below AL

Model PWS on
annual tap sampling
and not doing POU
sampling

p_tap_annual

Once a year







Model PWS on
triennial reduced tap
sampling and not
doing POU sampling

Every 3
years







p_tap_triennial













Final LCRI Economic Analysis

4-51





October 2024


-------
CWS Cost Per Activity

NTNCWS Cost
Per Activity

Conditions for Cost to
Apply to a Model PWS

Lead 90th „ .... 2
„ Other Conditions2
- Range

Frequency
of Activity







Model PWS on nine-
year reduced tap
sampling and not
doing POU sampling

p_tap_nine

Every 9
years

Above AL

All model PWSs not
doing POU sampling

Twice a
year

q) Oversee the customer-initiated lead sampling program

Total hours per probability per
household a customer requests a
sample multiplied by the system labor
rate.

(hrs_cost_request_oversee_op*pp_cus
tomer_request*(pws_pop/numb_hh)*rat
e op)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a year

r) Ship tap sample monitoring materials and instructions to participating households for
customer-initiated lead sampling program5

Number of requested samples
multiplied by the total of the hours per
sample to provide instructions times the
system labor rate, plus the cost of
materials per sample.

pp_customer_request*(pws_pop/numb
_hh)*((hrs_discuss_samp_op*rate_op)
+cost 5 It samp)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a year

s) Collect lead tap samples for customer-initiated lead sampling program

The number of requested samples per
system multiplied by the hours per
sample and the system labor rate.

pp_customer_request* pws_pop/numb
_hh)*((hrs_pickup_samp_op*rate_op)+
cost pickup samp)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a year

t) Analyze lead tap samples in-house or commercially for customer-initiated lead sampling
program5

The number of requested samples
multiplied by the probabilities for a
sample analyzed in house and a
sample analyzed in a commercial lab
times the different labor and material
cost burdens for each type of analysis.

((pp_customer_request*(pws_pop/num
b_hh)*pp_lab_samp)*((hrs_analyze_sa
mp_op*rate_op)+cost_lab_lt_samp))+
((pp customer request*(pws pop/num

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a year

Final LCRI Economic Analysis

4-52

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions2

Frequency
of Activity

b_hh)*pp_commercial_samp)*((hrs_an
alyze_ _op*rate_op)+cost_5_com
mercial lab))









u) Inform customers of lead tap sample results for customer-initiated lead sampling
program

The number of requested of samples
per system multiplied by the total of the
hours per sample times the system
labor rate plus the material cost per
sample.

(pp_customer_request* pws_pop/numb
_hh))*((hrs_inform_samp_op*rate_op)+
cost cust It)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a year

Acronyms: AL = action level; CWS = community water system; LSL = lead service line; NTNCWS = non-transient

non-community water system; POU = point-of-use; PWS = public water system.

Notes:

1	The data variables in the exhibit are defined previously in this section with the exception of:

•	numb_hh: the average number of people per household, which equals 2.53 (Section 4.3.5.1).

•	numb_reduced tap: the number of lead tap samples for system on reduced annual, triennial, or 9-year
monitoring (Section 4.3.2.1.1).

•	numb_samp_customer: the number of lead tap samples for system on standard 6-month tap monitoring
(Section 4.3.2.1.1).

•	p_tap_annual, p_tap_triennial, and p_tap_nine: likelihood a systems is collecting the reduced number of
lead tap samples on an annual, triennial, or 9-year frequency, respectively (Section 4.3.2.1.1).

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

2	Does not apply to CWSs serving < 3,300 people and all NTNCWSs that have selected POU as their compliance
option if they exceeded the lead AL. See Section 4.3.5 for additional detail. PWSs with lead content or unknown
lines are identified using the data variables and approach described in Chapter 3, Section 3.3.4.

3	In Arkansas, Louisiana, Mississippi, Missouri, North Dakota, and South Carolina the State sends sampling
instructions to the water systems and thus are assumed to incur the burden to update the sampling instruction in
lieu of the system (ASDWA, 2020a).

4	For modeling purposes, the EPA assumed that systems would report changes in sampling location during each
monitoring period.

5The burden and costs to provide sample bottles (cost_5_lt_samp) under activity h), conduct analyses under
activity k), and provide sampling results under activity p) are incurred by the State in Arkansas, Louisiana,
Mississippi, Missouri, and South Carolina (ASDWA, 2020a).

6 Only systems with 90th percentile values < 5 ng/L can quality for a 9-year monitoring waiver.

4.3.2.1.3 Lead Tap SamplinR PWS Unit Cost Estimation Example

This section provides examples of the estimation of the Lead Tap Sample Monitoring unit cost
calculations for each activity a) through p) that are presented in Section 4.3.2.1.2 and Exhibit 4-16 and
follows the same lettering system. These activities represent the routine lead tap monitoring
requirements and do not include the customer-initiated sampling program requirements.

For this example, the EPA is using data that describe a surface water CWS with the following attributes:

Final LCRI Economic Analysis

4-53

October 2024


-------
•	serves a population of 10,001 to 50,000;

•	has LSLs in the distribution system;

•	has CCT in place;

•	is on a triennial Lead Tap Monitoring schedule;

•	has a 90th percentile at or below the AL; and

•	is not conducing POU monitoring.

Model PWSs within the SafeWater LCR model are assigned either a 0 (no) or 1 (yes) for a number of
system characteristics at the start of analysis, including LSL status. As described in Chapter 3, Section
3.3.4.1 (Exhibit 3-10), this model PWS has a 32.2 percent chance of having LSLs. As shown in Exhibit 4-4,
the likelihood of this model PWS having a 90th percentile initially at or below the AL under the final LCRI
is 79 percent under the low cost scenario and 61.1 percent under the high cost scenario. Given that the
model PWS has a 90th percentile at or below the AL, the model PWS has a 98 percent likelihood of being
on a triennial lead tap sample monitoring schedule (see Exhibit 3-39 in Chapter 3).

a)	Update Sampling Instructions and Submit to the State

The model PWS would begin by updating their sampling instructions to reflect the requirements in the
final LCRI. The estimation of this cost is represented by the following expression, which can be found in
the first row under the heading "Update sampling instructions for lead tap sampling and submit to the
State" in Exhibit 4-16:

Cost to update sampling instructions = hrs_devel_samp_op * rate_op
where:

•	hrs_devel_samp_op is the number of hours a system will require to update sampling
instructions (see Section 4.3.2.1.2, activity a)).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

Cost to update sampling instructions = (2 hrs * $42.68/hr) = $85.37

The model PWS will incur this $85.37 cost to update its sampling instructions in Year 4.

b)	Contact Homes to Establish a New 100 percent LSL Tap Sampling Pool

Next, the example system would contact homes to establish a new 100 percent LSL tap sampling pool.
The estimation of this cost is represented by the following expression:

Cost to contact homes = hrs_add_lsl_samp_op * rate_op (equation provided in Exhibit 4-16).
where:

•	hrs_add_lsl_samp_op is the number of hours the system will require to contact homes with LSLs
to achieve a 100 percent LSL sampling pool (see Exhibit 4-11).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

Final LCRI Economic Analysis

4-54

October 2024


-------
Cost to contact homes = (60 hrs * $42.68/hr) = $2,561.07.

The model PWS will incur this $2,561.07 cost to contact homes to establish a new sampling pool once
within the first four years after promulgation.

c)	Update and Submit Tap Sampling Plan

Next, the example system would update and submit their revised tap sampling plan to the State. The
estimation of this cost is represented by the following expression:

Cost to submit tap sampling plan to the State = hrs_samp_plan_op*rate_op

where:

•	hrs_samp_plan_op is the number of hours the system will require to update and submit their
tap sampling plan to the State (see Section 4.3.2.1.2, activity c)).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

Cost to update and submit tap sampling plan to the State: (16 hrs * $42.68/hr) = $682.88

The model PWS will incur this $682.88 cost to update and submit their sampling plan to the State one
time during Year 4.

d)	Report Changes in Sampling Location to the State

Systems then report to the State on any changes in sampling location for lead tap sampling. The
estimation of this cost is represented by the following expression:

Cost to report changes in sampling location = hrs_chng_tap_op * rate_op

where:

•	hrs_chng_tap_op is the number of hours the system will require to report a change in tap
locations to the State (see Section 4.3.2.1.2, activity d)).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

Cost to report changes in sampling location = (3 hrs * $42.68/hr) = $128.05

The model PWS will incur this $128.05 cost to report changes in sampling location once per sampling
period, or every three years for this example system on triennial monitoring.

e)	Confer with the State on Initial Lead Monitoring Data and Status under LCRI

Systems will confer with their States to discuss their requirements with the State based on their most
recent two six-month monitoring periods.

Cost to confer with the State = hrs_initial_tap_confer_op * rate_op
where:

Final LCRI Economic Analysis

4-55

October 2024


-------
•	hrs_initial_tap_confer_op is the number of hours the system will require to report a confer with
the State on their initial monitoring data and status under the 2021 LCRR (see Section 4.3.2.1.2,
activity e)).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

Cost to confer with the State = (2 hr * $42.68/hr) = $85.36

The model PWS will incur this $85.36 cost to confer with the State one-time during Year 4.

f)	Recruit Household Volunteers

Systems also recruit household volunteers for the Lead Tap Sample Monitoring program for each round
of sampling. The number of required samples is inflated to include those not collected, rejected, and
invalidated to ensure that the cost reflects the additional burden that must occur to meet the sampling
requirement. The estimation of this cost is represented by the following expression:

Cost to recruit household volunteers = [numb_reduced_tap + numb_reduced tap * (1 -
pp_hh_return_samp) + pp_samp_reject + pp_samp_invalid)] * hrs_samp_volunt_op * rate_op

where:

•	numb_reduced_tap is the number of reduced tap samples required per system (i.e., number of
customers from whom a system must obtain samples for systems on reduced Lead Tap Sample
Monitoring (see Exhibit 4-9)).

•	1 - pp_hh_return_samp is the likelihood that a volunteer household will not collect the sample
for Lead Tap Sample Monitoring (see Section 4.3.2.1.2, activity f)).

•	pp_samp_reject is the likelihood that a sample will be rejected by the system following lead tap
sample monitoring but prior to sample analysis (see Section 4.3.2.1.2, activity f)).

•	pp_samp_invalid is the likelihood that a lead sample will be deemed invalid by the State (see
Section 4.3.2.1.2, activity f)).

•	hrs_samp_volunt_op is the number of hours per customer to obtain volunteer customers for
Lead Tap Sample Monitoring samples (see Section 4.3.2.1.2, activity f)).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

Cost to recruit household volunteers = [30 samples + 30 samples * (1 - 0.9) + 0.05 + 0.006)] * (1 hr *
$42.68/hr) = $1,480.30.

The model PWS will incur this $1,480.30 cost to recruit household volunteers once per sampling period,
or in this example once every three years.

g)	Offer an Incentive to Households for Participation

Systems that offer an incentive, do so to encourage participation in the lead tap sampling program.
Seventy-five percent of systems are expected to offer an incentive. The number of households is
assumed to equal the number of required samples. This number is not inflated by the number of
samples rejected or deemed invalid because the EPA assumed that incentives are only provided to

Final LCRI Economic Analysis

4-56

October 2024


-------
customers that collect a sample that is not later rejected or invalidated. The estimation of this cost is
represented by the following expression:

Cost to offer incentives = numb_reduced_tap * cost_incentive

where:

•	numb_reduced_tap is the number of reduced tap samples required per system (i.e., number of
customers from whom a system must obtain tap samples) for systems on reduced lead tap
sample monitoring (see Exhibit 4-9).

•	cost_incentive is the cost per customer for an incentive to participate in the sampling program
(see Section 4.3.2.1.2, activity g)).

The variable cost_incentive ranges from $25 to $100 for the example PWS (see Section 4.3.2.1.2, activity
g)). In the case of this example, the EPA assumed a value of $50.

Cost to offer incentives = (30 samples * $50/sample) = $1,500

The model PWS will incur this $1,500 cost to offer incentives once per sampling period, or in this
example, once every three years.

h) Ship Sample Material and Instructions

Systems then ship the lead tap sampling sample materials and instructions to the participating
households. The number of required samples is inflated to include those not collected, rejected, and
invalidated to ensure that the cost reflects the additional burden that would occur to meet the sampling
requirement. The estimation of this cost is represented by the following expression:

Cost to deliver sample material and instructions = [numb_reduced_tap + numb_samp_customer * (1 -
pp_hh_return_samp) + pp_samp_reject + pp_samp_invalid)] * (hrs_discuss_samp_op * rate_op +
cost_5_lt_samp)

where:

•	numb_reduced_tap is the number of reduced tap samples required per system for systems {i.e.,
number of customers from whom a system must obtain tap samples) on reduced Lead Tap
Sample Monitoring (see Exhibit 4-9).

•	1 - pp_hh_return_samp is the likelihood that a volunteer household will not collect the sample
for Lead Tap Sample Monitoring (see Section 4.3.2.1.2, activity f)).

•	pp_samp_reject is the likelihood that a sample will be rejected following lead tap sample
monitoring (see Section 4.3.2.1.2, activity f)).

•	pp_samp_invalid is the likelihood that a lead sample will be deemed invalid (see Section
4.3.2.1.2, activity f)).

•	hrs_discuss_samp_op is the number of hours per volunteer household to discuss and deliver
Lead Tap Sample Monitoring sample instructions (see Section 4.3.2.1.2, activity h)).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

Final LCRI Economic Analysis

4-57

October 2024


-------
•	cost_5_lt_samp is the material cost excluding consumables for in-house analyses for Lead Tap
Sample Monitoring (i.e., test kit and shipping to customers, and cost to travel to pick-up bottles)
(see Exhibit 4-12).

Cost to deliver sample material and instructions = [30 samples + 30 samples * ((1 - 0.9) + 0.05 + 0.006)] *
((0.25 hrs * $42.68/hr) + $8.96/sample) = $680.81

The model PWS will incur this $680.81 cost to ship materials and instructions once per sampling period,
or in this example once every three years.

i) Collect Lead Tap Samples

Systems then pick up lead tap samples from the participating households. The number of required
samples is inflated to include those rejected and invalidated to reflect the additional burden that must
occur to meet the sampling requirement. The estimation of this cost is represented by the following
expression:

Cost to pick up lead tap samples = [numb_reduced_tap + numb_reduced_tap * (pp_samp_reject +
pp_samp_invalid)] * ((hrs_pickup_samp_op * rate_op) + cost_pickup_samp)

where:

•	numb_reduced_tap is the number of reduced tap samples required per system for systems (i.e.,
number of customers from whom a system must obtain tap samples) on reduced Lead Tap
Sample Monitoring (see Exhibit 4-9).

•	1 - pp_hh_return_samp is the estimated likelihood that a volunteer household will not collect
the sample for Lead Tap Sample Monitoring (see Section 4.3.2.1.2, activity f)).

•	pp_samp_reject is the likelihood that a sample will be rejected following Lead Tap Sample
Monitoring (see Section 4.3.2.1.2, activity f)).

•	pp_samp_invalid is the likelihood that a lead sample will be deemed invalid (see Section
4.3.2.1.2, activity f)).

•	hrs_pickup_samp_op is the number of hours per sample for PWS staff to travel to the
customer's residence to pick up lead tap sample from customer (see Section 4.3.2.1.2, activity

i)).

•	cost_pickup_samp is the travel cost per sample for PWS to travel to the customer's residence to
pick up lead tap sample from customer (see Section 4.3.2.1.2, activity i)).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

Cost to pick up lead tap samples = [30 samples + 30 samples * (0.05 + 0.006)] * ((0.4 hrs * $42.68/hr) +
$5.75/sample) = $723.06.

The model PWS will incur this $723.06 cost to pick up lead tap samples once per sampling period, or in
this example once every three years.

Final LCRI Economic Analysis

4-58

October 2024


-------
j) Determine if a Lead Tap Sample Should be Rejected

Systems must determine if a lead tap sample collected by a household should be rejected and not
analyzed. The number of required samples is inflated to include those rejected and invalidated. The
estimation of this cost is represented by the following expression:

Cost to determine if a lead tap sample should be rejected = [numb_reduced_tap + numb_reduced_tap *
(pp_samp_reject + pp_samp_invalid)] * hrs_samp_reject_op * rate_op

where:

•	numb_reduced_tap is the number of reduced tap samples required per system (i.e., number of
customers from whom a system must obtain tap samples) for systems on reduced Lead Tap
Sample Monitoring (see Exhibit 4-9).

•	pp_samp_invalid is the likelihood that a lead sample will be deemed invalid (see Section
4.3.2.1.2, activity f)).

•	pp_samp_reject is the odds that a sample will be rejected following Lead Tap Sample Monitoring
(see Section 4.3.2.1.2, activity f)).

•	hrs_samp_reject_op is the number of hours per rejected sample for PWS staff to decide to
reject sample (see Section 4.3.2.1.2, activity j)).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

Cost to determine if a lead tap sample should be rejected = [30 samples + 30 samples * (0.05 + 0.006)] *
0.25 hrs * $42.68/hr = $338.06

The model PWS will incur this $338.06 cost to determine if lead tap samples should be rejected once per
sampling period, or in this example once every three years.

k) Analyze Lead Tap Samples

Systems then analyze the lead tap samples, either in-house or in a commercial laboratory. Systems
serving populations of 10,001 to 50,000 are assumed to use commercial labs. The number of samples is
inflated to include those invalidated, to ensure that the cost reflects the additional burden that must
occur to meet the sampling requirement. The estimation of this cost is represented by the following
expression:

Cost to analyze lead tap samples = [numb_reduced_tap + numb_reduced_tap * pp_samp_invalid] *
[(pp_commercial_samp * (hrs_analyze_samp_op * rate_op + cost_lab_lt_samp)) +
(pp_commercial_samp * cost_5_commercial_lab)]

where:

•	numb_reduced_tap is the number of reduced tap samples required per system {i.e., number of
customers from whom a system must obtain tap samples) for systems on reduced Lead Tap
Sample Monitoring (see Exhibit 4-9).

Final LCRI Economic Analysis

4-59

October 2024


-------
•	pp_samp_invalid is the likelihood that a lead sample will be deemed invalid (see Section
4.3.2.1.2, activity f)).

•	pp_commercial_samp is the likelihood that a sample will be analyzed in a commercial lab (see
Section 4.3.2.1.2, activity k)).

•	hrs_analyze_samp_op is the number of hours per sample it takes to analyze lead tap samples or
source water monitoring results (see Section 4.3.2.1.2, activity k)).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

•	cost_5_commercial_lab is the commercial laboratory cost per sample (see Section 4.3.2.1.2,
activity k)).

Cost to analyze lead tap samples = [30 samples + 30 samples * 0.006] * [(0 * (0 hrs * $0/sample + $0)) +
(1 * $57.20/sample)] = $1,726.30

The model PWS will incur this $1,726.30 cost to analyze lead tap samples once per sampling period, or in
this example once every three years.

I) Prepare and Submit Sample Invalidation Request to the State

The system must determine whether any of the samples may be invalid and submits the invalidation
request to the State. The estimation of this cost is represented by the following expression:

Cost to prepare and submit sample invalidation request = numb_reduced_tap * pp_samp_invalid *
hrs_samp_invalid_op * rate_op.

where:

•	numb_reduced_tap is the number of reduced tap samples required per system for systems (i.e.,
number of customers from whom a system must obtain tap samples) on reduced Lead Tap
Sample Monitoring (see Exhibit 4-9).

•	pp_samp_invalid is the likelihood that a lead sample will be deemed invalid (see Section
4.3.2.1.2, activity f)).

•	hrs_samp_invalid_op is the number of hours per invalidated samples to submit sample
invalidation request (see Section 4.3.2.1.2, activity I)).

•	rate_op is the hourly rate for PWS staff (see Chapter 3, Section 3.3.11.1).

Cost to prepare and submit sample invalidation request = 30 samples * 0.006 * 2 hrs * 42.68/hrs =
$15.37.

The model PWS will incur this $15.37 cost to prepare and submit a sample invalidation request once per
sampling period, or in this example once every three years.

m) Inform Customers of Results

After the sampling, systems inform customers of results of the Lead Tap Sampling Monitoring collected
at their household. The estimation of this cost is represented by the following expression:

Final LCRI Economic Analysis

4-60

October 2024


-------
Cost to inform customers of results = numb_reduced_tap * (hrs_inform_samp_op * rate_op +
cost_cust_lt)

where:

•	numb_reduced_tap is the number of reduced tap samples required per system (i.e., number of
customers from whom a system must obtain tap samples) for systems on reduced Lead Tap
Sample Monitoring (see Exhibit 4-9).

•	hrs_inform_samp_op is the number of hours per sample to inform customers of lead results
(see Section 4.3.2.1.2, activity m)).

•	rate_op is the hourly rate for PWS staff (see Section 3.3.11.1).

•	cost_cust_lt is the mailing cost per sample to inform customers of lead results (see Section
4.3.2.1.2, activity m)).

Cost to inform customers of results = 30 samples * (0.11 hrs * $42.68/hr + $0.72/sample) = $162.46

The model PWS will incur this $162.46 cost to inform customers of results once per sampling period, or
in this example once every three years.

n) Certify to the State that Results were Reported

Systems then certify to the State that the Lead Tap Sample Monitoring results were reported to the
customer. The estimation of this cost is represented by the following expression:

Cost to certify that results were reported = hrs_cert_cust_lt_op * rate_op

where:

•	hrs_cert_cust_lt_op is the number of hours to certify to State that Lead Tap Sample Monitoring
results were reported to customers (see Section 4.3.2.1.2, activity m)).

•	rate_op is the hourly rate for PWS staff (see Section 3.3.11.1).

Cost to certify that results were reported = 0.66 hrs * $42.68/hr = $28.17

The model PWS will incur this $28.17 cost to certify that results were reported once per sampling
period, or in this example once every three years.

o) Submit Renewal of Nine-Year Monitoring Waiver Application

Systems on nine-year sampling schedules would also be required to submit renewal of their nine-year
monitoring waiver application, but this would not apply in the case of this example system because as
discussed in Section 3.3.7.1, the EPA assumed only a subset of systems serving 1,000 or fewer would
qualify for this waiver.

p) Submit Monitoring Results and 90th Percentile Calculations to the State

Finally, systems will submit their lead monitoring results and 90th percentile calculations to their State.
The estimation of this cost is represented by the following expression:

Final LCRI Economic Analysis

4-61

October 2024


-------
Cost to draft and submit report on results = hrs_annual_lt_op * rate_op
where:

•	hrs_annual_lt_op is the number of hours it takes to draft and report lead results and 90th
percentile calculations (see Section 4.3.2.1.2, activity p)).

•	rate_op is the hourly rate for PWS staff (see Section 3.3.11.1).

Cost to draft and submit report on results = 3.13 hrs * $42.68/hr = $133.59

The model PWS will incur this $133.59 cost to draft and submit a report on results once per sampling
period or, in this example, once every three years.

Total One-Time Costs and Per Sampling Period Costs

The total one-time cost for the model PWS for activities a) through p) is $3,414.68 and the reoccurring
cost that the PWS will incur once every three years is $6,916.17.

The lead tap water sampling costs of each model PWS in the SafeWater LCR model will vary depending
on the characteristics of the model PWS. For example, if a model PWS with all of the attributes listed
above had a 90th percentile above the AL, the model PWS sampling costs would be quite different.
Instead of conducting one round of sampling every three years (i.e., triennial sampling), the model PWS
would conduct sampling every six months. In addition, instead of taking 30 samples each sampling
period, the model PWS will be required to take 60 samples each sampling period (see
numb_samp_customer in Exhibit 4-9). Finally, the model PWS would conduct customer-initiated
sampling.

4.3.2.2 PWS Lead Water Quality Parameter Monitoring

Lead WQP monitoring is required for all systems serving more than 10,000 people with CCT (except
systems that meet the criteria in 40 CFR 141.81(b)(3) or "b3" systems) and those serving 50,000 or fewer
people without CCT that exceed the lead AL of 10 ng/L. WQP samples are collected at representative
sites throughout the distribution system (also referred to as tap samples) and at each entry point to the
distribution system. Systems must conduct WQP monitoring prior to the installation of CCT and after
CCT installation. The State may designate optimal water quality parameters (OWQPs) after the
installation of CCT. Systems with CCT must continue to maintain WQPs at or above minimum values or
within OWQP ranges designated by the State. Under the final LCRI, systems with CCT that have a single
sample above 10 ng/L must conduct WQP monitoring in the distribution system at or near the site with
the high result and determine if problems with CCT contributed to elevated lead. See 4.3.3.3 for a
discussion of inputs related to this requirement.

The remainder of this section is divided into four subsections:

•	4.3.2.2.1: Baseline Corrosion Control Treatment

•	4.3.2.2.2: Initial Monitoring Schedules

•	4.3.2.2.3: Number of Samples

Final LCRI Economic Analysis

4-62

October 2024


-------
•	4.3.2.2.4: Lead WQP Monitoring Activities

Exhibit 4-37 at the end of Section 4.3.2.2 is a summary exhibit that explains how the cost inputs are
modeled by the SafeWater LCR model.

4.3.2.2.1 Baseline Corrosion Control Treatment

WQP monitoring requirements vary for systems with and without CCT and by type of CCT. To estimate
costs associated with WQP monitoring, the EPA identified systems with and without CCT, as described in
Chapter 3, Section 3.3.3. For those with CCT, the EPA estimated the percentage of systems that
currently have one of the three types of CCT used in the cost model:

•	Modify pH (pbaseph),

•	Add P04 without pH post-treatment (pbasepo4), and

•	Add P04 and modify pH (pbasephpo4).

To develop these percentages, the EPA reviewed the treatment process codes reported for each system
with a reported treatment objective code of "C" for corrosion control in the SDWIS/Fed fourth quarter
2020 frozen dataset. The EPA considered systems to:

•	Have pH adjustment if they had a reported treatment process of: pH adjustment; pH
adjustment, post; or pH adjustment, pre.

•	Use a phosphate-based inhibitor if they had a reported treatment process of: inhibitor,
polyphosphate; inhibitor, orthophosphate; inhibitor, bimetallic phosphate; or inhibitor,
hexametaphosphate.

•	Have both types of treatment if they had at least one of the treatment processes for both pH
adjustment and phosphate-based inhibitor.

•	Have only one of these treatments if they had one of the treatment processes for pH
adjustment but none for a phosphate-based inhibitor or vice versa.

The results of this review are presented in Exhibit 4-17. Eighty-six percent of systems serving 3,300 or
fewer people and 90 percent of systems serving more than 3,300 people reported a process code that
indicated the use of pH adjustment and/or phosphate inhibitor.

Exhibit 4-17: Baseline Percentage of Systems Modifying pH and/or Adding PO4

System Size
(Population Served)

Add PO4 and
Modify pH

Modify pH

Add P04

Total

< 3,300

10%

46%

30%

86%

> 3,300

22%

36%

32%

90%

Since these percentage values will be used to assign CCT process type to systems already known to have
CCT in place, the values in Exhibit 4-17 were normalized to represent the percent of systems with CCT
using pH adjustment, phosphate inhibitor, or both phosphate inhibitor and pH adjustment. For example,

Final LCR! Economic Analysis

4-63

October 2024


-------
of the 86 percent of systems serving 3,300 or fewer people that reported a treatment process of
phosphate and/or pH adjustment, 11 percent reported both a phosphate inhibitor and pH process code
(10 percent/86 percent). These adjusted or normalized percentages are used in the cost model and are
shown in Exhibit 4-18 below.

Exhibit 4-18: Normalized Baseline Percentage of Systems Modifying pH and/or Adding PO4

System Size
(Population Served)

Add PO4 and
Modify pH

Modify pH

Add P04

Total

pbasephpo4

pbaseph

pbasepo4

< 3,300

11%

54%

35%

100%

> 3,300

24%

40%

36%

100%

4.3.2.2.2	Initial MonitorinR Schedules

As described in Section 3.3.8.2, systems with CCT can qualify for reduced WQP monitoring in the
distribution system under the final LCRI if they are in compliance with State set OWQP ranges and do
not exceed the final AL of 10 ng/L. The number of consecutive monitoring periods in which a system
meets these criteria determines if a system will collect two samples at a reduced number of sites in the
distribution system on a semi-annual or annually monitoring schedule. Under the LCRI, systems can no
longer qualify for triennial WQP tap monitoring.

The EPA assumed only systems serving more than 10,000 people with CCT would qualify for reduced
distribution system monitoring because these systems are required to continue WQP monitoring to
demonstrate compliance with their OWQPs unlike smaller systems with CCT or systems without CCT.
Section 3.3.8.2 in Chapter 3 also provides the EPA's approach for determining the estimated percentage
of systems with CCT in each size category that would be on one of three WQP distribution monitoring
schedules at the start of rule implementation based on historical SDWIS/Fed data. These percentages
are provided in Chapter 3 in Exhibit 3-49 and Exhibit 3-50 for ground water and surface water CWSs with
CCT, respectively and in Exhibit 3-51 and Exhibit 3-52 for ground water and surface water NTNCWSs with
CCT, respectively.

4.3.2.2.3	Number of Samples

Exhibit 4-19 provides the minimum number of WQP distribution system samples for CWSs and
NTNCWSs on standard and reduced monitoring. These are from the pre 2021 LCR, which have not been
modified by the 2021 LCRR or the final LCRI with one exception. As discussed in Section 4.3.3.3.3, under
the final LCRI, systems with a lead tap sample result above 10 ng/L must conduct WQP monitoring in the
distribution system at or near the site with the high lead result. If an existing WQP site does not meet
these criteria, the system must identify a new WQP monitoring site and those with CCT must use it for
future sampling in addition to the existing number of WQP sites (numb_enhance_wqp or
numb_reduced_wqp) shown in Exhibit 4-19. Refer to Section 4.3.3.3.3 for a more detailed discussion.

Final LCRI Economic Analysis

4-64

October 2024


-------
Exhibit 4-19: Minimum Number of WQP Distribution Samples for Systems on Standard or

Reduced Monitoring



Standard Monitoring

Reduced Monitoring

System Size (Population
Served)

Number of
Sites

Number of Samples
(2 per site)

Number of
Sites

Number of Samples
(2 per site)





n umb_enhance_ wqp



numb_reduced_wqp



A

B = A*2

C

D = C*2

<500

1

2

1

2

501-3,300

2

4

2

4

3,301-10,000

3

6

3

6

10,001-100,000

10

20

7

14

>100,000

25

50

10

20

Notes: The required minimum number of WQP samples under the LCRI is the same as under the pre-2021 LCR and
2021 LCRR.

A&B: Specifies the number of samples collected in the distribution system for CWSs and NTNCWSs on standard
monitoring during each 6-month period for systems serving > 10,000 people with CCT and systems serving <
50,000 people without CCT during those monitoring periods in which they have a lead or copper ALE and each
subsequent monitoring period until they no longer have an ALE for two consecutive monitoring periods. Systems
must collect 2 samples per site and Column B reflects the input used in the cost model.

C&D: Specifies the reduced number of samples collected in the distribution system for CWSs and NTNCWSs on
reduced monitoring for systems subject to WQP monitoring. Systems on reduced monitoring may be sampling on a
6-month or annual frequency. (See "WQP Schedules_CWS_LCRI_Final.xlsx" and "WQP

Schedules_NTNCWS_LCRI_Final.xlsx" for initial WQP monitoring schedules.) Systems must collect 2 samples per
site and Column D reflects the input used in the cost model.

Systems must also collect WQP samples at each entry point to the distribution system. The number of
entry point samples, which corresponds to the SafeWater LCR model data input numb_ep_wqp, is as
follows:

•	Systems without CCT serving 50,000 or fewer people must collect 2 samples from each entry
point to the distribution system during each 6-month monitoring periods following a lead or
copper ALE. Under the LCRI, they must continue this monitoring until they no longer have an
ALE during two consecutive 6-month monitoring periods.

•	Systems with CCT must collect 1 sample per entry point every 2 weeks. This applies to all
systems serving more than 50,000 except "b3" systems and those serving 10,001 to 50,000
people with CCT. It also applies to systems serving 10,000 or fewer people with CCT during each
6-month monitoring periods following a lead or copper ALE and subsequent monitoring periods
until they no longer have an ALE for two consecutive monitoring periods.

There are no reduced monitoring provisions for WQPs collected at entry points, as was true under the
pre-2021 LCR and 2021 LCRR.

The estimated number of entry points per system, which corresponds to the SafeWater LCR model input
numb_ep, is provided in Chapter 3, Section 3.3.6.

Final LCRI Economic Analysis

4-65

October 2024


-------
4.3.2.2.4 Lead WQP MonitorinR Activities

The EPA has developed water system costs for five lead WQP monitoring activities as shown in Exhibit
4-20. The exhibit provides the unit burden and/or cost for each activity. The assumptions used in the
estimation of the unit burden and costs follow the exhibit. The last column provides the corresponding
SafeWater LCR model data variable in red/italic font.

Exhibit 4-20: PWS Lead WQP Monitoring Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

v) Collect lead WQP

Burden per sample per PWS

Burden

samples from the

0.5 hrs (distribution)

hrs_wqp_op

distribution system







Cost per sample

Cost



No CCT: $2.66 (CWS & NTNCWS)

No CCT: cost_wqp_material



pH adjustment:

pH: cost_wqp_material_ph



• $1.70 to $2.66 (CWS);





• $2.66 (NTNCWS)





Orthophosphate:

Orthophosphate:



• $2.66 to $2.82 (CWS)

cost_wqp_material_ortho



• $2.66 (NTNCWS)



w) Analyze lead WQP

In-House Burden per sample

In-House Burden

samples from the

No CCT: 0.15 hrs (CWS & NTNCWS)

No CCT: hrs_wqp_analyze_dist_op

distribution system

pH adjustment:

pH: hrs_wqp_analyze_ph_op



• 0.15 to 0.46 hrs (CWS)





• 0.15 hrs (NTNCWS)





Orthophosphate:

Orthophosphate:



• 0.15 to 1.34 hrs (CWS)

hrs_wqp_an alyze_ orth o_ op



• 0.15 hrs (NTNCWS)





In-House Cost per sample

In-House Cost



No CCT: $0.63 (CWS & NTNCWS)

No CCT: cost_wqp_analyze



pH adjustment:

pH: cost_wqp_ph_analyze



• $0.63 to $0.98 (CWS)





• $0.63 (NTNCWS)





Orthophosphate:

Orthophosphate:



• $0.63 to $1.07 (CWS)

cost_ wqp_ orth o_ an alyze



• $0.63 (NTNCWS)





Commercial Cost per sample

Commercial Cost



No CCT: $27.24 to 30.55 (CWS & NTNCWS)

No CCT: cost_lab_wqp



pH adjustment: $27.24 to 30.55 (CWS &

pH: cost_lab_ph_wqp



NTNCWS)

Orthophosphate:



Orthophosphate: $60.34 to $61.89 (CWS &

cost_lab_ortho_wqp



NTNCWS)



Final LCR! Economic Analysis

4-66

October 2024


-------
Activity

Unit Burden and/or Cost

SafeWaterLCR Data Variable

x) Collect lead WQP

Burden per sample

Burden

samples from entry

0.4 hrs for 80 percent of ground water

hrs_ep_wqp_op

points

PWSs1







Cost



Cost per sample

cost_ep_wqp_material



No CCT: $2.66 (CWS & NTNCWS)

cost_ep_wqp_ph_material



pH adjustment:





• $1.70 to $2.66 (CWS);





• $2.66 (NTNCWS)

cost_ep_wqp_ortho_material



Orthophosphate:





• $2.66 to $2.82 (CWS)





• $2.66 (NTNCWS)



y) Analyze lead WQP

In-House Burden per sample

In-House Burden

samples from entry

No CCT: 0.15 hrs (CWS & NTNCWS)

hrs_wqp_analyze_ep_op

points

pH adjustment:

hrs_wqp_an alyze_ph_ ep_ op



• 0.15 to 0.46 hrs (CWS)





• 0.15 hrs (NTNCWS)





Orthophosphate:

hrs_wqp_an alyze_ orth o_ ep_ op



• 0.15 to 1.34 hrs (CWS)





• 0.15 hrs (NTNCWS)





In-House Cost per sample

In-House Cost



No CCT: $0.63 (CWS & NTNCWS)

cost_ wqp_analyze_ep



pH adjustment:

cost_ wqp_an alyze_ph_ ep



• $0.63 to $0.98 (CWS)





• $0.63 (NTNCWS)





Orthophosphate:

cost_ wqp_analyze_ orth o_ep



• $0.63 to $1.07 (CWS)





• $0.63 (NTNCWS)





Commercial Cost per sample

Commercial Cost



No CCT:

cost_lab_wqp_ep



• $29.28 to $30.21 (CWS)

cost_lab_wqp_ph_ep



• No CCT: $30.55 (NTNCWS)

cost_lab_wqp_ortho_ep



pH adjustment:





• $30.58 to $33.30 (CWS)





• $33.93 (NTNCWS)





Orthophosphate:





• $61.90 to $63.49 (CWS)





• $63.84 (NTNCWS)



z) Report lead WQP

No CCT: 4 hrs/PWS

Burden

sampling data and

With CCT: 5 hrs/PWS

hrs_ report_ wqp_op

compliance with





OWQPs to the State





Acronyms: CCT = corrosion control treatment; CWS = community water system; NTNCWS = non-transient non-
community water system; PWS = public water system; WQP = water qualify parameter.

Source: "WQP Analytical Burden and Costs_Final.xlsx."

Notes:

Final LCRI Economic Analysis

4-67

October 2024


-------
1 The EPA assumed the burden to collect WQP samples to be 0.4 hours for all surface water systems and 20
percent of ground water systems based on the 2022 Disinfectants/Disinfection Byproducts, Chemical, and
Radionuclides Rules ICR (Renewal), Exhibit 9 (WQP Monitoring - Monitoring, Burden, and Cost Assumptions)
(USEPA, 2022a).

v) Collect lead WQP samples from the distribution system (hrs_wqp_op, cost_wqp_material,
cost_wqp_material_ph, cost_wqp_material_ortho). Systems subject to lead WQP monitoring
requirements must conduct WQP monitoring in the distribution system. The EPA assumed systems
will incur a burden of 0.5 hours to collect each distribution WQP sample (hrs_wqp_op). This
assumption is based on the 2022 Disinfectants/Disinfection Byproducts, Chemical', and Radionuclides
Rules ICR (Renewal), Exhibit 9 (WQP Monitoring - Monitoring, Burden, and Cost Assumptions)
(USEPA, 2022a).

Material costs for sample collection are for sample bottles. All systems subject to WQP
requirements are assumed to conduct pH analyses in-house and for each sample will incur the cost
for a 250-mL bottle in which the sample is collected. For systems using a commercial laboratory, all
other bottle costs are included in the lab cost. Systems conducting in-house analysis of all WQPs
(i.e., CWSs serving more than 100,000 people) will incur additional bottle costs for other analytes.

Exhibit 4-21 and Exhibit 4-22 provides the materials cost associated with sample collection by CCT
status and type for CWSs and NTNCWSs, respectively. The EPA's assumptions for each of these
inputs are detailed in the exhibit notes.

Exhibit 4-21: CWS Material Costs Associated with Distribution System Sample Collection

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_wqp_material

cost_wqp_material_ph

cost_wqp_material_ortho

A

B

C

<50,000

$2.66

$2.66

$2.66

50,001-100,000

$0

$2.66

$2.66

> 100,000

$0

$1.70

$2.82

Acronyms: CCT = corrosion control treatment

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "Non-Labor Cost_CWS_LCRR_LCRI."

Notes:

General: All CWSs subject to WQP monitoring analyze pH in-house, so the likelihood CWSs will conduct pH
analyses in-house or pp_lab_samp is 100 percent. CWSs serving 100,000 or fewer people are assumed to use
commercial laboratories to analyze other parameters for alkalinity and/or orthophosphate so the likelihood a
system will use a commercial lab or pp_commercial_samp is 100 percent for these parameters. The commercial
laboratory cost includes sample bottles. CWSs serving more than 100,000 people are assumed to analyze all WQPs
in-house. For these systems, pp_lab_samp is 100 percent and pp_commercial_samp is 0 percent.

A: Systems without CCT sample pH and alkalinity at entry points and within the distribution system. The EPA
assumed no costs for systems serving > 50,000 people without CCT because they are b3 systems (16 in total) and
are not subject to WQP requirements. Costs for systems serving < 50,000 people is the cost of a 250-mL bottle in
which the pH sample will be collected.

B: Systems using pH adjustment for CCT sample pH and alkalinity at entry points and within the distribution
system. Costs for systems serving < 100,000 people is the cost of a 250-mL bottle in which the pH sample is
collected. The bottle for the alkalinity sample is included in the commercial lab cost. Costs for systems serving >

Final LCRI Economic Analysis

4-68

October 2024


-------
100,000 people is for one 500-ml bottle to collect a sample for pH and alkalinity together. These large systems are
assumed to receive a discount on sample bottles because they buy in bulk.

C: Systems using orthophosphate treatment sample pH, alkalinity, and orthophosphate at entry points and within
the distribution system. Costs for systems serving < 100,000 people is the cost of a 250-mL bottle in which the pH
sample will be collected, with all other bottles being provided by the commercial lab and included in the
commercial lab cost. Costs for systems serving > 100,000 people is for: one 250-ml bottle for orthophosphate, and
one 500-ml bottle to collect a sample for pH and alkalinity together. These large systems are assumed to receive a
discount on sample bottles because they buy in bulk.

Exhibit 4-22: NTNCWS Material Costs Associated with Distribution System Sample Collection

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_wqp_material

cost_wqp_material_ph

cost_wqp_material_ortho

A

B

C

<50,000

$2.66

$2.66

$2.66

50,001-100,000

$0

$2.66

$2.66

100,001-1,000,000

$0

$2.66

$2.66

>1,000,000







Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "Non-Labor Cost_NTNCWS_LCRR_LCRI."
Notes:

General: All NTNCWSs serving 50,001 to 1 million people are assumed to have CCT. No NTNCWS serves > 1 million
people. All NTNCWSs subject to WQP monitoring analyze pH in-house, so the likelihood NTNCWSs will conduct pH
analyses in-house or pp_lab_samp is 100 percent. All NTNCWSs are assumed to use commercial laboratories to
analyze other parameters for alkalinity and/or orthophosphate so the likelihood a system will use a commercial lab
or pp_commercial_samp is 100 percent for these parameters. The commercial laboratory cost includes sample
bottles. Thus, the EPA assumed no NTNCWS would buy bottles in bulk.

A: Systems without CCT sample pH and alkalinity samples at entry points and within the distribution system. All
NTNCWSs serving > 50,000 people are assumed to have CCT. Cost for systems serving < 50,000 people is the cost
of a 250-mL bottle in which the pH sample will be collected.

B: Systems using pH adjustment for CCT sample pH and alkalinity at entry points and within the distribution
system. The cost is for a 250-mL bottle in which the pH sample will be collected. pH, alkalinity, and orthophosphate
at entry points and within the distribution system. The cost is for a 250-mL bottle in which the pH sample will be
collected. The EPA assumed no NTNCWS would buy bottles in bulk because with the exception of pH, all analyses
are conducted by commercial laboratories who provide sample bottles as part of their services.

w) Analyze lead WQP samples from the distribution system. Systems will also incur burden and costs
to analyze WQP samples collected in the distribution system. CWSs serving 100,000 or more people
are assumed to analyze all samples in-house. All CWSs serving 100,000 or fewer people and all
NTNCWSs are assumed to analyze pH in-house and to use a commercial laboratory to analyze other
WQPs such as alkalinity or orthophosphate. Exhibit 4-23 and Exhibit 4-24 provide the analytical
burden for CWSs and NTNCWSs to conduct in-house analyses, respectively. Exhibit 4-25 and Exhibit
4-26 provide the in-house analytical costs for CWSs and NTNCWSs, respectively. Lastly, Exhibit 4-27
and Exhibit 4-28 provide the commercial costs per sample for CWSs and NTNCWSs, respectively.
Detailed assumptions are provided in the notes to each exhibit.

Final LCRI Economic Analysis

4-69

October 2024


-------
Exhibit 4-23: CWS In-House WQP Analytical Burden for Distribution System Samples

(hrs/sample)



Without CCT

With pH Adjustment

With Orthophosphate

System Size
(Population Served)

hrs_ wqp_an alyze_ dist_ op

hrs_ wqp_ analyze_ph_ op

hrs_wqp_an alyze_ orth o_op



A

B

C

<50,000

0.15

0.15

0.15

50,001-100,000

0

0.15

0.15

>100,000

0

0.46

1.34

Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "ln-House_Burden_LCRR_LCRI."

Notes:

General: Burden estimates are the average of estimates provided by three laboratories. All CWSs subject to WQP
monitoring will analyze pH in-house, so the likelihood CWSs will conduct pH analyses in-house or pp_lab_samp is
100 percent. CWSs serving 100,000 or fewer people are assumed to use commercial laboratories to analyze other
parameters for alkalinity and/or orthophosphate so the likelihood a system will use a commercial lab or
pp_commercial_samp is 100 percent for these parameters. CWSs serving more than 100,000 people are assumed
to analyze all WQPs in-house. For these systems, pp_lab_samp is 100 percent and pp_commercial_samp is 0
percent.

A: Systems without CCT sample pH and alkalinity. Assumed no burden for systems serving > 50,000 people without
CCT because they are b3 systems (16 in total) and not subject to WQP requirements. The burden estimate for
systems serving < 50,000 people is to analyze pH in-house.

B: Systems using pH adjustment for CCT sample pH and alkalinity. The burden estimate for systems serving <
100,000 people is to analyze pH in-house and for those serving > 100,000 people to analyze pH and alkalinity in-
house.

C: Systems using orthophosphate treatment sample pH, alkalinity, and orthophosphate. The burden estimate for
systems serving < 100,000 of is to analyze pH in-house and for those serving > 100,000 people to analyze pH,
alkalinity, and orthophosphate in-house.

Exhibit 4-24: NTNCWS In-House WQP Analytical Burden for Distribution System Samples

(hrs/sample)

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

hrs_ wqp_an alyze_ dist_ op

hrs_ wqp_ analyze_ph_ op

hrs_ wqp_ an alyze_ ortho_op

A

B

C

<50,000

0.15

0.15

0.15

50,001-100,000

0

0.15

0.15

100,001-1,000,000

0

0.15

0.15

>1,000,000







Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "ln-House_Burden_LCRR_LCRI."

Notes:

General: All NTNCWSs serving 50,000 to 1 million people are assumed to have CCT; no NTNCWS serves > 1 million
people. Burden is based on estimates from three laboratories. All NTNCWSs subject to WQP monitoring will
analyze pH in-house, so the likelihood NTNCWSs will conduct pH analyses in-house or pp_lab_samp is 100 percent.
All NTNCWSs are assumed to use commercial laboratories to analyze other parameters for alkalinity and/or
orthophosphate so the likelihood a system will use a commercial lab or pp_commercial_samp is 100 percent for
these parameters.

Final LCRI Economic Analysis

4-70

October 2024


-------
A: Systems without CCT sample pH and alkalinity. The EPA assumed no costs for systems serving > 50,000 people
without CCT because all NTNCWSs are assumed to have CCT. The burden estimate for systems serving < 50,000 is
to analyze pH in-house.

B: Systems using pH adjustment for CCT sample pH and alkalinity. The burden estimate is to analyze pH in-house.
C: Systems using orthophosphate treatment sample pH, alkalinity, and orthophosphate. The burden estimate for
all NTNCWSs serving < 50,000 is to analyze pH in-house.

Exhibit 4-25: CWS In-House WQP Analytical Cost for Distribution System Samples ($/sample)

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_ wqp_ analyze

cost_ wqp_ph_ an alyze

cost_wqp_ortho_analyze

A

B

C

<50,000

$0.63

$0.63

$0.63

50,001-100,000

$0

$0.63

$0.63

>100,000

$0

$0.98

$1.07

Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "Non-Labor Cost_CWS_LCRR_LCRI."

Notes:

General: The exhibit presents in-house consumable costs for pH, alkalinity, and orthophosphate that are based on
the average of three vendor quotes. All CWSs subject to WQP monitoring will analyze pH in-house, so the
likelihood CWSs will conduct pH analyses in-house or pp_lab_samp is 100 percent. CWSs serving 100,000 or fewer
people are assumed to use commercial laboratories to analyze other parameters for alkalinity and/or
orthophosphate so the likelihood a system will use a commercial lab or pp_commercial_samp is 100 percent for
these parameters. CWSs serving > 100,000 people are assumed to analyze all WQPs in-house. For these systems,
pp_lab_samp is always 100 percent and pp_commercial_samp is always 0 percent.

A: Systems without CCT sample pH and alkalinity. Assumed no costs for systems serving > 50,000 people without
CCT because they are b3 systems (16 in total) and not subject to WQP requirements.

B: Systems using pH adjustment for CCT sample pH and alkalinity. The consumables cost for systems serving <
100,000 is to analyze pH in-house and for those serving > 100,000 people is to analyze pH and alkalinity in-house.
C: Systems using orthophosphate treatment sample pH, alkalinity, and orthophosphate. The consumables cost for
systems serving < 100,000 people is to analyze pH in-house and for those serving > 100,000 people is to analyze
pH, and alkalinity, and orthophosphate in-house.

Exhibit 4-26: NTNCWS In-House WQP Analytical Cost for Distribution System Samples

($/sample)

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_ wqp_ analyze

cost_ wqp_ph_ an alyze

cost_wqp_ortho_analyze

A

B

C

<50,000

$0.63

$0.63

$0.63

50,001-100,000

$0

$0.63

$0.63

100,001-1,000,000

$0

$0.63

$0.63

>1,000,000







Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx,"worksheet "Non-Labor Cost_NTNCWS_LCRR_LCRI."
Notes:

General: The exhibit presents in-house consumable costs for NTNCWSs that are based on the average of three
vendor quotes. All NTNCWSs serving 50,000 to 1 million people are assumed to have CCT; no NTNCWS serves > 1

Final LCRI Economic Analysis

4-71

October 2024


-------
million people. All NTNCWSs subject to WQP monitoring will analyze pH in-house, so the likelihood NTNCWSs will
conduct pH analyses in-house or pp_lab_samp is 100 percent. All NTNCWSs are assumed to use commercial
laboratories to analyze other parameters for alkalinity and/or orthophosphate so the likelihood a system will use a
commercial lab or pp_commercial_samp is 100 percent for these parameters.

A: Systems without CCT sample pH and alkalinity. The EPA assumed no costs for systems serving > 50,000 people
without CCT because all NTNCWSs are assumed to have CCT. The consumables cost for systems serving < 50,000
people is to analyze pH in-house.

B: Systems using pH adjustment sample pH and alkalinity. All system subject to WQP monitoring analyze pH in-
house. The consumables cost for all NTNCWSs is to analyze pH in-house.

C: Systems using orthophosphate treatment sample pH, alkalinity, and orthophosphate. The consumables cost for
all NTNCWSs is to analyze pH in-house.

Exhibit 4-27: CWS Commercial WQP Analytical Cost for Distribution System Samples

($/sample)

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_lab_wqp

cost_lab_ph_ wqp

cost_lab_ortho_wqp

A

B

C

<500

$30.55

$30.55

$61.89

501-3,300

$28.60

$28.60

$60.99

3,301-10,000

$27.96

$27.96

$60.74

10,001-50,000

$27.24

$27.24

$60.34

50,001-100,000

$0.00

$27.34

$60.45

>100,000

$0.00

$0.00

$0.00

Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "Non-Labor Cost_CWS_LCRR_LCRI."

Notes:

General: The exhibit presents commercial laboratory costs, based on quotes from seven laboratories, for alkalinity
and orthophosphate including shipping the sample to the laboratory. CWSs serving < 100,000 people will use
commercial laboratories for these analyses. CWSs serving > 100,000 people are assumed to conduct all WQP
analyses in-house and thus, will incur no commercial costs. All systems are assumed to conduct pH analyses in-
house, which results in no commercial costs. Note that in general the costs decrease as the size of the water
system increases. This is because larger water systems are sending more samples to the laboratory per shipment
and thus incur a lower per shipping sample cost.

A: Systems without CCT sample pH and alkalinity. The EPA assumed no costs for systems serving > 50,000 people
without CCT because they are b3 systems (16 in total) and not subject to WQP requirements. The commercial cost
for systems serving < 50,000 people is for alkalinity analyses of $26.43 based on the average of prices from seven
laboratories plus shipping costs to the laboratory based on the weight of the number of filled samples. Refer to the
source listed above for additional detail.

B: Systems using pH adjustment for CCT sample pH and alkalinity. The commercial cost for systems serving <
100,000 people is for alkalinity analyses and shipping the samples to the laboratory (see note A).

C: Systems using orthophosphate treatment sample pH, alkalinity, and orthophosphate. The commercial cost for
systems serving < 100,000 is for alkalinity of $26.43 and orthophosphate analysis of $33.29 and shipping the
samples to the laboratory (refer to the source above for additional detail).

Final LCRI Economic Analysis

4-72

October 2024


-------
Exhibit 4-28: NTNCWS Commercial WQP Analytical Cost for Distribution System Samples

($/sample)

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_lab_wqp

cost_lab_ph_ wqp

cost_lab_ortho_wqp

A

B

C

<500

$30.55

$30.55

$61.89

501-3,300

$28.60

$28.60

$60.99

3,301-10,000

$27.96

$27.96

$60.74

10,001-50,000

$27.24

$27.24

$60.34

50,001-100,000

$0.00

$27.34

$60.45

100,001-1,000,000

$0.00

$27.24

$60.34

>1,000,000







Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "Non-Labor Cost_NTNCWS_LCRR_LCRI."
Notes:

General: The exhibit presents commercial costs for alkalinity and orthophosphate. Alkalinity costs are based on the
average of quotes from seven laboratories; those for orthophosphate are based on seven laboratory quotes. All
NTNCWSs serving 50,000 to 1 million people are assumed to have CCT; no NTNCWS serves > 1 million people. All
NTNCWSs are assumed to conduct pH analyses in-house and to use commercial laboratories for other analyses.
A: Systems without CCT will sample pH and alkalinity. The EPA assumed no costs for NTNCWSs serving > 50,000
people without CCT because all NTNCWSs are assumed to have CCT. All NTNCWSs are assumed to analyze pH in-
house but to use commercial laboratories for all other analyses. Note that in general the costs decrease as the size
of the water system increases. This is because larger water systems are sending more samples to the laboratory
per shipment and thus incur a lower per shipping sample cost.

B: Systems using pH adjustment for CCT sample pH and alkalinity. The commercial cost for NTNCWSs is for
alkalinity analyses.

C: Systems using orthophosphate treatment sample pH, alkalinity, and orthophosphate. The commercial cost for
NTNCWSs is for alkalinity and orthophosphate analyses.

x) Collect lead WQP samples from entry points (hrs_ep_wqp_op, cost_ep_wqp_material,

cost_ep_wqp_ph_material, cost_ep_wqp_ortho_material). Systems will also collect WQP samples
at each entry point to the distribution system. The EPA assumed the burden to collect WQP samples
(hrs_ep_wqp_op) to be:

•	0 hours for all surface water systems and 20 percent of ground water systems because they
are already collecting entry point samples to comply with other drinking water regulations.

•	0.4 hours for the remaining 80 percent of ground water systems.

These estimates are based on the 2022 Disinfectants/Disinfection Byproducts, Chemicaland
Radionuclides Rules ICR (Renewal), Exhibit 9 (WQP Monitoring - Monitoring, Burden, and Cost
Assumptions) (USEPA, 2022a).

The EPA assumed that systems will analyze for the same WQPs in entry points samples as
distribution samples and incur the same material costs (i.e., bottle costs) as detailed in activity v).
Even though burden and costs inputs are identical, the EPA used different data variable names for
entry point samples for modeling flexibility in the SafeWater LCR model. The input values and

Final LCR! Economic Analysis

4-73

October 2024


-------
corresponding data variable IDs for entry point samples are provided in Exhibit 4-29 and Exhibit 4-30
for CWSs and NTNCWSs, respectively.

Exhibit 4-29: CWS Material Costs Associated with Entry Point Sample Collection

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_ep_wqp_material

cost_ep_wqp_ph_material

cost_ ep_ wqp_ orth o_materi al

A

B

C

<50,000

$2.66

$2.66

$2.66

50,001-100,000

$0

$2.66

$2.66

>100,000

$0

$1.70

$2.82

Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "Non-Labor Cost_CWS_LCRR_LCRI."
Notes: The input values in this exhibit are identical to Exhibit 4-21. Refer to the exhibit notes for Exhibit 4-21 for
detailed assumptions.

Exhibit 4-30: NTNCWS Material Costs Associated with Entry Point Sample Collection

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_ep_wqp_material

cost_ep_wqp_ph_material

cost_ ep_ wqp_ orth o_materi al

A

B

C

<50,000

$2.66

$2.66

$2.66

50,001-100,000

$0

$2.66

$2.66

100,001-1,000,000

$0

$2.66

$2.66

>1,000,000







Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "Non-Labor Cost_NTNCWS_LCRR_LCRI."
Notes: The input values in this exhibit are identical to Exhibit 4-22. Refer to the exhibit notes for Exhibit 4-22 for
detailed assumptions.

y) Analyze lead WQP samples from entry points. Systems are required to analyze the same WQPs in
entry points samples as distribution system samples and incur the same in-house burden and
material (i.e., bottle) cost as detailed in activity w). They will also have the same commercial costs
but different shipping costs due to differences in the number of entry point samples being shipped
for analysis compared to the number of distribution system samples. The input values with
corresponding SafeWater LCR model entry point data variables are provided in the following
exhibits:

•	Exhibit 4-31 and Exhibit 4-32 for CWS and NTNCWS in-house analytical burden, respectively,

•	Exhibit 4-33 and Exhibit 4-34 for CWS and NTNCWS in-house analytical cost, respectively, and

•	Exhibit 4-35 and Exhibit 4-36 for CWSs and NTNCWS commercial analyses, respectively.

Final LCR! Economic Analysis

4-74

October 2024


-------
Exhibit 4-31: CWS In-House WQP Analytical Burden for Entry Point Samples (hrs/sample)



Without CCT

With pH Adjustment

With Orthophosphate

System Size
(Population Served)

hrs_ wqp_an alyze_ ep_op

hrs_wqp_analyze_ph_ep_op

hrs_ wqp_ an alyze_ ortho_op



A

B

C

<50,000

0.15

0.15

0.15

50,001-100,000

0

0.15

0.15

>100,000

0

0.46

1.34

Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "ln-House_Burden_LCRR_LCRI."

Notes: The input values in this exhibit are identical to Exhibit 4-23. Refer to the exhibit notes for Exhibit 4-23 for
detailed assumptions.

Exhibit 4-32: NTNCWS In-House WQP Analytical Burden for Entry Point Samples (hrs/sample)

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

hrs_ wqp_an alyze_ ep_op

hrs_wqp_analyze_ph_ep_op

hrs_ wqp_ an alyze_ ortho_op

A

B

C

<50,000

0.15

0.15

0.15

50,001-100,000

0

0.15

0.15

100,001-1,000,000

0

0.15

0.15

>1,000,000







Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "ln-House_Burden_LCRR_LCRI."

Notes: The input values in this exhibit are identical to Exhibit 4-24. Refer to the exhibit notes for Exhibit 4-24 for
detailed assumptions.

Exhibit 4-33: CWS In-House WQP Analytical Cost for Entry Point Samples ($/sample)

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_ wqp_ an alyze_ep

cost_ wqp_ an alyze_ph_ ep

cost_ wqp_ an alyze_ ortho_ep

A

B

C

<50,000

$0.63

$0.63

$0.63

50,001-1,000,000

$0

$0.63

$0.63

>1,000,000

$0

$0.98

$1.07

Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet, "Non-Labor Cost_NTNCWS_LCRR_LCRI."
Notes: The input values in this exhibit are identical to Exhibit 4-25. Refer to the exhibit notes for Exhibit 4-25 for
detailed assumptions.

Final LCRI Economic Analysis

4-75

October 2024


-------
Exhibit 4-34: NTNCWS In-House WQP Analytical Cost for Entry Point Samples ($/sample)

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_ wqp_ an alyze_ep

cost_ wqp_an alyze_ph_ ep

cost_ wqp_ an alyze_ ortho_ep

A

B

C

<50,000

$0.63

$0.63

$0.63

50,001-100,000

$0

$0.63

$0.63

100,001-1,000,000

$0

$0.63

$0.63

>1,000,000







Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "Non-Labor Cost_NTNCWS_LCRR_LCRI."
Notes: The input values in this exhibit are identical to Exhibit 4-26. Refer to the exhibit notes for Exhibit 4-26 for
detailed assumptions.

Exhibit 4-35: CWS Commercial WQP Analytical Cost for Entry Point Samples ($/sample)

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_lab_wqp_ep

cost_lab_wqp_ph_ep

cost_lab_wqp_ortho_ep

A

B

C

<100

$30.21

$33.30

$63.49

101-500

$29.95

$32.82

$63.23

501-1,000

$29.39

$32.82

$62.68

1,001-3,300

$29.49

$32.00

$62.78

3,301-10,000

$29.52

$32.61

$62.98

10,001-50,000

$29.28

$31.84

$62.56

<50,000

$0

$30.58

$61.90

50,001-100,000

$0

$0

$0

>100,000

$0

$0

$0

Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "Non-Labor Cost_CWS_LCRR_LCRI."

Notes:

General: The exhibit presents commercial laboratory costs, based on quotes from seven laboratories, for alkalinity
and orthophosphate including shipping the sample to the laboratory. CWSs serving < 100,000 people will use
commercial laboratories for these analyses. CWSs serving > 100,000 people are assumed to conduct all WQP
analyses in-house and thus, will incur no commercial costs. All systems are assumed to conduct pH analyses in-
house, which results in no commercial costs. Note that in general the costs decrease as the size of the water
system increases. This is because larger water systems are sending more samples to the laboratory per shipment
and thus incur a lower per shipping sample cost.

A: Systems without CCT sample pH and alkalinity. The EPA assumed no costs for systems serving > 50,000 people
without CCT because they are b3 systems (16 in total) and not subject to WQP requirements. The commercial cost
for systems serving < 50,000 people is for alkalinity analyses of $26.43 based on the average of prices from seven
laboratories plus shipping costs to the laboratory based on the weight of the number of filled samples. Refer to the
source listed above for additional detail.

B: Systems using pH adjustment for CCT sample pH and alkalinity. The commercial cost for systems serving <
100,000 people is for alkalinity analyses and shipping the samples to the laboratory (see note A).

C: Systems using orthophosphate treatment sample pH, alkalinity, and orthophosphate. The commercial cost for
systems serving < 100,000 is for alkalinity of $26.43 and orthophosphate analysis of $33.29 and shipping the
samples to the laboratory (refer to the source above for additional detail).

Final LCRI Economic Analysis

4-76

October 2024


-------
Exhibit 4-36: NTNCWS Commercial WQP Analytical Cost for Entry Point Samples ($/sample)

System Size
(Population Served)

Without CCT

With pH Adjustment

With Orthophosphate

cost_lab_wqp_ep

cost_lab_wqp_ph_ep

cost_lab_wqp_ortho_ep

A

B

C

<50,000

$30.55

$33.93

$63.84

50,001-100,000

$0

$33.93

$63.84

100,001-1,000,000

$0

$0

$0

>1,000,000







Acronyms: CCT = corrosion control treatment.

Source: File "WQP Analytical Burden and Costs_Final.xlsx," worksheet "Non-Labor Cost_NTNCWS_LCRR_LCRI."
Notes:

General: No NTNCWS serves > 1 million people. The exhibit presents commercial laboratory costs, based on quotes
from seven laboratories, for alkalinity and orthophosphate including shipping the sample to the laboratory. All
NTNCWSs are assumed to use commercial laboratories for these analyses.

A: Systems without CCT sample pH and alkalinity. All NTNCWSs serving 50,000 to 1 million people are assumed to
have CCT. The commercial cost for systems serving < 50,000 people is for alkalinity analyses of $26.43 based on the
average of prices from seven laboratories plus shipping costs to the laboratory based on the weight of the number
of filled samples. Refer to the source listed above for additional detail.

B: Systems using pH adjustment for CCT sample pH and alkalinity. The commercial cost for systems serving <
100,000 people is for alkalinity analyses and shipping the samples to the laboratory (see note A).

C: Systems using orthophosphate treatment sample pH, alkalinity, and orthophosphate. The commercial cost for
systems serving < 100,000 is for alkalinity of $26.43 and orthophosphate analysis of $33.29 and shipping the
samples to the laboratory. Refer to the source listed above for additional detail.

z) Report lead WQP sampling data and compliance with OWQPs to the State (hrs_report_wqp_op).

Systems are required to report their WQP results and for those systems where OWQPs have been
set to demonstrate compliance with those OWQPs every six months. The EPA estimated systems
with CCT and without CCT would require 5 hours and 4 hours, respectively. The estimated reporting
burden for systems with CCT is based on the WQP Reporting (Annual) burden in Exhibit 9 of the
2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR (Renewal)

(USEPA, 2022a). The EPA assumed systems without CCT would incur a lower burden because they
would be reporting less entry point monitoring data than those with CCT that must conduct entry
point monitoring biweekly. These systems without CCT are also not determining compliance with
OWQPs.

Exhibit 4-37 shows the SafeWater LCR model cost estimation approach for water system lead WQP
monitoring activities including additional cost inputs that are required to calculate these costs.

Final LCR! Economic Analysis

4-77

October 2024


-------
Exhibit 4-37: PWS Lead WQP Monitoring Cost Estimation in SafeWater LCR by Activity1





Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

v) Collect lead WQP samples from the distribution system

Number of samples multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(numb enhance wqp*((hrs wqp op*rate op)+cost wqp material))





Model PWSs serving <50,000
without CCT



Number of samples multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(numb_enhance_wqp*((hrs_wqp_op*rate_op)+cost_wqp_material_ph))

Cost applies as
written to
NTNCWSs.

Above AL

Model PWSs serving <10,000
with pH adjustment

pbaseph

Twice per
year

Number of samples multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(numb_enhance_wqp*((hrs_wqp_op*rate_op)+cost_wqp_material_ortho))





Model PWSs serving <10,000
with PCM or both PCM and pH
adjustment

pbasepo4, pbasephpo4



Number of samples multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(numb_enhance_wqp*((hrs_wqp_op*rate_op)+cost_wqp_material_ph))

Cost applies as
written to
NTNCWSs.

All

Model PWSs serving >10,000
with pH adjustment that do not
qualify for reduced WQP
monitoring

Twice per
year

Number of samples multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(numb_enhance_wqp*((hrs_wqp_op*rate_op)+cost_wqp_material_ortho))





Model PWSs serving >10,000
with PO4 or both PO4 and pH
adjustment that do not qualify
for reduced WQP monitoring

pbasepo4, pbasephpo4,
1 - (p_wqp_annual +
p wqp six red)

Twice per
year

Final LCR! Economic Analysis

4-78

October 2024


-------
Conditions for Cost to Apply to a
Model PWS

CWS Cost Per Activity	NTNCWS Cost	Frequency

Per Activity Lead 90*	Other Conditions	of Activity







Model PWSs serving >10,000
with pH adjustment on six-
month reduced WQP
monitoring

Twice a
year







pbaseph, p_wqp_six_red



Number of samples multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(numb_reduced_wqp*((hrs_wqp_op*rate_op)+cost_wqp_material_ph))

Cost applies as
written to
NTNCWSs.

All

Model PWSs serving >10,000
with pH adjustment on annual
WQP monitoring

pbaseph, p_wqp_annual

Once a
year

Number of samples multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(numb_reduced_wqp*((hrs_wqp_op*rate_op)+cost_wqp_material_ortho))

Cost applies as
written to
NTNCWSs.



Model PWSs serving >10,000
with PCM or both PCM and pH
adjustment on six-month
reduced sample WQP
monitoring

pbasepo4, pbasephpo4,
p_wqp_six_red

Twice a
year





All

Model PWSs serving >10,000
with PO4 or both PO4 and pH
adjustment on annual WQP
monitoring

pbasepo4, pbasephpo4,
p_wqp_annual

Once a
year

Final LCRI Economic Analysis

4-79

October 2024


-------
CWS Cost Per Activity

w) Analyze lead WQP samples from the distribution system

There are different labor (burden) and material costs for a sample analyzed in
house and a sample analyzed using a commercial lab. The in-house analysis
costs are calculated using the number of required samples per system
multiplied by the percentage of samples analyzed in house times the system
labor rate, plus the material cost of the in-house analysis per sample. The
commercial lab analysis costs are calculated using the number of required
samples per system multiplied by the percentage of samples analyzed in a
commercial lab times the system labor rate, plus the material cost of the
commercial lab analysis per sample.

(((numb_enhance_wqp*pp_lab_samp)*((hrs_wqp_analyze_dist_op*rate_op)+
cost_wqp_analyze))+((numb_enhance_wqp*pp_commercial_samp)*cost_lab_
wqp))

There are different labor (burden) and material costs for a sample analyzed in
house and a sample analyzed using a commercial lab. The in-house analysis
costs are calculated using the number of required samples per system
multiplied by the percentage of samples analyzed in house times the system
labor rate, plus the material cost of the in-house analysis per sample. The
commercial lab analysis costs are calculated using the number of required
samples per system multiplied by the percentage of samples analyzed in a
commercial lab times the system labor rate, plus the material cost of the
commercial lab analysis per sample.

(((numb_enhance_wqp*pp_lab_samp)*((hrs_wqp_analyze_ph_op*rate_op)+c
ost_wqp_ph_analyze))+((numb_enhance_wqp*pp_commercial_samp)*cost_la
b_ph_wqp))

There are different labor (burden) and material costs for a sample analyzed in
house and a sample analyzed using a commercial lab. The in-house analysis
costs are calculated using the number of required samples per system
multiplied by the percentage of samples analyzed in house times the system
labor rate, plus the material cost of the in-house analysis per sample. The
commercial lab analysis costs are calculated using the number of required
samples per system multiplied by the percentage of samples analyzed in a
commercial lab times the system labor rate, plus the material cost of the
commercial lab analysis per sample.	

Final LCRI Economic Analysis



Conditions for Cost to Apply to a
Model PWS



NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

Cost applies as
written to
NTNCWSs.

Model PWSs serving <50,000
without CCT

Twice a
year

Model PWSs serving <10,000
with pH adjustment

pbaseph

Model PWSs serving <10,000
with PCM or both PCM and pH
adjustment

pbasepo4, pbasephpo4

4-80

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

(((numb_enhance_wqp*pp_lab_samp)*((hrs_wqp_analyze_ortho_op*rate_op)
+cost_wqp_ortho_analyze))+((numb_enhance_wqp*pp_commercial_samp)*c
ost_lab_ortho_wqp))









There are different labor (burden) and material costs for a sample analyzed in
house and a sample analyzed using a commercial lab. The in-house analysis
costs are calculated using the number of required samples per system
multiplied by the percentage of samples analyzed in house times the system
labor rate, plus the material cost of the in-house analysis per sample. The
commercial lab analysis costs are calculated using the number of required
samples per system multiplied by the percentage of samples analyzed in a
commercial lab times the system labor rate, plus the material cost of the
commercial lab analysis per sample.

(((numb_enhance_wqp*pp_lab_samp)*((hrs_wqp_analyze_ph_op*rate_op)+c
ost_wqp_ph_analyze))+((numb_enhance_wqp*pp_commercial_samp)*cost_la
b ph wqp))

Cost applies as
written to
NTNCWSs.

All

Model PWSs serving >10,000
with pH adjustment that do not
qualify for reduced WQP
monitoring

pbaseph;

1- (p_wqp_annual +
p_ wqp_six_red)

Twice a
year

There are different labor (burden) and material costs for a sample analyzed in
house and a sample analyzed using a commercial lab. The in-house analysis
costs are calculated using the number of required samples per system
multiplied by the percentage of samples analyzed in house times the system
labor rate, plus the material cost of the in-house analysis per sample. The
commercial lab analysis costs are calculated using the number of required
samples per system multiplied by the percentage of samples analyzed in a
commercial lab times the system labor rate, plus the material cost of the
commercial lab analysis per sample.

(((numb_enhance_wqp*pp_lab_samp)*((hrs_wqp_analyze_ortho_op*rate_op)
+cost_wqp_ortho_analyze))+((numb_enhance_wqp*pp_commercial_samp)*c
ost lab ortho wqp))





Model PWSs serving >10,000
with PCM or both PCM and pH
adjustment that do not qualify
for reduced WQP monitoring

pbasepo4; pbasephpo4;
1 - (p_wqp_annual +
p_ wqp_six_red)

Twice a
year

Final LCRI Economic Analysis

4-81

October 2024


-------
Conditions for Cost to Apply to a
Model PWS

CWS Cost Per Activity	NTNCWS Cost	Frequency

Per Activity Lead 90*	Other Conditions	of Activity

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor
and material cost burdens for each type of analysis.

(((numb_reduced_wqp*pp_lab_samp)*((hrs_wqp_analyze_ph_op*rate_op)+co
st_wqp_ph_analyze))+((numb_reduced_wqp*pp_commercial_samp)*cost_lab
_ph_wqp))

Cost applies as
written to
NTNCWSs.

All

Model PWSs serving >10,000
with pH adjustment on six-
month reduced sample WQP
monitoring

pbaseph, p_wqp_six_red

Twice a
year

Model PWSs serving >10,000
with pH adjustment on annual
WQP monitoring

pbaseph, p_wqp_annual

Once a
year

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor
and material cost burdens for each type of analysis.

(((numb_reduced_wqp*pp_lab_samp)*((hrs_wqp_analyze_ortho_op*rate_op)

+cost_wqp_ortho_analyze))+((numb_reduced_wqp*pp_commercial_samp)*co

st_lab_ortho_wqp))

Cost applies as
written to
NTNCWSs.

All

Model PWSs serving > 10,000
with PCM or both PCM and pH
adjustment on six-month
reduced WQP monitoring

pbasepo4, pbasephpo4,
p_wqp_six_red

Twice a
year

Model PWSs serving > 10,000
with PO4 or both PO4 and pH
adjustment on annual WQP
monitoring

pbasepo4, pbasephpo4,
p_wqp_annual

Once a
year

Final LCRI Economic Analysis

4-82

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

x) Collect lead WQP samples from entry points

The number of entry points per system multiplied by the number of samples,
then multiplied by the total of the labor hours per sample times the system
labor rate, plus the cost per sample.

((numb ep*numb ep wqp)*((hrs ep wqp op*rate op)+cost ep wqp materia

0)





Model PWSs serving <50,000
without CCT

Twice a
year

The number of entry points per system multiplied by the number of samples,
then multiplied by the total of the labor hours per sample times the system
labor rate, plus the cost per sample.

((numb_ep*numb_ep_wqp)*((hrs_ep_wqp_op*rate_op)+cost_ep_wqp_ph_ma
terial))

Cost applies as
written to
NTNCWSs.

Above AL

Model PWSs serving <10,000
with pH adjustment

pbaseph

Every 2
weeks

The number of entry points per system multiplied by the number of samples,
then multiplied by the total of the labor hours per sample times the system
labor rate, plus the cost per sample.

((numb_ep*numb_ep_wqp)*((hrs_ep_wqp_op*rate_op)+cost_ep_wqp_ortho_
material))





Model PWSs serving <10,000
with PCM or both PCM and pH
adjustment

pbasepo4, pbasephpo4



The number of entry points per system multiplied by the number of samples,
then multiplied by the total of the labor hours per sample times the system
labor rate, plus the cost per sample.

((numb_ep*numb_ep_wqp)*((hrs_ep_wqp_op*rate_op)+cost_ep_wqp_ph_ma
terial))

Cost applies as
written to
NTNCWSs.

All

Model PWSs serving > 10,000
with pH adjustment

pbaseph

Every 2
weeks

The number of entry points per system multiplied by the number of samples,
then multiplied by the total of the labor hours per sample times the system
labor rate, plus the cost per sample.

((numb_ep*numb_ep_wqp)*((hrs_ep_wqp_op*rate_op)+cost_ep_wqp_ortho_
material))





Model PWSs serving > 10,000
with PCM or both PCM and pH
adjustment

pbasepo4, pbasephpo4



Final LCRI Economic Analysis

4-83

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

y) Analyze lead WQP entry point samples

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor
and material cost burdens for each type of analysis.

((((numb_ep*numb_ep_wqp)*pp_lab_samp)*((hrs_wqp_analyze_ep_op*rate_
op)+cost_wqp_analyze_ep))+(((numb_ep*numb_ep_wqp)*pp_commercial_sa
mp)* cost_lab_wqp_ep))

Cost applies as
written to
NTNCWSs.

Above AL

Model PWSs serving <50,000
without CCT

Twice a
year

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor
and material cost burdens for each type of analysis.

((((numb_ep*numb_ep_wqp)*pp_lab_samp)*((hrs_wqp_analyze_ph_ep_op*ra
te_op)+cost_wqp_analyze_ph_ep))+(((numb_ep*numb_ep_wqp)*pp_commer
cial samp)*cost lab wqp ph ep))

Cost applies as
written to
NTNCWSs.

Above AL

Model PWSs serving <10,000
with pH adjustment

pbaseph

Every two
weeks

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor
and material cost burdens for each type of analysis.

((((numb_ep*numb_ep_wqp)*pp_lab_samp)*((hrs_wqp_analyze_ortho_ep_op
*rate_op)+cost_wqp_analyze_ortho_ep))+(((numb_ep*numb_ep_wqp)*pp_co
mmercial samp)*cost lab wqp ortho ep))



Above AL

Model PWSs serving <10,000
with PCM or both PCM and pH
adjustment

pbasepo4, pbasephpo4



The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor
and material cost burdens for each type of analysis.

((((numb_ep*numb_ep_wqp)*pp_lab_samp)*((hrs_wqp_analyze_ph_ep_op*ra
te_op)+cost_wqp_analyze_ph_ep))+(((numb_ep*numb_ep_wqp)*pp_commer
cial samp)*cost lab wqp ph ep))

Cost applies as
written to
NTNCWSs.

All

Model PWSs serving >10,000
with pH adjustment

pbaseph

Every two
weeks

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor
and material cost burdens for each type of analysis.

((((numb_ep*numb_ep_wqp)*pp_lab_samp)*((hrs_wqp_analyze_ortho_op*rat
e_op)+cost_wqp_analyze_ortho_ep))+(((numb_ep*numb_ep_wqp)*pp_comm
ercial samp)*cost lab wqp ortho ep))





Model PWSs serving >10,000
with PCM or both PCM and pH
adjustment

pbasepo4, pbasephpo4



Final LCRI Economic Analysis

4-84

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

z) Report lead WQP sampling data and compliance with OWQPs to the State







Model PWSs serving <50,000
without CCT







Above AL

Model PWSs serving <10,000
with pH adjustment

pbaseph



The labor hours for reporting per system multiplied by the labor rate.
(hrs_report_wqp_op*rate_op)

Cost applies as
written to
NTNCWSs.



Model PWSs serving <10,000
with PO4 or both PCM and pH
adjustment

pbasepo4, pbasephpo4

Twice a
year







Model PWSs serving >10,000
with pH adjustment







All

pbaseph









Model PWSs serving >10,000
with PO4 or both PCM and pH
adjustment









pbasepo4, pbasephpo4



Acronyms: AL = action level; CCT = corrosion control treatment; CWS = community water system; NTNCWS = non-transient non-community water system;

OWQP = optimal water quality parameter; PO4 = orthophosphate; PWS = public water system; WQP = water quality parameter.

Note:

1The data variables in the exhibit are defined previously in this section with the exception of:

•	numb_enhance_wqp: number of distribution system samples for systems on standard WQP monitoring (Section 4.3.2.2.3).

•	numb_ep: number of entry points per systems (Section 4.3.2.2.3).

•	numb_reduced_wqp\ number of distribution system samples for systems on reduced WQP monitoring (Section 4.3.2.2.3).

•	pbaseplr. likelihood a system has an existing CCT of modify pH (Section 4.3.2.2.1).

•	pbasepo4: likelihood a system has existing CCT of adding PO4 without pH post treatment (Section 4.3.2.2.1).

•	pbasephpo4\ likelihood a system has existing CCT of adding PO4 with modify pH (Section 4.3.2.2.1).

•	P-wqP-six-red, p_wqp_annual\ likelihood a system is on reduced distribution system monitoring schedule at a semi-annual or annual schedule,
respectively (Section 4.3.2.2.2).

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

Final LCRI Economic Analysis

4-85

October 2024


-------
4.3.2.3 PWS Copper Water Quality Parameter Monitoring

This discussion of copper WQP monitoring costs for water systems is presented in the following
subsections:

•	4.3.2.3.1: Applicability and Likelihood of a Copper ALE

•	4.3.2.3.2: Copper WQP Monitoring Activities

4.3.2.3.1 Applicability and Likelihood of a Copper ALE

The SafeWater LCR models Copper WQP Monitoring separately from the Lead WQP Monitoring. The
frequency of Lead WQP Monitoring depends on the lead 90th percentile, with all systems above the AL
and all systems serving more than 10,000 people with CCT, and those serving more than 50,000 people
without CCT105 conducting Lead WQP Monitoring. Copper WQP Monitoring is required when a system
exceeds the copper AL. To not double count the cost of WQP monitoring for systems experiencing both
a copper ALE and a lead ALE simultaneously, the SafeWater LCR models the costs of Copper and Lead
WQP Monitoring separately and restricts Copper WQP Monitoring to systems with a copper ALE only
and lead 90th percentile not greater than the lead AL.

Note that the cost inputs used to estimate WQP costs in response to a copper ALE are identical to those
incurred in response to a lead ALE with the following exceptions:

•	The likelihood of a system exceeding the copper AL only, which corresponds to p_copper_ale, is
used in lieu of a system's lead 90th percentile level.

•	Systems are not assumed to be on reduced WQP distribution system monitoring in response to
a copper ALE, and all systems are assumed to be on a 6-month standard monitoring schedule.
Thus, the data inputs associated with reduced monitoring are not applicable. These include the
reduced number of WQP monitoring samples per distribution sample site (numb_reduced_wqp),
and the likelihood that a system will be on a 6-month (p_wqp_six_red) or annual
(p_wqp_annual) WQP sampling schedule.

Exhibit 4-38 and Exhibit 4-39 provide the likelihood that a CWS and NTNCWS, respectively, will exceed
the copper AL of 1.3 mg/L, but not the final lead AL of 10 ng/L (p_copper_ale). In each exhibit, the
estimated percentages are provided for each of the 9 size categories and two source water types used in
the cost model. For systems without CCT, the EPA derived the percentages from SDWIS/Fed 90th
percentile results from 2012 - 2020106 as follows:

105	All systems serving more than 50,000 people are required to have CCT and to conduct WQP monitoring with the
exception of systems that have naturally non-corrosive water, i.e., "b3" systems. Refer to Chapter 3, Section 3.3.3
for the EPA's approach for deriving the number of "b3" systems (assumed to be 16 CWSs).

106	The EPA expanded the analysis period from the proposed LCRI EA (USEPA, 2023c) of 2017 - 2020 to be more
consistent with other analyses that use a nine-year analysis period.

Final LCRI Economic Analysis

4-86

October 2024


-------
•	Step 1: For each year during 2012 - 2020, the EPA identified the number of CWSs with a
reported copper 90th percentile value above the copper AL and no reported lead 90th percentile
above the AL107 for the nine size categories and two source types.

•	Step 2: The EPA divided the number of systems in step 1 by the number of CWSs in each size and
source strata to develop a percentage. Each percentage was divided by 100 to derive the
likelihood.

The EPA used the same approach to develop the estimated percent of NTNCWSs with independent
copper ALEs. Chapter 3, Exhibit 3-34 and Exhibit 3-35 provide the results of this analysis for CWSs and
NTNCWSs, respectively. Note that for modeling purposes, the EPA assumed that no system with CCT
would have a copper ALE and thus, would have a likelihood of 0 percent. The EPA made this simplifying
assumption because only approximately 1 percent of CWSs and approximately 3 percent of NTNCWSs
with CCT were estimated to have a copper only ALE. Exhibit 4-38 and Exhibit 4-39 provide the likelihoods
used in the SafeWater LCR model for the data variable p_copper_ale. As shown in these exhibits, no
CWS or NTNCWS serving more than 50,000 people is assumed to have a copper only ALE and be subject
to copper WQP monitoring. However, as discussed in Section 4.3.2.2, these systems are assumed to
conduct lead WQP monitoring with the exception of those designated as "b3" systems.

Exhibit 4-38: Estimated Likelihood a CWS Will Have a Copper Only ALE (2012 - 2020)

System Size
(Population Served)

p_copper_ale

with CCT1

without CCT2

Ground Water

Surface Water

Ground Water

Surface Water

<100

0.000

0.000

0.003

0.010

101-500

0.000

0.000

0.004

0.007

501-1,000

0.000

0.000

0.004

0.006

1,001-3,300

0.000

0.000

0.005

0.003

3,301-10,000

0.000

0.000

0.003

0.002

10,001-50,000

0.000

0.000

0.001

0.001

50,001-100,000

0.000

0.000

0.000

0.000

100,001-1,000,000

0.000

0.000

0.000

0.000

>1,000,000

0.000

0.000





Acronyms: CCT = corrosion control treatment.

Source:

SDWIS/Fed fourth quarter frozen data set, current through December 31, 2020. Also see "CWS Inventory
Characteristics Final.xlsx" for additional detail.

107 As noted in Chapter 3, Section 3.3.5.4 that details this analysis, the EPA did not adjust the lead 90th percentile
values to consider the change in the sampling protocol for systems with LSLs. Nor did the EPA account for the
proposed LCRI's lower lead AL of 10 ng/L. Thus, the EPA's approach will likely overestimate the percentage of
systems having a copper only ALE because more of these systems may also have a lead ALE under the proposed
LCRI compared to under the 2021 LCRR.

Final LCRI Economic Analysis

4-87

October 2024


-------
Notes:

General: The EPA estimated that 16 CWSs are b3 systems, serve 50,001 - 1 million people, and have no CCT. No b3
systems serve more than 1 million people and is indicated using gray shading.

1	Note that for modeling purposes, the EPA assumed that no system with CCT would have a copper ALE and thus,
would have a likelihood of 0 percent.

2	For each year during 2012 - 2020, the EPA identified the number of CWSs with a reported copper 90th percentile
value above the copper AL and no reported lead 90th percentile values above the AL for the 9 size categories and
two source types. The EPA then divided the number of systems by the number of CWSs in each size and source
strata to develop a percentage. Each percentage was divided by 100 to derive the likelihood. Note that all systems
serving > 50,000 people without CCT (16 systems) are categorized as "b3" systems and have no copper ALEs (see
Section 3.3.3 in Chapter 3 for additional detail).

Exhibit 4-39: Estimated Likelihood a NTNCWS Will Have a Copper Only ALE (2012 - 2020)

System Size
(Population Served)

p_copper_ale

with CCT1

without CCT2

Ground Water

Surface Water

Ground Water

Surface Water

<100

0.000

0.000

0.006

0.003

101-500

0.000

0.000

0.006

0.012

501-1,000

0.000

0.000

0.006

0.012

1,001-3,300

0.000

0.000

0.008

0.017

3,301-10,000

0.000

0.000

0.001

0.007

10,001-50,000

0.000

0.000

0.015

0.007

50,001-100,000



0.000





100,001-1,000,000



0.000





>1,000,000









Acronyms: CCT = corrosion control treatment.

Source:

SDWIS/Fed fourth quarter frozen data set, current through December 31, 2020. Also see "NTNCWS Inventory

Characteristics_Final.xlsx" for additional detail.

Notes:

General: Two NTNCWSs serve 50,001 - 1 million people and each have CCT. No NTNCWS serves > 1 million people.

1	Note that for modeling purposes, the EPA assumed that no system with CCT would have a copper ALE and thus,
would have a likelihood of 0 percent.

2	For each year during 2012 - 2020, the EPA identified the number of NTNCWSs with a reported copper 90th
percentile value above the copper AL and no reported lead 90th percentile values above the AL for the 9 size
categories and two source types. The EPA then divided the number of systems by the number of NTNCWSs in each
size and source strata to develop a percentage. Each percentage was divided by 100 to derive the likelihood.

4,3.2.3.2 Copper WQP Monitoring Activities

The activities, unit burden and costs, and data variables used to estimate copper WQP monitoring costs
are identical to those for lead, as shown in Exhibit 4-40, with the exception that they are triggered in
response to a copper ALE.

Final LCRI Economic Analysis

4-88

October 2024


-------
Exhibit 4-40: PWS Copper WQP Monitoring Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

aa) Collect copper WQP samples
from the distribution system

Same as Exhibit 4-20, activity v).

bb) Analyze copper WQP
samples from the
distribution system

Same as Exhibit 4-20, activity w).

cc) Collect copper WQP samples
from entry points

Same as Exhibit 4-20, activity x).

dd) Analyze copper WQP
samples from entry points

Same as Exhibit 4-20, activity y).

ee) Report copper WQP
sampling data and
compliance with OWQPs to
the State

Same as Exhibit 4-20, activity z).

Acronyms: ALE = action level exceedance; OWQP = optimal water qualify parameter; WQP = water qualify
parameter.

Source: "WQP Analytica Burden and Costs_Final.xlsx."

Exhibit 4-41 shows the SafeWater LCR model cost estimation approach for system WQP monitoring
activities in response to a copper ALE including additional cost inputs that are required to calculate
these costs.

Final LCR! Economic Analysis

4-89

October 2024


-------
Exhibit 4-41: PWS Copper WQP Monitoring Cost Estimation in SafeWater LCR by Activity1





Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency of
Activity

aa) Collect copper WQP samples in the distribution system

Number of samples multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(numb_enhance_wqp*((hrs_wqp_op*rate_op)+cost_wqp_material))





Model PWSs serving
<50,000 without CCT and
have a copper ALE

p_copper_ale



Number of samples multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(numb_enhance_wqp*((hrs_wqp_op*rate_op)+cost_wqp_material_ph))

Cost applies
as written to
NTNCWSs.

At or below
AL

Model PWSs serving
<10,000 that have pH
adjustment and a copper
ALE

p_copper_ale, pbaseph

Twice per
event

Number of samples multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(numb_enhance_wqp*((hrs_wqp_op*rate_op)+cost_wqp_material_ortho))





Model PWSs serving
<10,000 that have PCM or
both PCM and pH adjustment
and a copper ALE

p_copper_ale, pbasepo4,
pbasephpo4



Final LCR! Economic Analysis

4-90

October 2024


-------
CWS Cost Per Activity

bb) Analyze copper WQP samples from the distribution system

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor and
material cost burdens for each type of analysis.

(((numb_enhance_wqp*pp_lab_samp)*((hrs_wqp_analyze_dist_op*rate_op)+cost
_wqp_analyze))+((numb_enhance_wqp*pp_commercial_samp)*cost_lab_wqp))

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor and
material cost burdens for each type of analysis.

(((numb_enhance_wqp*pp_lab_samp)*((hrs_wqp_analyze_ph_op*rate_op)+cost
_wqp_ph_analyze))+((numb_enhance_wqp*pp_commercial_samp)*cost_lab_ph_

wqp))	

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor and
material cost burdens for each type of analysis.

(((numb_enhance_wqp*pp_lab_samp)*((hrs_wqp_analyze_ortho_op*rate_op)+co
st_wqp_ortho_analyze))+((numb_enhance_wqp*pp_commercial_samp)*cost_lab
ortho wqp))	

Final LCRI Economic Analysis

4-91

Conditions for Cost to Apply to a
Model PWS

Lead 90th -
Range

Other Conditions

Model PWSs serving
<50,000 without CCT and
have a copper ALE

p_copper_ale

Cost applies
as written to
NTNCWSs.

At or below
AL

Model PWSs serving
<10,000 that have pH
adjustment and a copper
ALE

Twice per
event

p_copper_ale, pbaseph

Model PWSs serving
<10,000 that have PCM or
both PCM and pH adjustment
and have a copper ALE

p_copper_ale, pbasepo4,
pbasephpo4	

October 2024


-------
CWS Cost Per Activity

cc) Collect copper WQP samples from entry points

The number of entry points per system multiplied by the number of samples, then
multiplied by the total of the labor hours per sample times the system labor rate,
plus the cost per sample.

((numb_ep*numb_ep_wqp)*((hrs_ep_wqp_op*rate_op)+cost_ep_wqp_material))

The number of entry points per system multiplied by the number of samples, then
multiplied by the total of the labor hours per sample times the system labor rate,
plus the cost per sample.

((numb_ep*numb_ep_wqp)*((hrs_ep_wqp_op*rate_op)+cost_ep_wqp_ph_materi
a!))

The number of entry points per system multiplied by the number of samples, then
multiplied by the total of the labor hours per sample times the system labor rate,
plus the cost per sample.

((numb_ep*numb_ep_wqp)*((hrs_ep_wqp_op*rate_op)+cost_ep_wqp_ortho_mat
erial))

Final LCRI Economic Analysis

4-92

Conditions for Cost to Apply to a
Model PWS

Lead 90th -
Range

Other Conditions

Cost applies
as written to
NTNCWSs.

At or below
AL

Model PWSs serving
<50,000 without CCT and
have a copper ALE

p_copper_ale

Model PWSs serving
<10,000 that have pH
adjustment and a copper
ALE

p_copper_ale, pbaseph

Model PWSs serving
<10,000 that have PCM or
both PCM and pH adjustment
and have a copper ALE

p_copper_ale, pbasepo4,
pbasephpo4	

Every 2
weeks per
event

October 2024


-------
CWS Cost Per Activity

dd) Analyze copper WQP samples from entry points

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor and
material cost burdens for each type of analysis.

((((numb_ep*numb_ep_wqp)*pp_lab_samp)*((hrs_wqp_analyze_ep_op*rate_op)

+cost_wqp_analyze_ep))+(((numb_ep*numb_ep_wqp)*pp_commercial_samp)*

cost_lab_wqp_ep))

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor and
material cost burdens for each type of analysis.

((((numb_ep*numb_ep_wqp)*pp_lab_samp)*((hrs_wqp_analyze_ph_ep_op*rate_
op)+cost_wqp_analyze_ph_ep))+(((numb_ep*numb_ep_wqp)*pp_commercial_sa
mp)*cost_lab_wqp_ph_ep))

The number of samples multiplied by the probabilities for a sample analyzed in
house and a sample analyzed in a commercial lab times the different labor and
material cost burdens for each type of analysis.

((((numb_ep*numb_ep_wqp)*pp_lab_samp)*((hrs_wqp_analyze_ortho_ep_op*rat
e_op)+cost_wqp_analyze_ortho_ep))+(((numb_ep*numb_ep_wqp)*pp_commerci
al_samp)*cost_lab_wqp_ortho_ep))

Final LCRI Economic Analysis

4-93

NTNCWS
Cost Per
Activity

Conditions for Cost to Apply to a
Model PWS

Lead 90th -
Range

Other Conditions

Frequency of
Activity

Model PWSs serving
<50,000 without CCT and
have a copper ALE

p_copper_ale

Cost applies
as written to
NTNCWSs.

At or below
AL

Model PWSs serving
<10,000 that have pH
adjustment and a copper
ALE

p_copper_ale, pbaseph

Every two
weeks per
event

Model PWSs serving
<10,000 that have PCM or
both PCM and pH adjustment
and have a copper ALE

p_copper_ale, pbasepo4,
pbasephpo4

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency of
Activity

ee) Report copper WQP sampling data and compliance with OWQPs to the State







Model PWSs serving
<50,000 without CCT and
have a copper ALE









p_copper_ale



The labor hours for reporting per system multiplied by the labor rate.
(hrs_report_wqp_op*rate_op)

Cost applies
as written to
NTNCWSs.

At or below
AL

Model PWSs serving
<10,000 that have pH
adjustment and a copper
ALE

p copper ale, pbaseph

Twice per
event







Model PWSs serving
<10,000 that have PCM or
both PO4 and pH adjustment
and have a copper ALE









p_copper_ale, pbasepo4,
pbasephpo4



Acronyms: AL = action level; ALE = action level exceedance; CCT = corrosion control treatment; CWS = community water system; NTNCWS = non-transient non-
community water system; PO4 = orthophosphate; OWQP = optimal water quality parameter; PWS = public water system; WQP = water quality parameter.

Note:

1The data variables in the exhibit are defined previously in this section with the exception of:

•	numb_enhance_wqp: number of distribution system samples for systems on standard WQP monitoring (Section 4.3.2.2.3).

•	numb_ep: number of entry points per systems (Section 4.3.2.2.3).

•	numb_ep_wqp\ number of entry points samples per systems (Section 4.3.2.2.3).

•	numb_reduced_wqp\ number of distribution system samples for systems on reduced WQP monitoring (Section 4.3.2.2.3).

•	pbaseplr. likelihood a system has an existing CCT of modify pH (Section 4.3.2.2.1).

•	pbasepo4: likelihood a system has existing CCT of adding PO4 without pH post treatment (Section 4.3.2.2.1).

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

Final LCRI Economic Analysis

4-94

October 2024


-------
4.3.2.4 PWS Source Water Monitoring

This discussion of source water monitoring costs for water systems is presented in the following
subsections:

•	4.3.2.4.1: Applicability and Required Number of Samples

•	4.3.2.4.2: Source Water Monitoring Activities

4.3.2.4.1	Applicability and Required Number of Samples

Under the final LCRI, CWSs and NTNCWSs must sample at each entry point if the system experiences a
significant source water change and/or has not already conducted source water monitoring for a
previous lead or copper ALE. The likelihood of a significant source change or ALE, as well as the required
number of source water samples, are described below.

Applicability

Section 3.3.9.1 in Chapter 3 provides the EPA's approach for using historical SDWIS/Fed data to estimate
the likelihood that systems would have a source change in any given year (p_source_chng) of 0.0388 for
all CWSs and 0.0278 percent for all NTNCWSs. The EPA developed a second related data input,
p_source_sig, using the same data set to estimate the likelihood that a source change would be a
significant change, i.e., one in which a system changed its primary source. The EPA set p_source_sig to
0.001 for CWSs and NTNCWSs. The likelihoods p_source_chng and p_source_sig are multiplied to
determine the joint likelihood that a system that makes a source change will be required to take
additional actions such as source water monitoring.

Lead and/or Copper ALE

Under the LCRI, the SafeWater LCR model assigns the source water monitoring burden and costs
described in Section 4.3.2.4.2 to any system the first time they exceed the lead and/or copper AL. A
discussion of the EPA's approach for estimating the likelihood a system will initially have a lead ALE
under the final LCRI is provided in Section 3.3.5.1 of Chapter 3, with the estimated percentages provided
in Exhibit 3-26. The likelihood a system will have a copper ALE is provided in Section 4.3.2.3.1. Note that
this approach may result in an overestimation of cost because the final LCRI allows systems to forego
source water monitoring if they previously sampled source water in response to an ALE, the State has
not required source water treatment, and they have not added any new water sources that change their
primacy source type. For modeling purposes no system is assumed to have source water treatment.

Number of Samples

The rule does not specify that systems must collect multiple samples per entry point. Thus, for the cost
model the agency assumed each system would collect one sample per entry point (numb_st_sample).
See Section 3.3.6.1 of Chapter 3 for a discussion of how the EPA estimated the number of entry points
per CWS and NTNCWS.

4.3.2.4.2	Source Water MonitorinR Activities

The EPA has developed system costs for three source water monitoring activities as shown in Exhibit
4-42. The exhibit provides the unit burden and/or cost for each activity. The assumptions used in the

Final LCRI Economic Analysis

4-95

October 2024


-------
estimation of each activity follows the exhibit. The last column provides the corresponding SafeWater
LCR model data variable in red/italic font. In a few instances, some of these activities are conducted by
the State instead of the water system. These activities are identified in the exhibit and further explained
in the exhibit notes.

Exhibit 4-42: PWS Source Monitoring Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

ff) Collect source water sample

Burden

0.5 hrs/sample

Burden

hrs_source_op



Cost

$1.12/sample for CWSs serving > 100K

Cost

cost_source_moter/o/2

gg) Analyze source water sample

In-House Burden

0.44 hrs/sample for CWSs serving > 100K

In-House Burden

hrs_analyze_samp_op1



In-House Cost

$3.92/sample for CWSs serving > 100K

In-House Cost
cost_source_analyze1



Commercial Cost

$31.00/sample for CWSs serving < 100K
and NTNCWSs

Commercial Cost
cos^source1

hh) Report source water

monitoring results to the State

1 hour/report

hrs_report_source_op1

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; PWS = public

water system.

Sources:

ff), hh): 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR (Renewal), Exhibit 15
(USEPA, 2022a); "Lead Analytical Burden and Costs_Final.xlsx," worksheets "Source_Collect_Analyze_CWS" and
"Source_Collect_Analyze_NTNCWS."

gg): See file "Lead Analytical Burden and Costs_Final.xlsx," worksheets "Source_Collect_Analyze_CWS" and

"Source_Collect_Analyze_NTNCWS."

Note:

1The burden and costs for these activities are incurred by the State in Arkansas, Louisiana, Mississippi, Missouri,
and South Carolina.

ff) Collect source water sample (hrs_source_op, cost_source_material). CWSs and NTNCWSs with a
significant source change and/or in response to their first lead or copper ALE will incur a burden of
0.5 hours to collect one source water sample at each entry point (hrs_source_op). This estimate is
based on the system burden for source water sample collection in the 2022

Disinfectants/Disinfection Byproducts, Chemical', and Radionuclides Rules ICR (Renewal), Exhibit 15
(USEPA, 2022a).

Based on input from laboratories, the EPA assumed only CWSs serving more than 100,000 people
will conduct analyses in-house, i.e., pp_lab_samp is 1 for CWSs serving more than 100,000 people
and 0 for all other CWSs and NTNCWSs. Conversely, the assigned likelihood a system will use a

Final LCR! Economic Analysis

4-96

October 2024


-------
commercial laboratory, i.e., pp_commercial_samp is 0 for CWSs serving more than 100,000 people
and 1 for all other systems.

Commercial laboratories provide bottles as part of their services. Thus, CWSs serving 100,000 or
fewer people and NTNCWSs will not incur bottle costs. CWSs serving more than 100,000 people are
assumed to purchase a 250-ml sample bottle in bulk at a per bottle cost of $1.12 based on quotes
from six vendors (cost_source_material).

gg) Analyze source water samples (hrs_analyze_samp_op, cost_source_analyze, cost_source). The EPA

assumed only CWSs serving more than 100,000 people will conduct analyses in-house and require
0.44 hours based on quotes from three laboratories. They will also incur in-house consumable costs
of $3.92 based on information from three vendors (cost_source_analyze). CWSs serving 100,000 or
fewer people and NTNCWSs will incur a cost of $31.00 per sample to have a commercial laboratory
conduct the lead analyses (cost_source). This includes $23.50 for the lead analysis based on quotes
from seven laboratories and a cost to ship a sample to the laboratory of $7.50.

hh) Report source water monitoring results to the State (hrs_report_source_op). Water systems are
required to report their source water monitoring results to the State. The EPA assumed that both
CWSs and NTNCWSs would report electronically and would not incur costs for paper, an envelope,
or postage. The agency did not have specific data on the time it would take to develop and submit a
report for the source water sampling results. Instead, in order to estimate this value, the EPA
employed the general assumption that a water system would require twice the time to prepare the
report as that needed for the State to review the report. The EPA used an estimate from Exhibit 48
of the 2022 Disinfectants/Disinfection Byproducts, Chemicaland Radionuclides Rules ICR (Renewal)
indicating a State burden of 0.5 hours for review of a "Source Water Monitoring Letter" submitted
from a water system (USEPA, 2022a). Therefore, the EPA assumed that systems would incur an
average burden of 1 hour to produce and submit this report for each monitoring period in which
they conduct source water monitoring (hrs_report_source_op).

Exhibit 4-43 provides the SafeWater LCR model cost estimation approach for system source water

monitoring activities including additional cost inputs required to calculate these costs.

Final LCRI Economic Analysis

4-97

October 2024


-------
Exhibit 4-43: PWS Source Water Monitoring Cost Estimation in SafeWater LCR by Activity1





Conditions for Cost to







Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th -
Range

Other Conditions2

Frequency
of Activity

ff) Collect source water sample3





All

Model PWSs with a
significant change in
source water

p_source_chng *
p source sig

Once per
event

The number of entry points per system
multiplied by the number of samples,
then multiplied by the total of the labor
hours per sample times the system labor
rate, plus the cost per sample.

((numb_ep*numb_st_sample)*((hrs_sour
ce_op*rate_op)+cost_source_material))

Cost applies as
written to
NTNCWSs.

At or below
AL

Model PWSs with a
copper ALE but no
lead ALE

p_copper_ale

One time





Above AL

All model PWSs



gg) Analyze source water samples3





All

Model PWSs with a
significant change in
source water

p_source_chng *
p source sig

Once per
event

The number of samples multiplied by the
probabilities for a sample analyzed in
house and a sample analyzed in a
commercial lab times the different labor
and material cost burdens for each type
of analysis.

((pp_lab_samp*(numb_ep*numb_st_sam

ple))*((hrs_analyze_samp_op*rate_op)+

cost_source_analyze))+((pp_commercial

_samp*(numb_ep*numb_st_sample))*co

st_source)

Cost applies as
written to
NTNCWSs.

At or below
AL

Model PWSs with a
copper ALE but no
lead ALE

p_copper_ale

One time





Above AL

All model PWSs



Final LCR! Economic Analysis

4-98

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th -
Range

Other Conditions2

Frequency
of Activity

hh) Report source water monitoring results to the State3





All

Model PWSs with a
significant change in
source water

p_source_chng *
p source sig

Once per
event

The total reporting hours per system
multiplied by the labor rate.

(hrs_report_source_op*rate_op)

Cost applies as
written to
NTNCWSs.

At or below
AL

Model PWSs with a
copper ALE but no
lead ALE

p_copper_ale

One time





Above AL

All model PWSs



Acronyms: AL = action level; ALE = action level exceedance; CWS = community water system; NTNCWS = non-

transient non-community water system; PWS = public water system.

Notes:

1The data variables in the exhibit are defined previously in this section with the exception of:

•	numb_ep: number of entry points per systems (Section 4.3.2.2.3).

•	numb_st_samp: number of samples per entry point for source water monitoring (Section 4.3.2.4.1).

•	p_copper_ale: likelihood a system with exceed the copper AL but not the lead AL (Section 4.3.2.3.1).

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

•	p_source_chng: Likelihood a system will have a source change (Chapter 3, Section 3.3.9.1).

•	p_source_sig: Likelihood that the system will have a significant change in which it changes its primary
source, e.g., for ground water to surface water (Chapter 3, Section 3.3.9.2).

2The likelihoods of p_source_chng and p_source_sig are multiplied to determine the joint likelihood that a system
that makes a source change will be required to conduct source water monitoring.

3The burden and cost to provide sample bottles (cost_source_material) under activity ff), conduct analyses under
activity gg), and report source water sample results to the system under activity hh) are incurred by the State in
Arkansas, Louisiana, Mississippi, Missouri, and South Carolina (ASDWA, 2020a).

4.3.2.5 CWS School and Child Care Facility Lead Sampling Costs

The final LCRI requires CWSs to implement a public education and lead in drinking water testing
program at public and private K-12 schools and licensed child care facilities. CWSs must collect five
samples per tested school (numb_samp_five) and two samples at each tested child care facility
(numb_samp_two). The rule splits this testing program into two phases. The first testing phase occurs at
elementary schools and child care facilities during the first 5 years of rule implementation, which is
assumed to occur in Years 4 through 8 of the 35-year period of analysis. During the first five-year cycle,
systems must schedule and conduct testing at 20 percent of elementary schools and 20 percent of child
care facilities (pp_mand_twenty_partic) per year such that each would be tested once in the five-year
period. CWSs may count any refusals or non-responses as part of the 20 percent. The final LCRI requires
water system to make at least two attempts to schedule sampling. The EPA assumed all elementary

Final LCRI Economic Analysis

4-99

October 2024


-------
schools and child facilities would accept sampling. CWSs are also required to annually provide secondary
schools with information on how to request sampling and must sample if requested by the school. In
Years 9 onward, CWSs are only required to test elementary schools, secondary schools, and child care
facilities that request testing. The EPA assumed 5 percent of elementary schools and child care facilities
would request testing each year, starting in Year 9 and 5 percent of secondary schools would request
testing each year, starting in Year 4 (pp_voluntary_partic).

Exhibit 3-57 in Chapter 3 provides the estimated number of public elementary schools, public secondary
schools, private elementary schools, private secondary schools, and child care facilities served by CWSs
for States, territories, and the Navajo Nation. Exhibit 3-58 is a continuation of Exhibit 3-57 and includes
the number of schools and child care facilities per CWS population served for each State. The SafeWater
LCR model applies the number of schools per CWS population served per State, to estimate the number
of:

•	Public elementary schools per system that corresponds to SafeWater LCR model data variable,
numb_elem_schools pub (see Column K of Exhibit 3-57 for public elementary schools per person
served by a CWS);

•	Private elementary schools per system that corresponds to SafeWater LCR model data variable,
numb_elem_schools priv (see Column N of Exhibit 3-57 for private elementary schools per
person served by a CWS);

•	Public secondary schools per system that corresponds to SafeWater LCR model data variable,
numb_second_schools pub (see Column J of Exhibit 3-57 for public secondary schools per person
served by a CWS);

•	Private secondary schools per system that corresponds to SafeWater LCR model data variable,
numb_elem_schools priv (see Column M of Exhibit 3-57 for private secondary schools per person
served by a CWS); and

•	Child care facilities per system that corresponds to SafeWater LCR model data variable,
numb_daycares (see Column O of Exhibit 3-57 for child care facilities per person served by a
CWS).

As previously discussed in Section 3.3.10.2, States with existing lead in drinking water programs at
schools and/or child care facilities that are at least as stringent as the final LCRI requirements can waive
these requirements for CWSs. In addition, CWSs can receive waivers to sample in schools and child care
facilities during the first 5-year testing cycle if the facility has been sampled between January 1, 2021,
and the LCRI compliance date. Exhibit 3-71 provides the percentage of elementary schools and child care
facilities for which CWSs will receive a waiver for the first five-year cycle or the entire testing program.
For additional detail, refer to Chapter 3, Section 3.3.10.2 and the file, "School_Child Care
lnputs_Final.xlsx" (available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov) for
additional information.

The requirements and associated costing inputs are described in more detail for the first testing cycle in
Section 4.3.2.5.1 and upon request program in Section 4.3.2.5.2.

Final LCRI Economic Analysis

4-100

October 2024


-------
4.3.2.5.1 First Five-Year TestinR Cycle

The EPA has developed system burden and costs to implement a lead in drinking water testing program
at elementary schools and child care facilities for the first five-year testing cycle as shown in Exhibit
4-44. The exhibit provides the unit burden and/or cost for each activity. The assumptions used in the
estimation of each activity follows the exhibit. The last column provides the corresponding SafeWater
LCR model data variable in red/italic font. In a few instances, some of these activities are conducted by
the State instead of the CWS. These activities are identified in the exhibit and further explained in the
exhibit notes.

Exhibit 4-44: CWS School and Child Care Facility Sampling Unit Burden and Cost Estimates for

the First Five-Year Testing Cycle (Years 4 - 8)

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

ii) Create a list of schools and child
care facilities served by the CWS
and submit to the State (one-
time)

0.08 hrs/school or child care
facility

hrs_school_identify_op

jj) Develop lead outreach materials
for schools and child care
facilities (one-time)

7 hrs/CWS

hrs_ devel_pe_school_ op

kk) Prepare and distribute initial
letters explaining the sampling
program and the EPA's 3Ts
Toolkit (one-time)

Burden

0.05 to 0.11 hrs/school or child
care facility

Cost

$0.47 to $0.72/ school or child
care facility

Burden

hrs_school_letter_op
Cost

cost_school_letter

II) Contact elementary school or
child care facility to determine
and finalize its sampling schedule
(one-time) or contact secondary
school to offer sampling (annual)

School

0.5 hrs/elementary school (one-
time)

0.05 to 0.11/secondary school
(annual)

School Cost

$0.47 to $0.72/secondary school

School

hrs_school_call_op (elementary)

hrs_school_annual_contact_op

(secondary)

Cost

cost_school_annual_contact
(secondary)



Child Care Facilitv
1 hr/child care facility

Child Care Facilitv
hrs childcare call op

mm) Contact school or child care
facility to coordinate sample
collection logistics

0.25 hrs/school or child care
facility

hrs_school_coor_sample_op

nn) Conduct walkthrough at school or
child care facility before the start
of sampling

Burden

1.40 to 1.71 hrs/school or child
care facility

Burden

hrs_ walkthrough_school_op



Cost

$5.75 to $10.24/school or child
care facility

Cost

cost_walkthrough_school

Final LCR! Economic Analysis

4-101

October 2024


-------
Activity

Unit Burden and/or Cost

SafeWaterLCR Data Variable

oo) Travel to collect samples

Burden

0.40 to 0.71 hrs/school or child
care facility

Cost

$5.75 to $10.24/school or child
care facility

Burden

hrs_travel_samp_school_op
Cost

cost_travel_samp_school

pp) Collect samples

Burden

0.17 hrs/sample
Cost

$1.12/sample for CWSs serving
> 100,000 people

Burden

hrs_collect_samp_school_op
Cost

cost_collect_samp_school1

qq) Analyze samples

In-House Analysis (CWSs > 100K
only)

Burden: 0.44 hrs/sample
Cost: $3.92/sample

Commercial Analysis
$31.00/sample

In-House Analysis
hrs_ an alyze_samp_ op1
costjabj^samp1

Commercial Analysis
cost commercial lab1

rr) Provide sampling results to tested
facilities

Burden

0.05 to 0.11 hrs/tested facility
Cost

$0.72/ tested facility

Burden

hrs_inform_samp_pe_school_op
Cost

cost_inform_samp_pe_school

ss) Discuss sampling results with the
school and child care facility

1 hr/school or child care facility

hrs_ result_ discuss_ op

tt) Conduct detailed discussion of
high sampling results with
schools and child care facilities

5 hr/sample

Burden

hrs_school_help_op

uu) Report school and child care
facility sampling results to the
State

Burden
12 hrs/CWS

Burden

hrs_report_sch_cc_results_op

vv) Prepare and provide annual
report on school and child care
facility sampling program to the
State

Burden

1 to 8 hrs/CWS
Cost

$0.72/CWS

Burden

hrs_annual_report_school_prepare_op
Cost

cost_ann ual_report_school_dist

Acronyms: AL = action level; 3Ts Toolkit = "3Ts for Reducing Lead in Drinking Water Toolkit"; CWS = community
water system; PWS = public water system.

Source: "School_Child Care lnputs_Final.xlsx." Other data sources are provided following this exhibit for each

activity, as applicable.

Note:

1The burden and costs for these activities are incurred by the State in Arkansas, Louisiana, Mississippi, Missouri,
and South Carolina.

ii) Create a list of schools and child care facilities served by CWS and submit to the State

(hrs_school_identify_op). The EPA assumed all CWSs would incur a burden at the start of the
program to create a contact list of schools and child care facilities in their service area and spend an

Final LCRI Economic Analysis

4-102

October 2024


-------
average of 5 minutes (0.08 hours) per school or child care facility. The EPA assumed a system can
use its customer database and obtain needed additional information online. Although systems
serving more than 10,000 people may spend less time to identify each facility, they are assumed to
use additional hours to create an electronic tracking system. Thus, the EPA also applied the 0.08
hours per facility to these larger systems.

jj) Develop lead outreach for schools and child care facilities (hrs_devel_pe_school_op). The EPA

assumed all CWSs would spend 7 hours to prepare outreach materials that describe the importance
of lead testing and the systems lead in drinking water testing program and submit these materials
for State review. The burden estimate of 7 hours is based on the hours to prepare additional
brochure language from Exhibit 33a of the 2022 Disinfectants/Disinfection Byproducts, Chemical,
and Radionuclides Rules ICR (Renewal) (USEPA, 2022a).

kk) Prepare and distribute the initial letters (hrs_school_letter_op, cost_school_letter). The EPA

assumed all CWSs would incur a one-time burden at the start of the program to prepare and
distribute an initial letter explaining the sampling program and providing a link to the EPA's "3Ts for
Reducing Lead in Drinking Water Toolkit" (3Ts Toolkit) (USEPA, 2018). The EPA estimated on average
systems serving 3,300 or fewer people would spend 1 hour per 9 letters (0.05 hours) and those
serving more than 3,300 people would spend 1 hour per 20 letters (0.11 hours) per school or child
care facility (hrs_school_letter_op). This estimate is based on the burden for a system to inform
customers of their lead testing results as documented in the 2022 Disinfectants/Disinfection
Byproducts, Chemicaland Radionuclides Rules ICR (Renewal), Exhibit 29 (Note G) (USEPA, 2022a).
Note that the EPA conservatively assumed that systems will send letters to every school. However,
the system may be able to send a letter to a single school district instead of individual schools as a
cost savings.

CWSs will also incur paper ($0,019), ink ($0.06), envelope ($0,092), and first class ($0.55) or bulk
rate postage costs ($0,299) to distribute the letter (cost_school_letter). CWSs serving 100,000 or
fewer people will incur total materials cost per letter of $0.72 and those serving more than 100,000
will incur a total cost of $0.47 due to the bulk postage rate discount.

II) Contact elementary school or child care facility to determine and finalize its sampling schedule
(hrs_school_call_op, hrs_childcare_call_op) and contact secondary schools to offer sampling
(hrs_school_annual_contact_op, cost_school_annual_contact). The EPA assumed CWSs would
coordinate with each elementary school or child care facility at the start of the program to plan
when each facility will be sampled. The EPA estimated CWSs would require two phone calls to reach
the appropriate person at an average of 15 minutes (0.25 hours) per call for a total of 0.5 hours per
school for this one-time activity. The EPA assumed CWSs would require additional time to contact
each child care facility at the start of the program to plan when each will be sampled. Some licensed
day cares are home-based facilities that may not have additional support staff and may require
multiple calls to reach the needed individual. The EPA estimated CWSs would require four calls at an
average of 15 minutes per call for a total of 1 hour for this one-time activity.

CWSs will send a letter to each secondary schools annually starting in Year 4 of the analysis period
explaining the sampling program, asking if the school wants to have their taps tested, and providing
health information on lead and a link to the EPA's 3Ts (USEPA, 2018). The EPA estimated on average
systems serving 3,300 or fewer people would spend 1 hour per 9 letters (0.05 hours) and those

Final LCRI Economic Analysis

4-103

October 2024


-------
serving more than 3,300 people would spend 1 hour per 20 letters (0.11 hours) per school
(hrs_school_letter_op). This estimate is based on the burden for a system to inform customers of
their lead testing results as documented in the 2022 Disinfectants/Disinfection Byproducts,

Chemical, and Radionuclides Rules ICR (Renewal), Exhibit 29 (Note G) (USEPA, 2022a). Note that the
EPA conservatively assumed that systems will send letters to every school. However, the system
may be able to send a letter to a single school district instead of individual schools as a cost savings.

CWSs will also incur paper ($0,019), ink ($0.06), envelope ($0,092), and first class ($0.55) or bulk
rate postage costs ($0,299) to distribute the letter (cost_school_letter). CWSs serving 100,000 or
fewer people will incur total materials cost per letter of $0.72 and those serving more than 100,000
will incur a total cost of $0.47 due to the bulk postage rate $0.47 per letter.

mm) Contact school or child care facility to coordinate sample collection logistics

(hrs_school_coor_sample_op). The EPA assumed CWSs will spend an average of 15 minutes (0.25
hours) calling each school or child care facility to coordinate sample collection logistics including
scheduling a walkthrough.

nn) Conduct walkthrough at school or child care facility before the start of sampling

(hrs_walkthrough_school_op, cost_walkthrough_school). The EPA assumed CWSs will conduct a
walkthrough with each school or child care facility to become familiar with the facility and to
identify sampling sites. The EPA assumed the following burden, which includes travel time roundtrip
to each facility plus one hour spent conducting the walkthrough (hrs_walkthrough_school_op) given
the equations below:

•	CWSs serving 100,000 or fewer people: 1.40 hours = ((5.0 miles * 2)/25 miles per hr) + 1 hr

•	CWSs serving 100,001 to 1,000,000 people: 1.51 hours = ((6.4 miles * 2)/25 miles per hr) + 1
hr

•	CWSs serving more than 1,000,000 people: 1.71 hours = ((8.9 miles * 2)/25 miles per hr) + 1
hr.

These estimates are based on census data and zip codes from the 2006 Community Water System
Survey, assumed the following one-way driving distances for CWSs: 5.0 miles for those serving <
100,000 people, 6.4 miles for those serving 100,001 - 1,000,000 people, and 8.9 miles for those
serving greater than 1,000,000 people. For additional detail on how these estimates were derived,
see "Estimated Driving Distances_Final.xlsx" EPA assumed an average speed of 25 miles per hour.

Systems will also incur travel costs to conduct this walkthrough (cost_walkthrough_school) as
follows:

•	CWSs serving 100,000 or fewer people: $5.75 = (5.0 miles * 2) * $0,575 per mile

•	CWSs serving 100,001 to 1,000,000 people: $7.36 = (6.4 miles * 2) * $0,575 per mile

•	CWSs serving more than 1,000,000 people: $10.24 = (8.9 miles * 2) * $0,575 per mile.

Final LCRI Economic Analysis

4-104

October 2024


-------
The EPA assumed a mileage cost of $0,575 per mile using the 2020 federal reimbursement rate from
the United States General Services Administration (GSA) (available in the docket at EPA-HQ-OW-
2022-0801 at www.regulations.gov).

oo) Travel to collect samples (hrs_travel_samp_school_op, cost_travel_samp_school). The EPA

assumed CWSs will incur burden to travel to a school or child care facility to collect samples
(hrs_travel_samp_school_op). The EPA assumed CWSs serving 100,000 or fewer people will spend
0.40 hours traveling round trip, those serving 100,001 to 1 million people will spend 0.51 hours, and
those serving more than 1 million people will spend 0.71 hours. The EPA used the same assumptions
as those used to develop hrs_walkthrough_school_op that is discussed in activity nn) above
excluding the 1 hour to conduct a walkthrough.

CWSs will also incur vehicle costs for this roundtrip travel (cost_travel_samp_school). The EPA used
the same assumptions as those for cost_walkthrough_school that is discussed in activity nn). The
EPA assumed the following costs: $5.75 for CWSs serving 100,000 or fewer people, $7.36 for those
serving 100,001 to 1 million people, and $10.24 for those serving more than 1 million people.

pp) Collect samples (hrs_collect_samp_school_op, cost_collect_samp_school). The final LCRI requires
CWSs to provide instructions to facilities on how to identify outlets for sampling at least 30 days
prior to sampling. For cost modeling purposes, the EPA assumed CWSs would collect the samples
and would require 10 minutes (0.17 hours) per sample (hrs_collect_samp_school_op). This estimate
is based on the assumption that the sample locations will be in the same building and the CWS has
previously conducted a walkthrough to identify sampling locations.

Based on information from laboratories, only CWSs serving more than 100,000 people are assumed
to conduct in-house analyses for lead; whereas those serving 100,000 or fewer people will use a
commercial lab. Bottles are supplied by the commercial lab. Thus, CWSs serving more than 100,000
people will incur a $1.12 per 250-mL wide mouth bottle based on the bulk discount costs from six
sources (refer to "Lead_WQP_Sample Bottle Costs_Final.xlsx" for additional detail).

qq) Analyze samples (hrs_analyze_samp_op, cost_lab_lt_samp, cost_commercial_lab). CWSs will incur
the same burden and cost to analyze the school and child care facility lead samples as they do
analyzing compliance lead tap samples. Therefore, the EPA used the same cost data variables for
both in-house and commercial laboratory analysis of lead tap samples. Specifically, CWSs serving
more than 100,000 people will incur a burden of 0.44 hours per sample (hrs_analyze_samp_op) and
a cost of $3.92 per sample (cost_lab_lt_samp) to analyze lead samples in-house. For these systems
the likelihood that a sample will be analyzed in-house is 1 (pp_lab_samp_school) and the likelihood
that the sample will be analyzed commercially is 0 (pp_commercial_samp_school). CWSs serving
100,000 or fewer will use a commercial lab at a cost of $23.50 per sample and a cost of $7.50 to ship
the sample to the lab for a total cost of $31.00 per sample (cost_commercial_lab). For these systems
pp_lab_samp_school is 0 and pp_commercial_samp_school is 1. See Section 4.3.2.1.2, activity k) for
additional detail.

rr) Provide sampling results to tested facilities (hrs_inform_samp_pe_school_op,

cost_inform_samp_pe_school). CWSs must provide sampling results to each tested facility. The EPA
assumed systems will spend 0.05 hours or 1 hour per 20 letters for systems serving 3,300 or fewer
people and 0.11 hours or 1 hour per 9 letter for systems serving more than 3,300 people

Final LCRI Economic Analysis

4-105

October 2024


-------
(hrs_inform_samp_pe_school_op). This estimate is based on the burden for a system to inform
customers of their lead testing results as documented in the 2022 Disinfectants/Disinfection
Byproducts, Chemicaland Radionuclides Rules ICR (Renewal), Exhibit 29 (Note G) (USEPA, 2022a).
The EPA also assumed CWSs will incur a material cost of $0.72/letter

(cost_inform_samp_pe_school). The EPA assumed information will be provided via mail (1 page of
information, double-sided). Material costs include paper ($0,019), ink ($0.06), envelope ($0,092),
and first class postage ($0.55). CWSs will provide the results of all testing to the State within 30 days
of receiving the results, which is discussed under activity uu) below. Systems must also provide
these results to local and State health departments within 30 days of receiving the results. The
burden and cost for this activity is captured in the data variables, hrs_hc_op and cost_hc (see
Section 4.3.6.2, activity I)).

ss) Discuss sampling results with the school and child care facilities (hrs_result_discuss_op). Although
not explicitly required under the final LCRI, the EPA assumed CWSs will incur additional burden to
discuss the sampling results with each school and child care facility at an average burden of 1 hour
per tested facility.

tt) Conduct detailed discussion of high sampling results with schools and child care facilities

(hrs_school_help_op). Although not explicitly required under the final LCRI, for each sample result
over the AL, the EPA assumed CWSs will spend approximately 5 hours discussing in greater detail
the sampling result(s) and the 3Ts Toolkit (USEPA, 2018). The estimate includes time for the system
to explain the relevant portions of the 3Ts Toolkit and to address any follow-up questions that the
school or child care facility might have after the initial discussion.

uu) Report school and child care facility sampling results to the State (hrs_report_sch_cc_results_op).

Under the final LCRI, CWSs will be required to provide school and child care facility testing results to
their State within 30 days of receiving the analytical results. The EPA assumed that CWSs will sample
a portion of schools and child care facilities each month and would require 1 hour each month. For
smaller CWSs with few schools and child care facilities, this may be an overestimation of the burden.
In addition, the EPA assumed systems would email the sampling results and incur no non-labor
costs.

vv) Prepare and provide an annual report on testing program to the State

(hrs_annual_report_school_prepare_op, cost_annual_report_school_dist). CWSs are required to
prepare and provide an annual report to the State regarding their testing program at schools and
child care facilities. The report certifies that the CWS made a good faith effort to identify all schools
and child care facilities they serve, summarizes all sampling activities conducted at schools and child
care facilities in a system's service area, and documents attempts that resulted in no response. Every
five years, the system must include any updates to the list of schools or child cares facilities or
confirmation of no change and is provided to the State. CWSs must keep documentation regarding
schools and child care facilities that are non-responsive or decline to participate in the testing
program. For modeling purposes, the EPA assumed all schools and child care facilities would elect to
participate in the testing program because the testing if free and they would want to better
understand their potential sources of lead in drinking water.

Final LCRI Economic Analysis

4-106

October 2024


-------
The EPA assumed systems would incur the following burden to prepare and distribute the annual
report (hrs_annual_report_school_prepare_op):

•	CWSs serving 100,000 or fewer people will incur a burden of 1 hour to prepare this report.
This effort is similar to the estimated burden for a system of this size to report lead tap
results and the 90th percentile calculation.

•	CWSs serving more than 100,000 people will be conducting sampling at a much larger
number of schools and child care facilities per year than smaller systems. CWSs serving more
than 100,000 people are assumed to have sophisticated tracking systems that can be used to
generate their reports. The EPA assumed systems serving 100,001 to 1,000,000 will require 2
hours and those serving more than 1,000,000 will require 8 hours to prepare the annual
report.

Systems also will incur mailing costs for paper ($0,019), ink ($0.06), envelope ($0,092), and first-class
postage ($0.55) to send a report to the State (cost_annual_report_school_dist). The material cost for
this report is $0.72.

Exhibit 4-45 provides the SafeWater LCR model cost estimation approach for each activity under the first
testing cycle including additional cost inputs required to calculate these costs.

Final LCR! Economic Analysis

4-107

October 2024


-------
Exhibit 4-45: CWS School and Child Care Facility First Five-Year Testing Cycle Cost Estimation in SafeWater LCR by Activity1,2





Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

ii) Create a list of schools and child care facilities and submit to the State

The number of public elementary schools per system population times
the system population multiplied by the hours to identify each facility
and the system labor rate.





All model PWSs



numb_elem_schools_pub*pws_pop*(hrs_school_identify_op*rate_op)









The number of public secondary schools per system population times
the system population multiplied by the hours to identify each school
and the system labor rate.





All model PWSs



numb_second_schools_pub*pws_pop*(hrs_school_identify_op*rate_o
p)









The number of child care facilities per system population times the
system population multiplied by the hours to identify each facility and
the system labor rate.

numb_daycares*p\NS_pop*(hrs_school_identify_op*rate_op)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

One time

The number of private elementary schools per system population
times the system population multiplied by the hours to identify each
facility and the system labor rate.





All model PWSs



numb_elem_schools_priv*pws_pop*(hrs_school_identify_op*rate_op)









The number of private secondary schools per system population times
the system population multiplied by the hours to identify each school
and the system labor rate.

numb second schools priv*pws pop*(hrs school identify op*rate o
P)





All model PWSs



Final LCR! Economic Analysis

4-108

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

jj) Develop lead outreach materials for schools and child care facilities

The hours per system to develop the lead outreach materials times the
system labor rate, plus the cost of materials.

(hrs_devel_pe_school_op*rate_op)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

One time

kk) Prepare and distribute initial letters explaining the sampling program and the 3Ts Toolkit

The number of public elementary schools per system population times
the system population multiplied by the hours per school times the
system labor rate, plus the cost of materials.





All model PWSs



numb_elem_schools_pub*pws_pop*((hrs_school_letter_op*rate_op)+c
ost_school_letter)









The number of public secondary schools per system population times
the system population multiplied by the hours per school times the
system labor rate, plus the cost of materials.





All model PWSs



numb_second_schools_pub*pws_pop*((hrs_school_letter_op*rate_op)
+cost_school_letter)







One time

The number of child care facilities requirements per system population
times the system population multiplied by the hours per facility times
the system labor rate, plus the cost of materials.

Cost does not
apply to
NTNCWSs.

All

All model PWSs



numb_daycares*pws_pop*((hrs_school_letter_op*rate_op)+cost_scho
ol_letter)









The number of private elementary schools per system population
times the system population multiplied by the hours per school times
the system labor rate, plus the cost of materials.





All model PWSs



numb_elem_schools_priv*pws_pop*((hrs_school_letter_op*rate_op)+c
ost_school_letter)









Final LCRI Economic Analysis

4-109

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

The number of private secondary schools per system population times
the system population multiplied by the hours per school times the
system labor rate, plus the cost of materials.

numb_second_schools_priv*pws_pop*((hrs_school_letter_op*rate_op)
+cost_school_letter)





All model PWSs



II) Contact elementary school or child care facility to determine and finalize its sampling schedule and contact secondary school to
offer sampling

The number of public elementary schools that do not meet waiver
requirements per system population times the system population
multiplied by the hours per school times the system labor rate, plus the
cost of materials.





All model PWSs

One time

pp_pub_elem_mand_waiver

*numb elem schools pub*pws pop*(hrs school call op*rate op)









The number of public secondary schools that do not meet waiver
requirements per system population times the system population
multiplied by the hours per school times the system labor rate, plus the
cost of materials.

pp_pub_second_onreq1_waiver
*numb_second_schools_pub*pws_pop*
((hrs_school_annual_contact_op *rate_op)+
cost_school_annual_contact)

Cost does not
apply to
NTNCWSs.

All

All model PWSs



The number of child care facilities that do not meet waiver
requirements per system population times the system population
multiplied by the hours per facility and the system labor rate.

pp_childcare_mand_waiver
*numb_daycares*(hrs_childcare_call_op*rate_op)





All model PWSs

One time

Final LCRI Economic Analysis

4-110

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

The number of private elementary schools that do not meet waiver
requirements per system population times the system population
multiplied by the hours per school times the system labor rate, plus the
cost of materials.





All model PWSs



pp_priv_elem_mand_waiver

*numb_elem_schools_priv*pws_pop*(hrs_school_call_op*rate_op)









The number of private secondary schools that do not meet waiver
requirements per system population times the system population
multiplied by the hours per school times the system labor rate, plus the
cost of materials.

pp_priv_second_onreq1jwaiver
*numb_second_schools_pub*pws_pop*
((hrs_school_annual_contact_op *rate_op)+
cost_school_annual_contact)





All model PWSs

Once a
year

mm) Contact school or child care facility to coordinate sample collection logistics

20% of public elementary schools that do not meet waiver
requirements multiplied by the hours per school and the system labor
rate.

(pp_pub_elem_mand_waiver

*numb_elem_schools_pub*pws_pop*pp_mand_twenty_partic)*(hrs_sc
hool coor sample op*rate op)





All model PWSs



5% of public secondary schools that do not meet waiver requirements
multiplied by the hours per school and the system labor rate.

(pp_pub_second_onreq1_waiver
*numb_second_schools_pub *pws_pop *
pp_voluntary_partic)*(hrs_school_coor_sample_op*rate_op)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a
year

20% of child care facilities that do not meet waiver requirements
multiplied by the hours per facility and the system labor rate.

(pp_childcare_mand_waiver

*numb_daycares*pws_pop*pp_mand_twenty_partic)*(hrs_school_coor
_sample_op*rate_op)





All model PWSs



Final LCRI Economic Analysis

4-111

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

20% of private elementary schools that do not meet waiver
requirements multiplied by the hours per school and the system labor
rate.

(pp_priv_elem_mand_waiver

*numb_elem_schools_priv*pws_pop*pp_mand_twenty_partic)*(hrs_sc
hool coor sample op*rate op)





All model PWSs



5% of private secondary schools that do not meet waiver requirements
multiplied by the hours per school and the system labor rate.









(pp_priv_second_onreq1jwaiver

*numb_second_schools_priv*pws_pop*

pp_voluntary_partic)*(hrs_school_coor_sample_op*rate_op)





All model PWSs



nn) Conduct walkthrough at school or child care facility before start of sampling

20% of public elementary schools that do not meet waiver
requirements multiplied by the hours per school times the system labor
rate, plus the cost of materials.

(pp_pub_elem_mand_waiver

*numb_elem_schools_pub*pws_pop*pp_mand_twenty_partic)*((hrs_w
alkthrough_school_op*rate_op)+cost_walkthrough_school)





All model PWSs



5% of public secondary schools that do not meet waiver requirements
multiplied by the hours per school times the system labor rate, plus the
cost of materials.

(pp_pub_second_onreq1_waiver

*numb_second_schools_pub*pws_pop*pp_voluntary_partic)*((hrs_wal
kthrough_school_op*rate_op)+cost_walkthrough_school)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a
year

20% of child care facilities that do not meet waiver requirements
multiplied by the hours per facility times the system labor rate, plus the
cost of materials.

(pp_childcare_mand_waiver

*numb_daycares*pws_pop*pp_mand_twenty_partic)*((hrs_walkthroug
h_school_op*rate_op)+cost_walkthrough_school)





All model PWSs



Final LCRI Economic Analysis

4-112

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

20% of private elementary schools that do not meet waiver
requirements multiplied by the hours per school times the system labor
rate, plus the cost of materials.

(pp_priv_elem_mand_waiver

*numb_elem_schools_priv*pws_pop*pp_mand_twenty_partic)*((hrs_w
alkthrough_school_op*rate_op)+cost_walkthrough_school)





All model PWSs



5% of private secondary schools that do not meet waiver requirements
multiplied by the hours per school times the system labor rate, plus the
cost of materials.

(pp_priv_second_onreq1jwaiver

*numb_second_schools_priv*pws_pop*pp_voluntary_partic)*((hrs_wal
kthrough_school_op*rate_op)+cost_walkthrough_school)





All model PWSs



oo) Travel to collect samples

20% of public elementary schools that do not meet waiver
requirements multiplied by the hours per school times the system labor
rate, plus the cost of materials.

(pp_pub_elem_mand_waiver

*numb_elem_schools_pub *pws_pop*pp_mand_twenty_partic) *( (hrs_tr
avel_samp_school_op*rate_op)+cost_travel_samp_school)





All model PWSs



5% of public secondary schools that do not meet waiver requirements
multiplied by the hours per school times the system labor rate, plus the
cost of materials.

(pp_pub_second_onreq1_waiver

*numb_second_schools_pub*pws_pop*pp_voluntary_partic)*((hrs_trav
el_samp_school_op*rate_op)+cost_travel_samp_school)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a
year

Final LCRI Economic Analysis

4-113

October 2024


-------


Conditions for Cost to Apply to a Model PWS

CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

20% of child care facilities that do not meet waiver requirements
multiplied by the hours per facility times the system labor rate, plus the
cost of materials.

(pp_childcare_mand_waiver

*numb_daycares*pp_mand_twenty_partic)*((hrs_travel_samp_school_
op*rate_op)+cost_travel_samp_school)





All model PWSs



20% of private elementary schools that do not meet waiver
requirements multiplied by the hours per school times the system labor
rate, plus the cost of materials.

(pp_priv_elem_mand_waiver

*numb_elem_schools_priv*pws_pop*pp_mand_twenty_partic)*((hrs_tr
avel_samp_school_op*rate_op)+cost_travel_samp_school)





All model PWSs



5% of private secondary schools that do not meet waiver requirements
multiplied by the hours per school times the system labor rate, plus the
cost of materials.

(pp_priv_second_onreq1jwaiver

*numb_second_schools_priv*pws_pop*pp_voluntary_partic)*((hrs_trav
el_samp_school_op*rate_op)+cost_travel_samp_school)





All model PWSs



Final LCRI Economic Analysis

4-114

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

pp) Collect samples3

20% of public elementary schools that do not meet waiver
requirements multiplied by the number of samples per school, is
multiplied by the number of hours per school times the system labor
rate, plus the material cost.

(pp_pub_elem_mand_waiver

*numb_elem_schools_pub*pws_pop*pp_mand_twenty_partic)*numb_
samp_five)*((hrs_collect_samp_school_op*rate_op)+cost_collect_sam
p school)





All model PWSs



5% of public secondary schools that do not meet waiver requirements
multiplied by the number of samples per school, is multiplied by the
number of hours per school times the system labor rate, plus the
material cost.

(pp_pub_second_onreq1_waiver

*numb_second_schools_pub*pws_poprpp_voluntary_partic)*numb_s

amp_five)*((hrs_collect_samp_school_op*rate_op)+cost_collect_samp

_school)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a
year

20% of child care facilities that do not meet waiver requirements
multiplied by the number of samples per facility, is multiplied by the
number of hours per facility times the system labor rate, plus the
material cost.





All model PWSs



(pp_childcare_mand_waiver

*numb_daycares*pp_mand_twenty_partic)*numb_samp_two)*((hrs_co
llect_samp_school_op*rate_op)+cost_collect_samp_school)









20% of private elementary schools that do not meet waiver
requirements multiplied by the number of samples per school, is
multiplied by the number of hours per school times the system labor
rate, plus the material cost.

(pp_priv_elem_mand_waiver

*numb_elem_schools_priv*pws_pop*pp_mand_twenty_partic)*numb_
samp_five)*((hrs_collect_samp_school_op*rate_op)+cost_collect_sam
p school)





All model PWSs



Final LCRI Economic Analysis

4-115

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

5% of private secondary schools that do not meet waiver requirements
multiplied by the number of samples per school, is multiplied by the
number of hours per school times the system labor rate, plus the
material cost.

(pp_priv_second_onreq1jwaiver

*numb_second_schools_priv*pws_popl*pp_voluntary_partic)*numb_sa
mp_five)*((hrs_collect_samp_school_op*rate_op)+cost_collect_samp_
school)

Cost does not
apply to
NTNCWSs.

All

All model PWSs



qq) Analyze samples3

The number of required samples per public elementary school
multiplied by 20% of elementary schools that do not meet waiver
requirements per year times by the probabilities for a sample analyzed
in house and a sample analyzed in a commercial lab times the
different labor and material cost burdens for each type of analysis.









((((pp_pub_elem_mand_waiver

*numb_elem_schools_pub*pws_pop*pp_mand_twenty_partic)*numb_
samp_five)*pp_lab_samp_school) ((hrs_analyze_samp_op*rate_op)+
cost_lab_lt_samp))+((((pp_pub_elem_mand_waiver
*numb_elem_schools_pub*pws_pop*pp_mand_twenty_partic)*numb_
samp_five)*pp_commercial_samp_school)*cost_commercial_lab)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a
year

The number of required samples per public secondary school
multiplied by 5% of secondary schools that do not meet waiver
requirements per year times by the probabilities for a sample analyzed
in house and a sample analyzed in a commercial lab times the
different labor and material cost burdens for each type of analysis.

((((pp_pub_second_onreq 1_ wai ver

*numb_second_schools_pub*p\NS_pop*pp_voluntary_partic)*numb_sa
mp_five)*pp_lab_samp_school)*((hrs_analyze_samp_op*rate_op)+co
st_lab_lt_samp))+((((pp_pub_second_onreq1_waiver
*numb_second_schools_pub*p\NS_pop*pp_voluntary_partic)*numb_sa
mp five)*pp commercial samp school)*cost commercial lab)





All model PWSs



Final LCRI Economic Analysis

4-116

October 2024


-------




Conditions for Cost to Apply to a Model PWS

CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

The number of required samples per child care facility multiplied by
20% of elementary schools that do not meet waiver requirements per
year times by the probabilities for a sample analyzed in house and a
sample analyzed in a commercial lab times the different labor and
material cost burdens for each type of analysis.

Cost does not







((((pp_childcare_mand_waiver

*numb_daycares*pws_pop*pp_mand_twenty_partic)*numb_samp_five
)*pp_lab_samp_school)*((hrs_analyze_samp_op*rate_op)+cost_lab_lt
_samp))+((((p_childcare_mand_waiver

*numb_daycares*pws_pop*pp_mand_twenty_partic)*numb_samp_five
)*pp commercial samp school)*cost commercial lab)

apply to
NTNCWSs.

All

All model PWSs



The number of required samples per private elementary school
multiplied by 20% of elementary schools that do not meet waiver
requirements per year times by the probabilities for a sample analyzed
in house and a sample analyzed in a commercial lab times the
different labor and material cost burdens for each type of analysis.









((((pp_priv_elem_mand_waiver

*numb_elem_schools_priv*pws_pop*pp_mand_twenty_partic)*numb_
samp_five)*pp_lab_samp_school)*((hrs_analyze_samp_op*rate_op)+
cost_lab_lt_samp))+((((pp_priv_elem_mand_waiver
*numb_elem_schools_priv*pws_pop

*pp_mand_twenty_partic) *numb_samp_Hve)*pp_commercial_samp_s
chool)*cost commercial lab)

Cost does not
apply to
NTNCWSs.

All

All model PWSs



The number of required samples per private secondary school
multiplied by 5% of secondary schools that do not meet waiver
requirements per year times by the probabilities for a sample analyzed
in house and a sample analyzed in a commercial lab times the
different labor and material cost burdens for each type of analysis.









((((pp_priv_second_onreq1_waiver
*numb_second_schools_priv*pws_pop

*pp_voluntary_partic)*numb_samp_five)*pp_lab_samp_school)*((hrs_
analyze_samp_op*rate_op)+cost_lab_lt_samp))+((((pp_priv_second_o
nreq1_waiver *numb_second_schools_priv*pws_pop
*pp_voluntary_partic)*numb_samp_five)*pp_commercial_samp_schoo
l)*cost commercial lab)





All model PWSs



Final LCRI Economic Analysis

4-117

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

rr) Provide sampling results to tested facilities

20% of public elementary schools that do not meet waiver
requirements multiplied by the hours per school times the system labor
rate, plus the cost of materials.

(pp_pub_elem_mand_waiver

*numb_elem_schools_pub *pws_pop*pp_mand_twenty_partic) *( (hrsjn
form_samp_pe_school_op*rate_op)+cost_inform_samp_pe_school)

Cost does not
apply to
NTNCWSs.

All

All model PWSs



5% of public secondary schools that do not meet waiver requirements
multiplied by the hours per school times the system labor rate, plus the
cost of materials.

(pp_pub_second_onreq1_waiver

*numb_second_schools_pub*pws_pop*pp_voluntary_partic)*((hrs_info
rm samp pe school op*rate op)+cost inform samp pe school)





All model PWSs



20% of child care facilities that do not meet waiver requirements
multiplied by the hours per facility times the system labor rate, plus the
cost of materials.

(pp_childcare_mand_waiver

*numb_daycares*pws_pop *pp_mand_twenty_partic) *( (hrs_inform_sa
mp_pe_school_op*rate_op)+cost_inform_samp_pe_school)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a
year

20% of private elementary schools that do not meet waiver
requirements multiplied by the hours per school times the system labor
rate, plus the cost of materials.

(pp_priv_elem_man d_ wai ver

*numb_elem_schools_priv*pws_pop*pp_mand_twenty_partic)*((hrs_in
form_samp_pe_school_op*rate_op)+cost_inform_samp_pe_school)





All model PWSs



5% of private secondary schools that do not meet waiver requirements
multiplied by the hours per school times the system labor rate, plus the
cost of materials.

(pp_priv_second_onreq1jwaiver

*numb_second_schools_priv*pws_pop*pp_voluntary_partic)*((hrs_info
rm_samp_pe_school_op*rate_op)+cost_inform_samp_pe_school)





All model PWSs



Final LCRI Economic Analysis

4-118

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

ss) Discuss sampling results with the school and child care facilities

20% of public elementary schools that do not meet waiver
requirements multiplied by the hours per school times the system labor
rate.

(pp_pub_elem_mand_waiver

*numb_elem_schools_pub*pws_pop*pp_mand_twenty_partic)*(hrs_re
suit discuss op*rate op)





All model PWSs



5% of public secondary schools that do not meet waiver requirements
multiplied by the hours per school times the system labor rate.









(pp_pub_second_onreq1_waiver
*numb_second_schools_pub *pws_pop *pp_mand_
voluntary partic)*(hrs result discuss op*rate op)





All model PWSs



20% of child care facilities that do not meet waiver requirements
multiplied by the hours per facility times the system labor rate.

Cost does not





Once a
year

(pp_childcare_mand_waiver

*numb_daycares*pws_pop*pp_mand_twenty_partic)*(hrs_result_discu
ss_op*rate_op)

apply to
NTNCWSs.

All

All model PWSs

20% of private elementary schools that do not meet waiver
requirements multiplied by the hours per school times the system labor
rate.

(pp_priv_elem_mand_waiver

*numb_elem_schools_priv*pws_pop*pp_mand_twenty_partic)*(hrs_re
sult_discuss_op*rate_op)





All model PWSs



5% of private secondary schools that do not meet waiver requirements
multiplied by the hours per school times the system labor rate.









(pp_priv_second_onreq1jwaiver

*numb_second_schools_priv*pws_pop*pp_voluntary_partic)*(hrs_resu
lt_discuss_op*rate_op)





All model PWSs



Final LCRI Economic Analysis

4-119

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

tt) Conduct detailed discussion of high sampling results with school and child care facilities

20% of public elementary schools that do not meet waiver
requirements multiplied by the number of required samples per system
above the AL multiplied by the hours per sample times the system
labor rate.

(pp_pub_elem_mand_waiver

*numb_elem_schools_pub*pws_pop*pp_mand_twenty_partic)*
((pp_above_al_bin_three*numb_samp_five)*(hrs_school_help_op*rate
QP))	

5% of public secondary schools that do not meet waiver requirements
multiplied by the number of required samples per system above the AL
multiplied by the hours per sample times the system labor rate.

(pp_pub_second_onreq1_waiver

*numb_second_schools_pub*pws_pop*pp_voluntary_partic)*
((pp_above_al_bin_three*numb_samp_five)*(hrs_school_help_op*rate
QP))	

20% of child care facilities that do not meet waiver requirements
multiplied by the number of required samples per system above the AL
multiplied by the hours per sample times the system labor rate.

(pp_childcare_mand_waiver

*numb_daycares*pws_pop*pp_mand_twenty_partic)*
((pp_above_al_bin_three*numb_samp_five)*(hrs_school_help_op*rate
QP))	

20% of private elementary schools that do not meet waiver
requirements multiplied by the number of required samples per system
above the AL multiplied by the hours per sample times the system
labor rate.

(pp_priv_elem_man d_ wai ver

*numb_elem_schools_priv*pws_pop*pp_mand_twer>ty_partic)*
((pp_above_al_bin_three*numb_samp_five)*(hrs_school_help_op*rate
QP))	

Cost does not
apply to
NTNCWSs.

At or
below the
AL

All model PWSs

All model PWSs

All model PWSs

All model PWSs

Once a
year

Final LCRI Economic Analysis

4-120

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

5% of private secondary schools that do not meet waiver requirements
multiplied by the number of required samples per system above the AL
multiplied by the hours per sample times the system labor rate.









pp_priv_second_onreq1jwaiver

*numb_second_schools_priv*pws_pop* pp_voluntary_partic)*
((pp_above_al_bin_three*numb_samp_five)*(hrs_school_help_op*rate
op))





All model PWSs



20% of public elementary schools that do not meet waiver
requirements multiplied by the number of required samples per system
above the AL multiplied by the hours per sample times the system
labor rate.

(pp_pub_elem_mand_waiver

*numb_elem_schools_pub*pws_pop*pp_mand_twenty_partic)*

((pp_above_al_bin_one*numb_samp_five)*(hrs_school_help_op*rate_

op))





All model PWSs



5% of public secondary schools that do not meet waiver requirements
multiplied by the number of required samples per system above the AL
multiplied by the hours per sample times the system labor rate.

(pp_pub_second_onreq1_waiver

*numb_second_schools_pub*pws_pop* pp_voluntary_partic)*

((pp_above_al_bin_one*numb_samp_five)*(hrs_school_help_op*rate_

op))

Cost does not
apply to
NTNCWSs.

Above the
AL

All model PWSs

Once a
year

20% of child care facilities that do not meet waiver requirements
multiplied by the number of required samples per system above the AL
multiplied by the hours per sample times the system labor rate.









(pp_childcare_mand_waiver

*numb_daycares*pws_pop*pp_mand_twenty_partic)*
((pp_above_al_bin_one*numb_samp_five) (hrs_school_help_op*rate_
op))





All model PWSs



Final LCRI Economic Analysis

4-121

October 2024


-------




Conditions for Cost to Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

20% of private elementary schools that do not meet waiver
requirements multiplied by the number of required samples per system
above the AL multiplied by the hours per sample times the system
labor rate.

(pp_priv_elem_mand_waiver

*numb_elem_schools_priv*pws_pop*pp_mand_twenty_partic)*

((pp_above_al_bin_one*numb_samp_five)*(hrs_school_help_op*rate_

op))





All model PWSs



5% of private secondary schools that do not meet waiver requirements
multiplied by the number of required samples per system above the AL
multiplied by the hours per sample times the system labor rate.









(pp_priv_second_onreq1jwaiver

*numb_second_schools_priv*pws_pop* pp_voluntary_partic)*

((pp_above_al_bin_one*numb_samp_five)*(hrs_school_help_op*rate_

op))





All model PWSs



uu) Report school and child care facility sampling results to the State

The total hours per system multiplied by the system labor rate.

Cost does not







(hrs_report_sch_cc_results_op*rate_op)

apply to
NTNCWSs.

All

All model PWSs

Once a year

vv) Prepare and provide annual report on school and child care facility sampling to State

The total hours per system multiplied by the system labor rate, plus the
materials cost.

(hrs_annual_report_school_prepare_op*rate_op)+cost_annual_report
_school_dist

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a
year

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; PWS = public water system.
Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of:

• numb_daycares, numb_elem_schools_priv, numb_elem_school_pub, numb_second_schools_priv, numb_second_school_pub: Number of child care
facilities, number of private elementary schools, number of public elementary schools, number of public secondary schools, and number of private
secondary schools, respectively that are served by CWSs (Section 3.3.10.1).

Final LCRI Economic Analysis

4-122

October 2024


-------
•	pp_childcare_mand_waiver, pp_priv_elem_mand_waiver, pp_pub_elem_mand_waiver, pp_pub_second_onreql_waiver,
pp_priv_second_onreql_waiver. States that qualify to waiver child care facilities, private K-12 schools, and public K-12 schools for the first testing
phase (Section 3.3.10.2).

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

2The first testing cycle is assumed to occur in Years 4 through 8 at elementary schools and child care facilities.

3 The burden and costs to provide sample bottles (cost_collect_samp_school) under activity pp) and conduct analyses under activity qq) are incurred by the
State in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina (ASDWA, 2020a).

Final LCRI Economic Analysis

4-123

October 2024


-------
4,3.2.5,2 Second Five-Year Testing Cycle On

Under the final LCRI, after CWSs complete one 5-year cycle of testing at elementary schools and child
care facilities, testing at these facilities is on request only. In addition, CWSs are only required to test
those secondary schools that request testing. The EPA assumed that 5 percent of elementary and
secondary schools, and licensed child care facilities per year would elect to participate in the sampling
program (pp_voluntary_partic). This estimate is based on the EPA's discussions with GCWW about their
school testing program (available in the docket at EPA-HQ-OW-2022-0801). GCWW indicated that they
had a low response rate from schools under their initial program that involved sending out letters to
school districts offering to assist schools in testing their drinking water outlets for lead, which is similar
to the on request program requirements.

The EPA has developed system burden and costs for 12 activities the agency has identified as necessary
to implement the on request program for drinking water testing at schools and child care facilities as
shown in Exhibit 4-46. The exhibit provides the unit burden and/or cost for each activity. The
assumptions used in the estimation of each activity follows the exhibit. The last column provides the
corresponding SafeWater LCR model data variable in red/italic font. In a few instances, some of these
activities are conducted by the State instead of the CWS. These activities are identified in the exhibit and
further explained in the exhibit notes.

Exhibit 4-46: CWS School and Child Care Facility Sampling Unit Burden and Cost Estimates

under the Second Five-Year Testing Cycle On

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

ww) Update the list of

schools and child care
facilities and submit to
the State (every five
years)

0.08 hrs/school or child care facility

hrs_sch ool_ iden tify_ op

xx) Contact school and child
care facilities to offer
sampling

Burden

0.05 to 0.11 hrs/school or child care
facility

Cost

$0.47 to $0.72

Burden

hrs_school_annual_contact_op
Cost

cost_school_annual_contact

yy) Contact the school or
child care facility to
coordinate sample
collection logistics

0.25 hrs/school or child care facility

hrs_school_coor_sample_op

zz) Conduct walkthrough at
school or child care
facility before the start
of sampling

Burden

1.40 to 1.71 hrs/school or child care
facility

Cost

$5.75 to $10.24/school or child care
facility

Burden

hrs_ walkthrough_school_op
Cost

cost_walkthrough_school

Final LCRI Economic Analysis

4-124

October 2024


-------
Activity

Unit Burden and/or Cost

SafeWaterLCR Data Variable

aaa) Travel to collect samples

Burden

0.40 to 0.71 hrs/school or child care
facility

Cost

$5.75 to $10.24/school or child care
facility

Burden

hrs_travel_samp_school_op
Cost

cost_travel_samp_school

bbb) Collect samples

Burden

0.17 hrs/sample
Cost

$1.12/sample for CWSs serving >
100K

Burden

hrs_ collect_samp_school_ op
Cost

cost_collect_samp_school1

ccc) Analyze samples

In-house Analysis (CWSs > 100K
only)

Burden: 0.44 hrs/sample
Cost: $3.92/sample

Commercial Analysis
$31.00/sample

In-House Analysis
hrs_ an alyze_samp_ op1
costjabj^samp1

Commercial Analysis
cost commercial lab1

ddd) Provide sampling results
to tested facilities

Burden

0.05 to 0.11 hrs/tested facility
Cost

$0.72/ tested facility

Burden

hrs_inform_samp_pe_school_op
Cost

cost_inform_samp_pe_school

eee) Discuss sampling results
with the school and child
care facility

1 hr/school or child care facility

hrs_ result_ discuss_ op

fff) Conduct detailed
discussion of high
sampling results with
schools and child care
facilities

5 hr/sample

Burden

hrs_school_help_op

ggg) Report school and child
care facility sampling
results to the State

Burden
12 hrs/CWS

Burden

hrs_report_sch_cc_results_op

hhh) Prepare and provide

annual report on school
and child care facility
sampling program to the
State

Burden

1 to 8 hrs/CWS
Cost

$0.72/CWS

Burden

hrs_annual_report_school_prepare_op
Cost

cost_ann ual_report_school_dist

Acronyms: AL = action level; CWS = community water system; PWS = public water system.

Source: "School_Child Care lnputs_Final.xlsx." Other data sources are provided following this exhibit for each

activity, as applicable.

Note:

1The burden and costs for these activities are incurred by the State in Arkansas, Louisiana, Mississippi, Missouri,
and South Carolina.

ww) Update the list of schools and child care facilities and submit to the State

(hrs_school_identify_op). The EPA assumed all CWSs would incur a burden every five years to

Final LCRI Economic Analysis

4-125

October 2024


-------
update the contact list of schools and child care facilities in their service area and spend an average
of 5 minutes (0.08 hours) per school or child care facility. The EPA assumed a system can use its
customer database and obtain needed additional information online. Although systems serving
more than 10,000 people may spend less time to identify each facility, they are assumed to use
additional hours to create an electronic tracking system. Thus, the EPA also applied the 0.08 hours
per facility to these larger systems.

xx) Contact schools and child care facilities to offer sampling (hrs_school_annual_contact_op,
cost_school_annual_contact). CWSs will annually send a letter to each public and private
elementary and secondary schools and child care facility explaining the sampling program, asking if
the school/child care facility wants to have their taps tested, and providing health information on
lead and a link to the 3Ts (USEPA, 2018). The EPA estimated on average systems serving 3,300 or
fewer people would spend 1 hour per 9 letters (0.05 hours) and those serving more than 3,300
people would spend 1 hour per 20 letters (0.11 hours) per school or child care facility
(hrs_school_letter_op). This estimate is based on the burden for a system to inform customers of
their lead testing results as documented in the 2022 Disinfectants/Disinfection Byproducts,
Chemical, and Radionuclides Rules ICR (Renewal), Exhibit 29 (Note G) (USEPA, 2022a). Note that the
EPA conservatively assumed that systems will send letters to every school. However, the system
may be able to send a letter to a single school district instead of individual schools as a cost savings.

CWSs will also incur paper ($0,019), ink ($0.06), envelope ($0,092), and first class ($0.55) or bulk
rate postage costs ($0,299) to distribute the letter (cost_school_letter). CWSs serving 100,000 or
fewer people will incur total materials cost per letter of $0.72 and those serving more than 100,000
will incur a total cost of $0.47 due to the bulk postage rate $0.47 per letter.

yy) Contact the school or child care facility to coordinate sample collection logistics

(hrs_school_coor_sample_op). The EPA assumed CWSs would incur the same average burden as
under the first testing cycle to call each facility to coordinate sample collection logistics including
scheduling a walkthrough. The average time spent per call to coordinate sample collection logistics
is 15 minutes (0.25 hours). See Section 4.3.2.5.1, activity mm) for additional detail.

zz) Conduct walkthrough at school or child care facility before the start of sampling

(hrs_walkthrough_school_op, cost_walkthrough_school). The EPA assumed CWSs will incur the
same burden and costs as under the first testing cycle to conduct a walkthrough with each school
or child care facility to become familiar with the facility and to identify sampling sites. The burden
and cost for the CWS to complete this task is 1.40 hours and $5.75 for CWSs serving 100,000 or
fewer people, 1.64 hours and $7.36 for those serving 100,001 to 1 million people, and 1.71 hours
and $10.24 for those serving more than 1 million people. See Section 4.3.2.5.1, activity nn) for
additional detail.

aaa) Travel to collect samples (hrs_travel_samp_school_op, cost_travel_samp_school). The EPA

assumed CWSs will incur the same burden and costs as under the first testing phase to travel to a
school or child care facility to collect samples of 0.40 hours and $5.75 for CWSs serving 100,000 or
fewer people, 0.51 hours and $7.36 for those serving 100,001 to 1 million people, and 0.71 hours
and $10.24 for those serving more than 1 million people. See Section 4.3.2.5.1, activity oo) for
additional detail.

Final LCRI Economic Analysis

4-126

October 2024


-------
bbb) Collect samples (hrs_collect_samp_school_op, cost_collect_samp_school). The EPA assumed

CWSs will require the same per sample burden and cost to collect samples as under the first testing
phase of 10 minutes (0.17 hours) for all system sizes and bottle cost of $1.12 that applies only to
CWSs serving more than 100,000 people. See activity pp) for additional detail.

ccc) Analyze samples (hrs_analyze_samp_op, cost_lab_lt_samp, cost_commercial_lab). Under the on
request program phase, CWSs will incur the same burden and cost to analyze lead samples in-
house or to use a commercial laboratory as lead tap sampling and the first testing phase of the
school and child care facility sampling program. Specifically, CWSs serving more than 100,000
people will incur a burden of 0.44 hours per sample (hrs_analyze_samp_op) and cost of $3.92 per
sample (cost_lab_lt_samp) to analyze lead samples in-house. CWSs serving 100,000 or fewer
people will use a commercial lab inclusive of shipping samples to the laboratory of $31.00 per
sample (cost_commercial_lab). See Section 4.3.2.5.1, activity qq) for additional detail.

ddd) Provide sampling results to tested facilities (hrs_inform_samp_pe_school_op,

cost_inform_samp_pe_school). CWSs will incur the same burden and costs as under the first
testing phase to provide sampling results to tested facilities. The CWS burden is 0.05 to 0.11 hours
and $0.72 per tested facility. See Section 4.3.2.5.1, activity rr) for additional detail.

eee) Discuss sampling results with the school and child care facility (hrs_result_discuss_op). CWSs will
continue to incur burden to discuss the sampling results with each school and child care facility
under the on request program phase. The EPA assumed the same average burden of 1 hour per
tested facility as under the mandatory program.

fff) Conduct detailed discussion of high sampling results with the school and child care facility

(hrs_school_help_op). For each lead sample result over the AL, the EPA assumed CWSs would incur
the same burden of 5 hours to work with each school or child care facility as under the first phase of
testing. See Section 4.3.2.5.1, activity tt) for additional detail).

ggg) Report school and child care facility sampling results to States. CWSs must also continue to report
school and child care facility testing results to the States within 30 days of learning the results. The
EPA estimated an annual burden of 12 hours (i.e., 1 hour per monthly report). See Section
4.3.2.5.1, activity uu) for additional detail.

hhh) Prepare and provide an annual report on testing program to the State

(hrs_annual_report_school_prepare_op, cost_annual_report_school_dist). CWSs must continue
to prepare and distribute an annual report regarding their testing program at schools and child care
facilities at an estimated burden of 1 hour for CWSs serving 100,000 or fewer people, 2 hours for
those serving 100,001 to 1 million people, and 8 hours for CWSs serving more than 1 million
people, and a cost of $0.72 for all system sizes. See Section 4.3.2.5.1, activity vv) for additional
detail.

Exhibit 4-47 shows the SafeWater LCR model cost estimation approach for activities under the second

five-year cycle including additional cost inputs required to calculate these costs.

Final LCR! Economic Analysis

4-127

October 2024


-------
Exhibit 4-47: CWS School and Child Care Facility Second Five-Year Testing Cycle Cost Estimation in SafeWater LCR by Activity1,2





Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

ww) Update the list of schools and child care facilities and submit to the State

The number of public elementary schools per system population times the system
population multiplied by the hours to identify each facility and the system labor rate.

numb_elem_schools_pub*pws_pop*(hrs_school_identify_op*rate_op)





All model PWSs



The number of public secondary schools per system population times the system
population multiplied by the hours to identify each school and the system labor rate.

numb_second_schools_pub*pws_pop*(hrs_school_identify_op*rate_op)





All model PWSs



The number of child care facilities per system population times the system population
multiplied by the hours to identify each facility and the system labor rate.

numb_daycares*pws_pop*(hrs_school_identify_op*rate_op)

Cost does
not apply to
NTNCWSs.

All

All model PWSs

Every five
years

The number of private elementary schools per system population times the system
population multiplied by the hours to identify each facility and the system labor rate.

numb_elem_schools_priv*pws_pop*(hrs_school_identify_op*rate_op)





All model PWSs



The number of private secondary schools per system population times the system
population multiplied by the hours to identify each school and the system labor rate.

numb_second_schools_priv*pws_pop*(hrs_school_identify_op*rate_op)





All model PWSs



Final LCR! Economic Analysis

4-128

October 2024


-------
CWS Cost Per Activity

xx) Contact schools and child care facilities to offer sampling

The number of public elementary schools that do not meet waiver requirements per
system population times the system population multiplied by the hours per school
times the system labor rate, plus the cost of materials.

p_pub_elem_onreq_waiver_LCRrnumb_elem_schools_pub*pws_pop*

((hrs school annual contact op *rate op)+ cost school annual contact)	

The number of public secondary schools that do not meet waiver requirements per
system population times system population multiplied by the hours per school times
the system labor rate, plus the cost of materials.

p_pub_second_onreq2on_waiver_LCRI*numb_second_schools_pub*pws_pop*

((hrs school annual contact op *rate op)+ cost school annual contact)	

The number of child care facilities that do not meet waiver requirements per system
population times the system population multiplied by the hours per facility times the
system labor rate, plus the cost of materials.

p_childcare_onreq_waiver_LCRrnumb_daycares*pws_pop*
((hrs_school_annual_contact_op *rate_op)+ cost_school_annual_contact)
The number of private elementary schools that do not meet waiver requirements per
system population times the system population multiplied by the hours per school
times the system labor rate, plus the cost of materials.

p_priv_elem_onreq_waiver_LCRrnumb_elem_schools_priv*pws_pop*
((hrs_school_annual_contact_op *rate_op)+ cost_school_annual_contact)
The number of private secondary schools that do not meet waiver requirements per
system population times the system population multiplied by the hours per school
times the system labor rate, plus the cost of materials.

p_priv_second_onreq2on_waiver_LCRrnumb_second_schools_priv*pws_pop*
((hrs_school_annual_contact_op *rate_op)+ cost_school_annual_contact)

Final LCRI Economic Analysis

4-129



Conditions for Cost to Apply to a
Model PWS



NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

All model PWSs

Cost does
not apply to
NTNCWSs.

All

All model PWSs

All model PWSs

Once a
year

All model PWSs

All model PWSs

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

yy) Contact the school or child care facility to coordinate sample collection logistics

5% of public elementary schools that do not meet waiver requirements multiplied by
the hours per school and the system labor rate.

(p_pub_elem_onreq_waiver_LCRrnumb_elem_schools_pub*pws_pop*pp_voluntary
_partic)*(hrs_school_coor_sample_op*rate_op)





All model PWSs



5% of public secondary schools that do not meet waiver requirements multiplied by
the hours per school and the system labor rate.

(p_pub_second_onreq2on_waiver_LCRI*numb_second_schools_pub*pws_pop*pp_v
oluntary_partic)*(hrs_school_coor_sample_op*rate_op)





All model PWSs



5% of child care facilities that do not meet waiver requirements multiplied by the
hours per facility and the system labor rate.

(p_childcare_onreq_waiver_LCRrnumb_daycares*pws_pop*pp_voluntary_partic)*(h
rs_school_coor_sample_op*rate_op)

Cost does
not apply to
NTNCWSs.

All

All model PWSs

Once a
year

5% of private elementary schools that do not meet waiver requirements multiplied by
the hours per school and the system labor rate.

(p_priv_elem_onreq_waiver_LCRI*numb_elem_schools_priv*pws_pop*pp_voluntary
_partic)*(hrs_school_coor_sample_op*rate_op)





All model PWSs



5% of private secondary schools that do not meet waiver requirements multiplied by
the hours per school and the system labor rate.

(p_priv_second_onreq2on_waiver_LCRrnumb_second_schools_priv*pws_pop*pp_v
oluntary_partic)*(hrs_school_coor_sample_op*rate_op)





All model PWSs



Final LCRI Economic Analysis

4-130

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

zz) Conduct walkthrough at school or child care facility before start of sampling

5% of public elementary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.

(p_pub_elem_onreq_waiver_LCRrnumb_elem_schools_pub*pws_pop*pp_voluntary
_partic)*((hrs_walkthrough_school_op*rate_op)+cost_walkthrough_school)





All model PWSs



5% of public secondary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.

(p_pub_second_onreq2on_waiver_LCRI*numb_second_schools_pub*pws_pop*pp_v
oluntary_partic)*((hrs_walkthrough_school_op*rate_op)+cost_walkthrough_school)





All model PWSs



5% of child care facilities that do not meet waiver requirements multiplied by the
hours per facility times the system labor rate, plus the cost of materials.

(p_childcare_onreq_waiver_LCRrnumb_daycares*pws_pop*pp_voluntary_partic)*((
hrs_walkthrough_school_op*rate_op)+cost_walkthrough_school)

Cost does
not apply to
NTNCWSs.

All

All model PWSs

Once a
year

5% of private elementary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.

(p_priv_elem_onreq_waiver_LCRI*numb_elem_schools_priv*pws_pop*pp_voluntary
_partic)*((hrs_walkthrough_school_op*rate_op)+cost_walkthrough_school)





All model PWSs



Final LCRI Economic Analysis

4-131

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

5% of private secondary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.

(p_priv_second_onreq2on_waiver_LCRrnumb_second_schools_priv*pws_pop*pp_v
oluntary_partic)*((hrs_walkthrough_school_op*rate_op)+cost_walkthrough_school)





All model PWSs



aaa) Travel to collect samples

5% of public elementary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.

(p_pub_elem_onreq_waiver_LCRrnumb_elem_schools_pub*pws_pop*pp_voluntary
_partic)*((hrs_travel_samp_school_op*rate_op)+cost_travel_samp_school)





All model PWSs



5% of public secondary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.

(p_pub_second_onreq2on_waiver_LCRI*numb_second_schools_pub*pws_pop

*pp_voluntary_partic)*((hrs_travel_samp_school_op*rate_op)+cost_travel_samp_sch

ool)

Cost does
not apply to
NTNCWSs.

All

All model PWSs

Once a
year

5% of child care facilities that do not meet waiver requirements multiplied by the
hours per facility times the system labor rate, plus the cost of materials.

(p_childcare_onreq_waiver_LCRrnumb_daycares*pp_voluntary_partic)*((hrs_travel
_samp_school_op*rate_op)+cost_travel_samp_school)





All model PWSs



Final LCRI Economic Analysis

4-132

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

5% of private elementary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.

(p_priv_elem_onreq_waiver_LCRI*numb_elem_schools_priv*pws_pop*pp_voluntary
_partic)*((hrs_travel_samp_school_op*rate_op)+cost_travel_samp_school)





All model PWSs



5% of private secondary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.









(p_priv_second_onreq2on_waiver_LCRrnumb_second_schools_priv*pws_pop

*pp_voluntary_partic)*((hrs_travel_samp_school_op*rate_op)+cost_travel_samp_sch

ool)





All model PWSs



bbb) Collect samples3

5% of public elementary schools that do not meet waiver requirements multiplied by
the number of samples per school, is multiplied by the number of hours per school
times the system labor rate, plus the material cost.

(p_pub_elem_onreq_waiver_LCRrnumb_elem_schools_pub*pws_pop*pp_voluntary

_partic)*numb_samp_five)*((hrs_collect_samp_school_op*rate_op)+cost_collect_sa

mp_school)





All model PWSs



5% of public secondary schools that do not meet waiver requirements multiplied by
the number of samples per school, is multiplied by the number of hours per school
times the system labor rate, plus the material cost.

(p_pub_second_onreq2on_waiver_LCRrnumb_second_schools_pub*pws_popl*pp_
voluntary_partic)*numb_samp_five)*((hrs_collect_samp_school_op*rate_op)+cost_c
ollect samp school)

Cost does
not apply to
NTNCWSs.

All

All model PWSs

Once a
year

5% of child care facilities that do not meet waiver requirements multiplied by the
number of samples per facility, is multiplied by the number of hours per facility times
the system labor rate, plus the material cost.





All model PWSs



(p_childcare_onreq_waiver_LCRrnumb_daycares*pp_voluntary_partic)*numb_samp
two)*((hrs collect samp school op*rate op)+cost collect samp school)









Final LCRI Economic Analysis

4-133

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

5% of private elementary schools that do not meet waiver requirements multiplied by
the number of samples per school, is multiplied by the number of hours per school
times the system labor rate, plus the material cost.

(p_priv_elem_onreq_waiver_LCRI*numb_elem_schools_priv*pws_pop*pp_voluntary

_partic)*numb_samp_five)*((hrs_collect_samp_school_op*rate_op)+cost_collect_sa

mp_school)





All model PWSs



5% of private secondary schools that do not meet waiver requirements multiplied by
the number of samples per school, is multiplied by the number of hours per school
times the system labor rate, plus the material cost.

(p_priv_second_onreq2on_waiver_LCRrnumb_second_schools_priv*pws_popl*pp_
voluntary_partic)*numb_samp_five)*((hrs_collect_samp_school_op*rate_op)+cost_c
ollect_samp_school)





All model PWSs



ccc) Analyze samples3

The number of required samples per public elementary school that do not meet
waiver requirements multiplied by 5 percent of elementary schools per year times the
probabilities for a sample analyzed in house and a sample analyzed in a commercial
lab times the different labor and material cost burdens for each type of analysis.

((((p_pub_elem_onreq_waiver_LCRI*numb_elem_schools_pub*pws_pop*pp_volunta
ry_partic) *numb_samp_five)*pp_lab_samp_school)*( (hrs_analyze_samp_op*rate_op
)+cost_lab_lt_samp))+((((p_pub_elem_onreq_waiver_LCRrnumb_elem_schools_pu
b*pws_pop*pp_voluntary_partic)*numb_samp_five)*pp_commercial_samp_school)*c
ost commercial lab)

Cost does
not apply to
NTNCWSs.

All

All model PWSs

Once a
year

The number of required samples per public secondary school that do not meet
waiver requirements multiplied by 5 percent of elementary schools per year times the
probabilities for a sample analyzed in house and a sample analyzed in a commercial
lab times the different labor and material cost burdens for each type of analysis.

((((p_pub_second_onreq2on_waiver_LCRI*numb_second_schools_pub*p\NS_pop*pp
_voluntary_partic)*numb_samp_five)*pp_lab_samp_school)*((hrs_analyze_samp_op
*rate op)+cost lab It samp))+((((p pub second onreq2on waiver LCRI*numb sec





All model PWSs



Final LCRI Economic Analysis

4-134

October 2024


-------




Conditions for Cost to Apply to a
Model PWS

CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

ond_schools_pub*p\NS_pop*pp_voluntary_partic)*numb_samp_five)*pp_commercial_
samp_school)*cost_commercial_lab)









The number of required samples per child care facility that do not meet waiver
requirements multiplied by 5 percent of elementary schools per year times the
probabilities for a sample analyzed in house and a sample analyzed in a commercial
lab times the different labor and material cost burdens for each type of analysis.

((((p_childcare_onreq_waiver_LCRI*numb_daycares*pws_pop*pp_voluntary_partic)*
numb_samp_five)*pp_lab_samp_school)*((hrs_analyze_samp_op*rate_op)+cost_lab
_lt_samp))+((((p_childcare_onreq_waiver_LCRI*numb_daycares*pws_pop*pp_volunt
ary partic)*numb samp five)*pp commercial samp school)*cost commercial lab)





All model PWSs



The number of required samples per private elementary school that do not meet
waiver requirements multiplied by 5 percent of elementary schools per year times the
probabilities for a sample analyzed in house and a sample analyzed in a commercial
lab times the different labor and material cost burdens for each type of analysis.

((((p_priv_elem_onreq_waiver_LCRrnumb_elem_schools_priv*pws_pop*pp_volunta
ry_partic) *numb_samp_five)*pp_lab_samp_school)*( (hrs_analyze_samp_op*rate_op
)+cost_lab_lt_samp))+((((p_priv_elem_onreq_waiver_LCRrnumb_elem_schools_pri
v*pws_pop

*pp_voluntary_partic)*numb_samp_five)*pp_commercial_samp_school)*cost_comm
ercial lab)

Cost does
not apply to
NTNCWSs.

All

All model PWSs



The number of required samples per private secondary school that do not meet
waiver requirements multiplied by 5 percent of elementary schools per year times the
probabilities for a sample analyzed in house and a sample analyzed in a commercial
lab times the different labor and material cost burdens for each type of analysis.









((((p_priv_second_onreq2on_waiver_LCRrnumb_second_schools_priv*pws_pop
*pp_voluntary_partic)*numb_samp_five)*pp_lab_samp_school)*((hrs_analyze_samp
_op*rate_op)+cost_lab_lt_samp))+((((p_priv_second_onreq2on_waiver_LCRI*numb_
second_schools_priv*pws_pop

*pp_voluntary_partic)*numb_samp_five)*pp_commercial_samp_school)*cost_comm
ercial lab)





All model PWSs



Final LCRI Economic Analysis

4-135

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

ddd) Provide sampling results to tested facilities

5% of public elementary schools multiplied that do not meet waiver requirements by
the hours per school times the system labor rate, plus the cost of materials.

(p_pub_elem_onreq_waiver_LCRrnumb_elem_schools_pub*pws_pop*pp_voluntary
_partic)*((hrs_inform_samp_pe_school_op*rate_op)+cost_inform_samp_pe_school)





All model PWSs



5% of public secondary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.

(p_pub_second_onreq2on_waiver_LCRI*numb_second_schools_pub*pws_pop*pp_v

oluntary_partic)*((hrs_inform_samp_pe_school_op*rate_op)+cost_inform_samp_pe_

school)

Cost does
not apply to
NTNCWSs.

All

All model PWSs

Once a
year

5% of child care facilities that do not meet waiver requirements multiplied by the
hours per facility times the system labor rate, plus the cost of materials.

(p_childcare_onreq_waiver_LCRrnumb_daycares*pws_pop*pp_voluntary_partic)*((
hrs_inform_samp_pe_school_op*rate_op)+cost_inform_samp_pe_school)





All model PWSs



5% of private elementary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.

(p_priv_elem_onreq_waiver_LCRrnumb_elem_schools_priv*pws_pop*pp_voluntary
_partic)*((hrs_inform_samp_pe_school_op*rate_op)+cost_inform_samp_pe_school)





All model PWSs



Final LCRI Economic Analysis

4-136

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

5% of private secondary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate, plus the cost of materials.









(p_priv_second_onreq2on_waiver_LCRrnumb_second_schools_priv*pws_pop*pp_v
oluntary_partic)*((hrs_inform_samp_pe_school_op*rate_op)+cost_inform_samp_pe_
school)





All model PWSs



eee) Discuss sampling results with the school and child care facility

5% of public elementary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate.

(p_pub_elem_onreq_waiver_LCRrnumb_elem_schools_pub*pws_pop*pp_voluntary
partic)*(hrs result discuss op*rate op)





All model PWSs



5% of public secondary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate.

(p_pub_second_onreq2on_waiver_LCRI*numb_second_schools_pub*pws_pop*pp_v
oluntary_partic)*(hrs_result_discuss_op*rate_op)

Cost does
not apply to
NTNCWSs.

All

All model PWSs

Once a
year

5% of child care facilities that do not meet waiver requirements multiplied by the
hours per facility times the system labor rate.

(p_childcare_onreq_waiver_LCRrnumb_daycares*pws_pop*pp_voluntary_partic)*(h
rs_result_discuss_op *rate_op)





All model PWSs



5% of private elementary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate.

(p_priv_elem_onreq_waiver_LCRrnumb_elem_schools_priv*pws_pop*pp_voluntary
_partic)*(hrs_result_discuss_op*rate_op)





All model PWSs



Final LCRI Economic Analysis

4-137

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

5% of private secondary schools that do not meet waiver requirements multiplied by
the hours per school times the system labor rate.

(p_priv_second_onreq2on_waiver_LCRrnumb_second_schools_priv*pws_pop*pp_v
oluntary_partic)*(hrs_result_discuss_op*rate_op)





All model PWSs



fff) Conduct detailed discussion of high sampling results with schools and child care facilities

5% of public elementary schools that do not meet waiver requirements multiplied by
the number of required samples per system above the AL multiplied by the hours per
sample times the system labor rate.

(p_pub_elem_onreq_waiver_LCRI*numb_elem_schools_pub*pws_pop*pp_voluntary
partic)* ((pp above al bin three*numb samp five)*(hrs school help op*rate op))





All model PWSs



5% of public secondary schools that do not meet waiver requirements multiplied by
the number of required samples per system above the AL multiplied by the hours per
sample times the system labor rate.

(p_pub_second_onreq2on_waiver_LCRI*numb_second_schools_pub*pws_pop*pp_v
oluntary_partic)*

((pp_above_al_bin_three*numb_samp_five)*(hrs_school_help_op*rate_op))

Cost does
not apply to
NTNCWSs.

At or below
the AL

All model PWSs

Once a
year

5% of child care facilities that do not meet waiver requirements multiplied by the
number of required samples per system above the AL multiplied by the hours per
sample times the system labor rate.

(p_childcare_onreq_waiver_LCRrnumb_daycares*pws_pop*pp_voluntary_partic)*
((pp above al bin three*numb samp five)*(hrs school help op*rate op))





All model PWSs



5% of private elementary schools that do not meet waiver requirements multiplied by
the number of required samples per system above the AL multiplied by the hours per
sample times the system labor rate.

(p_priv_elem_onreq_waiver_LCRrnumb_elem_schools_priv*pws_pop*pp_voluntary
partic)* ((pp above al bin three*numb samp five)*(hrs school help op*rate op))





All model PWSs



5% of private secondary schools that do not meet waiver requirements multiplied by
the number of required samples per system above the AL multiplied by the hours per
sample times the system labor rate.





All model PWSs



Final LCRI Economic Analysis

4-138

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

(p_priv_second_onreq2on_waiver_LCRI*numb_second_schools_priv*pws_pop*pp_v
oluntary_partic)*

((pp above al bin three*numb samp five)*(hrs school help op*rate op))









5% of public elementary schools that do not meet waiver requirements multiplied by
the number of required samples per system above the AL multiplied by the hours per
sample times the system labor rate.





All model PWSs



(p_pub_elem_onreq_waiver_LCRrnumb_elem_schools_pub*pws_pop*pp_voluntary
partic)* ((pp above al bin one*numb samp five)*(hrs school help op*rate op))









5% of public secondary schools that do not meet waiver requirements multiplied by
the number of required samples per system above the AL multiplied by the hours per
sample times the system labor rate.

(p_pub_second_onreq2on_waiver_LCRI*numb_second_schools_pub*pws_pop*pp_v
oluntary_partic)*

((pp above al bin one*numb samp five)*(hrs school help op*rate op))





All model PWSs



5% of child care facilities that do not meet waiver requirements multiplied by the
number of required samples per system above the AL multiplied by the hours per
sample times the system labor rate.

(p_childcare_onreq_waiver_LCRrnumb_daycares*pws_pop*pp_voluntary_partic)*
((pp above al bin one*numb samp five) (hrs school help op*rate op))

Cost does
not apply to
NTNCWSs.

Above the AL

All model PWSs

Once a year

5% of private elementary schools that do not meet waiver requirements multiplied by
the number of required samples per system above the AL multiplied by the hours per
sample times the system labor rate.





All model PWSs



(p_priv_elem_onreq_waiver_LCRrnumb_elem_schools_priv*pws_pop*pp_voluntary
partic)* ((pp above al bin one*numb samp five)*(hrs school help op*rate op))









5% of private secondary schools that do not meet waiver requirements multiplied by
the number of required samples per system above the AL multiplied by the hours per
sample times the system labor rate.

(p_priv_second_onreq2on_waiver_LCRrnumb_second_schools_priv*pws_pop*pp_v
oluntary_partic)*

((pp above al bin one*numb samp five)*(hrs school help op*rate op))





All model PWSs



Final LCRI Economic Analysis

4-139

October 2024


-------




Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

ggg) Report school and child care facility sampling results to the State

The total hours per system multiplied by the system labor rate.
(hrs_report_sch_cc_results_op*rate_op)

Cost does
not apply to
NTNCWSs.

All

All model PWSs

Once a
year

hhh) Prepare and provide annual report on school and child care facility sampling to the State

The total hours per system multiplied by the system labor rate, plus the materials
cost.

(hrs_annual_report_school_prepare_op*rate_op)+cost_annual_report_school_dist)

Cost does
not apply to
NTNCWSs.

All

All model PWSs

Once a
year

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; PWS = public water system.
Notes:

1The data variables in the exhibit are defined previously in this section with the exception of:

•	numb_daycares, numb_elem_schools_priv, numb_elem_school_pub, numb_second_schools_priv, numb_second_school_pub\ Number of child care
facilities, number of private elementary schools, number of public elementary schools, number of public secondary schools, and number of private
secondary schools, respectively that are served by CWSs (Chapter 3, Section 3.3.10.1).

•	p_childcare_onreq_waiver_LCRI, p_priv_elem_onreq_waiver_LCRI, p_pub_elem_onreq_waiver_LCRI, p_pub_second_onreq2on_waiver_LCRI,
p_priv_second_onreq2on_waiver_LCRI: States and schools/child care facilities, private K-12 schools, and public K-12 schools for the second period on
request program (Section 3.3.10.2).

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

2	The second five-year testing cycle is assumed to start in Year 9.

3	The burden and costs to provide sample bottles (cost_collect_samp_school) under activity bbb) and conduct analyses under activity ccc) are incurred by the

State in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina (ASDWA, 2020a).

Final LCRI Economic Analysis

4-140

October 2024


-------
4.3.2.6 Estimate of PWS National Sampling Costs

Exhibit 4-48 shows the total estimated national sampling costs, under the low and high scenarios,
discounted at 2 percent for the 2021 LCRR and the final LCRI. The annual monetized incremental
sampling costs range from $32.0 million and $32.6 million in 2022 dollars. Note, the more aggressive SL
replacement requirements under the final LCRI, combined with the higher likelihood of PWSs having an
ALE under the high scenario under both the 2021 LCRR and final LCRI, results in the high scenario lead
tap sampling incremental cost being lower than the low scenario. See Section 4.2.2 for more detail on
the factors that produce the difference between the low and high modeling scenarios.

Exhibit 4-48: Estimated National Annualized Sampling Costs - 2 Percent Discount Rate

(millions of 2022 USD)





Low Estimate



High Estimate





Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Lead Tap Sampling

$102.2

$98.7

-$3.5

$108.1

$103.5

-$4.6

Lead Water Quality
Parameters Monitoring

$16.6

$35.6

$19.0

$20.1

$40.6

$20.5

Copper Water Quality
Parameters Monitoring

$0.2

$0.2

$0.0

$0.2

$0.3

$0.1

Source Water Monitoring

$0.0

$0.0

$0.0

$0.0

$0.1

$0.1

School Sampling

$15.0

$31.5

$16.5

$15.2

$31.7

$16.5

Total Annual Sampling
Costs

$134.0

$166.0

$32.0

$143.6

$176.2

$32.6

4.3.3 PWS Corrosion Control Costs

PWSs may be required to install CCT, re-optimize their existing CCT, or perform a distribution system
and site assessment (DSSA)108 adjustment to their CCT under the final LCRI. CCT installation and re-
optimization are required based on the system's lead 90th percentile range. The likelihood of a model-
PWS exceeding the lead AL of 10 ng/L for the low and high cost scenarios is in Exhibit 4-4. The DSSA
adjustment to CCT is prompted by a requirement under the final LCRI where systems are required to
take certain actions when individual lead tap samples are greater than 10 ng/L.

Any changes to the status of a system's CCT may result in technology related costs (capital and/or
O&M), as well as ancillary costs for data submission, consultation, and CCT studies. This section presents
the following CCT-related costs:

•	4.3.3.1: CCT Installation

•	4.3.3.2: Re-optimization

108 This was previously known as "find-and-fix" under the 2021 LCRR.

Final LCRI Economic Analysis

4-141

October 2024


-------
• 4.3.3.3: DSSA Costs

• 4.3.3.4: System Lead CCT Routine Costs

Each subsection presents capital and O&M costs followed by ancillary costs. Note that PWS costs for
monitoring of CCT effectiveness (i.e., lead tap and WQP monitoring) has already been presented in
Sections 4.3.2.1 and 4.3.2.2, respectively. Also note that WWTP costs to address increased phosphorus
loadings are presented in Section 4.5.

All CCT-related capital and O&M costs are calculated using the EPA's WBS cost models, which are
described in Section 4.2.2.3 and detailed in Technologies and Costs for Corrosion Control to Reduce Lead
in Drinking Water (USEPA, 2023b). WBS capital cost equations are a function of design flow (DF), and
WBS O&M cost equations are a function of average daily flow (ADF). These flows are estimated based
on the system's retail population served. As explained in Chapter 3, Section 3.3.6 DF and ADF for the
system are divided by the average number of entry points per system to calculate flow per entry
point.109 These entry point flow values are used in the WBS cost equations. CCT-related capital and O&M
costs per entry point are summed for all entry points to produce the CCT-related capital and O&M costs
for the system. As noted in Section 4.2.2.3, the EPA recognizes uncertainty in CCT capital and O&M cost
equations by varying the WBS model inputs (e.g., fiberglass storage tank vs. more expensive stainless
steel construction) to create "low" and "high" cost equations. Low CCT cost equations are used for the
low cost scenario, and high CCT cost equations are used for the high cost scenario. These equations can
be found in the Technologies and Costs for Corrosion Control to Reduce Lead in Drinking Water (USEPA,
2023b).

In order to estimate CCT installation, re-optimization, or DSSA costs, the SafeWater LCR model requires
an estimate of the pH of the model-PWS's pre-regulatory compliance (or baseline) finished water. Using
data from the Six-Year 3 Review Information Collection Request (ICR) Dataset, the EPA developed
triangular distributions based on the minimum, mode, and maximum of baseline pH levels (converted to
logio values) for model-PWSs with and without existing pH adjustment. PFAS treatment is not expected
to adversely impact pH as most systems were projected to use granular activated carbon, which does
not impact pH. There is a pH impact from ion exchange, but only for a brief period after start-up. For
each distribution, the EPA estimated distribution quartile threshold pH values and quartile midpoint
values. Then the EPA estimated system-weighted averages of the midpoint values to derive the final set
of distributions for ground water and surface water systems with and without baseline pH adjustment
shown in Exhibit 4-49.

109 In the case of some very large systems (VLSs), the EPA knows the flows at each of its entry points (EPs) and each
EP's pH level. The SafeWater LCR model uses these data to calculate the CCT installation and O&M costs for each
EP for these VLSs.

Final LCRI Economic Analysis

4-142

October 2024


-------
Exhibit 4-49: Distribution of Baseline Finished Water pH by Source Water Type and pH

Adjustment Status

Likelihood

Finished Water pH

PWSs without pH Adjustment in
Place

PWSs with pH Adjustment in
Place

Groundwater

Surface Water

Groundwater

Surface Water

10%

5.1

5.6

6.3

6.3

15%

5.9

6.3

6.8

6.8

25%

6.6

6.8

7.3

7.2

25%

7.3

7.4

7.8

7.7

15%

8.0

7.9

8.3

8.3

10%

8.6

8.4

8.8

8.9

Acronyms: PWS = public water system.

The EPA then used the estimates in Exhibit 4-49 to develop the pH distribution for model-PWSs with 1)
no CCT installed and 2) orthophosphate (P04) treatment installed. The EPA assumes that model-PWSs
with no CCT would have pH of at least 7.0. Therefore, the EPA truncated the values for "PWSs without
pH Adjustment in Place" which resulted in the distribution for the variable baselineph_wocct. Likewise,
the EPA used the estimates in Exhibit 4-49 to develop the pH distribution for model-PWSs with P04 in
place without pH adjustment. The EPA assumes that model-PWSs with only P04 installed would have pH
of at least 6.3. Therefore, the EPA truncated the values for "PWSs without pH Adjustment in Place"
which resulted in the distribution for the variable baseline_woph. The distributions for both
baselineph_wocct and baselineph_woph are provided in Exhibit 4-50.

Exhibit 4-50: Distribution of Finished Water pH by Source Water Type for Model-PWSs
without pH Adjustment in Place by CCT Status

Probability

Finished Water pH

Probability

Finished Water pH

PWSs without CCT in
Place baselineph_wocct

PWSs with just P04 in
Place baselineph_woph

Groundwater

Surface Water

Groundwater

Surface Water







25%

6.3

6.3

50%

7.0

7.0

25%

6.6

6.8

25%

7.3

7.4

25%

7.3

7.4

15%

8.0

7.9

15%

8.0

7.9

10%

8.6

8.4

10%

8.6

8.4

Acronyms: CCT = corrosion control treatment; P04 = orthophosphate; PWS = public water system.

The EPA then used the estimates in Exhibit 4-49 to develop the pH distribution for model-PWSs with pH
adjustment in place by CCT status. The EPA assumed that model-PWSs with P04 and pH adjustment
could have any of the baseline pH levels associated with "PWSs with pH adjustment in place." Therefore,
no adjustment to the pH distribution for "PWSs with pH Adjustment in Place" was required to develop
the distribution for the variable baselineph_wpo4ph. However, the EPA determined that PWSs with only
pH adjustment installed would have a pH of at least 8.2. Therefore, the EPA truncated the values for

Final LCRI Economic Analysis

4-143

October 2024


-------
"PWSs with pH Adjustment in Place" which resulted in the distribution for the variable baseline_wph.
The distributions for both baselineph_wpo4ph and baselineph_wph are provided in Exhibit 4-51.

Exhibit 4-51: Distribution of Finished Water pH by Source Water Type for Model-PWSs with

pH Adjustment in Place by CCT Status

Probability

Finished Water pH

Probability

Finished Water pH

PWSs with P04 and pH
Adjustment in Place

baselineph_ wpo4ph

PWSs with only pH
Adjustment in Place

baselineph_ wph

Groundwater

Surface Water

Groundwater

Surface Water

10%

6.3

6.3







15%

6.8

6.8







25%

7.3

7.2







25%

7.8

in

75%

8.2

8.2

15%

8.3

8.3

15%

8.3

8.3

10%

8.8

8.9

10%

8.8

8.9

Acronyms: P04 = orthophosphate; PWS = public water system.

In order to determine the cost of re-optimizing CCT or undertaking pH adjustment triggered by DSSA
requirements, for model-PWSs with existing P04 treatment installed, the SafeWater LCR model needs an
estimate of the model-PWS's baseline dose of P04. Using data from the Six-Year 3 Review ICR Dataset,
the EPA developed a triangular distribution based on the minimum (0.05 mg/L), mode (1.4 mg/L), and
maximum (4 mg/L) of reported baseline P04 doses. For ease of modeling CCT unit costs, the EPA limited
the number of potential baseline P04 doses to four ranges and represented each range by its median as
shown in Exhibit 4-52, columns (a), (b), and (c). Using the triangular distribution, the EPA determined the
likelihood of a model-PWS having a baseline P04 dose in each range as shown in column (d). The EPA
assumed this likelihood applied to model-PWSs serving 50,000 or fewer people with no LSLs. The EPA
then assumed that these smaller systems, that have LSLs, will be less likely than same size systems
without LSLs to have P04 doses in the lowest of the four ranges, since LSLs, when present, represent the
greatest contributor of lead in a home's drinking water. A study published by the American Water Works
Association (AWWA) Research Foundation "Contributions of Service Line and Plumbing Fixtures to Lead
and Copper Rule Compliance Issues" (Sandvig et al., 2008) estimates that 50 percent to 75 percent of
lead in drinking water comes from LSLs.110 Since LSLs represent a more significant lead challenge, it is
expected that systems would need higher orthophosphate doses to reduce lead levels. The EPA
modeled this assumption by decreasing the likelihood of having a dose of 0.525 mg/L by 50 percent and

110 While removal of LSLs is critical to reducing lead in drinking water, premise plumbing materials also continue to
be a source of lead in drinking water (Elfland, 2010; Kimbrough, 2007; Rockey et al., 2021). In addition, premise
plumbing materials can be a source of particulate lead. For example, brass particles and lead solder particles were
identified as the cause of severe tap water contaminations during three field investigations in North Carolina and
Washington, D.C. (Triantafyllidou and Edwards, 2012). This means that even where systems remove all LSLs, CCT
must be continued because of the lead and copper sources that will remain in the premise plumbing of consumers'
homes and other buildings (USEPA, 2020b), and in lead connectors. Systems without LSLs can exceed the lead
action level, for example, due to the corrosion of premise plumbing containing lead.

Final LCR! Economic Analysis

4-144

October 2024


-------
increasing the likelihood of having the next highest dose, 1.5 mg/L, by an equivalent amount (see
column (e)).

The EPA also made adjustments to implement its assumption that larger systems have a higher
probability of higher doses than small systems with similar LSL status (see columns (f) and (g)), since the
distribution systems are larger and more complex. Finally, the EPA assumed, for modeling purposes, that
a dose of 3.2 mg/L will result in optimized CCT and that no model-PWS in the baseline has fully
optimized CCT as a conservative estimate. Therefore, the likelihood of a model-PWS having a baseline
dose of 3.2 mg/L is set to zero and the likelihoods of the other doses is normalized so that the sum of
the percentage values equal 100 percent. The final baseline P04 doses, and their likelihoods, are
provided in Exhibit 4-53.

Final LCRI Economic Analysis

4-145

October 2024


-------
Exhibit 4-52: Derivation of Baseline PO4 Dose by System Size and LSL Status

P04 Dose Range

P04 Dose Range

P04 Dose

Likelihood

Likelihood

Likelihood

Likelihood

Minimum

Maximum

Range Median

<50,000 people

<50,000 people

> 50,000 people

>50,000 people

(mg/L)

(mg/L)

(mg/L)

No LSL

LSL

No LSL

LSL

(a)

(b)

(c)

(d)

(e)

(f)

(g)

0.05

<1

0.525

7.9%

4.0%

4.0%

0%

> 1

<2

1.5

48.1%

52.1%

28.0%

32.0%

>2

<3.2

2.65

38.6%

38.6%

43.4%

43.4%

>3.2

4

3.6

5.4%

5.4%

24.7%

24.7%

Exhibit 4-53: Baseline PO4 Doses by System Size and LSL Status Used in Cost Modeling

P04 Dose

Normalized

Normalized

Normalized

Normalized

Range

Likelihood

Likelihood

Likelihood

Likelihood

Median

<50,000 people

<50,000 people

> 50,000 people

>50,000 people

(mg/L)

No LSL

LSL

No LSL

LSL

0.525

8%

4%

5%

0%

1.5

51%

55%

37%

42%

2.65

41%

41%

58%

58%

Final LCRI Economic Analysis

4-146

October 2024


-------
4.3.3.1 CCT Installation

PWSs without CCT may be required to install CCT under the final LCRI if they exceed the lead AL.111 Costs
related to CCT installation are categorized as follows:

•	Capital and operations and maintenance costs (see Section 4.3.3.1.1).

•	Ancillary costs (see Section 4.3.3.1.2).

4.3.3.1.1 Capital and Operation and Maintenance CCT Installation Costs
Under the final LCRI, a PWS that installs CCT will choose among three technology options.

•	Add P04 and pH post-treatment

•	Add P04 and modify pH

•	Modify pH

The EPA assumed that model-PWSs with a baseline pH (baselineph_wocct) equal to or greater than 7.2,
but less than 8.4, will choose to add P04 and conduct pH post-treatment, while those with pH below 7.2
will choose to add P04 and modify pH. For model PWSs that add P04 with pH post-treatment, the EPA
assumed that the P04 dose is equal to 3.2 mg/L and post-treatment will maintain the current pH level
(baselineph_wocct).112 For model-PWSs that add P04 and adjust pH, the EPA assumes the same P04 dose
of 3.2 mg/L. In addition, the EPA assumes the model PWS will adjust their pH from their starting pH
(baselineph_wocct) to 7.2. The EPA assumes that model-PWSs with a baseline pH greater that 8.4 will
choose to modify pH and not add P04.

The SafeWater LCR model uses the WBS unit cost functions (see Technologies and Costs for Corrosion
Control to Reduce Lead in Drinking Water (USEPA, 2023b), along with the entry point (EP) flow values, to
calculate the capital and O&M costs for CCT installation at each entry point to the distribution system
(EP). All of the WBS capital cost equations are a function of DF, and all WBS O&M costs are a function of
ADF.113 Since CCT is conducted at the model-PWS's EPs, the SafeWater LCR model calculates the DF and
ADF of each EP. For all model-PWSs except some very large systems114 (see Section 4.2.3), the EPA does
not know the number of people, and hence, flow, associated with individual EPs. Therefore, in the
absence of this information, the SafeWater LCR model calculates the EPs flows assuming they are equal
to:

111	EPA assumed that CWSs serving 50,000 or more people will have already installed CCT except for a very small
number of "b3" systems (16), which are assumed to have naturally non-corrosive water and never be required to
install CCT. The 2021 LCRR provides flexibility to CWSs serving 3,300 or fewer people and all NTNCWSs by allowing
them to choose among replace all LSLs, install POU treatment, install/re-optimize CCT, or replace all lead bearing
plumbing if they exceed the lead AL.

112	The addition of P04 lowers pH levels so post-treatment is conducted to maintain pH levels.

113	See Chapter 3, Section 3.3.6 for a description of how the EPA estimates PWS design and ADFs.

114	In the case of some very large systems (VLSs), the EPA collected additional EP level data on flows and pH level
(See Appendix B, Section B.2.3). The SafeWater LCR model uses these data to calculate the CCT installation and
O&M costs for each EP for these VLSs.

Final LCRI Economic Analysis

4-147

October 2024


-------
Entry Point Design Flow = PWS Design Flow / PWS Number of EPs

Entry Point Average Daily Flow = PWS Average Daily Flow / PWS Number of EPs

The model-PWS capital and O&M cost of CCT installation at each EP is then multiplied by the number of
EPs. The cost models, and their inputs, for calculating the capital and O&M cost of CCT installation are:

•	P04 and pH post-treatment

P04 dose = 3.2

Current pH: = baselineph_wocct
Ending pH = baselineph_wocct

•	Add P04 and modify pH

P04 dose = 3.2

Current pH = baselineph_wocct
Ending pH = 7.2

•	Modify pH

P04 dose = 3.2

Current pH = baselineph_wocct
Ending pH = 9.2

In addition to the capital and O&M cost of CCT installation, model-PWSs also face an ancillary CCT study
cost associated with CCT installation. This cost is discussed in the next section.

4.3.3.1.2 Ancillary CCT Installation Costs

The EPA has developed system costs for an ancillary activity associated with CCT installation as shown in
Exhibit 4-54. The exhibit provides the unit burden and/or cost for the activity. The assumptions used in
the estimation of the unit burden and cost follow the exhibit. The last column provides the
corresponding SafeWater LCR model data variable in red/italic font.

Exhibit 4-54: PWS CCT Installation-Related Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

a) Conduct a study

Study

•	No LSLs and PWSs serving < 10,000
people with LSLs (coupon testing):
$30,372

•	With LSLs (harvested pipe loop testing):
$307,744 for 10,001 - 50,000 people;
$376,685 for > 50,000 people

cost_cct_study_dem

Acronyms: CCT = corrosion control treatment; LSL = lead service line.

Note: Activity b), "Install CCT Treatment (PO4, PO4 with post treatment, pH adjustment, or modify pH)" was
previously discussed in Section 4.3.3.1.1.

Final LCR! Economic Analysis

4-148

October 2024


-------
a) Conduct a study (cost_cct_study_dem). The EPA assumed States will require all systems to conduct
either harvested pipe loop testing or a coupon study prior to CCT installation. The SafeWater LCR
model uses the following set of assumptions:

•	Systems required to conduct a CCT study will use a contractor.

•	Systems without LSLs and systems serving 10,000 or fewer people with LSLs will use a
coupon study at an estimated cost of $30,372 for systems of all sizes.

•	Systems with LSLs will incur a cost of $307,744 for those serving 10,001 to 50,000
people and $376,685 for those serving more than 50,000 people for harvested pipe loop
testing.

The development of harvested pipe loop and coupon test study costs are detailed in Technologies and
Costs for Corrosion Control to Reduce Lead in Drinking Water (USEPA, 2023b).

Exhibit 4-55 details how the data variables are used to estimate system ancillary activities related to CCT
Installation.

Exhibit 4-55: PWS Ancillary CCT Installation Cost Estimation in SafeWater LCR by Activity1





Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

a) Conduct a CCT study

Material cost per system for the
marginal contractor cost, with the
difference between coupon testing
and harvested pipe loop testing
reflected in the stratification of the
data by system LSL status.

Cost applies as
written to
NTNCWSs.

Above AL

Model PWSs without
CCT that conducts a
study on CCT
installation

One time

cost_cct_study_dem









Acronyms: AL = action level; CCT = corrosion control treatment; CWS = community water system; LSL = lead service

line; NTNCWS = non-transient non-community water system; PWS = public water system.

Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of:

• rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

4.3.3.2 Re-optimization of Existing Corrosion Control Treatment

PWSs that have previously implemented CCT may be required to re-optimize their treatment if they
exceed the lead AL again. Costs related to CCT re-optimization are categorized as follows:

•	Capital and operations and maintenance costs (see Section 4.3.3.2.1).

•	Ancillary costs (see Section 4.3.3.2.2).

Final LCR! Economic Analysis

4-149

October 2024


-------
4.3.3.2.1 Capital and Operation and Maintenance CCT Re-optimization Costs

Estimating the cost of existing CCT

While the EPA knows which model-PWSs currently have CCT installed, the EPA does not know which CCT
technology they have installed. Therefore, when the SafeWater LCR model develops the model-PWSs, it
assigns a CCT technology to each model-PWS known to have CCT in place.115 The input parameters used
in the WBS models that calculate existing CCT O&M costs for the three modeled CCT technologies, are:

•	Add P04 with PH Post Treatment.

o P04 Dose = baselinepo4dose
o Starting pH: baselineph_woph
o Ending pH: baselineph_woph

•	Modify pH.

o Starting pH: baselineph_wph - 0.5
o Ending pH: pH: baselineph_wph

•	Add P04 and Modify PH.

o P04 Dose = baselinepo4dose
o Starting pH: baselineph_wpo4ph- 0.5
o Ending pH: baselineph_wpo4ph

Estimating the cost of re-optimizing existing CCT

The EPA assumed that if a model-PWS must re-optimize its CCT under the final LCRI, it will achieve the
following standards based on its existing CCT technology (which was described above):

•	Add P04 and pH post-treatment.

o Increase P04 dose to 3.2 mg/L.
o Maintain existing pH.

•	Add P04 and modify pH.

o Increase P04 dose of 3.2 mg/L.
o Maintain pH at a minimum of 7.2.

•	Modify pH.

o Maintain pH at 9.2.

To calculate the cost to re-optimize CCT, the SafeWater LCR model first calculates the annual O&M cost
of treating to the above assumed standards (P04 dose and/or pH level) as if no CCT was installed. To do
so, the SafeWater LCR model uses the following parameters and WBS cost functions:

•	Add P04 and pH post-treatment.

115 See derivation file "Baseline CCT Characteristics.xlsx" for the baseline parameters and their likelihoods.

Final LCRI Economic Analysis

4-150

October 2024


-------
P04 Dose = 3.2 mg/L
Beginning pH = baselineph_woph
Ending pH = baselineph_woph

•	Add P04 and modify pH.

P04 Dose = 3.2 mg/L

Beginning pH = baselineph_wpo4ph- 0.5

Ending pH = the greater of baselineph_wpo4ph or 7.2

•	Modify pH.

Beginning pH = baselineph_wph -0.5
Ending pH = 9.2

The SafeWater LCR model then subtracts the model-PWS's existing CCT annual O&M cost from the new
annual O&M cost to calculate the share of the model-PWS's annual CCT O&M costs attributable to the
final LCRI CCT requirements. These O&M costs, combined with the annualized capital cost to retrofit the
CCT system based on the new parameters, described above, equal the model PWS's total annual capital
and O&M cost for CCT adjustment. The following section discusses additional ancillary costs associated
with CCT adjustment.

4.3.3.2.2 Ancillary CCT Re-optimization Costs

The EPA has developed system ancillary costs for an ancillary activity associated with CCT re-
optimization as shown in Exhibit 4-56. The exhibit provides the unit burden and/or cost for the activity.
The assumptions used in the estimation of the unit burden follow the exhibit. The last column provides
the corresponding SafeWater LCR model data variable in red/italic font.

Exhibit 4-56: PWS CCT Ancillary Re-optimization Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

c) Revise CCT study

Svstems with ALE

•	No LSLs and PWSs serving <
10,000 people with LSLs (coupon
test): $30,372

•	With LSLs (harvested pipe loop
testing): $307,744 for 10,001 -
50,000 people;

$376,685 for > 50,000 people

cost_cct_study_dem

Acronyms: ALE = action level exceedance, CCT = corrosion control treatment; LSL = lead service line.
Source: Technologies and Costs for Corrosion Control to Reduce Lead in Drinking Water (USEPA, 2023b).
Note: Activity d), "Re-optimize existing CCT" was previously discussed in Section 4.3.3.2.1.

c) Revise CCT study (cost_cct_study_dem). The EPA assumed States will require all systems to conduct
a study prior to CCT re-optimization.

• Systems will use a contractor to conduct a study.

Final LCRI Economic Analysis

4-151

October 2024


-------
• Systems with an ALE will conduct a demonstration study (cost_cct_study_dem). Specifically,
systems:

o Without LSLs and systems serving 10,000 or fewer people with LSLs will do a coupon
study at an estimated cost of $30,372 for all sizes.

o With LSLs will do a harvested pipe loop at an estimated cost of $307,744 for systems
serving 10,001 to 50,000 people and $376,685 for those serving more than 50,000
people.

The development of harvested pipe loop and coupon test study costs are detailed in Technologies and
Costs for Corrosion Control to Reduce Lead in Drinking Water (USEPA, 2023b).

Exhibit 4-57 shows the SafeWater LCR model cost estimation approach for system ancillary CCT re-
optimization study activities including additional cost inputs required to calculate these costs.

Exhibit 4-57: PWS CCT Ancillary Re-optimization Cost Estimation in SafeWater LCR by Activity1





Conditions for Cost to Apply to a
Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th -
Range

Other Conditions

Frequency
of Activity

c) Revise CCT study

Material cost per system for the
marginal contractor cost, with the
difference between coupon testing and
harvested pipe loop testing reflected in
the stratification of the data by system
LSL status.

Cost applies as
written to
NTNCWSs.

Above AL

Model PWS re-
optimizing CCT

One time

cost_cct_study_dem









Acronyms: AL = action level; CCT = corrosion control treatment; CWS = community water system; LSL = lead service

line; NTNCWS = non-transient non-community water system; PWS = public water system.

Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of:

• rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

Final LCR! Economic Analysis

4-152

October 2024


-------
4.3.3.3 DSSA Costs

Under the final LCRI, PWSs must take DSSA corrective actions whenever an individual lead tap water
sample exceeds 10 ng/L. The likelihood that a sample would exceed 10 ng/L is provided in Exhibit 4-58
with the corresponding SafeWater input names shown in red.

Exhibit 4-58: Likelihood of an Individual Lead Sample Result Above 10 |ig/L

LSL Status

P90 >15 ng/L

12 ng/L < P90 <
15ng/L

10 ng/L < P90 <
12 Hg/L

5 |ig/L < P90 <
10 ng/L

P90 < 5 ng/L

pp90above
allO_l

pp90above
all0_2

pp90above
all0_3

pp90above
all0_4

pp90above
all0_5

Has LSLs

25.2%

16.8%

13.8%

6.5%

1.8%

No LSLs

22.2%

23.1%

21.1%

6.5%

0.5%

Acronyms: ALE = action level exceedance; LSL = lead service line.
Source: Likelihood_Sample_Above_AL_LCRI_DSSA_Final.xlsx.

Note: This information also is provided as Exhibit 3-33 in Chapter 3.

The EPA assumed in the SafeWater LCR model that in response to individual lead tap water samples
above 10 ng/L, model-PWSs will take progressively more stringent corrective actions. These assumed
actions are:

1.	First sampling period with one or more individual tap water samples above 10 ng/L - model-
PWS will investigate the cause but not take any corrective action.

2.	Second sampling period with one or more individual tap water samples above 10 ng/L - model-
PWS will perform spot flushing once in the distribution system.

3.	Third sampling period with one or more individual tap water samples above 10 ng/L - model-
PWS will increase the pH level at one EP.

4.	Fourth sampling period with one or more individual tap water samples above 10 ng/L - model-
PWS will increase the pH at all other EPs (if more than one).

These corrective actions are not meant to encompass the entire suite of DSSA compliance options but
rather provide a representation of typical actions a PWS might take to correct reoccurring individual
lead tap samples over 10 ng/L.

4.3.3.3.1 Cost of Spot Flushing an Entry Point

In response to a second sampling period with at least one lead tap sample greater than 10 ng/L, the EPA
assumed, in the Safe Water LCR model, that systems will perform spot flushing. Spot flushing involves
crews opening hydrants in the area of the tap monitoring result to bring in fresh water and eliminate
potential issues with elevated water age, which could cause the water to be more corrosive. The
assumptions for spot flushing are consistent with the Technology and Cost Document for the Revised
Total Cotiform Rule (USEPA, 2012b). See Exhibit 4-59 for the PWS unit burden and cost for spot flushing
with detailed assumptions in the notes.

Final LCRI Economic Analysis

4-153

October 2024


-------
Exhibit 4-59: PWS Burden and Cost to Flush as DSSA Response (2020$)

System Size
(Population Served)

Burden (hrs per system)

(hrsJush_wqp_op)

Cost ($ per system)

(cost_flush_ wqp)

A

B

<1,000

4

$131.03

1,001-3,300

4

$195.89

3,301-50,000

8

$195.89

>50,000

8

$260.75

Source: Technology and Cost Document for the Revised Total Coliform Rule (USEPA, 2012b;
"Likelihood_Sample_Above_AL_LCRI_DSSA_Final.xlsx"). Costs have been updated to 2020$.

Notes:

A: Assumes that each spot flushing response is a one-half day event. Assumes 1-person crew for systems
serving 3,300 or fewer people and 2-person crew for those serving > 3,300 people.

B: Estimate is based on value of flushed water and cost of flushed water disposal, i.e., dichlorination. Where
LCRI system size categories do not match those used in the technology and cost document for the Revised
Total Coliform Rule (RTCR), the EPA used the closest category.

4.3.3.3.2 Cost of pH Adjustment

In response to a third sampling period with at least one lead tap sample greater than 10 ng/L, the EPA
assumed, in the Safe Water LCR model, that a model-PWS will increase its pH at one EP if it has
optimized CCT in place. The EPA assumed the model-PWS will achieve the following standards:

•	If model-PWS has used P04 for its corrosion inhibitor, then the system will maintain its pH at a
minimum of 7.5 instead of 7.2.

•	If a model-PWS modified pH for corrosion control, it will maintain its pH at 9.4 instead of 9.2.

To calculate the cost to increase pH in response to individual tap samples above 10 ng/L, the SafeWater
LCR model first calculates the total annual O&M cost for treating to the DSSA standards listed above as if
no CCT was installed. The SafeWater LCR model also calculates the capital cost to retrofit the CCT system
for additional pH adjustment. To do so, the SafeWater LCR model uses the following parameters and
WBS cost functions:

•	If the model-PWS has P04 treatment installed and its baselineph_woph < 7.5:

Add P04 and Modify pH
P04 Dose = 3.2

Starting pH: baselineph_woph
Ending pH: 7.5

•	If the model-PWS has P04 treatment installed and its baselineph_woph > 7.5:

Add P04with pH Post Treatment
P04 Dose = 3.2

Starting pH: baselineph_woph
Ending pH: baselineph_woph

•	If the model-PWS has pH adjustment installed:

Modify pH

Final LCRI Economic Analysis

4-154

October 2024


-------
Starting pH: = baselineph_woph - 0.5
Ending pH: 9.4

The SafeWater LCR model then subtracts the model-PWS's current CCT annual O&M cost from the new
DSSA annual O&M cost to calculate the share of model-PWS's annual CCT O&M costs attributable to
DSSA actions. These O&M costs combined with the annualized capital cost to retrofit the CCT system
based on the new parameters, described above, equal the model-PWS's total annual capital and O&M
cost of DSSA. Additional ancillary costs associated with DSSA are discussed in the following section.

In the fourth sampling period with one or more individual tap water samples above 10 ng/L, the model-
PWS will increase the pH at all other EPs (if the model-PWS has more than one EP). This calculation is
the same as described for the Year 3 DSSA pH adjustment except that the calculation is made for all
entry points.

4.3.3.3.3 Ancillary DSSA Costs

The EPA developed ancillary costs associated with a system's DSSA responses to a lead tap result above
10 ng/L as shown in Exhibit 4-60. The exhibit provides the unit burden and/or cost for each activity. The
assumptions used in the estimation of each activity follows the exhibit. The last column provides the
corresponding SafeWater LCR model data variable in red/italic font. In a few instances, some of these
activities are conducted by the State instead of the water system. These activities are identified in the
exhibit and further explained in the exhibit notes.

Exhibit 4-60: PWS Ancillary DSSA Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

e) Contact customers and collect
follow-up tap sample

Burden per sample
CWSs: 3.4 to 3.7 hrs
NTNCWSs: 0.5 hrs

Burden

hrs_samp_above_al_op



Costs per sample
CWSs: $5.75 to $13.09
NTNCWSs: $0

Cost

cost_samp_ above_ al

f) Analyze follow-up lead tap

In-house Analysis (CWSs > 100K onlv)

In-house Analysis

sample

Burden: 0.44 hrs/sample
Cost: $3.92

hrs_ an alyze_samp_ op1
costjabj^samp1



Commercial Analysis
$32.20

Commercial Analysis
cos^commercialjab1

Final LCR! Economic Analysis

4-155

October 2024


-------
Activity

Unit Burden and/or Cost

SafeWaterLCR Data Variable

g) Collect distribution system

Burden per sample per PWS with

Burden

WQP sample

CCT

hrs_ wqp_dssa_op



0.5 hrs







Cost



Cost for per sample per PWS with

pH: cost_wqp_material_ph



CCT





pH adjustment:





• $1.70 to $2.66 (CWS);

Orthophosphate:



• $2.66 (NTNCWS)

cost_wqp_material_ortho



Orthophosphate:





• $0.63 to $1.07 (CWS)





• $2.66 (NTNCWS)



h) Analyze distribution system

In-House Burden per sample per

In-House Burden

WQP sample

PWS with CCT

pH adjustment:



pH adjustment:

hrs_wqp_analyze_ph_op



• 0.15 to 0.46 hrs (CWS)





• 0.15 hrs (NTNCWS)

Orthophosphate:



Orthophosphate:

hrs_wqp_an alyze_ orth o_ op



• 0.15 to 1.34 hrs (CWS)





• 0.15 hrs (NTNCWS)





In-House cost per sample per PWS

In-House Cost



with CCT

pH adjustment:



pH adjustment:

cost_ wqp_ph_an alyze



• $0.63 to $0.98 (CWS)





• $0.63 (NTNCWS)





Orthophosphate:

Orthophosphate:



• $0.63 to $1.07 (CWS)

cost_ wqp_ orth o_ an alyze



• $0.63 (NTNCWS)





Commercial cost per sample per

Commercial Cost



PWS with CCT

pH: cost_lab_ph_wqp



pH adjustment: $27.24 to 30.55

Orthophosphate:



(CWS & NTNCWS)

cost_lab_ortho_wqp



Orthophosphate: $60.34 to $61.89





(CWS & NTNCWS)



i) Review incidents of

CWSs: 4 to 30 hrs/system

hrs_deter_dssa_op

systemwide event and other

NTNCWSs: lto 14 hrs/system



system conditions





j) Consult with the State prior to

2 hrs per system with CCT

hrs_consult_dssa_op

making CCT changes





k) Report follow-up sample

2 hrs/PWS serving < 50,000 people;

hrs_report_ dssa_ op

results and overall DSSA

4 hrs/PWS serving > 50,000 people



responses to the State





Acronyms: CCT = corrosion control treatment; CWS = community water system; DSSA = Distribution System and
Site Assessment; NTNCWS = non-transient non-community water system; PWS = public water system; WQP =
water quality parameter.

Sources: Data sources for each activity are provided following this exhibit.

Note:

Final LCRI Economic Analysis

4-156

October 2024


-------
1 In Arkansas, Louisiana, Mississippi, Missouri, and South Carolina, the State pays for the cost of bottles, shipping,
analysis, and providing sample results to the system (ASDWA, 2020a). Thus, the State will incur the burden and
cost for these activities in lieu of the system.

e) Contact customers and collect follow-up tap samples (hrs_samp_above_al_op,

cost_samp_above_al). CWSs and NTNCWSs will incur burden and costs to contact customers and
collect a follow-up tap sample at each compliance sampling location116 that had a result above 10
Hg/L. Exhibit 4-58 in Section 4.3.3.3 provides the likelihood a system will have a single sample above
10 ng/L for each of the five lead 90th percentile classifications. Also refer to Chapter 3, Section
3.3.5.3 for a detailed discussion of the EPA's approach for developing these percentages. For
modeling purposes, the EPA assumed all customers would respond to the water system and agree
to have a follow-up sample collected.

Exhibit 4-61 provides the burden or labor associated with these activities for CWSs and Exhibit 4-62
provides the associated costs. Burden and cost estimates for NTNCWSs follow the exhibits.

Note that the required notification to the customer of the original sample result above 10 ng/L that
triggered the additional sampling is captured under the Lead Tap Sampling Costs using
hrs_inform_samp_op and cost_cust_lt for CWSs and hrs_NTNCWS_inform_samp_op and
cost_NTNCWS_inform_lt for NTNCWSs. See Section 4.3.2.1.2, activity m).

Exhibit 4-61: Burden (hours) for CWSs to Contact Customers and Collect Tap Samples for
Locations with a Lead Tap Sample > 10 |ig/L (hrs_samp_above_al_op)

System Size
(Population Served)

Phone Call

Site Visit

Total

Travel (Round-
Trip)

Look for Lead
Sources

Sample
Collection

A

B

C

D

E=A:D

<100,000

0.5

0.40

2

0.5

3.4

100,001-1,000,000

0.5

0.51

2

0.5

3.5

>1,000,000

0.5

0.71

2

0.5

3.7

Source: "Likelihood_Sample_Above_AL_DSSA_LCRI_Final.xlsx."

Notes:

General: This requirement applies to all CWSs that have any sample > 10 ng/L.

A: Assumed systems would spend 0.5 hours to contact customer to coordinate site visit and to discuss possible
causes of the high tap sample value.

B: Based on census data and zip codes from the 2006 Community Water System Survey, assumed the following
one-way driving distances for CWSs: 5.0 miles for those serving < 100,000 people, 6.4 miles for those serving
100,001 - 1,000,000, and 8.9 miles for those serving > 1,000,000. See file, "Estimated Driving Distances_Final.xlsx,"
available in the docket under EPA-HQ-OW-2022-0801 at www.regulations.gov," for additional detail. The EPA
assumed an average speed of 25 miles per hour and two times distance for round-trip travel.

C: Assumed systems will spend 2 hours on average to look for lead sources in premise plumbing and service line.

116 Some systems conduct free lead testing at the request of the customer. The EPA encourages, but does not
require, systems to conduct DSSA activities if a customer requested tap sample result exceeds 10 ng/L.

Final LCRI Economic Analysis

4-157

October 2024


-------
D: Assumed same burden as used for systems to collect a lead and copper source water sample, see Section
4.3.2.4.2, activity ff) for detail.

Exhibit 4-62: Costs for CWSs to Contact Customers and Collect Tap Samples for Locations with
a Lead Tap Sample > 10 |ig/L (cost_samp_above_al)

System Size
(Population Served)

Phone Call

Site Visit

Total

Travel (Round-
Trip)

Look for Lead
Sources

Sample
Collection

A

B

C

D

E=A:D

<100,000

$0.00

$5.75

$0.00

$0.00

$5.75

100,001-1,000,000

$0.00

$7.36

$0.00

$2.85

$10.21

>1,000,000

$0.00

$10.24

$0.00

$2.85

$13.09

Source: "Likelihood_Sample_Above_AL_LCRI_DSSA_Final.xlsx."

Notes:

General: This requirement applies to all CWSs that have any sample > 10 ng/L.

A&C: Assumed to have no non-labor costs.

B: Based on census data and zip codes from the 2006 Community Water System Survey, assumed the following
one-way driving distances for CWSs: 5.0 miles for those serving < 100,000 people, 6.4 miles for those serving
100,001 - 1,000,000, and 8.9 miles for those serving > 1,000,000. See file, "Estimated Driving Distances_Final.xlsx,"
available in the docket under EPA-HQ-OW-2022-0801 at www.regulations.gov," for additional detail. Assumed cost
of $.5754 per mile using the 2020 reimbursement from https://www.gsa.gov/travel/plan-book/transportation-
airfare-pov-etc/privately-owned-vehicle-mileage-rates/pov-mileage-rates-archived#auto. Accessed 1/17/2022.
D: Based on information from laboratories, only CWSs serving > 100,000 people are assumed to conduct in-house
analyses for lead whereas those serving < 100,000 people will use a commercial lab and bottles are supplied by the
commercial lab. The average cost of a 1-liter wide mouth bottle assuming a bulk discount rate based on six sources
is $2.85. See "Lead_WQP_Sample Bottle Costs," worksheet "Average Bottle Costs" for additional information.

NTNCWSs will also be required to collect a follow-up sample but will incur a different burden and cost
from CWSs because they do not serve homeowners and thus, are not required to conduct a separate
site visit. The EPA assumed NTNCWSs will incur a burden 0.5 hours per follow-up sample
(hrs_samp_above_al_op), which is the same burden as that used to collect a lead and copper source
water sample and is based on the 2022 Disinfectants and Disinfection Byproducts, Chemical, and
Radionuclides Rules ICR, Exhibit 15 (USEPA, 2022a). In addition, NTNCWSs will incur no bottle costs to
collect the sample because the EPA assumed all NTNCWSs will use a commercial lab in which bottles are
included as part of the laboratory fee. Thus, cost_samp_above_al is $0.

f) Analyze follow-up tap samples (hrs_analyze_samp_op, cost_lab_lt_samp, cost_commercial_lab).

As previously presented in Section 4.3.2.1.2, activity k), the EPA assumed CWSs serving more than
100,000 people will conduct lead analyses in-house and require 0.44 hours per sample based on
estimates provided by three laboratories (hrs_analyze_samp_op). These systems will also incur
consumable costs of $3.92 per sample based on information from three vendors (cost_lab_lt_samp).
The remaining CWSs and all NTNCWSs are assumed to use a commercial laboratory and incur a cost
to ship the sample to the lab of $8.70 and an analytical cost of $23.50 per lead sample analysis
based on quotes from seven laboratories for a total per sample cost of $32.20

Final LCRI Economic Analysis

4-158

October 2024


-------
(cost_commercial_lab). See "Lead Analytical Burden and Costs_Final.xlsx," worksheet "Tap_
Collect_Analyze_CWS_LCRI" for additional information.

g) Collect distribution system WQP sample (hrs_wqp_dssa_op, cost_wqp_material_ph,

cost_wqp_material_ortho). Systems with CCT must collect one distribution sample at or near the
site where the high lead sample was collected within five days of learning of the lead results. Thus,
the EPA assumed the timing of this monitoring may not coincide with their Total Coliform Rule (TCR)
samples and systems would incur a burden of 0.5 hours to collect the WQP sample
(hrs_wqp_dssa_op). The EPA uses the same SafeWater LCR model data variables and input values
for the burden and cost associated with WQP distribution system sample collection as described in
Section 4.3.2.2.4 and Exhibit 4-21 (CWSs) and Exhibit 4-22 (NTNCWSs) for this activity.

If an existing WQP site does not meet these criteria, the system must identify a new monitoring site.
Only systems with CCT must use the new site for future WQP distribution system sampling. The EPA
has capped the additional number of WQP sample sites that must be added in response to DSSA
investigations to twice the standard number of required WQP sample sites. For example, as
discussed in Section 4.3.2.2.3, systems serving 10,001 to 50,000 people must conduct monitoring
from 10 sites if they are on standard monitoring (numb_enhance_wqp). For DSSA distribution
monitoring, no more than 10 additional sites would be added for systems on standard or reduced
monitoring.117

For CWSs, the EPA estimated the likelihood a WQP site will need to be added (pp_overlap_dssa) in
Exhibit 4-63. This likelihood is used to determine the number of sites added to a CWS's WQP sample
collection and analysis each year (numb_wqp_sites_added). NTNCWSs have limited distribution
systems and the EPA assumed these systems with CCT will not add new WQP sites.

Exhibit 4-63: Likelihood a CWS Will Add a WQP Sampling Site in Response to the DSSA

System Size
(Population Served)

Tap Samples sites
(standard number)

WQP sites
(standard number)

Percent of WQPs
compared to Tap
Sites

Likelihood a CWS
will add a WQP site

pp_ overlap_ dssa

A

B

C= B/A*100

D

<100

5

1

20.0%

0.0%

101-500

10

1

10.0%

0.0%

501-1,000

20

2

10.0%

0.0%

1,001-3,300

20

2

10.0%

20.0%

3,301-10,000

40

3

7.5%

20.0%

117 Systems subject to lead or copper WQP monitoring as discussed in Sections 4.3.2.2 and 4.3.2.3, respectively,
must collect two samples from the number of sites specified in the rule. As discussed in Section 4.3.2.2.3,
numb_enhance_wqp represents the standard number of WQP tap samples that must be collected at each site for
systems on standard monitoring. In the SafeWater LCR model, Yi numb_enhance_wqp represents the maximum
number of samples that could be added under the DSSA requirements for systems with CCT because only one
sample would be required at each site. This applies to systems with CCT on standard or reduced WQP tap
monitoring.

Final LCR! Economic Analysis

4-159

October 2024


-------
System Size
(Population Served)

Tap Samples sites
(standard number)

WQP sites
(standard number)

Percent of WQPs
compared to Tap
Sites

Likelihood a CWS
will add a WQP site

pp_ overlap_ dssa

A

B

C= B/A*100

D

10,001-100,000

60

10

16.7%

20.0%

>100,000

100

25

25.0%

10.0%

Source: "Likelihood_Sample_Above_AL_LCRI_DSSA_Final.xlsx."

Notes:

A: See Exhibit 4-9.

B: See Exhibit 4-19.

D: The EPA assumed for CWSs with CCT serving:

•	<1,000 people, the distribution system is not extensive and the WQP sampling location would be at or
near the sampling site with the lead result above 10 ng/L. Thus, these systems would have a zero
likelihood of adding a new WQP site.

•	> 1,000 people, the EPA divided the minimum required number of WQP sites (Column B) by the number
of tap sites (Column A). The EPA assumed the higher the ratio, the more likely a system would be to have
a WQP sampling site at or near a required tap sampling site with lead values greater than 10 ng/L and the
lower the likelihood a system would add a new WQP sampling site. Specifically, the EPA assumed those
with a ratio of < 20 percent (those serving 1,001 -100,000 people) would have a 0.2 likelihood of adding a
new WQP site. The EPA assumed those with a ratio of > 20 percent would have a lower likelihood of 0.1 of
adding a new WQP site (those serving > 100,000 people).

h)	Analyze distribution system WQP sample (hrs_wqp_analyze_ph_op, hrs_wqp_analyze_ortho_op,
cost_wqp_ph_analyze, cost_wqp_ortho_analyze, cost_lab_ph_wqp, cost_lab_ortho_wqp).

Systems with CCT must collect the same WQPs as discussed in Section 4.3.2.2.4 for lead WQP
monitoring. Specifically, systems using pH adjustment must sample for pH and alkalinity, those using
orthophosphate treatment must sample for pH, alkalinity, and orthophosphate. Thus, the EPA used
the same SafeWater LCR model data variables and input values for WQP sample analysis as
described in Section 4.3.2.2.4 for lead WQP monitoring. See Exhibit 4-23 and Exhibit 4-24 for the
analytical burden for CWSs and NTNCWSs to conduct in-house analyses, respectively
(hrs_wqp_analyze_ph_op, hrs_wqp_analyze_ortho_op). See Exhibit 4-25 and Exhibit 4-26 for the in-
house analytical costs for CWSs and NTNCWSs, respectively (cost_wqp_ph_analyze,
cost_wqp_ortho_analyze). See Exhibit 4-27 and Exhibit 4-28 for the commercial costs per sample for
CWSs and NTNCWSs, respectively (cost_lab_ph_wqp, cost_lab_ortho_wqp).

i)	Review incidents of systemwide events and other system conditions (hrs_deter_dssa_op). Under
the final LCRI, systems must determine if a CCT change is needed following lead tap sample result(s)
above 10 ng/L. For the purposes of this cost analysis, the EPA assumed that systems will assess
distribution system operations and determine if there could have been factors that contributed to
deteriorating water quality and elevated lead levels. Exhibit 4-64 provides the estimated burden for
CWSs and NTNCWSs to conduct this assessment. The estimates are based on comparable activities
and burden estimates for CWSs and NTNCWSs to conduct level 1 assessments following non-acute
TCR violations. Additional detail on the derivation of these burdens is provided in "Likelihood
_Sample_Above_AL_LCRI_DSSA_Final.xlsx," in worksheet, "Distribution_System_Assessment."

Final LCRI Economic Analysis

4-160

October 2024


-------
Exhibit 4-64: PWS Burden to Conduct Distribution System Assessment

System Size

CWS Burden to Conduct

NTNCWS Burden to Conduct

(Population Served)

Assessment (hrs/system)

Assessment (hrs/system)



hrs_deter_dssa_op

<1,000

4

1

1,001-3,300

6

1

3,301-10,000

8

4

10,001-50,000

10

5

50,001-100,000

13

6

>100,000

30

14

Source: Technology and Cost Document for the Final Revised Total Coliform Rule (USEPA, 2012b); Economic
Analysis for the Final Revised Total Coliform Rule (USEPA, 2012a) (available in the docket at EPA-HQ-OW-2022-
0801 at www.regulations.gov). Derived in "Likelihood_Sample_Above_AL_LCRI_DSSA_Final.xlsx," worksheet,
"Distribution_System_Assessment."

j) Consult with the State prior to making CCTchanges (hrs_consult_dssa_op). Systems with CCT that
have at least one sample > 10 ng/L must consult with their State prior to making any CCT changes.
The EPA assumed a 2 hour consultation burden that is consistent with other types of consultations
and is based on the estimated burden for systems to consult with their State on public education
activities from pg. 60 of the Economic and Supporting Analyses: Short-Term Regulatory Changes to
the Lead and Copper Rule (USEPA, 2007).

k) Report follow-up sample results and overall DSSA responses to the State (hrs_report_dssa_op).

PWSs will incur burden to provide the results of tap and WQP monitoring results, and any
distribution system management actions or CCT adjustments made to fix the cause of sample results
above 10 ng/L to their State. The EPA assumed the systems will require 2 hours and 4 hours to
prepare the annual report for systems serving 50,000 or fewer and those serving more than 50,000
people, respectively. The EPA assumed systems would not incur a separate cost for generating a
physical report because systems would provide this information electronically to their State.

Systems must also provide this information to the health departments. The EPA assumed that
systems would incorporate the DSSA results into a larger report that includes outreach information
and school sampling results (CWSs only). The burden and material cost of the report is captured
under the cost to distribute the outreach, which corresponds to data inputs hrs_hc_op and cost_hc.
See Section 4.3.6.2, activity I).

Exhibit 4-65 provides the SafeWater LCR model cost estimation approach for system ancillary DSSA

activities including additional cost inputs that are required to calculate the total costs.

Final LCR! Economic Analysis

4-161

October 2024


-------
Exhibit 4-65: PWS Ancillary DSSA Cost Estimation in SafeWater LCR by Activity1-2





Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost Per Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

e) Contact customers and collect follow-up tap samples-









The number of required samples per system above the AL
multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(pp_above_al_bin_three*numb_samp_customer)*((hrs_sam
p above al op*rate op)+cost samp above al)





PWSs not on reduced
tap sampling and not
doing POU sampling

1 - (p_tap_annual +
p_tap_triennial +
p_tap_nine)

Twice a
year



Cost applies as written to NTNCWSs.

At or
below AL

PWSs on annual
reduced tap sampling
and not doing POU
sampling

p tap annual

Once a
year

The number of required samples per system above the AL
multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(pp_above_al_bin_three*numb_reduced_tap)*((hrs_samp_a
bove_al_op*rate_op)+cost_samp_above_al)





PWSs on triennial
reduced tap sampling
and not doing POU
sampling

p_tap_triennial

Every 3
years

The number of required samples per system above the AL
multiplied by the total of the hours per sample times the
system labor rate, plus the cost of materials per sample.

(pp_above_al_bin_one*numb_samp_customer)*((hrs_samp
above al op*rate op)+cost samp above al)

Cost applies as written to NTNCWSs.

Above AL

All PWSs with at least
one sample > 10 |jg/L

Twice a
year

Final LCR! Economic Analysis

4-162

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost Per Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

f) Analyze follow-up lead tap sample3

The number of samples multiplied by the likelihoods for a
sample analyzed in house and a sample analyzed in a
commercial lab times the different labor and material cost
burdens for each type of analysis.





PWSs is not on
reduced tap sampling
and not doing POU
sampling

1 - (p_tap_annual +
pjtapjtrienniai +
p_tap_nine)

Twice a
year

(((pp above al bin three*numb samp customer)*pp lab









samp)*((hrs_analyze_samp_op*rate_op)+cost_lab_lt_samp)

)+(((pp_above_al_bin_three*numb_samp_customer)*pp_co

mmercial_samp)*cost_commercial_lab)

Cost applies as written to NTNCWSs.

At or
below AL

PWSs on annual
reduced tap sampling
and not doing POU
sampling

p_tap_annual

Once a
year

The number of samples multiplied by the likelihoods for a
sample analyzed in house and a sample analyzed in a
commercial lab times the different labor and material cost
burdens for each type of analysis.

(((pp_above_al_bin_three*numb_reduced_tap)*pp_lab_sam
p)*((hrs_analyze_samp_op*rate_op)+costJabJt_samp))+(((
pp_above_al_bin_three*numb_reduced_tap)*pp_commercia
1 samp)*cost commercial lab)





PWSs on triennial
reduced tap sampling
and not doing POU
sampling

pjtapjtrienniai

Every 3
years

The number of samples multiplied by the likelihoods for a
sample analyzed in house and a sample analyzed in a
commercial lab times the different labor and material cost
burdens for each type of analysis.

(((pp_above_al_bin_one*numb_samp_customer)*pp_lab_sa
mp)*((hrs_analyze_samp_op*rate_op)+cost_lab_lt_samp))+
(((pp_above_al_bin_one*numb_samp_customer)*pp_comm
ercial samp)*cost commercial lab)

Cost applies as written to NTNCWSs.

Above AL

All PWSs with at least
one sample > 10 |jg/L

Twice a
year

Final LCRI Economic Analysis

4-163

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost Per Activity

Lead 90th
- Range

Other Conditions

Frequency
of Activity

g) Collect distribution system WQP sample

The number of required samples per system above the AL
multiplied by the total of hours per sample times the system
labor rate, plus the material cost per sample. A system with
CCT only needs to collect an additional WQP monitoring
sample if there is not existing WQP monitoring done near
the site of the above the AL tap sample.

Cost does not apply to NTNCWSs.

All

PWSs with existing
CCT of pH and not
doing POU sampling

pbaseph

Once per
event

numb_wqp_sites_added

*((hrs_ wqp_dssa_op *rate_op)+cost_ wqp_material_ph)









The number of required samples per system above the AL
multiplied by the total of hours per sample times the system
labor rate, plus the material cost per sample. A system with
CCT only needs to collect an additional WQP monitoring
sample if there is not existing WQP monitoring done at or
near the site of the above the AL tap sample.

Cost does not apply to NTNCWSs.

All

PWSs with existing
CCT of PO4 or both
PO4 and pH
adjustment and not
doing POU sampling

Once per
event

numb_wqp_sites_added*((hrs_wqp_dssa_op*rate_op)+cost
wqp material ortho)





pbasepo4,
pbasephpo4,



h) Analyze distribution system WQP sample

The number of samples multiplied by the likelihoods for a
sample analyzed in house and a sample analyzed in a
commercial lab times the different labor and material cost
burdens for each type of analysis.

A system with CCT only needs to collect an additional WQP
monitoring sample if there is not existing WQP monitoring
done near the site of the above the AL tap sample.

((numb_wqp_sites_added*pp_lab_samp)*((hrs_wqp_analyz
e_ph_op*rate_op)+cost_wqp_ph_analyze))+((numb_wqp_si
tes added*pp commercial samp)*cost lab ph wqp)

Cost does not apply to NTNCWS

All

PWS with existing CCT
of pH and not doing
POU sampling

pbaseph

Once per
event

Final LCRI Economic Analysis

4-164

October 2024


-------
CWS Cost Per Activity

NTNCWS Cost Per Activity

Conditions for Cost to
Apply to a Model PWS

Lead 90th „ ....

„ Other Conditions
- Range

Frequency
of Activity

The number of samples multiplied by the likelihoods for a
sample analyzed in house and a sample analyzed in a
commercial lab times the different labor and material cost
burdens for each type of analysis.

A system with CCT only needs to collect an additional WQP
monitoring sample if there is not existing WQP monitoring
done near the site of the above the AL tap sample.

((numb_wqp_sites_added*pp_lab_samp)*((hrs_wqp_analyz
e_ortho_op*rate_op)+cost_wqp_ortho_analyze))+((numb_w
qp sites added*pp commercial samp)*cost lab ortho wq
P)

Cost does not apply to NTNCWS

All

PWSs with existing
CCT of PO4 or both
PO4 and pH
adjustment and not
doing POU sampling

pbasepo4,
pbasephpo4

Once per
event

i) Review incidents of systemwide event and other system conditions

The labor hours for review per system multiplied by the
system labor rate.

(hrs_deter_dssa_op*rate_op)

Cost applies as written to NTNCWSs.

All

All PWSs with at least
one sample > 10 |jg/L

Once per
event

j) Consult with the State prior to making CCT changes

The labor hours per system multiplied by the system labor
(hrs_consult_dssa_op*rate_op)

Cost applies as written to NTNCWSs.

All

All PWSs where a
second sampling
period has at least one
sample > 10 |jg/L

Once per
event

k) Report follow-up sample results and overall DSSA responses to the State

Hours for reporting multiplied by the system labor rate.
(hrs_report_dssa_op*rate_op)

Cost applies as written to NTNCWSs.

All

All PWSs with at least
one sample > 10 |jg/L

Once per
event

Acronyms: AL = action level; CCT = corrosion control treatment; CWS = community water system; NTNCWS = non-transient non-community water system; PO4

= orthophosphate; POU = point-of-use; PWS = public water system; WQP = water quality parameter.

Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of:

• pbaseph, pbasepo4, and pbasephpo4: Likelihood system has pH adjustment, orthophosphate, or pH adjustment and orthophosphate for their CCT
(Section 4.3.2.2.1).

Final LCRI Economic Analysis

4-165

October 2024


-------
•	pp_lab_samp and pp_commercial_samp: Likelihood that system will use in-house laboratory or commercial laboratory, respectively (Section
4.3.2.1.2).

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

2Systems on 9-year monitoring schedules cannot have any lead or copper in their entire distribution system including all buildings they serve and thus, none
should have any samples above 10 ng/L and be subject to distribution system and site assessment requirements.

3 The burden and costs to provide sample bottles (cost_samp_above_al) under activity e) and conduct analyses under activity f) are incurred by the State in
Arkansas, Louisiana, Mississippi, Missouri, and South Carolina (ASDWA, 2020a).

Final LCRI Economic Analysis

4-166

October 2024


-------
4.3.3.4 System Lead CCT Routine Costs

The EPA developed routine costs associated with CCT as shown in Exhibit 4-66. The exhibit provides the
unit burden each activity. The assumptions used in the estimation of each activity follows the exhibit.
The last column provides the corresponding SafeWater LCR model data variable in red/italic font.

Exhibit 4-66: PWS CCT Routine Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

1) Review CCT guidance

• 1 hr/system with CCT serving
> SOK/update1

hrs_rev_cct_op

m) Provide WQP data to the State
and discuss during sanitary
survey

• 1.5 to 3 hrs/system with CCT
per sanitary survey2

hrs_sanit_surv_op

n) Notify and consult with the
State if CCT actions are
required in response to source
water change

•	10 to 22 hrs/system on
reduced tap monitoring

•	6 to 12 hrs/system on
standard tap monitoring

hrs_coop_source_chng_red_op
hrs_coop_source_chng_rout_op

o) Notify and consult with the
State if CCT actions are
required in response to
treatment change

46 to 84 hrs/system

hrs_ coop_treat_ chng_ op

Acronyms: CCT = corrosion control treatment; WQP = water quality parameters.

Sources:

frequency of CCT guidance updates is assumed to be every 5 years.

2Sanitary surveys are conducted at least every 5 years for NTNCWSs and every 3 years for CWSs except where
ground water CWSs meet special performance criteria and are permitted to conduct sanitary surveys every 5 years
(p_spec_req).

I) & m): "CCT Study and Review Costs_Final.xlsx."

n): "Likelihood_SourceChange_Final.xlsx."

o): "Likelihood_TreatmentChange_Final.xlsx;" ASDWA, 2024.

Note:

0):	For the proposed LCRI EA, the EPA assumed a different burden for systems on standard and reduced
monitoring. For the final LCRI EA, the EPA used estimates from the ASDWA 2024 CoSTS model (ASDWA, 2024) and
assumed systems and States would incur the same burden to provide a report and conduct a review, respectively,
regardless of the system's monitoring schedule.

1)	Review CCT guidance (hrs_rev_cct_op). The EPA assumed that States will review new guidance and
determine applicability for systems serving 50,000 or fewer people.118 However, the EPA assumed
that systems serving more than 50,000 people will review the new CCT guidance themselves to
determine if CCT adjustment is needed and spend 1 hour on this review. The EPA assumed a
relatively small burden because the revised guidance is expected to include an executive summary
that can be used by large systems to quickly assess if new information is applicable to their system.
The EPA also assumed that the burden for systems to discuss updated guidance with the State is

118 See data input hrs_cct_reviewJs in Section 4.4.3.4, activity g).

Final LCRI Economic Analysis

4-167

October 2024


-------
already accounted for in the estimated burden to review CCT during the sanitary survey
(hrs_sanit_surv_op). See activity m) below.

m) Provide WQP data to the State and discuss during sanitary survey (hrs_sanit_surv_op). Systems
will incur burden to gather and submit non-compliance data (e.g., process control data, other WQP
data) and meet with their State during the sanitary survey to determine if CCT is still optimized. The
EPA assumed that documents are submitted electronically or provided on-site. The EPA assume 0.5
- 2 hours depending on system size for gathering and submitting data to the State, and 1 hour to
discuss this information as well as any relevant updated CCT guidance during the sanitary survey, as
shown in Exhibit 4-67.

Exhibit 4-67: Estimated PWS Burden to Gather Data and Review CCT-Related Data during
Sanitary Survey to Determine if CCT Is Still Optimized

System Size (Population Served)

SafeWater LCR Data Variable:
hrs_sanit_surv_op

<1,000

1.5

1,001-10,000

2.0

10,001-100,000

2.5

>100,000

3.0

Source: "CCT Study and Review Costs_Final.xlsx.

In addition to the unit costs, the SafeWater LCR model requires the frequency of the sanitary survey
as an input to calculate total costs for this activity. The required frequency of sanitary surveys is
based on system size and water type as follows:

•	The minimum frequency for all NTNCWSs is once every 5 years.

•	The minimum frequency for surface water CWSs is once every 3 years.

•	The minimum frequency for ground water CWSs is 3 years but can be extended to 5 years if
systems provide 4-log treatment of viruses (using inactivation, removal, or a State-approved
combination of these technologies) before or at the first customer or have an outstanding
performance record (e.g., past sanitary surveys with no significant deficiencies).

To determine the percent of ground water systems that meet the criteria for a minimum frequency
of 5 years (p_spec_req), the EPA used Exhibits 4.2 and 4.3 from the Economic Analysis for the Final
Ground Water Rule (USEPA, 2006a) that provide information on the estimated percentage of ground
water systems meeting the 4-log removal criteria. These estimates are presented in Exhibit 4-68.
These may be an underestimation because this approach does not capture systems with
outstanding performance that would also qualify for a 5-year sanitary survey frequency.

Final LCR! Economic Analysis

4-168

October 2024


-------
Exhibit 4-68: Estimated Percent of Ground Water CWSs Achieving 4-log Virus Inactivation



Data from the Economic Analysis for the Final
Ground Water Rule (USEPA, 2006a)









Total No. of
Entry Points with
4-log
Inactivation



Estimated GW

Percent of

System Size
(Population
Served)

Total No. of

GW CWSs

Average No. of
Entry points per
system

CWSs getting 4-
Log In at all
Entry Points

GW CWSs
getting 4-log

(p_spec_req)



A

B

C

D = B/C

<
q"

II

LU

<100

12,843

3,996

1.3

3,074

23.9%

101-500

14,358

8,873

1.6

5,546

38.6%

501-1,000

4,649

3,547

2

1,774

38.2%

1,001-3,300

5,910

5,378

2.4

2,241

37.9%

3,301-10,000

2,884

3,547

3.2

1,108

38.4%

10,001-50,000

1,445

3,856

5.6

689

47.7%

50,001-100,000

168

583

11.3

52

31.0%

100,001-1,000,000

103

545

12.4

44

42.7%

> 1,000,000

3

34

11.4

3

100.0%

Total

42,363

30,359



14,531

34.3%

Acronyms: CWS = community water system; GW = ground water.

Source: "CCT Study and Review Costs_Final.xlsx."

Notes:

A: Economic Analysis for the Final Ground Water Rule, Exhibit 4.2, Columns F plus K (USEPA, 2006a).

B: Economic Analysis for the Final Ground Water Rule, Exhibit 4.3, Column H (USEPA, 2006a).

C: Economic Analysis for the Final Ground Water Rule, Exhibit 4.3, Column A (USEPA, 2006a).

D: Assumed that systems that provide 4-log inactivation do so at all entry points in their system, and these systems

have the same number of entry points as other systems.

n) Notify and consult with the State on required actions in response to source water change

(hrs_coop_source_chng_red_op, hrs_coop_source_chng_rout_op). Systems are required to seek
prior approval before making any source water changes and to consult with the State on needed
responses including the possibility of CCT installation. The likelihood of a system changing source
(p_source_chng) is discussed in Chapter 3, Section 3.3.9.1 with estimated percentages for CWSs and
NTNCWSs presented in Exhibit 3-53 and Exhibit 3-54, respectively. Exhibit 4-69 below provides the
estimated system burden to report the source change and consult with the State for systems on
reduced and standard tap monitoring. Note that the EPA estimated fewer hours for consultation for
systems on standard monitoring because they are in more frequent contact with the State
compared to those on reduced monitoring.

Final LCRI Economic Analysis

4-169

October 2024


-------
Exhibit 4-69: Estimated Hours per System to Report and Consult on Source Water Change

Hours for systems on reduced monitoring to
report a source change

(hrs_coop_source_chng_red_op)

Hours for systems on standard
monitoring to report a source change

(hrs_coop_source_chng_rout_op)

A

B

Minimum

Maximum

Most Likely

Minimum

Maximum

Most Likely

10

22

10

6

12

6

Source: "Likelihood_SourceChange_Final.xlsx."

Notes:

A: Applies to systems that are conducting reduced lead tap monitoring less frequently than every 6 months. The
estimates are based on input received from North Carolina and Indiana in response to a 2016 ASDWA
questionnaire regarding potential 2021 LCRR requirements. A copy of the questionnaire and each State's
responses are available in the docket under EPA-HQ-OW-2022-0801 at www.regulations.gov. North Carolina
estimated 2 hours to review a change in source from ground water to another ground water source and 3 hours
for surface water source changes or surface water/ground water mixing. Indiana estimated 6 hours to review a
change to a similar source and 20 hours to review a change to a dissimilar source. The EPA used the average of the
two State estimates of 2 and 6 hours (4 hours), doubled to 8 hours for systems, for the minimum and most likely
value. The EPA set the most likely equal to the minimum because fewer than 1 percent of systems made more
significant sources changes during 2013 - 2016. For the maximum, the EPA assumed the 20 hours were more
reflective of the system burden to prepare needed documentation. To each estimate, the EPA assumed an
additional 2 hours for consultation with the State on needed action in response to the source change.

B: Applies to systems conducting standard lead tap monitoring every six months under the final LCRI. Because
these systems are in more frequent contact with the State, the EPA assumed 50 percent of the burden estimated
to prepare and submit the documentation for hrs_coop_source_chng_red_op or 50 percent of 8 hours for the
minimum and most likely and 50 percent of 20 hours for the maximum plus an additional 2 hours for consultation.
This equals a total burden of 6 hours for the minimum and most likely and 12 hours for the maximum.

o) Notify and consult with the State on required actions in response to treatment change

(hrs_coop_treat_chng_op). Systems are required to seek prior approval before making any long-
term treatment changes to ensure that corrosion control is maintained. The estimated likelihood of
a system changing treatment in a given year of 4.2 percent for all CWSs and 3.2 percent for all
NTNCWSs (p_treat_change) is discussed in Chapter 3, Section 3.3.9.3 with percentages for CWSs and
NTNCWSs presented in Exhibit 3-55 and Exhibit 3-56, respectively. Exhibit 4-70 below provides the
burden for systems to report the change and consult with the State. Note that for the proposed
LCRI EA (USEPA, 2023c), the EPA assumed a different burden for systems on standard and reduced
monitoring. For the final rule, the EPA increased the burden estimates based on the ASDWA 2024
CoSTS model (ASDWA, 2024) and assumed systems and States would incur the same burden to
report the change conduct a review, regardless of the system's monitoring schedule.

Final LCRI Economic Analysis

4-170

October 2024


-------
Exhibit 4-70: Estimated Hours per System to Report and Consult on Treatment Change

System Size
(Population Served)

Hours for systems to report a treatment
change and consult with the State

(hrs_coop_treat_chng_ op)

<100

46

101-500

46

501-1,000

46

1,001-3,300

46

3,301-50,000

84

>50,000

82

Source: "Likelihood_TreatmentChange_Final.xlsx;"; ASDWA, 2024.

Note: The estimates are based on ASDWA's 2024 CoSTS (ASDWA, 2024).

Exhibit 4-71 details how the data variables are used to estimate routine system activities related to CCT.

Exhibit 4-71: PWS Lead CCT Routine Cost Estimation in SafeWater LCR by Activity1





Conditions for Cost to Apply to a Model PWS

CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency of
Activity

1) Review CCT guidance

Hours per system multiplied by
the system labor rate.

(hrs_rev_cct_op * rate_op)

Cost applies
as written to
NTNCWSs.

All

Model PWSs with CCT
serving >50,000 people

Once per

Sanitary

Survey2

m) Provide WQP data to the State and discuss during sanitary survey2

Hours per system multiplied by
the system labor rate.

(hrs_sanit_surv_op * rate_op)

Cost applies
as written to
NTNCWSs.

All

Model PWSs with CCT

Once per

Sanitary

Survey2

n) Notify and consult with the State on response to a change in source water

The total hours per system
multiplied by the system labor
rate.

(hrs_coop_source_chng_rout_op*
rate_op)



At or below
AL

Model PWS that is not on
reduced tap sampling with
a change in source water

1 - (p_tap_annual +
p_tap_triennial +
p_tap_nine) *
p_source_chng





Cost applies
as written to
NTNCWSs.

Above AL

Model PWSs with a change
in source water
p source chng

Once per event

Final LCR! Economic Analysis

4-171

October 2024


-------




Conditions for Cost to Apply to a Model PWS

CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions

Frequency of
Activity

The total hours per system
multiplied by the system labor
rate.

(hrs_coop_source_chng_red_op*r
ate_op)



At or below
AL

Model PWS that is on
reduced tap sampling with
a change in source water
(p_tap_annual +
p_tap_triennial +
p_tap_nine) *
p_source_chng



o) Notify and consult with the State on response to a change in water treatment

The total hours per system
multiplied by the system labor
rate.

(hrs_coop_treat_chng_
op*rate_op)

Cost applies
as written to
NTNCWSs.

All systems

Model PWS with a change
in treatment
p_treat_change

Once per Event

Acronyms: CCT = corrosion control treatment; CWS = community water system; NTNCWS = non-transient non-

community water system; PWS = public water system.

Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of:

•	p_tap_annual, p_tap_triennial, and p_tap_nine: Likelihood a system will qualify to collect the reduced
number of lead tap samples at an annual, triennial, and nine-year frequency, respectively (Chapter 3,
Section 3.3.7.3).

•	p_source_chng\ Likelihood that a system will change sources in a given year (Chapter 3, Section 3.3.9.1).

•	p_treat_change: Likelihood that a system will change treatment in a given year (Chapter 3, Section
3.3.9.3).

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

4.3.3.5 Estimate of PWS National Corrosion Control Treatment Costs

Exhibit 4-72 shows the estimated national costs of CCT under the low and high cost scenarios, for the
2021 LCRR, the final LCRI, and the incremental cost, discounted at 2 percent. The monetized incremental
annual CCT costs range from $39.1 million to $45.1 million in 2022 dollars. The CCT Operation and
Maintenance (Existing) category in these exhibits are the EPA's estimate of the ongoing cost of
operating corrosion control at PWSs where CCT was in place at the beginning of the period of analysis.119

119 For additional context the average CCT cost per household for large systems (serving 10,000 or more people) is
$10.56 per year.

Final LCRI Economic Analysis

4-172

October 2024


-------
Exhibit 4-72: Estimated National Annualized Corrosion Control Costs - 2 Percent Discount Rate (millions of 2022 USD)





Low Estimate





High Estimate





Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

CCT Operations and Maintenance
(Existing)

$458.2

$458.2

$0.0

$458.1

$458.1

$0.0

CCT Related Sanitary Survey and
Source or Treatment Change
Notification Activities

CCT Installation

$2.5
$19.6

$5.1
$45.1

$2.6
$25.5

$2.5
$50.1

$5.1
$83.8

$2.6
$33.7

CCT Installation Ancillary Activities

$6.2

$4.2

-$2.0

$10.9

$6.4

-$4.5

CCT Re-Optimization

CCT Re-Optimization Ancillary
Activities

Distribution System and Site
Assessment (DSSA)

Ancillary DSSA Activities

$39.2
$5.9

$4.8
$15.6

$32.6
$7.6

$15.0
$23.3

-$6.6
$1.7

$10.2
$7.7

$82.7
$14.6

$10.6
$18.3

$71.7
$13.5

$27.2
$27.1

-$11.0
-$1.1

$16.6
$8.8

Total Annual Corrosion Control
Technology Costs

$552.0

$591.1

$39.1

$647.8

$692.9

$45.1

Acronyms: CCT = corrosion control treatment; LCRI = Lead and Copper Rule Improvements; USD = United States dollar.

Final LCRI Economic Analysis

4-173

October 2024


-------
4.3.4 PWS Service Line Inventory and Replacement Costs

This section provides burden and cost estimates for inventory related activities and SLR-related activities
under the final LCRI as follows:

•	Section 4.3.4.1: Service Line Inventory, provides inputs related to classifying connector material
in the updated LCRR initial inventory, preparing and submitting annual inventory updates, and
inventory validation.

•	Section 4.3.4.2: Service Line Replacement Plan, provides inputs related to the development of a
service line replacement plan.

•	Section 4.3.4.3: Physical Lead Service Line Replacement, provides inputs for replacements of
LSLs and GRR service lines.

•	Section 4.3.4.4: Ancillary Lead Service Line Replacement Activities, includes inputs for activities
that are not related to the service line inventory nor physical replacements.

National annualized inventory and LSLR-related costs are presented at a 2 percent discount rate in
Section 4.3.4.5.

4.3.4.1 Service Line Inventory

The discussion of service line inventory costs for water systems is presented in three subsections as
follows:

•	4.3.4.1.1: Updating the LCRR Initial Inventory to Include Connector Materials

•	4.3.4.1.2: Inventory Updates

•	4.3.4.1.3: Inventory Validation

Exhibit 4-83 at the end of Section 4.3.4.1 is a summary exhibit that indicates how the cost inputs are
modeled by the SafeWater LCR model. Note that the 2021 LCRR required systems to prepare an initial
inventory by October 16, 2024, which is prior to the EA's period of analysis. Therefore, the cost for
preparing the LCRR initial inventory, that does not include connector material information, is not
included in the final LCRI analysis in this section nor the pre-2021 LCR baseline cost analysis in Appendix
B. The LCRR initial inventory costs are outside of the period-of-analysis for the EA.

4.3.4.1.1 UpdatinR the LCRR Initial Inventory to Include Connector Materials

The EPA has developed system costs for activities associated with the review of records for connector
materials as shown in Exhibit 4-73. The assumptions used in the estimation of each activity follows the
exhibit. The last column provides the corresponding SafeWater LCR model data variable in red/italic
font.

Final LCRI Economic Analysis

4-174

October 2024


-------
Exhibit 4-73: PWS Service Line Inventory Connector Review Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

a) Conduct records review for
connector materials

0.5 to 7,599 hours per CWS per year

hrs_updated_initial_inv_op

b) Compile and submit
connector updated LCRR
initial inventory (baseline
inventory1) to the State

1 to 4 hrs /CWSs; 3.75 to 15
hrs/NTNCWS

hrs_report_ updated_initial_in v_op

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.

Sources: Data sources for each activity are provided following this exhibit.

Note:

1 Section 141.84(a)(2) of the final LCRI states: "All water systems must develop an updated initial inventory, known
as the "baseline inventory."

Under the final LCRI, all water systems must update the LCRR initial inventory with information on
connectors and submit it to the State three years after the publication of the final rule (or Year 3 of the
rule analysis period). To develop the connector updated initial inventory (referred to in the final LCRI as
the "baseline inventory"120), water systems must review any information listed below that describes
connector material and location:

•	All construction and plumbing codes, permits, and records or other documentation that indicate
the service line and connector materials used to connect structures to the distribution system.

•	All water system records on service lines and connectors, including distribution system maps
and drawings, recent or historical records on each service connection and connector, meter
installation records, historical capital improvement or master plans, and standard operating
procedures.

•	All records of inspections in the distribution system that indicate the material composition of
the service connections and connectors that connect a structure to the distribution system.

Water systems must include each connector in their service line inventory and categorize it as lead, non-
lead, unknown, or no connector present. If systems have already reviewed applicable records and
categorized each connector in their inventory by material type and location, they are not required to re-
review records.

Key assumptions for estimating the burden and cost for systems to conduct this records review and
prepare and submit the updated LCRR initial inventory under the final LCRI are as follows:

•	All CWSs, regardless of the extent of lead content service lines121 and unknowns, will incur
burden to review records for connector material and submit the updated LCRR initial inventory.

120	Note § 141.84(a)(2) of the final LCRI states: "All water systems must develop an updated initial inventory,
known as the "baseline inventory." Systems must submit the baseline inventory to the State by the compliance
date in § 141.80(a)(3)." The EPA is using the term "updated LCRR initial inventory" in place of "baseline inventory"
in the EA given the potential for confusion with the economic analysis concept of the baseline.

121	As described in Chapter 3, Section 3.3.4.1, the EPA defines "lead content service lines" as those with lead lines,
GRR, lead connectors, and galvanized previously downstream of lead connectors.

Final LCRI Economic Analysis

4-175

October 2024


-------
CWSs will not incur additional burden to make the inventory publicly accessible since this was
required under the 2021 LCRR.

•	The burden to review records for connector material is similar to the burden to review records
for the initial inventory because the records required to be reviewed are similar. However, the
EPA assumes that the review for connector material will be less burdensome due to several
factors that will be described in this section.

•	NTNCWSs will already have documentation of connector material because they own their own
service lines, but will require time to gather the information and prepare a package for the
State.

The EPA developed unit burden estimates separately for records review (activity a) for CWSs only) and
reporting (activity b) for CWSs and NTNCWSs) as described below. Note that all calculations and
assumptions are documented in the derivation file, "LCRI Updated Initial Inventory with
Connectors_Final.xlsx" available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

a) Conduct records review for connector materials (hrs_updated_initial_inv_op)

As noted previously, the EPA assumed that the unit burden for records review under the final LCRI is
similar to the burden for records review under the 2021 LCRR. Therefore, the EPA used an estimate of
the burden for the 2021 LCRR records review as a starting value for the burden associated with
conducting the LCRI required connector material records review. In the Economic Analysis for the Final
Lead and Copper Rule Revisions (hereafter referred to as the "Final 2021 LCRR EA") (USEPA, 2020), the
EPA estimated the burden for CWSs to conduct their records review based on limited information from
water systems and States but since that time, new information has become available from the following
key data sources:

•	CDM Smith. 2022. Considerations when Costing Lead Service Line Identification and
Replacement American Water Works Association. This report includes responses from a survey
of AWWA members regarding costs for developing a service line inventory. A total of 34 systems
responded to the survey, representing 23 States and a wide range of system sizes.

•	Liggett J. et al. 2022. Service Line Material Identification: The Experiences from North American
Water Systems. American Water Works Association. This study provides results from 11 case
studies of systems that have already completed their service line inventory.

•	Responses to the EPA questions regarding the time needed to develop and maintain the LCR
Service Line Inventory (USEPA, 2023e). This questionnaire was sent to nine systems in early
2023; EPA received responses from Grand Rapids, Ml; Pittsburgh, PA; and Cincinnati, OH.

As the first step in this analysis, the EPA used these data sources to develop a revised unit burden
(hours/CWS) for records review under the 2021 LCRR as follows:

•	The EPA compiled system-level estimates of cost or burden for reviewing records (hours or $/SL)
in the worksheet "Records Rev per SL Cost Input" in the derivation file "LCRI Updated Initial
Inventory with Connectors_Final.xlsx." The EPA included findings from the CDM Smith Report
(CDM Smith, 2022) for systems reporting new and previous records review efforts to represent
the range of real-world scenarios. Where the data were reported as burden (hours), the EPA

Final LCRI Economic Analysis

4-176

October 2024


-------
used estimated labor rates in Chapter 3, Section 3.3.11.1 to convert all data to cost per service
line reviewed.

•	The EPA reviewed data sources and removed values that included additional activities beyond
records review, such as field investigation.

•	The EPA analyzed the data set for outliers122 and removed them from the analysis.

The result was 17 individual estimates ranging from $0.01 to $16.22 per service line reviewed, with an
average labor cost of $3.76 per service line for the 2021 LCRR records review. The EPA used this value
for all system sizes. This cost represents a range of 4 to 7 minutes per record using the PWS labor rates
in Section 3.3.11.1.

Secondly, the EPA estimated the percentage of 2021 LCRR records review cost that will be incurred to
review records for connector material under the final LCRI. The EPA assumed that systems would spend
an average of 75% of the 2021 LCRR burden to review records again for the LCRI ($3.76/SL x 0.75 =
$2.82/SL) to account for the following factors:

•	Systems will have already identified key sources of information including plumbing codes and
construction standards under the 2021 LCRR.

•	Some systems will have digitized paper records under the 2021 LCRR allowing faster records
reviews for connector information under the final LCRI.

•	Systems are not required to review previous material evaluations under the LCRI as they were
under the 2021 LCRR.

•	The EPA service line inventory guidance (USEPA, 2022b) recommends that systems track
connector materials, thus some system may have already reviewed records for connectors when
preparing their inventory under the 2021 LCRR.

As the third step, the EPA evaluated which service lines would be exempt from the records review for
connector material. The EPA identified two scenarios under which systems could classify connector
materials without a records review: (1) where service lines were installed after the lead ban, and (2)
where the service lines are known to be lead and/or have lead connectors. The EPA's approach to
accounting for these scenarios is described as follows:

•	Systems can use the date of their local lead ban to identify buildings constructed after the lead
ban and assume those service line connectors to be non-lead. The EPA reviewed housing stock
data from the 2020 American Community Survey (U.S. Census Bureau, 2021) to estimate the
percent of 1-unit detached or attached buildings built in 1990 or later, recognizing that this is
conservative because local lead bans may have occurred earlier123. The EPA found that a total of

122	Values were determined to be outliers if they fell outside the upper and lower bounds determined by quartile 1
minus 1.5 time the interquartile range, and quartile 3 plus 1.5 times the interquartile range (Verardi and
Vermandele, 2018).

123	The 1986 Safe Drinking Water Act (SDWA) Amendments prohibited the use of pipe, solder, and flux that were
not "lead free" as defined in 1986 in new installations and repairs and directed states, as a condition of receiving
grants for the Public Water System Supervision (PWSS) program, to enforce the provision effective 24 months after

Final LCRI Economic Analysis

4-177

October 2024


-------
33% of housing in the United States was constructed 1990 or later (see the worksheet
"2020Physical Housing Char" in the derivation file "LCRI Updated Initial Inventory with
Connectors_Final.xlsx" for the data and calculations).

• For the second scenario, the EPA assumed that systems would classify the connector as lead for
service lines that are lead. Moreover, service lines that are reported as having a lead connector
in the 7th DWINSA124 would not require a records review because systems would already know
the connector material. The EPA identified the percent of service connections that are LSLs
(partial or full) and the percent that are lead connectors using the data presented in Section
3.3.4 of Chapter 3.

As the last step, the EPA used connection information from SDWIS/Fed, and PWS labor rates from
Chapter 3, Section 3.3.11.1, and the assumptions above to calculate an average unit burden per CWS
for records review for connector materials under the final LCRI. Exhibit 4-74 provides the calculations
and results of this analysis. Note that the final unit burden for records review in Column J (SafeWater
LCR model input hrs_updated_initial_inv_op) has been converted to an annual cost by taking the total
estimated burden in Column I and dividing it by 3 assuming that the burden is spread equally over the
first three years of the rule analysis period.

For NTNCWSs, the EPA assumes that they will not incur burden for this activity because they own their
own service lines and thus will already have documentation of connector material. However, the EPA
estimates that they will incur burden to compile the information, which is included with the reporting
costs in activity b).

June 19,1986, through state or local plumbing codes or other means (42 U.S. Code §300g-6(b)). Some states
adopted their own laws before the federal requirement. Appendix D of the EPA Service Line Inventory Guidance
(USEPA, 2022b) contains a summary of lead ban provisions by state. Nearly all of the states enacted the lead ban
by 1989, so the EPA searched for 1-unit detached or attached buildings built in 1990 or later.

124 Note that between the proposed and final LCRI EA, the EPA updated the service line material characterization
based on the results of the one-time update to the 7th DWINSA. For more information, see Sections 3.2.5 and
3.3.4. For the purposes of this EA, the term "7th DWINSA" includes results from the original survey and the one-
time update.

Final LCRI Economic Analysis

4-178

October 2024


-------
Exhibit 4-74: Estimated Unit Burden for CWSs to Review Records for Connector Material

System Size
(Population
Served)

System
Labor Rate
(S/hr)

Records Review for Connector Material (Burden / CWS)

Average
Number of
Service Lines
(SLs) /CWS

Percent of
SLs that
Were
Installed in
1990 or later

Percent of
SLs that are
Known LSLs

or Lead
Connectors

Average
Number of SL
Records/CWS
that Need to
be Reviewed
for Connector
Material

$ /SLfor
Records
Review for the
Initial
Inventory

$/SLfor
Records
Review for
Connector
Material

Total $/CWS
($2020) to

Review
Records for
Connector
Material

Estimated
Total Hrs/CWS

to Conduct
Records review
for Connector
Material

Estimated Annual

Burden
(hrs/CWS/yr) to
Conduct Records
Review for
Connector Material
(hrs_updated
initial_inv_op)

A

B

C

D

E = B*(l-C-D)

F

G = F*0.75

H = E*G

1 = H/A

J = 1/3

<100

$33.39

27

33.0%

0.29%

18

$3.76

$2.82

$51

2

0.5

101-500

$33.39

107

33.0%

0.29%

71

$3.76

$2.82

$200

6

2

501-1,000

$33.39

304

33.0%

0.29%

202

$3.76

$2.82

$571

17

6

1,001-3,300

$33.39

759

33.0%

0.29%

506

$3.76

$2.82

$1,427

43

14

3,301-10,000

$39.81

2,225

33.0%

4.08%

1,400

$3.76

$2.82

$3,948

99

33

10,001-50,000

$42.68

7,880

33.0%

4.51%

4,923

$3.76

$2.82

$13,887

325

108

50,001-100,000

$46.11

23,344

33.0%

4.89%

14,495

$3.76

$2.82

$40,891

887

296

100,001-1,000,000

$52.42

78,750

33.0%

3.52%

49,978

$3.76

$2.82

$140,990

2,690

897

>1,000,000

$52.42

670,176

33.0%

3.78%

423,602

$3.76

$2.82

$1,194,997

22,797

7,599

Acronyms: CWS = community water system; LSL = lead service line; SL = service line.

Notes:

General: Applies to all CWSs assuming none have done a review of records for connector materials that includes all of the material classification categories
under the LCRI (not required by any States). Assume that the estimated annual records review burden occurs each year in 2025, 2026, and 2027.

A.	PWS Labor Rates as presented in Chapter 3, Section 3.3.11.1.

B.	Chapter 3, Exhibit 3-14. Based on connection data from SDWIS 4th quarter 2020 frozen dataset, current through December 31, 2020. Adjusted for systems
serving < 100 people if the reported population was less than 25 or if the number of people per connection was less than 1 or greater than five.

C.	Derivation file "LCRI Updated Initial Inventory with Connectors_Final.xlsx", worksheet "2020Physical Housing Character." Based on the percent of single
family detached and attached homes in the U.S. that were built in 1990 or after, which is after the lead ban took effect (U.S. Census Bureau, 2021). Assumes
that all connectors installed after the lead ban took effect would be classified as "Never Lead."

D.	Percent of systems with lead content (p_lsl) times percent of service lines in those systems that are unknown (perc_lsl_known) times the percent of service
lines that are known lead (perc_lsl_known_lead) times the sum of the percent of known LSLs that are full or partial LSLs and lead connectors (pp_lsl_full +
pp_lsl_partial + pp_lsl_connector). See Chapter 3, Exhibits 3-10, 3-15, and 3-19 for these values. The EPA assumes that systems will be able to assign these
connectors as "lead" without needing to review records again.

Final LCRI Economic Analysis

4-179

October 2024


-------
F. Based on analysis of system-level burden and cost estimates per service line reviewed. See the derivation file "LCRI Updated Initial Inventory with
Connectors_Final.xlsx", worksheets "Records Rev per SL Cost Input" and "Records Rev Data Analysis."

Final LCRI Economic Analysis

4-180

October 2024


-------
b) Compile and submit connector updated LCRR initial inventory (baseline inventory) to the State
(hrs_report_updated_initial_inv_op)

The EPA assumed that in addition to the records review, CWSs will incur additional burden to compile
connector material information and submit an updated LCRR initial inventory to their State. The EPA
based this burden estimate on the burden to compile and submit the service line inventory under the
2021 LCRR, which was 10 to 40 hours depending on system size (USEPA, 2020). The EPA assumed this
burden for the final LCRI to be much smaller than the burden for compiling and submitting the LCRR
initial inventory because the inventory structure is already in place. The EPA assumed that systems will
add one column to their LCRR initial inventory to capture connector material information and resubmit
it. The EPA assumed that this effort to capture connector information and resubmit it will take
approximately 10 percent of the effort to compile and submit the 2021 LCRR initial inventory. See
Exhibit 4-75 for the estimated burden.

For NTNCWSs, the EPA assumes that they will already have documentation of connector material
because they own their own service lines, but will require time to gather the information and prepare a
package for the State. Consistent with assumptions for CWSs, the EPA used the burden estimate for
compiling and submitting the 2021 LCRR initial inventory, which was 5 to 20 hours per NTNCWS
depending on system size, as a starting point (USEPA, 2021). The EPA used a higher percentage for
NTNCWSs compared to CWSs because NTNCWSs need to compile information in addition to reporting.
The EPA assumed that NTNCWSs will incur 75 percent of the burden used for the 2021 LCRR inventory
to compile and submit an updated LCRR initial inventory with connector material under the final LCRI.
See Exhibit 4-75 for the estimated burden.

Exhibit 4-75: Estimated Unit Burden for CWSs and NTNCWSs to Compile and Submit the

Connector Updated LCRR Initial Inventory



Hours per system to compile and submit connector updated LCRR initial
inventory SafeWater LCR input (hrs_report_updated_initial_inv_op)

System Size (Population
Served)

CWS

NTNCWS



A

B

<100

1

3.75

101-500

1

3.75

501-1,000

1

3.75

1,001-3,300

1

3.75

3,301-10,000

2

7.5

10,001-50,000

2

7.5

50,001-100,000

4

15

100,001-1,000,000

4

15

>1,000,000

4

N/A

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.

Note: Applies to all CWSs and NTNCWSs. The EPA assumes these burdens are incurred in Year 3 of the rule analysis

period.

Sources:

Final LCRI Economic Analysis

4-181

October 2024


-------
A: The EPA assumed that in addition to the records review, CWSs will incur additional burden to compile connector
material information and submit an updated LCRR initial inventory) to their State. The EPA assumed this burden to
be much smaller than the burden for compiling and submitting the LCRR initial inventory because the inventory
structure is already in place. The EPA assumed that this effort will take approximately 10 percent of the effort to
compile and submit the 2021 LCRR initial inventory. See the derivation file "LCRI Updated Initial Inventory with
Connectors_Final.xlsx," worksheet "CWS Rep Updated Inv."

B: The EPA assumes that NTNCWSs will already have documentation of connector material because they own their
own service lines, but will require time to gather the information and prepare a package for the State. The EPA
assumes that preparing and submitting the connector updated LCRR initial inventory under the LCRI will take
approximately 75 percent of the burden to prepare the 2021 LCRR initial inventory, not including the hours to
prepare a tracking system for NTNCWSs with lead content service lines. See the derivation file "LCRI Updated Initial
Inventory with Connectors_Final.xlsx", worksheet "NTNCWS Rep Updated Inv."

4.3.4.1.2 Inventory Updates

Under the 2021 LCRR, systems are required to update their initial inventory and submit annual updates
to their State beginning October 16, 2025. Under the final LCRI, systems must continue making annual
updates, tracking both changes in service lines and connector materials, starting the year after they
submit their connector updated LCRR initial inventory. The connector updated LCRR initial inventory is
expected to be due on October 16, 2027, so the first annual update under the final LCRI would be by
October 16, 2028.

Under the 2021 LCRR and continued under the final LCRI, inventory updates must reflect replacements
of lead or GRR service lines and service line material inspections. Under the final LCRI, inventory updates
must also reflect lead connector replacements. Under both the 2021 LCRR and final LCRI, the inventory
must also be updated with any other resource, information, or identification method allowed or
required by the State to assess service line and connector materials. Moreover, water systems must
identify service line and connector materials as they are encountered in the course of normal operations
(e.g., checking service line materials when reading water meters or performing maintenance activities)
and use this information to update their inventory including the associated addresses.

Under the final LCRI, lead connectors must be replaced when encountered during planned or unplanned
water system infrastructure work unless the connector is not under the control of the system. Because
connectors are not required to be identified or replaced on a specific schedule under the final LCRI, the
EPA assumes that they will most often be identified and replaced in tandem with service line inspections
and replacement. The EPA assumes that the incremental burden for updating the connector material
while updating the service line material in the inventory would be minimal. Thus, the EPA assumes that
the system burden and costs associated with inventory updates are similar under the final LCRI and the
2021 LCRR and presents a single unit cost in this section for updating the service line inventory.

The activities associated with the update of the service line inventory are shown in Exhibit 4-76. The
assumptions used in the estimation of each activity follows the exhibit. The last column provides the
corresponding SafeWater LCR model data variable in red/italic font.

Final LCRI Economic Analysis

4-182

October 2024


-------
Exhibit 4-76: PWS Service Line Inventory Update Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

c) Identify material for
unknown service lines

$35.94 to $52.55 per unknown service
line investigated

cost_update

d) Report annual inventory
updates to the State

1 hr per CWS or NTNCWS per year for
systems with lead content and/or
unknowns

hrs_report_inv_op

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.
Sources: Data sources for each activity are provided following this exhibit.

Key assumptions for estimating the burden and cost for inventory updates are as follows:

•	NTNCWSs will not incur burden or cost to update their inventory since they own their own
service lines are not expected to have any unknowns. NTNCWSs with lead content service lines
will incur burden, however, to submit annual inventory updates to reflect SLRs.

•	Burden and costs to identify service lines material only apply to CWSs with unknown service
lines. CWSs that did not report any unknowns (e.g., systems with all non-lead or systems with a
mix of lead and non-lead) are expected to have good data on their service line material and do
not need to conduct investigations.

•	CWSs will use a combination of two methods to determine the service line material of
unknowns: (1) identify material during normal operation as required under the final LCRI, and
(2) conduct field investigations.

•	Field investigations will focus on the customer-owned portion of the service line.

•	The EPA assumes that CWSs with unknowns will investigate them at an average rate of 10
percent per year starting in Year 1 of the period of analysis. This assumption is based on the
requirement that all lead status unknown service lines be identified by the mandatory SLR
deadline, which is 10 years unless the State has set a shorter schedule or approved a deferred
rate under the final LCRI.

This section provides burden and cost estimates separately for (c) identifying service line material for
unknowns for CWSs only and (d) reporting inventory updates for CWSs and NTNCWSs.

c) Identify material for unknown service lines (cost_update)

As noted previously, the EPA estimates that systems will use a combination of two methods to identify
unknowns: (1) identify materials during normal operation, and (2) perform field investigations. These
methods are described below, followed by the approach for combining the costs of the two methods to
produce a weighted average cost per service line.

Final LCRI Economic Analysis

4-183

October 2024


-------
(1)	Normal Operations

To estimate the unit burden for identifying service line material during normal operation, the EPA used
information from the CDM Smith report (CDM Smith, 2022). In Section 3.3.3, CDM Smith noted that
"The study team assumed that the additional time required above and beyond the time already devoted
to the inspection would be an estimated 30 minutes of staff time to see the material, record the
information, take photographs and then update the inventory per service line." The EPA used this
estimate of 30 minutes, or 0.5 hours per service line for systems serving < 100,000 people. For systems
serving > 100,000 people, the EPA assumed CWSs would use an automated process to update the
inventory such as the enhanced work order process developed by the Pittsburg Water and Sewer
Authority (PWSA) to capture and upload service line material data during meter replacements (USEPA,
2023e). The EPA estimated that use of an automated process would require less burden at
approximately 10 minutes, or 0.2 hours per service line.

To estimate the percent of unknows that could be identified during normal operations, the EPA
evaluated the frequency of activities that could expose service line material. The 2022 inventory
guidance manual (USEPA, 2022b) identifies the following opportunities for data collection under normal
operations:

•	Water meter reading

•	Water meter repair or replacement

•	Service line repair or replacement

•	Water main repair or replacement

•	Backflow prevention inspections

•	Other street repair or capital projects with open cut excavation

To estimate the frequency of these events, the EPA made the following assumptions:

•	Meters are replaced an average of 6 percent per year based on an average meter lifespan of
approximately 15-20 years based on information from the city of Pasadena, Texas (Pasadena, no
date).

•	Water mains are replaced at a typical rate of 1 percent per year based on total installed length
(Folkman, 2018), and that these replacements result in an opportunity to inspect a proportional
percent of service connections.

The EPA made a conservative assumption that an additional 0.5 percent of service lines are exposed
during water meter reading, service line repair or replacement, backflow prevention inspections, and
other street repair or capital projects with open cut excavations. Thus, the total percent of service
connections that could be inspected each year during normal operations is estimated at 7.5 percent.

(2)	Field investigations

To estimate the costs for field investigations, the EPA used a 3-step process as described below.

Final LCRI Economic Analysis

4-184

October 2024


-------
Step 1: Identify commonly used field investigation methods: The EPA reviewed information on the use
of service line investigation methods as presented in the EPA service line inventory survey (USEPA,
2023e), the CDM Smith Report (CDM Smith, 2022), and Bukhari et al. 2020. The following methods were
identified as the most common methods used by water systems:

•	Visual inspection by customer inside the house

•	Visual inspection by water system personnel inside the house

•	Vacuum excavation

•	Mechanical excavation

Closed circuit television (CCTV) inspection and water quality sampling were not used because of their
potential limitations in definitively identifying non-lead service lines (USEPA, 2022b). Predictive
modeling was not included due to questions regarding its availability to large numbers of systems.

Step 2: Compile cost data from the literature and calculate an average unit cost per investigation
method. The EPA compiled cost estimates for each service line investigation method from the EPA
service line inventory survey (USEPA, 2023e), the CDM Smith Report (CDM Smith, 2022), Bukhari et al.,
2020; and Hensley et al., 2021. The EPA made adjustments to account for non-labor costs of a lead swab
for visual inspection and providing a filter for excavation techniques as required by the final LCRI. The
EPA also adjusted the cost for mechanical excavation to account for the scenario where the water
system finds an LSL or GRR service line during excavation and replaces it at that time. The EPA estimates
that this scenario will occur approximately half of the time based on information provided by the
Pittsburgh Water and Sewer Authority for the EPA service line inventory survey (USEPA, 2023e). When
systems find a lead or GRR service line during mechanical excavation, the EPA assumes that the system
will replace it and the cost for excavation is incorporated into the SLR cost. As shown in Exhibit 4-77, the
cost for individual investigation methods ranges from $40.50 to $777.11.

Final LCRI Economic Analysis

4-185

October 2024


-------
Exhibit 4-77: Average Cost per Service Line Investigated for Four Investigation Methods

Investigation Method

Range of $/SL
Material
Investigated from
the Literature
and Survey

Average $/SL
Material
Investigated from
the Literature and
Survey1

Additional
Non-Labor
Costs

Additional Non-
Labor Costs Notes

Cost
Adjustment

Cost Adjustment
Notes

Average
$/SL
Investigated

Visual inspection by
customer (mail
campaign) inside the
house

$8.44-$115.88

$40.50









$40.50

Visual inspection by
water system
personnel inside the
house

$17.29-$103.79

$56.25

$4.50

Lead swabs 2





$60.75

Vacuum Excavation

$210.80 - $450

$307.60

$64.21

Cost for filter and
door hanger3





$371.81

Mechanical Excavation

$700 - $2,190

$1,490.00

$64.21

Cost for filter and
door hanger3

50%

Reduced to 50% of
average cost to
account for LSLs
being found and
replaced 4

$777.11

Acronyms: SL = service line.

Notes:

1	Based on data presented in the CDM Smith Report (CDM Smith, 2022), Bukhari et al., 2020; Hensley et al., 2021; and the EPA service line inventory survey. For
detailed estimates per system, see the derivation file "Inventory Updates and Validation_Final.xlsx", worksheet "Unit Costs per Field Method."

2	The cost of a lead swab that could be used to test the pipe material was reported in CDM Smith 2022.

3	Systems are required to provide filters under the final LCRI after disturbance due to inventorying, which the EPA assumed would occur during a vacuum or
mechanical excavation. The EPA assumed that systems will provide a door hanger to alert customers of potential temporary elevated lead levels after the
disturbance, as recommended in the EPA inventory guidance (USEPA, 2022b) and required under the LCRI (cost = $0.21/hanger, see derivation file "Public
Education lnputs_CWS_Final.xlsx", worksheet "Service Line Disturbances." The cost to develop the public education materials is accounted for in this
worksheet). The EPA assumed minimal burden (not included) for system personnel to deliver door hangers because they are already on or near the customer's

Final LCRI Economic Analysis

4-186

October 2024


-------
property. The cost of the filter is SafeWater variable costJlter_hh and is estimated to be $64 (see Technologies and Costs for Corrosion Control to Reduce Lead
in Drinking Water (USEPA, 2023b)).

4 Adjustment to mechanical excavation cost to account for the assumption that systems are doing SLR if a lead or GRR service line is found. When lead or GRR
service lines are found, the EPA assumes that systems would replace them and the investigation costs for mechanical excavation would be incorporated into
the SLR cost. The EPA assumed LSLs and GRR service lines are found 50 percent based on information provided by the Pittsburgh Water and Sewer Authority
for the EPA LCRR survey (USEPA, 2023e).

Final LCRI Economic Analysis

4-187

October 2024


-------
Step 3: Create a decision tree to develop a weighted average unit cost ($/SL) for all investigation
methods. The EPA used information in Hensley et al. (2021) and case study examples in CDM Smith
(2022) Tables 3-11, 3-12 to develop a least cost decision tree for systems serving <1,000, 1,001 - 10,000,
and > 10,000 people. Results are shown in Exhibit 4-78 and assumptions are provided below the exhibit.

The EPA recognizes that there are areas of uncertainty in this approach. For example, this approach
does not capture instances where more than one visual inspection method is needed to identify the
material of the service line. On the other hand, the EPA did not include water quality sampling and
predictive modeling, which may be used to screen large numbers of service lines at a lower cost
compared to visual field inspection. The EPA also recognizes that systems may be able to identify all of
their unknowns during normal operation and not do any field investigations.

Exhibit 4-78: Least-Cost Decision Tree and Weighted Average Cost for Field Investigations

System Size
(Population Served)

Field Investigation Method
Decision Tree1

Percent of Service Lines
Identified by Each
Investigation Type
(sum to 100%)

$/SL Line
Investigated Per
Method

Weighted
Average $/SL Line
Investigated

< 1,000

Visual inspection by customer (mail
campaign) inside the house

40

$40.50

$160.10

< 1,000

Visual inspection by water system
personnel inside the house

45

$60.75

< 1,000

Mechanical Excavation

15

$777.11

1,001-10,000

Visual inspection by customer (mail
campaign) inside the house

45

$40.50

$138.82

1,001-10,000

Visual inspection by water system
personnel inside the house

40

$60.75

1,001-10,000

Vacuum Excavation

5

$371.81

1,001-10,000

Mechanical Excavation

10

$777.11

>10,000

Visual inspection by customer (mail
campaign) inside the house

50

$40.50

$117.55

>10,000

Visual inspection by water system
personnel inside the house

35

$60.75

>10,000

Vacuum Excavation

10

$371.81

>10,000

Mechanical Excavation

5

$777.11

Acronyms: SL = service line.

Notes:

1 The decision tree assumes that systems will use the lower cost methods as much as possible then move to more
expensive methods, consistent with the approach presented Hensley et al. (2021), Figure 6. Smaller systems will
use more simple methods compared to larger systems. The EPA assumed that systems serving 1,000 or fewer
people would not use vacuum excavation because it requires special equipment or a contractor, but that CWSs
serving more than 1,000 people would use this method as a cheaper alternative to mechanical excavation.

As the final step for estimating the unit cost for identifying unknown service material, the EPA combined
the results from method (1) normal operations and method (2) field investigations to produce an overall

Final LCRI Economic Analysis

4-188

October 2024


-------
weighted average unit cost to identify unknowns. As noted previously, the EPA estimated that 7.5
percent of unknown service lines can be identified each year during normal operations. Although not
required under the final LCRI, the EPA anticipates that systems will also perform field investigations of
unknowns to expedite SLR and meet the lead and GRR SLR schedule. The EPA estimates that systems will
conduct field investigation of an additional 2.5 percent of unknowns each year, for a total of 10 percent
of unknowns identified each year using both methods.

Exhibit 4-79 shows the weighted average unit costs for identifying unknowns service lines. The EPA
assumed that systems will begin these updates in Year 1 of the analysis period. The EPA used the
estimates of unknown service lines based on the 7th DWINSA as presented in Chapter 3, Section 3.3.4.
The EPA recognizes that this estimate of unknowns is conservatively high because it represents 2021
data, prior to when systems were required to review records for their 2021 LCRR initial inventory.

Exhibit 4-79: Weighted Average Unit Cost ($/SL) for Identifying service line material of
"Unknowns" for the Inventory Updates (CWSs only)

System Size
(Population Served)

Unit Burden for Collecting

Inventory Information
during Normal Operations
(hrs/SL)

Average Unit Cost for
Conducting Field
Investigation of SL Material
($/SL)

Weighted average unit cost

($/SL) for identifying
unknowns for the inventory
update
(cost_update)

A

B

C = (A* Labor
rate*0.75)+(B*0.25)

<100

0.5

$159.73

$52.55

101-500

0.5

$159.73

$52.55

501-1,000

0.5

$159.73

$52.55

1,001-3,300

0.5

$138.32

$47.23

3,301-10,000

0.5

$138.32

$49.63

10,001-50,000

0.5

$116.92

$45.39

50,001-100,000

0.5

$116.92

$46.68

100,001-1,000,000

0.2

$116.92

$35.94

>1,000,000

0.2

$116.92

$35.94

Acronyms: SL = service line.

Notes:

Applies to CWSs with unknown service lines and the number of unknowns in those systems as reported in the 7th
DWINSA. The EPA assumes that systems investigate 10 percent of their unknowns to update their inventory each
year starting in Year 1 of the analysis period.

Sources: "Inventory Updates and Validation_Final.xlsx," worksheet "CWS Inventory Update Est."

A: For systems serving <100,000 people, based on CDM Smith report. For systems serving > 100,000 people, the
EPA assumed CWSs would use an automated process to update the inventory such as the enhanced work order
process developed by the Pittsburg Water and Sewer Authority (PWSA) to capture and upload service line material
data during meter replacements (USEPA, 2023e). The EPA estimated that use of an automated process would
require less burden at approximately 10 minutes per service line (0.2 hours).

B: Exhibit 4-78.

Final LCRI Economic Analysis

4-189

October 2024


-------
C: Assume that 75 percent of unknowns are identified during normal operation and 25 percent are identified
through field investigations of service line material. Labor burden for normal operations is converted to $2020
using the PWS labor rates in Section 3.3.11.1.

d) Report annual inventory updates to the State (hrs_report_inv_op)

The EPA assumed that CWSs and NTNCWSs will incur burden to submit annual inventory updates to
their State. The EPA assumed this burden to be small because systems will be updating their inventory
throughout the year as they investigate unknowns and replace lead and GRR service lines. The EPA
estimated that CWSs and NTNCWSs would spend 1 hour per year submitting the updated inventory to
their State via email (thus no no-labor costs). Note that this burden applies to systems with lead content
and/or unknown service lines, not just systems with unknowns as with the previous section because
systems with lead content and no unknowns would still need to submit inventory updates as they
replace LSLs and GRR service lines. See Chapter 3, Section 3.3.4 for the estimates of systems with lead
content and/or unknown service lines.

4.3.4.1.3 Inventory Validation

The EPA has developed system costs for activities associated with the validation of non-lead service lines
as shown in Exhibit 4-80. The last column provides the corresponding SafeWater LCR model data
variable in red/italic font.

Exhibit 4-80: PWS Inventory Validation Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

e) Conduct field investigations
for inventory validation

$432.56 per non-lead service line
validated

cost_valid

f) Report validation results to
the State

1 hr per CWS or NTNCWS

hrs_valid_report_op

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.
Sources: Data sources for each activity are provided following this exhibit.

All water systems must validate the accuracy of non-lead service line categorization in the inventory.
125Water systems must identify a validation pool of non-lead service lines, excluding service lines
identified as non-lead through (1) records indicating the service line was installed after the effective
date of the Federal, State, or local lead ban, (2) a two-point visual inspection, or (3) previously replaced
lead or GRR service lines.126 Systems must select a random sample of non-lead service lines from the
validation pool that meet the minimum requirements in Exhibit 4-81. Each service line must be validated

125	The EPA is finalizing a flexibility for systems to be able to make a written request to the State to approve a
waiver of the inventory validation requirements if the system has completed validation efforts that are at least as
stringent as the LCRI requirements. This EA may be overestimating validation costs to the degree that States waive
the requirement.

126	In the proposed LCRI, all non-lead service lines identified through records review were excluded from the
validation pool. For the proposed LCRI EA, the validation pool consisted of only the unknowns found to be non-
lead. In addition, the EPA assumed NTNCWSs would have no service lines of unknown materials and thus did not
include them in the proposed LCRI EA.

Final LCRI Economic Analysis

4-190

October 2024


-------
by a minimum of two-point visual inspection. Where ownership is shared, the water system must
conduct at least one visual inspection on each portion of the service line. Where ownership is shared
and only one portion of the service line is included in the validation pool, systems need to conduct at
least one point of visual inspection on the unconfirmed portion of the service line.

Exhibit 4-81: Minimum Non-lead Service Line Validation Requirements of the Final LCRI

Size of Validation Pool

Number of Validations Required
(per system)

Fewer than 1,500

20% of validation pool

1,500-2,000

322

2,001-3,000

341

3,001-4,000

351

4,001-6,000

361

6,001-10,000

371

10,001-50,000

381

>50,000

384

The EPA estimated the number of validations required and the unit cost per validation (in dollars per
service line or $/SL). Key assumptions are below:

•	Burden and cost for validation apply to all systems.

•	Systems will conduct validation in Year 8 of the period of analysis.

•	The validation pool consists of two kinds of non-lead service lines:

o Part 1: Unknown service lines reported in the 7th DWINSA that are found to be non-lead
through inventory updates, except those determined to be non-lead though a two-point
visual inspection.

o Part 2: Non-lead services lines reported in the 7th DWINSA that were installed before the
Federal, State, or local lead ban and are not previously replaced LSL or GRR service lines.

•	All validations will be done using field investigations at two points along the service lines.

The following sections show the unit cost ($/SL) for conducting field investigations for validation (activity
e)) and the burden (hours/CWS) for reporting validation results (activity f)).

e) Conduct field investigations for inventory validation (cost_valid)

To estimate the cost of validation, the EPA used a two-step process: Step 1, determine the size of the
validation pool and the corresponding number of validations required per system and Step 2, estimate
the unit cost ($/SL) per validation.

Final LCRI Economic Analysis

4-191

October 2024


-------
Step 1: Determine the size of the validation pool and the corresponding number of validations required
per system. The EPA estimates the number of non-lead service lines in the validation pool to be the sum
of two parts as described below. The EPA then uses the total non-lead service lines in the validation pool
to determine the minimum number of validations required as shown in Exhibit 4-81. Part 1 of the
validation pool: Unknown Service Lines that are found to be non-lead (applies to CWSs only): As
previously stated, the EPA assumed NTNCWSs will have no unknown service lines. Thus, the EPA used
the following assumptions to estimate this portion of the validation pool for CWSs only:

•	Systems will investigate an average of 10 percent of their unknowns each year as part of
inventory updates starting at Year 1 of the LCRI rule period of analysis. See Section 4.3.4.1.2 for
the rationale for this assumption.

•	Validation will occur in Year 8 of the period of analysis. The percent of all unknowns that are
investigated by the time the system begins validation is 10 percent x 7 years = 70 percent.

•	The proportion of unknowns investigated that are found to be non-lead is one minus the
estimated proportion of unknowns found to be lead as presented in Section 3.3.4, Column E of
Exhibit 3-11.

•	Unknowns that were inspected using vacuum or mechanical excavation would be inspected at
two or more locations and thus, would be excluded from the validation pool.127 Based on the
estimated mix of field inspection methods shown in Exhibit 4-78, the EPA calculated that 15
percent of unknowns were inspected using vacuum or mechanical excavation and would be
excluded from the validation pool.

•	In summary, for each CWS in the SafeWater LCR model, Part 1 of the validation pool is equal to:
o The number of unknown service lines from the 7th DWINSA, multiplied by

o The percent of all unknown service lines that are investigated (70 percent), multiplied by

o The percent of unknowns that are found to be non-lead (1 - perc_unknown_lead, as
presented in Section 3.3.4, Exhibit 3-15, Column E), multiplied by

o The percent of unknowns that were investigated by methods other than vacuum or
mechanical excavation (85 percent).

Part 2 of the validation pool: Non-lead service lines installed before the lead ban (CWSs and
NTNCWSs):

For each CWS, the EPA used the following assumptions to estimate this part of the validation pool:

•	Thirty-three percent of service lines were installed as new construction after the Federal lead
ban (assumed to be 1990) based on a review of housing stock data. (For additional details and

127 For the proposed LCRI, the EPA estimated that all unknowns found to be non-lead would be identified as non-
lead by visual identification at one point during normal operation or field inspection. Under the final LCRI, the EPA
is assuming that unknowns that were inspected using vacuum or mechanical excavation would be inspected at two
or more locations and thus, would be excluded from the validation pool. The EPA conservatively assumes that
remaining unknowns found to be lead would still need two points of visual inspection for validation.

Final LCRI Economic Analysis

4-192

October 2024


-------
references for this assumption, see Section 4.3.4.1.1, Exhibit 4-74, Column C.) These service lines
are excluded from the validation pool based on the rule requirements. The remaining 67
percent of non-lead service lines are assumed to be installed before the lead ban and are
included in the validation pool.

•	CWSs have been replacing LSLs at a rate of 1 percent per year as part of normal infrastructure
programs between 1991 and 2020, and approximately 28 percent of those would be full
replacements. (For additional details and references for this assumption, see Section 3.3.4.1.2,
Step 7). Thus, the EPA assumed that of the remaining non-lead service lines that could be in the
validation pool, approximately 8 percent (1%/year x 29 years x 28% = 8%) would be replaced
LSLs and excluded from the validation pool. The remaining 92 percent are assumed to be
something other than replaced LSLs and would be included in the validation pool.

•	In summary, for each CWS in the SafeWater LCR model, Part 2 of the validation pool is equal to:
o The number of non-lead service lines per CWS from the 7th DWINSA, multiplied by

o The percent of service lines installed before the Federal Lead Ban (67 percent), multiplied by

o The percent of service lines that are not previously replaced LSLs or GRR (92 percent).

For each CWS, the EPA added the results of Parts 1 and 2 to determine the total size of the validation
pool. Based on this total, the EPA determined the required number of non-lead service lines that must
be validated using the final LCRI requirements shown in Exhibit 4-81.

For each NTNCWS the EPA used the following assumptions to estimate this part of the validation pool:

•	An estimated 97.5 percent of NTNCWSs have all non-lead service lines as presented in Chapter
3, Section 3.3.4.2.1.

•	For the estimated 2.5 percent of NTNCWSs that have lead content service lines, assume that all
service lines in are lead content. This is consistent with the estimates presented in Chapter 3,
Exhibit 3-23 with one exception: For NTNCWSs serving 10,000 to 50,000 people, the EPA
estimates that between 50 and 100 percent of service lines in those systems are lead content.
The EPA made a simplifying assumption that this value is 100 percent for the validation analysis
to simplify the SafeWater LCR modeling process and recognizing that uncertainty in this
assumption would have a very small/de minimums impact on total validation costs.

•	Use the first reported date in SDWIS/Fed to identify the proportion of NTNCWSs installed after
the lead ban. The EPA recognizes uncertainty in this assumption in that the first reported date
could be before or after service line installation; however, the first reported date is when the
facility was considered a public water system, so it is logical to assume that this date closely
approximates when the utility installed its water service lines. Using SDWIS/Fed 4th Quarter
2020 data, an estimated 9,326 NTNCWSs have a first reported date after 1990, which is 54
percent of the total number of NTNCWSs (9,326/17,418 total NTNCWSs = 54%). Thus, the
estimated percentage of NTNCWSs installed before the lead ban and potentially in the
validation pool is 1 - 0.54, or 46 percent.

•	In summary, for each NTNCWS in the SafeWater LCR Model, the validation pool is equal to:

Final LCRI Economic Analysis

4-193

October 2024


-------
o The percent of NTNCWSs with non-lead service lines (97.5 percent), multiplied by:

o The percent of NTNCWSs installed before the lead ban (46 percent), multiplied by:

o The number of service lines per NTNCWS from SDWIS/Fed 4th quarter 2020.

The EPA conservatively assumed that none of the non-lead service lines in NTNCWSs are previously
replaced LSLs. For each NTNCWS, the EPA used the results of Part 2 to determine the required number
of non-lead service lines that must be validated using the final LCRI requirements shown in Exhibit 4-81.

Step 2: Estimate the unit cost per validation.

Validation must be done at a minimum of two points along the service line and must include the system-
owned and customer-owned portions where ownership is split. Where ownership is shared and only one
portion of the service line is included in the validation pool, systems need to conduct at least one point
of visual inspection on the unconfirmed portion of the service line. The EPA made a conservative
assumption that all validations would require two points of visual inspection to meet final LCRI
requirements. This approach overestimates situations where validation is needed, for example, just on
the customer-owned portion of the service line at one point.

The EPA estimated that the field investigation methods used for validation would be the combination of
(1) vacuum excavation at one location along the service line, and (2) visual inspection by water system
personnel at the second location inside the house. The EPA did not consider mechanical excavation due
to its high cost compared to vacuum excavation and the fact that water systems have significant time to
find vacuum inspection contractors or purchase equipment prior to the validation due date (i.e., Year 10
of the analysis period). This assumption could result in an underestimate of unit costs when mechanical
excavation is used. However, this approach would overestimate unit costs when vacuum excavation can
be used for both points of a visual inspection, or when systems can use in-home inspection and a second
inspection that does not involve excavation such as visual inspection in the meter pit or curb box.

The derivation of the unit cost for validations is shown in Exhibit 4-82. Note that the EPA assumed the
same average unit cost for validation at CWSs and NTNCWSs regardless of system size.

Exhibit 4-82: Unit Cost ($/SL) for Validation

Unit Cost for Vacuum
Excavation

Unit Cost for Visual Inspection
by Water System Personnel
Inside the House

Total Unit Cost per Validation
(cost_valid)

A

B

C=A+B

$371.81

$60.75

$432.56

Sources: Derivation file "Inventory Updates and Validation_final.xlsx," worksheet "Unit Cost for Valid."
Notes: Applies to all CWSs and NTNCWSs regardless of system size.

A: Exhibit 4-77, row 3, includes cost for filter and door hanger.

B: Exhibit 4-77, row 2, include cost for lead swab that could be used to test the pipe material.

Final LCRI Economic Analysis

4-194

October 2024


-------
f) Report validation results to the State (hrs_report_valid_op)

The EPA assumes that CWSs and NTNCWSs will summarize the results of validation (e.g., how many
service lines were confirmed non-lead or if any were found to be lead or GRR service lines) in email
communications and that this will take approximately 1 hour.

Exhibit 4-83 provides the SafeWater LCR model costing approach for these inventory-related activities
including additional cost inputs that are required to calculate the total costs.

Exhibit 4-83: Lead Service Line Inventory Cost Estimation in SafeWater LCR by Activity1





Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions2

Frequency
of Activity

a) Conduct records review of connector materials

The total hours per system multiplied by
the system labor rate.

hrs_updated_initial_inv_op*rate_op

Cost applies
as written to
NTNCWSs.

All

All Model PWS

Once a
year for first
three years

b) Compile and submit connector updated LCRR initial inventory (baseline inventory) to the
State

The total hours per system multiplied by
the system labor rate.

hrs_report_updated_initial_inv_op*rate_op

Cost applies
as written to
NTNCWSs.

All

All Model PWS

Once a
year for first
three years

c) Identify material for unknown service lines

The cost per service line multiplied by the
number unknown lines identified.

num_unknown_resolved*cost_update

Cost does not
apply

NTNCWSs.

All

All Model PWS with
service lines of
unknown content

Once a
year for first
10 years

d) Report annual inventory updates to the State

The total hours per system multiplied by
the system labor rate.

hrs_report_inv_op*rate_op

Cost applies
as written to
NTNCWSs.

All

All Model PWS with
service lines of lead or
unknown content

Once a
year for first
10 years

e) Conduct field investigations for inventory validation

The cost per service line multiplied by the
number of lines validated.

num_lsl_ validated*cost_ valid

Cost applies
as written to
NTNCWSs.

All

All Model PWS

Once

f) Report validation results to the State

The total hours per system multiplied by
the system labor rate.

hrs_valid_report_op*rate_op

Cost applies
as written to
NTNCWSs.

All

All Model PWS

Once

Final LCR! Economic Analysis

4-195

October 2024


-------
Acronyms: CWS = community water system; LSL = lead service line; NTNCWS = non-transient non-community

water system; PWS = public water system.

Notes:

1	The data variables in the exhibit are defined previously in this section with the exception of:

• rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

2	PWSs with lead content or unknown lines are identified using the data variables and approach described in
Chapter 3, Section 3.3.4.

4.3.4.2 Service Line Replacement Plan

This section summarizes the EPA's cost estimate for the SLR plan that must be completed by all systems
with lead, GRR, or service lines of unknown material at the start of the rule and updated annually
thereafter. It also provides activities for periodic review and consultation with the State on the subset of
systems that are seeking or have been approved for a deferred SLR. Exhibit 4-84 provides the unit
burden and/or cost for these activities. The assumptions used in the estimation of the unit burden and
cost follow the exhibit. The last column provides the corresponding SafeWater LCR model data variable
in red/italic font.

Exhibit 4-84: PWS SLR Plan Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

g) Develop initial SLR plan and submit
to the State for review (one-time)

12 to 36 hrs/CWS;
12 hrs/NTNCWS

hrs_slr_plan_op

h) Identify funding options for full
SLRs

68 to 170 hrs/CWS

hrs_fin_op_op

i) Include information on deferred
deadline and associated
replacement rate in the SLR plan
(one-time)1

3 to 9 hrs/CWS seeking a deferral;
3 hrs/NTNCWS seeking a deferral

hrs_slr_plan_defer_op

j) Update SLR plan annually or certify
no changes

2 to 4 hrs/CWS;
2 hrs/NTNCWS

hrs_slr_plan_ update_ op

k) Provide an updated

recommendation of the deferred
deadline and associated
replacement rate1

3 to 9 hrs/CWS on a deferred SLR rate;
3 hrs/NTNCWS on a deferred SLR rate

hrs_defer_update_op

Acronyms: CWS = community water system; SLR = service line replacement; NTNCWS = non-transient non-
community water system.

Sources: Data sources for each activity are provided following this exhibit.

Notes:

1 Only applies to those systems eligible for and requesting a deferred SLR deadline.

Final LCR! Economic Analysis

4-196

October 2024


-------
g) Develop initial SLR Plan and submit to State for review (hrs_slr_plan_op). All systems with lead,

GRR, and/or unknown service lines128 must develop a plan for their SLR program that includes the

following elements:

•	A strategy for determining the composition of lead status unknown service lines in its inventory.

•	A strategy for informing consumers and customers before a full or partial SLR.

•	Procedures for coordinating the full SLR.

•	A funding strategy for conducting SLR that includes ways to accommodate customers that are
unable to pay to replace the portion of the service line they own.

•	A strategy to prioritize LSR based on factors including, but not limited to, known lead and GRR
service lines and community specific factors.

•	A procedure for consumers and customers to flush service lines and premise plumbing of
particulate lead following a disturbance or post-replacement.

•	A communication strategy to inform both consumers and owners of rental properties with LSLs,
GRR service lines, and service line of unknown material about the replacement program.

•	Identification of any State and local laws and water tariff agreements relevant to the water
system's ability to gain access to conduct full replacement.

•	For systems that identify lead-lined galvanized service lines in their inventory, a strategy to
determine the extent of the use of lead-lined galvanized service lines in the distribution system.

Also see activity i) for additional requirements for systems eligible for and requesting a deferred
deadline.

The estimated burden is provided in Exhibit 4-85. The EPA assumed systems would require twice the
burden to prepare the plan as for the State to review it. The State burden (hrs_slr_plan_js) is based
on the ASDWA CoSTS model that assumes 6 hours for States to review the plan for small CWSs and
NTNCWSs, 10 for medium CWSs, and 18 for large CWSs (ASDWA, 2020b; 2024).129 ASDWA's
estimates remained the same in their 2024 CoSTS model. See data variable hrs_slr_plan_js in Section
4.4.4.2, activity d) for assumptions used to derive that input. Note that the EPA developed a separate
estimate for identifying funding options for SLR as described in the next activity.

128	Section 3.3.4.1.1 in Chapter 3 presents the estimated percent of systems with known or potential lead content
service lines. Note that the EPA grouped all systems with lead content together, so the values in Section 3.3.4.1.1
likely overestimate the percent and number of CWSs with known or potential lead and GRR service lines because
they include lead connectors and galvanized pipe previously downstream of lead connectors.

129	The EPA assumed large, medium, and small systems in ASDWA's CoSTS model corresponded to those size
categories defined in the pre-2021 LCR as systems serving 50,000 or more people, 3,301 to 50,000 people, and
3,300 or fewer people, respectively.

Final LCRI Economic Analysis

4-197

October 2024


-------
Exhibit 4-85: Estimated Burden for Systems with Lead, GRR, and/or Unknown Service Lines to

Develop an SLR Plan

System Size
(Population Served)

hrs_slr_plan_op

CWSs

NTNCWSs

<3,300

12

12

3,301-10,000

20

12

10,001-50,000

20

12

>50,000

36

12

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.
Source: "LSLR Ancillary Costs_Final.xlsx."

h) Identify funding options for full SLRs (hrs_fin_op_op). The EPA assumes that CWSs with lead,
GRR, and/or unknown service lines will incur additional burden to identify and evaluate funding
options for SLR due to the complexities of financing SLR on private property. The burden for
financial planning and identifying funding options for SLR was estimated at 400 to 1,100 hours
per system in the proposed LCRI EA.

The proposed rule estimate was based on the estimate used for financial planning and
identifying funding options from the Final 2021 LCRR EA (USEPA, 2020). Since the 2021 LCRR
rule analysis was finalized, the EPA has provided additional technical assistance and guidance on
funding sources for SLR. Specifically, In January 2023, the EPA announced the "Lead Service Line
Replacement Accelerators" initiative (USEPA, 2023f). This major initiative is providing targeted
technical assistance services to help underserved communities access funds from the BIL for the
replacement of lead pipes that pose risks to the health of children and families. In December
2023, the EPA launched a new website titled "Identifying Funding Sources for Lead Service Line
Replacement", available online at https://www.epa.gov/ground-water-and-drinking-
water/identifving-funding-sources-lead-service-line-replacement. The EPA has also published
SLR financing case studies, available online at https://www.epa.gov/ground-water-and-drinking-
water/lslr-financing-case-studies. which were last updated in February 2024.

For the final LCRI EA, the EPA re-evaluated the previously identified sub-activities which make
up the total burden estimate for identifying funding options in light of these new funding
resources and technical support. This effort resulted in the EPA's holding one sub-activity
constant, reducing two activities and eliminating the remainder of the sub-activities. In
particular, the EPA maintained the burden from the proposed LCRI EA for evaluating potential
legal considerations regarding the use of funding on private property. The agency reduced the
burden for identifying and evaluating funding sources by half given the now available resources
provide by the EPA. The agency removed all other financing steps including meeting with
potential funding source, compiling a preliminary financing sub-activities, conducting public
meetings and outreach, conducting a consumer income survey, and submitting pre-application
for funding sources. These sub-activities were removed to avoid double counting the burden in
activity g) for initial SLR plan development, and because they were beyond the scope of the
initial SLR plan development. These updates resulted in a revised estimate of 68 to 170 hours as

Final LCRI Economic Analysis

4-198

October 2024


-------
a one time burden for CWS with lead, GRR, and/or unknown service lines to identify SLR funding
options under the final LCRI and the 2021 LCRR (baseline). See Exhibit 4-86 for the breakdown of
hours per system size category and activity.

Exhibit 4-86: PWS Burden to Identify Funding Options for SLRs

Planning Activity

Estimated Burden

CWSs serving:

<10,000

10,001-
100,000

>100,000

Legal considerations for funding options.

•	Determine if statutes/regulations prohibit/restrict a public system
from paying for SLRs on private property (i.e., using public funds for
private purposes).

•	Determine statutes/regulations prohibit/ restrict type of
funding used for SLRs and if so, do they apply to system type
(public vs. private) and SLR type (on public or private property).

8

16

20

Identify potential funding sources.

•	Consider grants, loans, or bonds or a combination; also consider
other govt, support for low income homeowner-owned segments
(e.g., HUD). Assume small systems have assistance identifying
options.

•	Include State-specific options (such as MA's interest-free LSLR
program through their State revolving fund (SRF)).

20

30

50

Evaluate funding sources.

Determine if project meets criteria, funding and project timeline are
compatible, impact on user charges, additional engineering or special
studies required, affordability for users.

40

50

100

Totals

68

96

170

Acronyms: CWS = community water system; LSLR = lead service line replacement; SLR = service line replacement.
Source: "LSLR Ancillary Costs_Final.xlsx."

i) Include information on deferred deadline and associated replacement rate in the SLR plan
(hrs_slr_plan_defer_op). Systems eligible for and requesting a deferred SLR deadline must also
include in their initial SLR plan the following: (1) documentation of the system's eligibility for a
deferred deadline; (2) documentation detailing the system's request for completing mandatory SLR
under a deferred deadline, including the annual number of replacements required, the length of
time (in years and months), the date of completion, and the associated cumulative average
replacement rate considered to be the fastest feasible but no slower than the replacement rate
corresponding to 39 annual replacements per 1,000 service connections; and (3) information
supporting the system's request for a deferred deadline and why replacing lead and GRR service
lines at a faster rate is not feasible. The EPA assumed that this burden would be 25 percent of the
burden to develop the initial SLR plan, which is based on the ASDWA 2020 and 2024 CoSTS models
(ASDWA 2020b; 2024). See Exhibit 4-87 below.

Final LCRI Economic Analysis

4-199

October 2024


-------
Exhibit 4-87: Estimated Additional Burden for the Initial SLR Plan Development for Systems

Requesting a Deferred SLR Rate

System Size
(Population Served)

hrs_slr_plan_defer_op

CWSs

NTNCWSs

3,300

3

3

3,301-10,000

5

3

10,001-50,000

5

3

>50,000

9

3

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.
Source: "LSLR Ancillary Costs_Final.xlsx."

Notes: This additional burden only applies to systems requesting a deferred SLR rate.

j) Update SLR plan annually or certify no changes (hrs_slr_plan_update_op). All systems with lead,
GRR, and/or unknown service lines must either: 1) update their SLR plan annually to include any
changes that affect the system's ability to conduct mandatory full SLR, such as updates to relevant
regulations (e.g., State or local government laws associated with utility access), a new strategy for
identifying materials of unknown service lines based on inventory validation, or lessons learned
from risk communication efforts in the community; or 2) submit a certification of no change. The
EPA assumed the majority of systems over time will not need to update their SLR program but
instead will provide a certification of no change. Water systems may cease annual certifications to
the State when there are no lead, GRR, and unknown service lines left in the inventory. The EPA
assumed a lower burden for this requirement than needed to develop the initial SLR plan, which is in
line other certification burden estimates using in this LCRI analysis, as shown in Exhibit 4-88 below.

Exhibit 4-88: Estimated Annual Burden for Systems to Update the SLR Plan or Certify No

Changes

System Size
(Population Served)

hrs_slr_plan_update_op

CWSs

NTNCWSs

<3,300

2

2

3,301-10,000

3

2

10,001-50,000

4

2

>50,000

4

2

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.

Source: "LSLR Ancillary Costs_Final.xlsx."

Notes: Systems with lead, GRR, or unknowns must annually update their SLR plan if they have a significant
change or must instead certify to the State that they have no changes.

k) Provide an updated recommendation of the deferred deadline and associated replacement rate
(hrs_defer_update_op). Systems with deferred deadlines, in addition to annual updates, must every

Final LCRI Economic Analysis

4-200

October 2024


-------
three years after the initial submission of the plan, update their replacement plan with the latest (1)
documentation of the system's eligibility for a deferred deadline that shows that replacing 10
percent of the total number of known lead and GRR service lines (based on the replacement pool)
results in the annual number of replacements per 1,000 service connections to exceed 39; (2)
documentation detailing the system's request for completing mandatory SLR under a deferred
deadline; and (3) information supporting the system's determination that the mandatory SLR rate is
not feasible to meet and why replacing lead and GRR service lines at a faster rate is not feasible. .
The EPA assumed that this burden would be 25 percent of the burden to develop the initial SLR plan,
consistent with the burden for activity i) and shown in Exhibit 4-87.

Exhibit 4-89 provides the SafeWater LCR model costing approach for this activity including additional

cost inputs that are required to calculate the total costs.

Exhibit 4-89: LSLR Plan Cost Estimation in SafeWater LCR by Activity1





Conditions for Cost to Apply
to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th -
Range

Other Conditions2

Frequency of
Activity

g) Develop initial SLR plan and submit to State for review

The total hours per system
multiplied by the system labor rate.

(hrs_slr_plan_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

Model PWSs with service
lines of lead, GRR,
and/or unknown service
lines

One time

h) Identify funding options for full SLRs

The total hours per system
multiplied by the system labor rate.

(hrs_fin_op_op*rate_op)

Cost does not
apply to
NTNCWSs.

All

Model PWSs with service
lines of lead, GRR,
and/or unknown service
lines

One time

i) Include information on deferred deadline and associated replacement rate in the SLR plan

The total hours per system
multiplied by the system labor rate.

(hrs_slr_plan_defer_op *rate_op)

Cost applies as
written to
NTNCWSs.

All

Model PWSs seeking a
deferral

One time

j) Update SLR plan annually or certify no changes

The total hours per system
multiplied by the system labor rate.

(hrs_slr_plan_update_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

Model PWSs with service
lines of lead, GRR,
and/or unknown service
lines

Annually
Starting in
Year 4

k) Provide an updated recommendation of the deferred deadline and associated replacement
rate

The total hours per system
multiplied by the system labor rate.

(hrs_defer_update_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

Model PWSs on a
deferred SLR rate

Year 6 and

triennially

thereafter

Acronyms: CWS = community water system; GRR = galvanized requiring replacement; LSL = lead service line; SLR =
service line replacement; NTNCWS = non-transient non-community water system; PWS = public water system.

Final LCR! Economic Analysis

4-201

October 2024


-------
Notes:

1	The data variables in the exhibit are defined previously in this section with the exception of:

• rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

2	PWSs with lead content or unknown lines are identified using the data variables and approach described in
Chapter 3, Section 3.3.4.

4.3.4.3 Physical Service Line Replacements

The final LCRI requires water systems to fully replace all lead and GRR service lines within 10 years
unless the State has set a shorter schedule or approved a deferred deadline130. As discussed in Chapter
3, Section 3.3.4.3, several States already require PWSs to replace service lines with lead content. Since
these requirements already exist, these State-required replacements131 are not included in the cost or
benefits of the final LCRI. For each PWS in a State with an existing SLR requirement, SafeWater LCR first
calculates the number of SLs that would need to be replaced each year under the final LCRI absent any
State requirement. These are known as the PWS's Federal SLRs. Next, SafeWater LCR calculates the
number of SLs that would need to be replaced each year under the State requirements, absent any
federal requirement. These are known as the PWS's State SLRs. SafeWater LCR then determines the
PWS's Total SLR as the maximum of the Federal or State Replacements. Finally, SafeWater LCR calculates
the PWS's SLRs due to the final LCRI as the difference between the PWS's Total SL replacements and the
PWS's State SL replacements. Only the SL replacements due to the final LCRI are included in the cost and
benefit estimates of the final rule. However, the PWS's Total SL replacements are tracked as they count
towards the PWS's SL replacement requirement and total lines replaced in the system (i.e., some
systems under more strict SL removal requirements may finish before the final LCRI 10 year deadline).

This section summarizes the EPA's cost estimates for physical replacement of service lines. Exhibit 4-90
summarizes cost estimate ranges for the physical full replacement and partial replacement of LSLs and
the physical replacement of GRR service lines.

Exhibit 4-90: PWS LSLR Cost Estimates

Activity

Cost Estimate Range (2020$)

SafeWater LCR Data Variable

1) Systems replace lead or GRR service
lines

Full: $6,507 - $8,519;
Partial: $1,920 - $5,400;
GRR: $1,920 - $5,400

cost_lslr_lsl_reg_mand_pws;
cost_ lslr_parti al_ reg_p ws;
cost Islr gal prev Isl reg pws

Acronyms: GRR = galvanized requiring replacement.
Source: "LSLR Unit Cost Final.xlsx"

I) Systems replace service lines (cost_lslr_reg_mand_pws; cost_lslr_partial_reg_pws:

cost_lslr_gal_prev_lsl_reg_pws). The EPA has developed estimates for a low and a high cost

130	The 2021 LCRR and final LCRI both require water systems to replace lead connectors when encountered during
normal operation. The EPA assumed that the incremental cost to meet this requirement is minimal because it is
standard practice for water systems to use new connectors when replacing water mains or service lines.

131	The States of Illinois, Michigan, New Jersey, and Rhode Island have passed State laws and regulations requiring
mandatory service line replacement independent of their tap monitoring results.

Final LCRI Economic Analysis

4-202

October 2024


-------
scenario based on reported project data in the 7th DWINSA for full and partial replacements. These
estimates are based on the 25th and 75th percentile data from 33 DWINSA reported projects. Note
the estimated full and partial replacement costs include the cost of replacing the lead connectors.
The detailed methodology for estimating the SLR unit costs is provided in Appendix A, Section A.2.
See Section 4.2.2.2 for a discussion of how EPA modeled uncertainty in service line replacement unit
costs using the 25th and 75th percentile.

Exhibit 4-91 provides the SafeWater LCR model costing approach for this activity including additional
cost inputs that are required to calculate the total costs.

Exhibit 4-91: Lead Service Line Replacement Cost Estimation in SafeWater LCR by Activity1





Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS
Cost Per
Activity

Lead 90th -
Range

Other Conditions2

Frequency
of Activity

1) Systems replace lead and GRR service lines

The sum of the number of lines replaced
for each category of possible types of
replacement multiplied by the costs per
type of replacement.

(num_lslr_lsl_replace*cost_lslr_lsl_reg_ma
nd_pws)+(num_lslr_partial_replace*cost_ls
lr_partial_reg_pws)+(num_lslr_gal_prev_lsl
_replace*cost_lslr_gal_prev_lsl_reg_pws)

Cost applies
as written to
NTNCWS.

All

Model PWS with
known or potential
lead content

Once a year3

Acronyms: AL = action level; CWS = community water system; LSL = lead service line; SLR = service line
replacement; NTNCWS = non-transient non-community water system; PWS = public water system.

Notes:

1	The data variables in the exhibit are defined previously in this section.

2	PWSs with lead content or unknown lines are identified using the data variables and approach described in
Chapter 3, Section 3.3.4.1.

3	Replacement of lines occurs on an annual time step. Most lines are replaced in the period defined by the
proposed LCRI, but some additional replacement occurs in the periods past the LCRI deadline based on systems
meeting the deferred replacement requirements of the LCRI.

4.3.4.4 Ancillary Service Line Replacement Activities

The EPA has developed system costs for ancillary activities associated with SLR, as shown in Exhibit 4-92.
The exhibit provides the unit burden and/or cost for each activity. The assumptions used in the
estimation of each activity follows the exhibit. The last column provides the corresponding SafeWater
LCR model data variable in red/italic font. In a few instances, some of these activities are conducted by
the State instead of the water system. These activities are identified in the exhibit and further explained
in the exhibit notes.

Final LCRI Economic Analysis

4-203

October 2024


-------
Exhibit 4-92: PWS SL Replacement Ancillary Unit Burden and Cost Estimates

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

m) Contact customers and conduct
site visits prior to SLR

Burden per replaced service line
1.70 to 2.07 hrs

Cost per replaced service line
$11.64 to $16.13/replaced LSL

Burden

hrs_replaced_lsl_contact_op
Cost

cost_replaced_lsl_contact

n) Deliver filters and 6 months of
replacement cartridges at time
of SLR

$64.00/replaced service line

cost_filter_hh

o) Collect tap sample post-SLR

Burden per sample
CWSs: 0.9 to 1.2 hrs
NTNCWSs: 0.5 hrs

Cost per sample per CWS
Travel: $5.75 to $10.24
Bottle: $0 to $2.85

Burden

hrs_collect_lsl_lslr_op
Cost

cost_pickup_samp
cos^otherj^samp1

p) Analyze post-SLR tap sample

In-house Analysis (CWSs > 100K onlv)

In-house Analvs/s

Burden: 0.44 hrs/sample
Cost: $3.92

Commercial Analyses
$32.20/sample

hrs_ an alyze_ Isljsl^op1
costjabjsljslr1

Commercial Analysis
cost commercial Isl Islr1

q) Inform customers of tap sample
result

Burden

CWSs: 0.05 -0.11 hrs/sample
NTNCWSs: 1 hr/system

Cost

CWSs: $0.72/sample
NTNCWSs: $0.079/system

Burden

hrs_inform_samp_op
hrs_ n tn cws_ cust_lslr_op

Cost

cost_cust_lslr
cost ntncws cust Islr

r) Submit annual report on SLR
program to State

1 to 8 hrs/CWS
1 hr/NTNCWS

hrs_ report_ lcr_ op

Acronyms: CWS = community water system; HH = household; SL = service lines; SLR = service line replacement;

NTNCWS = non-transient non-community water system.

Sources:

m) & r):"LSLR Ancillary Costs_Final.xlsx.

n): Technologies and Costs for Corrosion Control to Reduce Lead in Drinking Water (USEPA, 2023b)

o) - q) "Lead Analytical Burden and Costs_Final.xlsx."

Note:

1The burden and costs for these activities are incurred by the State in Arkansas, Louisiana, Mississippi, Missouri,

and South Carolina (ASDWA, 2020a).

m) Contact customers and conduct site visits prior to SLR (hrs_repiaced_lsl_contact_op,

cost_repiaced_lsi_contact). CWSs will incur burden and costs to coordinate with customers prior to
replacing the SLs. The estimated burden and costs are provided in Exhibit 4-93 and Exhibit 4-94,
respectively.

Final LCRI Economic Analysis

4-204

October 2024


-------
Exhibit 4-93: Estimated Burden Associated with Contacting Customers and Site Visit Prior to
LSLR (hours/replaced SL) (hrs_replaced_lsl_contact_op)



Upfront Contact

Site Visit Travel





System Size
(Population Served)

Phone Call

Prepare
Letter

Miles
one way

Time one
way (hrs)

Time
Roundtrip
(hrs)

On-Site
Review

Total
Burden



A

B

C

D

E = D*2

F

G =
A+B+E+F

<3,300

0.25

0.05

5.0

0.20

0.40

1

1.70

3,301-100,000

0.25

0.11

5.0

0.20

0.40

1

1.76

100,001-1,000,000

0.25

0.11

6.4

0.26

0.51

1

1.87

>1,000,000

0.25

0.11

8.9

0.36

0.71

1

2.07

Source: "LSLR Ancillary Costs_Final.xlsx," worksheet "Customer Coordination."

Notes:

A & B: For each SLR, the EPA assumed a system will first contact customers twice. These contacts are to coordinate
a site visit to confirm the presence of a SL requiring replacement prior to the actual replacement of the line that
are found to be lead. The EPA assumed the system first calls the customer (15 minutes per customer) and then
sends a certified letter. Burden to prepare the letter is 1 hour 20 letters for systems serving 3,300 or fewer people
and 1 hour per 9 letters for those serving more than 3,300 people, based on the 2022 Disinfectants and
Disinfection Byproducts, Chemical, and Radionuclides Rules ICR (Renewal), Exhibit 29 - Notification of Sampling
Results for Customers Whose Taps Are Sampled (Note G) (USEPA, 2022a).

C - E: Based on census data and zip codes from the 2006 Community Water System Survey. See file "Estimated
Driving Distances_Final.xlsx." EPA assumed an average speed of 25 miles per hour, round trip.

F: Includes 1 hour for on-site visual inspection. Assumed no testing.

Exhibit 4-94: Estimated Non-Labor Costs Associated with Contacting Customers and Site Visit
Prior to SLR ($/replaced SL) (cost_replaced_lsl_contact)

System Size
(Population Served)

Mailing Costs

Vehicle O&M

Total Cost

Certified
Mail

Paper

Envelopes

Miles
Roundtrip

2016
Mileage
Rate

Cost per
Trip

A

B

C

D

E

F = D * E

G =
A+B+C+F

<100,000

$5.80

$0,017

$0,076

10

$0,575

$5.75

$11.64

100,001-1,000,000

$5.80

$0,017

$0,076

12.8

$0,575

$7.36

$13.25

> 1,000,000

$5.80

$0,017

$0,076

17.8

$0,575

$10.24

$16.13

Source: "LSLR Ancillary Costs_Final.xlsx," worksheet "Customer Coordination."

Notes:

A: Includes certified mail cost ($3.55), emailed signature receipt ($1.70), and first class postage ($0.55). See
https://pe.usps.eom/Archive/NHTML/DMMArchive20201018/Noticel23.htm#_c037 (accessed 1/5/22).
B&C: Based on quotes from 3 vendors. See file, "General Cost Model lnputs_Final.xlsx" for additional detail.

Final LCRI Economic Analysis

4-205

October 2024


-------
D: Vehicle O&M based on 25 mph and Federal reimbursement rate of $0,575 (2020 mileage rate. See
https://www.gsa.gov/travel/plan-book/transportation-airfare-pov-etc/privately-owned-vehicle-mileage-rates/pov-
mileage-rates-archived#auto. Accessed 1/17/2022.

n) Deliver filters and 6 months of replacement cartridges at time of SLR (cost_filter_hh). Systems
must provide a pitcher filter (i.e., pour through filter) or point-of-use device that is certified to
remove lead to each resident following any lead or GRR SLR. The EPA assumed that the pitchers and
filters delivered to each resident to use for six months following a replacement will cost $64 on
average (including shipping and filter replacement). See Technologies and Costs for Corrosion
Control to Reduce Lead in Drinking Water (USEPA, 2023b) for additional detail.

o) Collect tap sample post-SLR (hrs_collect_lsl_lslr_op, cost_pickup_samp, cost_other_lt_samp). All

systems must collect one sample following replacement of each lead or GRR service line
(numb_samp_lslr). Burden and costs for this activity are different than mandatory tap sampling
because the system collects the sample after replacement as opposed to the tap sampling program
in which the customer collects the sample. Exhibit 4-95 and Exhibit 4-96 provide the estimated CWS
burden and cost to collect these samples. A discussion of the burden and costs to NTNCWSs follow
these exhibits.

Exhibit 4-95: CWS Unit Burden to Collect Post-SLR Tap Sample

System Size
(Population Served)

Burden (hrs/Sample)

Round-trip travel to
customer's home

Sample Collection
Burden

Total Sample Collection
Burden

hrs_collect_lsl_lslr_op

A

B

C = A+B

<100,000

0.40

0.5

0.9

100,001-1,000,000

0.51

0.5

1.0

>1,000,000

0.71

0.5

1.2

Source: "Lead Analytical Burden and Costs_Final.xlsx," worksheet, "LSLR_CollectAnaly_CWS_LCRR_LCRI."

Notes:

A: Based on census data and zip codes from the 2006 Community Water System Survey, the EPA assumed the
following one-way driving distances for CWSs: 5.0 miles serving < 100,000 people, 6.4 miles serving 100,001 - 1M,
and 8.9 miles for > 1M. These distances were doubled to estimate roundtrip mileage. See file, "Estimated Driving
Distance_Final.xlsx" for additional detail on how these estimates were derived. The EPA assumed an average
speed of 25 miles per hour.

B: The EPA assumed the same collection burden following LSLR as for source water sample collection, which is
based on the 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR (Renewal), Exhibit
15, Average Labor Hrs. for Collection (Per Sample) (USEPA, 2022a).

Exhibit 4-96: CWS Non-labor Unit Cost to Collect Post-SLR Tap Sample



Cost (hrs/Sample)

System Size
(Population Served)

Round-trip travel to customer's
home

Bottle Cost



cost_pickup_samp

cost_other_lt_samp



A

B

<100,000

$5.75

$0.00

Final LCRI Economic Analysis

4-206

October 2024


-------
System Size
(Population Served)

Cost (hrs/Sample)

Round-trip travel to customer's
home

Bottle Cost

cost_pickup_samp

cost_other_lt_samp

100,001-1,000,000

$7.36

$2.85

>1,000,000

$10.24

$2.85

Source: "Lead Analytical Burden and Costs_Final.xlsx," worksheet, "LSLR_CollectAnaly_CWS_LCRR_LCRI."

Notes:

A: Based on census data and zip codes from the 2006 Community Water System Survey, assumed the following
one-way driving distances for CWSs: 5.0 miles serving < 100,000 people, 6.4 miles serving 100,001 - 1M, and 8.9
miles for > 1M. These distances were doubled to estimate roundtrip mileage. See file, "Estimated Driving
Distance_Final.xlsx" for additional detail on how these estimates were derived. The EPA assumed an average
speed of 25 miles per hour and used the Federal reimbursement rate of $0,575 (2020 mileage rate).

B: Bottles are included as part of the commercial laboratory fee. Only CWSs serving more than 100,000 people are
assumed to conduct analyses in-house for lead. For a detailed discussion of the assumptions used to estimate
bottle costs, see file "Lead Analytical Burden and Costs_Final.xlsx," worksheet, "Sample Kit_Bottle_$-"

NTNCWSs will not incur the burden or costs to travel to a customer's house to collect a sample. In
addition, NTNCWSs do not incur bottle costs because laboratories provide the 1-liter bottle as part
of their commercial laboratory fee. Thus, they will only incur a burden of 0.5 hours per sample and
$0 costs associated with sample collection.

p) Analyze post-LSLR tap sample (hrs_analyze_lsl_lslr_op, cost_lab_lsl_lslr, cost_commercial_lsl_lslr).

As previously discussed in Section 4.3.2.1.2, activity k), the EPA assumed CWSs serving more than
100,000 people will conduct lead analyses in-house and require 0.44 hours per sample based on
estimates provided by three laboratories (hrs_analyze_lsl_lslr_op). These systems will also incur
consumable costs of $3.92 per sample based on information from three vendors (cost_lab_lsl_lslr).
The remaining CWSs and all NTNCWSs are assumed to use commercial laboratories and incur a cost
of $23.50 per lead sample based on quotes from seven laboratories plus a per sample shipping cost
of $8.70 for a total per sample cost of $32.20 (cost_commercial_lsl_lslr). Note that although the data
variable names are different, the unit costs for lead sample analysis are the same as for lead tap
sampling as presented in Section 4.3.2.1.2.

q) Inform customers of tap sample result (hrs_inform_samp_op, cost_cust_lslr,

hrs_ntncws_cust_lslr_op, cost_ntncws_cust_lslr). Systems must notify their customers of their lead
analytical results from the sample collected following SLR. The EPA made the following assumptions
regarding the burden and/or costs for this notification:

• CWSs of all sizes will send the results to their customers at a per sample burden of 0.05
hours (1 hour per 20 letters) for CWSs serving 3,300 or fewer people and 0.11 hours (1 hour
per 9 letters) for CWSs serving > 3,300 people (hrs_inform_samp_op) and a cost of $0.72
(cost_cust_lslr). These inputs are the same as those used for the tap sampling program. The
burden estimates are based on the public education burden for systems to notify occupants
of results estimated in the 2022 Disinfectants/Disinfection Byproducts, Chemical, and
Radionuclides Rules ICR (Renewal), Exhibit 29 (Note G) (USEPA, 2022a). Systems are also
assumed to mail the post-LSLR sample results. The cost consists of postage ($0.55), paper
($0,019), ink ($0.06), and envelope ($0,092) for a total cost of $0.72/sample. See file,

Final LCRI Economic Analysis

4-207

October 2024


-------
"General Cost Model lnputs_Final.xlsx" for additional information on the sources of these
estimates.

•	NTNCWSs are assumed to notify the people they serve electronically and through posting.
The EPA assumed all NTNCWSs of all size categories will spend 0.5 hours to develop/send e-
mail and an additional 0.5 hours to post the notification publicly for a total of burden 1 hour
per system (hrs_ntncws_cust_lslr_op). In addition, NTNCWSs will incur material costs for
paper posting of $0,079 based on quotes from three vendors (cost_ntncws_cust_lslr). See
derivation file, "General Cost Model lnputs_Final.xlsx" for quotes.

r) Submit annual report on SLR program to State (hrs_report_lcr_op). No later than 30 days after the
end of each replacement program year, systems must submit SLR program information to their
State including the location of each lead and GRR service line and lead connector replaced, the
number of unknown service lines determined to be non-lead, the number of unknown service lines
remaining, their replacement schedule, and other information as required under § 141.90(e). The
EPA assumed that systems will submit this information in the form of an annual report. The EPA
estimated that burden would be higher as system size increases to account for larger the number of
SLRs. The EPA estimated the following burden for CWSs to prepare and submit their annual report:

•	CWSs serving 3,300 or fewer and NTNCWSs: 1 hour.

•	CWSs serving 3,301 to 10,000 people: 2 hours.

•	CWSs serving 10,001 to 100,000 people: 4 hours.

•	CWSs serving more than 100,000 people: 8 hours.

Exhibit 4-97 provides the SafeWater LCR model cost estimation approach for PWS ancillary LSLR
activities including additional cost inputs that are required to calculate these costs.

Final LCR! Economic Analysis

4-208

October 2024


-------
Exhibit 4-97: Service Line Inventory Ancillary Cost Estimation in SafeWater LCR by Activity1





Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th -
Range

Other
Conditions2

Frequency
of Activity

m) Contact customers and conduct site visits prior to SLR

The number of lines replaced multiplied by
the total of the hours per lead line
replacement times the system labor rates,
plus the material cost.

num_lsl_replace *

(hrs_replaced_lsl_contact_op * rate_op +
cost_replaced_lsl_contact)

Cost does not
apply to
NTNCWSs.

All

Model PWSs with
service lines of
lead content

Once a year

n) Deliver filters and 6 months of replacement cartridges at time of SLR

The number of lines replaced multiplied by
the material cost.

num_lsl_replace*cost_filter_hh

Cost applies as
written to
NTNCWSs.

All

Model PWSs with
service lines of
lead content

Once a year

o) Collect tap sample post-SLR3

The number of samples per replaced lead
line multiplied by the number of lines
replaced, multiplied by the total of the hours
per lead line replacement times the system
labor rates, plus the material cost.

(numb_samp_lslr*num_lsl_replace)*((hrs_co
llect_lsl_lslr_op*rate_op)+cost_other_lt_sam
p+cost_pickup_samp)

Cost applies as
written to
NTNCWSs.

All

Model PWSs with
service lines of
lead content

Once a year

p) Analyze post-LSLR tap sample3

The number of samples multiplied by the
probabilities for a sample analyzed in house
and a sample analyzed in a commercial lab
times the different labor and material cost
burdens for each type of analysis.

(((numb_samp_lslr*num_lsl_replace)*pp_lab
_samp)*((hrs_analyze_lsl_lslr_op*rate_op)+
cost_lab_lsl_lslr))+( ((numb_samp_lslr*num_l
sl_replace)*pp_commercial_samp)*cost_co
mmercialjsljslr)

Cost applies as
written to
NTNCWSs.

All

Model PWSs with
service lines of
lead content

Once a year

Final LCR! Economic Analysis

4-209

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost
Per Activity

Lead 90th -
Range

Other
Conditions2

Frequency
of Activity

q) Inform customers of the tap sample result 1

The number of lines replaced multiplied by
the total of the hours per line replacement
times the system labor rates, plus the
material cost.

num_lsl_replace*((hrs_inform_samp_op*rat
e_op)+cost_cust_lslr)

The total hours
per system times
the system labor
rates, plus the
material cost.

(hrs_ntncws_cust
_lslr_op*rate_op)+
cost ntncws cust
_l sir)

All

Model PWSs with
service lines of
lead content

Once a year

r) Submit annual report on SLR program to State

The total hours for reporting per system
multiplied by the system labor rate.

(hrs_report_lcr_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

Model PWSs that
are replacing lead
service lines or
GRR

Once a year

Acronyms: AL = action level; CWS = community water system; GRR = galvanized requiring replacement; LSL = lead
service line; LSLR = lead service line replacement; NTNCWS = non-transient non-community water system; PWS =
public water system.

Notes:

1	The data variables in the exhibit are defined previously in this section with the exception of:

•	pp_lab_samp: Likelihood of in-house analysis (Section 4.3.2.1.2, activity k)).

•	pp_commercial_samp\ Likelihood of commercial lab analysis (Section 4.3.2.1.2, activity k)).

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

2	PWSs with lead content or unknown lines are identified using the data variables and approach described in
Chapter 3, Section 3.3.4.

3	The burden and costs to provide sample bottles (cost_other_lt_samp) under activity o) and conduct analyses
under activity p) are incurred by the State in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina
(ASDWA, 2020a).

4.3.4.5 Estimate of national service line testing and replacement costs

Exhibit 4-98 shows the estimated annualized national cost, under the low and high cost scenarios, of
developing the SL inventory, and conducting the required SLR programs under the 2021 LCRR, the final
LCRI, and the monetized incremental cost discounted at 2 percent. The incremental annual costs range
from $1.2 billion to $1.6 billion in 2022 dollars. Eighty-eight percent of the costs associated with the SL
replacement program is spent on actual SLRs.

Final LCRI Economic Analysis

4-210

October 2024


-------
Exhibit 4-98: Estimated National Annualized Lead Service Line Replacement Costs - 2 Percent

Discount Rate (millions of 2022 USD)



Low Estimate





High Estimate





Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

LSL Inventory

$60.9

$78.6

$17.7

$60.9

$78.4

$17.5

System SLR Plan

$3.1

$17.0

$13.9

$5.8

$17.0

$11.2

System SLR

$12.1

$1,104.7

$1,092.6

$44.7

$1,610.0

$1,565.3

SLR Ancillary
Activities

$8.5

$58.7

$50.2

$13.1

$58.5

$45.4

Total PWS SLR
Program Costs

$84.6

$1,259.0

$1,174.4

$124.5

$1,763.9

$1,639.4

Household SLR Costs

$8.1

$0.0

-$8.1

$26.4

$0.0

-$26.4

Total Annual Lead
Service

Replacement Costs

$92.7

$1,259.0

$1,166.3

$150.9

$1,763.9

$1,613.0

Acronyms: LCRI = Lead and Copper Rule Improvements; LSL = lead service line; PWS = public water system; SLR =
service line replacement; USD= United States dollar.

Notes: The EPA in the Final 2021 LCRR EA (USEPA, 2020) assumed that the cost of customer-side SLRs made under
the goal-based replacement requirement would be paid for by households. The agency also assumed that system-
side SLRs under the goal-based replacement requirement and all SLRs (both customer-side and systems-side)
would be paid by the PWS under the 3 percent mandatory replacement requirement. The EPA made these
modeling assumptions based on the different levels of regulatory responsibility systems faced operating under a
goal-based replacement requirement versus a mandatory replacement requirement. While systems would not be
subject to a potential violation for not meeting the replacement target under the goal-based replacement
requirement, under the 3 percent mandatory replacement requirement the possibility of a violation could
motivate more systems to meet the replacement target even if they had to adopt customer incentive programs
that would shift the cost of replacing customer-side service lines from customers to the system. To be consistent
with these 2021 LCRR modeling assumptions, under the final LCRI, the EPA assumed that mandatory replacement
costs would fall only on systems. Therefore, the negative incremental values reported for the "Household SLR
Costs" category do not represent a net cost savings to households. They represent an assumed shift of the
estimated SLR costs from households to systems. The EPA has insufficient information to estimate the actual SLR
cost sharing relationship between customers and systems at the national level of analysis. The EPA also recognizes
that the cost estimates shown may overestimate the annualized costs due to differences in timing between LCRI
SLR requirements and activities in the SafeWater model. The final LCRI defines the first mandatory service line
replacement year as being from the compliance date of the rule to the end of the next calendar year. For this
economic analysis, EPA divides the period of analysis into 12-month periods, representing one year, beginning on
the effective date of the rule. Thus, the SafeWater model predicts that costs will be incurred on a slightly earlier
schedule than is required in the final LCRI resulting in an overestimate of annualized costs.

4.3.5 PWS POU-Related Costs

Under the final LCRI, CWSs serving 3,300 or fewer people and NTNCWSs with a lead 90th percentile
above the AL of 10 ng/L must evaluate and recommend to their State which compliance option they will
implement from among CCT installation/re-optimization, or the compliance alternatives POU device
installation and maintenance or replacement of lead-bearing materials. For modeling purposes, the EPA
assumed that systems would choose the least costly option from among the first two alternatives.

Final LCRI Economic Analysis

4-211

October 2024


-------
Because of the wide variety of plumbing configurations in individual homes and buildings among
different water systems, it was not possible to estimate burden and costs for replacing lead-bearing
materials. The SafeWater LCR model calculates the annualized cost the system will face under these two
options and selects the least costly alternative.

Those systems approved for the POU provision must develop a plan and implement the program. Note
that once the POU option is started, the system must continue to implement this program even if they
no longer exceed the lead AL in the future.

In addition to the cost to provide and maintain POU devices and educate customers on them, systems
have associated ancillary public education and sampling costs. POU-related costs are grouped into two
subsections:

•	4.3.5.1: POU Device Installation and Maintenance

•	4.3.5.2: POU Ancillary Activities

In addition, Section 4.3.5.3 provides the national annualized POU costs under the low cost and high cost
scenarios at a 2 percent discount rate.

4.3.5.1 POU Device Installation and Maintenance

All costs in this category are grouped into one activity: a) provide, monitor, and maintain POU devices.

a) Provide, monitor, and maintain POU devices (annual_pou_cost_hh). CWSs approved for the POU

program must provide one POU device at each household they serve and continue to maintain the
device. The EPA determined the average number of households per system, which is equivalent to
the number of POU devices by dividing the retail population served by all systems in each of the four
size categories serving 3,300 people or fewer people (pws_pop) by the average number of people
per household (2.53 (numb_hh)) and then dividing by the number of systems per size category, as
shown in Exhibit 4-99. Note that a CWS that serves a non-residential building would also need to
provide a POU device for each tap used for cooking and/or drinking. For modeling purposes, the EPA
did not account for POU devises at non-residential buildings within the CWS service area, which will
underestimate costs for the CWSs that serve non-residential buildings.

Exhibit 4-99: Average Number of Households and POU Devices per CWS

System Size
(Population
Served)

# of Systems

Retail Population

Households (HH)
per System Size
Category

Average HH per
System (Equals
Number of POU
Devices)

pws_pop

A

B

C = B/2.53

<

u"

II

Q

<100

11,732

708,236

279,935

24

101-500

15,084

3,830,126

1,513,884

100

501-1,000

5,330

3,931,488

1,553,948

292

Final LCR! Economic Analysis

4-212

October 2024


-------
System Size
(Population
Served)

# of Systems

Retail Population

Households (HH)
per System Size
Category

Average HH per
System (Equals
Number of POU
Devices)

pws_pop

A

B

C = B/2.53

<

u"

II

Q

1,001-3,300

7,967

15,218,647

6,015,275

755

Acronyms: HH = household; POU = point-of-use.

Notes:

A,B: SDWIS/Fed fourth quarter 2020 "frozen" data set that includes information reported through December 31,
2020.

C: 2.53 is the average number of people per household for the year 2020 (U.S. Census Bureau, 2020). Table AVG1.
Average Number of People Per Household, By Race and Hispanic Origin, Marital Status, Age, And Education of
Householder: 2020. This corresponds to SafeWater data variable: numb_hh.

NTNCWSs must provide a POU device on each tap used for cooking and/or drinking water consumption.
Exhibit 4-100 provides the estimated number of POU devices per NTNCWS based on 11 types of
NTNCWS service categories classified under five Internal Plumbing Code (IPC) categories (business,
industrial, residential, daycare, and school). Two estimates are provided, a minimum that excludes the
installation of POU on bathroom taps and a maximum that includes bathroom taps. Additional detail on
the EPA's approach is provided in "POU lnputs_Final.xlsx."

Exhibit 4-100: Minimum and Maximum Estimated Number of Taps Requiring POU Devices per

NTNCWS

System Size
(Population Served)

Minimum Number of POU Devices

Maximum Number of POU Devices

numb_pou

A

B

<100

3

9

101-500

5

23

501-1,000

9

54

1,001-3,300

16

121

3,301-10,000

41

427

10,001-50,000

150

1,452

50,001-100,000

114

594

100,001-1,000,000

306

1,666

> 1,000,000





Acronyms: POU = point-of-use.

Source: "POU lnputs_Final.xlsx."

Notes:

A: The minimum number of POU devices is based on the weighted average of the number of taps excluding
bathrooms. See Table A-l in "POU lnputs_Final.xlsx," worksheet "NTNCWS_Cost Modeljnputs."

Final LCRI Economic Analysis

4-213

October 2024


-------
B: The maximum number of POU devices is based on the weighted average of number of taps including
bathrooms. See Table A-2 in "POU lnputs_Final.xlsx," worksheet "NTNCWS_Cost Modeljnputs."

The number of POU devices (numb_pou) is multiplied by the unit cost of the POU device installation and
maintenance (annual_pou_cost_hh) to produce the total cost. The EPA used a modified version of the
WBS model to calculate unit costs for POU devices that specifically remove lead. The WBS model
includes the following cost components of a complete POU program:

•	POU device purchase, and scheduling and installation labor;

•	Labor for POU device maintenance; and

•	Materials (replacement filters) for POU device maintenance.

The EPA assumed 25 percent of households receive countertop units and 75 percent receive faucet
mount units. The associated annual average cost is $104 per household per year. The derivation of this
unit cost (annual_pou_cost_hh) is shown in detail in Technologies and Costs for Corrosion Control to
Reduce Lead in Drinking Water (USEPA, 2023b).

Exhibit 4-101 provides the SafeWater LCR model costing approach for installation and maintenance of
POU devices including additional cost inputs that are required to calculate these costs.

Exhibit 4-101: Point-of-Use Device Installation and Maintenance Cost Estimation in SafeWater

LCR by Activity1





Conditions for Cost
to Apply to a Model
PWS



CWS Cost Per Activity

NTNCWS Cost Per
Activity

Lead 90th -
Range1

Other
Conditions

Frequency
of Activity

a) Provide, monitor, and maintain POU devices

Households per system multiplied by the
unit cost of the POU device installation and
maintenance.

(pws_pop/numb_hh)*annual_pou_cost_hh

The number of
POU devices per
system multiplied
by the unit cost of
the POU device
installation and
maintenance.

numb_pou*annual_
pou cost hh

All

Model PWS
installing a
POU device

Once per
year

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; POU = point-

of-use; PWS = public water system.

Notes:

1 Once the POU option is started in response to exceeding the lead AL, systems must continue to implement this
program regardless of their subsequent lead 90th percentile levels. POU installation occurs once with O&M costs
continuing annually.

Final LCR! Economic Analysis

4-214

October 2024


-------
4.3.5.2 POU Ancillary Activities

The EPA has developed costs for one-time ancillary PWS activities related to POU program development
and on-going ancillary activities as shown in Exhibit 4-102. The exhibit provides the unit burden and/or
cost for each activity. The assumptions used in the estimation of each activity follows the exhibit. The
last column provides the corresponding SafeWater LCR model data variable in red/italic font. In a few
instances, some of these activities are conducted by the State instead of the water system. These
activities are identified in the exhibit and further explained in the exhibit notes.

Exhibit 4-102: PWS Ancillary POU-Related Burden and Cost Estimates1

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

b) Develop POU plan and submit
to the State (one-time)2

178 to 328 hrs for CWSs;
148 to 388 hrs for NTNCWSs

hrs_pou_plan_dev_op

c) Develop public education
materials and submit to the
State (one-time)

7 hrs per CWS and NTNCWS

hrs_pe_pou_op

d) Print POU education materials

Burden

0.0025 hrs/sample per CWS
1 hr/NTNCWS

Cost

$0,079 sample per CWS and
NTNCWS

Burden

hrs_print_pe_pou_op
hrs_ntncws_distr_pe_pou_op

Cost

cost_print_pe_pou
cost_ntncws_distr_pe_pou

e) Obtain households for POU
monitoring

0.5 hrs per sample for CWSs only

hrs_samp_volunt_pou_op

f) Deliver POU monitoring
materials and instructions to
participating households

Burden

0.25 hrs/sample per CWS
Cost

$8.77 sample per CWS
$0 per NTNCWS

Burden

hrs_discuss_samp_op
Cost

cost_pou_samp3

g) Collect tap samples after POU
installation

CWS

Burden: 0.40 hrs/sample
Cost: $5.75

NTNCWS
0.5 hrs/sample

CWS

hrs_pickup_samp_ op
cost_pickup_ samp

NTNCWS
hrs_source_op

h) Determine if sample should be
rejected and not analyzed

0.25 hrs/rejected sample for CWSs only

hrs_samp_reject_op

i) Analyze POU tap samples

In-House Burden
N/A

In-House Cost
N/A

Commercial Analysis

$32.30/ sample per CWS and NTNCWSs

In-House Burden
hrs_analyze_samp_op3

In-House Cost
cost_lab_lt_samp3

Commercial Analysis
cost_commerical_lab3

Final LCR! Economic Analysis

4-215

October 2024


-------
Activity

Unit Burden and/or Cost

SafeWaterLCR Data Variable

j) Prepare and submit sample
invalidation request to the
State

2 hrs per sample per CWS and NTNCWS

hrs_samp_in valid_ op

k) Inform customers of POU tap
sample results

CWS

Burden: 0.05 hrs/sample
Cost: $0.72/sample

NTNCWS

Burden: 1 hr/sample
Cost: $0.079/sample

CWS

hrs_inform_samp_ op
cost_cust_lt

NTNCWS

hrs_n tncws_inform_samp_ op
cost ntncws cust It

1) Certify to the State that POU
tap results were reported to
customers

0.66 hrs/year per CWS;
0.66 to 1 hr/year for NTNCWS

hrs_cert_cust_lt_op

m) Prepare and submit annual
report on POU program to the
State

1 hr per CWS;

1 to 8 hrs per NTNCWS

hrs_pou_report_ann_prep_op

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; POU = point-

of-use.

Sources:

b)	& m) "POU lnputs_Final.xlsx", worksheets "CWS_Cost Model Inputs" and "NTNCWS_Cost Model Inputs",
worksheet, "POU Outreach."

c)	& d) Public Education lnputs_CWS_Final.xlsx; Public Education lnputs_NTNCWS_Update.xlsx.

e) -1): Lead Analytical Burden and Costs_Final.xlsx, worksheets "POU_Collect_Analyze_LCRR_LCRI" and

"POU_Sample_Report_LCRR_LCRI."

Notes:

1	Requirements apply only to CWSs serving 3,300 or fewer people and NTNCWSs that exceed the AL and have POU
provision and maintenance as their approved compliance option.

2	The rule does not explicitly include a POU plan. However, the EPA assumed most systems would prepare this plan
prior to implementing a POU program. This assumption may overestimate costs during the first year the program is
implemented.

3	In Arkansas, Louisiana, Mississippi, Missouri, and South Carolina, the State pays for the cost of bottles, shipping,
analysis, and providing sample results to the system (ASDWA, 2020a). Thus, the State will incur the burden and
cost for these activities in lieu of the system.

b) Develop POU plan and submit to the State (hrs_pou_plan_dev_op). Although not required under
the final LCRI, the EPA assumed that systems (i.e., CWSs serving 3,300 or fewer people and
NTNCWSs without CCT132) above the AL that select the POU option would develop a plan to provide
and maintain POU devices for lead removal. The EPA assumed the POU plan would include gathering
background information and identifying plan elements, customer participation (CWSs only),
installation, monitoring and maintenance, and logistics and administration. Each of these plan
elements are included in the overall burden estimate and provided in Exhibit 4-103 and Exhibit
4-104 for CWSs and NTNCWSs, respectively.

Additional detail on each of these plan elements is provided in the file, "POU lnputs_Final.xlsx."

132 The proposed LCRI does not prohibit systems with CCT from selecting the POU option. However, the EPA
assumed systems would re-optimize their CCT.

Final LCRI Economic Analysis

4-216

October 2024


-------
Exhibit 4-103: CWS Burden to Develop a POU Plan (hrs/system)
hrs_pou_plan_dev_op

System Size
(Population
Served)

Gather
Background
Information

Plan for
Customer
Participation

Plan for
Installation

Plan for
Monitoring &
Maintenance

Plan for
Logistics &
Administration

Total

A

B

C

D

E

F=A:E

<100

58

30

30

50

10

178

101-500

58

30

30

50

10

178

501-1,000

108

60

60

100

0

328

1,001-3,300

108

60

60

100

0

328

Source: "POU lnputs_Final.xlsx." This file provides the associated burden for the activities listed in Notes A - E.

Notes:

General:

1.	Under the final LCRI, the POU provision for CWSs is only available to those that serve 3,300 or fewer people.

2.	With the exception of reading the guidance (see note A) and planning for logistics and administration (see
note E), CWSs serving more than 500 people are assumed to incur twice the burden than those serving 500
or fewer people.

A: Includes read and understand "POU or point of entry (POE) Treatment Options for Small Drinking Water
Systems" (USEPA, 2006b) and identify considerations and options for their appropriate system type; prepare a
draft outline of plan elements and submit for management and State approval, as applicable; present a draft
outline of plan elements to system board/management bodies and incorporate feedback; and consult with a legal
expert on property liability and additional insurance.

B: Includes identifying in the plan the types of customer access and maintenance agreements needed and their
schedule for development; includes 2 hours of legal consultation.

C: Includes identifying the number of taps to treat and the schedule and customer priority for installation;
identifying whether vendors or licensed plumbers, and certified operators will install the units and how they will be
managed and tracked; and how and when arrangements for access to installation sites will occur and how they will
be managed and tracked.

D: Includes description of vendor responsibilities and utility responsibilities for monitoring and maintenance of the
POU units; unit maintenance frequencies and checklist for maintenance inspections; POU unit routine replacement
frequencies and protocol for emergency reporting of problems and response; and incorporation of rule-specific
monitoring requirements into the plan.

E: Includes description of contractual agreements and oversight responsibilities for lease agreements. Assumed
this primarily affects CWSs serving 500 and fewer people because they would not have available staff for
maintenance and monitoring of these units.

Exhibit 4-104: NTNCWS Burden to Develop a POU Plan (hours/system)

hrs_pou_plan_dev_op



Gather

Plan for

Plan for
Installation

Plan for

Plan for



System Size

Background

Customer

Monitoring &

Logistics &

Total

(Population Served)

Information

Participation

Maintenance

Administration





A

B

C

D

E

F=A:E

<500

58

0

30

50

10

148

501-10,000

108

0

60

50

10

228

10,001-50,000

208

0

120

50

10

388

50,001-1,000,000

108

0

60

50

10

228

Source: "POU lnputs_Final.xlsx."

Final LCRI Economic Analysis

4-217

October 2024


-------
Notes:

General: No NTNCWS serves more than 1 million people. Two NTNCWSs serve 50,001- 1,000,000 people. These
systems have fewer taps than the average estimated number for those serving 10,001 - 50,000 people. Thus, the
EPA assumed a similar burden for these two largest NTNCWSs as those serving 3,301 - 10,000 people.

A: Includes read and understand "POU or POE Treatment Options for Small Drinking Water Systems" (USEPA,
2006b) and identify considerations and options for their appropriate system type; prepare a draft outline of plan
elements and submit for management and State approval, as applicable; present a draft outline of plan elements
to governing bodies and incorporate feedback; and consult with a legal expert on property liability and additional
insurance.

B: Does not apply to NTNCWSs.

C: Includes identifying the number of taps to treat and the schedule for installation; identifying whether vendors or
licensed plumbers, and electricians will install the units and how these services will be provided; and how and
when arrangements for access to installation sites will occur and how they will be managed and tracked.

D: Includes description of vendor responsibilities for monitoring and maintenance of the POU units; unit
maintenance frequencies and checklist for maintenance inspections; POU unit routine replacement frequencies
and protocol for emergency reporting of problems and response; and incorporation of rule-specific monitoring
requirements into the plan.

E: Includes description of contractual agreements and oversight responsibilities for lease agreements.

c)	Develop public education materials and submit to the State (hrs_pe_pou_op). CWSs serving 3,300
or fewer people and NTNCWSs with a lead ALE that choose the POU option must implement the
POU program including providing public education on the maintenance and use of POU device to all
households they serve. The EPA assumed these systems will incur a one-time burden of 7 hours to
develop these public education materials and submit them to the State for review (hrs_pe_pou_op).
The burden estimate of 7 hours is based on the hours to prepare additional brochure language from
Exhibit 33a of the 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR
(Renewal) (USEPA, 2022a).

d)	Print POU education materials (hrs_print_pe_pou_op, cost_print_pe_pou,
hrs_ntncws_distr_pe_pou, cost_ntncws_distr_pe_pou). The EPA estimated CWSs serving 3,300 or
fewer people will require 0.0025 hours per household to print POU public education materials based
on assumptions for production labor used in the Economic and Supporting Analyses: Short-Term
Regulatory Changes to the Lead and Copper Rule, Exhibit 17 (USEPA, 2007). The EPA assumed that
this material would be provided in addition to the manufacturer's information that comes with the
POU device. The estimated cost for systems to print POU public education material per household is
$0,079 that is the cost of paper and ink. The EPA assumed that there will be no envelope or mailing
costs because public education materials will be provided when the system provides the POU
device. See "General Cost Model lnputs_Final.xlsx" for specific vendor paper and ink quotes. The
EPA assumed NTNCWSs would provide materials via email and post materials publicly with an
estimated burden of 0.5 hours to develop/send e-mail and an additional 0.5 hours to post the
materials, for a total of 1 hour (hrs_ntncws_distr_pe_pou_op). NTNCWSs will also incur a cost for
public education posted materials (cost_ntncws_distr_pe_pou) that will include paper and ink costs
of $0,079, which is the same case as that assumed for CWSs).

e)	Obtain households for POU monitoring (hrs_samp_volunt_pou_op). Under the POU program,
systems must sample one-third of locations with POU devices annually. For CWSs, the EPA assumed
customers can collect these samples. The EPA estimated that a CWSs will incur a burden of 0.5 hours
to obtain customers for POU sampling. The EPA also applied the same inflation percentages, from

Final LCRI Economic Analysis

4-218

October 2024


-------
the assumption associated with lead tap sampling, to the number of required POU samples to
account for the likelihood a customer does not collect the sample (10 percent, 1 -
pp_hh_return_samp), the sample is rejected (5 percent, pp_samp_reject), or invalidated (0.6
percent, pp_samp_invalid). Refer to Section 4.3.2.1.2, activity f) for additional detail.

f)	Deliver POU monitoring materials and instructions to participating households
(hrs_discuss_samp_op, cost_pou_samp). The EPA used the same data variables and inputs for CWSs
to discuss proper sampling procedures with customers of 0.25 hours per sample
(hrs_discuss_samp_op) as under the lead tap program. The EPA also assumed systems will incur the
same non-labor costs to provide a test kit to customers (cost_pou_samp) of $8.77 for CWSs serving
3,300 or fewer people as used for systems without LSLs under the tap sampling program
(cost_5_lt_samp). (See Exhibit 4-12.) EPA also applied the same inflation percentages to the number
of samples to account for the likelihood a customer does not collect the sample (1 -
pp_hh_return_samp), the sample is rejected (pp_samp_reject), or invalidated (pp_samp_invalid).
Refer to Section 4.3.2.1.2, activities f) and h) for additional detailed assumptions.

g)	Collect tap samples after POU installation (hrsjpickup_samp_op, cost_pickup_samp,
hrs_source_op). The EPA uses the same data variable and input for the burden and O&M cost for
CWSs serving 3,300 or fewer to travel to a customer's home to pick-up the collected sample of 0.40
hours (hrs_pickup_samp) and $5.75 (cost_pickup_samp). Refer to Section 4.3.2.1.2, activity i) for
additional detailed assumptions. The EPA also applied the same inflation percentages to the number
of samples to account for the likelihoods a customer would not collect the sample (1 -
pp_hh_return_samp), the sample is rejected (pp_samp_reject), or invalidated (pp_samp_invalid).

For NTNCWSs, the EPA uses the same data variable and input for the burden to collect POU sample
as a source water sample of 0.5 hours/sample (hrs_source_op). Refer to Section 4.3.2.4.2, activity ff)
for additional detailed assumptions. The EPA also inflated the number of samples to account for
invalidated samples (pp_samp_invalid).

CWSs and NTNCWSs must collect tap samples at one-third of the households or taps with POU
devices, respectively. See Exhibit 4-99 and Exhibit 4-100 for the estimated number of POU devices
for CWSs and NTNCWSs, respectively.

h)	Determine if samples should be rejected and not analyzed (hrs_samp_reject_op). The EPA used the
same data variable and input, of 0.25 hours per sample (hrs_samp_reject_op), for the CWS's burden
to review samples collected by customers to determine if they were collected properly or should be
rejected and not submitted for analysis. The EPA also applied the same inflation percentage of 5
percent to the number of samples to account for the likelihood a sample is rejected
(pp_samp_reject). Refer to Section 4.3.2.1.2 activity f) for additional detail of the likelihood a sample
will be rejected and activity j) for the burden to determine if a sample should be rejected.

i)	Analyze POU tap samples (hrs_analyze_samp_op, cost_lab_lt_samp, cost_commercial_lab). Based
on input from laboratories, the EPA assumed CWSs serving 3,300 or fewer people and all NTNCWSs
will use commercial labs for sample analysis; therefore, these systems will not incur any in-house
analytical burden (hrs_analyze_samp_op) or cost (cost_lab_lt_samp). Instead, these systems will
incur a cost of $32.20 per sample (cost_commercial_lab) to ship the POU tap sample to the lab
($8.70) and have it analyzed for lead by a commercial lab ($23.50). That cost corresponds to the

Final LCRI Economic Analysis

4-219

October 2024


-------
same cost input used for systems without LSLs under the lead tap sampling program
(cost_5_commercial_lab). Refer to Section 4.3.2.1.2, activity k) for additional detail. The EPA also
applied the same inflation percentage of 0.6 percent to the number of samples to account for the
likelihood a sample is invalidated (pp_samp_invalid). Refer to Section 4.3.2.1.2 activity f) for
additional detail of the likelihood a sample will be invalidated.

j) Prepare and submit sample invalidation request to the State (hrs_samp_invalid_op). The EPA used
the lead tap sampling data variable and input of 2 hours per request (hrs_samp_invalid_op) for the
burden for CWSs and NTNCWSs to prepare and submit a sample invalidation request to their State.
The EPA assumed that States will approve sample invalidation requests for the 0.6 percent of
samples for which systems will submit these requests (pp_samp_invalid). Refer to Section 4.3.2.1.2
activity f) for additional detail of the likelihood a sample will be invalidated and activity I) for the
burden to request that a sample be invalidated.

k) Inform customers of POU tap sample results (hrs_inform_samp_op, cost_cust_lt,

hrs_ntncws_inform_samp_op, cost_ntncws_cust_lt). The EPA uses the same data variables and
inputs for systems to provide the sampling results collected from POU taps as the lead tap sampling
program. CWSs must report individual lead sample results to customers who participated in the
sampling pool. The EPA estimates that CWSs will require an average of 0.05 hours per customer
(hrs_inform_samp_op). Systems are also assumed to mail these results at a cost of $0.72
(cost_cust_lt). For NTNCWSs, the EPA assumed the systems will deliver materials via email to all
customers and post in a public location at a burden of 1 hour for all system sizes
{hrs_ntncws_inform_samp_op). The EPA assumed NTNCWSs will incur paper and ink costs of $0,079
(cost_ntncws_cust_lt) to post the flyer. Refer to Section 4.3.2.1.2, activity m) for additional detailed
assumptions regarding these four data variables.

I) Certify to the State that POU tap monitoring results were reported to customers

(hrs_cert_cust_lt_op). For both the lead tap and POU monitoring programs, systems must prepare
and submit an annual certification to their State that they informed customers of their monitoring
results. For the POU certification, the EPA used the same data variable and input as used for the lead
tap sampling program. The EPA assumed a burden of 0.66 hours per year for CWSs and NTNCWSs
serving 50,000 or fewer people and 1 hour for those serving more than 50,000 people. Refer to
Section 4.3.2.1.2, activity n) for additional detailed assumptions.

m) Prepare and submit annual POU program Report to the State (hrs_pou_report_ann_prep_op).

Systems must prepare and submit a report of their POU program that includes monitoring results,
any corrective actions if the AL were exceeded, and if requested by the State, any maintenance
activities. The estimated burden and assumptions for CWSs and NTNCWSs are provided in Exhibit
4-105. The EPA assumed systems would submit this report electronically to the State and thus
would incur no paper or mailings costs.

Final LCRI Economic Analysis

4-220

October 2024


-------
Exhibit 4-105: PWS Annual POU Program Report Preparation and Submission Burden

System size
(Population Served)

CWSs

NTNCWSs

hrs_pou_report_ann_prep_op

A

B

<3,300

1

1

3,301-10,000

N/A

2

10,001-50,000

N/A

4

50,001-100,000

N/A

4

100,001-1,000,000

N/A

8

>1,000,000

N/A



Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.

Source: "POU lnputs_Final.xlsx."

Notes:

A, B: Assume reporting and recording keeping is similar to April 2006 EPA guidance on POU/POE devices (USEPA,
2006b).

B: No NTNCWSs serves more than 1 million people. Thus, the burden for this size category is 0.

Exhibit 4-106 provides the SafeWater LCR model cost estimation approach for system ancillary POU
system cost inputs including additional cost inputs that are required to calculate these costs.

Exhibit 4-106: PWS Point-of-Use Ancillary Costing Estimation in SafeWater LCR by Activity1,2,3





Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost Per
Activity

Lead 90th
- Range2

Other Conditions

Frequency
of Activity

b) Develop POU plan and submit to the State

The total hours per system
multiplied by the system labor rate.

(hrs_pou_plan_dev_op*rate_op)

Cost applies as written
to NTNCWSs.

Above AL

Model PWS
selecting POU
installation and
maintenance as
their compliance
option

One time

c) Develop public education materials and submit to the State for review

The total hours per system
multiplied by the system labor rate.

(hrs_pe_pou_op*rate_op)

Cost applies as written
to NTNCWSs.

Above AL

Model PWS
installing POU
device

One time

d) Print POU education material

Final LCR! Economic Analysis

4-221

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost Per
Activity

Lead 90th
- Range2

Other Conditions

Frequency
of Activity

The hours per household multiplied
by the system labor rate and the
material cost.

(pws_pop/numb_hh) *

((hrs_print_pe_pou_op*rate_op)+

cost_print_pe_pou)

The hours per system
multiplied by the
system labor rate and
the material cost.

C (hrs_ntncws_distr_pe
_pou_op*rate_op)+cos
t ntncws distr pe po
u)

Above AL

Model PWS
installing POU
device

Once a
year

e) Obtain households for POU Monitoring

One third of households per system
multiplied by the hours per sample
and the system labor rate. The
number of required samples
(assumed to be one per household)
is inflated to include those
unreturned, invalidated, and rejected
to ensure that the cost reflects the
additional burden that must occur to
meet the sampling requirement.

Cost does not apply to
NTNCWSs.

All

Model PWS
installing POU
device

Once a
year

(((1/3)*(pws_pop/numb_hh))+(((1/3)*
(pws_pop/numb_hh))*( 1-
pp_hh_return_samp))+(( (1/3) *(pws_
pop/numb_hh)) *pp_samp_in valid)+((
(1/3)*(pws_pop/numb_hh))*pp_sam
p_reject))*(hrs_samp_volunt_pou_o
p*rate_op)









f) Deliver POU monitoring materials and instructions to participating households4

One third of households per system
multiplied by the total of the hours
per sample to provide instructions
times the system labor rate, plus the
cost of materials per sample. The
number of required samples
(assumed to be one per household)
is inflated to include those
unreturned, invalidated, and
rejected, to ensure that the cost
reflects the additional burden that
must occur to meet the sampling
requirement.

Cost does not apply to
NTNCWSs.

All

Model PWS
installing POU
device

Once a
year

((((1/3)*(pws_pop/numb_hh)))+(((1/3
) *(pws_pop/numb_hh)) *pp_samp_in
valid)+(((1/3)*(pws pop/numb hh))*(
1-

pp_hh_return_samp))+(((1/3)*(pws_
pop/numb_hh)) *pp_samp_reject))*((
hrs_discuss_samp_op*rate_op)+cos
t pou samp)

















Final LCRI Economic Analysis

4-222

October 2024


-------
Conditions for Cost to
Apply to a Model PWS

CWS Cost Per Activity	NTNCWS Cost Per	Frequency

Activity	Lead 90th other Conditions of Activity

- Range'1

g) Collect tap samples after POU installation

One third of households per system
multiplied by the hours per sample
and the system labor rate. The
number of required samples
(assumed to be one per household)
is inflated to include those
unreturned, invalidated and rejected
to ensure that the cost reflects the
additional burden that must occur to
meet the sampling requirement.

((((1/3)*(pws_pop/numb_hh)))+(((1/3
) *(pws_pop/numb_hh)) *pp_samp_in
valid)+((( 1/3)*(pws pop/numb hh))*(
1-

pp_hh_return_samp))+(((1/3)*(pws_
pop/numb_hh))*pp_samp_reject))*((
hrs_pickup_samp_op*rate_op)+cost
_pickup_samp)

One third of the
number of POU
devices per system
multiplied by the total
of the hours per
system times the
system labor rate, plus
the material cost. The
number of required
samples is inflated to
include those
invalidated to ensure
that the cost reflects
the additional burden
that must occur to
meet the sampling
requirement.

(((1/3)*numb_pou)+(((
1/3)*numb_pou)*pp_s
amp_in valid) )*((hrs_so
urce_op*rate_op)+cost
pou samp)

All

Model PWS
installing POU
device

Once a
year

h) Determine if samples should be rejected and not analyzed

One third of households per system
with a sample expected to be
rejected (calculated by multiplying
the total number of required samples
by the likelihood of rejection)
multiplied by the hours per sample
and the system labor rate.

(((1/3)*(pws_pop/numb_hh))*pp_sa
mp_reject)*(hrs_samp_reject_op*rat
e op)

Cost does not apply to
NTNCWSs.

All

Model PWS
installing POU
device

Once a
year

Final LCRI Economic Analysis

4-223

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost Per
Activity

Lead 90th
- Range2

Other Conditions

Frequency
of Activity

i) Analyze POU tap samples3

1/3 of households per system
multiplied by the material cost of the
commercial lab analysis per sample.
All systems installing POUs are
assumed to use commercial labs for
sample analysis.

The number of samples (assumed to
be one per HH) is inflated to include
those invalidated, to ensure that the
cost reflects the additional burden
that must occur to meet the
sampling requirement.

((((1/3) *(pws_pop/numb_hh ))+(((1/3)
*(pws_pop/numb_hh))*pp_samp_inv
ali d))*cost_commercial_lab)

1/3 of the number of
POU devices per
system multiplied by
the material cost of the
commercial lab
analysis per sample.
All systems installing
POUs are assumed to
use commercial labs
for sample analysis.
The number of
required samples is
inflated to include
those invalidated to
ensure that the cost
reflects the additional
burden that must occur
to meet the sampling
requirement.

Systems will collect
one sample per POU
device.

(((1/3)*numb_pou)+(((
1/3)*numb_pou)*pp_s
amp_in valid)) *cost_co
mmercial lab

All

Model PWS
installing POU
device

Once a
year

j) Prepare and submit sample invalidation request to the State

1/3 of HHs per system where a
sample is expected to be invalid
(calculated by multiplying the total
number of required samples by the
likelihood of invalidation) multiplied
by the hours per sample and the
system labor rate.

(((1/3)*(pws_pop/numb_hh))*pp_sa

mp_invalid)*(hrs_samp_invalid_op*r

ate_op)

1/3 of the number of
POU devices per
system where a
sample is expected to
be invalid (calculated
by multiplying the total
number of required
samples by the
likelihood of
invalidation) multiplied
by the hours per
sample and the
system labor rate.

C C 1/3)*numb_pou)*pp_
samp_invalid) *(hrs_sa
mp_in valid_op *rate_op
)	

All

Model PWS
installing POU
device

Once a
year

Final LCRI Economic Analysis

4-224

October 2024


-------




Conditions for Cost to
Apply to a Model PWS



CWS Cost Per Activity

NTNCWS Cost Per
Activity

Lead 90th
- Range2

Other Conditions

Frequency
of Activity

k) Inform customers of POU tap sample results

1/3 of HHs per system multiplied by
the total of the hours per sample
times the system labor rate plus the
material cost per sample.

CC1/3) *(pws_pop/numb_hh))*( (hrsjnf
orm samp op*rate op)+cost cust 1
t)

The hours per system
multiplied by the
system labor rate, plus
the material cost.

(hrs_ntncws_inform_s

amp_op*rate_op)+

cost_ntncws_cust_lt

All

Model PWS
installing POU
device

Once a
year

1) Certify to State that POU tap sample results were reported to customers

The total hours per system to submit
certification multiplied by the system
labor rate.

Cost applies as written
to NTNCWSs.

All

Model PWS
installing POU
device

Once a
year

(hrs_cert_cust_lt_op*rate_op)









m) Prepare and submit annual POU program report to the State

The total hours reporting cost per
system multiplied by the system
labor rate.

(hrs_pou_report_ann_prep_op*rate_
op)

Cost applies as written
to NTNCWSs.

All

Model PWS
installing POU
device

Once a
year

Acronyms: AL = action level; CWS = community water system; HH = household; NTNCWS = non-transient non-

community water system; POU = point-of-use; PWS = public water system.

Notes:

1	The data variables in this exhibit are defined previously in this section with the exception of:

•	numb_pou\ Number of POU devices per PWSs that elects POU option (Section 4.3.5.1).

•	pp_commercial_samp\ Likelihood a sample will be analyzed by a commercial laboratory (Section 4.3.2.1.2,
activity k)).

•	pp_lab_samp: Likelihood a sample will be analyzed in-house (Section 4.3.2.1.2, activity k)).

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

2	Once the POU program is started in response to a lead ALE, systems must continue to implement this program
regardless of their subsequent lead 90th percentile levels.

3	For CWSs, the number of POU devices equals the number of households.

4	The burden and costs to provide sample bottles (cost_pou_samp) under activity f) and conduct analyses under
activity i) are incurred by the State in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina (ASDWA,
2020a).

4.3.5.3 Estimate of PWS National Point-of-Use Device Installation and Maintenance Costs

As shown in Exhibit 4-1, the estimated incremental annual costs of POU device installation and
maintenance range from $2.7 million to $3.7 at a 2 percent discount rate in 2022 dollars.

Final LCRI Economic Analysis

4-225

October 2024


-------
4.3.6 PWS Lead Public Education, Outreach, and Notification Costs

Systems will incur labor and non-labor costs to provide consumer notice related to individual lead and
copper tap results, to conduct education and outreach regardless of their lead 90th percentile level, and
to conduct public education requirements in response to a lead 90th percentile level exceedance. These
activities and associated costs are detailed in Sections 4.3.6.1 through 4.3.6.3, respectively. Systems with
multiple lead ALEs will be required to conduct additional public education activities. These activities and
associated costs are detailed in Section 4.3.6.4. Exhibit 4-119 provides the SafeWater LCR model cost
estimation approach for system lead public education and outreach costs for Sections 4.3.6.1 and
4.3.6.2 and is located at the end of Section 4.3.6.2. Similar exhibits for Sections 4.3.6.3 and 4.3.6.4 are
provided at the end of each section as Exhibit 4-126 and Exhibit 4-132, respectively. Section 4.3.6.5
provides the national annualized lead public education and outreach costs at a 2 percent discount rate.

Public education requirements for systems implementing a POU program were previously discussed in
Section 4.3.5.2 in activities d), e), and f).

4.3.6.1 Consumer Notice

Under the final LCRI, water systems must notify consumers of their individual lead and/or copper results
within three business days of learning the results, regardless if they above or below the AL. The EPA
assumed CWSs would use mail and NTNCWSs would use posting and electronic notification. For CWSs,
the EPA included the burden of 0.05 to 0.11 hrs and cost of $0.72 per notification as part of the Lead Tap
Sampling Costs using hrs_inform_samp_op and cost_cust_lt, respectively. Similarly, the EPA used the
burden of 1 hour and cost of $0.79 per monitoring period as part of the Lead Tap Sampling Costs using
NTNCWS_inform_samp_op and cost_NTNCWS_inform_lt for NTNCWSs.

Exhibit 4-107 provides the unit burden and/or cost for the CWS and NTNCWSs to submit a copy of the
consumer notification and a certification that the notification was distributed in a manner that meets
the rule requirements to their State. The assumptions used in the estimation of the unit burden follow
the exhibit. The last column provides the corresponding SafeWater LCR model data variable in red/italic
font.

Exhibit 4-107: PWS Burden for Consumer Notification of Lead and Copper Tap Sampling

Results

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

a) Develop lead consumer

notice materials and submit
to the State for review (one
time)

7 hours/PWS

hrs_consumer_notice_devel_op

b) Provide a copy of the
consumer notice and
certification to the State

0.08 hrs/customer contact

hrs_samp_n otice_ op

Source: "Public Education lnputs_CWS_Final.xlsx" and "Public Education lnputs_NTNCWS_Final.xlsx."

a) Develop lead consumer notice materials and submit to the State for review

(hrs_consumer_notice_devel_op). The EPA assumed that States will provide templates to CWSs and

Final LCRI Economic Analysis

4-226

October 2024


-------
NTNCWSs to develop consumer notice materials to include individual lead and copper tap results, an
explanation of the health effects of lead and copper, a list of steps consumers can take to reduce
exposure to lead and copper in drinking water, the maximum contaminant level goal and the AL for
lead and copper and the definitions for these two terms. The EPA also assumed that systems will
incur a burden of 7 hours to develop the lead consumer notice and submit it to their State for
review. The burden estimate is based on the hours to prepare additional brochure language from
Exhibit 33a of the 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules
ICR (Renewal) (USEPA, 2022a).

b) Provide a copy of the consumer notice to the State (hrs_samp_notice_op). CWSs and NTNCWSs
must submit a copy of the consumer notification and a certification that the notification was
distributed in a manner that meets the rule requirements to their State. The EPA assumed systems
would require 5 minutes or 0.083 hours to submit an electronic copy ($0) of this notice and
certification to the State (hrs_samp_notice_op).

4.3.6.2 Activities Regardless of Lead 90th Percentile Level

The EPA has developed CWS costs for activities associated with new public education requirements
under the final LCRI that are independent of a system's lead 90th percentile range, as provided in Exhibit
4-108. The exhibit provides the unit burden and/or cost for each activity. The assumptions used in the
estimation of the unit burden and/or cost follow the exhibit. The last column provides the
corresponding SafeWater LCR model data variable in red/italic font.

Note that the final LCRI would require enhanced outreach for water systems that do not meet their SLR
rate (see Section 4.3.4.3). However, the burden and cost associated with this outreach is not included in
the cost model because the EPA assumes full compliance with the regulation.

Exhibit 4-108: PWS Burden and Cost for Public Education Activities that Are Independent of

Lead 90th Percentile Levels

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

c) Update CCR language (one-
time)

0.5 hrs/CWS serving <3,300 people;
1 hr/CWS serving > 3,300 people

hrs_ update_ccr_op

d) Develop new customer
outreach plan (one-time)

4 hrs/CWS with LSL or GRR SLs serving
<50,000 people;

8 hr/CWS with LSLs or GRR SLs serving >
50,000 people

hrs_cust_plan_op

e) Develop approach for
improved public access to
lead health-related
information and tap sample
results (one-time)

10 to 40 hours/CWS

hrs_pub_ access_ op

f) Establish a process for public
access to information on
known or potential lead

5 hrs/CWS with lead content SLs serving
<3,300 people;

10 hrs/CWS with lead content SLs serving >
3,300 people

hrs_ access_lsl_ op

Final LCRI Economic Analysis

4-227

October 2024


-------
Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

content SL locations and tap
sample results (one-time)





g) Maintain a process for

public access to lead health
information, known or
potential lead content SL
locations, and tap sample
results

No LSLs

2 hrs/CWS serving < 3,300 people
4 hrs/CWS serving > 3,300 people

With LSLs

6 hrs/CWS serving < 3,300 people
12 hrs/CWS serving > 3,300 people

hrs_maint_lsl_op

h) Respond to customer
request for known or
potential lead content SL
information

0.05 hrs/request;
$0/request

hrs_ hh_reques t_ op;
cost_hh_request

i) Respond to requests from
realtors, home inspectors,
and potential home buyers
for known or potential lead
content SL information

0.05 hrs/request;
$0/request

hrs_other_request_op;
cost_other_request

j) Develop a list of local and
State health agencies

CWSs

0.08 hrs/ local and State health

hrs_hc_list_op

k) Develop lead outreach

materials for local and State
health agencies and submit
to the State for review (one-
time)

7 hrs/CWS

hrs_pub_devel_hc_op

1) Deliver lead outreach

materials for local and State
health agencies

CWSs

24 to 208 hrs/local and State health agency;
$71.65/ local and State health

hrs_hc_op;
cost_hc

m) Develop public education for
known or potential lead
content SL disturbances and
submit to the State (one-
time)

7 hrs/CWS with LSLs

hrs_pub_devel_wtr_op

n) Deliver public education for
SL disturbances

0.083 hours/delivery;
$0.21/delivery

hrs_pub_deliv_wtr_op;
cost_pub_deliv_wtr_ed

o) Deliver filters and 6 months
of replacement cartridges
during SL disturbances

$64.00/household

cost_filter_hh

p) Develop inventory-related
outreach materials and
submit to the State for
review (one time)

7 hours per system

hrs_pe_lsl_gen_develop_op

q) Distribute inventory-related
outreach materials

CWS

0.4426 to 0.0026/household per year
$0.35 to $0.48/household per year

NTNCWS

1 hr per system per year
$0.79 per system per year

CWS

hrs_pe_lsl_gen_dist_op
cost_pe_lsl_gen

NTNCWS

hrs_ntncws_pe_lsl_gen_dist_op
cost_ntncws_pe_lsl_op

Final LCRI Economic Analysis

4-228

October 2024


-------
Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

r) Provide translation services
for public education
materials

1.5 to 3.38 hrs/CWS per year
$200 to $800/CWS per year

hrs_translate_phone_op;
cost_translate_cws

s) Certify to the State that lead
outreach was completed

CWSs

2	hrs/CWS serving <50,000 people;

3	hrs/CWS serving > 50,000 people

NTNCWSs

0.66 hrs/NTNCWS serving <50,000 people;
1 hr/NTNCWS serving > 50,000 people

CWSs

hrs_pe_ certify_ quarterly_op
NTNCWSs

hrs_cert_outreach_annual_op

Acronyms: CCR = consumer confidence report; CWS = community water system; GRR = galvanized requiring

replacement; LSL = lead service lines; NTNCWS = non-transient non-community water system; SL = service line.

Sources:

c) - n): "Public Education lnputs_CWS_Final.xlsx."

o): Technologies and Costs for Corrosion Control to Reduce Lead in Drinking Water (USEPA, 2023b).

p -s): "Public Education lnputs_CWS_Final.xlsx;" "Public Education lnputs_NTNCWS_Final.xlsx."

c)	Update Consumer Confidence Report (CCR) language (hrs_update_ccr_op). The EPA is requiring
CWSs to update information about lead in the CCR. CWSs will incur a one-time burden

(hrs_update_ccr_op) to update their CCR with the revised lead health effects language and for
systems with LSLs and/or GRR service lines to further update their materials to include information
about a system's SLR program and opportunities to replace lead and GRR service lines. Systems with
lead and GRR service lines must also include information on how to access the service line inventory
and how to access the results of all tap sampling in the CCR. The EPA assumed for:

•	CWSs serving 3,300 or fewer, 50 percent will use CCRiWriter133 or a similar program to
update their CCR and will incur no additional burden because the standard text will already
be in the program. This percentage is based on current CCRiWriter users who are generally
small systems. All other CWSs serving 3,300 or fewer are assumed to incur 1 hour, giving an
average burden of 0.5 hours across all systems in this size category.

•	CWSs serving more than 3,300 people will not use CCRiWriter and will incur a burden of 1
hour.

d)	Develop new customer outreach plan (hrs_cust_plan_op). In response to final LCRI requirements,
CWSs with lead and GRR service lines will develop a new customer outreach plan. The EPA
estimated that systems serving 50,000 or fewer people will incur 4 hours of burden and those
systems serving more than 50,000 people will take 8 hours to develop the plan.

e)	Develop approach for improved public access to lead health-related information and tap sample
results (hrs_pub_access_op). CWSs will incur a one-time burden to develop improved public access
to lead data that includes lead health-related data and tap monitoring results. The EPA assumed
that systems serving 3,300 or fewer people with no existing system website will make data available

133 The CCRiWriter is a web-based program that allows water systems to enter data and generate their annual CCR.

Final LCRI Economic Analysis

4-229

October 2024


-------
for the public in hard copy form at the system office. Systems serving more than 3,300 will update
their existing websites. The one-time burden estimates are included in Exhibit 4-109.

Exhibit 4-109: One-Time Burden (per CWS) to Develop Approach for Improved Access to Lead

Information

System Size
(Population Served)

Hours to Develop Approach for
Improved Public Access to Lead
Data (all CWSs)

hrs_pub_access_op

<3,300

10

3,301-10,000

20

10,001-50,000

25

50,001-100,000

30

100,001-1,000,000

35

>1,000,000

40

Acronyms: CWS = community water system.

Source: "Public Education lnputs_CWS_Final.xlsx," worksheet, "Public Access."

f)	Establish a process for public access to information on known or potential lead content SL
locations (hrs_access_lsl_op). Under the final LCRI, CWSs must establish a way for customers and
the public to access information on potential lead content SLs. The EPA assumed that this will be a
one-time burden that applies to all CWSs with potential lead content SLs regardless of lead 90th
percentile level. The EPA assumed systems serving 3,300 or fewer with no existing system website
will make the information available in hard copy form at the system office and incur 5 hours to print
materials and set up a viewing location. The EPA assumed systems serving more than 3,300 people
will provide access to information about lead line locations and the replacement program by adding
content to an already existing website with links to materials and incur a burden of 10 hours per
system. Note that the hours associated with determining locations of potential lead content SLs and
establishing a replacement outreach program are described in Section 4.3.4.1.

g)	Maintain a process for public access to health information, known or potential lead content SL
locations, and tap sample results (hrs_maint_lsl_op). CWSs with potential lead content SLs would
also incur an annual burden to maintain a way for the public to access lead health and potential lead
content SL information. The EPA assumed that:

•	CWSs serving 3,300 or fewer people have no existing system website. Those without lead or
GRR service lines will require 2 hours to maintain lead-related data, such as lead sample
results in hard copy files. Those systems with lead or GRR service lines take an additional 4
hours to provide updated potential lead content SL locational information for a total annual
burden of 6 hours.

•	CWSs serving more than 3,300 people without lead or GRR service lines will require 4 hours
to update their website annually with lead-related data. Those with lead or GRR service lines
will require a total of 12 hours to update their website with health information, new
potential lead content SL locational information, and tap sample results.

Final LCRI Economic Analysis

4-230

October 2024


-------
h) Respond to customer requests for known or potential lead content SL information

(hrs_hh_request_op, cost_hh_request). CWSs will incur a per household burden to respond to
potential lead content SL information requests from homeowners and residents
(hrs_hh_request_op). The EPA assumed CWSs with potential lead content SLs will respond by phone
and spend an average of 3 minutes (0.05 hours) per request. The EPA assumed systems without
potential lead content SLs may still get inquiries, but the burden to be negligible. The EPA assumed
systems will not provide printed materials in response to these inquiries. Therefore, the cost to
respond to request from households (cost_hh_request) is $0.

The EPA estimated the likelihood that that a particular household in a system with potential lead
content SLs will request information about potential lead content SLs to be 0.0032 each year
(pp_hh_request_lslr). This was computed as a weighted average over the 32-year period from Year 4
through Year 35 of the analysis, as shown in Exhibit 4-110. Underlying this estimate are the
assumptions that these requests would come from 10 percent of households having young children
(under six years of age) present in each year in those systems having potential lead content SLs. As
shown in Exhibit 4-110, the EPA estimated that in Year 4, the likelihood that a household already has
children under the age of six is 0.11 (Column C, based on Columns A and B). The EPA also estimated
that the likelihood a new child will be born at a household each year for Years 5 through 35 is 0.0294
(Column E, based on Columns D and A). Column F (using the results in Columns C and E) shows the
calculation of the weighted average likelihood of a child under six being present in a given
household in each of the 32 years of the period of analysis. Lastly, Column G applies the assumption
that only 10 percent of those households will request LSL information.

Exhibit 4-110: Likelihood that a Resident Will Request Information about potential lead

content SLs

Total
Households
in the
United
States

Households
with
Children
under Six
Years Old

Likelihood a
Household
Has
Children
under Six
Years Old in
Year 4

Births
per year
in 2020

Likelihood of a

Birth per
Household per
Year in Years
5 to 35

32-Year
Weighted
Average
Likelihood a
Household has
Children
under Six
Years Old Each
Year

Likelihood that a
Household Having
Children under Six
Will Request
Potential Lead
Content SL
Information Each
Year

A

B

C = (B/A)

D

E = (D/A)

F =

(C+(31*E))/32

G = F*0.1

pp_hh_request_lslr

122,802,852

13,512,226

0.11

3,613,647

0.0294

0.03195

0.0032

Acronyms: SL = service line.

Sources:

A-D: Information is also documented in Public Education lnputs_CWS_Final.xlsx.

A-C: U.S. Census Bureau, 2019 American Community Survey 1-Year Estimates
(https://data.census.gov/cedsci/table?q=households%20and%20families&tid=ACSSTlY2019.S1101).
D: CDC. 2019. https://www.cdc.gov/nchs/nvss/births.htm. Accessed January 7, 2022.

Final LCRI Economic Analysis

4-231

October 2024


-------
i) Respond to requests from realtors, home inspectors, and potential home buyers for known or
potential lead content SL information (hrs_other_request_op, cost_other_request). CWSs with
potential lead content SLs must also respond to requests for potential lead content SL information
from other parties {e.g., realtors, home inspectors, and potential homebuyers). The EPA assumed
the same burden of 0.05 hours to respond to these requests by phone as assumed for responding to
a request from a homeowner (hrs_other_request_op). The EPA assumed systems without potential
lead content SLs may still get inquiries, but that the burden will be negligible. The EPA assumed
systems will not provide printed materials in response to these inquiries. Therefore, the material
cost to respond to other potential lead content SL information requests (cost_other_request) is $0.

The EPA conducted the following steps to determine the estimated number of requests that systems
will receive each year from other parties (numb_other_request).

1. Determined the percentage of households with children under the age of 6 that moved using
United States Census Bureau data from 2020, as shown in Exhibit 4-111.

Exhibit 4-111: Households (HHs) with Children under 6 and That Moved

Total number
of HHs

Total number
of HHs that
moved

Total HHs with
any children
under 6

Total HHs with
any children
under 6 that
moved

Percent of all
HHs that
moved

Percent of HHs
with any children
under 6 that
moved

A

B

C

D

E = (B/A)*100%

F = (D/A)*100%

48,493

5,019

14,080

1,873

10.35%

3.86%

Source: U.S. Census Bureau, Current Population Survey, 2020 Annual Social and Economic Supplement.
https://www.census.gov/data/tables/2020/demo/geographic-mobility/cps-2020.html

2. Multiplied the percentage of households with children under the age of 6 that moved by the
number of households per system. The EPA assumed that other parties would request LSL
information on 10 percent of the resulting number of households. The resulting number of
requests (numb_other_request) is provided in Exhibit 4-112.

Exhibit 4-112: Number of Potential Lead Content SL Information Requests from Realtors,
Home Inspectors, and Potential Home Buyers



Number of
CWSs

Total Population
Served

Average
Population per
CWS

Average
Households per
CWS

Number of requests
per CWS

System Size
(Population
Served)





C= B/A

D=C/2.53

E=D*3.86%*10%



A

B

pws_pop



numb_other_request

<100

11,732

708,236

60

24

0

101-500

15,084

3,830,126

254

100

0

501-1,000

5,330

3,931,488

738

292

1

1,001-3,300

7,967

15,218,647

1,910

755

3

Final LCRI Economic Analysis

4-232

October 2024


-------
System Size
(Population
Served)

Number of
CWSs

Total Population
Served

Average
Population per
CWS

Average
Households per
CWS

Number of requests
per CWS

A

B

C= B/A

D=C/2.53

E=D*3.86%*10%

pws_pop

numb_other_request

3,301-10,000

5,026

29,565,710

5,883

2,325

9

10,001-50,000

3,374

74,162,674

21,981

8,688

34

50,001-100,000

571

39,629,417

69,404

27,432

106

100,001-
1,000,000

421

99,359,362

236,008

93,284

360

>1,000,000

24

46,638,891

1,943,287

768,098

2,967

Acronyms: CWS = community water system.

Notes:

A, B: SDWIS/Fed, current through December 31, 2020 with an adjustment to systems serving <100. The EPA
increased the population to 25 for those systems reported in SDWIS/Fed as serving < 24 people. This resulted in an
increase in population from 701,258 to 708,236 for this size category.

D: Estimated as 2.53 people per household (numb_hh) for the year 2020 (U.S. Census Bureau, 2020). Table AVG1.
Average Number of People per Household, by Race and Hispanic Origin, Marital Status, Age, and Education of
Householder: 2020. https://www2.census.gov/programs-survevs/demo/tables/families/2020/cps-
2020/tabavgl.xls.

E: Assumes of the households with children ages 6 and under that moved, i.e., 3.86 percent (see Column F, Exhibit
4-111), 10 percent would request information.

j) Develop list of local and State health agencies (hrs_hc_list_op). All CWSs must conduct annual
outreach to State and local health agencies to discuss the sources of lead in drinking water, health
effects of lead, steps to reduce exposure to lead in drinking water, and information on DSSA
activities. The EPA expects CWSs will work with their State to conduct increased lead outreach to
health agencies. Systems will incur a one-time upfront burden to develop an initial list of local and
State health departments in their service area. The EPA assumed systems would require 5 minutes
for each health agency or 0.08 hours per agency, which is the same burden the EPA used to estimate
the burden to develop an initial contact list of schools and child care facilities for the lead in drinking
water testing program (hrs_school_identify_op) in activity ii) of Section 4.3.2.5.1. The burden per
health agency is multiplied by the number of health agencies (numb_ha +1), shown in Exhibit 4-113,
to develop the unit cost.

Exhibit 4-113: Estimated Number of Health Agencies

System Size
(Population served)

# of Organizations per system

numb_ha +1

<100,000

2

100,001 -1,000,000

3

>1,000,000

17

Final LCRI Economic Analysis

4-233

October 2024


-------
Source: "Public Education lnputs_CWS_Final.xlsx," worksheet, "Outreach to Health Depts." EPA assumed
each system would contact one additional State health agency.

k) Develop lead outreach materials for local and State health agencies and submit to the State for
review (hrs_pub_devel_hc_op). All CWSs are assumed to incur burden to develop lead outreach
materials for State review that will be distributed to local and State health agencies. The EPA
assumed systems will incur a burden of 7 hours, which is based on the hours to prepare additional
brochure language from Exhibit 33a of the 2022 Disinfectants/Disinfection Byproducts, Chemical,
and Radionuclides Rules ICR (Renewal) (USEPA, 2022a).

I) Deliver lead outreach to local and State health agencies (hrs_hc_op, cost_hc). CWSs must provide
the results of school testing to local and State health care agencies within 30 days of receiving the
results. The EPA assumed that a portion of schools and child care facilities will be tested each month
and therefore would report the results monthly. In addition, once a year the information to the local
and State health department would include the outreach materials developed under activity k), as
well as the results of any DSSA activities in response to a sample above 10 ng/L (as previously
discussed in Section 4.3.2.5 and Section 4.3.3.3.3, respectively). Systems will also incur annual
burden to make any necessary updates to the list of organizations. The resulting monthly burden
estimates for conducting outreach to health care agencies are provided in Exhibit 4-114.

Exhibit 4-114: Annual CWS Burden (per system) to Conduct Outreach to Local and State

Health Agencies

System Size
(Population served)

# of Organizations
per system

Production
Time per
organization

Distribute
Letters per
month

Update List
of

Organizations
(annual)

Total (Annual
Burden)



A

B

C = A*B

D

E = (C*12)+D



numb_ha +1







hrs_hc_op

<3,300

2

1

2

0

24

3,301-100,000

2

1

2

1

25

100,001-1,000,000

3

1

3

2

38

>1,000,000

17

2

34

2

410

Notes

A: See Exhibit 4-113.

B: The EPA assumed systems would require 1 hour and 2 hours each month to prepare a cover letter and assemble
the results of lead in drinking water testing at schools and child care facilities for systems serving 1 million people
or fewer and more than 1 million people, respectively. In addition, once per year, the information to local and
State health departments will also include DSSA.

D: The EPA assumed zero burden for systems serving 3,300 or fewer people. For CWSs serving 3,301 to 100,000
people, the EPA assumed an annual burden of 1 hour per system to update the list of organizations. For systems
serving more than 100,000 people, the EPA assumed an annual burden of 2 hours per system.

The EPA assumed systems will deliver the information to State and local health departments via certified
mail each month at an estimated cost of $5.97 per organization (cost_hc) per month that includes paper

Final LCRI Economic Analysis

4-234

October 2024


-------
($0,019), envelope ($0,092), ink ($0.06), and certified mail ($5.80) for a total annual cost of $71.65 per
health agency.

m) Develop public education materials for known or potential SL disturbances and submit to the State
(hrs_pub_devel_wtr_op). CWSs with lead, GRR, and unknown service lines must send public
education to customers and consumers when there is scheduled water-related work that could
result in disturbances of service lines and will incur a one-time burden to develop materials. The EPA
assumed:

•	All CWSs with lead, GRR, and unknown service lines will develop these materials.

•	The development of public education materials is similar across all types of public education
because systems will use EPA-developed templates and incur a burden of 7 hours, which is
based on the hours to prepare additional brochure language from Exhibit 33a of the 2022
Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR (Renewal) (USEPA,
2022a).

Under the final LCRI, outreach is also required due to disturbances to lead, GRR, or unknown service
lines during inventorying. The EPA assumed disturbances will occur when a system conducts
mechanical or vacuum excavation during the inventory process. Outreach activities are assumed to
include a door hanger and filter. These costs are captured in Section 4.3.4.1.2.

n) Deliver public education materials for SL disturbances (hrs_pub_deliv_wtr_op,

cost_pub_deliv_wtr_ed). CWSs that cause disturbances to a lead, GRR, or lead status unknown
service line will also incur an annual burden to deliver public education to impacted households
about the potential for elevated lead levels in drinking water as a result of the disturbance. The
annual burden to deliver public education (hrs_pub_deliv_wtr_op) is assumed to be 5 minutes per
delivery (0.083 hours). Systems are assumed to provide the messaging on door hangers that they
will distribute when they are in the area conducting work. The average cost of doorhangers is $0.21
based on quotes from three vendors (cost_pub_deliv_wtr_ed). See "Public Education
lnputs_CWS_Final.xlsx," worksheet "Service Line Disturbances" for specific quotes.

In Section 4.3.4.1.2, the EPA estimated the frequency of events that would result in exposed service
lines during normal operation (this input was used to estimate the proportion of unknowns in the
inventory that would be identified each year during normal operation). These events included meter
replacement (6 percent per year), water main replacement134 (1 percent per year), and other
activities including water meter reading, service line repair or replacement, backflow prevention
inspection, and other street repair or capital improvement projects (0.5 percent per year). As a
simplifying assumption, the EPA used the same total of 7.5 percent per year to estimate the percent
of lead, GRR, and unknown service lines that are disturbed each year and require delivery of public

134 In the final LCRI, the EPA specified that water systems must provide a filter (and 6 months of replacement
cartridges) and public education material not only if disturbance results from replacement of an inline water
meter, a water meter setter, or connector, but also if disturbance results from replacement of water main. The
EPA increased the estimate frequency of events that would result in a disturbance to reflect this rule change.

Final LCRI Economic Analysis

4-235

October 2024


-------
education material (perc_hh_water_wrk)135. This may be an overestimate because some activities
such as backflow prevention inspection and meter reading may not result in a disturbance.

o) Deliver filters and 6 months of replacement cartridges during SL disturbances (cost_filter_hh).

Similar to activity n) above, CWSs are required to provide filters and replacement cartridges
whenever there is a physical disturbance of a lead, GRR, or lead status unknown service line that
involves replacement of a meter, gooseneck, pigtail, or other connector. They also must be provided
when a physical disturbance results from the replacement of a water main whereby the service line
pipe is physically cut. As discussed in activity n), the EPA assumes the likelihood of these
disturbances to be 7.5 percent (perc_hh_water_wrk). The EPA assumed that the pitchers and filters
delivered to each resident to use for six months following a replacement will cost $64 on average
(including shipping and filter replacement). See Technologies and Costs for Corrosion Control to
Reduce Lead in Drinking Water (USEPA, 2023b) for additional detail.

The EPA assumes that the pitchers and POU filters delivered to each resident to use for six months
following SLR will cost $64 on average (including shipping and filter replacement). See Technologies
and Costs for Corrosion Control to Reduce Lead in Drinking Water (USEPA, 2023b) for additional
detail.

p) Develop inventory-related outreach materials and submit to the State for review

(hrs_pe_lsl_gen_develop_op). Under the final LCRI, CWSs and NTNCWSs must provide notification
to consumers served by lead, GRR, or service lines of unknown material. The notification includes
information on the health effects and sources of lead in drinking water (including LSLs), how to
access the SLR plan, how to have water tested for lead, actions consumers can take to reduce
exposure to lead, and information about the opportunities for SLR. In addition, the materials must
include instructions for consumers to notify the water system if they think the material
categorization is incorrect. CWSs and NTNCWSs will incur a one-time burden to develop these
outreach materials. The EPA assumed that systems will use EPA-developed templates as a starting
point for the notice but will adjust the template as needed to fit with specific system characteristics
resulting in an average burden of 7 hours per system. The 7 hour estimate comes from Exhibit 33a
of the 2022 Disinfectants/Disinfection Byproducts, Chemical and Radionuclides ICR (Renewal)
(USEPA, 2022a).

q) Distribute inventory-related outreach materials (hrs_pe_lsl_gen_dist_op; cost_pe_lsl_gen;

hrs_ntncws_pe_lsl_gen_dist_op; cost_ntncws_pe_lsl. CWSs will incur an annual cost to distribute
outreach materials to households served by lead, GRR, or unknown service lines. The EPA assumes
systems will use a combination of separate mailings and inserts that are part of the water bill (each
50 percent). The burden for CWSs to annually distribute these materials is provided in Exhibit 4-115.

The per household cost to distribute the materials in the water bill is $0.16 (cover letter: paper =
$0,019 + ink = $0.06; brochure: $0,019 + ink = $0.06). The per household mailing cost for systems
serving 500 or fewer people of $0.80 includes the material cost of $0.16 plus an envelope of $0,092

135 For the proposed LCRI EA, the EPA assumed that 5.9 percent of households will be impacted annually by water-
related work disturbances and would receive public education based on the estimated life of a water main, meter,
and other SLRs provided by Massachusetts Water Resources Authority. Utilizing these data, the EPA previously
assumed an average 17-year life of a meter, CWSs would replace a meter at an annual rate of 5.9 percent.

Final LCRI Economic Analysis

4-236

October 2024


-------
and $0.55 postage. The per household mailing cost for systems serving more than 500 people of
$0.55 includes the material cost of $0.16 plus an envelope of $0,092 and bulk rate postage of
$0,299. The EPA averaged the two methods for an estimated per household delivery cost of $0.48
and $0.35 for systems serving 500 or fewer people and more than 500 people, respectively. See
"General Cost Model lnputs_Final.xlsx" for additional information about paper, ink, envelope, and
postage costs.

Final LCRI Economic Analysis

4-237

October 2024


-------
Exhibit 4-115: CWS Annual Burden (per household) to Distribute General Inventory-related Outreach

System Size
(Population
Served)

Separate
mailing per
System

Bill Stuffer per
System

Subtotal per
System

Production (hrs
per HH)

Number of HH
per system

Separate/Bill
Stuffer (hrs per
HH)

Total (hrs per HH)
hrs_pe_lsl_gen_dist_op



A

B

C = (A+BJ/2

D

E

F = C/E

G = D + F

<100

15

6

10.5

0.0025

24

0.4401

0.4426

101-500

15

6

10.5

0.0025

100

0.1046

0.1071

501-1,000

25

10

17.5

0.0025

292

0.0600

0.0625

1,001-3,300

25

10

17.5

0.0025

755

0.0232

0.0257

3,301-10,000

120

30

75

0.0025

2,325

0.0323

0.0348

10,001-50,000

120

30

75

0.0025

8,688

0.0086

0.0111

50,001-100,000

120

30

75

0.0025

27,432

0.0027

0.0052

100,001-1,000000

120

30

75

0.0025

93,284

0.0008

0.0033

>1,000,000

120

30

75

0.0025

768,098

0.0001

0.0026

Source: "Public Education lnputs_CWS_Final.xlsx," worksheet, "Targeted Outreach."

Note:

A: The EPA assumption regarding the burden per system to conduct separate mailings.

B: The EPA assumption regarding the burden per system to mail materials with the water bill.

C: The EPA assumes that half of systems will conduct separate mailings and the other half will include targeted outreach materials with the water bill.

D: The EPA assumes 0.25 hours per 100 brochures for production. Estimate is based on assumptions for production labor used in the Economic and Supporting
Analyses: Short-Term Regulatory Changes to the Lead and Copper Rule (Exhibit 17).

E: "Public Education lnputs_CWS_Final.xlsx," worksheet, "Number of Households."

Final LCRI Economic Analysis

4-238

October 2024


-------
For NTNCWSs, the EPA assumed these systems will provide outreach via e-mail and public posting.
The EPA assumed a burden of 0.5 hour to develop/send e-mail for all system size categories and an
additional 0.5 hours to post the notification publicly for a total annual burden of 1 hour per system.
The EPA assumed that NTNCWSs will provide electronic notification and posting. Material costs of
$0.79 per system are for paper ($0,019) and ink ($0.06). See file, "General Cost Model
lnputs_Final.xlsx", worksheet "Paper_Envelopes" and worksheet, "Ink" for paper and ink based on
costs from three vendors, respectively.

r) Provide translation services for public education materials (hrs_translate_phone_op,
cost_translate_CWS)136. Under the final LCRI, water systems serving a large proportion of
consumers with limited English proficiency, as determined by the State, must include in all
public education materials listed under 40 CFR 141.85 information in the appropriate
language(s) regarding the importance of the materials. These systems must also either: 1)
include contact information for persons served by the water system to obtain a translated copy
of or translation assistance with the public education materials, or 2) pre-emptively provide the
public education materials in the appropriate language(s). In addition, the final LCRI requires, as
a condition of primacy, that States provide technical assistance to water systems in meeting the
requirement to provide translation assistance in communities with a large proportion of
consumers with limited English proficiency.

The EPA's approach for developing unit costs is to estimate phone and written translation labor
and non-labor costs for subsets of PWSs that must conduct public education under the final
LCRI. Note that the pre-2021 LCR and 2021 LCRR also required translation support in the case of
a lead ALE. The EPA did not estimate translation costs for the pre-LCR and 2021 LCRR, which
underestimates baseline costs, resulting in an overestimate of incremental translation costs
from both the pre-2021 LCR and 2021 LCRR baselines to the final LCRI. Throughout this analysis,
the EPA relied on data and assumptions related to the translation component of the CCR
Revisions rulemaking, which are presented in the document, Analysis of the Economic Impacts
of the Final Consumer Confidence Reports Rule Revisions (USEPA, 2024a), hereafter referred to
as the "Final CCR3 EA."

As the first step of this analysis, the EPA estimated the likelihood that the water systems serve a
large proportion of non-English speaking customers and require translation under the LCRI. The
EPA used a simplifying assumption that no NTNCWSs serve a large proportion of non-English
speaking customers because NTNCWSs are often businesses such as schools, factories, office
buildings, and hospitals and the organization would already have staff available to provide
translation services for employees if needed. For CWSs, the EPA assumed that they will provide
translation services if a system meets at least one of the following criteria: 1) at least 5 percent
of the system's total population served, or 2) at least 1,000 people served, have limited English
proficiency. This criterion is based on assumptions developed in the Final CCR3 EA (USEPA,
2024a). The Final CCR3 EA estimated the number of CWSs that would meet this threshold using

136 The EPA updated the estimates for translation between the proposed and final LCRI to reflect changes in the
proposed and Final CCR3 EA. In particular, changes reflect new information gathered through a review of 120
system websites and CCRs. Revisions to this analysis also reflect the final LCRI requirement that States must
provide translation if requested by the system as a condition of primacy.

Final LCRI Economic Analysis

4-239

October 2024


-------
data from the American Community Survey (ACS) 2016-2020 5-Year Estimates, which provided
the population of metropolitan areas with limited English proficiency (U.S. Census Bureau,
2022). The EPA utilized this same dataset but disaggregated the percentages into nine system
size categories for the LCRI instead of the four system size categories used in the Final CCR3 EA.
The percentages ranged from 7 percent of small systems serving 100 or fewer people to 100
percent for large systems serving more than 100,000 people. These percentages are provided in
Exhibit 4-116 and represent the likelihood that any translation assistance will be needed
(p_translation). Note that many States have not set a limited English proficiency threshold. If the
State sets a higher threshold, fewer systems would need to provide translation services. Thus,
assuming a 5 percent/1,000 person threshold for each State may be conservative and
overestimate the percent of systems needing to provide translation services.

Exhibit 4-116: Likelihood that the CWS Has a High Proportion of non-English Speaking

Customers

System Size (Population
Served)

SafeWater LCR Variable

p_translation

<100

7%

101-500

11%

501-1,000

14%

1,001-3,300

18%

3,301-10,000

28%

10,001-50,000

50%

50,001-100,000

94%

100,001-1,000,000

99%

>1,000,000

100%

The EPA then estimated the likelihood of CWSs that will use phone support instead of written
translations (p_translation_phone). This likelihood is based on recent research conducted for
the Final CCR3 EA whereby the EPA randomly selected 120 CWSs and reviewed both their
websites and most recent CCRs to investigate what methods (if any) these systems already
employ to provide meaningful access to consumers with limited English proficiency. From this
research, the Final CCR3 EA reported that among systems that already provide translation, an
estimated 70 percent of systems serving more than 100,000 people provided a translated
report. For systems serving 100,000 or fewer people, a majority, estimated also at 70 percent,
included a contact number for translation assistance. The EPA used the same estimates for the
Final LCRI EA.

With respect to whether the CWSs or States will be providing translation services, the EPA made
a simplifying assumption that small systems serving 10,000 or fewer people will rely on the
State, whereas systems serving more than 10,000 will provide their own translations based on
the analysis of 120 CCRs as described above. The EPA used the same assumption for phone
translation (p_translation_phone_cws) and written translation (p_translation_written_cws).
Note that the Final CCR3 EA used a more detailed analysis of each State's support; however, the

Final LCRI Economic Analysis

4-240

October 2024


-------
EPA expects translation support for LCRI to be more consistent across States given it is a
condition of primacy and therefore national level estimates are used in this cost analysis.

The final step in the analysis is to estimate unit labor and non-labor for each type of public
education required under 40 CFR 141.85 for phone and written translation. For the purposes of
modeling translation costs using the SafeWater LCR model, the EPA developed unit labor burden
and non-labor costs in the form of cost per CWS per year for the following public education
materials:

•	Public education to consumers served by lead, GRR, or unknown service lines once per
year;

•	Public education to all consumers twice per year for CWSs with a lead ALE; and

•	Public education materials twice per year for CWSs with multiple lead ALEs.

Note that the EPA assumed that translation services for notification of tap sample results (as
required under 40 CFR 141.85) would be negligible because the customers that receive these
notifications are a small subset of all customers. Moreover, CWSs will have already had contact
with the sampling population when providing sampling instructions. Similarly, the EPA assumes
that translation services for service line disturbances would be negligible since they apply to a
small portion (7.5 percent) of the subset of customers with a lead, GRR, or unknown service line.
The EPA also assumed that State and local health departments are not likely to request
translation of public education materials because the staff should be proficient in English in
order to perform the expectations of their jobs.

For the three types of public education materials listed in the bullets above, the EPA developed
unit burden for phone and written translation, as presented in Exhibit 4-117 and Exhibit 4-118,
respectively. Note that burdens are only for systems serving more than 10,000 people because
the State is assumed to provide all translation services (phone and written) for systems serving
10,000 or fewer people.

To determine the phone translation burden per year in Exhibit A-lllExhibit 4-118, the EPA
multiplied the estimated per-call duration by the estimated average number of calls per year.
The EPA estimated the per-call duration to be 15 to 30 minutes based on assumptions used in
the Final CCR3 EA (USEPA, 2024a) for small systems. The EPA used the average of these two
estimates (0.375 hours) for all system sizes. The EPA did not use the Final CCR 3 EA average call
duration of 0.5 hours for systems serving more than 100,000 people because the LCRI public
education materials are not expected to vary by size as was assumed for the CCRs.

The estimated number of calls per year depends on the type of public education materials. The
EPA estimated nine calls per year for CWSs serving 10,000 or more people with a lead ALE based
on data from the Final CCR3 EA on phone calls received by systems and States for translation
support of CCRs (USEPA, 2024a). The EPA assumed that the number of calls for CCR translation
would be the same as the number of calls for translation of public education materials following
a lead ALE because both communications must be delivered to all customers. For CWSs with
lead, GRR, or unknown service lines, the number of calls is estimated to be half, or four calls per
year since the public education materials will be delivered to only a subset of customers (those

Final LCRI Economic Analysis

4-241

October 2024


-------
served by a lead, GRR, or unknown SL). The EPA assumes that systems with multiple lead ALEs
will receive an additional nine calls beyond that estimated for a lead ALE in Column D of Exhibit
4-117. This estimate is based on the enhanced outreach required for systems with multiple lead
ALEs that could result in more customers becoming aware of the ALEs and requesting
translation assistance. The EPA assumed there are no non-labor costs to provide phone
translation, consistent with the assumptions made in the Final CCR3 EA (USEPA, 2024a).

Final LCRI Economic Analysis

4-242

October 2024


-------
Exhibit 4-117: Unit Burden for CWSs to Provide Phone Translation by Type of Public Education Material

System Size
(Population
Served)

LOE per
Translation

Public Education for Customers Served
by Lead, GRR, and Unknown SL

Public Education for All Customers in
CWSs with a Lead ALE

Public Education for All Customers in
CWSs with Multiple Lead ALEs

Average
Number of
Phone
Calls per
Year

Total Translation Burden
per CWS per Year
(SafeWater LCR Input:
hrs_translate_phone_op)

Average
Number of
Phone Calls
per Year

Total Translation Burden
per CWS per Year
(SafeWater LCR Input:
hrs_translate_phone_op)

Average
Number of
Phone Calls
per Year

Total Translation Burden
per CWS per Year
(SafeWater LCR Input:
hrs_translate_phone_op)

A

B

C=A*B

D

E = A*D

F

G = A *F

<100

0.375

0

0

0

0

0

0

101-500

0.375

0

0

0

0

0

0

501-1,000

0.375

0

0

0

0

0

0

1,001-3,300

0.375

0

0

0

0

0

0

3,301-10,000

0.375

0

0

0

0

0

0

10,001-50,000

0.375

4

1.5

9

3.375

9

3.375

50,001-100,000

0.375

4

1.5

9

3.375

9

3.375

100,001-
1,000,000

0.375

4

1.5

9

3.375

9

3.375

>1,000,000

0.375

4

1.5

9

3.375

9

3.375

Acronyms: ALE = action level exceedance; CWS = community water system; GRR = galvanized requiring replacement; SL = service line.

Notes:

General: The EPA assumes that for phone translation services, CWSs serving more than 10,000 people will provide phone translation, whereas the State will
provide phone translation for CWSs serving 10,000 or fewer people.

A: This is the average burden for a CWS to provide translation call-in support. The EPA assumed that these calls would be a duration of between 15 to 30
minutes, consistent with the assumptions for phone support for systems translating the CCR in the Final CCR3 EA (USEPA, 2024a).

B: The average number of calls per year for systems with lead, GRR or unknown service lines is estimated to be approximately half the number of calls estimated
for the Final CCR3 EA because the education materials will be delivered to a subset of customers as opposed to all customers.

D: The average number of calls per year for systems that have a lead ALE is assumed to be the same as the number of calls anticipated for the CCR because the
materials are being delivered to all customers. The estimate of nine calls per year is based on interviews with water systems related to phone calls requesting
translation of their CCRs that were conducted to develop costs for CCR3.

F: The additional average number of calls per year for systems with multiple lead ALEs is based on enhanced outreach and more customers potentially becoming
aware of the ALEs and requesting translation assistance. The EPA estimates that systems will receive double the calls requesting translation assistance based on
this enhanced outreach. The number of calls shown is the incremental calls beyond the number that is estimated for a lead ALE in Column D.

Final LCR! Economic Analysis

4-243

October 2024


-------
To determine the unit cost per CWS per year for written translation, the EPA estimated the cost
to translate each public education material multiplied by the number of languages and the
number of documents being translated per year. To estimate costs of translating one public
education document, the EPA reviewed public education templates and estimated that they are
1,000 words or less. The EPA then multiplied the 1,000 words by $0.20 per word, for a total of
roughly $200. This $0.20 per word cost estimate is consistent with responses provided by water
systems to the EPA during discussions of the cost to translate CCRs as part of developing the
costs estimates for the Final CCR3 EA (USEPA, 2024a). The EPA made a simplifying assumption
that the average cost for a contractor to translate one document would be the same regardless
of system size and type of public education material.

To estimate the number of languages needed, the EPA reviewed ACS data (U.S. Census Bureau,
2022). The EPA estimated that systems serving more than 50,000 people will likely translate
their materials into two languages because ACS data showed that systems of this size may have
multiple languages spoken in their service area. Specifically, almost 90 percent of CWSs serving
100,000 or more and almost 50 percent of CWSs serving between 50,000-100,000 serve
communities in which more than one language is spoken by at least 1,000 people or in five
percent of the households in the service area (U.S. Census Bureau, 2022).

As was true for phone support, the average number of public education materials being
translated per year depends on the type of material. The EPA assumed that systems with lead,
GRR or unknown service lines will translate one document for notification of SL materials each
year. CWSs that have a lead ALE are assumed to deliver public education materials that require
translations twice per year; thus, the EPA assumed two translations. Systems with multiple lead
ALEs may require additional translations to meet the requirements; therefore, the EPA assumed
two additional written translations per year for systems with multiple lead ALEs. Exhibit 4-118
provides the unit cost for CWSs to provide a written translation for those with lead, GRR, or
unknown service lines; a lead ALE; and multiple lead ALEs.

Final LCRI Economic Analysis

4-244

October 2024


-------
Exhibit 4-118: Unit Cost for CWSs to Provide Written Translation by Type of Public Education Material

System Size
(Population
Served)

Average
Cost per
Translated
PE
Material

Number
of

Languages

Public Education for Customers Served
by Lead, GRR, and Unknown SL

Public Education for All Customers in
CWSs with a Lead ALE

Public Education for All Customers in CWSs
with Multiple Lead ALEs

Annual Number
of PE Materials
Being Translated

Total Translation
Cost per CWS per

Year
(SafeWater LCR
Input:
cost_ translate_ cws)

Annual Number
of PE Materials
Being Translated

Total Translation
Cost per CWS per

Year
(SafeWater LCR
Input:
cost_translate_cws)

Annual Number of
PE Materials Being
Translated

Total Translation
Cost per CWS per

Year
(SafeWater LCR
Input:
cost_translate_cws)







C

D = A*B*C

E

F = A*B*E

G

H = A*B*G

<100

$200

1

0

$0

0

$0

0

$0

101-500

$200

1

0

$0

0

$0

0

$0

501-1,000

$200

1

0

$0

0

$0

0

$0

1,001-3,300

$200

1

0

$0

0

$0

0

$0

3,301-10,000

$200

1

0

$0

0

$0

0

$0

10,001-
50,000

$200

1

1

$200

2

$400

2

$400

50,001-
100,000

$200

2

1

$400

2

$800

2

$800

100,001-
1,000,000

$200

2

1

$400

2

$800

2

$800

>1,000,000

$200

2

1

$400

2

$800

2

$800

Acronyms: ALE = action level exceedance; CWS = community water system; GRR = galvanized requiring replacement; PE = public education.

Notes:

General: The EPA assumes that for written translation services, CWSs serving more than 10,000 people will provide written translation, whereas the State will
provide written translation for CWSs serving 10,000 or fewer people.

A: This is the estimated average cost for a the CWS to pay for contractor support to provide written translation service, based on a typical word count of public
education materials of 1,000 multiplied by $0.20 per word for translation services.

B. Assumes one language for systems serving 10,000 or fewer people and two languages for systems serving more than 10,000 based on data from the ACS,
which provided the population of metropolitan areas with limited English proficiency.

C: Assumes translation of one document per year for notification of SL material for customers served by a lead, GRR, or unknown service line.

E: Assumes one document per 6-month period for a total of two documents for CWSs with a lead ALE per year.

Final LCR! Economic Analysis

4-245

October 2024


-------
G: Assumes CWSs with multiple ALEs must produce one document per six-month period for a total of two additional documents per year. The number of
written translations shown is the incremental written translations beyond the number that is estimated for a lead ALE in Column E.

Final LCRI Economic Analysis

4-246

October 2024


-------
s) Certify to the State that lead outreach was completed (hrs_pe_certify_quarterly_op,

hrs_cert_outreach_annual_op). CWSs have quarterly, semi-annual, and annual public education
requirements in response to a lead ALE. Thus, CWSs must report the certification on a quarterly
basis. The EPA estimated an average 0.33 and 0.5 hours to review public education certifications
under the pre-2021 LCR based on data from North Carolina and Indiana, respectively. These are two
States that responded to an ASDWA survey about LCR implementation. The EPA took these
estimates to review a public education certification and doubled them because systems are
expected to incur a larger burden for developing the materials than the States to review them.137
These estimates were then multiplied by 0.75 to account for quarters in which there is less
information to report on the self-certification. Then the numbers were multiplied by four to account
for the quarterly frequency of the self-certification letter. The EPA assumed that each certification
for systems serving 50,000 or fewer people would require 0.5 hours or 2 hours annually (based on
the lower burden reported from North Carolina) and 0.75 hours/certification or 3 hours annually for
CWSs serving more than 50,000 people (based on the higher burden reported from Indiana).

NTNCWSs will also incur burden to certify to the State that they met their annual public education
and outreach requirements (hrs_cert_outreach_annual_op). The EPA assumed that NTNCWSs will
submit an annual certification to the State electronically and incur a burden of 0.66 hours for
systems serving 50,000 or fewer people and 1 hour for those serving more than 50,000 people.
Estimates are based on input from North Carolina (0.33 hours) and Indiana (0.5 hours), respectively,
for the burden to review the system's public education certification in response to a 2016 ASDWA
survey about LCR implementation. Estimates were doubled since systems are expected to incur a
larger burden to prepare the certification than needed for the State's review. A copy of the
questionnaire and each State's responses are available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

Note that this certification is assumed to include the consumer notice discussed in Section 4.3.6.1,
activities that are required when a system exceeds the lead AL that are described in Section 4.3.6.3, and
required activities when a system has multiple lead ALEs that are described in Section 4.3.6.4. In
addition, under the final LCRI, systems must resubmit copies of their public education and outreach
materials along with the certification.

Exhibit 4-119 provides details on how costs are calculated for PWS public education activities that apply
regardless of a system's lead 90th percentile level for activities a) through s) including additional cost
inputs that are required to calculate these costs.

137 Based Exhibit 35 and 48 of the 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules
ICR (Renewal) (USEPA, 2022a), the system burden to prepare the public education certification (referred to as the
Public Education letter) was 1 hour compared to 0.5 hours for the State review. The EPA increased the estimated
burden based on input from North Carolina and Indiana but retained the relationship that systems would incur
double the burden than the State.

Final LCRI Economic Analysis

4-247

October 2024


-------
Exhibit 4-119: PWS Lead Public Education Unit Costing Approach in SafeWater LCR by Activity1

CWS Cost Per Activity

NTNCWS Cost
Per Activity

Conditions for Cost to Apply to a
Model PWS

Frequency
of Activity





Lead-

go111 -

Range

Other Conditions2



a) Develop lead consumer notice materials and submit to the State for review

The total hours per system
multiplied by the system labor
rate.

(hrs_consumer_notice_devel_op*
rate op)

Cost applies as
written to
NTNCWS

All

All model PWSs

One time

b) Provide a copy of the consumer notice and certification to the State

The total hours per system
multiplied by the system labor
rate.

(hrs_samp_notice_op*rate_op)

Cost applies as
written to
NTNCWS

All

All model PWSs

Once per
event

c) Update CCR language

The total hours per system
multiplied by the system labor
rate.

(hrs update ccr op*rate op)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

One time

d) Develop new customer outreach plan

The total hours per system
multiplied by the system labor
rate.

(hrs cust plan op*rate op)

Cost does not
apply to
NTNCWSs.

All

Model PWSs with service lines of
lead or unknown content

One time

e) Develop approach for improved public access to lead health-related information and tap
sample results

The total hours per system
multiplied by the system labor
rate.

(hrs pub access op*rate op)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

One time

f) Establish a process for public access to information on known or potential lead content SL
locations and tap sample results

The total hours per system
multiplied by the system labor
rate.

(hrs access Isl op*rate op)

Cost does not
apply to
NTNCWSs.

All

Model PWSs with service lines of
lead or unknown content

One time

g) Maintain a process for public access on lead health information, known or potential lead
content SL locations, and tap sample results

The total hours per system
multiplied by the system labor
rate.

(hrs maint Isl op*rate op)

Cost does not
apply to
NTNCWSs.

All

All model PWSs

Once a
year

h) Respond to customer requests for known or potential lead content SL information

The number of requests from
homeowners and residents
multiplied by the total of the hours

Cost does not
apply to
NTNCWSs.

All

Model PWSs with service lines of
lead or unknown content

Once a
year

Final LCR! Economic Analysis

4-248

October 2024


-------
CWS Cost Per Activity

NTNCWS Cost
Per Activity

Conditions for Cost to Apply to a
Model PWS

Frequency
of Activity





Lead-
90th -
Range

Other Conditions2



per request times the system
labor rate, plus the material cost.

(pp_hh_request_lslr*(pws_pop/nu
mb_hh))*((hrs_hh_request_op*rat
e op)+cost hh request)









i) Respond to requests from realtors, home inspectors, and potential home buyers for known
or potential lead content SL information

The number of requests from
realtors, home inspectors, and
potential homebuyers multiplied
by the total of the hours per
request times the system labor
rate, plus the material cost.

numb_other_request*((hrs_other
_request_op*rate_op)+cost_other
request)

Cost does not
apply to
NTNCWSs.

All

Model PWSs with service lines of
lead or unknown content

Once a
year

j) Develop list of local and State health agencies

The number of State and local
health agencies per system times
the total hours per health agency
multiplied by the system labor
rate.

(numb_ha+1)*(hrs_hc_list_op*rat
e op)

Cost applies as
written to
NTNCWSs.

All

All model PWSs

One time

k) Develop lead outreach materials for local and State health agencies and submit to the State
for review

The total hours per system
multiplied by the system labor
rate.

(hrs_pub_devel_hc_op*rate_op)

Cost applies as
written to
NTNCWSs.

All

All model PWSs

One time

I) Deliver lead outreach to State and local health agencies

The number of State and local
health agencies per system times
the total hours per health agency
multiplied by the system labor
rate.

(numb_ha+1)*((hrs_hc_op*rate_o
p)+cost he)

Cost applies as
written to
NTNCWSs.

All

All model PWSs

Once a
year

m) Develop public education material for known or potential SL disturbances and submit to the
State

The total hours per system
multiplied by the system labor
rate.

(hrs_pub_devel_wtr_op*rate_op)

Cost does not
apply to
NTNCWSs.

All

Model PWSs with service lines of
lead or unknown content

One time

Final LCRI Economic Analysis

4-249

October 2024


-------
CWS Cost Per Activity

NTNCWS Cost
Per Activity

Conditions for Cost to Apply to a
Model PWS

Frequency
of Activity





Lead-
90th -
Range

Other Conditions2



n) Deliver public education for SL disturbances

The percentage of the
households in the system having
water work done multiplied by the
total of the hours per household
times the system labor rate, plus
the material cost.

((hh_remain_lsl+hh_unknown_re
main) *perc_hh_ water_ wrk) *( (hrs_
pub_deliv_wtr_op*rate_op)+cost_
pub deliv wtr ed)

Cost does not
apply to
NTNCWSs.

All

Model PWSs with service lines of
lead or unknown content

Once a
year

o) Deliver filters and 6 months of replacement cartridges during disturbances of SLs

The percentage of the
households in the system having
water work done multiplied by the
total material cost.

((hh_remain_isi+hh_unknown_re
main)*perc hh water wrk)*cost f
ilter_hh

Cost does not
apply to
NTNCWSs.

All

Model PWSs with service lines of
lead or unknown content

Once a
year

p) Develop inventory-related outreach materials and submit to the State for review

The total hours per system
multiplied by the system labor
rate.

hrs_pe_lsl_gen_develop_op*rate
op

Cost applies as
written to
NTNCWSs.

All

All Model PWS with service lines of
lead or unknown content

Once

q) Distribute inventory-related outreach materials

The number of remaining
households with LSLs or an
unknown line multiplied by the
hours per household and the
system labor rate, plus the
material cost per household.

(hh_remain_lsl +hh_unknown_re
main) *( (hrs_pe_isi_gen_dist_op *r
ate_op)+cost_pe_isi_gen)

The total hours
per system
multiplied by the
system labor
rate, plus the
material cost
per system.

(hrs_ntncws_pe
_lsl_gen_dist_o
p*rate_op)+cost
_ntncws_pe_lsl
gen

All

All Model PWS with service lines of
lead or unknown content

Once per
year

r) Provide translation services for public education materials

The total hours per system
multiplied by the system labor
rate, plus the material cost.

(hrs_transiate_phone_op*rate_op
)+cost translate cws

Cost does not
apply to
NTNCWSs.

Below
AL

All model PWSs providing
translation services either by
telephone or written

pjtranslation

Once a
year

The total hours per system
multiplied by the system labor
rate, plus the material cost.



Above
AL

p_translation_phone
p_translation_phone_cws



Final LCRI Economic Analysis

4-250

October 2024


-------
CWS Cost Per Activity

NTNCWS Cost
Per Activity

Conditions for Cost to Apply to a
Model PWS

Frequency
of Activity





Lead-
90th -
Range

Other Conditions2



(hrs_translate_ale_phone_op*rat
e op)+cost translate ale cws









The total hours per system
multiplied by the system labor
rate, plus the material cost.

(hrs_translate_ale_phone_op*rat
e op)+cost translate ale cws

Multiple
ALEs

s) Certify to State that lead outreach was completed

The total hours per system
multiplied by the system labor
rate.

(hrs_pe_certify_quarterly_op*rate
_op)

The hours per
system

multiplied by the
system labor
rate.

hrs_cert_outrea
ch_annual_op*r
ate op

All

All model PWSs

Once a
year

Acronyms: AL = action level; ALE = action level exceedance; CCR = consumer confidence report; CWS = community
water system; LSL = lead service line; NTNCWS = non-transient non-community water system; POU = point-of-use;
PWS = public water system; SL = service line.

Notes:

1	The data variables in the exhibit are defined previously in Sections 4.3.6.1 and 4.3.6.2 with the exception of:

• rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

2	PWSs with lead content or unknown lines are identified using the data variables and approach described in
Chapter 3, Section 3.3.4.

4.3.6.3 Public Education Activities in Response to Lead ALE

The final LCRI retains the public education requirements of the pre-2021 LCR for systems that exceed
the lead AL and includes the 2021 LCRR requirement for systems to update their mandatory public
education language. The EPA has developed system costs for these activities, as provided in Exhibit
4-120. The exhibit provides the unit burden and/or cost for each activity. The assumptions used in the
estimation of the unit burden follow the exhibit. The last column provides the corresponding SafeWater
LCR model data variable in red/italic font.

Exhibit 4-120: PWS Public Education Burden in Response to Lead ALE

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

t) Update mandatory
language for lead ALE
public education and
submit to the State for
review (one-time)

7 hrs per CWS and NTNCWS

hrs_pe_al_devel_op

Final LCRI Economic Analysis

4-251

October 2024


-------
Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

u) Deliver lead ALE public
education materials to all
customers

CWSs

0.0025 hours/household;
$0.27 to $0.40/CWS

NTNCWSs
1 hr/NTNCWS
$0.079/NTNCWS

CWSs

hrs_distr_edu_op;
cost_pe_lcr_delivery

NTNCWSs

hrs_n tncws_distr_ edu_ op;
cost_ntncws_pe_lcr_delivery

v) Post notice to website

0.5 hrs/CWSs serving > 50,000
people

hrs_web_op

w) Prepare press release

10 hrs/press release per CWS
serving > 3,300 people;
$0/press release

hrs_pr_op;
cost_pr

x) Contact public health
agencies to obtain
additional organizations
and update recipient list

0.5 hrs/CWSs serving <3,300
people;

1.5 hrs/CWSs serving 3,301 to
100,000 people;

2.5 hrs/CWS serving > 100,000
people

hrs_ha_op

y) Notify public health
agencies and other
organizations

0.0025 hours/organization/CWS;
$5.97/organization/CWS

hrs_ distr_ agenci es_pe_ op;
cost_pe_lead_ale

z) Consult with the State on
other public education
activities

2 hrs/CWS

hrs_ale_consult_op

aa) Implement other public
education activities

2.7 to 1,039.2 hrs/CWS;
$38.82 to $297,956/CWS

hrs_ale_other_op;
cost_ale_other

Acronyms: ALE = action level exceedance; CWS = community water system; NTNCWS = non-transient non-

community water system; PWS = public water system.

Sources:

t), u): "Public Education lnputs_CWS_Final.xlsx"; "Public Education lnputs_NTNCWS_Final.xlsx."
v)-aa): "Public Education lnputs_CWS_Final.xlsx."

t) Update mandatory language for lead ALE public education and submit to the State for review
(hrs_pe_al_devel_op). Under the final LCRI, CWSs and NTNCWSs with lead ALEs must update their
mandatory health effects language and include additional steps to reduce lead exposure from
drinking water such as the use of filters. The language must include an explanation that lead levels
may vary and therefore lead exposure is possible even when tap sampling results do not detect lead
at one point in time. For systems with lead, GRR, or unknown service lines, the materials must
include SLR and service line material identification opportunities, how to obtain a copy or view the
service line inventory and replacement plan, programs to assist with SLR, and the systems'
responsibility to replace their portion of the lead or GRR service line when the property owner
notifies them that the private-side portion is being replaced. The public education materials must
also include instructions for consumers to notify the water system if they think the material
classification is incorrect. The EPA assumed a one-time burden of 7 hours to update these materials.
This burden estimate is based on the hours to prepare additional brochure language from Exhibit
33a of the 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR
(Renewal) (USEPA, 2022a).

Final LCRI Economic Analysis

4-252

October 2024


-------
u) Deliver lead ALE public education materials to all customers (hrs_distr_edu_op,

cost_pe_lcr_delivery, hrs_ntncws_distr_edu_op, cost_ntncws_pe_lcr_delivery). The final LCRI
retains the prior public education requirements for CWSs to distribute public education to all
households they serve (see Exhibit 4-112 for the estimated number of households (numb_hh)). The
EPA estimates CWSs would require 15 minutes per 100 copies (0.0025 hours/household) to
distribute public education materials (hrs_distr_edu_op). The estimate is based on assumptions for
production labor used in the Economic and Supporting Analyses: Short-Term Regulatory Changes to
the Lead and Copper Rule, Exhibit 17 (USEPA, 2007). CWSs will also incur the following material cost
associated with delivery of annual lead PE in the water bill.138 The EPA assumed 50 percent of
systems will include lead public education in the water bill and only incur an additional cost for
paper ($0,019) and ink ($0.06). The other 50 percent will mail a pamphlet and incur costs for paper
($0,019), ink ($0.06), an envelope ($0,092), and postage ($0.55). Systems serving more than 500
people will deliver more than 200 pamphlets and qualify for bulk-rate postage ($0,299). Thus, the
average annual delivery cost per household (cost_pe_lcr_delivery) is $0.40 for. The cost formula is
shown below for:

•	CWSs serving < 500 people = ($0.019+0.06)*50%)) + (($0,019 + $0.06 + $0,092 + 0.55) *
50%) = $0.40.

•	CWSs serving > 500 people = ($0.019+0.06)*50%)) + (($0,019 + $0.06 + $0,092 + 0.299) *
50%) = $0.27.

The total burden per CWS is based on an estimated number of households, which is based on the
system's served population. This approach may miss bill-paying customers that reside outside the
water system service area and would underestimate the burden and cost.

The final LCRI also retains the prior public education requirements for NTNCWSs following a lead
ALE. NTNCWSs are subject to different requirements for public education delivery than a CWS and
can deliver material via email and public posting. The EPA assumed that NTNCWSs will deliver
materials via email and post materials publicly with an estimated burden of 0.5 hours to
develop/send e-mail and an additional 0.5 hours to post the materials, for a total of 1 hour
(hrs_ntncws_distr_edu_op). NTNCWSs will also incur a cost for public education posted materials
(cost_ntncws_pe_lcr_delivery) that will include paper costs of $0,019 and ink of $0.06 based on
costs from 3 vendors (see file "General Cost Model lnputs_Final.xlsx" for more detail).

v) Post notice to website (hrs_web_op). Each CWS serving more than 50,000 people with a lead ALE
must post public education materials on their website at an estimated annual burden of 0.5 hours
per system. This estimate is based on the burden to post a notice on a website used in the Economic
and Supporting Analyses: Short-Term Regulatory Changes to the Lead and Copper Rule (page 57)
(USEPA, 2007). Systems serving 50,000 or fewer people are not subject to this requirement.

w) Prepare press release (hrs_pr_op, cost_pr). The EPA assumed systems serving 3,300 or fewer will
not prepare a press release because they deliver notices to all households individually as allowed
under the rule. Systems serving more than 3,300 are estimated to require 5 hours per public

138 CWSs are also required to include a brief lead informational statement on or in each water bill. The EPA
assumed systems would incur negligible burden and no costs for this activity.

Final LCRI Economic Analysis

4-253

October 2024


-------
education event (two per year) for preparation and delivery to a total of 8 newspapers, radio
stations, or TV stations for a total burden of 10 hours. The EPA assumed systems will not incur any
material costs associated with these activities. For press releases (cost_pr), the EPA assumed that
newspapers, radio stations, or TV stations will run the press release materials as a public service
announcement (PSA), free of charge. In addition, systems are assumed to provide the press release
and certification electronically. For additional information, see the 2022 Disinfectants/Disinfection
Byproducts, Chemicaland Radionuclides Rules ICR (Renewal) (Exhibit 31 (Labor Hours per PSA))
(USEPA, 2022a).

x) Contact public health agencies to obtain additional organizations and update recipient list
(hrs_ha_op). CWSs must contact local health agencies to obtain a list of additional organizations
that serve at-risk populations. The estimated number of health agencies (numb_ha) is provided in
Exhibit 4-113. The EPA assumed that systems will elect to contact the public health agency by phone
or in-person and spend on average 30 minutes (0.5 hours) per health agency to obtain a list of
additional community-based organizations that should be contacted in response to a lead ALE
(hrs_ha_op). The EPA assumed this contact would result in additional burden to update the list of
organizations for systems serving more than 3,300 people. Specifically:

•	Systems serving 3,301 to 100,000 people would incur an additional annual burden
requirement of 1 hour per system to update the list of organizations for a total annual
burden of 1.5 hours.

•	Systems serving more than 100,000 people would incur an additional burden of 2 hours per
system to update the list of organizations for a total of 2.5 hours.

These estimates are based on Appendix H-3 in the Economic and Supporting Analyses: Short-Term
Regulatory Changes to the Lead and Copper Rule (USEPA, 2007).

y) Notify public health agencies and other organizations (hrs_distr_agencies_pe_op,

cost_pe_lead_ale). CWSs must provide public education materials to facilities that include but are
not limited to local public health agencies, schools, child care facilities, and medical providers that
offer services to pregnant people, children, and infants to better reach these at-risk populations and
their caregivers (numb_lcr_other_org). This input is provided in Exhibit 4-121.

•	Estimated hours to conduct outreach per organization. The EPA assumed systems would
require 15 minutes per 100 copies (0.0025 hours/organization) to produce the outreach for
public health agencies and other organizations in response to a lead ALE

(hrs_distr_agencies_pe_op). This estimate is based on assumptions for production labor used in
the Economic and Supporting Analyses: Short-Term Regulatory Changes to the Lead and Copper
Rule (Exhibit 17) (USEPA, 2007).

•	Notify public health agencies and other organizations. The EPA assumed CWSs will send one
pamphlet per health agency and other organizations and ask these organization to make copies.
The EPA assumed the information is delivered via certified mail at an estimated cost of $5.97
per organization. This total unit cost includes paper ($0,019), ink ($0.06), envelope ($0,092), and
certified mail ($5.80) (cost_pe_lead_ale).

Final LCRI Economic Analysis

4-254

October 2024


-------
Exhibit 4-121: Number of Local Health Agencies, Schools, Child Care Facilities, and Targeted
Medical Providers Proportionally Distributed by CWS Population Served

System Size
(Population
Served)

ft of
Systems

Population
Served

Number of
Agencies
Proportionally
Distributed

Number of
Agencies per
System

Number of
Agencies per
System (Rounded

Up to Nearest
Whole Number)

A

B

C

<

u"

II

Q

E

n umb_lcr_ other_ org

<100

11,732

708,236

2,173

0.2

1

101-500

15,084

3,830,126

11,752

0.8

1

501-1,000

5,330

3,931,488

12,063

2.3

3

1,001-3,300

7,967

15,218,647

46,695

5.9

6

3,301-10,000

5,026

29,565,710

90,716

18.0

19

10,001-50,000

3,374

74,162,674

227,553

67.4

68

50,001-100,000

571

39,629,417

121,595

213.0

213

100,001-1,000,000

421

99,359,362

304,864

724.1

725

>1,000,000

24

46,638,891

143,102

5,962.6

5,963

Total

49,529

313,044,551

960,513





Source: "Public Education lnputs_CWS_Final.xlsx," worksheet, "Pb ALE_Recipients," Table 2a.

Notes:

General: CWSs must provide lead public education materials to facilities that include but are not limited to local
public health agencies, schools, child care facilities, and medical providers that offer services to pregnant people,
children, and infants to better reach these at-risk populations and their caregivers. The estimates do not explicitly
include all groups that are required to receive public education, i.e., the Special Supplemental Nutrition Program
for Women, Infants, and Children (WIC) and Head Start, and public and private hospitals and clinics, family
planning centers, and local welfare agencies. Note the omission of some of the organizations that receive public
education will not impact the incremental costs of the final LCRI because this requirement is the same under the
pre-2021 LCR, 2021 LCRR, and final LCRI.

A&B: From SDWIS/Fed, current through December 31, 2020.

C: Assumes the number of local health agencies and community-based organizations is proportionally distributed
across the size categories.

z) Consult with State on other public education activities (hrs_ale_consult_op). CWSs will consult with
their State on other required public education activities conducted in response to a lead ALE and will
incur a burden of 2 hours per CWS. This assumption is based on the estimate to consult with the
State on public education activities used in the Economic and Supporting Analyses: Short-Term
Regulatory Changes to the Lead and Copper Rule, page 60 (USEPA, 2007).

aa) Implement other public education activities (hrs_ale_other_op, cost_ale_other). CWSs with a lead
ALE will also incur burden to implement other public education activities that use other delivery
methods to inform consumers about the health effects of lead and ways to mitigate their exposure.
Specifically, CWSs that exceed the lead ALE and serve more than 3,300 people must conduct

Final LCRI Economic Analysis

4-255

October 2024


-------
additional annual public education activities from a list specified in the rule in consultation with the
State until the system no longer has a lead ALE. CWSs serving 3,300 or fewer people must select one
activity. These activities and the EPA's burden assumptions are as follows:

•	Public Service Announcements (PSAs): Systems will require 10 hours to prepare and e-mail a
notification to newspapers and radio and TV stations.

•	Paid Ads: Systems will require 0.5 hours to coordinate paid advertisements, which will be based
on the information developed for the PSA. Thus, the EPA assumes minimal development
burden.

•	Public Display: Systems will post notices at local grocery stores, laundromats, or similar
establishments. Systems serving 500 or fewer people would need 5 such postings, and systems
serving between 501 and 10,000 people need 20 postings. Those serving 10,001 to 50,000
people need 100 postings, 50,001 to 100,000 need 200 postings, and 100,001 to 1,000,000 need
500 postings. It is assumed that it will take a system 1 hour to complete 5 postings.

•	Email Notification: Systems will have a preexisting list of customer e-mail addresses and incur a
burden of 1 hour.

•	Public Meetings: Systems will incur burden for pre-meeting logistical arrangements, preparation
of presentation/talking points, attending meeting, post-meeting activities (e.g., develop and
post meeting minutes). Estimates for each of these meeting components and the total
estimated burden are included in Exhibit 4-122, Column E.

•	Material to Multifamily homes and institutions: Systems will require 0.0025 hours/household,
which is 15 minutes per 100 copies. This is multiplied by the average number of households per
CWS (numb_hh) and the percentage of total occupied housing units that are multi-family units
(13.1 percent).

The EPA assumed that each activity has an equal likelihood of being selected and thus, the average
burden is used for hrs_ale_other_op. Burden estimates for systems serving more than 3,300 are
multiplied by three because the rule requires these systems to conduct three activities whereas
CWSs serving 3,300 or fewer people are required to conduct one activity. Burden estimates are
included in Exhibit 4-123.

Exhibit 4-122: System Burden for Public Meetings

System Size
(Population
Served)

Pre-meeting

logistical
arrangements

Preparation of
presentation/
talking points

Attend
meeting

Post meeting,
including notes

Total

A

B

C

D

E = A:D

<3,300

2

2

2

0

6

3,301-10,000

6

14

6

0

26

10,001-50,000

10

38

12

8

68

50,001-100,000

20

50

12

6

88

Final LCRI Economic Analysis

4-256

October 2024


-------
System Size
(Population
Served)

Pre-meeting

logistical
arrangements

Preparation of
presentation/
talking points

Attend
meeting

Post meeting,
including notes

Total

A

B

C

D

E = A:D

>100,000

20

50

30

28

128

Source: "Public Education lnputs_CWS_Final.xlsx."

Notes:

The EPA based estimates on the Economic and Supporting Analyses: Short-Term Regulatory Changes to the Lead
and Copper Rule (USEPA, 2007), Appendix Exhibits H-14 through H-17. This EA did not provide estimates for
systems serving <3,300 people so the EPA adjusted the burden used for systems serving 3,301 to 10,000 people
downward to develop the burden estimates for system serving < 3,300 people. See notes A - D for additional
detail.

A: Includes burden to select date, research and select site, negotiate with site for use, publicize meeting, set up
room including electronics (microphones, sound system, and presentation).

B: Includes burden to prepare a 30-minute presentation (30-50 slides) including consultation with health experts
and technical personnel as necessary, to receive feedback from management, and to practice presentation.
C: Estimate is based on DC Water (formerly called DC WASA): 1.5 hour open house, 1 hour presentation/Q&A, 15
minutes before and after, for a total of 3 hours, attended by two system representatives.

D: Includes burden to prepare and review meeting transcript or notes and follow up with attendees as appropriate.

Exhibit 4-123: System Burden for Additional Public Education Activities after a Lead ALE

System Size
(Population
Served)

PSA

Paid
Ads

Public
Display

Email
Notification

Public
Meetings

Delivery to
all

Households

Material to
Multifamily
homes and
institutions

Average Burden
for Additional
Activities (per
system)

A

B

C

D

E

F

G

H

hrs_ale_other_op

<100

10

0.5

1

1

6

0.1

0.01

2.65

101-500

10

0.5

1

1

6

0.3

0.03

2.68

501-1,000

10

0.5

4

1

6

1

0.10

3.19

1,001-3,300

10

0.5

4

1

6

2

0.26

3.38

3,301-
10,000

10

0.5

4

1

26

6

0.80

20.62

10,001-
50,000

10

0.5

20

1

68

22

3.00

53.24

50,001-
100,000

10

0.5

40

1

88

69

9.46

93.23

100,001-
1,000,000

10

0.5

100

1

128

233

32.18

216.38

> 1,000,000

10

0.5

100

1

128

1,920

264.99

1,039.17

Sources/Assumptions:

Notes:

* General: The targeted customer contact is listed in the rule but was assumed not to be selected because those
subsets of the population (e.g., pregnant women and children) are contacted through other public education
recipients, such as doctors, schools, and child care facilities.

Final LCRI Economic Analysis

4-257

October 2024


-------
A: Based on the 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR (Renewal)
(Exhibit 31 (Labor Hours per PSA)) (USEPA, 2022a).

B: The EPA assumed a half hour to develop ad material with assistance from news outlet.

C: The EPA assumed systems will provide an increasingly larger number of postings per systems size and each

would require one hour per five postings.

D: Based on the Economic and Supporting Analyses: Short-Term Regulatory Changes to the Lead and Copper Rule
(Appendix Exhibit H-12) (USEPA, 2007).

E: See Exhibit 4-122.

F: Estimate is based on assumptions for production labor used in the Economic and Supporting Analyses: Short-
Term Regulatory Changes to the Lead and Copper Rule (Exhibit 17) (USEPA, 2007).

G: Includes multi-family unit burden and not institutions. The USEPA (2008a) CWS public education guidance does
not discuss distributing information to institutions. Also, other public education requirements already include
distribution to several organizations (e.g., WIC, hospitals, medical clinics, pediatricians, family planning centers,
etc.). Multi-family units (in buildings with 10 or more units) represent 13.8 percent of the total occupied housing
units according to the 2019 American Community Survey (ACS) from the Census Bureau. (U.S. Census Bureau,
2019). 2019 data were used rather than 2020 data because the Census only released experimental estimates for
the 2020 ACS due to COVID that impacted their data collection efforts.

These other public education activities have associated non-labor costs:

•	Paid Ads: The EPA obtained estimates to run an ad from nine newspapers - three small, three
medium, and three large based on circulation size, as shown in Exhibit 4-124. The last column
provides the average cost based on circulation size. The EPA assumed that smaller systems will
use small, local newspaper, whereas larger systems will use newspapers with wider circulation.

•	Public Meetings: Includes the cost of a single-page handout ($0,079 = $0,019 for paper + $0.06
for ink) multiplied by the average number of households per system.

•	Delivery to all households: Includes the cost of postage ($0.55 for < 200 mailings) or ($0,299 for
bulk rate of > 200 mailing), paper ($0,019), ink ($0.06) and envelopes ($0,092). These costs are
multiplied by the average number of households per CWS.

•	Material to Multifamily homes and institutions: Includes postage ($0.55), paper ($0,019), ink
($0.06), and envelopes ($0,092) per multifamily home. The bulk postage rate ($0,299) is used for
systems mailing more than 200 pieces. These costs are multiplied by the average number of
households per CWS and percentage of total occupied housing units that are multi-family units
(13.1 percent).

Exhibit 4-124: Cost for Paid Ads (2021$)

Newspaper

Circulation Size
Category

1/8 page

Average Cost per
Circulation Size
Category

Bozeman Daily Chronicle (Bozeman, MT)

Small

$215.80

$250

Wayne Independent (Honesdale, PA)

Small

$360.00

Daily Astorian (Astoria, OR)

Small

$175.00

Final LCRI Economic Analysis

4-258

October 2024


-------
Newspaper

Circulation Size
Category

1/8 page

Average Cost per
Circulation Size
Category

Milwaukee Journal Sentinel (Milwaukee, Wl)

Medium

$1,294.00

$1,888

StarTribune (Minneapolis, MN)

Medium

$1,990.00

Miami Herald (Miami, FL)

Medium

$2,380.00

Chicago Tribune (Chicago, IL)

Large

$2,197.14

$4,328

LA Times (Los Angeles, CA)

Large

$1,517.25

Washington Post (Washington, DC)

Large

$9,270.00

Source: See file "Public Education lnputs_CWS_Final.xlsx, worksheet, "Pb ALE_Other Activity Detail", Table 2 for

conversion of pricing to 1/8 page.

Notes:

1.	Costs reflect non-Sunday rates, which are higher.

2.	The EPA assumed that the newspaper develops advertisement based on base content provided by system. Costs
reflect current costs per inch for 2021. The EPA also assumed that smaller systems will use small, local newspaper,
whereas larger systems will use newspapers with wider circulation. See the Economic and Supporting Analyses:
Short-Term Regulatory Changes to the Lead and Copper Rule (USEPA, 2007).

To estimate the non-labor costs for the other required activities in response to a lead ALE
(cost_ale_other), the EPA assumed that each of the seven activities had an equal likelihood of being
selected and summed the costs for each including those with $0 and divided by seven to get an average
activity cost. The EPA multiplied the average activity cost by three for CWSs serving more than 3,300
people because the rule requires them to conduct three activities as opposed to one for CWSs serving
3,300 or fewer people. The resulting inputs for cost_ale_other are included in Exhibit 4-125.

Final LCRI Economic Analysis

4-259

October 2024


-------
Exhibit 4-125: System Non-Labor Costs for Additional Public Education Activities after a Lead ALE

System Size
(Population Served)

PSA

Paid Ads

Public
Display

Email
Notification

Public
Meetings

Delivery to all
HHs

Material to
Multifamily homes
and institutions

Average Non-Labor
Costs for Additional
Activities
(per system)

A

B

C

D

E

F

G

H

cost_ale_other

<100

$0

$250

$0

$0

$1.89

$17.20

$2.37

$38.82

101-500

$0

$250

$0

$0

$7.93

$72.36

$9.99

$48.65

501-1,000

$0

$250

$0

$0

$23.03

$210.21

$29.01

$73.22

1,001-3,300

$0

$250

$0

$0

$59.65

$544.37

$75.12

$132.77

3,301-10,000

$0

$1,888

$0

$0

$183.68

$1,676.41

$231.34

$1,705.47

10,001-50,000

$0

$1,888

$0

$0

$686.35

$6,264.05

$864.44

$4,158.36

50,001-100,000

$0

$4,328

$0

$0

$2,167.15

$19,778.64

$2,729.45

$12,430.01

100,001-1,000,000

$0

$4,328

$0

$0

$7,369.42

$67,257.61

$9,281.55

$37,815.73

> 1,000,000

$0

$4,328

$0

$0

$60,679.72

$553,798.43

$76,424.18

$297,955.91

Notes:

General: The targeted customer contact is listed in the rule but was not included because the EPA assumed that subsets of the population (e.g., pregnant

women and children) are contacted through other public education recipients, such as doctors, schools, and child care facilities.

A: The EPA assumed that systems will deliver public education materials as a public service announcement (PSA), free of charge.

B: See file "Public Education lnputs_CWS_Final.xlsx, worksheet, "Pb ALE_Other Activity Detail", Table 2 for conversion of pricing to 1/8 page.

C, D: No additional cost expected.

E: Estimate includes the cost of a single-page handout (paper = $0,019 + ink = 0.06) multiplied by the average number of households per system. See "General
Cost Model lnputs_Final.xlsx" for additional information about costs for paper.

F: Estimate includes the cost of postage ($0.55), paper ($0,019), ink ($0.06), and envelopes ($0,067) multiplied by the average number of households per
system. The bulk rate for postage ($0,299) is used when a system mails more than 200 pieces. See "General Cost Model lnputs_Final.xlsx" for additional
information.

G: See "General Cost Model lnputs_Final.xlsx" for additional information about costs for postage, paper, and envelopes. Estimate includes multi-family unit
cost and not institutions. The USEPA (2008a) CWS public education guidance does not discuss distributing information to institutions. Also, other public
education requirements already include distribution to several organizations (e.g., WIC, hospitals, medical clinics, pediatricians, family planning centers, etc.).
Multi-family units (in buildings with 10 or more units) represent 13.8 percent of the total occupied housing units according to the 2019 ACS from the Census
Bureau. 2019 data were used rather than 2020 data because the Census only released experimental estimates for the 2020 ACS due to COVID that impacted
their data collection efforts.

Final LCRI Economic Analysis

4-260

October 2024


-------
Exhibit 4-126 provides details on how costs are calculated for PWS public education activities that occur
when a system has an ALE in activities t) through aa) including additional cost inputs that are required to
calculate these costs.

Exhibit 4-126: PWS Lead ALE Public Education Unit Costing Approach in SafeWater LCR by

Activity1

CWS Cost Per Activity

NTNCWS Cost Per Activity

Conditions for
Cost to Apply to
a Model PWS

Frequency
of Activity





Lead
90th -
Range

Other
Conditions



t) Update mandatory language for lead ALE public education and submit to State for review

The total hours per system
multiplied by the system labor
rate.

(hrs_pe_al_devel_op*rate_op)

Cost applies as written to
NTNCWSs.

Above
AL

All model
PWSs

One time

u) Deliver lead ALE public education materials to all customers

The number of households per
system multiplied by the total of
the hours per household times the
system labor rate, plus the
material cost.

(pws_pop/numb_hh)*( (hrs_distr_e
du_op*rate_op)+cost_pe_lcr_deliv
ery)

The hours per system multiplied by
the system labor rate, plus the
material cost.

((hrs_ntncws_distr_edu_op*rate_o
p)+cost_ntncws_pe_lcr_delivery)

Above
AL

All model
PWSs

Once a
year2

v) Post lead notice on website

The total hours per system
multiplied by the system labor
rate.

(hrs_web_op*rate_op)

Cost does not apply to NTNCWSs.

Above
AL

All model
PWSs serving
> 50,000
people

Once a
year2

w) Prepare a press release

The total hours per system
multiplied by the system labor
rate, plus the material cost.

hrs_pr_op * rate_op + cost_pr

Cost does not apply to NTNCWSs.

Above
AL

All model
PWSs serving
> 3,300 people

Twice a
year2

x) Contact public health agencies to obtain additional organizations and update recipient list

The number of health agencies
per system multiplied by the hours
per health agency and the system
labor rate.

Cost does not apply to NTNCWSs.

Above
AL

All model
PWSs

Once a
year2

numb ha*(hrs ha op*rate op)









Final LCR! Economic Analysis

4-261

October 2024


-------
CWS Cost Per Activity

NTNCWS Cost Per Activity

Conditions for
Cost to Apply to
a Model PWS

Frequency
of Activity





Lead
90th -
Range

Other
Conditions



y) Notify public health agencies and other organizations

The number of public health
agencies and other organizations
per system multiplied by the total
of the hours per agency and
organization times the system
labor rate, plus the material cost.

Cost does not apply to NTNCWSs.

Above
AL

All model
PWSs

Once a
year2

numb_lcr_other_org *
(hrs_distr_agencies_pe_op *
rate op + cost pe lead ale)









z) Consult with the State on other public education activities

The total consultation hours per
system multiplied by the system
labor rate.

(hrs ale consult op*rate op)

Cost does not apply to NTNCWSs.

Above
AL

All model
PWSs

Once a
year2

aa) Implement other public education activities

The total hours per system
multiplied by the system labor
rate, plus the material cost.

(hrs_ale_other_op*rate_op)+cost_
ale_other

Cost does not apply to NTNCWSs.

Above
AL

All model
PWSs serving
> 3,300 people

Once a
year2

Acronyms: AL = action level; ALE = action level exceedance; CCR = consumer confidence report; CWS = community
water system; LSL = lead service line; NTNCWS = non-transient non-community water system; POU = point-of-use;
PWS = public water system.

Notes:

1 The data variables in the exhibit are defined previously in Section 4.3.6.3 with the exception of:

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

•	The required number of samples (either numb_samp_customer for systems on standard monitoring or
numb_reduced_tap for systems on reduced monitoring) is based on the system's monitoring schedule.
See Section 4.3.2.1.1 for details on how the SafeWater LCR model determines monitoring schedule and
lead tap sampling requirements.

2A system can discontinue this requirement after it no longer exceeds the lead AL.

4.3.6.4 Public Education Activities in Response to Multiple Lead ALEs

The final LCRI requires water systems to develop a plan for making filters available if they have two lead
ALEs in a five-year period.139 Systems that have three or more lead ALEs in a rolling five-year period (i.e.,
have multiple lead ALEs) must make filters available and provide enhanced public education. The EPA
has developed system costs for these activities, as provided in Exhibit 4-127. The exhibit provides the
unit burden and/or cost for each activity. The assumptions used in the estimation of the unit burden
follow the exhibit. The last column provides the corresponding SafeWater LCR model data variable in

139 Under the proposed LCRI, systems were not required to develop their filter plan until they had three lead ALEs
in a five-year period.

Final LCRI Economic Analysis

4-262

October 2024


-------
red/italic font. Also refer to Chapter 3, Section 3.3.5.2 for a discussion of the likelihood a system will
have at least two lead ALEs and multiple lead ALEs.

Exhibit 4-127: PWS Public Education Burden in Response to Multiple Lead ALEs

Activity

Unit Burden and/or Cost

SafeWater LCR Data Variable

bb) Develop plan for making
filters available and submit to
the State for review

5.5 hrs per CWS and 3 hrs per
NTNCWS

hrs_temp_filter_plan_dev_ op

cc) Develop outreach materials
for systems with multiple
lead ALEs and submit to the
State for review

7 hrs per CWS and NTNCWS

hrs_devel_persist_ale_op

dd) Conduct enhanced public
education for systems with
multiple lead ALEs

CWSs

28.1 to 131.7/yr
$4.16 to $49,514/yr

hrs_deliv_persist_ale_op;
cost_ deliv_persist_ ale



NTNCWSs
1 hr/NTNCWS

hrs_n tncws_distr_pe_persist_ ale_ op

ee) Consult with State on filter
program for systems with
multiple lead ALEs

2 to 8 hrs per CWS and NTNCWS

hrs_consult_temp_pou_op

ff) Administer filter program for
systems with multiple lead
ALEs

CWS and NTNCWSs
0.167 hrs/filter

hrs_request_pou_op

gg) Make filters available due to
multiple lead ALEs

CWS and NTNCWSs
$64/filter

cost_temp_pou

Acronyms: ALE = action level exceedance; CWS = community water system; NTNCWS = non-transient non-
community water system; PWS = public water system.

Sources: "Public Education lnputs_CWS_Final.xlsx"; "Public Education lnputs_NTNCWS_Final.xlsx."

bb) Develop plan for making filters available and submit to the State for review

(hrs_temp_filter_plan_dev_op). After the second lead ALE, water systems must develop a plan that
describes which methods the system will use to make filters and replacement cartridges available
and document how the system will address barriers to customers obtaining filters. The plan is due to
the State within 60 days after of the second lead ALE. For CWSs, the EPA assumed that the plan will
be short and will be sent to the State via email. As part of planning, the EPA assumed that CWSs will
add information on how to obtain a filter to their water bill or on their website, will make filters
available at their office or other central location, and track filter distribution by adding a column to
their service line inventory. The estimated burden for this activity is 5.5 hours for all system sizes
and includes time to consult with internal staff and develop the plan (4 hrs), add instructions to the
water bill or website (1 hr based on the time estimated to update the CCR, hrs_update_ccr_op), and
modify their service line inventory to track filter distribution (assumed to be 0.5 hrs).

Final LCRI Economic Analysis

4-263

October 2024


-------
For NTNCWSs, the EPA assumes that they would provide filters at all taps as a simplifying
assumption. The EPA estimates that NTNCWSs will spend 2 hours developing a plan that describes
which methods the system will use to make filters and replacement cartridges available. The EPA
assumes that NTNCWSs will spend an additional 1 hour providing information on the filters via
email, based on hrs_ntncws_distr_pe_persist_ale_op, for a total of 3 hours for this activity for all
system sizes.

cc) Develop outreach materials for systems with multiple lead ALEs and submit to the State for review
(hrs_devel_persist_ale_op). CWSs and NTNCWSs that have at least three lead ALEs in a 5-year
period (i.e., have multiple lead ALEs) will incur a one-time burden of 7 hours to develop outreach
materials and submit a copy to their State for review. The burden estimate is based on the hours to
prepare additional brochure language from Exhibit 33a of the 2022 Disinfectants/Disinfection
Byproducts, Chemicaland Radionuclides Rules ICR (Renewal) (USEPA, 2022a). Although it is not
required, under the LCRI, for water systems with multiple lead ALEs to provide their outreach
materials for review, the EPA assumed systems would elect to provide these materials to their State.

dd) Conduct enhanced public education for systems with multiple lead ALEs (hrs_deliv_persist_ale_op,
cost_deliv_persist_ale, hrs_ntncws_distr_pe_persist_ale_op). CWSs with multiple lead ALEs must
conduct at least one enhanced community outreach activity every six months until they no longer
exceed the lead ALE three times in a rolling 5-year period. These activities include: Conducting a
public meeting; participating in a community event; contacting customers by phone, text message,
email, or doorhanger; conducting a social media campaign; or conducting other State-approved
methods.

To estimate the burden to CWSs with multiple lead ALEs to deliver enhanced outreach materials, the
EPA estimated the per system burden to conduct each of the seven specified activities in the rule
(excluding other State-approved methods). These estimates are provided in Exhibit 4-128. The EPA
assumed CWSs serving 3,300 or fewer people would have an equal likelihood of picking each activity
and averaged them to estimate the burden for these systems to deliver enhanced outreach. The EPA
assumed CWSs serving more than 3,300 people would elect not to contact customers using door
hangers due to the burden and cost to conduct this activity and used the average of the other six
delivery methods shown in Exhibit 4-128 to estimate the burden to deliver enhanced outreach. Also
refer to the notes below the exhibit for additional EPA assumptions. The resulting per system
burden is provided in Exhibit 4-129.

Exhibit 4-128: Community Water System Burden for Enhanced Outreach Following a
Minimum of 3 Lead Action Level Exceedances in a 5-Year Period (per system per 6-month

period)

Size Category
(Population
Served)

Public
Meeting

Community
Event

Contacting
customers
by phone

Contacting
customers
by text
message

Contacting
customers
by email

Contacting
customers
using door
hangers

Social
Media
Campaign



A

B

C

D

E

F

G

<100

6.0

6.0

6

1

1

2.4

76.0

Final LCRI Economic Analysis

4-264

October 2024


-------
Size Category
(Population
Served)

Public
Meeting

Community
Event

Contacting
customers
by phone

Contacting
customers
by text
message

Contacting
customers
by email

Contacting
customers
using door
hangers

Social
Media
Campaign

A

B

C

D

E

F

G

101-500

6.0

6.0

25

1

1

8.8

76.0

501-1,000

6.0

6.0

73

1

1

25.1

76.0

1,001-3,300

6.0

6.0

1

1

1

64.5

76.0

3,301-10,000

26.0

26.0

1

1

1

N/A

76.0

10,001-
50,000

68.0

68.0

1

1

1

N/A

76.0

50,001-
100,000

88.0

88.0

1

1

1

N/A

136

100,001-
1,000,000

128.0

128.0

1

1

1

N/A

136

>1,000,000

128.0

128.0

1

1

1

N/A

136

Source: "Public Education lnputs_CWS_Final.xlsx", worksheet, "Multiple ALEs."

Notes:

General: Assumes the EPA will have developed: 1) Key messaging document; 2) Sample social media posts for
Facebook and Twitter; 3) Social media graphics; and 4) Guidance with social media best practices.

A: From "Public Education lnputs_CWS_Final.xlsx," worksheet, "Pb ALE_Other Activity Detail", Column E of Table 1:
System Burden for Public Meetings.

B: The EPA assumed CWSs would incur the same burden to prepare for and attend a community event as a public
meeting.

C: For CWSs serving 3,300 or fewer people, the EPA assumed a phone call to a household would average 15
minutes. This burden is converted to a per system burden by multiplying the burden times the number of
households from "Public Education lnputs_CWS_Final.xlsx," worksheet, "Number of Households." For CWSs
serving more than 3,300 people, the EPA assumed they would use a robocalling service and would incur a burden
of 1 hour to coordinate with the company who is providing the service, as well as non-labor costs that are
presented in Exhibit 4-130, Column C.

D & E: The EPA assumed that systems would have a pre-existing list of customers' phone and e-mail addresses.
Estimate for email is based on the Economic and Supporting Analyses: Short-Term Regulatory Changes to the Lead
and Copper Rule (Appendix Exhibit H-12). The EPA assumed the same burden to send the information by text.
F: Burden to deliver door hangers. The EPA assumes systems would spend on average 5 minutes per household to
deliver a door hanger. This burden is converted to a per system burden by multiplying the burden times the
number of households from worksheet, "Number of Households." Based on this assumption, the EPA assumed
systems serving more than 3,300 people would not use this method because there are less burdensome
alternatives available. For CWSs serving 3,300 or fewer, the EPA added the burden to drive round trip to a
neighborhood at a speed of 25 miles per hour. The EPA estimated the one-way mileage to be 5 miles, or 10 miles
roundtrip. See file, "Estimated Driving Distances_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov. The EPA assumed one round trip for systems serving 500 or fewer people, 2 round trips for
systems serving 501 -1,000 people, and 4 round trips for those serving 1,001 - 3,300 people.

G: Refer to file, "Failure to Meet LSLR Goal_Final.xlsx," worksheet, "Social Media Campaign" for detailed
assumptions. The EPA assumed systems serving 10,000 or fewer people would incur the same burden as those
serving, 10,001 - 50,000 people.

Final LCRI Economic Analysis

4-265

October 2024


-------
Exhibit 4-129: Estimated Average Annual Burden to Conduct Enhanced Outreach for CWSs

with Multiple Lead ALEs (per system)

Size Category (Population Served)

Average Burden per 6-month Period

Average Burden per Year

(SafeWater LCR Input:
hrs_deliv_persist_ale_op)

<100

14.1

28.1

101-500

17.7

35.4

501-1,000

26.9

53.7

1,001-3,300

22.2

44.4

3,301-10,000

21.8

43.7

10,001-50,000

35.8

71.7

50,001-100,000

52.5

105.0

100,001-1,000,000

65.8

131.7

>1,000,000

65.8

131.7

Source: File "Public Education lnputs_CWS_Final.xlsx," worksheet, "Multiple Lead ALEs."

For NTNCWSs, the EPA assumed that systems would provide materials via email and that they would
have a pre-existing list of customer e-mail addresses resulting in 1 hour of estimated burden
(hrs_ntncws_distr_pe_persist_ale_op). This estimate is based on the Economic and Supporting Analyses:
Short-Term Regulatory Changes to the Lead and Copper Rule (Appendix Exhibit H-12).

To determine the cost for CWSs to deliver the enhanced public education materials, the EPA developed
corresponding costs for the activities presented in Exhibit 4-128, which are shown in Exhibit 4-130
below. The EPA applied the same approach for estimating the system cost as the system burden.
Specifically, for CWSs serving 3,300 or fewer people, the EPA used the average of all seven activities to
estimate the non-labor costs for providing enhanced outreach. For those serving more than 3,300
people, the EPA used the average non-labor costs of all activities excluding door hangers. Also refer to
the notes below the exhibit for additional EPA assumptions. The resulting per system non-labor costs to
provide enhanced outreach is provided in Exhibit 4-131.

Final LCRI Economic Analysis

4-266

October 2024


-------
Exhibit 4-130: Community Water System Non-Labor Cost for Enhanced Outreach Following a
Minimum of 3 Lead Action Level Exceedances in a 5-Year Period (per system per 6-month

period)

Size Category
(Population
Served)

Public
Meeting

Community
Event

Contacting
customers by
phone

Contacting
customers
by text
message

Contacting
customers
by email

Contacting
customers
using door
hangers

Social
Media
Campaign

A

B

C

D

E

F

G

<100

$1.89

$1.89

$0.00

$0

$0

$10.80

$0

101-500

$7.93

$7.93

$0.00

$0

$0

$27.00

$0

501-1,000

$23.03

$23.03

$0.00

$0

$0

$73.23

$0

1,001-3,300

$59.65

$59.65

$44.02

$0

$0

$182.86

$0

3,301-10,000

$183.68

$183.68

$135.55

$0

$0

N/A

$0

10,001-
50,000

$686.35

$686.35

$463.07

$0

$0

N/A

$0

50,001-
100,000

$2,167.15

$2,167.15

$1,234.45

$0

$0

N/A

$300

100,001-
1,000,000

$7,369.42

$7,369.42

$3,889.93

$0

$0

N/A

$300

>1,000,000

$60,679.72

$60,679.72

$28,189.18

$0

$0

N/A

$300

Source: "Public Education lnputs_CWS_Final.xlsx", worksheet, "Multiple Lead ALEs."

Notes:

General: Assumes the EPA will have developed: 1) Key messaging document; 2) Sample social media posts for
Facebook and Twitter; 3) Social media graphics; and 4) Guidance with social media best practices.

A: From "Public Education lnputs_CWS_Final.xlsx", worksheet, "Pb ALE_Other Activity Detail," Column E of Table 5:
System Cost for Additional Public Education Activities after a Lead ALE.

B: The EPA assumed CWSs would incur the same costs to prepare for a community event as a public meeting.
C- E: The EPA assumed CWSs serving more than 3,300 people would use a robocalling service. The average cost
from three companies (see file, "Robocall Pricing Estimates.xlsx") is multiplied by the number of households from
Public Education lnputs_CWS_Final.xlsx," worksheet "Number of Households to develop a per system cost.

F: Cost to deliver door hangers is the cost of the door hanger of $0.21 (see worksheet, Service Line Disturbances,
Table 2b) times the number of households from the worksheet, "Number of Households." EPA assumed systems
serving more than 3,300 people would not use this method because there are less costly alternatives available. For
CWSs serving 3,300 or fewer, the EPA added the cost to drive round trip to a neighborhood at a mileage
reimbursement rate of $0,575 (2020 mileage rate). The EPA estimated the one-way mileage to be 5 miles. See file,
"Estimated Driving Distances_Final.xlsx," available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov. The EPA assumed 1 round trip for systems serving 500 or fewer people, 2 round trips for
systems serving 501 -1,000 people, and 4 round trips for those serving 1,001 - 3,300 people.

G: Refer to file, "Failure to Meet LSLR Goal_Final.xlsx," worksheet, "Social Media Campaign" for detailed
assumptions.

Final LCRI Economic Analysis

4-267

October 2024


-------
Exhibit 4-131: Estimated Average Annual Non-Labor Costs to Conduct Enhanced Outreach for

CWSs with Multiple Lead ALEs (per system)

Size Category (Population Served)

Average Cost per 6-month Period

Average Annual Cost
(SafeWater LCR Input:
cost_deliv_persist_ale)

<100

$2.08

$4.16

101-500

$6.12

$12.25

501-1,000

$16.22

$32.44

1,001-3,300

$46.99

$93.98

3,301-10,000

$83.82

$167.64

10,001-50,000

$305.96

$611.92

50,001-100,000

$978.12

$1,956.25

100,001-1,000,000

$3,154.80

$6,309.59

>1,000,000

$24,974.77

$49,949.54

Source: "Public Education lnputs_CWS_Final.xlsx", worksheet, "Multiple Lead ALEs."

For NTNCWSs, the EPA assumed no non-labor costs because systems would distribute their
enhanced outreach using email.

ee) Consult with the State on filter program for systems with multiple lead ALEs

(hrs_consult_temp_pou_op). CWSs and NTNCWSs will incur burden to consult with the State on
specific requirements for its filter program. The EPA estimated systems serving 3,300 or fewer
people will require 2 hours, those serving 3,301 to 10,000 people will require 6 hours, and those
serving more than 10,000 people will require 8 hours.

ff) Administer filter program for systems with multiple lead ALEs (hrs_request_pou_op). CWSs must
also make pitcher filters available. The EPA assumes that systems will make filters available for
pickup at a central location, and estimated burden to hand out filters and track who received them
to be 0.167 hours per filter. The EPA assumes NTNCWSs will incur the same burden to track
placement of filters on taps within their water system.

gg) Make filters available to multiple lead ALEs (cost_temp_pou). The EPA estimated the cost of a
pitcher filter to be $64. See Technologies and Costs for Corrosion Control to Reduce Lead in Drinking
Water (USEPA, 2023b) for additional detail. The EPA estimates that 20 percent of customers in CWSs
would request a filter. As a simplifying assumption, the EPA estimates that NTNCWSs with multiple
lead ALEs will provide filters at all of their taps.

Exhibit 4-132 provides details on how costs are calculated for PWS public education activities that occur
when a system has multiple ALEs for activities bb) through gg) including additional cost inputs that are
required to calculate these costs.

Final LCRI Economic Analysis

4-268

October 2024


-------
Exhibit 4-132: PWS Lead Multiple ALEs Public Education Unit Costing Approach in SafeWater

LCR by Activity1,2

CWS Cost Per Activity

NTNCWS Cost Per Activity

Conditions for
Cost to Apply to
a Model PWS

Frequency
of Activity





Lead
90th -
Range

Other
Conditions2



bb) Develop plan for making filters available and submit to the State for review

The total hours per system multiplied by
the system labor rate, plus the material
cost.

(.hrs_temp_filter_plan_dev_op*rate_op)

Cost applies as written to
NTNCWS.

Above
AL

All model
PWSs with at
least two lead
ALEs

Once

cc) Develop outreach materials for systems with multiple lead ALEs and submit to State for review

The total hours per system multiplied by
the system labor rate.

(.hrs_devel_persist_ale_op*rate_op)

Cost applies as written to
NTNCWS.

Above
AL

All model
PWSs with
multiple lead
ALEs

One time

dd) Conduct enhanced public education for systems with multiple lead ALEs

The total hours per system multiplied by
the system labor rate, plus the material
cost.

(.hrs_deliv_persist_ale_op*rate_op)+cos
t_deliv_persist_ale

The hours per system
multiplied by the system labor
rate, plus the material cost.

(hrs_ntncws_distr_pe_persist_
ale_op*rate_op)

Above
AL

All model
PWSs with
multiple lead
ALEs

Once a
year

ee) Consult with State on filter program due to multiple lead ALEs

The total hours per system multiplied by
the system labor rate, plus the material
cost.

(hrs_consult_temp_pou_op*rate_op)

Cost applies as written to
NTNCWS.

Above
AL

All model
PWSs with
multiple lead
ALEs

Once

ff) Administer filter program due to multiple lead ALEs

The total hours per filter multiplied by
the system labor rate, plus the material
cost.

num_temp_pou*(hrs_request_pou_op*r
ate_op)



Above
AL

All model
PWSs with
multiple lead
ALEs

Once a
year

gg) Make filters available due to multiple lead ALEs

The total hours per filter multiplied by
the material cost.

num_temp_pou*cost_temp_pou

Cost applies as written to
NTNCWS.

Above
AL

All model
PWSs with
multiple lead
ALEs

Once a
year

Final LCR! Economic Analysis

4-269

October 2024


-------
Acronyms: AL = action level; ALE = action level exceedance; CCR = consumer confidence report; CWS = community
water system; LSL = lead service line; NTNCWS = non-transient non-community water system; POU = point-of-use;
PWS = public water system.

Notes:

1	The data variables in the exhibit are defined previously in Section 4.3.6.4 with the exception of:

•	num_temp_pou is the number of temporary filters provided by systems with multiple ALEs.

•	rate_op: PWS hourly labor rate (Chapter 3, Section 3.3.11.1).

2	The likelihood a system will have at least two or multiple lead ALEs is described in Chapter 3, Section 3.3.5.2.

4.3.6.5 Estimate of National Lead Public Education and Outreach Costs

As shown in Exhibit 4-133, the incremental estimated annualized lead public education and outreach
costs range from $197.7 million to $230.1 million in 2022 dollars at a 2 percent discount rate.

Exhibit 4-133: Estimated National Annualized Public Education Costs - 2 Percent Discount

Rate (millions of 2022 USD)

Low Estimate

High Estimate



Baseline

LCRI

Incremental

Baseline

LCRI Incremental

General Lead in Drinking Water
Public Education

$67.2

$221.1

$153.9

$65.9

$220.1

$154.2

Public Education Required after an
ALE

$2.4

$5.4

$3.0

$6.2

$9.7

$3.5

Public Education, including filter
provision, after multiple ALEs

$0.0

$40.8

$40.8

$0.0

$72.4

$72.4

Total Annual Public Education
Costs

$69.6

$267.3

$197.7

$72.1

$302.2

$230.1

Acronyms: ALE = action level exceedance; LCRI = Lead and Copper Rule Improvements; USD = United States dollar.

4.3.7 Summary of PWS Costs

This section summarizes the PWS impacts and costs of the major rule components of the final LCRI,
including:

•	PWS counts and population affected by rule components;

•	national PWS costs by system category; and

•	household costs by CWS size and source water type.

4.3.7.1 PWS counts and population affected by rule components

Exhibit 4-134 shows the number of PWSs and the population affected by each major rule requirement
under the low and high cost scenarios, for the 2021 LCRR, the final LCRI and the increment. The table
also shows the number of lead and GRR service lines that are expected to be replaced.

Final LCRI Economic Analysis

4-270

October 2024


-------
Exhibit 4-134: Estimated System Counts and Population Impacted
(Over 35 Year Period of Analysis)

Low Estimate

High Estimate



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS Count

66,946

66,946

0

66,946

66,946

0

PWSs with SLR

25,425

25,823

398

25,501

25,823

322

Population impacted by
SLR

1,577,551

21,714,621

20,137,070

2,471,476

21,720,307

19,248,831

SLR

489,820

6,885,738

6,395,918

776,687

6,885,742

6,109,055

PWSs that Install CCT

834

3,822

2,988

1,626

5,540

3,914

Population Affected by
CCT Installation

4,339,763

8,606,323

4,266,560

9,746,077

14,735,535

4,989,458

PWSs that Re-Optimize
CCT

1,703

2,243

540

2,858

3,566

708

Population Affected by
CCT Re-Optimization

48,328,044

51,586,612

3,258,568

81,791,662

89,692,133

7,900,471

PWSs that Conduct
DSSA of CCT

2,314

4,998

2,684

4,139

7,505

3,366

Population Affected by
DSSA of CCT

49,783,958

55,458,627

5,674,669

87,752,600

98,002,604

10,250,004

PWSs that Install POU

1,273

2,406

1,133

2,769

4,066

1,297

Population Affected by
POU Installation

263,970

250,048

-13,922

578,331

474,266

-104,065

Acronyms: CCT = corrosion control treatment; DSSA = Distribution System and Site Assessment; LCRI = Lead and
Copper Rule Improvements; POU = point-of-use; PWS = public water system; SLR = service line replacement.

Final LCRI Economic Analysis

4-271

October 2024


-------
4.3.7.2 Estimated Cost per Public Water System by System Category

Exhibit 4-135 shows the estimated annualized national PWS low cost scenario estimates for the 2021
LCRR, the final LCRI, and the incremental costs by system type, primary source water, and system size
category for CWSs. The high cost scenario estimates for CWSs are shown in Exhibit 4-136. The same
information for the low and high cost scenarios for NTNCWSs are provided in Exhibit 4-137 and Exhibit
4-138, respectively.

Final LCRI Economic Analysis

4-272

October 2024


-------
Exhibit 4-135: Estimated Annualized Incremental Cost per CWS - Low Scenario (2022 USD)

Funding

Source Water Size

Mean

10th
Percentile

25th
Percentile

50th
Percentile

75th
Percentile

90th
Percentile

Private

Ground

<100

$2,519

$1,810

$1,891

$2,035

$3,056

$3,716

Private

Ground

101 to 500

$2,854

$1,629

$1,829

$1,929

$3,458

$4,742

Private

Ground

501 to 1,000

$3,539

$1,029

$1,134

$2,210

$4,632

$6,419

Private

Ground

1,001 to 3,300

$5,085

$1,722

$1,906

$2,024

$6,556

$9,431

Private

Ground

3,301 to 10,000

$22,860

-$367

$2,170

$12,163

$37,463

$66,220

Private

Ground

10,001 to 50,000

$151,929

$2,893

$10,985

$120,197

$223,328

$368,730

Private

Ground

50,001 to 100,000

$520,919

$2,004

$16,427

$543,510

$779,446

$1,069,141

Private

Ground

100,001 to 1,000,000

$852,161

$6,292

$38,359

$578,756

$1,130,427

$2,114,840

Private

Surface

<100

$2,520

$1,817

$1,897

$2,076

$3,060

$3,683

Private

Surface

101 to 500

$3,039

$1,779

$1,843

$1,985

$3,747

$5,163

Private

Surface

501 to 1,000

$3,744

$1,476

$1,851

$1,963

$4,764

$6,925

Private

Surface

1,001 to 3,300

$6,125

$1,884

$1,916

$2,591

$7,725

$10,294

Private

Surface

3,301 to 10,000

$27,577

$1,472

$2,314

$17,828

$43,837

$74,106

Private

Surface

10,001 to 50,000

$160,481

$3,160

$11,092

$107,870

$235,375

$409,499

Private

Surface

50,001 to 100,000

$473,773

$3,567

$115,893

$479,886

$729,827

$946,664

Private

Surface

100,001 to 1,000,000

$1,446,832

$9,035

$39,505

$880,444

$1,815,169

$2,727,948

Private

Surface

>1,000,000

$2,007,743

$1,399,180

$1,400,795

$2,444,699

$2,450,057

$2,650,827

Public

Ground

<100

$2,533

$1,938

$2,002

$2,137

$2,782

$3,538

Public

Ground

101 to 500

$3,037

$1,864

$1,942

$2,049

$3,510

$4,716

Public

Ground

501 to 1,000

$3,675

$1,479

$1,571

$2,100

$4,615

$6,697

Public

Ground

1,001 to 3,300

$5,322

$1,945

$1,994

$2,454

$6,421

$9,999

Public

Ground

3,301 to 10,000

$22,967

$601

$2,296

$13,824

$37,685

$65,058

Public

Ground

10,001 to 50,000

$130,277

$2,913

$12,782

$98,611

$195,252

$310,620

Public

Ground

50,001 to 100,000

$423,629

$2,273

$102,386

$458,218

$667,395

$849,936

Final LCRI Economic Analysis

4-273

October 2024


-------
Funding

Source Water

Size

Mean

10th
Percentile

25th
Percentile

50th
Percentile

75th
Percentile

90th
Percentile

Public

Ground

100,001 to 1,000,000

$1,384,551

$10,143

$224,794

$1,052,551

$1,695,260

$2,601,212

Public

Ground

>1,000,000

$1,696,007

$418,697

$420,267

$2,659,726

$2,723,821

$2,818,505

Public

Surface

<100

$2,459

$1,955

$2,032

$2,159

$2,716

$3,245

Public

Surface

101 to 500

$3,054

$1,898

$1,954

$2,093

$3,638

$4,799

Public

Surface

501 to 1,000

$3,845

$1,689

$1,964

$2,158

$4,601

$7,142

Public

Surface

1,001 to 3,300

$5,782

$1,918

$2,007

$2,544

$6,738

$10,133

Public

Surface

3,301 to 10,000

$26,364

$1,825

$2,384

$17,362

$43,850

$71,233

Public

Surface

10,001 to 50,000

$143,832

$6,831

$12,914

$107,196

$221,667

$345,924

Public

Surface

50,001 to 100,000

$417,167

$3,633

$30,743

$437,909

$663,263

$860,490

Public

Surface

100,001 to 1,000,000

$1,487,491

$11,540

$99,596

$1,107,748

$1,916,177

$3,441,197

Public

Surface

>1,000,000

$3,296,734

$388,527

$1,212,339

$2,518,070

$3,286,896

$6,576,694

Acronyms: USD = United States dollar.

Notes: System Category rows are not included for system categories that contain zero systems.

When evaluating the economic impacts on PWSs, the EPA uses the estimated PWS cost of capital to discount future costs (not the 2 percent discount rate used
to evaluate social costs and benefit), as this best represents the actual costs of compliance that water systems would incur over time. For more information on
cost of capital, see Section 4.2.3.3.

Final LCRI Economic Analysis

4-274

October 2024


-------
Exhibit 4-136: Estimated Annualized Incremental Cost per CWS - High Scenario (2022 USD)

Funding

Source Water

Size

Mean

10th
Percentile

25th
Percentile

50th
Percentile

75th
Percentile

90th
Percentile

Private

Ground

<100

$2,427

$1,430

$1,874

$1,969

$3,041

$3,887

Private

Ground

101 to 500

$2,796

$619

$1,821

$1,901

$3,593

$5,516

Private

Ground

501 to 1,000

$3,690

$998

$1,109

$1,929

$4,951

$7,625

Private

Ground

1,001 to 3,300

$5,439

$1,151

$1,893

$2,003

$7,125

$10,610

Private

Ground

3,301 to 10,000

$29,825

-$4,227

$2,180

$14,643

$47,781

$90,472

Private

Ground

10,001 to 50,000

$210,412

$2,911

$12,575

$147,788

$301,893

$485,814

Private

Ground

50,001 to 100,000

$732,647

$2,192

$137,451

$751,544

$1,135,675

$1,532,014

Private

Ground

100,001 to 1,000,000

$1,189,604

$6,016

$27,528

$783,023

$1,573,335

$2,727,490

Private

Surface

<100

$2,431

$1,526

$1,872

$1,977

$3,104

$3,888

Private

Surface

101 to 500

$2,903

$605

$1,818

$1,903

$3,810

$5,719

Private

Surface

501 to 1,000

$3,909

$1,045

$1,829

$1,931

$4,943

$7,493

Private

Surface

1,001 to 3,300

$6,725

$1,475

$1,892

$1,994

$8,056

$12,655

Private

Surface

3,301 to 10,000

$35,803

-$1,298

$2,246

$16,836

$54,668

$104,858

Private

Surface

10,001 to 50,000

$216,071

$3,011

$10,910

$130,403

$327,595

$565,149

Private

Surface

50,001 to 100,000

$664,920

$3,542

$112,602

$681,019

$1,026,824

$1,301,522

Private

Surface

100,001 to 1,000,000

$1,970,415

$7,941

$15,737

$1,302,530

$2,477,142

$3,839,989

Private

Surface

>1,000,000

$2,643,810

$1,795,314

$1,797,301

$3,172,334

$3,188,453

$3,440,641

Public

Ground

<100

$2,519

$1,669

$1,997

$2,090

$2,814

$3,766

Public

Ground

101 to 500

$3,130

$1,174

$1,929

$2,015

$3,619

$5,714

Public

Ground

501 to 1,000

$3,975

$1,458

$1,567

$2,117

$4,857

$8,136

Public

Ground

1,001 to 3,300

$6,080

$1,719

$1,983

$2,066

$6,911

$11,206

Public

Ground

3,301 to 10,000

$28,850

-$3,072

$1,932

$14,120

$45,553

$87,586

Public

Ground

10,001 to 50,000

$181,705

$2,922

$12,501

$130,318

$277,217

$461,521

Public

Ground

50,001 to 100,000

$599,079

$2,076

$102,910

$647,708

$940,339

$1,225,129

Final LCRI Economic Analysis

4-275

October 2024


-------
Funding

Source Water

Size

Mean

10th
Percentile

25th
Percentile

50th
Percentile

75th
Percentile

90th
Percentile

Public

Ground

100,001 to 1,000,000

$1,933,944

$9,289289

$300,590

$1,446,859

$2,378,070

$3,490,476

Public

Ground

>1,000,000

$2,250,767

$432,092

$433,958

$3,511,114

$3,659,136

$3,957,839

Public

Surface

<100

$2,383

$1,602

$1,995

$2,092

$2,737

$3,349

Public

Surface

101 to 500

$3,110

$1,103

$1,927

$2,024

$3,825

$5,723

Public

Surface

501 to 1,000

$3,987

$1,112

$1,939

$2,062

$5,022

$8,069

Public

Surface

1,001 to 3,300

$6,496

$1,775

$1,983

$2,295

$7,271

$11,782

Public

Surface

3,301 to 10,000

$33,286

$442

$2,326

$19,620

$54,550

$99,968

Public

Surface

10,001 to 50,000

$198,710

$4,992

$11,799

$148,064

$301,159

$501,063

Public

Surface

50,001 to 100,000

$579,102

$3,350

$23,030

$592,831

$933,176

$1,239,996

Public

Surface

100,001 to 1,000,000

$2,072,128

$11,278

$124,059

$1,545,207

$2,597,966

$4,929,156

Public

Surface

>1,000,000

$4,442,763

$392,442

$1,541,232

$3,422,413

$4,437,768

$8,865,809

Acronyms: USD = United States dollar.

Notes: System Category rows are not included for system categories that contain zero systems.

When evaluating the economic impacts on PWSs, the EPA uses the estimated PWS cost of capital to discount future costs (not the 2 percent discount rate used
to evaluate social costs and benefit), as this best represents the actual costs of compliance that water systems would incur over time. For more information on
cost of capital, see Section 4.2.3.3.

Final LCRI Economic Analysis

4-276

October 2024


-------
Exhibit 4-137: Estimated Annualized Incremental Cost per NTNCWS - Low Scenario (2022 USD)

Funding

Source
Water

Size

Mean

10th Percentile 25th Percentile

50th
Percentile

75th
Percentile

90th
Percentile

Private

Ground

<100

$765

$6

$76

$106

$163

$1,450

Private

Ground

100 to 500

$713

-$2

$91

$116

$166

$525

Private

Ground

500 to 1,000

$585

-$444

-$439

-$381

-$174

$1,400

Private

Ground

1,000 to 3,300

$709

-$202

$40

$77

$114

$538

Private

Ground

3,300 to 10,000

$750

-$1,003

-$989

-$891

-$722

$964

Private

Ground

10,000 to 50,000

$3,803

-$1,005

-$123

$49

$214

$5,761

Private

Surface

<100

$735

$31

$79

$109

$168

$1,054

Private

Surface

100 to 500

$598

-$191

-$188

-$129

-$60

$1,471

Private

Surface

500 to 1,000

$703

-$378

-$68

$82

$135

$1,588

Private

Surface

1,000 to 3,300

$735

-$377

-$172

$61

$102

$815

Private

Surface

3,300 to 10,000

$1,318

-$872

-$290

$63

$136

$477

Private

Surface

10,000 to 50,000

$3,058

-$1,436

-$625

$23

$164

$5,221

Private

Surface

100,000 to 1,000,000

$4,760

-$2,507

-$1,371

-$615

-$397

-$337

Public

Ground

<100

$741

$36

$52

$97

$146

$1,229

Public

Ground

100 to 500

$681

$24

$70

$103

$153

$595

Public

Ground

500 to 1,000

$665

-$257

-$251

-$189

$36

$1,181

Public

Ground

1,000 to 3,300

$840

-$139

$48

$72

$105

$590

Public

Ground

3,300 to 10,000

$1,369

-$746

-$730

-$639

-$456

$2,882

Public

Ground

10,000 to 50,000

$4,088

-$974

-$18

$61

$1,886

$5,822

Public

Surface

<100

$719

$40

$52

$96

$142

$508

Public

Surface

100 to 500

$726

-$99

-$77

-$32

$18

$711

Public

Surface

500 to 1,000

$810

-$218

-$9

$72

$122

$2,021

Public

Surface

1,000 to 3,300

$1,010

-$163

$22

$53

$94

$2,047

Final LCRI Economic Analysis

4-277

October 2024


-------
Source	50th	75th	90th

Funding	... .	Size	Mean 10th Percentile 25th Percentile

Water	Percentile Percentile Percentile

Public	Surface 3,300 to 10,000	$1,827	-$652	-$163	$71	$140	$4,092

Public	Surface 10,000 to 50,000	$4,671	-$1,153	-$437	$40	$147	$5,310

Public	Surface 50,000 to 100,000	$2,390	-$1,707	-$916	-$862	-$509	-$469

Acronyms: USD = United States dollar.

Notes: System Category rows are not included for system categories that contain zero systems. Under the final LCRI, PWSs without SL with lead content benefit
from significantly reduced tap water sampling requirements. Since NTNCWSs have very few SLs with lead content, the cost of replacing these SLs is more than
offset by the savings in tap water sampling costs. Therefore, a portion of NTNCWSs will see a decrease in compliance costs under the LCRI as compared to the
2021 LCRR.

When evaluating the economic impacts on PWSs, the EPA uses the estimated PWS cost of capital to discount future costs (not the 2 percent discount rate used
to evaluate social costs and benefit), as this best represents the actual costs of compliance that water systems would incur over time. For more information on
cost of capital, see Section 4.2.3.3.

Exhibit 4-138: Estimated Annualized Incremental Cost per NTNCWS - High Scenario (2022 USD)

10th	25th

Funding

Source Water

Size

Mean

Percentile

Percentile

50th Percentile 75th

Percentile 90th Percentile

Private

Ground

<100

$1,017

-$2

$72

$104

$165

$3,749

Private

Ground

100 to 500

$1,097

-$19

$91

$112

$175

$3,770

Private

Ground

500 to 1,000

$858

-$447

-$439

-$376

-$121

$3,196

Private

Ground

1,000 to 3,300

$1,056

-$414

$30

$68

$116

$3,796

Private

Ground

3,300 to 10,000

$776

-$1,027

-$997

-$937

-$727

$4,346

Private

Ground

10,000 to 50,000

$2,590

-$1,498

-$427

$43

$237

$5,763

Private

Surface

<100

$1,051

$14

$78

$106

$171

$5,210

Private

Surface

100 to 500

$1,059

-$200

-$188

-$148

-$60

$5,444

Private

Surface

500 to 1,000

$891

-$439

-$130

$72

$128

$2,740

Private

Surface

1,000 to 3,300

$869

-$458

-$234

$61

$101

$2,501

Private

Surface

3,300 to 10,000

$1,581

-$985

-$404

$46

$135

$3,212

Final LCRI Economic Analysis

4-278

October 2024


-------
10th	25th

Funding Source Water

Size

Mean

Percentile

Percentile

50th Percentile 75th Percentile 90th Percentile

Private

Surface

10,000 to 50,000

$2,624

-$1,467

-$586

$28

$167

$5,251

Private

Surface

100,000 to 1,000,000

$7,660

-$2,727

-$1,487

-$657

-$429

$6,330

Public

Ground

<100

$1,130

$33

$52

$96

$147

$5,669

Public

Ground

100 to 500

$961

-$5

$68

$98

$154

$2,566

Public

Ground

500 to 1,000

$1,069

-$258

-$251

-$191

$46

$4,705

Public

Ground

1,000 to 3,300

$1,099

-$188

$39

$64

$97

$2,430

Public

Ground

3,300 to 10,000

$1,171

-$764

-$734

-$656

-$438

$2,617

Public

Ground

10,000 to 50,000

$4,181

-$1,159

-$248

$59

$3,203

$6,784

Public

Surface

<100

$1,079

$36

$52

$95

$146

$3,718

Public

Surface

100 to 500

$1,132

-$106

-$95

-$52

$9

$5,323

Public

Surface

500 to 1,000

$1,118

-$234

-$18

$69

$121

$4,218

Public

Surface

1,000 to 3,300

$1,247

-$263

$12

$55

$103

$4,165

Public

Surface

3,300 to 10,000

$1,727

-$692

-$233

$46

$138

$4,409

Public

Surface

10,000 to 50,000

$4,573

-$1,209

-$512

$11

$170

$10,791

Public

Surface

50,000 to 100,000

$2,379

-$1,917

-$960

-$902

-$546

$7,170

Acronyms: USD = United States dollar.

Notes: System Category rows are not included for system categories that contain zero systems. Since NTNCWSs have very few SLs with lead content, the cost
of replacing these SLs is more than offset by the savings in tap water sampling costs. Therefore, a portion of NTNCWSs will see a decrease in compliance costs
under the LCRI as compared to the 2021 LCRR.

When evaluating the economic impacts on PWSs, the EPA uses the estimated PWS cost of capital to discount future costs (not the 2 percent discount rate used
to evaluate social costs and benefit), as this best represents the actual costs of compliance that water systems would incur over time. For more information on
cost of capital, see Section 4.2.3.3.

Final LCRI Economic Analysis

4-279

October 2024


-------
4.3.7.3 Household Costs by CWS Size and Source Water Type

The SafeWater LCR model calculates the annualized total cost per household assuming that all
regulatory costs are passed on to consumers.140 The SafeWater LCR model first calculates the cost per
gallon of water produced by the model PWS:

Cost per gallonCWs = Annualized Total CWS Cost / (Average Daily FlowCWs * 365 x 1,000)

It then multiplies this cost per gallon by the average annual household consumption (in gallons) to
determine the model PWS's average annual household cost:

Average Annual Household Cost = Annual Household Consumption * Cost per gallonCWs

Exhibit 4-139 and Exhibit 4-140 show the distribution of LCRI incremental annualized costs for CWS
households by primary water source and size category for the low and high scenarios, respectively.
Note: the percentiles represent the distribution of average household costs among CWSs in a category
not the distribution of costs across all households in a CWS category. The incremental annualized per
household cost estimates presented in Exhibit 4-139 and Exhibit 4-140 may overestimate actual costs
given the potential that systems could obtain grants to offset the cost of SLR or other LCRI related
activities through the Drinking Water State Revolving Fund or other funding sources.

140 Note that the EPA assumes that all SLR costs are borne by the PWS in the analysis of the proposed LCRI.

Final LCRI Economic Analysis

October 2024

4-280


-------
Exhibit 4-139: Estimated Annualized Incremental Cost per Household - Low Scenario (2022 USD)

10th

2Sff

50th

75"

Funding

Source Water

Size

Mean

Percentile

Percentile

Percentile

Percentile

90th Percentile

Private

Ground

<100

$67.10

$28.10

$39.80

$57.80

$89.00

$117.00

Private

Ground

100 to 500

$22.50

$6.40

$11.40

$19.40

$28.10

$43.50

Private

Ground

500 to 1,000

$4.60

$1.20

$1.60

$3.00

$6.10

$8.50

Private

Ground

1,000 to 3,300

$2.70

$0.60

$0.90

$1.60

$3.60

$4.80

Private

Ground

3,300 to 10,000

$8.50

-$0.20

$0.60

$5.00

$14.50

$25.00

Private

Ground

10,000 to 50,000

$6.50

$0.10

$0.60

$6.40

$11.20

$14.30

Private

Ground

50,000 to 100,000

$7.50

$0.00

$0.30

$8.70

$11.70

$13.90

Private

Ground

100,000 to 1,000,000

$4.70

$0.00

$0.20

$3.80

$8.50

$9.70

Private

Surface

<100

$59.20

$23.40

$32.80

$50.90

$78.60

$106.40

Private

Surface

100 to 500

$17.70

$5.60

$8.40

$15.00

$22.40

$33.70

Private

Surface

500 to 1,000

$4.30

$1.50

$1.90

$2.80

$5.20

$8.70

Private

Surface

1,000 to 3,300

$2.60

$0.60

$0.70

$1.40

$3.20

$4.60

Private

Surface

3,300 to 10,000

$9.70

$0.30

$0.80

$6.40

$15.30

$26.20

Private

Surface

10,000 to 50,000

$5.50

$0.20

$0.50

$4.70

$9.60

$13.00

Private

Surface

50,000 to 100,000

$7.00

$0.00

$2.00

$7.90

$10.90

$13.80

Private

Surface

100,000 to 1,000,000

$5.70

$0.00

$0.20

$6.10

$9.70

$12.10

Private

Surface

>1,000,000

$1.90

$1.30

$1.30

$2.40

$2.40

$2.60

Public

Ground

<100

$52.20

$23.40

$31.60

$43.50

$69.50

$93.90

Public

Ground

100 to 500

$14.80

$4.90

$7.40

$11.80

$18.60

$28.10

Public

Ground

500 to 1,000

$3.70

$1.20

$1.60

$2.50

$4.40

$6.70

Public

Ground

1,000 to 3,300

$2.00

$0.50

$0.70

$1.30

$2.50

$3.50

Public

Ground

3,300 to 10,000

$7.10

$0.20

$0.60

$4.30

$11.30

$19.30

Public

Ground

10,000 to 50,000

$4.50

$0.10

$0.50

$4.00

$7.30

$10.20

Public

Ground

50,000 to 100,000

$5.20

$0.00

$0.90

$6.00

$8.20

$9.90

Final LCRI Economic Analysis

4-281

October 2024


-------
Funding

Source Water

Size

Mean

10th
Percentile

25th
Percentile

50th
Percentile

75th
Percentile

90th Percentile

Public

Ground

100,000 to 1,000,000

$5.20

$0.00

$1.20

$6.30

$8.00

$9.60

Public

Ground

>1,000,000

$0.60

$0.30

$0.30

$0.80

$0.80

$0.90

Public

Surface

<100

$54.30

$21.00

$29.70

$52.50

$72.20

$90.30

Public

Surface

100 to 500

$12.60

$4.40

$6.30

$10.20

$15.50

$23.60

Public

Surface

500 to 1,000

$3.50

$1.30

$1.60

$2.40

$4.20

$6.40

Public

Surface

1,000 to 3,300

$2.00

$0.50

$0.70

$1.20

$2.30

$3.40

Public

Surface

3,300 to 10,000

$7.90

$0.50

$0.80

$5.30

$12.90

$20.60

Public

Surface

10,000 to 50,000

$5.00

$0.20

$0.60

$4.60

$8.40

$11.10

Public

Surface

50,000 to 100,000

$5.90

$0.00

$0.40

$6.50

$9.50

$11.80

Public

Surface

100,000 to 1,000,000

$6.50

$0.10

$0.50

$7.60

$10.00

$12.10

Public

Surface

>1,000,000

$2.40

$0.30

$0.60

$2.00

$2.40

$5.00

Acronyms: USD = United States dollar.

Notes: System Category rows are not included for system categories that contain zero systems.

When evaluating the economic impacts on households, the EPA uses the estimated PWS cost of capital to discount future costs (not the 2 percent discount
rate used to evaluate social costs and benefit) because this best represents the actual costs of compliance that water systems would incur over time. For more
information on cost of capital, see Section 4.2.3.3.

Final LCRI Economic Analysis

4-282

October 2024


-------
Exhibit 4-140: Estimated Annualized Incremental Cost per Household - High Scenario (2022 USD)

Funding

Source Water

Size

Mean

Percentile

Percentile

Percentile

Percentile

Percentile

Private

Ground

<100

$64.60

$25.50

$35.50

$55.40

$87.40

$115.80

Private

Ground

100 to 500

$22.00

$4.60

$9.40

$18.70

$27.70

$46.80

Private

Ground

500 to 1,000

$4.80

$1.00

$1.50

$2.90

$6.50

$11.00

Private

Ground

1,000 to 3,300

$2.80

$0.50

$0.80

$1.50

$3.70

$5.20

Private

Ground

3,300 to 10,000

$11.20

-$1.70

$0.60

$6.20

$19.50

$34.00

Private

Ground

10,000 to 50,000

$8.90

$0.10

$0.50

$8.00

$15.40

$20.40

Private

Ground

50,000 to 100,000

$10.60

$0.00

$0.10

$12.00

$16.70

$20.10

Private

Ground

100,000 to 1,000,000

$6.50

$0.00

$0.20

$6.10

$11.70

$13.80

Private

Surface

<100

$57.20

$20.90

$29.90

$49.30

$79.90

$108.10

Private

Surface

100 to 500

$16.70

$2.60

$6.90

$13.30

$21.20

$35.10

Private

Surface

500 to 1,000

$4.40

$1.20

$1.80

$2.70

$5.60

$9.70

Private

Surface

1,000 to 3,300

$2.80

$0.50

$0.70

$1.20

$3.40

$5.20

Private

Surface

3,300 to 10,000

$12.50

-$0.50

$0.70

$7.10

$20.30

$36.60

Private

Surface

10,000 to 50,000

$7.50

$0.10

$0.60

$4.90

$13.10

$18.20

Private

Surface

50,000 to 100,000

$9.80

$0.00

$2.20

$10.90

$15.30

$19.40

Private

Surface

100,000 to 1,000,000

$8.00

$0.00

$0.10

$8.50

$14.00

$16.90

Private

Surface

Greater than 1,000,000

$2.50

$1.60

$1.60

$3.20

$3.20

$3.40

Public

Ground

<100

$51.70

$22.20

$29.40

$44.40

$71.70

$92.10

Public

Ground

100 to 500

$15.00

$4.40

$6.40

$11.50

$18.80

$30.60

Public

Ground

500 to 1,000

$4.00

$1.20

$1.50

$2.50

$4.80

$8.20

Public

Ground

1,000 to 3,300

$2.30

$0.40

$0.70

$1.20

$2.70

$4.30

Public

Ground

3,300 to 10,000

$8.70

-$0.60

$0.50

$4.40

$15.00

$26.30

Public

Ground

10,000 to 50,000

$6.20

$0.10

$0.50

$5.70

$10.50

$14.40

Public

Ground

50,000 to 100,000

$7.30

$0.00

$1.50

$8.40

$11.70

$14.20

Final LCRI Economic Analysis

4-283

October 2024


-------
10th

2Sff

50th

75«

Funding

Source Water

Size

Mean

Percentile

Percentile

Percentile

Percentile

Percentile

Public

Ground

100,000 to 1,000,000

$7.20

$0.00

$2.00

$8.60

$11.00

$13.50

Public

Ground

Greater than 1,000,000

$0.80

$0.30

$0.30

$1.10

$1.10

$1.20

Public

Surface

<100

$52.90

$19.40

$28.50

$50.30

$71.00

$90.50

Public

Surface

100 to 500

$12.60

$3.80

$5.40

$9.80

$15.80

$25.50

Public

Surface

500 to 1,000

$3.60

$1.10

$1.50

$2.30

$4.60

$7.60

Public

Surface

1,000 to 3,300

$2.20

$0.40

$0.60

$1.20

$2.60

$4.00

Public

Surface

3,300 to 10,000

$9.90

$0.10

$0.70

$5.80

$17.00

$27.90

Public

Surface

10,000 to 50,000

$7.00

$0.20

$0.60

$6.20

$11.70

$16.00

Public

Surface

50,000 to 100,000

$8.20

$0.00

$0.40

$9.00

$13.50

$16.70

Public

Surface

100,000 to 1,000,000

$9.10

$0.00

$0.60

$10.50

$14.10

$17.00

Public

Surface

Greater than 1,000,000

$3.20

$0.30

$0.80

$2.60

$3.30

$6.90

Acronyms: USD = United States dollar.

Notes: System Category rows are not included for system categories that contain zero systems. When evaluating the economic impacts on households, the EPA
uses the estimated PWS cost of capital to discount future costs (not the 2 percent discount rate used to evaluate social costs and benefit) because this best
represents the actual costs of compliance that water systems would incur over time. For more information on cost of capital, see Section 4.2.3.3.

Final LCRI Economic Analysis

4-284

October 2024


-------
4.4 Estimating State (Primacy Agency) Costs

For many of the water system activities described in Section 4.3, the 56 primacy agencies141 (note: this
document uses "States" to refer generally to primacy agencies) will incur costs in the form of burden
(i.e., hours) to provide oversight and review. The State burden is multiplied by the labor rate ($/hr), as
presented in Chapter 3, Section 3.3.11.2 to estimate labor unit costs. The remainder of this section
mirrors that of Section 4.3 and is organized as follows:

•	4.4.1: State Implementation and Administrative Costs

•	4.4.2: State Sampling Related Costs

•	4.4.3: State CCT Related Costs

•	4.4.4: State Service Line Inventory and Replacement Related Costs

•	4.4.5: State POU Related Costs

•	4.4.6: State Public Education-Related Costs

Section 4.4.7 provides a summary of State costs affected by each major requirement for low and high
cost scenarios at a 2 percent discount rate.

Exhibit 4-141 provides an overview of the rule components, subcomponents, and activities for which the
EPA estimates State costs for the final LCRI. The derivation of unit burden is provided in each referenced
subsection. At the end of each subsection, the EPA provides a summary exhibit showing the SafeWater
LCR modeling approach for each State activity, as was done in Section 4.3 for PWSs. The SafeWater LCR
model uses the information from these exhibits to calculate total annualized State cost for each activity.
See Section 4.2 for detail on the cost modeling methodology.

As noted in Section 4.3, costs for State presented in this section are LCRI costs if no previous rule were in
place. The national costs of the final LCRI, or incremental costs, are the difference between the cost of
compliance with the final LCRI and the cost of compliance with the 2021 LCRR. These costs are
presented in Exhibit 4-1 at the 2 percent discount rate.142

Also as discussed throughout Section 4.4, many of the inputs have been modified to include information
provided by ASDWA in the February 20, 2020 version of their CoSTS model (ASDWA, 2020b) and/or in its

141	The 56 primacy agencies include 49 States (excluding Wyoming), Puerto Rico, Guam, United States Virgin
Islands, American Samoa, North Mariana Islands, and Navajo Nation. For cost modeling purposes, the EPA also
included the District of Columbia (D.C.) as a primacy agency when assigning burden and costs of the rule although
some of these costs are incurred by the actual primacy agency, EPA Region 3. Note that the EPA uses the "State" to
denote "primacy agency" in this economic analysis.

142	Note that the incremental national costs of the final LCRI when compared to the pre-2021 LCR have also been
computed and are provided in Appendix C. Appendix B, Section B.9 explains how the EPA developed the cost
values for the pre-2021 LCR, which were subtracted from the final LCRI costs to produce the incremental cost of
moving from the pre-2021 LCR to the final LCRI rule requirements.

Final LCRI Economic Analysis

4-285

October 2024


-------
January 31, 2024 CoSTS model.143 Both models include estimates of burden and/or cost to implement
the rule requirements based on the final 2021 LCRR for 49 States excluding Wyoming.

Exhibit 4-141: State Cost Components, Subcomponents, and Activities Organized by Section1

Component

Subcomponents

Activities2

4.4.1: State
Implementation and
Administrative Costs

4.4.1.1: State Start-up
Implementation and
Administrative Activities

a)	Adopt rule and develop program.

b)	Modify data management systems.

c)	Provide system training and technical
assistance.

d)	Provide staff training.

e)	Review and approve small system
flexibility option.

4.4.1.2: State Annual
Implementation and
Administrative Activities

f)	Coordinate with the EPA.

g)	Provide ongoing technical assistance.

h)	Report to SDWIS/Fed.

i)	Train staff for annual administration.

4.4.2: State Sampling
Related Costs

4.4.2.1: State Lead Tap Sampling
Costs

a)	Provide templates for revised sampling
instructions and conduct review.

b)	Review updated sampling plan.

c)	Review initial lead monitoring data and
prepare systems for status under the
LCRI.

d)	Review change in tap sample locations.

e)	Review 9-year monitoring waiver
renewal.

f)	Review sample invalidation requests.

g)	Review consumer notification
certifications.

h)	Review monitoring results and 90th
percentile calculations.

4.4.2.2: State Lead WQP Sampling
Costs

i) Review lead WQP sampling data and
compliance with OWQPs.

4.4.2.3: State Copper WQP
Monitoring Costs

j) Review copper WQP sampling data and
compliance with OWQPs.

4.4.2.4: State Source Water
Monitoring Costs

k) Review source water monitoring results.

143 ASDWA developed a model to estimate the increase in costs to States to implement the final 2021 LCRR
requirements, which they provided to the EPA as part of the public comment process on the proposed rulemaking
(referred to as the ASDWA 2020 CoSTS model) (ASDWA, 2020b). ASDWA prepared a similar model for the
proposed LCRI, which they included as part of their public comments on the proposed rule (ASDWA, 2024). ASDWA
subsequently provided slight modifications to their 2024 model in an email from ASDWA on April 19, 2024 to the
EPA. The EPA uses the term "ASDWA 2024 CoSTS model" to refer to the revised version of model. Copies of the
2020 and 2024 models are available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

Final LCRI Economic Analysis

4-286

October 2024


-------
Component

Subcomponents

Activities2



4.4.2.5: State School Sampling
Costs

1) Review list of schools and child care
facilities.

m) Provide templates on school and child

care facility testing program,
n) Review school and child care facility

testing program materials,
o) Review school and child care facility
sampling results after individual
sampling events,
p) Review annual reports on school and
child care facility lead in drinking water
testing program.

4.4.3: State CCT Related
Costs

4.4.3.1: CCT Installation

a)	Review CCT study and determine type of
CCT to be installed.

b)	Set OWQPs after CCT installation.

4.4.3.2: Re-optimization

c)	Review CCT study and determine
needed OCCT adjustment.

d)	Reset OWQPs after CCT re-optimization.

4.4.3.3: State DSSA Costs

e)	Consult with system prior to any DSSA
CCT adjustments.

f)	Review report on DSSA responses.

4.4.3.4: State Lead CCT Routine
Costs

g)	Review CCT guidance and applicability
to individual PWSs.

h)	Review water quality data with PWSs
during sanitary survey.

i)	Consult on required actions in response
to source water change.

j) Consult on required actions in response
to treatment change.

4.4.4: State Service Line
Inventory and
Replacement Related
Costs

4.4.4.1: SL Inventory Costs

a)	Review connector updated LCRR initial
inventory (baseline inventory).

b)	Review annual service line inventory
updates.

c)	Review inventory validation report.

4.4.4.2: SLR Plan Review Costs

d)	Review initial SLR plan.

e)	Review information on deferred
deadline and associated replacement
rate in the SLR plan and determine
fastest feasible rate.

f)	Review annually updated SLR plan or
certification of no change.

g)	Conduct triennial review of water
system updated recommended deferred
deadline and associated replacement
rate and determine fastest feasible rate.

4.4.4.3: SLR Report Review Costs

h) Review annual SLR program report.

4.4.5: State POU Related
Costs

4.4.5.1: One-Time POU Program
Costs

a)	Review POU plan.

b)	Provide templates for POU outreach
materials.

c)	Review POU public education materials.

Final LCRI Economic Analysis

4-287

October 2024


-------
Component

Subcomponents

Activities2



4.4.5.2: Ongoing POU Program
Costs

d)	Review sample invalidation request for
POU monitoring.

e)	Review customer notification
certifications.

f)	Review annual POU program report.

4.4.6: State Public
Education-Related Costs

4.4.6.1: Consumer Notice

a)	Provide templates for consumer notice
materials.

b)	Review lead consumer notice materials.

c)	Review copy of the consumer notice and
certification.

4.4.6.2: Activities Regardless of
the Lead 90th Percentile Level

d)	Provide templates for updated CCR
language.

e)	Provide templates for local and State
health department lead outreach.

f)	Review lead outreach materials for local
and State health departments.

g)	Participate in joint communication
efforts with local and State health
departments.

h)	Provide templates for service line
disturbance outreach materials.

i)	Review public education materials for
service line disturbances.

j) Provide templates for inventory-related
outreach materials.

k) Review inventory-related outreach
materials.

1) Provide technical assistance to PWSs for
public education materials.

m) Review public education certifications.

4.4.6.3: Public Education Activities
in Response to Lead ALE

n) Provide templates for updated public
education materials for systems with a
lead ALE.

o) Review revised lead language for

systems with a lead ALE.
p) Consult with CWS on other public

education activities in response to lead
ALE.

4.4.6: State Public
Education-Related Costs
(continued)

4.4.6.4: Public Education Activities
in Response to Multiple Lead ALEs

q) Review plan for making filters available,
r) Provide templates for systems with

multiple lead ALEs.
s) Review outreach materials provided by

systems with multiple lead ALEs.
t) Consult on filter program for systems
with multiple lead ALEs.

Acronyms: ALE = action level exceedance; CCR = Consumer Confidence Report; CCT = corrosion control treatment;
CWS = community water system; DSSA = Distribution System and Site Assessment; LCRI = Lead and Copper Rule
Improvements; OCCT = optimal corrosion control treatment; OWQPs = optimal water quality parameters; POU =
point-of-use; PWS = public water system; SDWIS/Fed = Safe Drinking Water Act Information System/Federal
version; SL = service line; SLR = service line replacement; WQP = water quality parameter.

Notes:

Final LCRI Economic Analysis

4-288

October 2024


-------
1 States will also incur burden for recordkeeping activities under the final LCRI, such as retaining records of
decisions, supporting documentation, technical basis for decisions, and documentation submitted by the system.
The EPA has included burden for recordkeeping with each activity when applicable as opposed to providing
separate burden estimates.

2The EPA assigned a unique letter of identification (ID) for each activity under a given rule component. Activities
are generally organized with upfront, one-time activities first followed by ongoing activities. Note that these
activities are different than the activities identified for PWSs in Exhibit 4-6.

4.4.1 State Implementation and Administrative Costs

States will incur both one-time and annual burden to implement and administer the new requirements.
These one-time activities and associated SafeWater LCR model cost inputs are described in Sections
4.4.1.1. Ongoing activities and associated cost inputs are provided in Section 4.4.1.2.144

Note that State burden estimates for responding to specific requirements of the final LCRI (e.g., review
changes in a system's treatment, consult with systems, etc.) are presented in the sections for those
particular rule requirements.

4.4.1.1 State Start-up Implementation and Administrative Activities

The EPA estimated that States will incur burden from conducting upfront, administrative activities to
implement the final LCRI. These activities are not directly required by specific provisions of the final
LCRI; however, they are necessary for States to ensure that the provisions are properly carried out.

The EPA has identified and developed costs for five start-up implementation and administration
activities as shown in Exhibit 4-142. The last column provides the data variable used in the SafeWater
LCR model in red/italic font. Each of these costs occur during Years 1 through 5 of the 35-year period of
analysis. Additional assumptions related to each activity follow the exhibit. These burdens are based on
the ASDWA 2020 and 2024 CoSTS models (ASDWA, 2020b; 2024).

144 Also note that the EPA recognizes uncertainty in the burden estimates for State oversight and administration.
As noted throughout this chapter, the EPA based several costing inputs on the ASDWA 2020 and/or 2024 CoSTS
models. The EPA carefully reviewed both models and the proposed rule State burden estimates. In general, the
EPA opted to use the more conservative, or higher, burden estimate among these relevant sources. Through this
approach, the EPA intended to help account for additional LCRI State activities that are necessary for effective rule
implementation and oversight but not required by the final rule and, therefore, not explicitly included in the
SafeWater LCR cost model. With this strategy the agency has captured some portion of the additional State burden
not directly associated with the regulatory requirements of the final LCRI but acknowledges that additional burden
may still exist to the State programs. Further, ASDWA acknowledged in its comments to EPA on the proposed rule
that its 2024 model may underestimate the dedicate staff time needed to handle calls from consumers, the media,
and other State level staff resulting from the increase in the number of public notifications. If this is the case, then
EPA's State cost estimates may also be underestimated.

Final LCRI Economic Analysis

4-289

October 2024


-------
Exhibit 4-142: State Administration Activities and Unit Burden Estimates (Occur during Years

1 through 5)

Activity

Unit Burden

SafeWater LCR Data Variable

a) Adopt rule and develop program

640 hrs/State

hrs_adopt_ruleJs

b) Modify data management systems

740 hrs/State

hrs_modify_dsJs

c) Provide system training and technical
assistance

800 hrs/State

hrsjnitial_taJs

d) Provide staff training

196 hrs/State

hrs_trainjmpJs

e) Review and approve small system
flexibility option

6 hrs per CWSs serving
<3,300 and all NTNCWSs

hrs_smJex_optionJs

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.

Sources: ASDWA 2020 and 2024 CoSTS models (ASDWA, 2020b; 2024). Also see, "Administrative Burden and

Costs_Final.xlsx" for more detailed information on deriving the estimated burden based on ASDWA's 2020 and

2024 CoSTS models.

Note: Costs occur during the first five years of rule implementation (Years 1 through 5) (ASDWA, 2020b; 2024).

These costs apply to 49 States (excluding Wyoming), D.C, Puerto Rico, Guam, United States Virgin Islands,

American Samoa, North Mariana Islands, and Navajo Nation

a)	Adopt rule and develop program (hrs_adopt_ruleJs). The EPA assumed States would incur a
burden of 640 hours per year during Years 1 through 5 to adopt the rule that includes preparation of
a primacy package and to develop their program for the LCRI. This estimate is based on ASDWA's
projection in CoSTS, worksheet "Reg. Start-up" that State would require 3,200 hours over a 5-year
period (ASDWA, 2020b; 2024). ASDWA's estimate remained the same in their 2024 CoSTS model.

b)	Modify data management systems (hrs_modify_dsJs). The EPA assumed States will modify the
data management system in-house and incur an annual burden of 740 hours for Years 1 through 5.
This estimate is based on ASDWA's projection in CoSTS, worksheet "Reg. Start-up" that State would
require 3,700 hours over a 5-year period (ASDWA, 2020b; 2024). ASDWA's estimates remained the
same in their 2024 CoSTS model.

c)	Provide system training and technical assistance (hrs_initial_taJs). The EPA assumed States would
incur an annual burden of 800 hours per year during Years 1 through 5 to provide initial system
training and technical assistance related to the LCRI. This estimate is based on ASDWA's projection
in CoSTS, worksheet "Reg. Start-up" that State would require 4,000 hours over a 5-year period
(ASDWA, 2020b; 2024). ASDWA's estimates remained the same in their 2024 CoSTS model.

d)	Provide staff training (hrs_trainjmpJs). In CoSTS, worksheet "Reg. Start-up," ASDWA provided the
estimated burden for States to provide four types of staff training on the proposed 2021 LCRR and
proposed LCRI related to: 1) LSL inventories and replacement, 2) CCT, 3) public education, and 4)
sampling and simultaneous compliance (ASDWA, 2020b; 2024). ASDWA developed different burden
estimates for this training burden for different State sizes, as shown in Exhibit 4-143. The EPA used
the weighted average divided over a 5-year period of 196 hours as the proposed LCRI burden for
each of the States included in the SafeWater LCR model would incur during Years 1 through 5.
ASDWA's estimates remained the same in their 2024 CoSTS model.

Final LCRI Economic Analysis

4-290

October 2024


-------
Exhibit 4-143: Estimated Burden for States to Provide Staff Training during Years 1 through 5

State Size

# of States

Burden per State

Large

9

2,000

Medium

20

1,000

Small

20

500

Weighted Average



980

5-year weighted average



196

Sources: ASDWA, 2020b; 2024, worksheet Reg. Start-Up.

Note: The EPA assumed the four types of training would occur over a 5-year period.

e) Review and approve small system flexibility option (hrs_sm_flex_option Js). States will incur
burden to review and approve the compliance option recommended by CWSs serving 3,300 or
fewer and all NTNCWSs that exceed the lead AL, which is assumed to occur in Year 4 of the 35-year
analysis period. The EPA assumed a burden of 6 hours based on the burden for States to review and
track a system's selected compliance option from the ASDWA 2024 CosTS model, section "Small
System Flexibility"(ASDWA, 2024). This is an increase from 5 hours used in the proposed rule EA
(USEPA, 2023c) that was based on the ASDWA 2020 CoSTS model (ASDWA, 2020b).

4.4.1.2 State Annual Implementation and Administrative Activities

In addition to one-time, upfront activities, States will incur burden to conduct annual activities to
administer the LCRI. The EPA has identified and developed costs for four annual administration activities
as shown in Exhibit 4-144. The exhibit provides the unit burden estimate for each activity and additional
burden for new SDWIS/Fed reporting requirements under the LCRI. The last column provides the
corresponding SafeWater LCR model data variable. A more detailed explanation of how the EPA derived
the inputs are provided in text that follows the exhibit.

Exhibit 4-144: State Annual Administration Activities and Unit Burden Estimates

Activity

Unit Burden
(hours/State)

SafeWater LCR Data Variable

f) Coordinate with the EPA

1,040

hrs_coord_epaJs

g) Provide ongoing technical assistance

2,367

hrs_taJs

h) Report to SDWIS/Fed

1,560

hrs_sdwisJs

i) Train staff for annual administration

104

hrs_train_annJs

Per State Total

5,071



Sources:

f),	h), and i): "Administrative Burden and Costs_Final.xlsx." Unit burdens are based on implementation burden
estimated for the EPA's 2012, Economic Analysis for the Final Revised Total Coliform Rule, Exhibit 7.4, available in
the docket.

g):	ASDWA 2020 and 2024 CoSTS models (ASDWA, 2020b; 2024) and "Administrative Burden and Costs_Final.xlsx."

f) Coordinate with the EPA (hrs_coord_epaJs). States must coordinate with their particular EPA
Regional office to be certain that their program is consistent with federal requirements. The EPA

Final LCRI Economic Analysis

4-291

October 2024


-------
estimated an annual burden of 1,040 hours based on the Economic Analysis for the Final Revised
Total Cotiform Rule, Exhibit 7.4 (USEPA, 2012a). ASDWA agreed with this estimate in their 2024
CoSTS model (ASDWA, 2024).

g)	Provide ongoing technical assistance (hrs_taJs). The EPA determined the on-going tracking and
follow-up per system estimates provided in the ASDWA 2020 and 2024 CoSTS models (ASDWA,
2020b; 2024) for LSL inventory and replacement, tap sampling, sample site assessment, public
notification and public education, and lead testing in schools and child care facilities as follows:

1.	Determined the per system burden estimates separately for 12 categories that included
small, medium, and large CWSs with and without lead and GRR service lines and the same
categories for NTNCWSs because the estimates and rule applicability vary by system size,
system type, LSL/GRR service line status.

2.	Multiplied the per system estimate by the number of systems in each of the 12 categories
based on the system inventory information provided in Chapter 3.

3.	Summed the burden for the four system type and LSL/GRR service line status categories145
to derive a total burden by size category.

4.	Divided each burden by the 49 States used in the ASDWA CoSTS model to derive a total
burden by size category.

5.	Determined the weighted average across the size categories.

6.	Divided the burden in step 5 by five because the estimates are provided for a 5-year period.

In determining the per system burden estimates, the EPA reviewed both the ASDWA 2020 and
2024 CoSTS models. In the instances where the burdens differed between the ASDWA 2020 CoSTS
and ASDWA 2024 CoSTS models, the EPA used the higher of the two to provide a more
conservative estimate. Note that the EPA did not include ASDWA's estimates for reporting or re-
evaluation activities in the ongoing technical assistance burden because they are included in other
data variables, violations, nor compliance estimates because the EPA assumed full compliance for
cost modeling purposes. Also, the ongoing technical assistance burden does not include estimates
from the "TL" or "CCT" worksheets because they are one-time activities and the EPA has
accounted for their burden in other activities.146

h)	Report to SDWIS/Fed (hrs_sdwisJs). The EPA assumed States will require 1,000 hours to meet the
requirements of the pre-2021 LCR and an additional burden of 560 hours (or 0.25 full time
equivalents) to meet the additional requirement for the LCRI for a total annual burden of 1,560
hours. The EPA is proposing to modify the reporting requirements under the LCRI. Specifically, the
EPA is proposing to require States to report to SDWIS/Fed for each water system:

145	The EPA split CWSs and NTNCWSs into those with LSLs, GRR, and unknown service lines versus those will all
non-lead based on estimates from the 7th DWINSA. See Chapter 3, Section 3.3.4 for additional details.

146	TL refers to the burden needed to review a system's latest rounds of compliance monitoring to determine their
requirements under the 2021 LCRR. The TL does not apply under the proposed LCRI, but States will incur burden to
review the system's latest rounds of compliance monitoring using input hrs_initial_tap_revJs that is described in
Section 4.4.2.1 activity c).

Final LCRI Economic Analysis

4-292

October 2024


-------
•	The number of lead, GRR, and lead status unknown service lines and the number of lead
connectors and non-lead service lines.

•	The number and type of service lines replaced and the cumulative average replacement rate
calculation.

•	The 90th percentile lead level if it is above the AL within 15 days following the end of each
applicable tap monitoring period or within 24 hours of receiving notification of an ALE from a
water system, whichever is earlier.

•	The completion date for systems on a deferred deadline and an explanation why a faster rate is
not feasible.

The EPA based the burden estimate on the Economic Analysis for the Final Revised Total Coliform
Rule, Exhibit 7.4 (USEPA, 2012a). ASDWA agreed with this estimate in their 2024 CoSTS model
(ASDWA, 2024).

i) Train staff for annual administration (hrs_train_annJs). The EPA assumed States will incur annual
burden to continue to train staff related to annual administration. The EPA estimated an annual
burden of 104 hours based on the Economic Analysis for the Final Revised Total Coliform Rule,
Exhibit 7.4 (USEPA, 2012a). ASDWA agreed with this estimate in the 2024 CoSTS model (ASDWA,
2024).

Exhibit 4-145 provides details on how costs are calculated for State administrative and rule
implementation activities a) through i) including additional cost inputs that are required to calculate
these costs.

Exhibit 4-145: State Administration and Rule Implementation Cost Estimation in SafeWater

LCR (by Activity)1

State Cost Per Activity for
CWSs

State Cost Per Activity for
NTNCWSs

Conditions for
Cost to Apply
to a State

Frequency
of Activity





Lead
90th-
Range

Other
Conditions



a) Adopt rule and develop program

The hours per State multiplied by
the State labor rate.

(hrs_adopt_ruleJs*rateJs)

Cost applies as written to States for
NTNCWSs.

All

All States

Annually
for first 5
years

b) Modify data management systems

The hours per State multiplied by
the State labor rate.

(hrs_modify_dsJs*rateJs)

Cost applies as written to States for
NTNCWSs.

All

All States

Annually
for first 5
years

c) Provide system training and technical assistance

The hours per State multiplied by
the State labor rate.

(hrsjnitialjaJs*rateJs)

Cost applies as written to States for
NTNCWSs.

All

All States

Annually
for first 5
years

Final LCR! Economic Analysis

4-293

October 2024


-------
State Cost Per Activity for
CWSs

State Cost Per Activity for
NTNCWSs

Conditions for
Cost to Apply
to a State

Frequency
of Activity





Lead
90th-
Range

Other
Conditions



d) Provide staff training

The hours per State multiplied by
the State labor rate.

(hrs_train_impJs*rateJs)

Cost applies as written to States for
NTNCWSs.

All

All States

Annually
for first 5
years

e) Review and approve small system flexibility option2

The hours per system multiplied
by the State labor rate.

(hrs_sm_flex_optionJs*rateJs)

Cost applies as written to States for
NTNCWSs.

Above
AL

CWSs

without

CCT

serving <
10,000 and
NTNCWSs

2

One time

f) Coordinate with the EPA

The hours per State multiplied by
the State labor rate.

(hrs_coord_epaJs*rateJs)

Cost does not apply as written to
States for NTNCWSs.

All

All States

Annually

g) Provide ongoing technical assistance

The hours per State multiplied by
the State labor rate.

(hrsjaJs*rateJs)

Cost does not apply as written to
States for NTNCWSs.

All

All States

Annually

h) Report to SDWIS/Fed

The hours per State multiplied by
the State labor rate.

(hrs_sdwisJs*rateJs)

Cost does not apply as written to
States for NTNCWSs.

All

All States

Annually

i) Train staff for annual administration

The hours per State multiplied by
the State labor rate.

(hrsjrain_annJs*rateJs)

Cost does not apply as written to
States for NTNCWSs.

All

All States

Annually

Acronyms: CCT = corrosion control treatment; CWS = community water system; NTNCWS = non-transient non-

community water system.

Notes:

1	Costs are applied per State as opposed per system. The data variables in the exhibit are defined previously in
Section 4.4.1 with the exception of:

• rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

2	Applies to CWSs serving < 3,300 people and all NTNCWSs that exceed the AL.

4.4.2 State Sampling Related Costs

This section provides State unit burden related to lead tap sampling, lead WQP monitoring, copper WQP
monitoring, source water monitoring, and school testing in Sections 4.4.2.1 through 4.4.2.5,
respectively. As noted in Subsections 4.4.2.1, 4.4.2.4, and 4.4.2.5, as well as Section 4.4.5 that pertains
to the POU program and Section 4.3.4.4 that pertains to SLR, five States incur the cost of bottles,

Final LCRI Economic Analysis

4-294

October 2024


-------
analysis, and providing lead sample results to the system (ASDWA, 2020a). In addition, six States also
incur the burden and cost to update lead tap sampling instructions (see Sections 4.3.2.1.2 and 4.4.2.1).
Note that there may be additional State laboratories that incur some analytical and reporting burden
and costs in lieu of the system that would result in an underestimation of State costs.

4.4.2.1 State Lead Tap Sampling Costs

The EPA has identified and developed costs for eight State oversight and review activities associated
with lead tap sampling conducted by water systems as shown in Exhibit 4-146. The exhibit provides the
unit burden for each activity. The assumptions used in the estimation of the unit burden follow the
exhibit. The last column provides the corresponding SafeWater LCR model data variable in red/italic
font.

Exhibit 4-146: State Lead Tap Sampling Burden Estimates

Activity

Unit Burden

SafeWater LCR Data Variable

a) Provide templates for revised sampling
instructions and conduct review (one-
time)

0.75 to 1 hr/PWS

hrs_rev_sampJs1

b) Review updated sampling plan

PWSs without LSLs
2 to 4 hrs/PWS

PWSs with LSLs
4 to 10 hrs/PWS

hrs_rev_samp_planJs

c) Review initial lead monitoring data and
prepare systems for status under LCRI

2 to 4 hrs/PWS

hrs_initial_tap_revJs

d) Review change in tap sample locations2

2 hrs/CWS

hrs_chng_tapJs

e) Review 9-year monitoring waiver
renewal

0.5 hrs/PWS for those with 9-
year monitoring waiver

hrs_renew_nineJs

f) Review sample invalidation requests

2 hrs/invalidation request

hrs_sampjnvalidJs

g) Review consumer notification
certifications

0.33 to 0.5 hrs/certification

hrs_cert_custjtJs

h) Review monitoring results and 90th
percentile calculations3

PWSs without LSLs
0.5 to 2 hrs/PWS

PWSs with LSLs
0.63 to 2.5 hrs/PWS

hrs_annualjtJs

Acronyms: CWS = community water system; LCRR = Lead and Copper Rule revisions; LSL = lead service line; PWS =
public water system.

Source: "Lead Analytical Burden and Costs_Final.xlsx."

Notes:

1	As previously discussed in Section 4.3.2.1.2, in Arkansas, Louisiana, Mississippi, Missouri, North Dakota, and
South Carolina the State sends sampling instructions to the water systems and thus are assumed to incur the
burden to update the sampling instruction in lieu of the system (ASDWA, 2020a).

2	Applies to CWSs only. The EPA assumed 0 hours for NTNCWSs because they collect their own samples from
sampling locations under their control and thus, are unlikely to change sampling sites and submit documentation
to the State for review.

Final LCR! Economic Analysis

4-295

October 2024


-------
3 As previously discussed in Section 4.3.2.1.2, in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina the
State pays for the cost of bottles, analysis, and providing sample results to the system (ASDWA, 2020a). Thus, the
State will incur the burden and cost for these activities in lieu of the system. In this instance, the system burden to
provide monitoring results and 90th percentile calculations is applied to these States and hrs_annual_ltJs would be
0. Instead, they will incur the system burden of hrs_annual_lt_op (see 4.3.2.1.2, activity p)).

a)	Provide templates for revised sampling instructions and conduct review (hrs_rev_samp Js). All

CWSs and NTNCWSs must update their sampling instructions to be consistent with updated tap
sampling procedures. Systems are assumed to use an EPA template provided by the State as the
basis for this update. The EPA estimates States will incur a one-time burden per system of 0.75
hours to 1 hour to provide each water system with the template and to review the system's updated
sampling instructions. This estimate is based on responses provided by North Carolina and Indiana
of 0.25 and 0.5 hours, respectively, on the estimated time needed to update sampling instructions
based on a template. The EPA used this estimate as the hours needed to provide the templates to
the water systems. The EPA also assumed the States would not be reviewing extensive changes to
the sampling instructions and would require 0.5 hours on average for this review.

b)	Review updated sampling plan (hrs_rev_samp_plan Js). All CWSs and NTNCWSs must submit a
sampling plan to the State under the final LCRI. States will incur a one-time burden to review the
sampling plans submitted by all systems. The EPA estimated States will require 2 hours, 3 hours, and
4 hours for systems without LSLs serving 3,300 or fewer people; 3,301 to 100,000 people; and more
than 100,000 people, respectively. The EPA assumed States will incur a higher burden for reviewing
sampling plans for systems with LSLs of 4 hours, 8 hours, and 10 hours for systems serving 3,300 or
fewer people; 3,301 to 100,000 people; and more than 100,000 people, respectively. The estimates
for this input are based on the ASDWA 2020 CoSTS model, section "Tap Sampling" (ASDWA, 2020b).
The EPA did not use ASDWA's size categories for medium and large systems, which are systems
serving 3,301-50,000 and systems serving more than 50,000 people, respectively. The EPA
intentionally applied the burdens for large systems to those serving more than 100,000 people
because of the difference in the number of required samples compared to systems serving 100,000
or fewer people.147 Further, the EPA also did not use estimates from the ASDWA 2024 CoSTS model
to be more conservative because estimates in the 2020 model were lower.

c)	Review initial lead monitoring data and prepare systems for status under the LCRI
(hrsjnitial_tap_revJs). The EPA estimates States incur a one-time upfront burden per system to
review their latest two rounds of compliance monitoring data to determine the system's status
under the rule and prepare it for any new requirements. The EPA estimated States will require 2
hours, 3 hours, and 4 hours for small, medium, and large systems to review the information based
on the ASDWA 2024 CoSTS model, section "Tap Sampling" (ASDWA, 2024), which the EPA assumed
to be systems serving 3,300 or fewer people; 3,301 to 50,000 people; and more than 50,000 people,
respectively. These estimates were increased from 2.1 hours per system in the proposed LCRI EA.

147 As shown in Exhibit 4-9Exhibit 4-9, the minimum required number of samples for a system on standard
monitoring is 100 for those serving more than 100,000 people compared to 60 samples for those serving 10,001 to
100,000 people.

Final LCRI Economic Analysis

4-296

October 2024


-------
The EPA used ASDWA 2024 CoSTS model estimates because they are more conservative than the
proposed rule estimates.

d)	Review change in tap sample locations (hrs_chng_tapJs). The EPA estimates States will spend 2
hours per CWS to review reported changes in tap sample locations and other updates to sampling
plans between monitoring periods, starting in Year 5. The burden estimate is based on that provided
in the 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR (Renewal),
Exhibit 48 (Move Tap Sampling Location) (USEPA, 2022a). This estimate was doubled from the
proposed rule based on the ASDWA 2024 CoSTS model, section "Tap Sampling" (ASDWA, 2024), to
be more conservative and to include the review of other updates to sampling plans. The EPA
assumed this review to be negligible for NTNCWSs because they collect their own samples from
sampling locations under their control and thus, are unlikely to change sampling sites and submit
documentation to the State for review. Note that this assumption would underestimate burden in
those instances in which a NTNCWS had to change sampling sites (e.g., the site no longer meets the
tiering criteria because the LSL was removed). However, the EPA anticipates that once all LSLs are
removed, a NTNCWS' sampling plan would remain fairly static. The EPA also assumed that a CWS'
sampling locations are more likely to change than a NTNCWS' due a turnover in customer
participation.

e)	Review 9-year monitoring waiver renewal (hrs_renew_nineJs). The EPA estimated States will
require 0.5 hours per CWS or NTNCWS on a 9-year tap monitoring schedule to review its 9-year
monitoring waiver renewal request.148 This estimate is based on the 2022 Disinfectants/Disinfection
Byproducts, Chemicaland Radionuclides Rules ICR (Renewal), Exhibit 48 (Monitoring Waiver
Application) (USEPA, 2022a).

f)	Review sample invalidation requests (hrs_samp_invalidJs). The EPA estimated that States will
incur 2 hours per sample invalidation request from a CWS or NTNCWS based on Indiana's estimate
of 2 hours to review this request in response to a 2016 ASDWA questionnaire. As discussed in
4.3.2.1.2, activity f)f), the EPA estimates that 0.6 percent of samples will be invalidated annually
(pp_samp_invalid). ASDWA agreed with this estimate in their 2024 CoSTS model (ASDWA, 2024).

g)	Review consumer notification certifications (hrs_cert_custjt Js). The EPA estimated States will
require 0.33 hours to 0.5 hours to review each CWS or NTNCWS's certification that monitoring
results were reported to the consumer based on North Carolina and Indiana's estimates for this
review, respectively, in response to a 2016 ASDWA questionnaire. The questionnaire and each
State's responses are available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

h)	Review monitoring results and 90th percentile calculations (hrs_annualjtJs). The EPA estimated
the burden for States to review monitoring results and lead 90th percentile calculations. This
information is provided in Exhibit 4-147 for States to review information submitted by CWSs and
NTNCWSs with and without LSLs with more detailed assumptions provided in the exhibit notes. The
EPA doubled these estimates from those used for the proposed LCRI EA to be more conservative
and to better align the estimates with the higher estimates provided in the ASDWA 2024 CoSTS

148 Systems serving 3,300 or fewer can apply for 9-year waivers if they can demonstrate their entire system
including all buildings they serve are free of lead and copper. However, the EPA assumed that only those systems
serving 1,000 people or fewer will meet the waiver requirements. For the rationale, see Chapter 3, Section 3.3.7.1.

Final LCRI Economic Analysis

4-297

October 2024


-------
model, section "Tap Sampling" (ASDWA, 2024). The EPA applied the ASDWA 2024 CoSTS model
estimate for large systems (2 hrs) to large systems without LSLs, increasing the burden from 1 hour
under the proposed rule. The EPA doubled the burden for all remaining system categories and
maintained the assumption from the proposed LCRI EA that the burden for systems with LSLs would
be 25 percent higher than those without LSLs.

Exhibit 4-147: Burden to Review Lead Tap Sampling Results and 90th Percentile Level

System Size
(Population Served)

Review Lead Tap Sampling Results and 90th Percentile Calculation
(hrs/system/monitoring period)

hrs_annual_ltJs

A

B=A*1.25

No LSL

LSL

<3,300

0.5

0.63

3,301-10,000

1

1.25

10,001-100,000

1.5

1.88

> 100,000

2

2.5

Source: 2022 Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR (Renewal), Exhibit 48
(Tap Sample Calcs) (USEPA, 2022a). These estimates were doubled in the final rule based on ASDWA's 2024 CoSTS
model, section "Tap Sampling" (ASDWA, 2024).

Note: For systems with LSLs, the EPA assumed States would require an additional burden of 25 percent because
LSLs systems must also provide documentation under the final LCRI if they are unable to collect all of their samples
from LSL sites.

Exhibit 4-148 shows the SafeWater LCR model costing approach for these State lead tap sampling
activities including additional cost inputs required to calculate these costs.

Final LCRI Economic Analysis

4-298

October 2024


-------
Exhibit 4-148: State Lead Tap Sampling Unit Cost Estimation in SafeWater LCR by Activity1,2

State Cost Per Activity for CWSs

State Cost Per Activity for NTNCWSs

Conditions for Cost to Apply to a State

Frequency
of Activity





Lead 90th -
Range

Other Conditions2



a) Provide templates for revised sampling instructions and conduct review

The hours per system multiplied by
the State labor rate.

(hrs_rev_sampJs*rateJs)

Cost applies as written to States for NTNCWSs.

All

All States

One time

b) Review updated sampling plan

The hours per system multiplied by
the State labor rate.

(hrs_rev_samp_planJs*rateJs)

Cost does not apply to States for NTNCWSs.

All

All States with model PWSs
with LSLs

One time

c) Review initial lead monitoring data and prepare systems for status under the LCRI

The hours per system multiplied by
the State labor rate.

(hrs_initial_tap_revJs*rateJs)

Cost applies as written to States for NTNCWSs.

All

All States

One time

d) Review change in tap sampling locations

The hours per system multiplied by
the State labor rate.

(hrs_chng_tapJs*rateJs)

Cost does not apply to States for NTNCWSs.

At or below
AL

States with any model PWSs
not on reduced tap sampling
and not doing POU sampling

1 - (p_tap_annual +
p_tap_triennial + p_tap_nine)

Twice a year

States with any model PWSs
on reduced annual tap
sampling and not doing POU
sampling

p_tap_annual

Once a year

Final LCR! Economic Analysis

4-299

October 2024


-------
State Cost Per Activity for CWSs

State Cost Per Activity for NTNCWSs

Conditions for Cost to Apply to a State

Lead 90th - „ .... 2
„ Other Conditions2
Range

Frequency
of Activity







States with any model PWSs
on reduced triennial tap
sampling and not doing POU
sampling

p tap triennial

Every 3
years

The hours per system multiplied by
the State labor rate.

(hrs_chng_tapJs*rateJs)

Cost does not apply to States for NTNCWSs.

At or below AL

States with any model PWSs
on reduced nine year sampling
and not doing POU sampling

pjap_nine

Every 9
years

Above AL

States with any model PWSs
not doing POU sampling

Twice a year

e) Review 9-year monitoring waiver renewal3

The hours per system multiplied by
the State labor rate.

(hrs_renew_nineJs*rateJs)

Cost applies as written to States for NTNCWSs.

At or below AL3

States with any model PWSs
on reduced nine-year sampling
and not doing POU sampling

pjap_nine

Every 9
years

f) Review sample invalidation request

The number of samples
determined to be invalid multiplied
by the hours per sample per
system and the State labor rate.

(numb_samp_customer*pp_samp_
in valid) *(hrs_samp_in valid Js *rate_
js)

Cost applies as written to States for NTNCWSs.

At or below AL

States with any model PWSs
not on reduced tap sampling
and not doing POU sampling

1 - (pjap_annual +
pjapjrienniai + pjap_nine)

Twice a year

Final LCRI Economic Analysis

4-300

October 2024


-------
State Cost Per Activity for CWSs

State Cost Per Activity for NTNCWSs

Conditions for Cost to Apply to a State

Frequency
of Activity





Lead 90th -
Range

Other Conditions2









States with any model PWSs
on reduced annual tap
sampling and not doing POU
sampling

Once a year







p tap annual



The number of samples
determined to be invalid multiplied
by the hours per sample per
system and the State labor rate.

Cost applies as written to States for NTNCWSs.

At or below AL

States with any model PWSs
on reduced triennial tap
sampling and not doing POU
sampling

Every 3
years

(numb_reduced_tap*pp_samp_inv
alid)*(hrs samp invalidJs*rateJs)





pjtapjtrienniai









States with any model PWSs
on reduced nine year sampling
and not doing POU sampling

Every 9
years







p_tap_nine



The number of samples
determined to be invalid multiplied
by the hours per sample per
system and the State labor rate.

Cost applies as written to States for NTNCWSs.

Above AL

States with any model PWSs
not doing POU sampling

Twice a year

(numb_samp_customer*pp_samp_
in valid) *(hrs_samp_in valid Js *rate_
is)









Final LCRI Economic Analysis

4-301

October 2024


-------
State Cost Per Activity for CWSs

State Cost Per Activity for NTNCWSs

Conditions for Cost to Apply to a State

Frequency
of Activity





Lead 90th -
Range

Other Conditions2



g) Review consumer notification certifications







States with any model PWSs
not on reduced tap sampling
and not doing POU sampling

Twice a year







1 - (p_tap_annual +
p_tap_triennial + p_tap_nine)



The hours per system multiplied by
the State labor rate.

(hrs_cert_cust_ltJs*rateJs)



At or below
AL

States with any model PWSs
on reduced annual tap
sampling and not doing POU
sampling

p tap annual

Once a year



Cost applies as written to States for NTNCWSs.



States with any model PWSs
on reduced triennial tap
sampling and not doing POU
sampling

p tap triennial

Every 3
years







States with any model PWSs
on reduced nine year sampling
and not doing POU sampling

Every 9
years







p_tap_nine







Above AL

States with any model PWSs
not doing POU sampling

Twice a year

Final LCRI Economic Analysis

4-302

October 2024


-------
State Cost Per Activity for CWSs

State Cost Per Activity for NTNCWSs

Conditions for Cost to Apply to a State

Frequency
of Activity





Lead 90th -
Range

Other Conditions2



h) Review monitoring results and 90th percentile calculations4







States with any model PWSs
not on reduced tap sampling
and not doing POU sampling

Twice a year







1 - (p_tap_annual +
p_tap_triennial + p_tap_nine)



The hours per system multiplied by
the State labor rate.

(hrs_annual_ltJs*rateJs)



At or below AL

States with any model PWSs
on reduced annual tap
sampling and not doing POU
sampling

p tap annual

Once a year



Cost applies as written to States for NTNCWSs.



States with any model PWSs
on reduced triennial tap
sampling and not doing POU
sampling

p tap triennial

Every 3
years







States with any model PWSs
on reduced nine year sampling
and not doing POU sampling

Every 9
years







p_tap_nine



Acronyms: AL = action level; CWS = community water system; LSL = lead service line; NTNCWS = non-transient non-community water system; POU = point-of-

use; PWS = public water system.

Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of the following:

•	numb_reduced_tap\ Number of tap samples for systems on reduced lead tap monitoring that include systems with lead 90th percentile values < 10
Hg/L and which are sampling less frequently than semi-annually (Section 4.3.2.1.1).

•	numb_samp_customer: Number of tap samples for systems on standard lead tap monitoring that include some systems with 90th percentile values <
10 ng/L and all systems > 10 ng/L (Section 4.3.2.1.1).

•	pp_samp_invalid: Likelihood that a lead sample will be deemed invalid (Section 4.3.2.1.2, activity f)).

•	p_tap_annual: Likelihood a system will qualify to collect the reduced number of lead tap samples at an annual frequency (Section 4.3.2.1.1).

•	p_tap_triennial: Likelihood a system will qualify to collect the reduced number of lead tap samples at a triennial frequency (Section 4.3.2.1.1).

•	p_tap_nine: Likelihood a system will qualify to collect the reduced number of lead tap samples at a nine-year frequency (Section 4.3.2.1.1).

Final LCRI Economic Analysis

4-303

October 2024


-------
• rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

2	Does not apply to CWSs serving < 3,300 people and all NTNCWSs that have selected POU provision and maintenance as their compliance option if they
exceeded the lead AL. See Section 4.3.5 for additional detail. PWSs with lead content or unknown lines are identified using the data variables and approach
described in Chapter 3, Section 3.3.4.

3	Only systems with 90th percentile values < the AL of 10 ng/L can quality for a 9-year monitoring waiver.

4	As previously discussed in Section 4.3.2.1.2, in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina the State pays for the cost of bottles, shipping,
analysis, and providing sample results to the system (ASDWA, 2020a). Thus, the State will incur the burden and cost for these activities in lieu of the system. In
this instance, the system burden to provide monitoring results and 90th percentile calculations is applied to these States and hrs_annual_ltJs would be 0.
Instead, they will incur the system burden of hrs_annual_lt_op (see Section 4.3.2.1.2, activity p)).

Final LCRI Economic Analysis

4-304

October 2024


-------
4.4.2.2 State Lead WQP Sampling Costs

The EPA has developed State costs for the review of lead WQP monitoring data submitted by systems
serving 50,000 or fewer people with a lead ALE and all systems serving more than 10,000 people with
CCT,149 as shown in Exhibit 4-149. The exhibit provides the unit burden. The assumptions used in the
estimation of the unit burden follow the exhibit. The last column provides the corresponding SafeWater
LCR model data variable in red/italic font.

Exhibit 4-149: State Lead WQP Monitoring Burden Estimates

Activity

Unit Burden

SafeWater LCR Data
Variable

i) Review lead WQP
sampling data and
compliance with OWQPs

No CCT: 5 hrs/system/6-month monitoring period;
With CCT: 8.5 hrs/system/6-month monitoring period

hrs_wqpJs

Acronyms: CCT = corrosion control treatment; OWQP = optimal water quality parameter; WQP = water quality
parameter.

Source: "WQP Analytical Burden and Costs_Final.xlsx."

i) Review lead WQP sampling data and compliance with OWQPs (hrs_wqpJs). States will review a
system's WQP monitoring data collected from entry points and within the distribution system. The
EPA assumed States will incur a burden of 5 hours per system during each 6-month monitoring
period for systems without CCT. This estimate is based on the average of responses provided by
North Carolina and Indiana to a 2016 ASDWA survey question regarding the hours to review WQP
monitoring data of 6 and 4 hours, respectively. A copy of the questionnaire and each State's
responses are available in the docket at EPA-HQ-OW-2022-0801. The EPA assumed States will set
OWQPs for all systems with CCT and will incur an additional 3.5 hours per 6-month monitoring
period to review compliance with OWQPs for a total of 8.5 hours.

Exhibit 4-150 shows the SafeWater LCR model costing approach for this State lead WQP monitoring
activity. As shown in the exhibit, the SafeWater LCR model relies upon additional inputs, such the
likelihood a system has a certain type of CCT in place, to estimate total costs. A description of the data
variables and section where they are described in more detail are provided in the footnote to the
exhibit.

149 All systems serving more than 50,000 people except those with naturally non-corrosive water (i.e., "b3"
systems) are required to have CCT. Also, the LCRI strengthens the requirements for CWSs serving 10,001 to 50,000
with CCT to require them to continue to conduct WQP monitoring regardless of the lead or copper AL. Previously,
these systems were only required to conduct WQP monitoring during the monitoring periods in which they had a
lead or copper ALE, unless required by the State.

Final LCRI Economic Analysis

4-305

October 2024


-------
Exhibit 4-150: State Lead WQP Monitoring Cost Estimation in SafeWater LCR by Activity1

State Cost Per
Activity for
CWSs

State Cost Per Activity for
NTNCWSs

Conditions for Cost to Apply
to a State

Frequency
of Activity





Lead 90th
Range

Other Conditions



i) Review lead WQP sampling data and compliance with OWQPs







States with any PWSs
serving <50,000 and
without CCT







Above AL

States with any PWSs
serving <10,000 and having
pH adjustment in place

pbaseph



The hours per
system multiplied
by the State labor
rate.

(hrs_wqpJs*rate
J's)

Cost applies as written to States for
NTNCWSs.



States with any PWSs
serving <10,000 and having
PO4 or both PO4 and pH
adjustment in place

pbasepo4, pbasephpo4

Twice a
year





All

States with any PWSs
serving >10,000 and
having pH adjustment in
place

pbaseph









States with any PWSs
serving >10,000 and
having PCM or both PCM
and pH adjustment in place









pbasepo4, pbasephpo4



Acronyms: AL = action level; CCT = corrosion control treatment; CWS = community water system; NTNCWS = non-
transient non-community water system; OWQP = optimal water quality parameter; PO4 = orthophosphate; PWS =
public water system; WQP = water quality parameter.

Notes:

The data variables in the exhibit are defined previously in this section with the exception of:

•	pbaseph, pbasepo4, and pbasephpo4: Likelihood system has pH adjustment, orthophosphate, or pH
adjustment and orthophosphate for their CCT (Section 4.3.2.2.1).

•	rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

4.4.2.3 State Copper WQP Monitoring Costs

The EPA has developed State costs for the review of copper WQP monitoring data per 6-month
monitoring period as shown in Exhibit 4-151. The exhibit provides the unit burden. The assumptions
used in the estimation of the unit burden follow the exhibit. The last column provides the corresponding
SafeWater LCR model data variable in red/italic font. Note that the data variable is the same as for
reviewing lead WQP data.

Final LCR! Economic Analysis

4-306

October 2024


-------
Exhibit 4-151: State Copper WQP Monitoring Burden Estimates

Activity

Unit Burden

SafeWater LCR Data
Variable

j) Review copper WQP
sampling data and
compliance with OWQPs

No CCT: 5 hrs/system/6 month monitoring
period;

With CCT: 8.5 hrs/system/6 month monitoring
period

hrs_wqpJs

Acronyms: CCT = corrosion control treatment; OWQP = optimal water quality parameter; WQP = water quality
parameter.

Source: "WQP Analytica Burden and Costs_Final.xlsx."

j) Review copper WQP sampling data and compliance with OWQPs (hrs_wqpJs). As stated in Section
4.3.2.3, the SafeWater LCR models copper WQP monitoring separately from lead WQP monitoring
to avoid double counting the cost of WQP monitoring for systems experiencing a copper ALE and a
lead ALE simultaneously. The SafeWater LCR model restricts copper WQP monitoring to systems
serving 50,000 or fewer people without CCT that do not exceed the lead AL but exceed the copper
AL of 1.3 mg/L. See Exhibit 4-38 and Exhibit 4-39 in Section 4.3.2.3.1 for the likelihood a system has
a copper only ALE (p_copper_ale)150for CWSs and NTNCWSs, respectively. The unit burden for States
to review sampling data and compliance with OWQPs (hrs_wqp_js) is identical to that used for State
Lead WQP Monitoring of 5 hours and 8.5 hours per system per 6-month monitoring period for
systems without CCT and with CCT, respectively (see Section 4.4.2.2, activity i)).

Exhibit 4-152 shows the SafeWater LCR model costing approach for this State copper WQP monitoring
activity. As shown in the exhibit, the SafeWater LCR model relies upon additional inputs that include the
likelihood a system has a certain type of CCT in place and as discussed above, the likelihood a system has
a copper ALE. A description of the data variables and section where they are described in more detail
are provided in footnote 1 to the exhibit.

150 As described in Section 4.3.2.3.1, the EPA assumed all systems with CCT would have sufficient CCT such that
none would have a copper ALE. Because all systems serving 50,000 or more people have CCT (except for 16 "b3"
systems), SafeWater LCR does not assign any copper WQP costs to systems serving more than 50,000 people.

Final LCR! Economic Analysis

October 2024

4-307


-------
Exhibit 4-152: State Copper WQP Monitoring Cost Estimation in SafeWater LCR by Activity1

State Cost Per
Activity for CWSs

State Cost Per Activity for NTNCWSs

Conditions for Cost to
Apply to a State

Frequency
of Activity





Lead
90th-
Range

Other
Conditions



j) Review copper WQP sampling data and compliance with OWQPs

The hours per
system multiplied by
the State labor rate.

(hrs_wqpJs*rateJs)

Cost applies as written to States for
NTNCWSs.

At or
below AL

States with any
model PWSs
serving <50,000,
without CCT,
and having a
copper ALE

p copper ale

Twice a year

The hours per
system multiplied by
the State labor rate.

(hrs_wqpJs*rateJs)

Cost applies as written to States for
NTNCWSs.

At or
below AL

States with any
model PWSs
serving >10,000,
having pH
adjustment in
place, and
having a copper
ALE

p_copper_ale,
pbaseph

Twice a year







States with any
model PWSs
serving >10,000,
having PCM or
both PO4 and
pH adjustment
in place, and
having a copper
ALE









p_copper_ale,

pbasepo4,

pbasephpo4



Acronyms: AL = action level; ALE = action level exceedance; CCT = corrosion control treatment; CWS = community
water system; NTNCWS = non-transient non-community water system; OWQP = optimal water quality parameter;
PO4 = orthophosphate; PWS = public water system; WQP = water quality parameter.

Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of:

•	p_copper_ale: Likelihood that a system exceeds the copper AL but not the lead AL (Section 4.3.2.3.1).

•	rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

4.4.2.4 State Source Water Monitoring Costs

The EPA has developed State costs to review source water monitoring data as shown in Exhibit 4-153.
The exhibit provides the unit burden. The assumptions used in the estimation of the unit burden
following the exhibit. The last column provides the corresponding SafeWater LCR model data variable in
red/italic font.

Final LCR! Economic Analysis

4-308

October 2024


-------
Exhibit 4-153: State Source Monitoring Burden Estimates

Activity

Unit Burden

SafeWater LCR Data
Variable

k) Review source water monitoring
results

0.5 hrs/system/monitoring period in which
source water samples are collected

hrs_sourceJs

Source: " Lead Analytical Burden and Costs_Final.xlsx," worksheet, "Source_Reporting_Review."

Notes: As previously discussed in Section 4.3.2.4.2 in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina
the State pays for the cost of bottles, analysis, and providing sample results to the system (ASDWA, 2020a). Thus,
the State will incur the burden and cost for these activities in lieu of the system. In these States, because the State
is reporting the results, the burden to review the results (hrs_sourceJs) is 0. Instead, the system burden to report
the results (hrs_report_source_op) is applied to these States (see Section 4.3.2.4.2, activity hh)).

k) Review source water monitoring results (hrs_sourceJs). States will incur burden to review
source water monitoring results submitted by water systems. The EPA estimates that the State
will incur 0.5 hours per system per monitoring period in which the system conducts source
water monitoring (hrs_source_js). The burden estimate is based on the State review burden for
a source water monitoring letter in the 2022 Disinfectants/Disinfection Byproducts, Chemical,
and Radionuclides Rules ICR (Renewal), Exhibit 48 (USEPA, 2022a).

Exhibit 4-154 details how the data variables are used to estimate State source water monitoring unit
costs. As shown in the exhibit, the SafeWater LCR model relies upon additional inputs, such the
likelihood a system has changed its source. A description of the data variables and section where they
are described in more detail in the footnote 1 to the exhibit.

Exhibit 4-154: State Source Water Monitoring Cost Estimation in SafeWater LCR by Activity1

State Cost Per Activity for
CWSs

State Cost Per Activity for
NTNCWSs

Conditions for Cost
to Apply to a State

Frequency
of Activity





Lead 90th
- Range

Other
Conditions



k) Review source water monitoring results2





All

States with any
model PWSs
with a
significant
change in
source water

p_source_sig *
p source chng

3

Once a year

The hours per system multiplied
by the State labor rate.

(hrs_sourceJs*rateJs)

Cost applies as written to
States for NTNCWSs.

At or below
AL

States with any
model PWSs
with a copper
ALE

p copper ale

One time





Above AL

All States with
PWSs that
have not
conducted prior
source water
monitoring



Final LCR! Economic Analysis

4-309

October 2024


-------
Acronyms: AL = action level; ALE = action level exceedance; CWS = community water system; NTNCWS = non-

transient non-community water system; PWS = public water system.

Notes:

1	The data variables in the exhibit are defined previously in this section with the exception of:

•	p_source_chng: Likelihood a system will have a source change (Chapter 3, Section 3.3.9.1).

•	p_source_sig: Likelihood that the system will have a significant change in which it changes its primary
source, e.g., for ground water to surface water (Chapter 3, Section 3.3.9.2).

•	p_copper_ale: Likelihood that a system exceeds the copper AL but not the lead AL (Section 4.3.2.3.1).

•	rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

2	As previously discussed in Section 4.3.2.4.2 in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina the
State pays for the cost of bottles, shipping, analysis, and providing sample results to the system (ASDWA, 2020a).
Thus, the State will incur the burden and cost for these activities in lieu of the system. In these States, because the
State is reporting the results, the burden to review the results (hrs_sourceJs) is 0. Instead, the system burden to
report the results (hrs_report_source_op) is applied to these States (see Section 4.3.2.4.2, activity hh)).

3	The likelihoods of p_source_chng and p_source_sig are multiplied to determine the joint likelihood that a system
that makes a source change will be required to take additional action such as source water monitoring.

4.4.2.5 State School Sampling Costs

The EPA has developed burden for one-time State activities for oversight of CWSs' lead in drinking water
testing programs at schools and child care facilities as shown in Exhibit 4-155. The exhibit provides the
unit burden for each activity. The assumptions used in the estimation of the unit burden follow the
exhibit. The last column provides the corresponding SafeWater LCR model data variable in red/italic
font. Note that the one-time activities are assumed to occur in Year 4 and the on-going activities to
occur under the first and subsequent five-year testing cycles starting in Year 4 onward.

Exhibit 4-155: State School Sampling Burden Estimates

Activity

Unit Burden

SafeWater LCR Data
Variable

1) Review list of schools and child care
facilities (every 5 years starting in Year 4)

3 hrs/CWS

hrs_rev_school_listJs

m) Provide templates on school and child
care facility testing program (one time)

0.25 to 0.5 hrs/CWS

hrs_temp_schoolJs

n) Review school and child care facility
testing program materials (one time)

1 hr/CWS serving < 50,000;
3 hrs/CWS serving > 50,000

hrs_rev_schoolJnfoJs

o) Review school and child care facility

sampling results after individual sampling
events

6 hrs/CWS/year

hrs_sch_cc_results_reviewJs

p) Review annual reports on school and child
care facility lead in drinking water testing
program

1 hr/CWS/year

hrs_annual_report_schoolJs

Acronyms: CWS = community water system.

Source: "School_Child Care lnputs_Final.xlsx."

Note:

1 As previously discussed in Section 4.3.2.5 in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina the
State pays for the cost of bottles, shipping, and analyses associated with lead testing (ASDWA, 2020a). Thus, the
State will incur the burden and cost for these activities under the testing program at schools and child care
facilities.

Final LCR! Economic Analysis

4-310

October 2024


-------
I) Review list of schools and child care facilities (hrs_rev_school_listJs). The EPA estimated that
States will review the initial list of schools and licensed child care facilities served by each CWS and
the list updates every five years. The EPA assumed States will incur a burden of 3 hours per CWS per
review based on the ASDWA 2020 CoSTS model, section "Lead Testing in Schools" (ASDWA, 2020b).
The EPA did not use estimates from the ASDWA 2024 CoSTS model because they were less
conservative than those provided in the 2020 model.

m) Provide templates on school and child care facility testing program (hrs_temp_school Js). CWSs
must notify each school and child care facility they serve about the testing program. The EPA
assumed States would provide a template to assist CWSs in developing these materials. The EPA
assumed States would incur a similar burden to provide these templates as other outreach materials
of 0.25 to 0.5 hours per system. The burden estimates are based on North Carolina and Indiana's
response to a 2016 ASDWA survey regarding the burden to provide a sampling instruction template
of 0.25 hours and 0.5 hours per template, respectively. The questionnaire and each State's
responses are available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

n) Review school and child care facility testing program materials (hrs_rev_school_infoJs). The EPA

estimated that States will incur a one-time burden to review school and child care facility testing
program materials. The EPA assumed CWSs serving 50,000 or fewer people will rely mainly on the
template, and States will require 1 hour for review. The EPA assumed that systems serving more
than 50,000 people will adapt the template and the States will require more time (3 hours) to
review these materials. This estimate is based on the ASDWA 2024 CoSTS model (ASDWA, 2024).
This is an increase in the burden that was used in the proposed LCRI EA of 0.5 and 2 hours for
systems serving 50,000 or fewer people and more than 50,000 people, respectively.

o) Review school and child care facility sampling results after individual sampling events. U nder the
final LCRI, CWSs will be required to provide school and child care facility testing results to their State
within 30 days of receiving the analytical results (hrs_report_sch_cc_results_op). The EPA assumed
that CWSs will sample a portion of schools and child care facilities each month and would require 1
hour each month or 12 hours annually. The EPA estimated States will require half the burden (or 6
hours) per CWS per year to review the monitoring results as the burden required for a water system
to prepare and email the sampling results.

p) Review annual reports on school and child care facility lead in drinking water testing program
(hrs_annual_report_schoolJs). The EPA estimated States will require 1 hour per CWS to review the
system's annual report (hrs_annual_report_schoolJs). This burden is based on the ASDWA 2020
CoSTS model, section "Lead Testing in Schools" (ASDWA, 2020b).151 This may overestimate the
burden in Years 14 onward because systems will likely be reporting on their testing program for
fewer schools and child care facilities. Note that an estimate for this review was not explicitly
included in the ASDWA 2024 CoSTS model.

151 Refer to footnote 8.

Final LCRI Economic Analysis

4-311

October 2024


-------
Exhibit 4-156 provides details on how costs are calculated for State school and child care facility
sampling-related costs including additional cost inputs that are required to calculate these costs.

Exhibit 4-156: State School and Child Care Facility Sampling Cost Estimation in SafeWater LCR

by Activity1,2

State Cost Per Activity for CWSs

State Cost Per Activity for
NTNCWSs

Conditions for
Cost to Apply to a
State

Frequency
of Activity





Lead
90th-
Range

Other
Conditions



1) Review list of schools and child care facilities

The hours per system multiplied by
the State labor rate.

(hrs_rev_school_listJs*rateJs)

Cost does not apply to States for
NTNCWSs.

All

All States

Every five
years

m) Provide templates on school and child care facility testing program

The hours per system multiplied by
the State labor rate.

(hrs_temp_schoolJs*rateJs)

Cost does not apply to States for
NTNCWSs.

All

All States

One time

n) Review school and child care facility testing program materials

The hours per system multiplied by
the State labor rate.

(hrs_rev_schoolJnfoJs*rateJs)

Cost does not apply to States for
NTNCWSs.

All

All States

One time

o) Review school and child care facility sampling results

The hours per system multiplied by
the State labor rate.

(hrs_sch_cc_results_reviewJs*rate
is)

Cost does not apply to States for
NTNCWSs.

All

All States

Once a
year

p) Review annual reports on school and child care facility lead in drinking water testing
program

The hours per system multiplied by
the State labor rate.

(hrs_annual_report_schoolJs*rateJ
s)

Cost does not apply to States for
NTNCWSs.

All

All States

Once a
year

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.
Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of the following:

• rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

2 As previously discussed in Section 4.3.2.5 in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina the
State pays for the cost of bottles, shipping, and analyses associated with lead testing (ASDWA, 2020a). Thus, the
State will incur the burden and cost for these activities for the first and subsequent 5-year cycles of the testing
program at schools and child care facilities.

Final LCR! Economic Analysis

4-312

October 2024


-------
4.4.3 State CCT Related Costs

State oversight and review activities related to CCT are grouped into four major subcomponents:

•	CCT Installation

•	Re-optimization

•	Distribution System and Site Assessment

•	Routine

Unit costs and modeling assumptions for each activity related to these four subcomponents are
presented in Sections 4.4.3.1 through 4.4.3.4, respectively.

4.4.3.1 State CCT Installation Costs

The EPA has developed State cost for two one-time activities associated with CCT installation as shown
in Exhibit 4-157. The exhibit provides the unit burden for each activity. The assumptions used in the
estimation of the unit burden follow the exhibit. The last column provides the corresponding SafeWater
LCR model data variables in red/italic font.

Exhibit 4-157: State CCT Installation Related Burden Estimates

Activity

Unit Burden

SafeWater LCR Data Variable

a) Review CCT study and
determine type of CCT to
be installed

27 to 52 hrs/system

hrs_review_cct_study_leadJs

b) Set OWQPs after CCT
installation

2 to 12 hrs/system serving <
50,000 people

hrs_set_owqpJs

Acronyms: CCT = corrosion control treatment; LSL = lead service line; OWQP = optimal water quality parameter.
Source: a), b): "CCT Study and Review Costs_Final.xlsx."

a) Review CCT study and determine type of CCT to be installed (hrs_review_cct_study_leadJs). States
will incur burden to review a system's CCT study. The EPA based the estimated burden on those
provided in the ASDWA 2024 CoSTS model, section "CCT" (ASDWA, 2024). For the proposed LCRI EA
(USEPA, 2023c), the estimated burden was based on responses from North Carolina to a 2016
questionnaire provided by ASDWA (available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov) and included different estimates for systems with and without LSLs. Exhibit
4-158 provides the data variable and input values associated with this activity.

Exhibit 4-158: Estimated Burden for States to Review Initial CCT Study

System Size
(Population Served)

Review CCT Study Report (hrs/system)

(hrs_review_cct_studyjeadJs)

<500

27

501-3,300

27

3,301-10,000

52

Final LCRI Economic Analysis

4-313

October 2024


-------
System Size
(Population Served)

Review CCT Study Report (hrs/system)

(hrs_review_cct_study_leadJs)

10,001-50,000

52

>50,000

N/A

Acronyms: CCT = corrosion control treatment.

Source: "CCT Study and Review Costs_Final.xlsx;" ASDWA, 2024.

Notes:

With the exception of b3 systems, serving > 50,000 people were already required to conduct a CCT
study and install CCT under the LCR.

b) Set OWQPs after CCT installation (hrs_set_owqpJs). The EPA assumed that States will incur burden
to set OWQPs after systems install CCT. The EPA based its estimate on responses from North
Carolina to a 2016 questionnaire provided by ASDWA. The questionnaire and North Carolina's
responses are available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov. Exhibit
4-159 provides the data variable and input values associated with this activity and detailed
assumptions in the notes.

Exhibit 4-159: Estimated Burden for State Review to Set OWQPs

System Size (Population Served)

Set OWQPs1

(hrs_set_owqpJs)

<500

2

501-3,300

5

3,301-50,000

12

>50,0002

N/A

Acronyms: OWQP = optimal water quality parameters.

Source: "CCT Study and Review Costs_Final.xlsx."

Notes:

1	In response to a 2016 ASDWA questionnaire (docket EPA-OW-HQ-2022-0801 at www.regulations.gov). North
Carolina estimated a burden of 2 hours for systems serving < 500 people to 12 hours for systems serving
10,001 to 50,000 people to set OWQPs. The EPA assumed a burden within this range of 5 hours for those
serving 501 to 3,300 people and 12 hours for those serving 3,301 to 10,000 people.

2	With the exception of "b3" systems, serving > 50,000 people were already required to conduct a CCT study
and install CCT under the LCR and States would have already set OWQPs.

Exhibit 4-160 provides the SafeWater LCR model costing approach for the two State activities related to
CCT Installation including additional cost inputs that are required to calculate total costs.

Final LCR! Economic Analysis

4-314

October 2024


-------
Exhibit 4-160: State CCT Installation Cost Estimation in SafeWater LCR by Activity1

State Cost Per Activity for
CWSs

State Cost Per Activity for
NTNCWSs

Conditions for Cost to
Apply to a State

Frequency
of Activity





Lead 90th
- Range

Other Conditions



a) Review CCT study and determine type of CCT to be installed

The hours per system multiplied
by the State labor rate.

(hrs_review_cct_study_leadJs*
rateJs)

Cost applies as written to
States for NTNCWSs.

Above AL

States with any
model PWSs
without CCT
conducting a study
on the installation
of CCT

One time

b) Set OWQPs after CCT installation

The hours per system multiplied
by the State labor rate.

Cost applies as written to
States for NTNCWSs.

Above AL

States with any
model PWSs

One time

(hrs_set_owqpJs*rateJs)



installing CCT



Acronyms: AL = action level; CCT = corrosion control treatment; CWS = community water system; NTNCWS = non-

transient non-community water system; PWS = public water system.

Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of:

• rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

4.4.3.2 State CCT Re-optimization Costs

The EPA has identified and developed State costs for two oversight and review activities associated with
a system's re-optimization of existing CCT, as shown in Exhibit 4-161. The exhibit provides the unit
burden for each activity. The assumptions used in the estimation of the unit burden follow the exhibit.
The last column provides the corresponding SafeWater LCR model data variable in red/italic font.

Exhibit 4-161: State CCT Re-Optimization-Related Burden Estimates

Activity

Unit Burden

SafeWater LCR Data Variable

c) Review CCT study and
determine needed OCCT
adjustment

28 to 50 hrs/system

hrs_review_cct_study_leadJs

d) Reset OWQPs after CCT re-
optimization

2 to 20 hrs/system

hrs_reset_owqpJs

Acronyms: CCT = corrosion control treatment; LSL = lead service line; OWQP = optimal water quality parameter.
Source: "CCT Study and Review Costs_Final.xlsx."

c) Review CCT study and determine needed OCCT adjustment (hrs_review_cct_study_ieadJs). States
will incur burden to review the revised CCT study for PWSs with existing CCT when they exceed the
AL. The EPA based its estimates on those provided in the ASDWA 2024 CoSTS model, section "CCT"
(ASDWA, 2024). For the proposed LCRI EA (USEPA, 2023c), the EPA based the burden estimates on

Final LCRI Economic Analysis

4-315

October 2024


-------
responses from North Carolina to a 2016 questionnaire provided by ASDWA (available in the docket
at EPA-HQ-OW-2022-0801 at www.regulations.gov) and included different estimates for systems
with and without LSLs. The estimated burden to review a revised study is provided in Exhibit 4-162.

Exhibit 4-162: Estimated Burden for States to Review a Revised CCT Study and Determine

Needed CCT Adjustment

System Size (Population Served)

Review Revised CCT Study Report
(hrs/system)

(hrs_review_cct_study_leadJs)

<500

28

501-3,300

28

3,301-10,000

50

10,001-50,000

50

>50,000

50

Acronyms: CCT = corrosion control treatment.
Source: "CCT Study and Review Costs_Final.xlsx."

d) Reset OWQPs after CCT re-optimization hrs_reset_owqpJs). States will need to reset OWQPs after
the system re-optimizes its CCT. For systems serving 50,000 or fewer people, the EPA assumed this
burden is the same as the burden to set OWQPs for the first time (2 to 12 hours, data variable
hrs_set_owqpJs as presented in Exhibit 4-159). For systems serving more than 50,000 people, the
EPA assumed a burden of 20 hours for States to reset OWQPs due to the larger size and relative
complexities of these systems.

Exhibit 4-163 details how the data variables are used to estimate State activities related to CCT re-
optimization including additional cost inputs that are required to calculate the total costs.

Exhibit 4-163: State CCT Re-optimization Cost Estimation in SafeWater LCR by Activity1

State Cost Per Activity for CWSs

State Cost Per Activity for
NTNCWSs

Conditions for Cost to
Apply to a State

Frequency
of Activity





Lead 90th -
Range

Other
Conditions



c) Review CCT study and determine needed OCCT adjustment

The hours per system multiplied by
the State labor rate.

(hrs_review_cct_study_leadJs*rate
J's)

Cost applies as written to
States for NTNCWSs.

Above AL

States with
model PWS
conducting a
study prior to
re-optimizing
CCT

One time

d) Reset OWQPs after CCT re-optimization

The hours per system multiplied by
the State labor rate.

(hrs_reset_owqpJs*rateJs)

Cost applies as written to
States for NTNCWSs.

Above AL

States with
model PWS
re-optimizing
CCT

One time

Acronyms: AL = action level; CCT = corrosion control treatment; CWS = community water system; NTNCWS = non-
transient non-community water system; OCCT = optimal corrosion control treatment; OWQP = optimal water
quality parameters; PWS = public water system.

Final LCR! Economic Analysis

4-316

October 2024


-------
Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of:

• rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

4.4.3.3 State Distribution System and Site Assessment Costs

The EPA developed State costs to related to DSSA activities as shown in Exhibit 4-164. The exhibit
provides the unit burden for each activity. The assumptions used in the estimation of the unit burden
follow the exhibit. The last column provides the corresponding SafeWater LCR model data variable in
red/italic font.

Exhibit 4-164: State DSSA Burden Estimates

Activity

Unit Burden

SafeWater LCR Data Variable

e) Consult with system prior to any
DSSA CCT adjustments

2 hrs per PWS

hrs_consult_dssaJs

f) Review report on DSSA responses

1	hr/PWS serving < 50,000 people;

2	hrs/PWS serving > 50,000 people

hrs_report_dssaJs

Acronyms: CCT = corrosion control treatment; DSSA = Distribution System and Site Assessment; PWS = public
water system.

Source: "Likelihood_Sample_Above_AL_LCRI_DSSA.xlsx."

e)	Consult with system prior to any DSSA CCT adjustment (hrs_consuit_dssa Js). Systems with CCT
that have at least one sample > 10 ng/L must consult with their State prior to making any CCT
changes. The EPA assumed States will incur a 2 hour burden per system that is consistent with other
types of consultations, e.g., State consultation prior to a change in source or treatment.

f)	Review report on DSSA responses (hrs_report_dssaJs). States will incur burden to review the
system's report that provides the results of tap and WQP monitoring, a distribution system
assessment, and recommended corrective actions (i.e., DSSA responses) if a system has one or more
samples above 10 ng/L in a given year. The EPA assumed the State will require 1 hour and 2 hours to
review the report submitted by systems serving 50,000 or fewer and those serving more than
50,000 people, respectively.

Exhibit 4-165 provides details on how total costs for the final LCRI are calculated for this activity
including additional cost inputs that are required to calculate the total costs.

Final LCRI Economic Analysis

4-317

October 2024


-------
Exhibit 4-165: State CCT DSSA Cost Estimation in SafeWater LCR by Activity1,2

State Cost Per Activity for
CWSs

State Cost Per Activity for
NTNCWSs

Conditions for Cost
to Apply to a State

Frequency
of Activity





Lead 90th
- Range

Other
Conditions



e) Consult with system prior to any DSSA CCT adjustments

The hours per system multiplied
by the State labor rate.

(hrs_consult_dssaJs*rateJs)

Cost applies as written to States
for NTNCWSs.

All

All States with
model PWS with
at least one
sample > 10

hq/l

Once a
year

f) Review report on DSSA responses

The hours per system multiplied
by the State labor rate.

(hrs_report_dssaJs*rateJs)

Cost applies as written to States
for NTNCWSs.

All

All States with
model PWS with
at least one
sample > 10
ms/l

Once a
year

Acronyms: CWS = community water system; CCT = corrosion control treatment; DSSA = Distribution System and
Site Assessment; NTNCWS = non-transient non-community water system; PWS = public water system.

Notes:

1	The data variables in the exhibit are defined previously in this section with the exception of:

• rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

2	As previously discussed in Section 4.3.3.2.2 in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina the
State pays for the cost of bottles, shipping, and analyses (ASDWA, 2020a). Thus, the State will incur the burden and
cost for these activities.

4.4.3.4 State Lead CCT Routine Costs

The EPA developed State costs to review and consult on system's activities related to review of CCT
guidance, submitted water quality data during the sanitary survey, and the notification of a source or
treatment change as shown in Exhibit 4-166. The exhibit provides the unit burden for each activity. The
assumptions used in the estimation of the unit burden follow the exhibit. The last column provides the
corresponding SafeWater LCR model data variable in red/italic font.

Exhibit 4-166: State CCT Installation Related Burden Estimates

Activity

Unit Burden

SafeWater LCR Data Variable

g) Review CCT guidance and
applicability to individual
PWSs

40 hrs/State/update

hrs_cct_reviewJs

h) Review water quality data
with PWSs during sanitary
survey

2 to 5 hrs/system/sanitary survey

hrs_sanit_survJs

i) Consult on required
actions in response to
source water change

•	6 to 12 hrs/system on reduced
tap monitoring

•	4 to 7 hrs/system on standard
tap monitoring

hrs_coop_source_chng_redJs
hrs_coop_source_chng_routJs

Final LCR! Economic Analysis

4-318

October 2024


-------
Activity

Unit Burden

SafeWaterLCR Data Variable

j) Consult on required
actions in response to
treatment change

46 to 84 hrs/system

hrs_coop_treat_chng_ js

Acronyms: CCT = corrosion control treatment; PWS = public water system.

Sources:

g), h): "CCT Study and Review Costs_Final.xlsx."
i): "Likelihood_SourceChange_Final.xlsx."
j): "Likelihood_TreatmentChange_Final.xlsx."

g)	Review CCT guidance and applicability to individual PWSs (hrs_cct_reviewJs). States will incur
burden to review updated EPA guidance, identify changes that could affect their systems, prepare a
memo to communicate changes to State surveyors, and be available to answer questions

(hrs_cct_reviewJs) at an estimated burden of 40 hours total. The estimate is based on the ASDWA
2024 CoSTS model, section "CCT" (ASDWA, 2024). The EPA assumed this guidance will be updated
every 5 years. For the proposed LCRI EA (USEPA, 2023c), the EPA used an estimate of 16 hours based
on Indiana's response to a 2016 ASDWA questionnaire.

h)	Review water quality data with PWSs during sanitary survey (hrs_sanit_survJs). States will also
incur burden to review water quality data during sanitary surveys with water systems that have CCT.
Exhibit 4-167 provides the data variables and input values associated with this review.

Exhibit 4-167: Estimated State Burden to Review CCT-Related Data during Sanitary Survey

System Size
(Population Served)

State Burden (hrs / system)

(hrs_sanit_survJs)

<1,000

2

1,001-10,000

3

10,001-100,000

4

>100,000

5

Note:

The EPA assumed that State burden for reviewing CCT-related non-compliance data would be
twice that of the system burden to gather the data (see data variable: hrs_sanit_surv_op in
Section 4.3.3.4, activity m)) plus 1 hour to discuss the sanitary survey.

The minimum sanitary survey frequency is every 3 years for surface water CWSs and every 5 years
for NTNCWSs. The minimum frequency for ground water CWSs is also every 3 years except for the
subset that can meet certain treatment or performance criteria. For these systems, the minimum
frequency can be extended to every 5 years. Refer to Section 4.3.3.4, activity m) for the likelihood a
ground water system will meet these treatment or performance criteria (p_spec_req).

i) Consult on required actions in response to source water change (hrs_coop_source_chng_red Js,
hrs_coop_source_chng_routJs). Systems are required to seek prior approval before making any
source water changes and to consult with the State on needed responses. Exhibit 4-168 provides the
estimated State burden estimate of 6 to 12 hours (6 hours most likely) per system per monitoring
period for systems on reduced monitoring and an estimate of 4 to 7 hours (4 hours most likely) per
system per monitoring period for system on standard monitoring for this review and consultation,

Final LCRI Economic Analysis

4-319

October 2024


-------
which is based on input received from North Carolina and Indiana in response to a 2016 ASDWA
questionnaire regarding potential 2021 LCRR requirements. North Carolina estimated 2 hours to
review a change in source from ground water to another ground water source and 3 hours for
surface water source changes or surface water/ground water mixing. Indiana estimated 6 hours to
review a change to a similar source and 20 hours to review a change to a dissimilar source. The
questionnaire and each State's responses are available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

The estimated burden for States to consult with systems in response to source change depends on
the system's lead tap monitoring and reporting frequency as follows:

•	For systems monitoring less frequently than every 6 months (hrs_coop_source_chng_redJs),
the EPA used the average of the two estimates of 2 and 6 hours (4 hours) for the minimum and
most likely value. The EPA set the most likely equal to the minimum because less than 1 percent
of systems made more significant sources changes during 2013-2020. For the maximum, the EPA
assumed the 20 hours were more reflective of the system burden to prepare needed
documentation and instead set the State burden to equal 50 percent of that estimated for the
system (50 percent of 20 hours). Additionally, the EPA assumed States would incur an additional
2 hour burden to consult with the system on needed actions in response to the source change
for a total burden of 6 hours for the minimum and most likely and 12 hours for the maximum.

•	For systems monitoring every 6 months, the EPA assumed 50 percent of the burden estimated
for hrs_coop_source_chng_routJs for the review portion because the State is already reviewing
data semi-annually as opposed to annually and an additional 2 hours for the consultation. For
the minimum and most likely the burden equals 2 hours for the review plus 2 hours for the
consultation for a total of 4 hours. For the maximum, the burden equals 5 hours for the review
plus 2 hours for the consultation for a total of 7 hours. Exhibit 4-168.

Exhibit 4-168: Estimated Hours per System for State to Consult on Source Water Change

Hrs per system per monitoring period

hrs_coop_source_chng_re
dJs

hrs_coop_source_chng_routJs



Minimum

Maximum

Most
Likely

Minimum

Maximum

Most
Likely

6

12

6

4

7

4

Source: "Likelihood_SourceChange_Final.xlsx."

As discussed in Section 3.3.8.1, the EPA used historical data from SDWIS/Fed to estimate the
likelihood that 3.43 percent of CWSs and 1.58 percent of NTNCWSs would have a source change in
any given year (p_source_change).

j) Consult on required actions in response to treatment change (hrs_coop_treat_chng_ js;

hrs_coop_treat_chng_ js). Systems are also required to seek prior approval before making any long-
term treatment changes and to consult with the State on needed responses. The likelihood of a
system changing treatment in any given year of 4.2 percent for all CWSs and 3.2 percent for all

Final LCRI Economic Analysis

4-320

October 2024


-------
NTNCWSs (p_treat_change) is discussed in Section 3.3.8.3 with estimated percentages for CWSs and
NTNCWSs presented in Exhibit 3-55 and Exhibit 3-56, respectively. Exhibit 4-169 below provides the
estimated State burden for this review and consultation, which is based on burden estimates
provided by the ASDWA 2024 CoSTS model, section "CCT" (ASDWA, 2024). The EPA assumed the
burden for States to review and consult on the treatment change to be the same as the burden
needed for water systems to report the change and consult with the States on needed actions. The
EPA also assumed based on the ASDWA 2024 CoSTS model that burden would not differ based on
the system's monitoring schedule.152

Exhibit 4-169: Estimated Hours per System for State to Consult on Treatment Change

System Size
(Population Served)

Hrs per system per monitoring
period

hrs_coop_treat_chng_ js

<100

46

101-500

46

501-1,000

46

1,001-3,300

46

3,301-10,000

84

10,001-50,000

84

>50,000

82

Source: "Probability_TreatmentChange_Final.xlsx."

Exhibit 4-170 details how the data variables are used to estimate State activities related to CCT re-
optimization including additional cost inputs that are required to calculate the total costs.

Exhibit 4-170: State CCT Re-optimization Cost Estimation in SafeWater LCR by Activity1

State Cost Per Activity for
CWSs

State Cost Per
Activity for
NTNCWSs

Conditions for Cost to Apply
to a State

Frequen

cy of
Activity





Lead 90th -
Range

Other Conditions



g) Review CCT guidance and applicability to individual PWSs

The total hours multiplied by
the State labor rate.

(hrs_cct_reviewJs*rateJs)

Cost applies as written
to States for
NTNCWSs.

All

States with any model
PWSs with CCT

Every 5
years

h) Review water quality data with PWSs during sanitary survey

The hours per system
multiplied by the State labor
rate.

(hrs_sanit_survJs*rateJs)

Cost applies as written
to States for
NTNCWSs.

All

States with any model
PWSs that do not meet
the special requirements
to conduct the Sanitary
Survey at a reduced
interval

Every 3
years

152 For the proposed LCRI EA (USEPA, 2023c), the EPA based the State review and consultation burden on North
Carolina's response to a 2016 ASDWA questionnaire regarding possible 2021 LCRR requirements. In addition, the
EPA had assumed different burdens based on a system's monitoring schedule.

Final LCRI Economic Analysis

4-321

October 2024


-------
State Cost Per Activity for
CWSs

State Cost Per
Activity for
NTNCWSs

Conditions for Cost to Apply
to a State

Frequen

cy of
Activity





Lead 90th -
Range

Other Conditions









1 - p_spec_req









States with any model
PWSs that do meet the
special requirements to
conduct the Sanitary
Survey at a reduced
interval

Every 5
years







p_spec_req



i) Consult on required actions in response to source water change

The hours per system
multiplied by the State labor
rate.

(hrs_coop_source_chng_rout
Js*rateJs)

The hours per system
multiplied by the State labor
rate.

(hrs_coop_source_chng_red
Js*rateJs)

Cost applies as written
to States for
NTNCWSs.

At or below
AL

States with any model
PWSs not on reduced tap
sampling that have a
change in source water

1 - (p_tap_annual +
p_tap_triennial +
p_tap_nine);
p_source_chng

Above AL

States with any model
PWSs with a change in
source water

p_source_chng

At or below
AL

States with any model
PWSs on reduced tap
sampling that have a
change in source water

p_tap_annual,
p_tap_triennial,
p_tap_nine,
p_source_chng

j) Consult on required actions in response to treatment change

Final LCRI Economic Analysis

4-322

October 2024


-------
State Cost Per Activity for
CWSs

State Cost Per
Activity for
NTNCWSs

Conditions for Cost to Apply
to a State

Frequen

cy of
Activity





Lead 90th -
Range

Other Conditions



The hours per system
multiplied by the State labor
rate.



At or below
AL

States with any model
PWSs with a change in
treatment



(hrs_coop_treat_chng_
js*rateJs)





p_treat_change





Cost applies as written
to States for
NTNCWSs.

Above AL

States with any model
PWSs with a change in
treatment

p_treat_change

Once per
event

Acronyms: AL = action level; CCT = corrosion control treatment; CWS = community water system; NTNCWS = non-

transient non-community water system; PWS = public water system.

Note:

1The data variables in the exhibit are defined previously in this section with the exception of:

•	p_tap_annual, p_tap_triennial, and p_tap_nine: Likelihood a system will qualify to collect the reduced
number of lead tap samples at an annual, triennial, and nine-year frequency, respectively (Chapter 3,
Section 3.3.7).

•	p_source_chng\ Likelihood that a system will change sources in a given year (Chapter 3, Section 3.3.9.1).

•	p_spec_req\ Likelihood a ground water CWS will meet special conditions to conduct a sanitary survey
every 3 years vs. every 5 years (Section 4.3.3.4, activity m)).

•	p_treat_change: Likelihood that a system will change treatment in a given year (Chapter 3, Section
3.3.9.3).

•	rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

4.4.4 State Service Line Inventory and Replacement Related Costs

States will incur burden to conduct oversight activities related to systems' SL inventory and replacement
programs. Section 4.4.4.1 describes oversight activities associated with the SL inventory and outreach.
Section 4.4.4.2 includes activities to review the SLR plan and periodic re-evaluation of SLR rates for
systems eligible for a deferred deadline. Section 4.4.4.3 includes the review of the annual SLR report.
Exhibit 4-177 at the end of Section 4.4.4.3 provides details on how costs are calculated for State service
line inventory and replacement activities a) through e) including additional cost inputs that are required
to calculate these costs.

4.4.4.1 SL Inventory Costs

The EPA has identified and developed State costs for activities associated with SL inventory
development as shown in Exhibit 4-171. The exhibit provides the unit burden for each activity. The
assumptions used in the estimation of the unit burden follow the exhibit. The last column provides the
corresponding SafeWater LCR model data variable in red/italic font.

Final LCR! Economic Analysis

4-323

October 2024


-------
Exhibit 4-171: State SL Inventory Burden Estimates

Activity

Unit Burden

SafeWater LCR Data Variable

a) Review connector updated LCRR
initial inventory (baseline inventory)
(one-time)

0.5 to 2 hrs/CWS or NTNCWS

hrs_updated_initial_inv_revJs

b) Review annual service line
inventory updates

0.5 hrs/CWS or NTNCWS

hrsjnv_update_revJs

c) Review validation report (one-time)

0.5 hrs per CWS

hrsjnv_valid_revJs

Acronyms: CWS = community water system; SL = service lines; NTNCWS = non-transient non-community water

system; PWS = public water system.

Sources:

a):	"LCRI Updated Initial Inventory with Connectors.xlsx"

b)	& c): "Inventory Updates and Validation.xlsx."

a)	Review connector updated LCRR initial inventory (baseline inventory)
(hrs_updated_initial_inv_revJs). Under the final LCRI, States will incur a one-time burden to review
the updated LCRR initial inventory that includes lead connector information. The EPA assumed
States would require 0.5, 1 hour, and 2 hours to conduct this review for CWSs and NTNCWSs serving
3,300 or fewer people; 3,301 to 50,000 people and more than 50,000 people, respectively. This is
half of the CWS burden to prepare and report an updated inventory with connectors
(hrs_report_updated_initial_inv_op).153

b)	Review annual service line inventory updates (hrs_inv_update_revJs). The EPA assumed States
will incur an annual burden to review CWS and NTNCWS updated inventories. The EPA estimated
the State will require 0.5 hours to review each update.

c)	Review inventory validation report (hrs_inv_valid_revJs). The EPA assumed States will incur a one-
time burden to review CWS and NTNCWS validation results. The EPA estimated the State will require
0.5 hours to review the validation results. ASDWA agreed with this estimate in their 2024 CoSTS
model (ASDWA, 2024).

4.4.4.2 SLR Plan Review Costs

The EPA has identified and developed State costs for activities associated with the review of the SLR
plan and annual SLR report as shown in Exhibit 4-172. The exhibit provides the unit burden for each
activity. The assumptions used in the estimation of the unit burden follow the exhibit. The last column
provides the corresponding SafeWater LCR model data variable in red/italic font.

153 The burden estimate in the proposed LCRI EA (USEPA, 2023c) of 1 hour per CWS and NTNCWS for all size
categories did not reflect the larger number of service lines for larger systems compared to smaller systems. The
former will have a more extensive inventory. Using half of the CWS estimate is an approach used for other State
inputs. The EPA used half the burden for CWSs to estimate the State review of NTNCWSs because the estimate for
NTNCWS to report their updated inventory with connectors (3.75 to 15 hrs) includes hours to compile the
connector information in addition to reporting.

Final LCRI Economic Analysis

4-324

October 2024


-------
Exhibit 4-172: State SLR Plan and Deferred Replacement Deadline Review Burden Estimates

Activity

Unit Burden

SafeWater LCR Data Variable

d) Review initial SLR plan (one-time)

6 to 18 hours/CWS
6 hours/NTNCWS

hrs_slr_planJs

e) Review information on deferred
deadline and associated replacement
rate in the SLR plan and determine
fastest feasible rate (one-time)

1.5 to 4.5 hrs/CWS seeking a
deferred SLR rate;
1.5 hrs/NTNCWS seeking a
deferred SLR rate

hrs_slr_plan_deferJs

f) Review annually updated SLR plan or
certification of no change

1 to 2 hrs/CWS;
1 hr/NTNCWS

hrs_slr_plan_updateJs

g) Conduct triennial review of water
system updated recommended
deferred deadline and associated
replacement rate and determine
fastest feasible rate

1.5 to 4.5 hrs/CWS on a deferred
SLR rate;

1.5 hrs/NTNCWS on a deferred SLR
rate

hrs_defer_updateJs

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; PWS = public
water system; SLR = service line replacement.

Source: "LSLR Ancillary Costs_Final.xlsx."

d)	Review initial SLR plan (hrs_slr_plan_js). States will incur burden to review the SLR plan that water
systems with lead, GRR, and/or unknown service lines must prepare (see activity g) in Section
4.3.4.2 for required elements of the plan). The State burden (hrs_slr_planJs) is based on the
ASDWA 2020 CoSTS model that assumed 6 hours for States to review the plan for small CWSs
(assumed to serve 3,300 or fewer people) and NTNCWSs, 10 hours for medium CWSs (assumed to
serve 3,301 to 50,000 people), and 18 hours for large CWSs (assumed to serve more than 50,000
people) (ASDWA, 2020b; 2024). ASDWA's estimates remained the same in their 2024 CoSTS model.

e)	Review information on deferred deadline and associated replacement rate in the SLR plan and
determine fastest feasible rate (hrs_slr_plan_deferJs). States will incur burden to conduct an
additional review for systems requesting a deferred replacement deadline in their initial SLR plan.
The State must determine whether the system's requested deferred deadline and associated
cumulative average replacement rate are the fastest feasible to conduct mandatory SLR. If the
requested rate is not the fastest feasible, the State must set a new deferred deadline and
replacement rate that is the fastest feasible for the system. The State must consider information
that includes, but is not limited to, the system's submissions of the service line inventory and
replacement plan and information collected from other water systems conducting mandatory SLR.
The EPA assumed that States would incur half the burden required for systems to prepare the
additional information, as shown in Exhibit 4-173 below.

Final LCRI Economic Analysis

4-325

October 2024


-------
Exhibit 4-173: Estimated Additional Burden for States to Review the Initial SLR Plan for
Systems Requesting a Deferred Replacement Deadline

System Size
(Population Served)

hrs_slr_plan_deferJs

CWSs

NTNCWSs

<3,300

1.5

1.5

3,301-10,000

2.5

1.5

10,001-50,000

2.5

1.5

>50,000

4.5

1.5

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.

Source: "LSLR Ancillary Costs_Final.xlsx."

Notes: This additional burden only applies to States reviewing the SLR plan for CWSs requesting a deferred
replacement deadline.

f) Review annually updated SLR plan or certification of no change (hrs_slr_plan_updateJs). All

systems with lead, GRR, and/or unknown service lines must either update their SLR plan annually,
starting in Year 2 (i.e., Year 5 of the period of analysis), to include any significant changes, such as
updates to relevant regulations, approach for identifying unknowns or submit a certification of no
change. The EPA assumed the majority of systems will not need to update their SLR program but
instead will provide a certification of no change. The EPA assumed that States would incur half the
burden required for systems to prepare the updated plan or certification, as shown in Exhibit 4-174
below.

Exhibit 4-174: Estimated Annual Burden for States to Review SLR Plan Updates or

Certifications of No Changes

System Size
(Population Served)

hrs_slr_plan_updateJs

CWSs

NTNCWSs

<3,300

1

1

3,301-10,000

1.5

1

10,001-50,000

2

1

>50,000

2

1

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.

Source: "LSLR Ancillary Costs_Final.xlsx."

Notes: Systems with lead, GRR, or unknowns must annually update their SLR plan if they have a significant change
or must instead certify to the State that they have no changes.

g) Conduct triennial review of water system updated recommended deferred deadline and

associated replacement rate and determine fastest feasible rate (hrs_defer_updateJs). By the end

of the fifth program year, and every three years thereafter, the State must review the system's
updated recommendation of the deferred deadline and associated replacement rate and determine

Final LCRI Economic Analysis

4-326

October 2024


-------
if a shorter deadline is feasible. If the requested rate is not the fastest feasible, the State must set a
new deferred deadline and replacement rate that is the fastest feasible for the system. The EPA
assumed that States would incur half the burden required for systems to prepare the update
required for their replacement plan, which equates to 1.5 hour for CWSs serving 3,300 or fewer
people and all NTNCWSs, 2.5 hours for CWSs serving 3,301 to 50,000 people, and 4.5 hours for
CWSs serving more than 50,000 people.

4.4.4.3 SLR Report Review Costs

The EPA has identified and developed State costs for an activity associated with the review of the annual
SLR report, as shown in Exhibit 4-175 . The exhibit provides the unit burden for each activity. The
assumptions used in the estimation of the unit burden follow the exhibit. The last column provides the
corresponding SafeWater LCR model data variable in red/italic font.

Exhibit 4-175: State Report Review Burden Estimates

Activity

Unit Burden

SafeWater LCR Data Variable

h) Review annual SLR program report

1 to 4 hours/CWS
1 hour/NTNCWS

hrs_reportJcrJs

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; SLR = service
line replacement.

h) Review annual SLR program report (hrs_report_lcr Js). States will incur burden to review annual
information submitted by water systems related to their SLR program, including the location of each
lead and GRR service line and lead connector replaced, the number of unknown service lines
determined to be non-lead, the number of unknown service lines remaining, their replacement
schedule, and other information as required under 40 CFR 141.90(e). This information is expected to
be in the form of an annual report. Exhibit 4-176 provides the estimated burden associated with this
review. For the proposed LCRI EA, the EPA assumed that the State review burden would be half of
the system burden to prepare the SLR program report (as presented in Section 4.3.4.4, activity r)).
For this final LCRI EA, the EPA used the burden estimates provided in the ASDWA 2024 CoSTS model
(ASDWA, 2024) for CWSs serving less than or equal to 3,300 people (1 hr), CWSs serving 50,001 to
100,000 (3 hrs), and all NTNCWS (1 hr). The EPA continued to use the estimated burden from the
proposed LCRI EA for all other CWS size categories because it was higher or the same as the ASDWA
2024 CoSTS model.

Exhibit 4-176: State Burden to Review System's Annual Service Line Replacement

Report (hrs per system)

System Size
(Population Served)

CWSs

NTNCWSs

SafeWater cost input ID: hrs_reportJcrJs



A

B

<3,300

1

1

3,301-10,000

1

1

Final LCRI Economic Analysis

4-327

October 2024


-------
System Size
(Population Served)

CWSs

NTNCWSs

SafeWater cost input ID: hrs_reportJcrJs



A

B

10,001-50,000

2

1

50,001 -100,000

3

1

>100,000

4

1

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.
Source: "LSLR Ancillary Costs_Updated.xlsx."

Exhibit 4-177 provides the SafeWater LCR model costing approach including additional cost inputs that
are required to calculate the total costs.

Exhibit 4-177: State Service Line Replacement Cost Estimation in SafeWater LCR by Activity1,2

State Cost Per Activity for CWSs

State Cost Per Activity
for NTNCWSs

Conditions for Cost to
Apply to a State

Frequency
of Activity





Lead 90th -
Range

Other
Conditions3



a) Review connector updated LCRR initial inventory (baseline inventory)

The hours per system multiplied by the
State labor rate.

hrs_updated_initial_inv_revJs*rateJs

Cost applies as written
to States for
NTNCWSs.

All

States with any
model PWSs

Once per
year for the
first three
years

b) Review annual service line inventory updates

The hours per system multiplied by the
State labor rate.

(hrs_inv_update_revJs*rateJs)

Cost applies as written
to States for
NTNCWSs.

All

States with any
model PWSs
with service
lines of lead or
unknown
content

Once per
year for the
first 10
years

c) Review inventory validation report

The hours per system multiplied by the
State labor rate.

(hrsjnv_valid_revJs*rateJs)

Cost does not apply to
States for NTNCWSs.

All

States with any
model PWSs
with service
lines of lead
content or
unknowns

One Time

d) Review initial SLR plan

The hours per system multiplied by the
State labor rate.

(hrs_slr_planJs*rateJs)

Cost applies as written
to States for
NTNCWSs.

All

States with any
model PWSs
with service
lines containing
lead content or
unknowns

One Time

e) Review information on deferred deadline and associated replacement rate in the SLR plan
and determine fastest feasible rate

The hours per system multiplied by the
State labor rate.

(hrs_slr_plan_defer js*rateJs)

Cost applies as written
to States for
NTNCWSs.

All

Model PWSs
seeking a
deferral.

One time

f) Review annually updated SLR plan or certification of no change

Final LCR! Economic Analysis

4-328

October 2024


-------
State Cost Per Activity for CWSs

State Cost Per Activity
for NTNCWSs

Conditions for Cost to
Apply to a State

Frequency
of Activity





Lead 90th -
Range

Other
Conditions3



The hours per system multiplied by the
State labor rate.

(hrs_slr_plan_update js*rateJs)

Cost applies as written
to States for
NTNCWSs.

All

Model PWSs
with service
lines of lead,
GRR, and/or
unknown
service lines

Year 5 and

annually

thereafter

g) Conduct triennial review of water system updated recommended deferred
associated replacement rate and determine fastest feasible rate

deadline and

The hours per system multiplied by the
State labor rate.

(hrs_slr_defer_updateJs*rateJs)

Cost applies as written
to States for
NTNCWSs.

All

Model PWSs on
a deferred SLR
rate

Year 8 and

triennially

thereafter

h) Review annual SLR program report

The hours per system multiplied by the
State labor rate.

(hrs_reportJcrJs*rateJs)

Cost applies as written
to States for
NTNCWSs.

All

States with any
model PWSs
that are
replacing lead
or GRR service
lines

Once a
year

Acronyms: AL = action level; CWS = community water system; GRR = galvanized requiring replacement; LSL = lead
service line; LSLR = lead service line replacement; NTNCWS = non-transient non-community water system; PWS =
public water system; SLR = service line replacement.

Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of the following:

• rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

2As previously discussed in Section 4.3.4.4, in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina the
State pays for the cost of bottles and shipping and conducting the analysis for samples following LSLR (ASDWA,
2020a). Thus, the State will incur the burden and cost for these activities.

3 PWSs with lead content or unknown lines are identified using the data variables and approach described in
Chapter 3, Section 3.3.4.

4.4.5 State POU Related Costs

States will incur both one-time and ongoing burden to conduct oversight activities related to systems'
POU programs. CWSs serving 3,300 or fewer people and NTNCWSs with a lead 90th percentile above the
AL must evaluate and recommend to their State which compliance alternative they plan to implement
that can include POU device installation and maintenance. State activities and associated SafeWater LCR
model cost inputs for one-time and ongoing activities are described in Sections 4.4.5.1 and 4.4.5.2,
respectively.

4.4.5.1 One-Time POU Program Costs

The EPA has developed costs for three one-time State activities related to POU program oversight as
shown in Exhibit 4-178. The exhibit provides the unit burden for each activity. The assumptions used in
the estimation of the unit burdens follow the exhibit. The last column provides the corresponding
SafeWater LCR model data variable in red/italic font.

Final LCR! Economic Analysis

4-329

October 2024


-------
Exhibit 4-178: State One-Time POU-Related Burden Estimates

Activity

Unit Burden

SafeWater LCR Data Variable

a) Review POU plan

37 to 67 hrs/CWS serving < 3,300;
29.5 to 67 hrs/NTNCWS

hrs_pou_plan_revJs

b) Provide templates for POU
outreach materials

0.25 to 0.5 hrs/CWS serving < 3,300 and
NTNCWS

hrs_temp_pouJs

c) Review POU public
education materials

0.5 hrs/CWS serving < 3,300;
0.5 to 2 hrs/NTNCWSs

hrs_review_pe_pouJs

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; POU = point-

of-use; PWS = public water system.

Source:

a):	"POU lnputs_Final.xlsx."

b)	& c): "Public Education lnputs_CWS_Final.xlsx"; "Public Education lnputs_NTNCWS_Final.xlsx."

Notes:

c):	States will only conduct these activities for the subset of CWS serving <3,300 people and NTNCWSs with a lead

ALE and for which POU provision and maintenance is their approved lead compliance option.

a) Review POU plan (hrs_pou_plan_revJs). As previously stated in Section 4.3.5.2, the rule does not
explicitly require systems to prepare a POU plan under the small system flexibility requirements.
However, the EPA assumed systems would prepare a plan and States would incur burden to review
water systems' POU plans. These assumptions are made given the desire in the EA to capture all
reasonable costs incurred by the impacted entities and the high likelihood that States will want to
have oversight on the human health protective actions being taken by a system in response to high
lead samples at households. The SafeWater LCR model assumes that these plans are developed by
CWSs serving 3,300 or fewer people and NTNCWSs that meet the following criteria: 1) have no CCT,
2) have a lead ALE, and 3) POU provision and maintenance is their approved compliance option. The
EPA assumed that State burden to review the plan is 25 percent of the PWS burden to prepare the
plan (hrs_pou_plan_dev_op), excluding the system's burden for board/management and legal
consultation.154 The State burden is provided in Exhibit 4-179. See Section 4.3.5.2, activity b) for
assumptions used to estimate the PWS burden. The EPA estimates NTNCWSs on average will have
more taps that will require POU devices than CWSs and thus they will require additional burden to
develop the plan and for the State to review the plan.

Exhibit 4-179: Estimated Hours for State Review of POU Plan (hrs/system)

System size
(Population Served)

CWSs

NTNCWSs

SafeWater LCR Data Variable: hrs_pou_plan_revJs

A

B

<500

37

29.5

501-3,300

67

42

154 For the proposed LCRI EA, the EPA estimated the State burden to be 50 percent of the burden estimated for a
water system to prepare the plan. For the final LCRI EA, EPA revisited its assumption due to the low estimated
burden of 2 to 4 hours for this review provided by ASDWA in its 2024 CoSTS model (ASDWA, 2024).

Final LCRI Economic Analysis

4-330

October 2024


-------
System size
(Population Served)

CWSs

NTNCWSs

SafeWater LCR Data Variable: hrs_pou_plan_revJs

A

B

3,301 to 10,000

N/A

42

10,001-50,000

N/A

67

50,001-1,000,000

N/A

42

>1,000,000

N/A



Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; POU = point-
of-use.

Source: "POU lnputs_Final.xlsx."

Notes:

A: The EPA assumed States will incur 25 percent of the burden as to review the plan as for water systems to
prepare the plan (see data variable hrs_pou_plan_dev_op in Section 4.3.5.2, activity b)), excluding the system's
burden for board/management and legal consultation.

B: No NTNCWSs serves more than 1 million people; thus, the burden for this size category is 0. The EPA estimates
that NTNCWSs serving 10,001 - 50,000 people have the highest estimated number of taps, will have a higher
burden to prepare the POU plan, and States will require additional burden to review the plan. See "POU
lnputs_Final.xlsx" for the approach for estimating the required number of POU devices.

b)	Provide templates for POU outreach materials (hrs_temp_pouJs). The EPA assumed that States
will provide templates to CWSs serving 3,300 or fewer people and NTNCWSs to develop POU
outreach materials that describe the POU program and proper use of the POU devices. The EPA
assumed States will incur a one-time burden of 0.25 to 0.5 hours to provide these templates based
on responses to an ASDWA survey regarding the burden to provide revised sampling instruction
templates from North Carolina and Indiana of 0.25 and 0.5 hours, respectively. The questionnaire
and each State's responses are available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

c)	Review POU public education materials (hrs_review_pe_pouJs). CWSs serving 3,300 or fewer
people with a lead ALE that selected the POU option must provide public education on the use of
POU device to all households they serve. NTNCWSs must provide this outreach to the consumers
they serve. The EPA estimated that States will incur a one-time burden to review these public
education materials of 0.5 hours for CWSs serving 3,300 or fewer people and NTNCWSs serving
50,000 or fewer. The EPA assumed States would require 2 hours to review these materials for
NTNCWSs serving more than 50,000 people. ASDWA agreed with this estimate in their 2024 CoSTS
model (ASDWA, 2024).

Exhibit 4-182 in Section 4.4.5.2 provides the SafeWater LCR model approach including additional cost
inputs that are required to calculate the total costs.

4.4.5.2 Ongoing POU Program Costs

The EPA has developed costs for three ongoing State activities related to POU program oversight as
shown in Exhibit 4-180. The exhibit provides the unit burden for each activity. The assumptions used in
the estimation of the unit burdens follow the exhibit. The last column provides the corresponding
SafeWater LCR model data variable in red/italic font.

Final LCR! Economic Analysis

4-331

October 2024


-------
Exhibit 4-180: State Ongoing POU-Related Burden Estimates

Activity

Unit Burden

SafeWaterLCR Data Variable

d) Review sample invalidation request for
POU monitoring

2 hrs/request

hrs_sampjnvalidJs

e) Review customer notification
certifications

0.33 to 0.5/certification

hrs_cert_custjtJs

f) Review annual POU program report

0.5 hrs/CWS;
0.5 to 4 hr/NTNCWS

hrs_pou_report_ann_revJs

Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system; POU = point-

of-use.

Sources:

d)	& e): "Lead Analytical Burden and Costs_Final.xlsx."

f): "POU lnputs_Final.xlsx."

d) Review sample invalidation request for POU monitoring (hrs_samp_invalidJs). Systems must
sample one-third of locations with POU devices annually. For CWSs, all households must have
POU devices, so sampling must occur at one third of households. The number of households per
system is estimated as the retail population (pws_pop) divided by the total number of
households per system of 2.53 (numb_hh). For NTNCWSs, the number of POUs is equivalent to
the number of taps used for drinking water consumption. See Section 4.3.5.1 for additional
details and values for these inputs. The EPA assumed that 0.6 percent of samples will be
invalidated, consistent with the assumption for other compliance tap sampling
(pp_sampjnvalid). See Section 4.3.2.1.2, activity f) for additional information. The EPA assumed
States will require 2 hours per sample invalidation request based on a 2016 ASDWA
questionnaire. The questionnaire and each State's responses are available in the docket at EPA-
HQ-OW-2022-0801 at www.regulations.gov. Note that ASDWA agreed with the estimate of 2
hours in their 2024 CoSTS model (ASDWA, 2024).

e)	Review customer notification certifications (hrs_cert_cust_lt Js). As discussed in Section 4.4.2.1,
the burden for States to review each system's certification that monitoring results were reported to
customers is 0.33 hours to 0.5 hours and is based on North Carolina and Indiana's estimates for this
review, respectively, in response to a 2016 ASDWA questionnaire. The EPA assumed this review has
the same burden regardless of whether the lead tap sample is collected at a site with or without a
POU device and thus used the same data variable and input. The questionnaire and each State's
responses are available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

f)	Review annual POU program report (hrs_pou_report_ann_revJs). States will incur burden to
review a system's annual report on its POU program that includes monitoring results and may
include corrective actions and routine maintenance activities. The EPA estimated that States will
incur 50 percent of the burden to review the plan as assumed for the system to prepare the plan
(hrs_pou_report_ann_prep_op). See Exhibit 4-181 for the estimated burden to review POU reports
for CWSs and NTNCWSs.

Final LCRI Economic Analysis

4-332

October 2024


-------
Exhibit 4-181: State Burden to Review Annual POU Program Report (hours/system)

System size
(Population Served)

CWSs

NTNCWSs

SafeWater Cost Model Input: hrs_pou_report_ann_revJs

A

B

<3,300

0.5

0.5

3,301-10,000

N/A

1

10,001-50,000

N/A

2

50,001-100,000

N/A

2

100,001-1,000,000

N/A

4

>1,000,000

N/A



Acronyms: CWS = community water system; NTNCWS = non-transient non-community water system.

Source: "POU lnputs_Final.xlsx."

Notes:

A & B: Estimated as 50 percent of system burden to prepare the report (hrs_pou_report_ann_prep_op). See
Section 4.3.5.2, activity m) for details. No NTNCWSs serves more than 1 million people. Thus, the burden for this
size category is 0.

Exhibit 4-182 provides the SafeWater LCR model costing approach for POU-related activities a) through
f) including additional cost inputs that are required to calculate the total costs.

Final LCR! Economic Analysis

4-333

October 2024


-------
Exhibit 4-182: State POU Cost Estimation in SafeWater LCR (by Activity)1,2

State Cost Per Activity for
CWSs

State Cost Per Activity for
NTNCWSs

Conditions for
Cost to Apply to
a State

Frequency
of Activity





Lead 90th
- Range

Other
Conditions



a) Review POU plan

The hours per system multiplied
by the State labor rate.

(hrs_pou_plan_revJs*rateJs)

Cost applies as written to States
for NTNCWSs.

Above AL

States with
model PWSs
installing
POU

devices or
conducting a
POU plan

One time

b) Provide templates for POU outreach materials

The hours per system multiplied
by the State labor rate.

(hrs_temp_pouJs*rateJs)

Cost applies as written to States
for NTNCWSs.

Above AL

States with
model PWSs
installing
POU

devices or
conducting a
POU
devices

One time

c) Review POU public education materials

The hours per system multiplied
by the State labor rate.

(hrs_review_pe_pouJs*rateJs)

Cost applies as written to States
for NTNCWSs.

Above AL

States with

any model

PWSs

installing

POU

devices

One time

d) Review sample invalidation request for POU monitoring

One third of households per
system where the sample is
determined to be invalid
(assume one sample per
household) multiplied by the
hours per sample per system
and the State labor rate.

(((1/3)*(pws_pop/numb_hh)*pp
_sampjn valid) *(hrs_sampjn va
lidJs*rateJs)

One third the number of POU
devices per system where the
sample is determined to be
invalid (assume one sample per
POU device) multiplied by the
hours per sample per system
and the State labor rate.

(((1/3)*numb_pou)*pp_sampJnv
alid)*(hrs_sampjnvalidJs*rateJ
s)

All

States with

any model

PWSs

installing

POU

devices

Once a year

e) Review customer notification certifications

The hours per system multiplied
by the State labor rate.

(hrs_cert_custjtJs*rateJs)

Cost applies as written to States
for NTNCWSs

All

States with

any model

PWSs

installing

POU

devices

Once a year

f) Review annual POU program report

The hours per system multiplied
by the State labor rate.

(hrs_pou_report_ann_revJs*rat
ejs)

Cost applies as written to States
for NTNCWSs

All

States with

any model

PWSs

installing

POU

devices

Once a year

Final LCR! Economic Analysis

4-334

October 2024


-------
Acronyms: AL = action level; CWS = community water system; NTNCWS = non-transient non-community water

system; POU = point-of-use; PWS = public water system.

Notes:

1 The data variables in the exhibit are defined previously in this section with the exception of:

•	numb_pou\ Number of POU devices per PWSs that elects POU option (Section 4.3.5.1).

•	pp_sampjnvalid: Likelihood that a lead sample will be deemed invalid (Section 4.3.2.1.2, activity f)).

•	rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

2As previously discussed in Section 4.3.5.2, in Arkansas, Louisiana, Mississippi, Missouri, and South Carolina the
State pays for the cost of bottles and shipping and conducting the analysis for samples following LSLR (ASDWA,
2020a). Thus, the State will incur the burden and cost for these activities.

4.4.6 State Public Education-Related Costs

States will incur burden to conduct oversight and review activities related to the public education
requirements of the final LCRI. These activities are broadly grouped into those related to: a consumer
notice in response to a single lead sample above 10 ng/L that are independent of a system's lead 90th
percentile level (see Section 4.4.6.1); independent of a system's lead 90th percentile level (see Section
4.4.6.2); conducted in response to a lead ALE (see Section 4.4.6.3); and required in response to multiple
lead ALEs (see Section 4.4.6.4). Exhibit 4-189 in Section 4.4.6.4 provides details on how costs are
calculated for State public education activities a) through o) in Sections 4.4.6.1 through 4.4.6.4 including
additional cost inputs that are required to calculate these costs.

Note that State public education activities associated with the POU program were previously discussion
in Section 4.4.5.

4.4.6.1 Consumer Notice

The EPA has developed State costs related to a system's consumer notice in response to a lead or
copper sample of any level, as shown in Exhibit 4-183. The exhibit provides the unit burden. The
assumptions used in the estimation of the unit burden follow the exhibit. The last column provides the
corresponding SafeWater LCR model data variable in red/italic font.

Exhibit 4-183: PWS Burden for Consumer Notification

Activity

Unit Burden

SafeWater LCR Data Variable

a) Provide templates for consumer
notice materials

0.25 to 0.5 hrs per PWS

hrs_consumer_notice_tempJs

b) Review lead consumer notice
materials

0.5 to 2 hours per PWS

hrs_consumer_notice_revJs

c) Review copy of the consumer notice
and certification

0.5 hrs/PWS per monitoring period

hrs_samp_noticeJs

Acronyms: PWS = public water system.

Source: "Public Education lnputs_CWS_Final.xlsx;" "Public Education lnputs_NTNCWS_Final.xlsx."

a) Provide templates for consumer notice (hrs_consumer_notice_tempJs). The EPA assumed that
States will provide templates to CWSs to develop consumer notice materials and will incur a one-
time burden of 0.25 to 0.5 hours to provide these templates. These estimates are based on
responses to a 2016 ASDWA survey regarding the burden to provide revised sampling instruction

Final LCRI Economic Analysis

4-335

October 2024


-------
templates from North Carolina and Indiana of 0.25 and 0.5 hours, respectively. The questionnaire
and each State's responses are available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

b)	Review lead consumer notice materials (hrs_consumer_notice_revJs). The EPA estimated that
States will incur a one-time burden to review public education material developed by CWSs that is
described in activity c). The EPA assumed systems serving 50,000 or fewer people will use the
template with minor changes. Thus, States will require minimal time to review the public education
materials of 0.5 hours per system. Systems serving more than 50,000 people will adapt the template
and States will require 2 hours per system to review these materials.

c)	Review a copy of the consumer notice and certification (hrs_samp_noticeJs). The EPA assumed
States will incur a burden of 0.5 hours per PWS per monitoring period to review a sample copy of
the consumer notification and a certification that the notification was distributed in a manner that
meets the rule requirements that must be submitted by CWSs and NTNCWSs. This estimate is based
on the ASDWA 2024 CoSTS model, section "Public Education & Notif." (ASDWA, 2024) and is an
increase from that used for the proposed LCRI EA (2023c) of 0.8 hours per system per sampling
period.

4.4.6.2 Activities Regardless of the Lead 90th Percentile Level

The EPA has developed system costs for activities associated with public education requirements under
the final LCRI that are independent of a system's lead 90th percentile status, as provided in Exhibit 4-184.
The exhibit provides the unit burden. The assumptions for the unit burden follow the exhibit. The last
column provides the corresponding SafeWater LCR model data variable in red/italic font.

Exhibit 4-184: State Burden for Public Education Activities that Are Independent of Lead 90th

Percentile Levels

Activity

Unit Burden

SafeWater LCR Data Variable

d) Provide templates for updated CCR
language (one-time)

0.25 to 0.5 hrs/CWS

hrs_temp_ccrJs

e) Provide templates for local and State
health departments lead outreach

0.25 to 0.5 hrs/CWS

hrs_pub_temp_hcJs

f) Review lead outreach materials for
State and local health departments

0.5 to 2 hrs/CWS

hrs_pub_rev_hcJs

g) Participate in joint communication
efforts with local and State health
departments

1 hr/CWS

hrs_hcJs

h) Provide templates for service line
disturbance outreach materials

0.25 to 0.5/CWS

hrs_wtr_tempJs

i) Review public education materials for
service line disturbances

0.5 to 2 hrs/CWS with LSLs

hrs_review_wtr_peJs

j) Provide templates for inventory-

related outreach materials (one-time)

0.25 to 0.5/CWS or
NTNCWS

hrs_pejsl_gen_tempJs

k) Review inventory-related outreach
materials (one-time)

0.5 to 2 hours/CWS or
NTNCWS

hrs_pe_lsl_revJs

Final LCRI Economic Analysis

4-336

October 2024


-------
Activity

Unit Burden

SafeWater LCR Data Variable

1) Provide technical assistance to PWSs
for public education materials

0.375 to 0.75 hours per

CWS per year

$200 to $400 per CWS per

year

hrs_translate_phoneJs
cost_translate_state

m) Review public education certifications

CWSs

1 to 1.5 hrs/CWS
NTNCWSs

0.33 to 0.5 hr/NTNCWS

CWSs

hrs_pe_certify_quarterlyJs
NTNCWSs

hrs_cert_outreach_annualJs

Acronyms: CCR = consumer confidence report; CWS = community water system; LSL = lead service lines; NTNCWS

= non-transient non-community water system.

Sources:

d) - i): "Public Education lnputs_CWS_Final.xlsx."

j) - m): "Public Education lnputs_CWS_Final.xlsx;" "Public Education lnputs_NTNCWS_Final.xlsx."

d)	Provide templates for updated CCR language (hrs_temp_ccrJs). The EPA assumed that States will
provide templates to CWSs to update their CCR language to include the revised mandatory health
effects language and for those with lead, GRR, and unknown service lines to further update their
materials to include information about the system's SLR program and opportunities to replace LSLs
and GRR service lines. In addition, CWSs that have LSLs, GRR, or service lines of unknown material
must also include information on how to access the SL inventory and how to access the results of all
tap sampling in the CCR. The EPA assumed States will incur a one-time burden of 0.25 to 0.5 hours
to provide these templates. These estimates are based on responses to an ASDWA survey regarding
the burden to provide revised sampling instruction templates from North Carolina and Indiana of
0.25 and 0.5 hours, respectively. This estimate is the same as the estimated burden to provide the
sampling template (hrs_rev_sampJs) as discussed in Section 4.4.2.1, activity a). The questionnaire
and each State's responses are available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

e)	Provide templates for local and State health departments lead outreach (hrs_pub_temp_hc Js).

The EPA assumed States will incur a one-time burden to provide templates to CWSs to develop
outreach materials that will be sent to State and local health departments. The EPA assumed States
will incur a one-time burden of 0.25 to 0.5 hours to provide these templates. These estimates are
based on responses to an ASDWA survey regarding the burden to provide revised sampling
instruction templates from North Carolina and Indiana of 0.25 and 0.5 hours, respectively. This
estimate is the same as the estimated burden to provide the sampling template
(hrs_rev_samp_/'sJ.The questionnaire and each State's responses are available in the docket at EPA-
HQ-OW-2022-0801 at www.regulations.gov.

f)	Review lead outreach materials for local and State health departments (hrs_pub_rev_hcJs). The

EPA estimated that States will incur a one-time burden to review public education material
developed by CWSs that is described in activity e). The EPA assumed systems serving 50,000 or
fewer people will use the template with minor changes. Thus, States will require minimal time to
review the public education materials of 0.5 hours per system. Systems serving more than 50,000

Final LCRI Economic Analysis

4-337

October 2024


-------
people will adapt the template and States will require more time to review these materials of 2
hours per system.

g)	Participate in joint communication efforts with local and State health departments (hrs_hcJs).

States will incur annual burden to participate in joint communication efforts with CWSs to provide
lead public education to health departments annually. The EPA assumed that water systems would
have the major role in this activity, but States would provide support to develop joint letters to be
sent to State and local health departments of 1 hour per system. ASDWA agreed with this estimate
in their 2024 CoSTS model (ASDWA, 2024).

h)	Provide templates for service line disturbance outreach materials (hrs_wtr_tempJs). The EPA

assumed that States will provide templates for CWSs with lead, GRR, or unknown service lines to
develop materials when a water system causes disturbances to service lines that can result from
scheduled water-related work. Under the final LCRI, these materials also apply when disturbances
occur during service line inventory investigations. These estimates are based on responses to an
ASDWA survey regarding the burden to provide revised sampling instruction templates from North
Carolina and Indiana of 0.25 and 0.5 hours, respectively.

i)	Review public education materials for service line disturbances (hrs_review_wtr_peJs). The EPA

estimated that States will incur a one-time burden to review public education material developed by
CWSs with lead, GRR, or unknown service lines for delivery during scheduled water-related work or
when disturbances occur during service line inventory investigations. The EPA assumed systems
serving 50,000 or fewer people will use the template with minor changes. Thus, States will require
minimal time to review the public education materials of 0.5 hours per system. Systems serving
more than 50,000 people will adapt the template and States will require 2 hours per system to
review these materials.

j) Provide templates for inventory-related outreach materials (hrs_pejsl_gen_temp Js). CWSs and
NTNCWSs with LSLs must provide notification to customers served by lead, GRR, or unknown service
lines regarding information on the health effects and sources of lead in drinking water (including
SLs), how to have water tested for lead, actions customers can take to reduce exposure to lead, and
information about the opportunities for SLR. The EPA estimates that States will incur a one-time
burden to provide a template for SLR outreach of 0.25 to 0.5 hours. The EPA assumed that the
burden to provide the outreach template would be the same as the burden to provide a template
for updated sampling instructions (hrs_rev_sampJs). The burden estimates are based on North
Carolina and Indiana's response to a 2016 ASDWA survey. The questionnaire and each State's
responses are available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

k) Review inventory-related outreach materials ((hrs_pe_LSL_revJs). States will incur one-time
burden to review the inventory-related outreach materials before they are made publicly available.
The EPA assumed CWSs serving 50,000 or fewer people will use the templates with minor
modification and thus, States will require minimal time to review the outreach materials of 0.5
hours per system. The EPA assumed that systems serving more than 50,000 people will adapt
template and States will require 2 hours per system to review these materials.

I) Provide technical assistance to PWSs for public education materials (hrs_translate_phone Js,
cost_translate_state). As previously discussed in Section 4.3.6.2, under activity r), the final LCRI

Final LCRI Economic Analysis

4-338

October 2024


-------
requires States, as a condition of primacy to provide translation support for public education
materials listed under § 141.85 if the water system is unable to do so. The EPA assumes that States
will provide translation support for small systems serving 10,000 or fewer people. As was true for
CWSs, the EPA assumes that the labor burden and non-labor costs for written or phone translation
support will depend on the CWS's public education category for which the State is providing support
(i.e., notification of service line material for CWS with lead, GRR, or unknown service lines; public
education for CWSs with lead ALEs; or public education material for CWSs with multiple lead ALEs).

Exhibit 4-185 shows the estimated burden for States to provide phone translation support by system
public education category for systems serving 10,000 or fewer people. The EPA used the same
approach to estimate the State burden to provide phone translation support for CWSs serving
10,000 or fewer people as that described for CWSs serving more than 10,000 people under activity
r) in Section 4.3.6.2 with the following exceptions. The EPA assumed States would provide support
for fewer phone calls due to the smaller number of people receiving public education materials in
systems serving 10,000 or fewer people compared to those serving more than 10,000 people. The
EPA estimated that States would annually receive on average one call per year for systems with
lead, GRR or unknown service lines, two calls for each CWS with a lead ALE, and an additional two
calls for each CWS with multiple lead ALEs. The EPA assumed there are no non-labor costs for States
to provide phone translation, which is consistent with the Final CCR3 EA (USEPA, 2024a).

Exhibit 4-186 provides the unit cost for States to provide a written translation for CWSs serving
10,000 or fewer people with lead, GRR, or unknown service lines; a lead ALE; and multiple lead ALEs.
The EPA used the same approach to estimate the State cost to provide written translation support
to CWSs serving 10,000 or fewer people as that described for CWSs serving more than 10,000
people under activity r) in Section 4.3.6.2.

Final LCRI Economic Analysis

4-339

October 2024


-------
Exhibit 4-185: Unit Burden per CWSfor States to Provide Phone Translation by Type of Public Education Material





Public Education for Customers Served

Public Education for All Customers in

Public Education for All Customers in





by Lead, GRR, and Unknown SL

CWSs with a Lead ALE

CWSs with Multiple Lead ALEs

System Size
(Population
Served)

LOE per
Translation

Average
Number of
Phone Calls
per Year

Total Translation Burden
per CWS per Year
(SafeWater LCR Input:
hrs_translate_phone_op)

Average
Number of
Phone Calls
per Year

Total Translation Burden
per CWS per Year
(SafeWater LCR Input:
hrs_translate_phone_op)

Average
Number of
Phone Calls
per Year

Total Translation Burden
per CWS per Year
(SafeWater LCR Input:
hrs_translate_phone_op)



A

B

C=A*B

D

E = A*D

F

G = A *F

<100

0.375

1

0.375

2

0.75

2

0.75

101-500

0.375

1

0.375

2

0.75

2

0.75

501-1,000

0.375

1

0.375

2

0.75

2

0.75

1,001-3,300

0.375

1

0.375

2

0.75

2

0.75

3,301-10,000

0.375

1

0.375

2

0.75

2

0.75

10,001-50,000

0.375

0

0

0

0

0

0

50,001-100,000

0.375

0

0

0

0

0

0

100,001-1,000,000

0.375

0

0

0

0

0

0

>1,000,000

0.375

0

0

0

0

0

0

Acronyms: ALE = action level exceedance; CWS = community water system; GRR = galvanized requiring replacement; SL = service line.

Notes:

General: The EPA assumes that for phone translation services, CWSs serving more than 10,000 people will provide phone translation (See Section 4.3.6.2, activity r)),
whereas the State will provide phone translation for CWSs serving 10,000 or fewer people.

A: This is the average burden per CWS for a State to provide translation call-in support. The EPA assumed that these calls would be a duration of between 15 to 30
minutes, consistent with the assumptions for phone support for small systems translating the CCR as presented in the Final CCR3 EA (USEPA, 2024a).

B: The average number of calls per year the State will support for each system with lead, GRR or unknown service lines is estimated to be one because the education
materials will be delivered to a subset of customers (those with lead, GRR, and unknown service lines) as opposed to all customers.

D: Assumes States will provide phone translation assistance for two calls per year for each system that has a lead ALE.

F: Assumes States will provide phone translation assistance for two additional average number of calls per year for each system with multiple lead ALEs. The two
additional call estimate is based on the EPA's assumption that enhanced outreach will result in more customers potentially becoming aware of the ALE and
requesting translation assistance. The number of calls shown is the increment beyond the number that is estimated for a lead ALE in Column D.

Final LCR! Economic Analysis

4-340

October 2024


-------
Exhibit 4-186: Unit Costs per CWS for States to Provide Written Translations by Type of Public Education Material

System Size
(Population
Served)

Average
Cost per
Translated
PE
Material

Number
of

Languages

Public Education for Customers
Served by Lead, GRR, and Unknown
SL

Public Education for All Customers
in CWSs with a Lead ALE

Public Education for All Customers
in CWSs with Multiple Lead ALEs

Annual
Number of PE
Materials Being
Translated

Total Translation
Cost per CWS per

Year
(SafeWater LCR
Input:
cost_translate_cws)

Annual
Number of
PE Materials

Being
Translated

Total Translation
Cost per CWS per

Year
(SafeWater LCR
Input:
cost_translate_cws)

Annual
Number of
PE Materials

Being
Translated

Total Translation
Cost per CWS Per

Year
(SafeWater LCR
Input:
cost_translate_cws)

A

B

C

D = A*B*C

E

F = A*B*E

G

H = A*B*G

<100

$200

1

1

$200

2

$400

2

$400

101-500

$200

1

1

$200

2

$400

2

$400

501-1,000

$200

1

1

$200

2

$400

2

$400

1,001-3,300

$200

1

1

$200

2

$400

2

$400

3,301-10,000

$200

1

1

$200

2

$400

2

$400

10,001-50,000

$200

1

0

0

0

$400

0

$0

50,001-100,000

$200

2

0

0

0

$800

0

$0

100,001-
1,000,000

$200

2

0

0

0

$800

0

$0

>1,000,000

$200

2

0

0

0

$800

0

$0

Acronyms: ALE = action level exceedance; CWS = community water system; GRR = galvanized requiring replacement; PE = public education; SL = service line.
Notes:

General: The EPA assumes that for written translation services, CWSs serving more than 10,000 people will provide written translation (See Section 4.3.6.2,
activity r)), whereas the State will provide written translation for CWSs serving 10,000 or fewer people.

A: This is the estimated average cost for the State to pay for contractor support to provide written translation service, based on a typical word count of public
education materials of 1,000 multiplied by $0.20 per word for translation services.

B. Assumes States will provide translated materials in one language for systems serving 10,000 or fewer people based on data from the ACS, which provided
the population of metropolitan areas with limited English proficiency.

C: Assumes translation of one document per year per CWS for notification of service line material for customers served by a lead, GRR, or unknown service line.
E: Assumes States will provide one translated document per 6-month period for a total of two documents for each CWS with a lead ALE per year.

G: Assumes States will provide each CWS with multiple ALEs one translated documents per 6-month period for a total of two additional documents per year
per CWS. The number of written translations shown is incremental to the estimate for a lead ALE in Column E.

Final LCR! Economic Analysis

4-341

October 2024


-------
m) Review public education certifications (hrs_pe_certify_quarteriyjs, hrs_cert_outreach_annuaiJs).

States will review each system's certification that they have met their public education and outreach
requirements including any conducted in response to a lead ALE. Under the final LCRI, systems must
resubmit copies of their public education and outreach materials along with the certification. CWSs
have quarterly, semi-annual, and annual public education requirements in response to a lead ALE
(see Section 4.3.6.3 for detailed requirements). Thus, CWSs must report the certification on a
quarterly basis. The EPA estimated a range from 0.33 to 0.5 hours to review public education
certifications under the pre-2021 LCR based on data from North Carolina and Indiana, respectively,
in response to an ASDWA survey about LCR implementation.155 These estimates were multiplied by
0.75 to account for quarters where there is less information to report on the self-certification. Then
the numbers were multiplied by four to account for the quarterly frequency of the self-certification
letter. The EPA assumed that the review of each certification for systems serving 50,000 or fewer
people would require 0.33 hours or 1 hour annually (based on the lower burden reported from
North Carolina) and 0.5 hours/certification or 1.5 hours annually for CWSs serving more than 50,000
people (based on the higher burden reported from Indiana).

NTNCWSs do not have quarterly public education requirements in response to a lead ALE. Instead,
they submit an annual certification only hrs_cert_outreach_annualJs. The EPA assumed States
would spend 0.33 to 0.5 hours per NTNCWS to review their annual certification based on the
estimates provided by North Carolina and Indiana.

The EPA assumed that a system's certification would not only include any outreach conducted in
response to a lead ALE but also include any public education activities described in Section 4.4.6.

4.4.6.3 Public Education Activities in Response to Lead ALE

The EPA has developed State costs for activities associated with public education requirements in
response to a lead ALE as provided in Exhibit 4-187. The exhibit provides the unit burden. The
assumptions for the unit burden follow the exhibit. The last column provides the corresponding
SafeWater LCR model data variable in red/italic font.

Exhibit 4-187: State Public Education Burden in Response to Lead ALE

Activity

Unit Burden

SafeWater LCR Data Variable

n) Provide templates for updated
public education materials for
systems with a lead ALE (one-time)

0.25 to 0.5/CWS or NTNCWS

hrs_ale_lang_tempJs

o) Review revised lead language for
systems with a lead ALE (one-time)

0.5 to 2 hrs/CWS or NTNCWS

hrs_ale_langjs

p) Consult with CWS on other public
education activities in response to
a lead ALE

2 hrs/CWS

hrs_ale_consultJs

Acronyms: ALE = action level exceedance; CWS = community water system; NTNCWS = non-transient non-
community water system.

155 The questionnaire and each state's responses are available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov

Final LCRI Economic Analysis

4-342

October 2024


-------
Sources:

n) & o): "Public Education lnputs_CWS_Final.xlsx"; "Public Education lnputs_NTNCWS_Final.xlsx."
p): "Public Education lnputs_CWS_Final.xlsx."

n) Provide templates for updated public education materials for systems with a lead ALE

(hrs_ale_lang_tempJs). The final LCRI requires systems with a lead ALE to update the mandatory
health effects language and include information on additional steps to reduce lead exposure from
drinking water such as the use of filters. For systems with lead, GRR, or unknown service lines, the
materials must include SLR and service line material identification opportunities, how to obtain a
copy or view the service line inventory and replacement plan, programs to assist with SLR, and the
systems' responsibility to replace their portion of the lead or GRR service line when the property
owner notifies them that the private-side portion is being replaced. The EPA assumed States will
incur a one-time burden of 0.25 to 0.5 hours to provide these templates. These estimates are based
on responses to an ASDWA survey regarding the burden to provide revised sampling instruction
templates from North Carolina and Indiana of 0.25 and 0.5 hours, respectively.

o) Review revised lead language for systems with a lead ALE (hrs_ale_lang_js). States will incur a
one-time burden to review each system's revised public education mandatory language in materials
that are delivered when a system has a lead ALE. The EPA assumed the same burden to review
public education language that is used for other types of public education. Specifically, systems
serving 50,000 or fewer people will use the template with only very minor changes and States will
require 0.5 hours per system for their review. Systems serving more than 50,000 people will adapt
the template and States will require 2 hours per system to review these materials.

p) Consult with CWS on other public education activities in response to a lead ALE

(hrs_ale_consultJs). States will consult with CWSs on other required public education activities
conducted in response to a lead ALE and will incur a burden of 2 hours per CWS. This assumption is
based on the estimate for systems to consult with their State on public education activities used in
the Economic and Supporting Analyses: Short-Term Regulatory Changes to the Lead and Copper Rule
(USEPA, 2007). ASDWA agreed with this estimate in their 2024 CoSTS model (ASDWA, 2024).

Exhibit 4-189 in Section 4.4.6.4 provides details on how costs are calculated for State public education
activities a) through o) including additional cost inputs that are required to calculate these costs.

4.4.6.4 Public Education Activities in Response to Multiple Lead ALEs

The EPA has developed State costs for activities associated with public education requirements in
response to multiple lead ALEs as provided in Exhibit 4-188. The exhibit provides the unit burden. The
assumptions for the unit burden follow the exhibit. The last column provides the corresponding
SafeWater LCR model data variable in red/italic font.

Final LCRI Economic Analysis

4-343

October 2024


-------
Exhibit 4-188: State Public Education Burden in Response to Multiple Lead ALE

Activity

Unit Burden

SafeWaterLCR Data Variable

q) Review plan for making filters
available

2 hours per CWS, 1 hr per
NTNCWS

hrs_temp_filter_plan_devJs

r) Provide templates for systems with
multiple lead ALEs

0.25 to 0.5 hrs per CWS and
NTNCWS with at least 3 lead
ALEs in 5 years

hrs_temp_persist_aleJs

s) Review outreach materials provided
by systems with multiple lead ALEs

0.5 to 2 hrs per CWS and
NTNCWS with at least 3 lead
ALEs in 5 years

hrs_review_pe_persist_aleJs

t) Consult on filter program for systems
with multiple lead ALEs (one-time)

2 to 8 hrs per CWS and
NTNCWS

hrs_consult_temp_pouJs_

Acronyms: ALE = action level exceedance; CWS = community water system; NTNCWS = non-transient non-
community water system.

Sources: "Public Education lnputs_CWS_Final.xlsx"; "Public Education lnputs_NTNCWS_Final.xlsx."

q) Review plan for making filters available (hrs_temp_filter_plan_devJs). Under the final LCRI, State
will incur a one-time burden to review filter plans that CWSs and NTNCWSs must develop after they
have two lead ALEs in a five-year period. As previously noted in 4.3.6.4, under the proposed LCRI
this plan was due after three lead ALEs in a five-year period. The EPA assumed that States would
incur half the burden required for systems to develop the plan, which is equivalent to 2 hours per
CWS plan, and 1 hour per NTNCWSs plan.

r) Provide templates for systems with multiple lead ALEs (hrs_temp_persist_aleJs). The final LCRI
requires CWSs and NTNCWSs that have at least three lead ALEs in a five-year period (i.e., have
multiple lead ALEs) to provide enhanced outreach. States will incur one-time burden of 0.25 to 0.5
hours to provide templates that will assist systems in developing their outreach materials, which is
the same burden used to provide templates for other public education and outreach materials.

s) Review outreach materials provided by systems with multiple lead ALEs (hrs_temp_persist_aleJs).

States will incur one-time burden to review the outreach materials developed by water systems with
multiple lead ALEs. The EPA assumed the same burden to review public education language that is
used for other types of public education. Specifically, systems serving 50,000 or fewer people will
use the template with only very minor changes and States will require 0.5 hours per system for their

Final LCRI Economic Analysis

4-344

October 2024


-------
review. Systems serving more than 50,000 people will modify the template to better fit the systems
needs and States will require 2 hours per system to review these materials.

t) Consult on filter program for systems with multiple ALEs (hrs_consult_temp_pouJs). States will
incur a one-time burden to consult with the CWSs and NTNCWSs on specific requirements for its
filter program. The EPA estimated systems serving 3,300 or fewer people will require 2 hours, those
serving 3,301 to 10,000 people will require 6 hours, and those serving more than 10,000 people will
require 8 hours.

Exhibit 4-189 provides details on how total costs for the final LCRI public education requirements are
calculated for activities a) through t) including additional cost inputs that are required to calculate the
total costs.

Exhibit 4-189: State Lead Public Education Cost Estimation in SafeWater LCR (by Activity)1,2

State Cost Per Activity for CWSs

State Cost Per
Activity for
NTNCWSs

Conditions for Cost to
Apply to a State

Frequency
of Activity





Lead 90th
- Range

Other Conditions3



a) Provide templates for consumer notice materials

The hours per system multiplied by
the State labor rate.

(hrs_consumer_notice_tempJs
*rateJs)

Cost applies as
written to States for
NTNCWSs.

All

All States

One time

b) Review lead consumer notice materials

The hours per system multiplied by
the State labor rate.

(hrs_consumer_notice_revJs
*rateJs)

Cost applies as
written to States for
NTNCWSs.

All

All States

One time

c) Review copy of the consumer notice and certification

The hours per system multiplied by
the State labor rate.

(hrs_samp_noticeJs*rateJs)

Cost applies as
written to States for
NTNCWSs.

All

All States

Once per
event

d) Provide templates for updated CCR language

The hours per system multiplied by
the State labor rate.

(hrsjemp_ccrJs*rateJs)

Cost does not
apply to States for
NTNCWSs.

All

All States

One time

e) Provide templates for local and State health departments lead outreach

The hours per system multiplied by
the State labor rate.

(hrs_pubjemp_hcJs*rateJs)

Cost does not
apply to States for
NTNCWSs.

All

All States

One time

f) Review lead outreach materials for local and State health departments

Final LCRI Economic Analysis

4-345

October 2024


-------
State Cost Per Activity for CWSs

State Cost Per
Activity for
NTNCWSs

Conditions for Cost to
Apply to a State

Frequency
of Activity





Lead 90th
- Range

Other Conditions3



The hours per system multiplied by
the State labor rate.

(hrs_pub_rev_hcJs*rateJs)

Cost does not
apply to States for
NTNCWSs.

All

All States

One time

g) Participate in joint communication efforts with local and State health departments

The hours per system multiplied by
the State labor rate.

(hrs_hcJs*rateJs)

Cost does not
apply to States for
NTNCWSs.

All

All States

Once per
year

h) Provide templates for service line disturbance outreach materials

The hours per system multiplied by
the State labor rate.

(.hrs_wtr_tempJs *rateJs)

Cost does not
apply States for
NTNCWSs.

All

States with any model
PWSs with service lines
of lead or unknown
content3

One time

i) Review public education materials for service line disturbances

The hours per system multiplied by
the State labor rate.

(hrs_review_wtr_peJs*rateJs)

Cost does not
apply States for
NTNCWSs.

All

States with any model
PWSs with service lines
of lead or unknown
content3

One time

j) Provide templates for inventory-related outreach materials

The hours per system multiplied by
the State labor rate.

(hrs_pejsl_genjempJs*rateJs)

Cost applies as
written to States for
NTNCWSs.

All

States with any model
PWSs with service lines
of lead or unknown
content3

One Time

k) Review inventory-related outreach materials

The hours per system multiplied by
the State labor rate.

(hrs_pejsl_revJs*rateJs)

Cost applies as
written to States for
NTNCWSs.

All

States with any model
PWSs with service lines
of lead or unknown
content3

One Time

1) Provide translation technical assistance to PWS for public education materials

The total hours per system multiplied
by the system labor rate, plus the
material cost.

(hrsJranslate_phoneJs*rateJs)+cos
t translate state



Below AL

States providing model
PWSs translation



The total hours per system multiplied
by the system labor rate, plus the
material cost.

(hrsJranslate_ale_phoneJs*rateJs)
+cost translate ale state

Cost does not
apply to
NTNCWSs

Above AL

services either by
telephone or written

pjranslation
p translation phone
1-

Once a
year

The total hours per system multiplied
by the system labor rate, plus the
material cost.

(hrsJranslate_ale_phoneJs*rateJs)
+cost translate ale state



Multiple
ALEs

pJranslation_phone_cw
s



m) Review public education certifications

Final LCRI Economic Analysis

4-346

October 2024


-------
State Cost Per Activity for CWSs

State Cost Per
Activity for
NTNCWSs

Conditions for Cost to
Apply to a State

Frequency
of Activity





Lead 90th
- Range

Other Conditions3



The hours per system multiplied by
the State labor rate.

(hrs_pe_certify_quarterlyJs*rateJs)

The hours per
system multiplied
by the State labor
rate.

(hrs_cert_outreach
annual js*rate js)

Above AL

All States

Once per
year4

n) Provide templates for updated public education materials for systems with a lead ALE5

The hours per system multiplied by
the State labor rate.

(hrs_ale_lang_tempJs*rateJs)

Cost applies as
written to States for
NTNCWSs.

Above AL

All States

One time

o) Review revised lead language for systems with a lead ALE5

The hours per system multiplied by
the State labor rate.

(hrs_alejangjs*rateJs)

Cost applies as
written to States for
NTNCWSs.

Above AL

All States

One time

p) Consult with CWS on other public education activities in response to lead ALE5

The hours per system multiplied by
the State labor rate.

(hrs ale consult js*rate js)

Cost does not
apply to States for
NTNCWSs.

Above AL

All States

Once a
year

q) Review plan for making filters available

The hours per system multiplied by
the State labor rate.

(hrsjemp_filter_plan_devJs*rateJs)

Cost applies as
written to States for
NTNCWSs.

Above AL

States with any model
PWSs with at least two
lead ALEs

One time

r) Provide templates for systems with multiple lead ALEs

The hours per system multiplied by
the State labor rate.

(hrsjemp_persist_aleJs*rateJs)

Cost applies as
written to States for
NTNCWSs.

Above AL

States with any model
PWSs with multiple
ALEs

One time

s) Review outreach materials provided by systems with multiple lead ALEs

The hours per system multiplied by
the State labor rate.

(hrs_review_pe_persist_aleJs*rateJ
s)

Cost applies as
written to States for
NTNCWSs.

Above AL

States with any model
PWSs with multiple
ALEs

One time

t) Consult on filter program for systems with multiple ALEs

The hours per system multiplied by
the State labor rate.

(hrs_consultJemp_pouJs*rateJs)

Cost applies as
written to States for
NTNCWSs.

Above AL

States with any model
PWSs with multiple
ALEs

One time

Acronyms: AL = action level; ALE = action level exceedance; CCR = consumer confidence report; CWS = community
water system; LSL = lead service line; NTNCWS = non-transient non-community water system; PWS = public water
system.

Final LCRI Economic Analysis

4-347

October 2024


-------
Notes:

1	State oversight burden and costs for systems with LSLs with the exception of those associated with service line
disturbances and implementing the POU program are included in Sections 4.4.4 and 4.4.5.1, respectively.

2	The data variables in the exhibit are defined previously in this section with the exception of:

• rateJs: State hourly labor rate (Chapter 3, Section 3.3.11.2).

3	PWSs with lead content or unknown lines are identified using the data variables and approach described in
Chapter 3, Section 3.3.4. PWSs with multiple ALEs are described in Chapter 3, Section 3.3.5.2.

4States will review certifications quarterly for CWSs that are providing public education in response to a lead ALE.
For modeling purposes, the State burden is estimated on an annual basis.

5 States can discontinue these activities when the system no longer has a lead ALE for one monitoring period.
4.4.7 Summary of Estimated State Costs

The estimated monetized incremental annual State costs range from $27.7 million to $25.8 million in
2022 dollars at a 2 percent discount rate, under the low and high cost scenarios respectively (see Exhibit
4-1).

4.5 Costs and Ecological Impacts Associated with Additional Phosphate Usage

Adding phosphate to lead content piping creates a protective inner coating that can inhibit lead
leaching. However, once phosphate is added to the PWS, some of this incremental loading remains in
the water stream as it flows into WWTPs downstream. This generates treatment costs for certain
WWTPs. In addition, at those locations where treatment does not occur, water with elevated
phosphorus concentrations may discharge to water bodies and induce certain ecological impacts. Due to
the fact that many water systems operate both the wastewater and drinking water systems, the EPA is
evaluating the costs of additional phosphate usage for informational purposes. These costs to WWTPs
and the downstream ecological impacts are not "likely to occur solely as a result of compliance" with the
final LCRI, and therefore are not costs considered as part of the HRRCA under SDWA, section
1412(b)(3)(C)(i)(lll).

4.5.1 Estimating the Costs of Increased Phosphorus Loadings

4.5.1.1 Incremental phosphorus loading to wastewater treatment plants

When PWSs add orthophosphate to their finished water for corrosion control purposes, some portion of
the orthophosphate added will reach downstream WWTPs. To estimate the potential fate of the
orthophosphate added at PWSs, the EPA developed a conceptual mass balance model, shown in Exhibit
4-190. The EPA applied this conceptual model to estimate the increase in loading at WWTPs (G in Exhibit
4-190), given an initial loading from corrosion control at water treatment plants (A in Exhibit 4-190). In
applying the model, the EPA used the assumptions shown in Exhibit 4-191 regarding the other sources
and losses of phosphorus (B through F in Exhibit 4-190).

Final LCRI Economic Analysis

4-348

October 2024


-------
Exhibit 4-190: Phosphorus Mass Balance Conceptual Model

Users

Drinking
Water
Treatment
Plant

Distribution System

Wastewater Treatment
Plant

Exhibit 4-191: Summary of Assumptions Used in Estimating Phosphorus Loading Increase

Phosphorus Source or Loss

Assumptions Used

Loss Due to Incorporation in
Distribution System Scale (B)

Assumed 0 percent based on data that P accounted for very little of the total
mass of the scale formed during pipe loop testing (Benjamin et al., 1990);
this assumption results in a conservative estimate of the incremental loading
[i.e., erring on the side of greater loading).

Loss to Distribution System
Leaks and Breaks (C)

Average = 57.42 gpd/connection; Warm Climate = 53.64 gpd/connection;
Cold Climate = 78.52 gpd/connection (Chastain-Howley et al., 2013).1

Loss to Outdoor or Other Uses
(D)

Average = 30 percent (USEPA, 2008b); Warm Climate = 67%; Cold Climate =
22% (Mayer et al., 1999).2

Baseline Residential Loading (E)

Not used; relevant only to calculating total loading, not incremental loading.

Loss to Sewer System Leaks and
Overflows (F)

Assumed 0 percent based on an estimate that that losses due to sewer
overflows and misconnections are relatively small (Comber et al., 2013); this
assumption results in a conservative estimate of the incremental loading
(i.e., erring on the side of greater loading).

Acronyms: P = phosphorus; gpd = gallons per day.

Notes:

1	With respect to temperature, systems were classified as one of two categories depending on whether their
location had an average annual temperature above or below 50°F (10°C).

2	Warm climate value reflects the upper bound of outdoor use reported for cities in hot climates; cold climate
value reflects the lower bound of outdoor use reported for cities in a cooler, wetter climates.

Specifically, the EPA adapted the conceptual mass balance model and the assumptions, shown in Exhibit
4-190 and Exhibit 4-191, respectively into Equation 1, and applied this equation in SafeWater LCR model
to estimate the incremental WWTP loading resulting from adding upstream orthophosphate at each
affected drinking water treatment plant.

Final LCRI Economic Analysis

4-349

October 2024


-------
Equation 1:

^incremental = 0-775 x Average Flow x P04 Dose — 0.061 x Connections x P04 Dose

Where:

Pincrementai = incremental WWTP loading in pounds per year measured as phosphorus
Average Flow = drinking water system average flow in thousand gallons per year
P04 Dose = incremental orthophosphate dosage in milligrams per liter (mg/L) as P04
Connections = drinking water system number of connections

The equation above incorporates the colder climate assumptions from Chastain-Howley et al. (2013) and
Mayer et al. (1999). Colder climates have greater losses to leaks and break, but a lower percentage of
losses of outdoor use. Warmer climates show the opposite pattern. The equation uses the colder
climate assumptions because, in combination, these assumptions result in an overall larger estimated
loading increases than the warm climate or average climate assumptions.156

4.5.1.2 Incremental phosphorus removal costs at wastewater treatment plants

WWTPs could incur costs because of upstream orthophosphate addition if they have permit discharge
limits for phosphorus parameters. Exhibit 4-192 shows data from the EPA's national pollutant discharge
elimination system (NPDES) on the status of WWTPs with respect to permit limits for phosphorus.

Exhibit 4-192: WWTP Status with Respect to Phosphorus Discharge Permit Limits

Year

Total Number of WWTPs

Nu mber of WWTPs with
Phosphorus Permit Limits

Percentage of WWTPs with
Phosphorus Permit Limits

2007

14,764

1,446

9.8%

2024

16,147

2,809

17.4%

Acronyms: WWTPs = wastewater treatment plants

Source: Based on national data from the EPA's Discharge Monitoring Report (DMR) "Water Pollutant Loading Tool"
using search criteria limiting results to the phosphorus parameter group and WWTPs only (USEPA, 2024b). Note
DMR Water Pollutant Loading Tool data is only available from 2007 onward.

As shown in Exhibit 4-192, the percentage of WWTPs with phosphorus limits has increased over time.
From 2007 to 2024, in annual percentage rate terms, the growth rate in the percentage of WWTPs with
phosphorus limits is 3.4 percent, calculated as follows:

17.4%\(1/17)

9.8% ) ~ 1

The EPA assumed this increase would continue as States transition from narrative to numerical nutrient
criteria and set numeric permits limits, especially for impaired waters. The EPA applied the growth rate
observed from 2007 to 2024 to estimate the anticipated percentage of WWTPs with phosphorus limits
in future years. The EPA estimated the percentage anticipated for a given year using Equation 2. The EPA

156 The derivation file "POTW P Loading Equations.xlsx" shows the detailed derivation of Equation 1 from the
assumptions identified in Exhibit 4-191, including conversion factors.

Final LCRI Economic Analysis

4-350

October 2024


-------
calculated the estimated percentage for each year of the analysis and applied these percentages in the
SafeWater LCR model as discussed below.

Equation 2:

%Y = %2024 x (1 + Rate)(r-2024)

Where:

Y = specific year being estimated

%Y = percentage of WWTPs anticipated to have phosphorus discharge limits in year Y
%2024 = percentage of WWTPs with phosphorus discharge limits in 2024, or 17.4%

Rate = historical annual percent growth rate observed from 2007 to 2024, or 3.4%

Note that Equation 2 results in an estimated 61 percent of WWTPs with phosphorus discharge limits
after 35 years.157 Applied as the percentage of WWTPs that need to take treatment actions, this
estimate is likely conservative particularly given the potential availability of alternative compliance
mechanisms, such as, individual facility variance and nutrient trading programs.

The specific actions a WWTP might need to take to maintain compliance with its NPDES phosphorus
limit will depend on the type of treatment present at the WWTP and the corresponding phosphorus
removal provided (if any). Assuming a phosphorus permit limit of 1 mg/L (as Total P) - the most
common limit observed in the source data for Exhibit 4-192 - it is likely that most of the WWTPs that
already have phosphorus limits have some type of treatment to achieve the limit. Technologies for
phosphorus removal from wastewater include the following (Jiang et al., 2004; USEPA, 2013; Rodgers,
2014; USEPA, 2021):

•	enhanced biological processes (e.g., those that rely on phosphate accumulating organisms);

•	chemical precipitation;

•	adsorptive media;

•	membrane processes;

•	various emerging or innovative technologies; and

•	treatment trains that combine one or more of the above.

Some treatment processes can accommodate incremental increases in influent loading and still maintain
their removal efficiency. Examples include enhanced biological processes (assuming they are not limited
by influent biological oxygen demand) and membrane processes. Such processes might not require
significant adjustment to maintain their existing phosphorus removal efficiency, given an incremental
increase.

Other treatment processes can require modification to their design or operation to maintain their
removal efficiency in the face of an influent loading increase. A specific example is chemical

157 The EPA estimated the percent of WWTPs with phosphorus discharge limits from 2027, the EPA's expected date
of compliance, to 2061.

Final LCR! Economic Analysis

4-351

October 2024


-------
precipitation, in which the dosage of chemical(s) added {e.g., ferric chloride, alum) is directly
proportional to the influent phosphorus concentration. If influent loading increases, treatment trains
relying on chemical precipitation would need to add more chemicals to maintain their efficiency of
phosphorus removal.

Data are not available to identify the specific WWTPs that might be affected by increased
orthophosphate loading or the burden associated with the phosphorus removal technologies in place at
these WWTPs. Therefore, the EPA estimated costs by assuming that, on average, these costs would be
similar to costs for a WWTP that uses ferric chloride for chemical precipitation to maintain 90 to 98
percent removal, and that has sufficient existing capacity to accommodate the increase in phosphorus
loading.

Specifically, the EPA used the assumptions shown in Exhibit 4-193 to derive a unit cost of $5.44 in 2022
dollars per pound of phosphorus for removing incremental phosphorus. This unit cost includes the cost
of additional chemical consumption and the operating cost of additional sludge processing and
disposal.158 This unit cost will overestimate costs for WWTPs that do not require significant operational
adjustment to maintain their existing phosphorus removal efficiency. That would include, for example,
WWTPs using enhanced biological processes that are not limited by biological oxygen demand. The unit
cost, however, assumes that existing chemical feed, solids separation, and sludge management
equipment has sufficient capacity. Therefore, it will underestimate costs for WWTPs that need to
expand their treatment process capacity or install additional treatment to handle the increased loading.

Exhibit 4-193: Summary of Assumptions Used in Estimating Phosphorus Removal Unit Cost

Assumption

Value Used

Sources

Unit cost for ferric
chloride

$0.11 per pound of
bulk solution

Average of vendor bids in Fredrick County (2014) and Bi-state
Commission (2014)1, escalated to 2016 dollars using Bureau of
Labor Statistics Producer Price Index for industrial chemicals

Ferric chloride

solution

concentration

40%

Consistent with vendor bids in Frederick County (2014) and Bi-
state Commission (2014)

Ferric chloride
solution bulk density

11.85 pounds per
gallon

Consistent with 40% ferric chloride solution concentration;
used to convert vendor bids in Bi-state Commission (2014)

Molar ratio required
for phosphorus
removal

2 moles iron per mole
of phosphorus

A molar ratio of 1.5 to 2:1 (iron-to-phosphorus) can achieve an
80 to 98% reduction in soluble phosphorus per USEPA (2010)

Unit cost for sludge
processing and
disposal

$336 per dry ton

Average of actual sludge management costs reported in
Stamford Water Pollution Control Authority (WPCA) (2013),
City of Seabrook (2016), Sloan et al. (2008), Center for Rural
Pennsylvania (2007), escalated to 2016 dollars using the
consumer price index

Sludge production
factor

10 grams per gram of
phosphorus removed

USEPA (2010)

158 The derivation file "POTW P Loading Equations.xlsx" shows the detailed derivation of this unit costs from the
assumptions identified in Exhibit 4-191 including conversion factors.

Final LCRI Economic Analysis

4-352

October 2024


-------
Note:1 The sources used are tabulations of bids received by water utilities from vendors bidding on contracts to
provide an annual supply of treatment chemicals. Using these sources ensures that the cost estimate reflects
prices charged to utility customers for typical quantities of ferric chloride solution with concentrations and other
specifications appropriate to the water treatment application the EPA is modeling.

Finally, the costs a WWTP could incur depend on the magnitude of the loading increase relative to the
specific WWTP's effluent permit limit. WWTPs whose current discharge concentrations are closer to
their limit are more likely to have to take action. However, WWTPs whose current concentrations are
well below their limit could incur costs if, for example:

1.	They are currently achieving their limit using a P removal technology.

2.	The P removal provided by that technology is significant.

3.	The P removal achieved by technology is sensitive to incremental P loading increases (e.g.,
chemical phosphorus removal).

Furthermore, future phosphorus limits could be more stringent than existing limits.

Therefore, the EPA assumed that any WWTP with a discharge limit for phosphorus parameters could
incur costs. Accordingly, in calculating costs in the SafeWater LCR model, the EPA used the anticipated
percentage of WWTPs with phosphorus discharge limits, calculated as shown in Equation 2, as the
likelihood that incremental orthophosphate loading from a drinking water system would reach a WWTP
with a limit. The EPA combined this likelihood and the unit cost estimated above with incremental
phosphorus loading to calculate incremental costs to WWTPs for each year of the analysis period. This
calculation is equivalent to that shown in Equation 3.

Incremental Costy = incremental cost to WWTPs in year Y

%Y = percentage of WWTPs anticipated to have phosphorus discharge limits in year Y, calculated
as shown in Equation 2

Unit Cost = incremental cost of treatment per pound of phosphorus, or $5.44 per pound
IPincrementai = incremental WWTP loading in pounds per year measured as total phosphorous
from all affected drinking water treatment plants

As shown in Exhibit 4-1, the estimated incremental annualized cost that WWTPs will incur to remove
additional phosphorous associated with the final LCRI ranges from $120,000 to $300,000 in 2022 dollars
at a 2 percent discount rate.

4.5.2 Ecological Impacts of Phosphorus Loadings

The ecological impacts of increased phosphorous loadings are highly localized: total phosphorus
loadings will depend on the amount and timing of the releases, characteristics of the receiving water

Final LCRI Economic Analysis	4-353	October 2024

Equation 3:

Incremental CostY = %y x Unit Cost x Pincremental

Where:


-------
body, effluent discharge rate, existing total phosphorus levels, and weather and climate conditions.
Unfortunately, detailed spatially explicit information on effluents and on receiving water bodies does
not exist in a form suitable for this analysis. Rather, to evaluate the potential ecological impacts of the
rule, the EPA developed approximate, national-level total phosphorous loading estimates, and evaluated
the significance of the loadings compared to other phosphorous sources in the terrestrial ecosystem.

4.5.2.1 Incremental total phosphorus loadings in water bodies

The SafeWater LCR model, using Equation 1 described above, estimated the total incremental
phosphorus loadings to reach WWTPs under the final LCRI. Exhibit 4-194 provides the estimated total
and increase in phosphorus loadings nationally for selected years after the LCRI goes into effect under
the low scenario. Exhibit 4-195 provides the same information for the high scenario. If the LCRI were not
to go into effect, by Year 5, under the 2021 LCRR, PWSs would have begun compliance with the 2021
LCRR and CCT treatment would be increased. This is why the incremental increase in loadings associated
with the LCRI are negative in Year 5.

Exhibit 4-194: Estimated Nationwide Annual Phosphorus Reaching WWTPs after
Implementation of the LCRI under Low Cost Scenario

2021LCRR
Final LCRI

Increase Under 2021 LCRR
Increase Under Final LCRI
Incremental Increase over 2021 LCRR

Thousands of Pounds of Phosphorous

YearO

Year 5

Year 15

Year 25

Year 35

6,279

6,819

6,970

7,113

7,255



6,279

7,512

7,680

7,775



540

691

835

976



-

1,234

1,401

1,497



(540)

542

567

520

Acronyms: LCRR = Lead and Copper Rule Revisions; LCRI = Lead and Copper Rule Improvements; WWTPs =
wastewater treatment plants.

Exhibit 4-195: Estimated Nationwide Annual Phosphorus Reaching WWTPs after
Implementation of the LCRI under High Cost Scenario

2021LCRR
Final LCRI

Increase Under 2021 LCRR
Increase Under Final LCRI
Incremental Increase over 2021 LCRR

Thousands of Pounds of Phosphorous

YearO

Year 5

Year 15

Year 25

Year 35

6,173

7,275

7,522

7,769

7,995



6,173

8,178

8,450

8,621



1,102

1,349

1,596

1,822



-

2,005

2,277

2,448



(1,102)

656

681

626

Acronyms: LCRR = Lead and Copper Rule Revisions; LCRI = Lead and Copper Rule Improvements; WWTPs =
wastewater treatment plants.

The EPA then adjusted these values for the expected treatment of influent at WWTPs. Based on the
Clean Watersheds Needs Survey, about 50 percent of facilities (36 percent of flow) have secondary
water treatment and 34 percent of facilities (57 percent of flow) have greater than secondary treatment
(USEPA, 2012a) that will reduce the amount of phosphorus reaching waterbodies. Estimates suggest
that secondary treatment may remove 20 to 75 percent of total phosphorus and greater than secondary
treatment may remove 90 to 95 percent (Metcalf and Eddy, 2003; Grady, 2011; USEPA, 2015b) of the

Final LCRI Economic Analysis

4-354

October 2024


-------
phosphorus reaching waterbodies. Thus, the EPA conservatively estimates that 36 percent of flow will
experience a 20 percent reduction in total phosphorus and 57 percent of the flow will experience a 90
percent reduction of total phosphorus, generating a flow-weighted average reduction in total
phosphorus levels of about 58.5 percent. Using these assumptions, the EPA estimated the amount of
total phosphorus that is expected to enter receiving waterways nationally as a result of the 2021 LCRR
and final LCRI under the low cost assumptions (Exhibit 4-196) and high cost assumptions (Exhibit 4-197)

Exhibit 4-196: Estimated Nationwide Annual Phosphorus Reaching Waterbodies after
Implementation of the LCRI under Low Cost Scenario

Thousands of Pounds of Phosphorous



YearO

Year 5

Year 15

Year 25

Year 35

2021LCRR

2,605

2,829

2,892

2,951

3,010

Final LCRI



2,605

3,116

3,186

3,226

Increase Under 2021 LCRR



224

287

346

405

Increase Under Final LCRI



-

512

581

621

Incremental Increase over 2021 LCRR



(224)

225

235

216

Acronyms: LCRR = Lead and Copper Rule Revisions; LCRI = Lead and Copper Rule Improvements.

Exhibit 4-197: Estimated Nationwide Annual Phosphorus Reaching Waterbodies after
Implementation of the LCRI under High Cost Scenario

Thousands of Pounds of Phosphorous



Year 0

Year 5

Year 15

Year 25

Year 35

2021LCRR

2,561

3,018

3,120

3,223

3,317

Final LCRI



2,561

3,393

3,505

3,576

Increase Under 2021 LCRR



457

559

662

756

Increase Under Final LCRI



-

832

945

1,015

Incremental Increase over 2021 LCRR



(457)

272

282

260

Acronyms: LCRR = Lead and Copper Rule Revisions; LCRI = Lead and Copper Rule Improvements.

To put these phosphorus loadings in context, estimates from the United States Geological Survey (USGS)
SPAtially Referenced Regression On Watershed attributes (SPARROW) model suggest that
anthropogenic sources deposit roughly 750 million pounds of total phosphorus per year (USEPA, 2019b).
Under the high cost scenario, this additional phosphorous loading is small, about 0.03 percent (260,000/
750,000,000) of the total phosphorous load deposited annually from all other anthropogenic sources.
Note that the EPA model assumes that once CCT is installed or re-optimized phosphate use remains
constant over the remainder of the period of analysis. Because most CCT implementation is carried out
prior to complete LSL removal and the model does not allow for reductions in the use of phosphate after
systems remove all their lead content service lines, the EPA's CCT cost estimates and phosphorus
loading estimates to both WWTPs and receiving waterbodies may be overestimated.

National average load impacts may obscure significant localized ecological impacts. The existing data do
not allow an assessment as to whether this incremental load will induce ecological impacts in particular
areas; however, localized impacts may occur in water bodies without restrictions on phosphate loadings,
or in locations with existing elevated phosphate levels. The next section describes potential ecological
impacts that could occur in receiving water bodies.

Final LCRI Economic Analysis

4-355

October 2024


-------
4.5.2.2 Ecological impacts of potential increases in phosphate loadings

Aquatic organisms rely on some amount of essential nutrients, including nitrogen and phosphorous, for
growth and survival. In many aquatic ecosystems, phosphorous is the limiting nutrient (USEPA, 2016a).
As a limiting nutrient, phosphorous frequently controls the growth rate of phytoplankton, bacteria, and
algae (USEPA, 2016a). Discharging excess phosphorous into waterbodies can therefore stimulate excess
plant and algae growth and, under certain circumstances, create undesirable ecological impacts.
Phosphorous in the environment can persist longer periods of time relative to nitrogen. Sediment-
bound phosphorous can persist unchanged and, when re-suspended back to the water column, can pose
renewed threats. Localized conditions will enhance or dissipate phosphorous problems.

Nutrient pollution causes eutrophication—that is, excessive plant and algae growth—in lakes, reservoirs,
streams, and estuaries throughout the United States. According to the EPA's 2012 National Lakes
Assessment, 40 percent of lakes in the United States have excess phosphorus (USEPA, 2016a). The EPA's
2008-2009 National Rivers and Streams Assessment found that 40 percent of river and stream miles
have nutrient pollution (USEPA, 2016b). The excessive growth of algae and phytoplankton can reduce
water clarity and light penetration, reducing the production of benthic plant growth (Lehtiniemi et al.,
2005). The reduction of benthic plants alters or destroys habitat that may be required or utilized by
other organisms, such as fish, benthic macroinvertebrates, amphibians, and more. Predators reliant
upon vegetation may have reduced predation success (Lehtiniemi et al., 2005). The excessive growth of
algae and phytoplankton eventually leads to mass mortality events, in which these microorganisms die
off rapidly. The decomposition of the additional biomass consumes oxygen in the water, creating
hypoxia, a state of low dissolved oxygen. Sufficiently low to no oxygen states can create dead zones, or
areas in the water where aquatic life cannot survive. Studies indicate that eutrophication can decrease
aquatic diversity for this reason (e.g., Dodds et al., 2009).

Eutrophication may also stimulate the growth of harmful algal blooms (HABs), or over-abundant algae or
cyanobacteria populations. Algal blooms can seriously harm the aquatic ecosystem by blocking sunlight
and creating diurnal swings in oxygen levels as a result of overnight respiration. Such conditions can
starve and deplete aquatic species. In addition, rapid photosynthesis may consume dissolved inorganic
carbon and elevate pH (Chislock et al., 2013). Altered pH levels in aquatic ecosystems can impact the
chemosensory abilities of aquatic species, potentially altering their behaviors and interactions with
other species (Turner & Chislock, 2010). Certain types of phosphorous-fueled cyanobacterial blooms
may produce toxins to both humans and aquatic life. These toxins include microcystins (liver toxins) and
neurotoxins. This issue is particularly prevalent in lakes or other slow-flowing water bodies. For
additional information on the human health risks when HABs result in drinking water cyanotoxin
exposure and the management tools available to PWSs see The EPA website:

https://www.epa.gov/ground-water-and-drinking-water/managing-cvanotoxins-public-drinking-water-
systems. HAB events have also directly or indirectly contributed to fish kill events by causing the
absorption or ingestion of toxins, or by creating conditions of limited sunlight and oxygen (Glibert et al.,
2005). In addition to lethal impacts on aquatic organisms, toxins produced by HABs can harm terrestrial
wildlife and livestock that are exposed to toxins in sufficient levels (Backer, 2002; Chislock et al., 2013).
Toxins are capable of bioaccumulating and transferring to higher trophic levels, killing birds, mammals,
and other wildlife that consume prey contaminated with toxins (Su et al., 2020). In marine
environments, HABs can impact or destabilize cultivated stocks of finfish or shellfish, potentially destroy
benthic habitat, and contribute to marine fish kills (Cloern, 2001). Overall, the effects of eutrophication

Final LCRI Economic Analysis

4-356

October 2024


-------
and HABs can alter higher trophic level communities (Jeppesen et al., 1997). Changes to community
composition can potentially degrade emergent ecosystem properties and impact overall ecosystem
function.

Finally, an increase in phosphorus loadings can lead to significant economic impacts and undesirable
aesthetic impacts. Research estimates significant economic costs of eutrophication, including recreation
and angling costs and property value costs (Dodds et al., 2009). Aesthetic impacts such as reduced water
clarity and an increase in foul-smelling odors may also arise, making water unsuitable for recreational
activities such as swimming, boating, and fishing (Dodds et al., 2009). Phosphorus additions can also
reduce the non-use (e.g., option, existence or bequest value) value of the water resource.

The seasonal Gulf of Mexico dead zone demonstrates a powerful example of the negative ecological
impacts that result from excessive nutrients. The Gulf of Mexico dead zone is the second largest in the
world, and results from the inflow of nutrients from the Mississippi River basin (Costa et al., 2023;
Louisiana Universities Marine Consortium, 2018). The dead zone begins in later summer, when the
water column stratifies, and the benthic water column layer becomes deprived of dissolved oxygen
(Costa et al., 2023; Louisiana Universities Marine Consortium, 2018). The dead zone persists for months
until weather changes provide stronger mixing of water and break up the layer stratification (Louisiana
Universities Marine Consortium, 2018). As of 2023, the five-year average size of the dead zone measures
just over 4,300 square miles (National Oceanic and Atmospheric Administration, 2023). The hypoxic area
represents millions of acres of possibly unsuitable habitat for some wildlife (National Oceanic and
Atmospheric Administration, 2023). While some species can leave the area as the hypoxic zone forms,
others cannot escape it and become stressed or die. The displacement or removal of species can impact
the trophic interactions within an ecosystem, creating impacts to other species, such as predators.
Models have demonstrated the ability of the dead zone to lower the reproductive capacity, increase
mortality, and alter the diet of finfish, as well as alter habitat use of shellfish (Craig & Crowder, 2005;
National Oceanic and Atmospheric Administration, 2020; Rose et al., 2018).

4.6 References

American Water Works Association (2017). Replacement and Flushing of Lead Service Lines.

ANSI/AWWA C810-17. Effective date: November 1, 2017.

https://engage.awwa.org/PersonifyEbusiness/Bookstore/Product-Details/productld/65628258

Association of State Drinking Water Administrators (ASDWA). 2020a. States LCRR Compliance
Monitoring Costs 03172020. May 17, 2020.

ASDWA. 2020b. Costs of States Transactions Study (CoSTS) for EPA's Proposed LCRR. February 6, 2020.

ASDWA. 2024. Costs of States Transactions Study (CoSTS). January 31, 2024. Initially submitted by
ASDWA as part of their comments on the proposed Lead and Copper Rule Improvements. Revised model
transmitted to Erik Helm, USEPA on April 18, 2024 from Kevin Letterly, ASDWA.

Backer, L. C. 2002. Cyanobacterial Harmful Algal Blooms (CyanoHABs): Developing a Public Health
Response. Lake and Reservoir Management, 18(1), 20-31. doi:10.1080/07438140209353926

Benjamin, M.M., S.H. Reiber, J.F. Ferguson, E.A. Vanderwerff, and M.W. Miller. 1990. Chemistry of
Corrosion Inhibitors in Potable Water. American Water Works Association Research Foundation.

Final LCRI Economic Analysis

4-357

October 2024


-------
Betanzo, Elin Warn, Safe Water Engineering, and Spieght, Vanessa. 2024. Lead Service Line Replacement
Costs and Strategies for Reducing Them. National Resources Defense Council. Submitted by NRDC as
part of their comment on the proposed LCRI.

Bi-state Commission. 2014. 2015 Water Treatment Chemicals - October 29, 2014 - Bid opening.

Bukhari, Z., S. Ge, S. Chiavari, and P. Keenan. 2020. Lead Service Line Identification Techniques. The
Water Research Foundation, Project No. 4693. Available at

https://www.waterrf.org/research/proiects/lead-service-line-identification-techniques

CDM Smith. 2022. Considerations when Costing Lead Service Line Identification and Replacement.
American Water Works Association. Submitted by AWWA as part of the Small Business Advocacy Review
(SBAR) comments.

Center for Rural Pennsylvania. 2007. Biosolids Disposal in Pennsylvania. November.

Chastain-Howley, A., G. Kunkel, W. Jernigan, and D. Sayers. 2013. Water loss: The North American
dataset. Journal AWWA 105(6): 57-60.

Chislock, M.F., E. Doster, R.A. Zitomer, and A.E. Wilson. 2013. Eutrophication: Causes, consequences,
and controls in aquatic ecosystems. Nature Education Knowledge 4(4):10.

City of Seabrook. 2016. Agenda Briefing: Bid Award for Disposal of Municipal Sludge Project 2016-05. 11
April.

Cloern, J. 2001. Our Evolving Conceptual Model of the Coastal Eutrophication Problem. Marine Ecology
Progress Series, 210, 223-253. doi:10.3354/meps210223

Comber, S., M. Gardner, K. Georges, D. Blackwood, and D. Gilmour. 2013. Domestic source of
phosphorus to sewage treatment works. Environmental Technology 34(9-12): 1349-1358.

Costa, H., E. Sprout, S. Teng, M. McDaniel, J. Hunt, D. Boudreau, . . . H. Hall (2023). Dead Zone.

Retrieved from https://education.nationalgeographic.org/resource/dead-zone/

Craig, J. K., and L.B. Crowder. (2005). Hypoxia-induced habitat shifts and energetic consequences in
Atlantic croaker and brown shrimp on the Gulf of Mexico shelf. Marine Ecology Progress Series, 294, 79-
94.

Dodds, W.K., W.W. Bouska, J.L. Eitzmann, T.J. Pilger, K.L. Pitts, A.J. Riley, J.T. Schloesser, and D.J.
Thornbrugh. 2009. Eutrophication of U.S. freshwaters: Analysis of potential economic damages.
Environmental Science Technology 43(1): 12-19.

Elfland, C., P. Scardina, and M. Edwards. (2010). Lead-contaminated water from brass plumbing devices
in new buildings. Journal AWWA, 102(11), 66-76. https://doi.Org/10.1002/j.1551-8833.2010.tbll340.x.

Folkman, S. 2018. Water Main Break Rates In the USA and Canada: A Comprehensive Study. Utah State
University, Buried Structures Laboratory.

Frederick County. 2014. Executive Summary, Mayor and Board of Aldermen, Re: Bid #14-135 (Chemicals
for Water Supply and Wastewater Treatment). 29 May.

Glibert, P.M., D.M. Anderson, P. Gentien, E. Graneli, and K.G. Sellner. 2005. The global, complex
phenomena of harmful algal blooms. Oceanography 18(2): 136-147.

Final LCRI Economic Analysis

4-358

October 2024


-------
Grady, C.P.L. 2011. Biological Wastewater Treatment. 3rd Edition.

Hensley, K., V. Bosscher, S. Triantafyllidou, and D.A. Lyttle. 2021. Lead Service Line Identification: A
Review of Strategies and Approaches. AWWA Water Science.

Jeppesen, E., J. P. Jensen, M. S0ndergaard, T. Lauridsen, L. J. Pedersen, & L. Jensen. (1997). Top-down
control in freshwater lakes: the role of nutrient state, submerged macrophytes and water depth.
Hydrobiologia, 342(0), 151-164. doi:10.1023/A:1017046130329

Jiang, F., M.B. Beck, R.G. Cummings, K. Rowles, and D. Russell. 2004. Estimation of Costs of Phosphorus
Removal in Wastewater Treatment Facilities: Construction De Novo. Water Policy Working Paper #2004-
010. June.

Kimbrough, D.E. (2007). Brass corrosion as a source of lead and copper in traditional and all-plastic
distribution systems. Journal AWWA, 99(8): 70-76. https://doi.Org/10.1002/j.1551-
8833.2007.tb08008.x.

Lehtiniemi, M., J. Engstrom-Ost, & M. Viitasalo. (2005). Turbidity decreases anti-predator behaviour in
pike larvae, Esox lucius. Environmental Biology of Fishes, 73(1), 1-8. doi:10.1007/sl0641-004-5568-4

Liggett, J., H. Baribeau, E. Deshommes, D.A. Lytle, S.V. Masters, Q. Muylwyk, and S. Triantafyllidou. 2022.
Service Line Material Identification: Experiences From North American Water Systems. Journal AWWA,
114: 8-19.

Louisiana Universities Marine Consortium. (2018). About Hypoxia. Retrieved from
https://gulfhypoxia.net/about-hypoxia/

Mayer, P.W., W.B. DeOreo, E.M. Opitz, J.C. Kiefer, W.Y. Davis, B. Dziegielewski, and J.O. Nelson. 1999.
Residential End Uses of Water. American Water Works Association Research Foundation.

Metcalf and Eddy. 2003. Wastewater Engineering: Treatment and Reuse. 4th Edition.

National Oceanic and Atmospheric Administration. (2020). Hypoxia Study Shows Fish Diet and
Catchability Changes in Gulf of Mexico Dead Zone. Retrieved from

https://coastalscience.noaa.gov/news/hypoxia-study-shows-fish-diet-and-catchability-changes-in-gulf-
of-mexico-dead-zone/

National Oceanic and Atmospheric Administration. (2023). Below Average Summer 2023 'Dead Zone'
Measure in Gulf of Mexico. Retrieved from https://coastalscience.noaa.gov/news/below-average-
summer-2023-dead-zone-measured-in-gulf-of-mexico/

Office of Management and Budget (OMB). (2003). Circular A-4. Obama White House Archives.
https://obamawhitehouse.archives.gov/omb/circulars a004 a-4/.

OMB. (2023). Circular A-4. November 9, 2023. Retrieved from https://www.whitehouse.gov/wp-
content/uploads/2023/ll/CircularA-4.pdf

Pasadena, Texas. Frequently Asked Questions - Water Meters. Frequentlv-Asked-Questions—Water-
Meters-PDF (pasadenatx.gov). No date.

Final LCRI Economic Analysis

4-359

October 2024


-------
Rockey, N.C., Shen, Y., Haig, S.J., Wax, M., Yonts, J., Wigginton, K.R., Raskin, L., and Olson, T.M. (2021).
Impact of service line replacement on lead, cadmium, and other drinking water quality parameters in
Flint, Michigan. Environmental Science: Water Research & Technology, 7(4): 797-808.
https://doi.org/10.1039/D0EW00975J.

Rodgers, M. 2014. Impact of Corrosion Control on Publicly Owned Treatment Works. Presented at
American Water Works Association Water Quality Technology Conference. New Orleans, Louisiana.
November 16-20.

Rose, K. A., S. Creekmore, P. Thomas, J.K. Craig, M.S. Rahman, & R.M. Neilan. (2018). Modeling the
Population Effects of Hypoxia on Atlantic Croaker (Micropogonias undulatus) in the Northwestern Gulf
of Mexico: Part 1—Model Description and Idealized Hypoxia. Estuaries and Coasts, 41(1), 233-254.
doi: 10.1007/sl2237-017-0266-6

Sandvig, A., P. Kwan, G. Kirmeyer, B. Maynard, D. Mast, R.R. Trussell, S. Trussell, A. Cantor, and A.
Prescott. 2008. Contribution of Service Line and Plumbing Fixtures to Lead and Copper Rule Compliance
Issues. Denver, Colo.: Awwa Research Foundation. Peer reviewed by AwwaRF Project Advisory
Committee.

Sloan, D.S., R.A. Pelletier, and M. Modell. 2008. Sludge management in the City of Orlando - It's
supercritical. Florida Water Resources Journal June: 46-54.

Stamford Water Pollution Control Authority (WPCA). 2013. Memorandum to SWPCA Board, Re: Sludge
Drying and Disposal. 7 January.

Su, R. C., C.M. Meyers, E.A. Warner, J.A. Garcia, J.M. Refsnider, A. Lad, . . . D.J. Kennedy. (2020). Harmful
Algal Bloom Toxicity in Lithobates catesbeiana Tadpoles. Toxins (Basel), 12(6).
doi:10.3390/toxinsl2060378

Triantafyllidou, Simoni & Edwards, Marc. (2012). Lead (Pb) in Tap Water and in Blood: Implications for
Lead Exposure in the United States. Critical Reviews in Environmental Science and Technology, 42: 1297-
1352. 10.1080/10643389.2011.556556.

Turner, A. M., & M. F. Chislock. (2010). Blinded by the stink: Nutrient enrichment impairs the perception
of predation risk by freshwater snails. Ecol Appl, 20(8), 2089-2095. doi:10.1890/10-0208.1

United States Census Bureau. 2019. 2019 American Community Survey. Available at
https://www.census.gov/acs/www/data/data-tables-and-tools/data-profiles/2019/.

United States Census Bureau. 2020. Table AVG1. Average Number Of People Per Household, By Race
And Hispanic Origin, Marital Status, Age, And Education of Householder: 2020. Available at
https://www2.census.gov/programs-surveys/demo/tables/families/2020/cps-2020/tabavgl.xlsx

United States Census Bureau. 2021. 2020 American Community Survey. Available at
https://www.census.gov/acs/www/data/data-tables-and-tools/data-profiles/2020/.

United States Census Bureau. 2022. DP02: Selected Social Characteristics in the United States. U.S.
Census Bureau, 2016-2020 American Community Survey 5-Year Estimates. Available at:
https://data.census.gov/table?t=Language+Spoken+at+Home&g=0100000US$1600000&tid=ACSDP5Y20
20.DP02.

Final LCRI Economic Analysis

4-360

October 2024


-------
United States Department of Treasury (USDT). (2024). State and Local Fiscal Recovery Funds: Public
Data, https://home.treasury.gov/policy-issues/coronavirus/assistance-for-state-local-and-tribal-
governments/state-and-local-fiscal-recovery-funds/public-data.

United States Environmental Protection Agency (USEPA). 2006a. Economic Analysis for the Final Ground
Water Rule. October 2006. Office of Water. EPA 815-R-06-014.

USEPA. 2006b. Point-of-Use or Point-of-Entry Treatment Options for Small Drinking Water Systems. April
2006. Office of Water. EPA 815-R-06-010. Available at https://www.epa.gov/sites/default/files/2015-
09/documents/guide smallsystems pou-poe june6-2006.pdf.

USEPA. 2007. Economic and Supporting Analyses: Short-Term Regulatory Changes to the Lead and
Copper Rule. September 2007. Office of Water. EPA-815-R0-7022. Available at
https://nepis.epa.gov/Exe/ZvPDF.cgi?Dockev=P100150Y.txt.

USEPA. 2008a. Lead and Copper Rule: Public Education & Other Public Information Requirements for
Community Water Systems. June 2008. Office of Water. EPA-816-F-08-019. Available at
https://nepis.epa.gov/Exe/ZvPDF.cgi?Dockev=60001Q8Y.txt.

USEPA. 2008b. Indoor Water Use in the United States. June 2008. EPA-832-F-06-004. Available at

https://www.epa.gov/sites/default/files/2017-Q3/documents/ws-facthseet-indoor-water-use-in-the-

us.pdf.

USEPA. 2009. 2006 Community Water System Survey Volume I: Overview. February 2009. Office of
Water. EPA 815-R-09-001. Available at https://www.epa.gov/dwreginfo/communitv-water-svstem-
survev.

USEPA. 2010. Nutrient Control Design Manual. August 2010. Office of Research and Development,
National Risk Management Research Laboratory - Water Supply and Water Resources Division.
EPA/600/R-10/100.

USEPA. 2012a. Economic Analysis for the Final Revised Total Coliform Rule. September 2012. Office of
Water. EPA 815-R-12-004. Available at

https://nepis.epa.gov/Exe/ZvPDF.cgi/P100PIVQ. PDF?Dockev=P100PIVQ. PDF.

USEPA. 2012b. Technology and Cost Document for the Final Revised Total Coliform Rule. December
2012. Office of Water. EPA-815-R-12-005. https://beta.regulations.gov/document/EPA-HQ-QW-2008-
0878-0299.

USEPA. 2013. Emerging Technologies for Wastewater Treatment and In-Plant Wet Weather
Management. March 2013. EPA 832-R-12-011.

USEPA. 2015a. Public Water System Supervision Program Information Collection Request (ICR) (Renewal).
October 2015. https://www.reginfo.gov/public/do/PRAViewlCRPref nbr=201510-2040-001.

USEPA. 2015b. Case Studies on Implementing Low-Cost Modifications to Improve Nutrient Reduction at
WWTPs. Draft report. 147 p.

USEPA. 2016a. National Lakes Assessment 2012: A Collaborative Survey of Lakes in the United States.
December 2016. EPA 841-R-16-113. Available at https://www.epa.gov/sites/production/files/2016-
12/documents/nla report dec 2016.pdf.

Final LCRI Economic Analysis

4-361

October 2024


-------
USEPA. 2016b. National Rivers and Streams Assessment 2008-2009 Fact Sheet.

https://www.epa.gov/sites/production/files/2016-

03/documents/fact sheet draft variation march 2016 revision.pdf.

USEPA. 2018. 3Tsfor Reducing Lead in Drinking Water in Schools and Child Care Facilities: A Training,
Testing, and Taking Action Approach (Revised Manual). October 2018. Office of Water. EPA 815-B-18-
007. Available at https://www.epa.gov/ground-water-and-drinking-water/3ts-reducing-lead-drinking-
water-toolkit.

USEPA. 2019. Estimated Total Nitrogen and Total Phosphorus Loads and Yields Generated within States.
Retrieved from: https://www.epa.gov/nutrient-policy-data/estimated-total-nitrogen-and-total-
phosphorus-loads-and-vields-generated-within.

USEPA. 2020. Economic Analysis for the Final Lead and Copper Rule Revisions. December 2020. Office of
Water. EPA 816-R-20-008.

USEPA. (2020b). Use of Lead Free Pipes, Fittings, Fixtures, Solder and Flux for Drinking Water; Final Rule.
Federal Register. 85 FR 54235. September 1, 2020.

USEPA. 2021. Innovative Nutrient Removal Technologies: Case Studies of Intensified or Enhanced
Treatment. August 2021. Office of Water. EPA 830-R-01-001.

https://www.epa.gov/system/files/documents/2022-08/innovative-nutrient-removal-technologies-
report-082721.pdf

USEPA. 2022a. Disinfectants/Disinfection Byproducts, Chemical, and Radionuclides Rules ICR (Renewal).
Submitted to OMB for review and approval on March 24, 2023. (EPA ICR Number 1896.12, OMB Control
Number 2040-0204).

USEPA. 2022b. Guidance for Developing and Maintaining a Service Line Inventory. August 2022. Office of
Water. EPA 816-B-22-001. Available at https://www.epa.gov/system/files/documents/2022-
08/lnventorv%20Guidance August%202022 508%20compliant.pdf.

USEPA. 2023a. Drinking Water Infrastructure Needs Survey and Assessment (2021 DWINSA): Seventh
Report to Congress. September 2023. Office of Water. EPA 810-R-23-001. Available at
https://www.epa.gov/system/files/documents/2023-
09/Seventh%20DWINSA September2023 Final.pdf.

USEPA. 2023b. Technologies and Costs for Corrosion Control to Reduce Lead in Drinking Water. August
2023. Office of Water.

USEPA. 2023c. Economic Analysis for the Proposed Lead and Copper Rule Improvements. November
2023. Office of Water. EPA 815-R-23-005.

USEPA. 2023d. National Primary Drinking Water Regulations: Lead and Copper Rule Improvements.
Proposed Rule. Federal Register. 88 FR 84878. December 6, 2023. Washington, D.C.: Government
Printing Office.

USEPA. 2023e. Questions Regarding the Time Needed to Develop and Maintain the LCR Service Line
Inventory. Questionnaire developed by EPA and sent to 9 systems in early 2023. Responses received
from Grand Rapids, Ml; Pittsburgh, PA; and Cincinnati, OH.

Final LCR! Economic Analysis

4-362

October 2024


-------
USEPA. 2023f. EPA Launches New Initiative to Accelerate Lead Pipe Replacement to Protect
Underserved Communities. Press Release. EPA Headquarters. Office of Water. Washington, DC.

Retrieved from https://www.epa.gov/newsreleases/epalaunches-new-initiative-accelerate-leadpipe-
replacement-protect-underserved.

USEPA. 2024a. Economic Analysis of the Final Consumer Confidence Reports Rule Revisions. Office of
Water.

USEPA. 2024b. Accessed 2024. Water Pollution Search, https://echo.epa.gov/trends/loading-tool/water-
pollution-search

Verardi, V., and C. Vermandele. 2018. Univariate and multivariate outlier identification for skewed or
heavy-tailed distributions. Stata Journal 18: 517-532.

Final LCRI Economic Analysis

4-363

October 2024


-------
5 Benefits Resulting from the Lead and Copper Rule Improvements

5.1 Introduction

Lead is a highly toxic pollutant that can damage neurological, cardiovascular, immunological,
developmental, and other major body systems (USEPA, 2024a). Children are at higher risk from the
effects of lead than adults, due to differences in their stage of brain development, body weight,
physiology, and behavior (USEPA, 2024a).

Although copper is essential to normal physiology, excess intake is also associated with several adverse
health outcomes (NRC, 2000). Most commonly, excess exposure to copper leads to gastrointestinal
symptoms such as nausea, vomiting, and diarrhea (NRC, 2000). In children with genetic disorders or
predispositions to accumulate copper, chronic exposure to non-physiological levels of this element can
result in liver damage.

Due to these serious adverse effects, the Lead and Copper Rule improvements (LCRI) are expected to
yield significant health benefits, which are described in this chapter and associated appendices. Some of
these benefits are expressed in monetary terms. Section 5.2 presents modeling results and limitations of
the analysis on the reduction of lead levels in water as a result of two interventions: 1) the removal of
lead service lines (LSLs) and 2) the introduction of corrosion control treatment (CCT). Section 5.3
discusses the assignment of drinking water concentrations to public water system (PWS) populations
and associated limitations. Sections 5.4, 5.5, and 5.6 focus on the methodology and assumptions for
quantifying the benefits of line removal and corrosion control interventions. Section 5.6.2 presents the
results of the quantified and monetized benefits. Section 5.7 provides a summary exhibit and outlines
the identified limitations and uncertainties in the benefits analysis and how they might affect the
estimated values presented in the chapter. Section 5.8 discusses the nonquantifiable benefits associated
with the regulatory requirements of the final LCRI. Section 5.9 discusses the potential climate disbenefits
from the operation of optimal corrosion control treatment (OCCT) at drinking water treatment facilities
and the use of construction and transport vehicles in the replacement of lead and galvanized requiring
replacement (GRR) service lines.

The United States Environmental Protection Agency (the EPA or the agency) quantitatively estimated
benefits using low and high benefit scenarios. The low and high scenarios are driven by the number of
PWSs that will exceed the lead action level (AL) under the revised tap sampling requirements of the final
LCRI.

The low and high scenarios are also defined by the concentration-response functions that characterize
how reductions in blood lead levels (BLLs) (caused by changes in lead exposure) translate into avoided
intelligence quotient (IQ) reductions, cases of attention deficit hyperactivity disorder (ADHD), and
cardiovascular disease (CVD) premature mortality. The specific concentration-response functions that
define the low and high scenarios are as follows:

•	For IQ in children Lanphear et al. (2005, errata 2019) is used for the high scenario estimate and
the low scenario estimate is based on Crump et al. (2013). See Section 5.5.1.

•	For avoided cases of ADHD in children the high scenario estimate is based on Froelich et al
(2009) and the low estimate is based on Ji et al. (2018). See Section 5.5.3.

Final LCRI Economic Analysis

5-1

October 2024


-------
• For CVD premature mortality in adults the high scenario estimate is based on Lanphear et al.
(2018) and the low scenario is based on Aoki et al. (2016). See Section 5.5.7.

Note that the fourth category of quantified benefits, reductions in lower birth weight in infants due to
mother's lead exposure, is estimated with a single function in both the high and low scenarios based on
Zhu et al. (2010). See Section 5.5.5.

The third factor that differentiates the estimated range of benefits is the use of low and high valuations
for the ADHD cost of illness, as described in Section 5.5.4.

Numerous other adverse health effects are associated with exposure to lead, many at low doses.
Appendices D and E contain additional information on the effects of lead and copper exposure.

Appendix D provides more detailed information on the six categories of health effects that the EPA and
the National Toxicology Program (NTP) have deemed to be associated with lead exposures:
cardiovascular effects, renal effects, reproductive and developmental effects, immunological effects,
neurological effects, and cancer. The adverse health effects associated with copper are summarized in
Appendix E. At sufficient exposures, copper has been associated with gastrointestinal effects in the
general population and with liver toxicity in susceptible individuals (e.g., individuals with Wilson's
Disease). The EPA anticipates that these adverse health effects will also be reduced due to the rule, but
they are not explicitly quantified in this analysis. Appendix F presents an additional sensitivity analysis on
the valuation of IQ estimates, for children up to age 7.

5.2 Baseline and Post-Rule Drinking Water Lead Exposures

This section discusses methods for estimating baseline (i.e., current) and post-rule exposures to lead
through drinking water. The EPA used the lead concentration of water drawn from the kitchen tap to
estimate exposure through drinking water under each of the potential CCT and LSL159 scenarios. No
national level dataset exists that incorporates sufficient detail regarding these scenarios, so the EPA
obtained datasets from multiple sources. This combined dataset has limitations, such as varying
sampling methods and locations. The EPA managed these limitations, first through data cleaning,
coding, model fitting, and selection. The EPA subsequently used this model to produce a simulated
dataset of lead concentrations under the different scenarios used to control for variation in the
combined dataset to the extent possible.

To estimate drinking water lead concentrations at the tap, the EPA obtained and assessed tap water
lead concentration data from utilities, the EPA's regional offices and Office of Research and
Development, and authors of published journal articles. These data include information about sampling
methods, locations, dates, and LSL status. The EPA further divided the lead tap concentration records
into CCT categories based on the locations and dates of samples, and known treatment and finished
water quality histories. The EPA combined these sources to produce a dataset (described in Section
5.2.1) for further analysis (Section 5.2.2). The EPA then fit a model to these data and subsequently used

159 Note that the EPA does not have sampling data from galvanized lines previously downstream of a lead line and
therefore requiring replacement (GRR). In the estimation on benefits the EPA assumes that water lead levels from
both LSLs and GRRs are equivalent.

Final LCRI Economic Analysis

5-2

October 2024


-------
the fitted model to simulate representative lead concentrations in PWSs. The resulting simulated
dataset of the tap sample lead concentrations was used to estimate BLLs (as described in Section 5.4).

Ideally, to determine the potential lead tap concentrations under the various CCT and LSL scenarios, a
researcher would analyze the variation of lead concentrations in tap samples nationwide across the
defined scenarios. However, due to the nature of the available data, the EPA's lead concentration data
were collected from different locations, with different methods, over multiple decades, and for different
purposes. Therefore, the interpretation of what is driving the tap sample lead concentration variation
within and across CCT and LSL scenarios becomes complicated. A good deal of the variation in the lead
concentration data may be due to the use of different sample collection methodologies and unequal
numbers of repeated samples at the same time and place. Therefore, rather than using summary
statistics from the original data directly, the EPA undertook a detailed analysis to understand the effects
of the LSL and CCT scenarios while statistically controlling for data collection artifacts that may have
contributed to variation in measurements of lead concentration at the tap (Sections 5.2.1).

The EPA implemented the following analyses of the aggregate dataset and adjusted the data to enhance
the quality. After compiling the water lead concentration dataset, the EPA statistically modeled the
relationship of LSLs and CCT with lead concentrations at the tap (Section 5.2.2). This model also related
lead concentration to the amount of water that had flowed from the tap after stagnation. Additionally,
the EPA incorporated methods to estimate the effects of different water systems, residences, and
sampling events at the same residence. The EPA incorporated terms into the model for the amount of
water, city water system, and residence to control for data collection artifacts from different studies, as
most cities were linked to a single study per city water system (Exhibit 5-1). The EPA similarly controlled
for differences among sampling events in the same homes and within studies.

The fitted model demonstrates that LSL status and CCT both affect lead concentration at the tap. The
presence of an LSL is associated with higher lead concentrations. In homes with any LSL (full or partial),
improved CCT is associated with lower lead concentrations. CCT has less of an effect in homes with no
LSL present. Assessment showed that seven combinations (i.e., scenarios) of LSL status and CCT had
predicted concentrations that were sufficiently distinct to warrant separate predictive modeling. These
seven scenarios were used to produce estimates of drinking water concentration. The EPA also used
information from the statistical model to simulate estimates of lead concentration for ten sequential
one-liter volumes drawn from a household tap after stagnation (Section 5.2.3). Given recent findings
(Urbanic et al. 2022) from the comparison of composite samples, which approximate lead exposure
given water use patterns at a residence, and profile samples, where a volume weighted average lead
concentration was calculated, at sites in two cities, the agency chose to use a volume weighted average
lead concentration calculated using data from the first 10 liters of profile data in approximating
exposure at the tap for this final LCRI benefits analysis. Throughout this document, the EPA uses
standard terminology to describe LSL status and CCT implementation. "LSL" indicates lead service lines
are present, "partial LSL" indicates some presence of lead in service lines (i.e., partial replacement), and
"no LSL" indicates no LSLs. For CCT, "none" indicates no CCT, "partial" represents systems that have
some CCT in place but are not optimized, and "representative" indicates a water chemistry that
exemplifies the optimal CCT currently in use (which can include some combination of higher phosphate
values or optimized pH levels). Water lead concentration prediction intervals overlapped completely for
all CCT scenarios in homes with no LSL (described in Section 5.2.3), so "combined" indicates pooled CCT
estimates representing all three states of CCT in non-lead service line households. Further details are

Final LCRI Economic Analysis

5-3

October 2024


-------
provided in Section 5.3. Additionally, all water modeling was conducted using data based on the
presence or absence of LSLs (Sections 5.2.1, 5.2.2, and 5.2.3). Galvanized service lines requiring
replacement (GRR) are also considered in the rule, and these are assigned a water concentration based
on the results of the LSL analysis (see Section 5.2.4).

5.2.1 Drinking Water Lead Concentration Profile Data

The EPA combined data from multiple sources for use in estimating lead concentration at the tap based
on LSL status and CCT implementation. For the LCRI, the EPA updated the analysis done for the 2021
LCRR with new data. In order to produce these updated values, we incorporated new data from the city
of Clarksburg, WV collected in October of 2021. This dataset was collected in fall to winter 2021, and
each sampling event consisted of 19 profile samples with 17 liters of water in each profile. The
Clarksburg data included 19 unique residences with lead lines, as well as 11 that have no known LSL
history and had copper lines at the time of sampling. The locations with no known LSL history and
copper lines were excluded. Ten of the 19 residences with LSL were sampled during two separate
sampling events. One of the sampling events was conducted directly after lead service line replacement
(LSLR) and could not be included due to the potential for elevated lead concentrations directly following
LSLR (McFadden et al. 2011, Sandvig et al. 2008).

The EPA also considered including datasets from Cleveland, OH, Chicago, IL, Kalamazoo, Ml, Parchment,
Ml, Flint, Ml, Galesburg, IL, and Sebring, OH. These data were collected between 2016 and 2021 but not
included in the original analysis. These other datasets had some data availability and study design issues
and therefore could not be included. The EPA's mixed model used for water concentration modeling
estimates random terms for multiple sampling events at the same location. Therefore, when no
locations in a city were sampled over multiple sampling events at different times, the lack of repeated
measurements caused issues in fitting the mixed model. This issue affected the Cleveland, Flint, Chicago,
Kalamazoo, Parchment, and Galesburg datasets. The other datasets were also missing information
regarding LSL and/or CCT status and history. Locations in Flint had previously undergone only partial
LSLR on the private side, but dates of that LSLR were not available, and the private-side replacements
often affected long portions of the service line. Chicago had no information regarding service line
material or replacement status associated with lead samples. The Sebring dataset held multiple
sampling events for three addresses taken while CCT optimization was occurring. While these locations
had LSLs at the beginning of the study, some Sebring locations underwent LSLR over the optimization
period, or had previous partial replacements. The dates for these changes in LSL status were not given
relative to the sampling dates. Additionally, the structure of the dataset suggested the non-detects may
have been replaced by a value of 0.005 mg/L (the practical quantitation limit for lead), but some values
were lower than this. Due to the combination of the lack of clarity regarding the LSLR history and non-
detect usage, we excluded these three locations. In addition to having only single samples per location
with no repeated sampling events, the Kalamazoo and Parchment locations sampled were not clearly
associated with any specific water system. This made it difficult to determine CCT status on any given
date, and also caused uncertainty regarding which system to assign observations to for estimation of the
system-level effect. Finally, CCT information about Galesburg was incomplete for the time of the study,
so the Galesburg data were not included in the updated regression.

Including the material from Clarksburg, WV, the data used in updating the regression analysis
represented 18,571 samples collected from 1,657 homes in 16 cities representing 15 city water

Final LCRI Economic Analysis

5-4

October 2024


-------
systems160 across the United States and Canada (Exhibit 5-1). Data included lead concentrations and
information regarding LSL status, location, and date of sample collection from seven municipal water
systems in the United States and eight in Canada between 1998 and 2021. The EPA chose to include
data from Canada because data from the United States were limited or nonexistent for certain types of
sites, such as sites without corrosion control after LSL removal or homes with LSLs but no CCT. Overall,
geometric mean concentrations were similar in the two countries (described in Section 5.2.1.2),
although there were not enough overlapping data to compare the geometric means for all combinations
of LSL and CCT status.

Exhibit 5-1: Tap Water Lead Concentration Sample Data: Source Citations, City Water System,
LSL and CCT Status Represented in the Data Source, and Number of Individual Sample Bottles

per Source*

Citation of Data Source3

City Water System
Represented by
Data Sourceb

LSL Status of Samples
by Data Source

CCT Status by Data
Source

Total Number
of Samples by
Data Sourcec

Camara et al., 2013

Halifax, NS

LSL, No LSL, Partial

Partial

16

The Cadmus Group, Inc.,
2007

Washington, D.C.

LSL, Partial

Partial

969

Campbell, 2016

Ottawa, ON

LSL, Partial

Representative

5,149

Clarksburg, 2021

Clarksburg, WV

LSL

Partial

532

Commons, 2011

Providence, Rld

LSL, Partial

Partial

169

Commons, 2014

Providence, Rld

Partial

Partial

40

Craik, 2016

Edmonton, AB

LSL, No LSL, Partial

None

967

Del Toral et al., 2013

Chicago, IL

LSL

Representative

695

Del Toral, 2016

Flint, IL

LSL, No LSL

Partial, Representative

3,678

Deshommes et al., 2016

Montreal, QC

LSL, No LSL, Partial

None

630

Desmarais et al., 2015

London, ON

LSL, Partial

None

1,430

EPCOR Water Services, 2008

Edmonton, AB

LSL

None

107

Hayes et al., 2014

Calgary, AB

LSL, No LSL

None

144

Muylwyk, 2016

Guelph, ON

LSL, No LSL, Partial

None

1,039

O'Brien & Gere, 2015

Providence, Rld

LSL, Partial

Partial

158

DC Water, 2016

Washington, D.C.

LSL, No LSL, Partial

Representative

1,391

Schock, 2016

Sebring, OH

LSL

Partial, None

825

Estes-Smargiassi et al., 2006

Boston, MA

LSL, Partial

Representative

50

Swertfeger et al., 2006
Desmarais et al., 2015
Triantafyllidou et al., 2015

Cincinnati, OH

LSL, No LSL, Partial

Partial, Representative

582

* The full analytical dataset is available in the docket for the rule under docket number EPA-HQ-OW-2022-0801 at
https://www.regulations.gov, file: "2023-05-25_PbProfileAbt_FittedModels_DataLCR.xlsx."

a Some of these citations contain data from multiple studies, including previously published and unpublished data.
b Some of these cities represent places where corrosion control levels changed in the same location over time, or
where LSLs were replaced.

160 This number included two cities, Providence and Cranston in Rhode Island, which have been re-categorized as a
single city water system to reflect their shared water source. Cranston is a consecutive system and receives its
water from Providence.

Final LCRI Economic Analysis

5-5

October 2024


-------
cThe number of samples is the number of individually measured water samples (i.e., bottles). The number of
profile sampling events is shown in Section 5.2.1.2.

d Cincinnati before 2006; Halifax and Providence/Cranston water systems were revised from "Representative" CCT
to "Partial" CCT based on public comment as well as peer review of Stanek et al., 2020. These changes were
applied to all of the following figures and tables.

5.2.1.1 Lead Concentration Profiles

Most data sources contained series, or "profiles," of water samples that were drawn from the same
kitchen tap after a whole-house stagnation period. Exhibit 5-2 shows the general sampling process as it
relates to portions of home plumbing, service line, and the connection to the city water main. In
general, the EPA does not believe that the water in water mains in the United States contain lead.161
Water can become contaminated during stagnation by lead leaching from LSL and home plumbing
containing lead. When the tap is turned on and water is drawn after stagnation, lead concentrations
may show peaks based on the amount and location of lead-bearing plumbing materials in contact with
the water between the tap and the water main. In other words, there may be considerable variation in
lead concentration measured in water samples drawn from a tap after a stagnation period; this variation
decreases as non-stagnant water from the main reaches the tap. Taps have different flow rates, and the
volume of water rather than the length of time was used to account for the position in tap sampling
series.

A "complete" profile includes consecutive measurements taken from the tap, through any peaks in lead
concentration, to a point where the lead concentration in water shows little to no further decrease.
Exhibit 5-3 displays an example of a complete profile of lead in tap water. Most of the primary data
sources, representing the 15 city water systems, contain profiles of varying levels of completeness
(Exhibit 5-4 and Exhibit 5-5). However, the sources also incorporate data regarding sample volume and
position in the profile series for each individual sample. The EPA used this information to calculate the
"profile liter" variable (Section 5.2.2) to control for variability in differences in profile position and
volume among samples within the fitted model and the following simulation.

Although these data represent a large portion of available data, they may not be nationally
representative with respect to the following factors: water chemistry and corrosion control practices;
service line length, materials, and scales; size, type, and location of internal piping and lead sources; the
type and number of residences with LSLs; and the relative contribution of particulate lead. These data
also do not incorporate water usage patterns within a home that could affect exposure, such as
dishwasher use, laundry, and showering. Some usage patterns may flush water lines and reduce
exposure to stagnant water through drinking and cooking. The following sections describe how the EPA
cleaned the data; coded and fit models to control for some of the variation in the existing dataset due to
water system, site, and sampling methods; and produced simulated values for use in BLL estimation.

161 The EPA does not believe that there are lead water mains in the country. Water mains are typically six to 16
inches in diameter whereas service lines have a smaller diameter. The common water main materials include
ductile iron, PVC, asbestos cement, HDPE, and concrete steel (Folkman, 2018). Lead service lines are two inches or
less in diameter (LSLR Collaborative, n.d.).

Final LCRI Economic Analysis

5-6

October 2024


-------
Exhibit 5-2: Diagram Showing Plumbing Where Water Can Become Contaminated with Lead

This exhibit shows a profile of multiple, one-liter samples. Although mixing occurs, the earliest samples drawn after
stagnation are representative of water in fixtures and home plumbing, while those that follow represent water
from service lines, and finally, the water main.

Exhibit 5-3: Example of a Complete Consecutive Liter Profile of Lead Concentrations in Tap
Water from a Location with a Lead Service Line

_Q
CL

30

20 -

10-

5	10

Cumulative Liters after last stagnation period

15

Source: Data from Commons (2011).

Final LCRI Economic Analysis

5-7

October 2024


-------
Note: Lead concentration is elevated in the first liter, lower in the second through third liters, highest from the
fourth to sixth liters, and zero after the seventh liter. Red dots represent lead (Pb) concentration plotted at the
midpoint of the cumulative volume of each sample ("profile liter"). The widths of the horizontal bars indicate the
total volume of each sample. The samples shown in this figure were from a residence with an LSL and
representative CCT in Providence, Rl (Commons, 2011).

5.2.1.2 Data Cleaning

The EPA cleaned and combined the datasets listed in Exhibit 5-1 by removing duplicate records, records
without water lead concentration values, and records that did not meet the criteria for inclusion in the
profile dataset. Only samples of known volume after stagnation periods of at least 30 minutes were
included in the profile data. Samples that were collected immediately after flushing events were
generally excluded, unless the flushing volume had also been recorded. Samples from known locations
other than kitchen taps, such as exterior spigots, were also excluded from the data. Concentration
records for homes that underwent partial or full LSLRs occasionally included a number of post-
replacement sampling profile series collected over several months to years after service line
replacement. In these cases, lead concentrations typically declined over subsequent sampling periods,
as residual lead in household plumbing was flushed. As there were too few cases of this post-
replacement sampling in the dataset to incorporate this effect in models, only the last profile after LSLR
was included in the analysis dataset. If elapsed time after an LSLR could not be determined for the post-
replacement samples, all samples after LSLR were included, which may increase the observed variability
in estimates of concentration after LSLR. An outlier for a site in Washington, D.C., was removed after
confirming with the data provider (personal communication, DC Water, May 2017) that the sample was
unlikely to be representative of concentrations in most homes. Other cases with concentrations higher
or lower than expected for particular CCT and LSL categories did not have clear reasons to exclude them,
such as suspect sample collection conditions or obvious particulate lead. These values were included for
the integrity of the dataset.

Before producing summary statistics or fitting models using the profile dataset, the EPA set all known
lead non-detects to 0.1 ng/L,162 and then log-transformed the lead concentrations. The summary tables
in Exhibit 5-4 and Exhibit 5-5 reflect the data cleaning steps.

Exhibit 5-4 and Exhibit 5-5 show the geometric mean, standard deviation (SD), and maximum lead
concentration for each combination of CCT and LSL status in the cleaned data after log-transformation
of lead concentrations for all 18,571 samples included in the model. Additional details regarding data
cleaning and categorization for specific datasets are contained in Appendix F, Section F-l of the
Economic Analysis for the Final Lead and Copper Rule Revisions (hereafter referred to as the "Final 2021
LCRR EA") (USEPA, 2020a). Exhibit 5-5 provides summary statistics by LSL and CCT status from the
existing data, ignoring differences in city water system, site, sampling event, and study sampling volume
methodology.

162 As the log of zero is not a real number, setting all known non-detects to a small, non-zero value allows log-
transformation of all results in a dataset. It was not possible to determine detection limits or all non-detects for all
included datasets, and known non-detects were identified from zeroes or missing values.

Final LCRI Economic Analysis

5-8

October 2024


-------
Exhibit 5-4: Summary Statistics for Tap Water Lead Concentrations by LSL and CCT Status Combinations, Country, and Citation









Geometric

Geometric
SD Lead

Arithmetic

Arithmetic

Number

Number

Number
of Sites

LSL

CCT

Country

Citation3

Mean Lead

Mean

SD Profile

of

of









(Hg/L)

Profile Literb

Liter

Samples

Profiles





USA

Schock, 2016

26.84

1.13

5.87

4.29

15

1

1







Craik, 2016

15.35

2.50

4.95

3.52

194

26

20







Deshommes et al., 2016

26.87

2.14

3.99

2.80

309

69

27

LSL

None

CND

Desmarais et al., 2015

16.43

2.21

4.00

2.29

1,062

133

11







EPCOR Water Services, 2008

21.45

1.93

3.87

3.26

107

26

11







Hayes et al., 2014

14.55

1.71

6.00

3.47

120

5

5







Muylwyk, 2016

16.63

2.56

1.00

0.50

248

124

123







The Cadmus Group Inc., 2007

9.81

3.30

11.95

7.82

895

41

36







Clarksburg, 2021

6.72

2.99

7.84

5.34

532

28

19







Commons, 2011

14.60

2.70

7.81

4.78

121

8

8







Del Toral, 2016

2.71

4.41

5.60

4.23

2,068

137

91

LSL

Partial

USA

O'Brien & Gere, 2015

14.77

2.99

1.77

2.14

133

46

7







DC Water, 2016

6.17

3.16

8.48

5.55

205

13

6







Schock, 2016

6.89

2.49

5.93

4.28

810

53

14







Swertfeger et al., 2006;





















Desmarais et al., 2015;

10.47

1.96

0.38

0.00

91

91

21







Triantafyllidou et al., 2015



















CND

Camara et al., 2013

16.30

1.93

2.00

1.20

8

2

2







Del Toral et al., 2013

8.00

2.03

6.19

3.70

695

57

32







Del Toral, 2016

2.81

3.07

6.50

4.73

1,270

80

47







DC Water, 2016

3.05

3.34

6.74

4.12

839

64

52

LSL

Representative

USA

Estes-Smargiassi et al., 2006

5.23

2.28

3.03

2.78

25

2

2







Swertfeger et al., 2006;





















Desmarais et al., 2015;

1.38

4.01

3.55

2.70

303

46

12







Triantafyllidou et al., 2015



















CND

Campbell, 2016

1.89

3.57

2.15

1.35

4,997

1,205

639





USA

None

—

—

—

—

—

—

—







Craik, 2016

6.81

5.71

1.96

1.13

451

122

116

Partial

None

CND

Deshommes et al., 2016

12.71

2.04

4.76

3.26

248

40

40







Desmarais et al., 2015

10.00

2.06

4.00

2.29

368

46

4







Muylwyk, 2016

7.53

4.86

1.00

0.50

341

171

169







The Cadmus Group, Inc., 2007

3.70

3.02

10.64

7.21

74

4

2







Commons, 2011

9.12

2.88

3.56

2.19

48

7

7







Commons, 2014

10.19

2.19

6.88

4.20

40

3

3

Partial

Partial

USA

O'Brien & Gere, 2015

8.84

2.57

1.70

2.11

25

9

1







DC Water, 2016

8.80

1.93

7.50

4.47

15

1

1

Final LCRI Economic Analysis

5-9

October 2024


-------
LSI-

CCT

Country

Citation3

Geometric
Mean Lead
(Hg/L)

Geometric
SD Lead

Arithmetic

Mean
Profile Literb

Arithmetic
SD Profile
Liter

Number

of
Samples

Number

of
Profiles

Number
of Sites







Swertfeger et al., 2006;
Desmarais et al., 2015;
Triantafyllidou et al., 2015

2.78

2.53

0.38

0.00

11

11

11

CND

Camara et al., 2013

18.44

1.27

2.00

1.29

4

1

1

Partial

Representative

USA

DC Water, 2016

1.95

2.45

7.30

4.60

266

19

19

Estes-Smargiassi et al., 2006

0.24

4.53

3.03

2.78

25

2

2

Swertfeger et al., 2006;
Desmarais et al., 2015;
Triantafyllidou et al., 2015

1.54

2.07

3.35

2.68

116

10

2

CND

Campbell, 2016

1.71

3.37

4.00

2.30

152

19

11

No LSL

None

USA

None

...

—

—

—

...

...

—

CND

Craik, 2016

0.82

9.18

1.97

1.13

322

85

85

Deshommes et al., 2016

3.31

3.72

5.31

4.04

73

12

10

Hayes et al., 2014

1.01

1.80

6.00

3.53

24

1

1

Muylwyk, 2016

1.24

3.15

1.00

0.50

450

225

224

No LSL

Partial

USA

Del Toral, 2016

1.74

3.71

8.33

6.85

222

11

7

DC Water, 2016

1.92

2.02

7.50

4.47

15

1

1

Swertfeger et al., 2006;
Desmarais et al., 2015;
Triantafyllidou et al., 2015

2.41

3.59

0.38

0.00

61

61

5

CND

Camara et al., 2013

1.42

2.98

2.00

1.29

4

1

1

No LSL

Representative

USA

Del Toral, 2016

0.66

1.81

6.87

4.87

118

7

4

DC Water, 2016

0.66

2.39

6.44

3.83

51

4

4

CND

None

—

—

—

—

—

—

—

Acronyms: CND = Canada; USA = United States of America; SD = standard deviation.

a Each citation contains data from a single city water system (Exhibit 5-1). Some citations have entries in multiple categories.

b Arithmetic mean and standard deviation (SD) of the "profile liter" term and number of individual sample bottles, profiles, and sites provide information
regarding some of the differences in sampling methods observed among studies. Studies with fewer samples, or with smaller sam pie volumes, have smaller
values of profile liter. Some studies always collected the same sample volume and others sampled to a particular point (e.g., until the water had run cold from
the tap).

Final LCRI Economic Analysis

5-10

October 2024


-------
Exhibit 5-5: Summary Statistics, Including Geometric Mean, Standard Deviation (SD),
Maximum Value, and Sample Size for Tap Water Lead Concentration Sample Data by LSL and

CCT Status Used in Statistical Modeling





Number of

Geometric
Mean Lead
Concentration
(Hg/L)

Geometric SD

Maximum Lead

LSL Status

CCT Status

Individual

Lead

Concentration





Samples

Concentration

(Hg/L)

LSL

None

2,055

17.79

2.27

170

Partial LSL

None

1,408

8.61

3.9

180

No LSL

None

869

1.15

5.32

119.3

LSL

Partial

4,863

5.15

4.02

2,970

Partial LSL

Partial

217

6.49

2.97

81.13

No LSL

Partial

302

1.86

3.6

36

LSL

Representative

8,129

2.37

3.60

714

Partial LSL

Representative

559

1.63

2.95

38.03

No LSL

Representative

169

0.66

1.98

12.3

Note: The table shows values based on the full dataset used for the analysis shown in Section 5.2.2, not just those
at a particular position in a sampling series. These values were not directly used in blood lead modeling, as they do
not adequately control for repeated sampling within sites and city water systems, or for differences in profile liter.

5.2.1.3 Coding

After cleaning the data as described above, the EPA added a centered "profile liter" term and contrast-
coded variables describing LSL and CCT for use in fitting models. The profile liter term controls for
differences in cumulative sample volume and sampling profile series position, as described in Section
5.2.1.1. Centering the intercept at the mean value of a profile liter for all samples allowed for improved
interpretation of interaction terms. Contrast codes likewise improve interpretability and ease of
projection from the fitted model, particularly when interactions are included.

To produce the centered profile liter term, the EPA calculated the midpoint of the cumulative sample
volume, as described in Section 5.2.1.1. Then, the mean of the original profile liter term was subtracted
from the profile liter term for all samples. This sets the intercept for the model to a profile liter of
approximately 4.5,163 this point is analogous to the fifth liter drawn from the tap after stagnation.

The EPA used the sample data's descriptive information on LSL status to generate two contrast variables
that allow for the statistical comparison of water lead concentrations between the three LSL scenarios
represented in the data ("LSL," "Partial," and "No LSL"). The "LSL (yes/no)" variable indicates lead
concentration samples that come from sites with a full LSL compared to samples from locations with no
LSL. "LSL (no/partial)" designates samples that come from sites with a partial LSL in place compared to
site samples that come from locations with no LSL. Used together in the statistical model, these
two variables allow the EPA to compare water lead concentrations in homes with no LSLs to
concentrations in homes with full LSLs and homes with partial LSLs. Exhibit 5-6 shows the numeric codes
used to describe LSL status in the analysis.

163 The mean of the original, un-centered profile liter term is 4.495. This has been rounded to 4.5, or the "fifth
liter" for readability throughout this document.

Final LCRI Economic Analysis

5-11

October 2024


-------
The EPA determined corrosion control levels for each lead concentration sample reported through
records of CCT practices and implementation dates, as well as water quality samples. CCT was coded as
a single contrast variable, "CCT (yes/no)/' marking those tap samples taken in the presence of corrosion
control against those sample taken at sites without CCT. This variable is used to quantify the difference
between water lead concentrations at sites with representative CCT, partial CCT, and no CCT. Exhibit 5-7
shows the codes used to represent CCT in the analysis.

Exhibit 5-6: Numeric Values Assigned to Two Discrete Contrast Variables Representing LSL
Status in the Estimated Drinking Water Lead Concentration Regression Model

LSL Status

LSL (yes/no)

LSL (no/partial)

LSL

-0.5

0

Partial

0

-0.5

No LSL

0.5

0.5

Exhibit 5-7: Numeric Values Assigned to a Discrete Contrast Variable Representing CCT Status
Use in the Estimated Drinking Water Lead Concentration Regression Model

CCT Status

CCT (yes/no)

None

-0.5

Partial

0

Representative

0.5

5.2.2 Drinking Water Lead Concentration Model Fitting and Selection

Next, the EPA developed a model to estimate typical lead concentrations for each LSL/CCT status
category, or "intervention category". The intervention category is confounded with differences in profile
liter of individual samples, as well as with numbers of sites and profiles from each city water system
(Exhibit 5-4, Exhibit 5-5). Therefore, geometric means cannot be directly compared across intervention
categories. Rather than selecting only samples from some common profile liter (e.g., the "first liter"),
and aggregating within sites and city water systems to produce a homogeneous subset of the dataset,
the EPA fit linear mixed-effects models with explicit terms to statistically control for the differences in
profile liter, city water system, site, and sampling event. This single-step meta-analysis allowed for the
greatest inclusion of available data, while limiting the effects of different methods.

The EPA fit multiple, nested, linear mixed-effects models (Equations 1-5, Exhibit 5-8) of tap water lead
concentration as predicted by LSL presence ("LSL" or "No LSL"), LSL extent ("None" or "Partial"), CCT
status, and profile liter. To simplify model fitting, these models assumed equal variance in lead
concentration among combinations of LSL and CCT status, profile liter, and sampling events. This
assumption means the model may slightly over- or under-estimate the variation in lead water
concentrations in some scenarios. For instance, increased variability may occur as lead flushes from
residential pipes after LSLR; or in cases of "partial" CCT, where poorly optimized and changing corrosion
control may interact unpredictably with pipe chemistry to produce more variable concentrations than
would be expected with fully optimized corrosion control.

Final LCRI Economic Analysis

5-12

October 2024


-------
The EPA compared several models, from simple to more complex, to find the best function for use in
predicting lead concentrations at the tap (Equations 1- 5). For all models, the EPA allowed the intercept
of the fitted equation (mean lead concentration at the fifth liter164 assuming no LSL or CCT; see
Section 5.2.1.2) to vary by sampling event and location, with each sampling event nested within a site
and each site nested within a city water system165. The EPA also considered models that accounted for
random variation in parameters, such as differences in length of a service line among sites, or specific
features that could change the effectiveness of CCT, related to event, site, and city water system, but
found that the collected dataset does not contain sufficient information to fit such models.

To describe the non-linear effects of the profile liter, the models include a natural cubic spline. The
spline models the effect of the profile liter as a curve, and allows the fitted lead concentration to
increase and decrease as water is drawn through the household and service line (Exhibit 5-2 and Exhibit
5-3). This spline included three interior knots and two boundary knots. Knots define points where
different pieces of the curve meet and allow the model to be fit with different "sub" curves for each
piece between the knots. The three interior knots correspond to the first 0.5, 4, and 8 liters after
beginning sampling and have curves at either end. Boundary knots correspond to 0.06 liters and 13 liters
after beginning sampling.166 These knots produce linear sections at either end of the curve where there
were few samples. Models were fitted with the Ime4 package in R (Bates, 2010; Bates et al., 2015;
Pinheiro et al., 2017; R Core Team, 2016) with full-information maximum likelihood (FIML) for model
comparisons and restricted maximum likelihood (REML) to produce final parameters after model
selection (Bolker et al., 2009).167

In the equations below, identifies a particular water sample, and "j" identifies a sampling event,
nested within a site and a city water system. /? refers to the coefficient for each parameter. Thus, /?0j is
the intercept for a particular sampling event j. f>s terms refer to matrices of spline coefficients (not
shown) for each model term that includes a spline. Thus, Psl is the matrix of spline coefficients for the
effect of the profile liter term alone.

The EPA selected the "Reduced spline model with CCT interactions" (Equation 2) to produce simulated
lead concentrations for use in the benefits analysis. Although the most complex "full spline" model
showed the best fit overall (Equation 1, Exhibit 5-8), the improvement in fit was small relative to the
increase in complexity, and a close examination of the fitted model suggested that the full spline model
over-fit specific study parameters and produced predictions that were likely unrealistic. The full model
projected a gradual rise in lead concentration after the service line peak for some intervention

164	The EPA centered the term for profile liter at its mean. As a result, the "fifth liter" occurs at the intercept, which
is close to the fifth liter of a 5- liter sampling series, or roughly 30 seconds of flushing the tap.

165	A 'nested' variable structure represents data where a particular factor level can occur only within a particular
level of another factor. This structure reflects the structure of sample concentration data. A single sampling event
could only occur within a particular site, and that site can only occur within a particular city.

166	Interior knot positions were chosen to represent potentially important transition points in the profile. The fitted
splines were compared against models that used standard quantile selection for interior knot position and were
found to produce similar estimates. Therefore, the knot positions chosen by transition point (i.e., faucet,
beginning, and end of largest service line-related lead increases) were retained for the final model.

167	Using REML reduces bias in the SD of random effects parameters but does not produce meaningful values of
log-likelihood, Akaike Information Criterion (AIC), or Bayesian Information Criterion (BIC) when comparing models
with different fixed effects. Therefore, the EPA used FIML for model comparisons and REML to fit the final model.

Final LCRI Economic Analysis

5-13

October 2024


-------
combinations, which is unlikely to be realistic given that the water represented in this tail of the profile
represents non-stagnated water from the system main. In addition, for homes with no LSLs, the full
model produced predictions of relatively high lead concentrations in homes with representative CCT,
and relatively low concentrations in homes with no CCT. Again, this is unlikely to represent the true
effects of CCT; therefore, the simpler model was selected for simulation.

Full spline model:

l°9e (

Pb

VŁ\

L hi

f30j + Spline (Pro file LiterL) * /?sl + /?, LSL {yes/no)t

+ /32LSL (no/partial)i + f>3CCTi + /34LSL (yes/no)^ * CCTt
+ PsLSL (no/partial)i * CCTt
+ Spline (Pro file Liter{) * /3s2LSL (yes/no)ii
+ Spline (Pro file Litert) * [>s5 LSL (no /partial) t
+ Spline(Profile Litert) * (Js4 CCTt
+ Spline(Profile Litert) * ps5LSL (yes/no)^ * CCTi
+ Spline (Pro file Litert) * /3s6LSL (no/partial)i * CCTt + Łt

Reduced spline model with CCT interactions:

/ Li a \

loge ( Pb	= [>0j + Spline (Pro file Liter•) * /?sl + p±LSL (yes/no)i

L ij

+ (32LSL (no/partiai)i + p3CCT(level)t
+ p4LSL (yes/no) t * CCTt + (35LSL (no/partial)t * CCTt
+ Spline(Profile LiterL) * (]s2LSL (yes/no)i
+ Spline(Profile Litert) * f?s5 LSL (no/partial)t + et

Reduced spline model without CCT interactions:

/ Li a \

loge ( Pb	= [>0j + Spline (Pro file Liter•) * /?sl

^ ij

+ piLSL(yes/no)i + (32LSL (no/partial)t + p3CCTt
+ Spline(Profile LiterL) * /3s2LSL (yes/no)t
+ Spline(Profile Litert) * f?s5 LSL (no/partial)t

+ Łi

(Equation 1)

Spline model with no interactions:

/ I10\

loge {^Pb —J = [>0j + Spline (Pro file Liter•) * /?s

+ p1LSL(yes/no)i + f>2LSL (no/partial)t + p3CCTt

+ ei

Linear model with no interactions

iPb^ =

loge (/

Lhj

Poj + P\ (Profile Litert)

+ /32LSL (yes/no)i + (J3LSL (no/partial)i + P4CCTt

+ Łi

(Equation 2)

(Equation 3)

(Equation 4)

(Equation 5)

Final LCRI Economic Analysis

5-14

October 2024


-------
Exhibit 5-8: Comparison of Tap Sample Lead Concentration Model Results Based on Maximum

Likelihood Estimators for Goodness of Fit

Model

DFa

Log-Likelihood

AIC

BIC

Full spline (Equation 1)

34

-20,425

40,918

41,184

Reduced spline with CCT interaction (Equation 2)

22

-20,515

41,075

41,247

Reduced spline without CCT interaction (Equation 3)

20

-20,535

41,110

41,267

Spline model with no interactions (Equation 4)

12

-20,641

41,306

41,400

Linear with no interactions (Equation 5)

9

-21,220

42,458

42,529

a Degrees of freedom (DF) are the number of parameters estimated for the model, including the variance for each
random effects level (sampling event is nested within the site nested within the city water system), fixed
coefficients, and the residual error. The other columns provide model fit statistics for comparing the fixed terms of
the model. AIC is Akaike's Information Criterion & BIC is Bayesian Information Criterion. For AIC and BIC, the
smaller numbers imply more preferred models; and for log-likelihood, the larger numbers imply a better fit to the
underlying data.

The reduced spline model with CCT interactions suggests that besides water system, residence
(sampling location), and sampling event, the largest effects on lead concentration come from LSLs and
the number of liters drawn since the last stagnation period (Exhibit 5-9).

Exhibit 5-9: Results from the Reduced Cubic Spline Interaction Model with CCT Interactions:
Fixed Effects and Random Effects for Sampling Event, Site, and City Water System

Fixed Effects

Pa

SEb

F (Type III

SS)C

Intercept

1.47

0.156

96

Cumulative volume (spline)

-

-

169

LSL (yes/no)

-1.03

0.104

98

LSL (no/partial)

-0.82

0.191

19

CCT (no/representative)

-0.74

0.163

21

LSL (yes/no)* CCT (no/representative)

0.95

0.181

28

LSL (no/partial)* CCT (no/representative)

1.79

0.32

31

LSL (yes/no)* Cumulative volume (spline)

-

-

-

LSL (no/partial)* Cumulative volume (spline)

-

-

-

Random Effectsd

N

SD

Sampling event in site in city water system

3,130

0.51

Site in city water system

1,657

1.06

City water system

15

0.59

Individual samples (Total N)e

18,571

—

Acronyms: N = number of observations, and SD = standard deviation.

a p, the unstandardized regression coefficient, provides the size and direction of the relationship between each
model term and log-transformed lead concentration, p for spline effects are too complex to show in this exhibit.
b SE shows the standard error estimated for each coefficient.

c F provides the F statistic for each coefficient after controlling for all other coefficients for type III sums of squares
(SS). Unbalanced sample sizes for random effects complicate accurate degrees of freedom (DF) calculations, and
no p-values are provided. However, larger F values indicate stronger effects.

d For random effects, N shows the number of groups at each level, and SD provides the attributable to that level of
random effect.

eTotal N is the number of individual sample bottles.

Final LCRI Economic Analysis

5-15

October 2024


-------
The fitted model for the reduced cubic spline interaction (Equation 2) was used to produce simulated
concentrations of lead at the kitchen tap that statistically control for variation in the sample dataset due
to differences in profile liter among studies, the city water system, the site, and the sampling event.
Additionally, the simulated concentrations incorporate variation in lead levels found among sampling
events, sites, and city water systems.

5.2.3 Simulated Drinking Water Lead Concentrations Based on Selected Model Fit

For use in blood lead modeling, the EPA produced 500,000 simulated lead concentrations based on the
final model (reduced spline with CCT interactions; Equation 2; Exhibit 5-9). These concentrations were
simulated for the first ten profile liter values taken after stagnation (Exhibit 5-10). This approximates
simulations often one-liter samples drawn after stagnation. The simulated dataset includes derived
concentrations for new cities, sites, and sampling events not found in the original dataset using
estimates of variability and uncertainty from the fitted model, and given information on LSL and CCT
status. While the simulated dataset includes variability similar to the original data, individual simulated
estimates are best thought of as central tendencies of possible concentration values given fitted model
parameters and estimated variance. The simulated results also incorporate the model assumptions of
equal variance in lead concentration among different scenarios, and equal variance over the range of
profile liters, as previously discussed in Section 5.2.2.

Final LCRI Economic Analysis

5-16

October 2024


-------
Exhibit 5-10: Estimates for the Simulated Data Showing the Relationship between Tap Lead
Concentration and Profile Liter for Each Combination of CCT and LSL Status

25

20-

15-

_a

CL

10

5-

0-

5	10	15

Liters after last stagnation period

20

i i
i-i
i i

N

i i
—

i i

L_ I

LSL:Yes
CCT: No

LSL: Yes
CCT: Partial

H

LSLYes
CCT:Rep.

LSL:Partial
CCT: No

N

LSL:Partial
CCT: Partial

LSL: Partial
CCT:Rep.

i i

I I

N

i i

LSL:No
CCT: No

LSL:No
CCT:Partial

_ LSL:No
i | CCT:Rep.

Note: Central estimates are solid lines, and 95 percent confidence intervals (CIs) (bootstrapped) are indicated by
shaded areas bounded by dotted lines. The highest concentrations occur, on average, roughly 5 liters after the last
stagnation period in homes with LSLs in place. Note that CIs can overlap somewhat even where there is a
significant effect of scenario (i.e., CCT and LSL presence). However, for scenarios with no LSLs, CIs for CCT scenarios
overlap almost completely.

Though CCT produced significant reductions in lead water concentrations, the simulated predictions for
sites with full LSL removals primarily overlapped for all CCT conditions in the final model Exhibit 5-9).
Therefore, the EPA used the pooled estimate for all CCT conditions in residences with no LSL in place
(this is referred to as "combined CCT"). Because of this overlap in the simulated data, the EPA was
unable to quantify the impacts of improvements in CCT status on non-LSL households using these data.

Exhibit 5-11: LSL and CCT Scenarios and Simulated Geometric Mean Tap Water Lead
Concentrations and Standard Deviations for the First Ten Liters Drawn after Stagnation for

Each Combination of LSL and CCT Status

LSL Status

CCT Status

Simulated Mean
of Log Lead
(W?/L)

Simulated SD" of
Log Lead

Simulated
Geometric Mean
of Lead
(Hg/L)

Simulated
Geometric SDź
of Lead

LSL

None

2.67

1,32

14.38

3.75

Final LCRI Economic Analysis

5-17

October 2024


-------
LSL Status

CCT Status

Simulated Mean
of Log Lead
(Hg/L)

Simulated SDa of
Log Lead

Simulated
Geometric Mean
of Lead
(Hg/L)

Simulated
Geometric SDa
of Lead

Partial LSL

None

1.92

1.33

6.85

3.77

No LSL

None

-0.19b

1.33b

0.83b

3.78b

LSL

Partial

2.07

1.33

7.93

3.77

Partial LSL

Partial

1.35

1.33

3.84

3.78

No LSL

Partial

-0.19b

1.33b

0.83b

3.78b

LSL

Representative

1.45

1.33

4.27

3.78

Partial LSL

Representative

0.76

1.33

2.14

3.78

No LSL

Representative

-0.19b

1.33b

0.83b

3.78b

Acronyms: LSL = lead service line; CCT = corrosion control treatment; SD = standard deviation.
a SD reflects "among-sampling event" variability.

b Bolded values show how simulated results were pooled to produce a common estimate for homes with no LSL
across CCT conditions.

Although the existing data did not provide enough information to estimate the effect of CCT where no
LSL were present (Exhibit 5-10 and Exhibit 5-11), the CCT status of the PWS is tracked in the analysis
regardless of LSL status. This is described in Section 5.3. Note in Exhibit 5-11 that the statistics describing
the distribution of tap water lead concentrations are the same for all three rows for "no LSLs/'
regardless of whether there is representative, partial, or no CCT. Effectively, in the primary analysis the
EPA did not quantify the incremental benefits of CCT when LSLs are absent. On the other hand, because
CCT is done on a system-wide basis, there are no incremental costs associated with providing CCT to
homes without LSLs when it is being provided for the entire system. The impact of CCT for these no LSL
homes likely varies by location depending on whether there are legacy system and/or household lead
solder or higher lead content brass parts.

5.2.4 Determination of GRR, and Point-of-Use and Pitcher Filter Water Lead Concentrations

In addition to modeled drinking water concentrations described above, the following assumptions are
made:

•	The EPA assumes that mean water lead concentrations resulting from galvanized service lines
previously downstream from an LSL are equivalent to the mean water concentration value for
partial LSL replacements as reported in Exhibit 5-11. This assumption may under or overestimate
the change in water lead concentration associated with GRR service line replacements.

•	A point-of-use (POU) device is a water treatment device physically installed or connected to a
single fixture, outlet, or tap to reduce or remove contaminants in drinking water. For the
purposes of subpart I of 40 CFR 141, it must be certified by an American National Standards
Institute (ANSI) accredited certifier to reduce lead in drinking water.

•	A pitcher filter means a non-plumbed water filtration device, which consists of a gravity feed
water filtration cartridge and a filtered drinking water reservoir, that is certified by an ANSI
accredited certifier to reduce lead in drinking water.

Final LCRI Economic Analysis

5-18

October 2024


-------
To estimate benefits, both "POU" devices and filters are considered. Pitcher filters are useful in some
cases, such as mitigating potential short-term increases in lead exposure. Due to the efficiency of these
filters, the EPA chose to assign the lowest-modeled water concentration (0.83 ng/L) to those households
using both pitcher filters and POU devices for the duration of their use, regardless of LSL and CCT status.
In doing so, the EPA assumes that POU devices and pitcher filters are properly used and maintained for
all drinking and cooking, and that the presence of treatment equipment provides the same reduction in
lead exposure as the removal of a lead service line. These assumptions may overestimate this reduction
in lead concentrations from the use of POU devices and pitcher filters.

5.2.5 Limitations of Baseline and Post-Rule Water Concentration Estimates

Although the EPA tried to account for and model variability in lead concentrations at the tap using all
available historical datasets that met inclusion criteria, the underlying data and chosen modeling
strategy have limitations. First, the datasets came from 16 water systems in the United States and
Canada (Exhibit 5-1, Exhibit 5-4, and Exhibit 5-5). Within the United States, datasets include only
samples from the EPA regions 1, 3, and 5. Therefore, the source data do not fully represent water
quality conditions, chemistry differences in pipe scale, possible seasonal differences in leaching, and
treatment practices across all the EPA regions. There was not enough information to include housing
age, which may be related to additional sources of lead. Additionally, the original studies (Exhibit 5-1,
Exhibit 5-4, and Exhibit 5-5) were conducted for different reasons by different entities, and sometimes
varied in their sampling methods. Both of these issues may limit generalizability of the data. See Exhibit
5-7.

The simulated concentrations statistically control for differences in methodology among studies by
standardizing the "profile liter" term and including random effects to control for repeated samples
within sampling event, site, and city water system. This approach is not equivalent to conducting a large
new study to collect consistent samples over a broader variety of water systems. As previously
discussed, using simulated concentrations also incorporates some assumptions, such as equal variance
in lead concentrations among different combinations of CCT and LSL status.

The resulting drinking water concentrations were minimally changed after incorporating the new data
which became available the 2021 LCRR analysis was conducted. For a further description of the
uncertainties and variabilities in the data, see Stanek et al. (2020).

5.3 Assignment of Drinking Water Lead Tap Concentrations to PWS Populations

This section first describes how the simulated drinking water concentrations described in Section 5.2 are
assigned to each type of PWS, and next describes how the number of people in each PWS are estimated
and tracked through the analysis period. Each tap water lead concentration displayed in Exhibit 5-11 is
assigned to the various LSL, POU, pitcher filter, and CCT scenarios under the final rule. Due to data
limitations, some scenarios have been assigned the same lead tap water concentration. As illustrated in
Exhibit 5-10, in the case where there is no LSL, confidence limits on modeled drinking water
concentrations, regardless of CCT status, all overlap. Therefore, as described in Section 5.2.1, these were
combined in the analysis. It is possible that given more data, one might expect to see lower drinking
water lead levels when CCT is optimized, however the available data did not allow the EPA to update

Final LCRI Economic Analysis

5-19

October 2024


-------
this assumption. For this reason, the EPA kept these scenarios separate in the benefits modeling,
including tracking the number of people in PWSs with this LSL/CCT status.168

Mapping Exhibit 5-11 drinking water concentrations to modeled benefit scenarios is illustrated in Exhibit
5-12.

Exhibit 5-12: Mapping Simulated Drinking Water Lead Tap Concentrations to

Benefit Scenarios

LSL Status

CCT Status

Geometric Mean Tap

Water Lead
Concentration3 (|ig/L)

Geometric SD

LSL

None

14.38

3.75

Partial LSL/GRR

None

6.85

3.77

No LSL

None

0.83b

3.78b

LSL

Partial

7.93

3.77

Partial LSL/GRR

Partial

3.84

3.78

No LSL

Partial

0.83b

3.78b

LSL

Representative

4.27

3.78

Partial LSL/GRR

Representative

2.14

3.78

No LSL

Representative

0.83b

3.78b

POU and pitcher filters0

0.83b

3.78b

a Simulated geometric mean water concentrations are based on available data for various LSL and CCT
scenarios, as described in Section 5.2.3.

b Bolded values show how simulated results were pooled to produce a common estimate for homes with
no LSLs across CCT conditions. Also, these "No LSL" values were used for POU lead tap concentrations.
cThis value is used for all POU and pitcher filters, for the duration of use.

The EPA estimated benefits under both low and high cost scenarios used in the final LCRI
that characterize some of the uncertainty in the cost estimates. The low and high cost scenarios differ in
their assumptions about the number of PWSs above the action level (AL) which impacts the number of
systems installing or re-optimizing CCT, the number of small systems selecting POU as a compliance
alternative, and the number of water systems that supply temporary filters (POU or pitcher filters) due
to multiple AL exceedances (ALEs). This difference in estimated ALE between the low and high scenario
affects the timing and proportions of PWS populations that move from pre-regulatory, or baseline,
water lead concentration values to new post-regulatory lead tap water concentration categories. Both
pre- and post-regulatory water lead concentrations are shown in Exhibit 5-12.169

The monetary benefits of the rule are modeled in the SafeWater Lead and Copper Rule (LCR) model. For
each model PWS, a population cohort is created in the SafeWater LCR model. Each simulated population
cohort for each PWS has an age distribution equal to that of the general population and is followed in

168	Note that these are also tracked in the cost side of the model (see Chapter 4).

169	Note additional differences between lower and higher scenarios assumptions with regard to concentration
response functions for the IQ, ADHD, and adult CVD premature mortality health endpoints also define the range
between the estimated low and high scenario benefits but do not affect population movement between categories
in the SafeWater LCR model. See Section 5.5 for additional detail.

Final LCRI Economic Analysis

5-20

October 2024


-------
the benefits analysis for 35 years. A new cohort of infants is added in each subsequent year of the 35-
year analysis based on birth rates from the 2020 United States Census. Thus, 35 cohorts of people are
modeled and can possibly accrue benefits due to implementation of the rule - one cohort for each year
of the analysis.

In the analysis, the EPA assumes that characteristics of households with LSLs have the same
characteristics as the general population in regards to the age and number of people living in the
home.170 Each statistical person within a model PWS in the SafeWater LCR model is initially assigned to
one of the simulated drinking water lead concentrations in Exhibit 5-11, depending on the CCT status
and number of LSLs assigned to that modeled system in the baseline. Depending on the rule
requirements, implementation schedule, and each year's tracked system level 90th percentile tap sample
lead concentration, a modeled PWS may experience changes in CCT, pitcher filter and POU status if the
system has an ALE. The modeled PWS will also experience a decrease in the number of LSL/GRRs apart
from an ALE given the proactive replacement requirements in the final LCRI. Based on these modeled
changes in CCT, pitcher filter, POU, and LSL/GRR status, specific proportions of the modeled population
within the system will be assigned to a new lead tap water concentration category representing the new
technology in place at the system in each year of the 35-year period of performance.

For a further discussion of how this is implemented in the SafeWater LCR model, see Section 4.3.5 and
the flow charts in Appendix B; and for the following discussions, see Section B.3, Estimating the Cost of
Compliance with the LCRI:

•	Small community water systems (CWSs) serving 3,300 or fewer people and all non-transient
non-community water systems (NTNCWSs) can choose between POU and CCT and are assumed
to choose whichever compliance option has the lowest cost.

•	How the SafeWater LCR model determines if a PWS installs/optimizes CCT or installs a POU
device.

Due to different assumptions, different numbers of people will experience benefits under the low- and
high-cost scenarios. The Safe Water LCR model tracks the number of people who move from one
treatment combination (i.e., beginning condition) to another (i.e., ending condition) over the 35-year
analysis period across all model PWS strata under the low- and high-cost scenarios, respectively. Each
population is assigned a concentration from Exhibit 5-12. In the case of a CCT installation or re-
optimization, the entire population of a model PWS will move to the new CCT status at the same time.
The EPA also assumes that the entire PWS moves to the drinking water lead concentration (from Exhibit
5-12), assigned to a POU device when this option is implemented by the PWS, which implies that
everyone is properly using the POU device. Thus, a corresponding change in the concentration of lead in

170 Note the EPA has insufficient data at the national level to model the potential correlations between at risk
populations and housing characteristics that might put individuals at higher risk of lead exposure like presence of
LSL, lead paint, and housing age. The case studies were conducted in the Environmental Justice analysis for the
LCRI found that older housing stock. Housing is associated with higher lead-dust and higher blood lead levels in
children, and potentially also with location of LSLs.

Final LCRI Economic Analysis

5-21

October 2024


-------
drinking water will occur for the entire PWS population in the year the change is implemented. Chapter
4 provides more detail on these assumptions.171

The portion of the population corresponding to the number of households undergoing service line
replacement each year will change to the lower drinking water lead concentrations and BLLs in the year
the LSL/GRR service line is replaced. To simplify the analysis, the EPA assumes no change in other
sources of lead exposure besides drinking water over the 35-year timeline used in the analysis. This
includes exposure to lead in drinking water not consumed in the home.

The EPA did not quantify the benefits of reduced lead exposure to individuals who reside and work in
buildings that do not have LSL/GRR service lines. These buildings, while not having an LSL/GRR service
line in place, may still contain leaded plumbing materials, including leaded brass fixtures and leaded
solder. The EPA expects that the final LCRI requirements will result in reduced lead exposure to the
occupants of these buildings as a result of improved monitoring and additional actions to optimize CCT.

In the final LCRI analysis, the EPA assumes there is no difference in the geometric mean water lead
concentration in households with no LSL, regardless of CCT status. In other words, for each of the
three scenarios of no LSL (i.e., no LSL - no CCT, no LSL - partial CCT, and no LSL - representative CCT),
the geometric mean water lead concentration is equivalent to 0.83 ng/L, which is likely to lead to an
underestimate of benefits. The EPA made this same assumption of a constant geometric mean water
lead concentration in households with no LSL, regardless of CCT status as part of the assessment of the
2021 LCRR, and as part of the 2021 LCRR analysis the EPA attempted to assess the potential benefits to
children in these non-LSL homes where CCT was improved. This approach assumed that there is still a
benefit to CCT in the absence of an LSL due to the potential for reduced leaching of lead from internal
fixtures such as lead solder on brass water faucets that would result in lower water lead concentrations.
The agency has determined that the data are too limited and the uncertainties too significant to include
this assessment in the quantified and monetized benefit estimates of the LCRI regulation. This analysis
was presented in Appendix G of the Final 2021 LCRR EA (USEPA, 2020a).

171 Under the final LCRI, PWSs with multiple ALEs must make temporary POU or pitcher filters available to all
customers with service lines with lead or potential lead content. The EPA has assumed that 100% of these
customers would pick-up and use these filters. This equates to between 9.7 and 17.1 million people who will use
temporary POU or pitcher filters until their service line is replaced or confirmed not to have lead content.

Final LCRI Economic Analysis	5-22	October 2024


-------
5.4 Methods for Estimating Blood Lead Levels

The EPA assessed benefits of the final LCRI in terms of avoided losses in IQ in children aged 0-7, which
required estimating blood lead levels in this age group using the SHEDS-Pb model (formerly known as
SHEDS-IEUBK) (Zartarian et al. 2017, 2023; Stanek et al. 2020). See Zartarian et al., (2023) for an updated
evaluation of this model. For the LCRI analysis the EPA used the SAS version of the model which was
tailored during the 2021 LCRR to handle several different drinking water inputs. Additionally, as
described in Section 5.5, the EPA estimated avoided cases of ADHD in children that result from
additional actions required under the final rule. The EPA also estimated the avoided lower birth weight
in infants due to reductions in their mother's lead exposure, and avoided cases of CVD premature
mortality in both adult men and women. For older children and adults, the EPA uses the All-Ages Blood
Lead Model (AALM), version 3 to estimate BLLs (USEPA, 2024b).

Section 5.4.1 describes methods used to estimate BLLs in young children. Section 5.4.2 describes the
estimation of BLLs in older children and adults.

5.4.1 Methods for Estimating Blood Lead Levels in Children Ages 0-7

This modeling approach follows that described in Stanek et al., (2020) with minor updates to drinking
water inputs, and soil and dust inputs to remain consist with other agency lead rulemakings. (Henning et
al. 2024; Zartarian et al. 2023). The model used for children is the SHEDS-Pb model, which couples two
existing models (The Stochastic Human Exposure and Dose Simulation (SHEDS)-Multimedia model
described briefly in Section 5.4.1.1 and the integrated exposure uptake biokinetic (IEUBK) model,
described briefly in Section 5.4.1.2). The SHEDS-Pb model has been evaluated against an empirical
model (Henning et al., 2024).

Estimating benefits of the final LCRI in children requires estimates of BLLs from ages Oto 7. Specifically,
to estimate the effects of lead exposures on IQ, estimates of BLLs in each year of life from ages 0 to 7
and lifetime BLLs to age 7 are needed.

The agency compiled available environmental lead concentration data across various media (i.e., soil,
air, food, and water). The lead concentration estimates for soil, air, and food in this analysis are held
constant in the blood lead modeling in order to represent background lead levels172, with the only
varying concentration being drinking water. In order to estimate the potential changes in BLLs that

172 With regard to adjusting baseline exposure over the 35-year period of analysis, in response to potential future
changes in the non-water exposure pathways, the EPA has found no way to extrapolate lead concentrations over
the period of analysis for the alternative media that would not introduce significant uncertainty into the BLL
estimation modeling likely outweigh any perceived improvement. National level projections of decreases in lead
dust and soil, and food concentrations over a decades long period are not available and projection of current
trends would introduce significant uncertainty. Using projected values from NHANES to adjust estimated BLLs post
modeling would also introduce significant uncertainty, as the rate of decrease in BLLs observed in NHANES has
slowed in recent years. Therefore, the EPA held constant the non-water sources of lead over the 35-year analysis
period. The EPA also notes that because of the log-linear shape of the concentration response function for IQ, and
that the concentration response functions show no evidence of a threshold below which effects cease, projected
policy scenario reductions in water lead concentrations that occur at lower baseline water lead levels (WLLs) would
result in larger changes in avoided IQ point, therefore the inclusion of lower baseline values would result in
increased benefit estimates associated with the final LCRI.

Final LCRI Economic Analysis

5-23

October 2024


-------
result from the final LCRI requirements, the EPA used several modeled/estimated drinking water lead
concentration values associated with drinking water system scenarios that represent possible
combinations of CCT, POU, pitcher filters, and LSL/GRR service line status, as described in Section 5.3
above and Exhibit 5-12. Following the methodology outlined in Stanek et al., (2020) the EPA used these
inputs to relate total lead absorbed into the body to a set of BLLs representing the different CCT, POU,
pitcher filter, and LSL/GRR service line scenarios.

This section begins with an overview of the two models used to produce these estimates, followed by a
description of the methods for coupling the models.

5.4.1.1	SHEDS-Multimedia Modeling

SHEDS-Multimedia is a probabilistic model that simulates aggregate (i.e., multimedia) and cumulative
population exposures to chemicals over space and time based on realistic activity patterns,
concentration distributions, and exposure factors. Full details of this model and the inputs have been
previously described (Zartarian et al., 2006, 2012, 2023; Xue et al., 2010, 2012a, 2012b, 2014a, 2014b;
Glen et al., 2012; USEPA, 2016, Stanek et al., 2020). A brief overview is provided in this section.

The SHEDS-Multimedia model has undergone numerous peer reviews and has been well-validated for
use in assessing exposures to diverse chemicals (USEPA, 2016). SHEDS-Multimedia provides exposure
estimates as a result of both dietary and residential exposures, and it can be used to estimate these
exposures by sex and age. SHEDS-Multimedia has the capability to assess exposures via ingestion,
inhalation, and dermal pathways. However, dermal exposures are not considered in the current analysis
because lead exposures through this pathway are assumed to be negligible.

SHEDS-Multimedia has several strengths that make it a powerful tool to assess chemical exposures, such
as the ability to consider correlated inputs (e.g., correlations between concentrations of contaminants in
dust and soil); and the use of two-stage Monte Carlo sampling, which allows variability in population
exposure and dose estimates, and uncertainty associated with different percentiles to be quantified.

5.4.1.2	lEUBKModel

The IEUBK model was developed as a simulation tool to predict BLLs in children from birth up to age 7
and thereby assist in the risk assessment of contaminated sites (USEPA, 1994). The model is intended to
"enable rapid calculations and recalculations of an extremely complex set of equations that includes
scores of exposure, uptake, and biokinetic parameters" (USEPA, 1994, p. 1-1). It provides an estimate of
the BLL for a population of similarly exposed children associated with specified concentrations of lead in
media (e.g., water, soil) in the child's environment (USEPA, 2007). In addition, the IEUBK model
estimates the probability that a population of similarly exposed children with a given exposure scenario
will have a BLL greater than a specified level. Users can modify inputs and assumptions within the model
(e.g., concentrations of lead in environmental media, intake rates for environmental media) to explore
the effects on children's BLLs.

The IEUBK model uses four main components to mathematically and statistically link environmental lead
exposure to children's BLLs: exposure, uptake, biokinetics, and variability (White et al., 1998). Exposures
are quantified by combining information on the concentration of lead in environmental media, the
amount of contact with the media (e.g., amount of drinking water ingested per day), and the duration of

Final LCRI Economic Analysis

5-24

October 2024


-------
the contact (e.g., number of days) (White et al., 1998). The environmental media included in the IEUBK
model are drinking water, soil, household dust, air, and food; exposure to lead based paint is assessed
via its contribution to household dust and soil concentrations (White et al., 1998). The uptake
component models the transfer of lead to the bloodstream (i.e., the absorption) after intake into the
child's body via inhalation or ingestion routes. In the present analysis, the EPA used information from
the IEUBK model on uptake and biokinetics only, as further described below in the SHEDS-Pb coupling
section.

5.4.1.3 Background Lead Exposure Inputs into SHEDS-Pb

Exhibit 5-13 to Exhibit 5-16 provide a summary of the main inputs for the SHEDS-Pb analyses, which
were previously published in the supplemental material of Zartarian et al. (2017) and a few updates are
noted in this section. The final LCRI analysis uses updated soil and dust concentrations. The soil and dust
concentrations are still based on the United States Department of Housing and Urban Development
(HUD, 2011) American Healthy Homes Survey I (AHHS) 2005-2006 data, and now also include the AHHS
II data (2018-2019) (USHUD, 2021). These concentrations are summarized in Exhibit 5-15 as the
geometric means and geometric standard deviations of the lognormally distributed data and are
consistent with Henning et al. 2024. The estimates for daily water consumption are used in conjunction
with the set of drinking water system scenario modeled lead concentrations in Exhibit 5-12. The other
levels, such as daily lead from food, dust, and soil are used as background and do not change across
drinking water system scenarios.

Exhibit 5-13: Summary of Daily Water Consumption Inputs for Drinking Water Consumption

in SHEDS-Pb Coupling (Zartarian et al., 2017)

Daily Water Consumption (mL/day)
NHANES3 2005-2012

Age
(years)

Age
(months)b

N

Mean

SD

50th
Percentile

Geometric
Mean

Geometric
SD

75th
Percentile

95th Percentile

99th
Percentile

06
months

0-6c

1,246

662

320

630

526

2.5

854

1,216

1,481

0

0-1 lc

2,618

581

349

532

410

3.0

806

1,172

1,489

1

12-23

1,792

247

247

219

151

3.3

306

690

1,148

2

24-35

1,948

300

312

251

176

3.4

360

909

1,424

3

36-47

1,272

316

313

257

193

3.1

398

917

1,640

4

48-59

1,358

320

333

261

197

3.2

404

874

1,434

5

60-71

1,196

364

366

303

213

3.5

447

1,037

1,802

6

72-83

1,306

377

353

332

228

3.5

480

1,067

1,601

Acronyms: N = number of observations, and SD = standard deviation.

Source: Zartarian et al. (2017).

Note: This exhibit summarizes drinking water consumption values that were used as inputs for the SHEDS-Pb
analysis. These values were previously published in the supplemental material of Zartarian et al. (2017).
a The National Health and Nutrition Examination Survey (NHANES) is a program of studies designed to assess the
health and nutritional status of adults and children in the United States. It provides nationally representative data
on the United States population, including estimates of drinking water consumption.
b Age in months was added to clarify the age ranges listed in years.

c Water consumption for 0-11 months was used in the modeling for 6-11 month-old infants.

Final LCRI Economic Analysis

5-25

October 2024


-------
Exhibit 5-14: Summary of Daily Inputs for Dietary Lead Intake (|ig/day) in SHEDS-Pb (Zartarian

et al. (2017))

Age
(years)

Age
(months)3

N

Mean

SD

Median

Geometric
Mean

Geometric
SD

75th
Percentile

95th
Percentile

99th
Percentile

06
months

0-6

1,072

0.7

0.98

0.3

0.27

4.75

0.91

2.71

3.47

1

12-23

2,226

2.58

1.84

2.17

2

2.16

3.41

5.83

7.63

2

24-35

1,788

3.44

2.03

3.06

2.85

1.94

4.49

7.23

8.46

3

36-47

1,160

3.54

2.06

3.18

2.98

1.89

4.63

7.26

8.43

4

48-59

1,240

3.57

2.16

3.18

3

1.87

4.55

7.25

8.63

5

60-71

1,066

3.85

2.18

3.43

3.31

1.77

4.83

7.86

9.52

6

72-83

1,086

3.8

2.02

3.51

3.29

1.76

4.84

7.55

8.3

Acronyms: N = number of observations, and SD = standard deviation.

Notes: This exhibit summarizes dietary lead intake values that were used as inputs for the SHEDS-Pb analysis.

These values were previously published in the supplemental material of Zartarian et al. (2017).

Data sources: United States Food and Drug Administration's (FDA's) Total Diet Study (TDS) 2007-2013 and recipe

mapping data from the Center for Food Safety and Applied Nutrition (CFSAN).

Method source: Xue et al., 2010. N=Number of Observations

a Age in months was added to clarify the age ranges listed in years. Data for 6-11 month-old infants not available.

Exhibit 5-15: Summary of Inputs for Soil and Dust Lead Concentration (|ig/gram) in SHEDS-Pb

Coupling (USHUD 2011, 2021)

Data Source

Housing Vintage

Dust Concentration
(GM, GSD)

Soil Concentration
(GM, GSD)



Pre-1940

177.98, 2.23

267.6, 3.37

AHHSI +AHHSII

1940-1959

114.20, 2.20

73.94, 2.88

1960-1977

75.40, 2.00

29.06, 2.47



Post-1978

52.87, 2.04

14.54, 2.46

Acronyms: N = number of observations; SD = standard deviation

Source: United States Department of Housing and Urban Development (USHUD, 2011) American Healthy Homes
Survey I (AHHS) 2005-2006 data, and the AHHS II data (2018-2019) (USHUD, 2021)Notes: This exhibit summarizes
soil and dust lead concentration values for the lognormal distribution for each of the four vintages of housing for
the combined survey data that were used as inputs for the SHEDS-Pb analysis. These values were updated from
Zartarian et al. (2017). They now include the United States Department of Housing and Urban Development

Final LCRI Economic Analysis

5-26

October 2024


-------
(USHUD, 2011) American Healthy Homes Survey I (AHHS) 2005-2006 data, and the AHHS II data (2018-2019)
(USHUD, 2021).

Exhibit 5-16: Summary of Daily Inputs for Soil/Dust Ingestion (mg/day) in SHEDS-Pb

(Ozkaynak et al., 2022)

Age

Soil/Dust

Mean

SD

50th Percentile

Geometric Mean

Geometric SD

95th Percentile

0-
-------
Applying these absorption fractions results in distributions of lead uptake by exposure/media route,
which can be summed across routes to give total lead uptake per day (ng/day). Next, the EPA used age-
based relationships derived from IEUBK to relate these lead uptakes (ng/day) to BLLs (ng/dL). The aim
was to develop a "reduced form" of the IEUBK model, allowing BLL distributions to be efficiently
estimated without having to apply the full version of the IEUBK model. Specifically, as described in
Zartarian et al. (2017, 2023) and Stanek et al. (2020) the EPA developed regression equations between
lead uptake and blood lead by running IEUBK with increasing amounts of lead intake. Since the
relationship between lead uptake and blood lead in IEUBK is not perfectly linear, SHEDS-Pb uses a
polynomial regression to address the slight departures from linearity, which represent the non-linear
binding of lead to red blood cells. Exhibit 5-18 shows age-specific regressions used to describe an age-
dependent relationship relating lead uptake to blood lead. The coefficients pertain to a third-order
polynomial regression of the form:

Blood lead (ng/dL) = p0 + Pi Uptake + p2 Uptake2 + p3 Uptake3 + e	(Equation 6)

Coefficients for the month that represents the midpoint of the age range of interest were used in the
analyses.

Exhibit 5-18: Age-Specific Polynomial Regressions Equations for Approximating IEUBK

(Zartarian et al., 2017)

IEUBK Age
Interval
(year)

Age
(months)3

Po

Pi

P2

P3

0.5-1

6-11

7.86E-03

5.47E-01

-1.31E-03

6.01E-6

1-2

12-23

-3.11E-04

4.47E-01

-6.37E-04

1.53E-6

2-3

24-35

1.23E-03

3.79E-01

-4.29E-04

8.45E-7

3-4

36-47

6.58E-04

3.55E-01

-3.71E-04

6.24E-7

4-5

48-59

6.36E-04

3.36E-01

-3.38E-04

5.44E-7

5-6

60-71

1.65E-03

3.13E-01

-2.78E-04

3.57E-7

6-7

72-83

1.32E-04

2.88E-01

-2.30E-04

3.08E-7

Source: Zartarian et al. (2017).

Notes: R2> 0.995. This exhibit summarizes the coefficients used for age-specific IEUBK modeling to predict BLLs.
a Age in months was added for consistency across input tables.

To account for biological variability, the EPA applied a biological variance correction factor of 0.185 for
1- to < 2-year-olds and 0.176 for 2- to < 7-year-olds to the predicted blood lead variance estimated by
the SHEDS-Pb model. Additional details about the calculation of these biological variance correction
factors can be found in Zartarian et al. (2017).

Zartarian et al. (2017) compared estimates generated using SHEDS-Pb to BLL estimates reported from
the National Health and Nutrition Examination Survey (NHANES) (2013-2014) and from the National
Human Exposure Assessment Survey (NHEXAS) and were shown to closely estimate these BLLs. For
further information on SHEDS-Pb model development and evaluation, refer to Zartarian et al.'s (2017)
paper, "Children's Lead Exposure: A Multimedia Modeling Analysis to Guide Public Health Decision-
Making." Additionally, Henning et al., (2024) further validated the SHEDS-Pb model using data from
NHANES 2011-2016.

Final LCRI Economic Analysis

5-28

October 2024


-------
5.4.1.5 Estimates of Pre- and Post-Rule Blood Lead Levels in Young Children

Exhibit 5-19 presents modeled SHEDS-Pb geometric mean BLLs in children by year of life. The BLLs in this
exhibit represent what children's BLLs would be if they lived under the corresponding drinking water
system scenario for their entire lives from birth to age 7. These BLLs are used as inputs for the
representative children in each corresponding PWS for the benefits modeling, and do not represent
weighted population estimates. In the SafeWater LCR model analyses of benefits, the EPA estimated
lifetime BLLs from these values by taking the average of the BLLs for each year of the child's life, up to
age 7, based on their drinking water system scenario status during each year. The age at implementation
of the rule was taken into account when calculating lifetime average BLLs. If, for example, the child is
age 3 at implementation of the rule, the EPA would calculate lifetime average BLLs by averaging 3 years
of pre-rule BLLs and 4 years of post-rule BLLs. Or, if the child is age 5 at implementation of the rule, the
EPA would calculate lifetime average BLLs by averaging 5 years of pre-rule BLLs and 2 years of post-rule
BLLs. The column labeled "Average" contains calculated average lifetime BLLs, assuming a child lived in
the corresponding LSL/GRR service line, CCT, POU, or pitcher filter scenario for their entire life.

Exhibit 5-19: Modeled SHEDS-Pb Geometric Mean (GM) Blood Lead Levels in Children for Each
Possible Drinking Water Lead Exposure Scenario for Each Year of Life





Water

Concentration

GM Blood Lead Level (|ig/dL)b for
Specified Year of Life

Lead
Service

Line
Status

Corrosion
Control
Treatment
Status

0-1=

1-2

2-3

3-4

4-5

5-6

6-7

Avg.c

LSL

None

14.38

4.94

2.74

2.82

2.71

2.78

2.95

2.61

3.08

Partial
LSL/GRR

None

6.85

3.12

1.98

2.01

2.01

2.01

2.08

1.84

2.15

No LSL

None

0.83

1.19

1.28

1.30

1.28

1.30

1.39

1.10

1.26

LSL

Partial

7.93

3.27

2.11

2.13

2.10

2.08

2.21

1.95

2.27

Partial
LSL/GRR

Partial

3.84

2.18

1.64

1.66

1.68

1.64

1.72

1.47

1.71

No LSL

Partial

0.83

1.19

1.28

1.30

1.28

1.30

1.39

1.10

1.26

LSL

Representative

4.27

2.36

1.72

1.73

1.74

1.73

1.80

1.53

1.80

Partial
LSL/GRR

Representative

2.14

1.65

1.47

1.45

1.47

1.46

1.51

1.28

1.47

No LSL

Representative

0.83

1.19

1.28

1.30

1.28

1.30

1.39

1.10

1.26

POU or pitcher filter

0.83

1.19

1.28

1.30

1.28

1.30

1.39

1.10

1.26

3 Blood lead levels for the first year of life are based on regression from IEUBK for 0.5- to 1-year-olds only.
b These values represent the blood lead for a child living with the LSL/CCT status in the columns to the left. Each
year blood lead corresponding to actual modeled child is summed and divided by 7 in the model to estimate
lifetime average blood lead.

Final LCR! Economic Analysis

5-29

October 2024


-------
c This column contains calculated average lifetime blood lead levels assuming a child lived in the corresponding
LSL/GRR service line, CCT, POU, or pitcher filter scenario for their entire life.

Changes in the geometric mean BLL averages in children ages 0-7 due to the anticipated changes in lead
drinking water concentration associated with the final LCRI regulatory requirements are summarized in
Exhibit 5-20.

Exhibit 5-20: Anticipated Decreases in Blood Lead Levels in Children

Pre-Rule Drinking Water

Post-Rule Drinking Water

Estimated
Decrease in
GM BLL
(Hg/L)

Water
Lead
Cone.
(Hg/L)

BLL Kictt

LSL Status

(Hg/L)

CCT Status

Water
Lead
Cone.
(Hg/L)

BLL LSL
(|ig/L) Status

CCT Status

14.38

3.08 LSL

None

0.83

1.26 No LSL

None

1.82

14.38

3.08 LSL

None

4.27

1.8 LSL

Representative

1.28

14.38

3.08 LSL

None

0.83

1.26 No LSL

Representative

1.82

14.38

3.08 LSL

None

0.83

1.26

POU or pitcher filter

1.82

6.85

2.15 Partial/GRR

None

0.83

1.26 No LSL

None

0.89

6.85

2.15 Partial/GRR

None

2.14

1.47 Partial

Representative

0.68

6.85

2.15 Partial/GRR

None

0.83

1.26 No LSL

Representative

0.89

6.85

2.15 Partial/GRR

None

0.83

1.26

POU or pitcher filter

0.89

0.83

1.26 No LSL

None

0.83

1.26 No LSL

Representative

0

0.83

1.26 No LSL

None

0.83

1.26

POU or pitcher filter

0

7.93

2.27 LSL

Partial

0.83

1.26 No LSL

Partial

1.01

7.93

2.27 LSL

Partial

4.27

1.8 LSL

Representative

0.47

7.93

2.27 LSL

Partial

0.83

1.26 No LSL

Representative

1.01

7.93

2.27 LSL

Partial

0.83

1.26

POU or pitcher filter

1.01

3.84

1.71 Partial/GRR

Partial

0.83

1.26 No LSL

Partial

0.45

3.84

1.71 Partial/GRR

Partial

2.14

1.47 Partial

Representative

0.24

3.84

1.71 Partial/GRR

Partial

0.83

1.26 No LSL

Representative

0.45

3.84

1.71 Partial/GRR

Partial

0.83

1.26

POU or pitcher filter

0.45

Final LCRI Economic Analysis

5-30

October 2024


-------
Pre-Rule Drinking Water

Post-Rule Drinking Water

Estimated
Decrease in
GM BLL
(Hg/L)

Water
Lead
Cone.
(Hg/L)

BLL KICtt

LSL Status

(Hg/L)

CCT Status

Water
Lead
Cone.
(Hg/L)

BLL LSL
(|ig/L) Status

CCT Status

0.83

1.26 No LSL

Partial

0.83

1.26 No LSL

Representative

0

0.83

1.26 No LSL

Partial

0.83

1.26

POU or pitcher filter

0

4.27

1.8 LSL

Representative

0.83

1.26 No LSL

Representative

0.54

4.27

1.8 LSL

Representative

0.83

1.26

POU or pitcher filter

0.54

2.14

1.47 Partial/GRR

Representative

0.83

1.26 No LSL

Representative

0.21

2.14

1.47 Partial/GRR

Representative

0.83

1.26

POU or pitcher filter

0.21

0.83

1.26 No LSL

Representative

0.83

1.26

POU or pitcher filter

0

5.4.2 Methods for Estimating Blood Lead Levels in Older Children and Adults

The EPA estimated the BLLs associated with exposure from drinking water in older children and adults.
The EPA estimated BLLs in older children and adults for each year of life, beginning at age 8 and ending
at age 79. The EPA assessed males and females.

5.4.2.1 Overview of the All Ages Lead Model

The EPA's All Ages Lead Model (AALM) tool relates exposure to lead from various exposure media over a
lifetime to lead levels in blood and other tissue (USEPA, 2024c). The AALM includes developments from
previous models, including the IEUBK model among other lead models. The tool consists of a lead
exposure model and a lead biokinetics model. User inputs for selected environmental media (soil, dust,
water, air, and food) are used in the exposure model to predict lead intake per day for a simulated
individual accounting for sex and age differences. The AALM tool produces an estimate of lead
concentration in various tissues and excreta, including estimate of blood lead levels over a lifetime. The
calculation for lead intake is generally summarized in the following equations:

Lead intake from air, dust, soil, and water:

INmedium = Y^t=i(P^me^umi ' fmediumi) • IRmedium Equation (7)

•	INmedium =Pb intake rate (figPB/day) for a specific environmental medium

•	Pbmediumi = exposure concentration (e.g., figPB/L water) in that medium for a given
exposure setting i

•	f medium = fraction of total intake of the medium that occurs in setting i

•	IRmedium = intake rate of medium (e.g., L water/day)

•

Final LCRI Economic Analysis

5-31

October 2024


-------
Lead intake from food and other gastrointestinal sources:

INmedium = Ylt=i(Pbintakei )	Equation (8)

• Pbintakei = rate ofPb intake (fig PB/day) entered for the medium for exposure setting i

Analysis in All Ages Lead Model (AALM)

The EPA released AALM version 2.0 for public use in September 2019. It was peer reviewed by the EPA's
Science Advisory Board (SAB), and their report was released in August 2020. The agency began updating
the AALM version 2.0 in 2021, which included updates to the user interface and to the code, including a
revised lung model. The beta version of the updated model was used by the EPA to estimate the blood
lead levels for the proposed LCRI. For the final LCRI modeling presented here, the EPA used AALM
version 3.0, which is publicly available and reflects updates resulting from the peer review process.
(USEPA, 2024b). The AALM model was evaluated for performance in predicting adult blood levels
against available data from exposed workers and was shown to perform reasonably well, although with
some variability (USEPA, 2019b).

For each combination of LSL and CCT status, the geometric mean water concentrations, as presented in
Exhibit 5-12, were entered as the lead concentration in water that a modeled individual, male or
females was exposed to throughout their lifetime from age 0 to 79 years. Exhibit 5-21 displays the
constant variables entered into the AALM for water, soil, dust, and air. The AALM will apply the intake or
concentration for all age years after the age where the value is entered, or until another value is
specified at another age. To complete this modelling, the intake and concentrations between ages were
interpolated. Lead intake from food was estimated in AALM using the recommended inputs in the EPA's
Technical Support Document for the All Ages Lead Model (AALM) version 3.0 - Parameters, Equations,
and Evaluations (2024), and intakes differed by sex. These inputs are based on lead contamination in
food and beverages, excluding drinking water. The values do not include contamination of food
introduced during food preparation (e.g. dust contaminated surfaces). The values in Exhibit 5-21 follow
the AALM guidance for determining model inputs for lead in food.

Exhibit 5-21: Constant Variables Entered into the AALM for Both Sexes

Media parameter (units)

Age (years)
value starts

Values
Entered

Data Source

Water intake (L/day)

0

0.2

AALM default values (Technical Support Document,
pp. 281-282) adapted from the 2011 Exposure
Factors Handbook, Table 3-1



0.25

0.3





1

0.35





10

0.45





15

0.55





25

0.7





50

1.04



Final LCRI Economic Analysis

5-32

October 2024


-------
Media parameter (units)

Age (years)
value starts

Values
Entered

Data Source

Soil

concentration (ng/g)

0

61.3365

Weighted average based on data of lead-based paint
homes from HUD's American Healthy Homes Survey
(AHHS) 1 and II Lead Findings report (Exhibit
5-15)(USHUD, 2011, USHUD, 2021)

Soil intake (g/day)

0

0

Ozkaynak et al. 2022, Table 2 (Geometric mean)



0.083

0





0.25

0





0.5

0





1

0.0048





2

3
6

11
16

0.012

0.014

0.013

0.0076

0.0023



Dust concentration (ng/g)

0

84.2943

Weighted average based on data of lead-based paint
homes from HUD's American Healthy Homes Survey
(AHHS) 1 and II Lead Findings report (Exhibit
5-15)(USHUD, 2011, USHUD, 2021)

Dust intake (g/day)

0

0.019

Ozkaynak et al. 2022, Table 2 (Geometric mean)



0.083

0.021





0.25

0.023





0.5

0.026





1

0.023





2

3
6

11
16

0.014

0.015

0.013

0.0088

0.0035



Air concentration (ng/m3)

0

0.01

The EPA's 2017 Proposed Modeling Approaches for a
Health-Based Benchmark for Lead in Drinking Water
and Cavender, 2013

Air ventilation rate (m3/day)

0

2.9

AALM default values based on the 2011 Exposure
Factors Handbook, ICRP (1994), and the EPA
Technical Review Workgroup ventilation rates
(Stifelman 2007, Brochu et al. 2006, IOM 2005,
Layton 1993)



1

5.2





5

8.8





10

15.3





15

17.9





25

19.9



Males Food Lead intake
(Hg/day)

0

2.2

The EPA's (2024) Technical Support Document for the
All Ages Lead Model (AALM) version 3.0 -
Parameters, Equations, and Evaluations



1

3



Final LCRI Economic Analysis

5-33

October 2024


-------
Media parameter (units)

Age (years)
value starts

Values
Entered

Data Source



2

3.7





3

5.3





4

5.6





5

5.9





6

6.2





7

6.6





8

6.9





9

7.3





10

7.7





15

10.8





20

11.1





25

11





30

10.8





40

10.5





50

10.1





60

9.8





70

9.5





80

9.1





90

8.8



Females Food Lead Intake
(Hg/day)

0

2.1

The EPA's (2024) Technical Support Document for the
All Ages Lead Model (AALM) version 3.0 -
Parameters, Equations, and Evaluations



1

2.9





2

3.8





3

5.4





4

5.7





5

6





6

6.4





7

6.7





8

7





9

7.3





10

7.6





15

9.2





20

8.8





25

8.7





30

8.5





40

8.3





50

8





60

7.7





70

7.5



Final LCRI Economic Analysis



5-34

October 2024


-------
Media parameter (units)

Age (years)
value starts

Values
Entered

Data Source



80

7.2



90

7

Lead concentrations in soil and dust were consistent for all age groups and calculated as a weighted
average based on prevalence data of lead-based paint in homes by construction year in select housing
reported in U.S. Department of Housing and Urban Development's (HUD) American Healthy Homes
Survey (AHHS) I and II Lead Findings (Exhibit 5-1517). Construction year of homes is categorized into four
bins: pre-1940, 1940-1959, 1960-1977, and 1978-2017, which is consistent with OCSPP methodology for
reconsideration of the DLHS/DLCL. For each bin of housing construction year, the percentage of total
housing was multiplied by the geometric dust concentration. Soil lead concentration was calculated with
the geometric mean soil concentration using the same methodology. Soil and dust intake rates by age
group up to age 21 were estimated by Ozkaynak et al. (2022).

AALM model simulations for adults do not account for higher historical lead exposures and long-term
bone accumulation that may have occurred prior to the baseline or regulatory scenarios. Also, with
regard to adjusting baseline exposure over the 35-year period of analysis, in response to potential future
changes in the non-water exposure pathways, the EPA has found no way to extrapolate lead
concentrations over the period of analysis for the alternative media that would not introduce significant
uncertainty into the BLL estimation modeling likely outweigh any perceived improvement. National level
projections of decreases in lead dust and soil, and food concentrations over a decades long period are
not available and projection of current trends would introduce significant uncertainty. Using projected
values from NHANES to adjust estimated BLLs post modeling would also introduce significant
uncertainty, as the rate of decrease in BLLs observed in NHANES has slowed in recent years. Therefore,
the EPA held constant the non-water sources of lead over the 35-year analysis period. The EPA also
notes that because of the log-linear shape of the concentration response function for CVD premature
mortality, and given that the concentration response functions show no evidence of a threshold below
which effects cease, projected policy scenario reductions in water lead concentrations that occur at
lower baseline WLLs would result in larger changes in avoided CVD premature mortality, therefore the
inclusion of lower baseline values would result in increased benefit estimates associated with exposed
adult populations in the final LCRI.

5.4.2.2 Estimates of Pre- and Post-Rule Blood Lead Levels in Adults

Exhibit 5-22 displays BLL estimates for adults by each LSL/GRR service line, POU, pitcher filter or CCT
status combination summarized by age groups. Note that when "No LSL" is the beginning or post-rule
state, 0.83 ng/L is the assumed concentration across all levels of CCT status (i.e., none, partial,
representative). The extent to which changes in CCT status make meaningful differences in lead
concentrations for those without LSLs cannot be determined from the results presented in Exhibit 5-22.

Final LCRI Economic Analysis

5-35

October 2024


-------
Exhibit 5-22: Estimates of Blood Lead Levels in Adults Associated with Drinking Water Lead

Exposures from LSL/CCT or POU Combinations

Lead
Service
Line
Status

Corrosion
Control
Treatment
Status

Sex

Geometric Mean Blood Lead Level (|ig/dL) for Specified Age Group1
in Years from the AALM







8-15

16-19

20-29

30-39

40-49

50-59

60-69

70-79



None

Male

1.33

1.28

1.70

1.82

1.92

1.98

1.36

1.94

LSL



Female

1.25

1.44

1.99

2.14

2.27

2.35

1.56

2.31

Partial

None

Male

1.03

1.00

1.30

1.35

1.37

1.39

1.36

1.34

LSL/GRR



Female

0.97

1.10

1.47

1.53

1.56

1.59

1.56

1.53

No LSL

None

Male

0.80

0.77

0.98

0.97

0.94

0.92

0.88

0.85

Female

0.74

0.83

1.06

1.03

1.00

0.98

0.94

0.91

LSL

Partial

Male

1.08

1.04

1.36

1.42

1.45

1.47

1.45

1.42

Female

1.01

1.15

1.55

1.62

1.66

1.70

1.67

1.65

Partial

Partial

Male

0.92

0.89

1.14

1.16

1.16

1.15

1.12

1.10

LSL/GRR

Female

0.85

0.96

1.26

1.28

1.28

1.28

1.25

1.22

No LSL

Partial

Male

0.80

0.77

0.98

0.97

0.94

0.92

0.88

0.85

Female

0.74

0.83

1.06

1.03

1.00

0.98

0.94

0.91

LSL

Representati

Male

0.93

0.90

1.16

1.19

1.19

1.19

1.16

1.13

ve

Female

0.87

0.98

1.29

1.32

1.32

1.32

1.29

1.27

Partial

Representati

Male

0.85

0.82

1.05

1.05

1.03

1.02

0.99

0.96

LSL/GRR

ve

Female

0.79

0.89

1.15

1.14

1.12

1.11

1.07

1.04

No LSL

Representati

Male

0.80

0.77

0.98

0.97

0.94

0.92

0.88

0.85

ve

Female

0.74

0.83

1.06

1.03

1.00

0.98

0.94

0.91

POU or pitcher filter

Male

0.80

0.77

0.98

0.97

0.94

0.92

0.88

0.85

Female

0.74

0.83

1.06

1.03

1.00

0.98

0.94

0.91

^he estimated values reported in this exhibit represent the mean BLL for the ages specified in the range. The
AALM reports age-specific yearly BLLs for each single year age that are used in the SafeWater LCR benefits model.
Note: This exhibit displays BLL estimates for adults by each LSL, POU, pitcher filter or CCT combination summarized
by age groups. The EPA assumes that GRR service line water lead concentrations are the same as "Partial LSL"
values, therefore, BLL values for partial LSLs also represent GRR BLLs. Note that BLLs by each year (not age group
average) are used in the analysis.

The estimated BLLs in Exhibit 5-22 are average adult BLLs given the corresponding estimated lead tap
water concentrations resulting from LSL/GRR service line, CCT, POU, and pitcher filter status at steady-
state, holding other exposures constant. In the SafeWater LCR model, water systems are tracked as they
move from one LSL/GRR service line, CCT, pitcher filter or POU status to another as a result of rule
implementation. The numbers of males and females in each age group served by those water systems
are proportional to the age/sex makeup of the United States population as a whole. Age specific yearly
BLLs are used in the benefit valuation modeling. Expected changes on average for all adults 40-80 due to
changes in water concentrations due to the rule are displayed in Exhibit 5-23.

Final LCR! Economic Analysis

5-36

October 2024


-------
Exhibit 5-23 shows the estimated change in average lifetime BLLs for adults who experience a change in
water lead concentration as a result of pitcher filter use, LSL/GRR service line removal and/or
installation of CCT or POU, rather than the set of initial LSL/GRR service line and CCT status
combinations. Expected changes on average for all adults 40-80 due to changes in water concentrations
due to the rule are displayed in Exhibit 5-23.

Exhibit 5-23: Estimated Lifetime Average Blood Lead Level Decrease for Adults Experiencing
Alternate LSL/GRR, CCT, pitcher filter and POU Status Combinations

Pre-Rule Drinking Water

Post-Rule Drinking Water

Estimated Decrease in the
Means of Blood Lead

Lead
Cone.
(Hg/L)

LSL Status

CCT Status

Lead
Cone.
(Hg/L)

LSL Status

CCT Status

FEMALE:
Ages 40-80
(Hg/dL)

MALE:
Ages 40-80
(Hg/dL)

14.38

LSL

None

0.83

No LSL

None

1.36

1.05

14.38

LSL

None

4.27

LSL

Representative

1.01

0.78

14.38

LSL

None

0.83

No LSL

Representative

1.36

1.05

14.38

LSL

None

0.83

POU or pitcher filter

1.36

1.05

6.85

Partial/GRR

None

0.83

No LSL

None

0.6

0.47

6.85

Partial/GRR

None

2.14

Partial

Representative

0.47

0.37

6.85

Partial/GRR

None

0.83

No LSL

Representative

0.6

0.47

6.85

Partial/GRR

None

0.83

POU or pitcher filter

0.6

0.47

0.83

No LSL

None

0.83

No LSL

Representative

0

0

0.83

No LSL

None

0.83

POU or pitcher filter

0

0

7.93

LSL

Partial

0.83

No LSL

Partial

0.71

0.55

7.93

LSL

Partial

4.27

LSL

Representative

0.37

0.28

7.93

LSL

Partial

0.83

No LSL

Representative

0.71

0.55

7.93

LSL

Partial

0.83

POU or pitcher filter

0.71

0.55

3.84

Partial/GRR

Partial

0.83

No LSL

Partial

0.3

0.23

3.84

Partial/GRR

Partial

2.14

Partial

Representative

0.17

0.13

3.84

Partial/GRR

Partial

0.83

No LSL

Representative

0.3

0.23

3.84

Partial/GRR

Partial

0.83

POU or pitcher filter

0.3

0.23

0.83

No LSL

Partial

0.83

No LSL

Representative

0

0

0.83

No LSL

Partial

0.83

POU or pitcher filter

0

0

4.27

LSL

Representative

0.83

No LSL

Representative

0.35

0.27

4.27

LSL

Representative

0.83

POU or pitcher filter

0.35

0.27

2.14

Partial/GRR

Representative

0.83

No LSL

Representative

0.13

0.1

Final LCRI Economic Analysis

5-37

October 2024


-------
Pre-Rule Drinking Water

Post-Rule Drinking Water

Estimated Decrease in the
Means of Blood Lead

Lead
Cone.
(Hg/L)

LSL Status

CCT Status

Lead
Cone.
(Hg/L)

LSL Status

CCT Status

FEMALE:
Ages 40-80
(Hg/dL)

MALE:
Ages 40-80
(Hg/dL)

2.14

Partial/GRR

Representative

0.83

POU or pitcher filter

0.13

0.1

0.83

No LSL

Representative

0.83

POU or pitcher filter

0

0

Acronyms: LSL = lead service line; CCT = corrosion control treatment; POU = point-of-use; GRR = galvanized
requiring replacement

5.5 Concentration Response Functions and Valuations used in the Estimation of Benefits to
Children and Adults

The EPA undertook a rigorous process to identify concentration response functions to quantify benefits.
This included reviewing all available studies which could be used to develop quantitative relationships
between changes in lead exposure and/or changes in blood lead levels and changes in health endpoints.
The EPA evaluated the studies for quality and potential biases. The EPA then developed a separate
report for each health endpoint. In addition to the quality review findings, each report provides
quantitative estimates, based on the identified functions, of potential changes in the health endpoint
and was reviewed by EPA experts and/or externally peer reviewed. For the final LCRI the EPA has relied
on concentration response functions for four quantified health endpoints that have been extensively
reviewed by the agency and in the case of reductions in IQ losses, low birth weight and cardiovascular
disease premature mortality, externally peer reviewed. Also, the approach used for IQ has been used in
multiple prior rulemakings and undergone SAB review.

As with costs, the EPA estimated both high and low benefit scenarios for each health endpoint that is
quantified. For lower birth weight, only one concentration response function was determined to be of
high-quality, so this is used in both the high and low benefit scenario calculations. For IQ, ADHD, and
CVD premature mortality, two or more functions were available, and the EPA selected the functions that
gave the highest and lowest health benefit estimates across most blood lead levels.173 For information
on the uncertainties associated with the use of the selected concentration response functions see
Section 5.7. The monetized benefit estimates provided in this chapter use the 2 percent discount rate as
prescribed by the Office of Management and Budget's updated Circular A-4 (OMB Circular A-4, 2023).174

5.5.1 Concentration-Response Functions for Lead and IQ

Previously, to estimate benefits supporting the 2021 LCRR, the EPA used a function based on Crump et
al. (2013) in the main analysis and explored the choice of two additional IQfunctions in the sensitivity
analysis. Both functions in the sensitivity used the corrected Lanphear et al. (2005) function, as reported

173	As some of the functions are not linear, there are cases where these functions may not always give the highest
or the lowest benefits.

174	Because the EPA provided benefit estimates discounted at 3 and 7 percent for the proposed LCRI based on OMB
guidance which was in effect at the time of the proposed rule analysis (OMB Circular A-4, 2003), the agency has
also calculated the benefit impacts at both the 3 and 7 percent discount rates. See Appendix F for results.

Final LCRI Economic Analysis

5-38

October 2024


-------
in Kirrane and Patel (2014): one with a low-dose linearization and the other without a low-dose
linearization. To estimate avoided IQ loss in children for the final LCRI, the EPA selected two
concentration-response functions. The low scenario benefits estimate is based on the study by Crump et
al. (2013). The EPA chose the corrected Lanphear et al (2005, erratum 2019) function without low-dose
extrapolation for the calculation of the high scenario benefit estimate for avoided IQ loss under the final
LCRI. These studies were included in the EPA's SAB review of the 2021 LCRR (USEPA, 2020b).

This section provides an overview of these two key studies. Additional details of Crump et al. (2013) and
Lanphear et al. erratum (2019) can be found in Appendix J of the Final 2021 LCRR EA (USEPA, 2020a),
which provides more in-depth summaries of the key studies used in the concentration-response
functions for the benefits analysis, as well as the Kirrane and Patel (2014) correction to the Lanphear et
al. (2005) results, which was conducted prior to the publication of the Lanphear erratum.

Lanphear et al. erratum (2019) conducted a pooled analysis of seven international cohort studies that
investigated the relationship between BLLs and full-scale IQ (the composite of verbal and performance
IQ scores) in children 5-10 years old. The pooled study sample comprised 1,333 children, with a lifetime
average BLL of 12.4 ng/dL. All the children underwent IQ testing with the Wechsler Intelligence Scale for
Children. The mean IQ in the study sample was approximately 93 points. Associations between IQ and
four different measures of BLLs in children were examined: concurrent (measurement obtained closest
to the IQtest), maximum (peak value at any time before the IQtest), early (mean BLL from 6 to 24
months of age), and lifetime (mean BLL from 6 months of age to concurrent). For each of these
measures, Lanphear et al. erratum (2019) estimated the relationship between BLLs and IQ by
constructing an adjusted log-linear model.

Results of the Lanphear et al. erratum (2019) study showed that all blood lead measures were
significantly associated with IQ loss, and were highly correlated with one another. Based on the R2
values for each regression model (data not presented in the paper), Lanphear et al. erratum (2019)
determined that concurrent BLLs were the strongest predictors of IQ, followed by lifetime average BLLs.

Exhibit 5-24 shows the beta estimates for the log-linear associations between each of the blood lead
measures examined in Lanphear et al. erratum (2019). The estimated decreases in IQ associated with
increases in concurrent BLLs from 2.4 to 10 ng/dL, 10 to 20 ng/dL, and 20 to 30 ng/dL were 3.8, 1.8, and
1.1 points, respectively. Consistent with the log-linear model, IQ deficits were greater at lower levels of
lead exposures.

Changes in IQ associated with changes in BLLs for the high benefits scenario were estimated using
Equation 9 below. Average BLLs for children age 0-7 (lifetime exposure) from the SHEDS-Pb modeling
were used as inputs to the equation.

Final LCRI Economic Analysis

5-39

October 2024


-------
(Equation 9)

Where:

(B = Corrected lifetime beta estimate from Lanphear et al. (-3.25)

PbB1 = Pre-rule BLL
PbB2 = Post-rule BLL

In their 2013 paper, Crump et al. had two aims: 1) to perform a reanalysis of the methods in Lanphear et
al. (2005), and 2) to conduct an independent analysis of the data from Lanphear et al. (2005). In the
reanalysis, Crump et al. (2013) identified a few minor errors in the original Lanphear et al. (2005) paper.
The correction of these minor errors resulted in slight changes to the regression coefficients but did not
affect the main conclusions of the paper. These errors were confirmed by the EPA in a reanalysis by
Kirrane and Patel (2014), which also reaffirmed that the main conclusions of Lanphear et al. (2005)
remained unchanged, and Lanphear et al. erratum (2019) confirmed this in an Erratum of the original
study. Kirrane and Patel (2014) additionally found that the early childhood blood lead measure had the
highest R2 value, though all R2 values were similar.

In their independent analysis, Crump et al. (2013) made changes to the dataset used for final analysis
(e.g., in selecting IQ measurements and defining blood lead measurements). Additionally, the authors
opted to add 1 to the BLLs before log-transformation so that IQ loss was equal to 0 when BLL was 0, as
shown in Equation 10.

Where:

(B =	Lifetime beta estimate from Crump et al. (2013) independent analysis (-3.25)

PbB1	= Pre-rule BLL

PbB2	= Post-rule BLL

Changes in IQ associated with changes in BLLs for the low benefits scenario were estimated using
Equation 10 based on the Crump independent analysis. As with the high benefit scenario, average BLLs
for children ages 0-7 from the SHEDS-Pb model were used as inputs to the Equation 10.

For both equations, the SHEDS-Pb model estimated pre and post rule BLLs in children ages 0-7 are
described in Section 5.4.

(Equation 10)

Final LCRI Economic Analysis

5-40

October 2024


-------
Exhibit 5-24: Comparison of Adjusted Coefficients from Lanphear et al. Erratum (2019) with
Those Obtained in the Kirrane and Patel (2014), and the Reanalysis and Independent Analysis

of Lanphear et al. (2005) by Crump et al. (2013)

BLL
Variable

Kirrane and Patel
(2014)

Lanphear et al.
Erratum (2019)

Crump et al. (2013)
Reanalysis In(BLL)

Crump et al. (2013)

Independent
Analysis ln(BLL+ 1)

P

(95% CI)

R2

P

(95% CI)

R2a

P

(95% CI)

R2

P

(95% CI)

R2

Early

-2.21
(-3.38, -1.04)

0.643

-2.21
(-3.38, -1.04)

n/a

-2.21
(-3.38, -1.03)

0.643

-2.46
(-3.82,-1.10)

0.659

Peak

-2.86
(-4.10,-1.61)

0.640

-2.86
(-4.10,-1.61)

n/a

-2.86
(-4.10,-1.61)

0.640

-2.48
(-3.83,-1.14)

0.656

Lifetime

-3.14
(-4.39, -1.88)

0.641

-3.25
(-4.51,-1.99)

n/a

-3.19
(-4.45, -1.94)

0.641

-3.25
(-4.66, -1.83)

0.659

Concurrent

-2.65
(-3.69,-1.61)

0.641

-2.65
(-3.69,-1.61)

n/a

-2.65
(-3.69,-1.61)

0.641

-3.32
(-4.55, -2.08)

0.658

Sources: Crump et al. (2013, Table 2 and Table 5), Kirrane and Patel (2014, Table 1), Lanphear et al. erratum (2019, Table 4).
a R2 not reported in Lanphear et al. erratum (2019); however, the paper reported that the concurrent BLL was the largest R2.
Notes: This table displays regression coefficients and R2 values for the Lanphear et al. erratum (2019) analysis, the Crump et al.
(2013) and Kirrane and Patel (2014) reanalysis of Lanphear et al. (2005), and the Crump et al. (2013) independent analysis of
Lanphear et al. (2005). This table summarizes the relationship between BLL and IQ loss across various blood lead metrics.

As can be seen in Exhibit 5-24, the R2 values are all similar: the strength of the relationship between BLLs
and IQ loss appears to be similar regardless of the blood lead metric used. Because lifetime average BLLs
are more reflective of the long-term changes in lead exposure anticipated under the final LCRI, the EPA
chose to model the benefits under both the low and high benefit scenarios based on lifetime BLLs rather
than concurrent BLLs.

No threshold has been identified for the neurological effects of lead (Schwartz and Otto, 1991; Budtz-
J0rgensen et al., 2013; Crump et al., 2013; USEPA, 2024). Therefore, the EPA assumes that there is no
threshold for this endpoint and quantified avoided IQ loss associated with all BLLs (Schwartz and Otto,
1991; Budtz-J0rgensen et al., 2013; Crump et al., 2013; USEPA, 2024). Budtz-J0rgensen et al. (2013), as

well as the smaller cohort study of Min et al. (2009), used more recent BLLs than those used in the
Crump and Lanphear analyses, and confirmed the results in Crump et al. (2013) and Lanphear et al.
erratum (2019). Additionally, in Min et al. (2009), the steeper slopes at lower BLLs without log-
transformation show increased IQ deficits, which provides additional evidence that reducing lead levels
in the lower range of average BLLs has a significant impact on preventing IQ loss.

5.5.2 Valuation of Avoided IQ Loss

The economics literature provides a robust basis for estimating the relationship between IQ change and
lifetime earnings. Because the literature relies on large datasets that are representative of the US
population, it is appropriate to use the results to infer subpopulation-level impacts (though individual-
level impacts) from changes in environmental policy, even when average impacts are very small in
magnitude. The estimated effects of IQon lifetime earnings are not predicated on a particular type or

Final LCRI Economic Analysis

5-41

October 2024


-------
pathway of pollutant exposure. Rather, they are broadly applicable to evaluating any type of policy
intended to improve children's cognitive development (Lin et al. 2018).

The value of an IQ point used in the main analysis (both high and low scenarios) is derived from the
EPA's (2019a) reanalysis of Salkever (1995), which estimates that a one-point change in IQ results in a
mean 1.9 percent change in lifetime earnings for males and a mean 3.4 percent change in lifetime
earnings for females. Lifetime earnings are estimated using the average of 10 American Community
Survey (ACS) single-year samples (2008 to 2017) and projected cohort life tables from the Social Security
Administration. Projected increases in lifetime earnings are then adjusted for direct costs of additional
years of education and forgone earnings while in school. The USEPA (2019) reanalysis of Salkever (1995)
estimates a mean change of 0.08 years of schooling per change in IQ point resulting from a reduction in
lead exposure for males and a mean change of 0.09 years of schooling for females. This approach was
reviewed by the EPA's SAB (USEPA, 2020b).

To estimate the uncertainty underlying the model parameters of the Salkever (1995) reanalysis, USEPA
(2019a) used a bootstrap approach to estimate a distribution of model parameters over 10,000
replicates (using random sampling with replacement). For each replicate, the net monetized value of a
one-point change in IQ is subsequently estimated as the gross value of an IQ point, less the value of
additional education costs and lost earnings while in school.

Based on the mean value of the 10,000 sampling iterations, the USEPA (2019) estimated that the change
in one IQ point discounted to age 7 is $42,226, in 2022 dollars, using a 2 percent discount rate. Note that
the EPA's use of the term "2 percent discount rate" with regard to the calculation of the IQ point high
and low values (which represent the present value of the change in lifetime earnings) is shorthand for a
declining discount rate which begins with a 2 percent discount rate for the years 2024-2079, a 1.9
percent discount rate used for the years 2080-2096, and a 1.8 percent discount rate used in years 2095-
2102. This declining rate structure was implemented to comply with updates to OMB Circular A-4
guidance which indicates that a declining discount rate may be used to capture the uncertainty in the
appropriate discount rate over long time horizons like lifetime labor force participation.175176

The Salkever IQ value is presented in 2022 dollars to be consistent with the cost estimates. As described
in Section 5.6, benefits are further discounted back to year one of the analysis and annualized within the
SafeWater LCR model. A summary of the Salkever component values, by sex, can be found in Exhibit
5-25.

175	The revised Circular A-4 discusses discounting over long time horizons (OMB Circular A-4 2023). As noted by
OMB in the updated Circular A-4, "[t]here are various reasonable approaches to long-term discounting that
account for uncertainty and other relevant factors, and therefore lead to dynamic discount rates over time." When
the time horizon of an analysis is sufficiently long (i.e., 2080 or beyond), use of a declining discount rate may be
appropriate to capture uncertainty in the discount rate over long time horizons.

176	Note that the declining discount rate structure was not used in the proposed rule calculation of IQ point values
and the EPA has continued to use the constant discount rate IQ point values in the 3 and 7 percent benefit
calculations found in Appendix F.

Final LCR! Economic Analysis

5-42

October 2024


-------
Exhibit 5-25 Updated Estimates for Lifetime Earnings, Additional Education Costs, and Lost
Earnings from Additional Education (2022 USD), discounted at 2 percent to age 7

Estimate

Updated Salkever Estimates

Male

Female

Male and

Female

Combined

1. Lifetime Earnings

$2,174,849

$1,424,497

-

2. IQ Effect

1.87%

3.41%

-

3. IQ Effect*Lifetime Earnings

$40,700

$48,559

$44,551

4. Additional Education Costs

$1,702

$1,940

$1,819

5. Lost Earnings (from additional
education)

$594

$415

$506

6. Value of an IQ Point (3 - (4+5))

$38,404

$46,204

$42,226

Note: The EPA uses of the term "2 percent discount rate" with regard to the calculation of the IQ point high
and low estimates is shorthand for a declining discount rate which begins with a 2 percent discount rate for
the years 2024-2079, a 1.9 percent discount rate used for the years 2080-2096, and a 1.8 percent discount
rate used in years 2095-2102. This declining rate structure was implemented to comply with updates to OMB
Circular A-4 guidance.

See Appendix F for a Sensitivity Analysis with an alternative value for IQ benefits based on Lin et. al.
(2018). For additional discussion of the methods, also see Appendix J of the Final 2021 LCRR EA (USEPA,
2020a) and Appendix A of USEPA (2024c).

5.5.3 Concentration-Response Function for Lead and ADHD

This is the first regulation in which the EPA has estimated benefits of avoided cases of ADHD associated
with reductions in lead exposure; as discussed below the approach for quantifying such benefits will
continue to evolve as our understanding of the potential relationship improves. As described in
Appendix D the USEPA (2024b) ISA strengthened the conclusions of the 2013 ISA and concluded that
there was a causal relationship between lead exposure and inattention, impulsivity, and hyperactivity in
children based on recent studies of children with group mean BLLs <5 ng/dL. The 2024 ISA states that
"prospective studies of ADHD, including a study of clinical ADHD that controlled for parental education
and SES [Socioeconomic status], although not quality of parental caregiving reported positive
associations" (USEPA, 2024b. p. IS-30).The causes of ADHD are not fully understood, but research
suggests a number of potential causes, including genetics, exposure to environmental toxins, prenatal
cigarette smoking or alcohol intake, and brain changes (Tripp and Wickens, 2009; Pliszka et al., 2007).
The EPA's 2013 lead ISA statedthat in children, "attention was associated with biomarkers of Pb
exposure representing several different lifestages and time periods. Prospective studies did not examine
a detailed Pb biomarker history, and results do not identify an individual critical lifestage, time period, or

Final LCRI Economic Analysis

5-43

October 2024


-------
duration of Pb exposure associated with attention decrements in children. Associations in prospective
studies for attention decrements with tooth Pb level, early childhood average and lifetime average
blood Pb levels point to an effect of cumulative Pb exposure." The 2024 ISA addresses the uncertainties
presented in the 2013 ISA by stating that "The largest uncertainty addressed by the recent evidence
base is the previous lack of prospective studies examining ADHD (Appendix 3.5.2.4-3.5.2.5). The bulk of
the recent evidence comprises prospective studies that establish the temporality of the association
between Pb [lead] exposure and parent or teacher ratings of ADHD symptoms and clinical ADHD. Across
studies, associations were observed with tooth Pb concentrations, childhood BLLs (<6 ng/dL), and with
maternal or cord BLLs (2-5 ng/dL)." The available studies relating blood lead to ADHD use one-time
BLLs, while it is possible that cumulative exposure is also important. However, one-time and cumulative
measures of BLLs in children are often correlated. Therefore, the EPA has chosen diagnosed cases of
ADHD as an endpoint in this benefits analysis, because literature exists linking ADHD diagnosis to these
monetizable outcomes. The larger body of literature on attention, impulsivity, and hyperactivity
symptoms in children supports this association. The EPA chose a higher and lower concentration-
response function for the estimates of avoided cases to partially address the uncertainty in the most
appropriate function to use in estimating avoided cases due to the rule. Additional future research will
help to further understand the critical exposure window (thus exposure metric), the mode of action of
lead in the development of ADHD and/or related symptoms, and the interplay with genetic factors and
exposures to other substances.

The approach used to quantify ADHD here is based on review and analysis that Abt Associates (Abt
Associates, 2022a) conducted under contract to the EPA.

For the LCRI, the EPA estimates the benefits based on avoided cases of ADHD in children due to the rule.
The EPA chose a higher and lower concentration-response function for the estimates of avoided cases to
partially address the uncertainty in the most appropriate function to use in estimating avoided cases
due to the final rule.

This section provides a brief overview of two studies that inform the high and low benefit estimates for
ADHD. Froehlich et al. (2009) forms the basis of the high benefits estimates, and Ji et al. (2018) forms
the basis of the low benefits estimates. The selection of these studies is summarized in a report
prepared for the EPA (Abt Associates, 2022a) Additionally, see Section 5.7.5 for a discussion on the
strengthened evidence addressing the uncertainty in the relationship between Pb and ADHD presented
in the 2024 Pb ISA.

Froehlich et al. (2009) aimed to investigate the associations between ADHD and childhood lead
exposures, both independently and in combination with prenatal tobacco exposures. The authors
analyzed data from 2001-2004 NHANES on 2,588 children aged 8 to 15 years old with complete
information on ADHD diagnosis, lead and tobacco exposures, and additional covariates. Children with
high serum cotinine levels (>10 ng/mL), were excluded from the study to prevent confounding of the
effects of secondhand tobacco exposure. In the main analyses, ADHD diagnosis in NHANES was based on
completion of the Diagnostic Interview Schedule for Children (DISC) by caregivers. The DISC is a
structured interview that contains questions on ADHD symptoms, onset, pervasiveness, and severity in
the last 12 months and uses DSM-IV177 criteria to diagnose ADHD. As a secondary outcome, the

177 Diagnostic and Statistical Manual of Mental Disorders

Final LCRI Economic Analysis

5-44

October 2024


-------
definition of ADHD diagnosis was expanded to capture children with ADHD who did not meet full DSM-
IV criteria due to appropriate medication treatment. In these secondary analyses, children that had a
caregiver report both a history of ADHD diagnosis and ADHD medication use in the past year were
additionally included in the analyses. The authors investigated variables that had previously been shown
to be associated with ADHD as potential confounders. In the secondary analyses, health insurance status
was also examined as a covariate. Logistic regression analyses were used to examine associations
between lead exposures and ADHD, adjusted for confounders that were confirmed to be significantly
associated with ADHD (x2 test, p < 0.2). The final logistic regression model was adjusted for sex, age,
race/ethnicity, preschool attendance, birth weight178, income/poverty ratio, maternal age at child's
birth, and both current secondhand and prenatal tobacco exposures (operationalized by serum cotinine
levels and via maternal report, respectively). Additional analyses were performed restricting the sample
to children with blood lead < 5 ng/dL. Joint toxicant (i.e., both lead and tobacco exposure) effects were
assessed by examining ADHD incidence at varying levels of co-exposures.

Froehlich et al. (2009) found that 8.7% of children studied met DSM-IV criteria for ADHD diagnosis.
Children in the highest tertile of lead exposure (>1.3 ng/dL) were 2.3 times more likely to be diagnosed
with ADHD (95% CI, 1.5-3.8) than children in the lowest tertile (0.2 to 0.8 ng/dL). The same adjusted
odds ratio (OR) was observed when restricting the sample to children with blood lead < 5 ng/dL. When
blood lead was logarithmically transformed and analyzed as a continuous variable, the adjusted OR for
ADHD diagnosis was 1.8 (95% CI, 1.2-2.8) given a one-unit increase in natural log blood lead179. The
significant association between lead exposures and ADHD remained when the definition of ADHD
diagnosis was expanded in the secondary analyses: the adjusted OR was 2.0 (95% CI, 1.3-3.0). Childhood
lead and prenatal tobacco exposures combined had a multiplicative effect on the risk of ADHD. Froehlich
et al. (2009) estimated that 25% of ADHD cases among U.S. children with blood lead > 1.3 ng/dL are
attributable to lead exposures, corresponding to approximately 598,000 cases.

Results of Froehlich et al. (2009) were consistent with prior studies that found a relationship between
childhood lead exposures and DSM-IV ADHD diagnosis. The use of a national, population-based sample
of children with low blood lead makes results generalizable to the U.S. population of children. The
possibility of residual confounding from unmeasured genetic and environmental confounders (e.g.,
prenatal alcohol exposure) or parental characteristics remains. Because of small sample sizes for each
subtype, the authors could not investigate associations between blood lead and specific ADHD subtypes.

Ji et al. (2018) investigated the relationship between early childhood exposure to lead (blood leads were
measured prior to age 4) and the risk of being diagnosed with ADHD using a prospective cohort design,
including effect modification by sex, maternal high density lipoprotein (HDL) levels, and stress during
pregnancy. Data from the Boston Birth Cohort were utilized in this study. The Boston Birth Cohort
includes mother-infant pairs enrolled at birth from the Boston Medical Center. Enrollment is on a rolling
basis since 1998, and at the time of this study 3098 mother-infant pairs had enrolled in the post-natal
follow-up study. After excluding mother-infant pairs due to missing data, lead measurements taken after
an ADHD diagnosis, incorrect measurement dates, age over 4 years at measurement, and lead levels
higher than 10 ng/dL, the final analysis including 1479 pairs.

178	Birth weight could be one pathway through which Pb exposure affects ADHD.

179	Per Joseph Braun, personal communication to Meghan Lynch

Final LCRI Economic Analysis

5-45

October 2024


-------
Data were collected using a questionnaire, electronic medical records, and maternal blood samples
obtained 24 to 72 hours after delivery. A questionnaire was used to collect data from mothers on
demographic characteristics, stress during pregnancy, and smoking status. Birthweight, gestational age,
parity, intrauterine infections, complications, child lead levels, and ADHD diagnostic codes were
obtained from electronic medical records. If a child had repeated lead measures, the earliest
measurement taken was used for analysis. If a child's electronic medical record contained a diagnostic
code for ADHD, the child was enrolled in the ADHD group. Children in the neurotypical group were not
diagnosed with any of the ADHD codes, nor were they diagnosed with autism spectrum disorder,
conduct disorders, developmental delays or intellectual disabilities, failure to thrive or congenital
anomalies. HDL and lead levels were measured in maternal blood samples taken between 24 to 72
hours after delivery.

To examine the concentration-response relationship between lead and ADHD diagnosis, the authors
used categorical and continuous multiple logistic regression, and adjusted for maternal age at delivery,
mode of delivery, maternal race/ethnicity and education, smoking status during pregnancy, intrauterine
infection, parity, child's sex, preterm birth, and birthweight in all models (except sex when it was
included as joint or interaction term in the models). Additional analyses were conducted to investigate
the effects of sex on the lead-ADHD relationship.

Ji et al. (2018) found elevated lead levels at 5-10 ng/dL were associated with a 66% increase in risk of an
ADHD diagnosis, OR=1.66 (95% CI, 1.0-2.56), compared to children with lead levels less than 5 ng/dL.
The natural log-transformed linear lead levels were associated with an increased risk of ADHD diagnosis
(OR=1.25, 95% CI, 1.01-1.56). In joint association analyses, the effects of lead on the risk of ADHD
diagnosis were attenuated in both stratified and joint effects models for females. For males, risk of
ADHD diagnosis was 2.5 times higher when lead levels were 5-10 ng/dL compared to lead levels <5
Hg/dL (OR=2.49, 95% CI, 1.46-4.26). Findings were similar in Cox proportional hazards models.

This main health impact function is applied to both the Froehlich et al. (2009) and Ji et al. (2018)
studies180. Regression coefficients ((Bs) are summarized below the equation.

AADHD=

V o -

Po

^ 	 p ^ g—pi[ln(Blood Pbi)—In^Blood Pbf*)] _|_

P 0

x pop

(Equation 11)

Where:

Po = Baseline rate of ADHD in the population of interest

= Beta estimate from study: 0.223 using Ji et al. (2018) or 0.588 using Froelich et al. (2009)
Blood Pbi= Initial blood lead (ng/dL)

Blood Pbf = Final blood lead (ng/dL)
pop = Number of children in the population of interest

180 A derivation of this function can be found in Abt Associates (2023).

Final LCRI Economic Analysis

5-46

October 2024


-------
Ji et al. (2018) measured early childhood BLLs, therefore, in the SafeWater LCR model analyses (see
Section 5.6) the blood lead outputs from the SHEDS-Pb models were used, as these are more relevant to
younger children. Benefits based on Ji et al. (2018) are captured at age 7, assuming all children over the
analysis period are diagnosed with ADHD at age 7. This is the basis of the low benefits estimates for
ADHD.

Froelich et al. (2009) measured BLLs in children ages 8-15. Therefore, output from the AALM model was
used in the SafeWater LCR model analyses to estimate BLLs in that age group. Benefits using Froelich et
al. (2009) are captured at age 11, assuming all children over the analysis period are diagnosed with
ADHD at age 11.

For both the high and low benefit calculations, the baseline rate of ADHD is assumed to be 9.6 percent
based on Danielson et al. (2018).181

5.5.4 Valuation of Avoided ADHD

This analysis applies a valuation for ADHD cases based on a study by Doshi et al. (2012) following a
similar approach to that used in the EPA's (2023a) Economic Analysis of Updated Soil Lead Guidance for
Sites and Facilities Being Addressed Under CERCLA and RCRA Authorities.

To value each case of ADHD avoided, the USEPA (2023a) applied the following values obtained from
Doshi et al. (2012) for annual per-person incremental costs in 2023 dollars covering the following cost
categories:

•	Children/Adolescent costs

o	Health care (patient); ages 0-21: $2,348

o	Health care (family); ages 0-18: $1,930

o	Productivity losses (family); ages 0-18: $326

o	Education; ages 5-18: $4,680

o	Justice system; adolescents aged 13-17: $362

•	Adult costs

o Health care (patient); ages 18-64: $2,680
o Health care (family); ages 19-44: $1,330
o Justice system; ages 18-28: $2,405

As described in Section 5.5.3 two different concentration response functions are used for the high and
low scenarios. Ji et al. (2018) measured early childhood BLLs. Benefits based on Ji et al. (2018) are
captured at age 7, assuming all children over the analysis period are diagnosed with ADHD at age 7. This
is the basis of the low benefits estimates for ADHD. Froelich et al. (2009) measured BLLs in children ages
8-15. Benefits using Froelich et al. (2009) are captured at age 11, assuming all children over the analysis
period are diagnosed with ADHD at age 11. Therefore, for the valuation in the low scenario, costs for

181 Note the EPA updated the baseline rate of ADHD based on Danielson et al. (2018). In the EPA assessment for
the "Updated Residential Soil Lead Guidance for CERCLA Sites and RCRA Corrective Action Facilities" the agency
used a baseline rate for ADHD of 10.2 percent from Xu et al. (2018).

Final LCR! Economic Analysis

5-47

October 2024


-------
children 0-6 are not included in the estimate. For the high scenario, costs for children 0-10 are not
included in the estimate.

There is uncertainty about what percent of ADHD cases persist into adulthood. Therefore, for the final
LCRI rule analysis, the EPA uses a high and low estimate of the ADHD cost of illness, based on a high and
low estimate of ADHD persistence into adulthood.

The high analysis assumes that 90 percent of childhood cases of ADHD persist into adulthood, based on
Sibley et al. (2022) and as used in USEPA (2024c). This assumption is used to adjust the healthcare and
justice system benefits realized at ages 18 and older for an avoided case of ADHD diagnosed in
childhood. The assumption is derived from Sibley et al. (2022)'s finding that 9.1 percent of childhood
cases (mean age 8 years) recovered from ADHD at the study's final 16-year follow up (mean age 25
years, sample size 558). Recovery was defined as a full remission of ADHD sustained for at least two
consecutive study assessments (conducted approximately every two years). However, the authors find
that most cases have ADHD symptoms and impairments that fluctuate over time, and only a small
percentage are stable into adulthood, either as persistent case or full recovery status. For example, at
the final 16-year follow-up, 39.7 percent of participants were categorized as having persistent ADHD
(defined using DSM-5 symptom thresholds) and 45.7 percent were categorized with partial remission.
These participants were comprised of a mix of those with stable persistence (10.8%) or partial remission
over all study time periods (15.6%), and a majority with fluctuating occurrence of symptoms over time
(63.8%).

In sum, while this analysis assumes that 90 percent of childhood ADHD diagnoses persist into adulthood,
only a fraction of those cases are likely to meet the full DSM diagnostic criteria and/or present stable
symptoms in each year of adulthood. Thus, the high analysis may potentially overestimate ADHD
benefits resulting from the final rule to the extent that these variances are not captured in the cost-of-
illness estimates for the value of an avoided case of ADHD.

The low estimate is based on Barbaresi et al. (2013) which reports a 29.3 percent persistence rate,
where persistence is defined according to the number of ADHD symptoms in adulthood that exceed two
standard deviations of the mean number of symptoms in non-ADHD controls. Barbaresi et al. (2019) is
based on a small sample size (367) and the population is nearly all white, and focused on Rochester,
Minnesota, which the authors describe as geographically isolated in southeastern Minnesota. The study
categorizes itself as the only study to not look at cases referred to a specialty treatment program. It is
possible this is an underestimate of persistence given that it excludes some cases of partial ADHD
symptoms, which are likely to yield social costs. Given the range of persistence into adulthood, the EPA
chose 29% as the lower bound.

The high and low net present value estimates of all avoided ADHD costs incurred through age 64 are
presented in Exhibit 5-26 in 2022 dollars. The values have been discounted back to age 7 for use with Ji
et al. and back to age 11 for use with Froelich et al. using a 2 percent discount rate. Once captured,
SafeWater further discounts back to the first analysis year.182

182 Because the EPA provided benefit estimates discounted at 3 and 7 percent for the proposed LCRI based on OMB
guidance which was in effect at the time of the proposed rule analysis (OMB Circular A-4, 2003), the agency has
also calculated the ADHD benefit impacts at both the 3 and 7 percent discount rates. In the calculation of these

Final LCRI Economic Analysis

5-48

October 2024


-------
Exhibit 5-26: Present Value of Avoided ADHD Cases 2022 USD, Per Case

Assumed Persistence of
ADHD Into Adulthood

Age at ADHD Diagnosis

2% Discount Rate

90%

11 (High- Froelich)

$184,149

29.3%

7 (Low- Ji)

$128,559

Note: The EPA uses of the term "2 percent discount rate" with regard to the calculation of the ADHD
high and low estimates is shorthand for a declining discount rate which begins with a 2 percent
discount rate for the years 2024-2079, a 1.9 percent discount rate used for the years 2080-2085. This
declining rate structure was implemented to comply with updates to OMB Circular A-4 guidance.

5.5.5 Concentration-Response Function for Lead and Birth Weight of Infants Born to Women of
Child-Bearing Age

In this analysis, women of childbearing age are represented by the population of women between the
ages of 17-45 years old. The EPA utilized the AALM to generate estimates of blood lead in women of
childbearing age. Zhu et al. (2010) was used to develop a concentration-response function for the birth
weight of children born to these women for both the high and low benefit scenarios as this was the only
study of suitable quality for benefits analysis (see Abt Associates, 2022b).183

Zhu et al.'s study, Maternal Low-Level Lead Exposure and Fetal Growth (2010), examined the
association between low-level (<10 ng/dL) lead exposure and decreased fetal growth, specifically
measures of birth weight, pre-term birth, and small for gestational age. In their retrospective cohort
study, Zhu et al. matched the blood lead records from New York State's Heavy Metals Registry (HMR)184
to birth certificate data for singleton births in the state of New York for 43,288 mother-infant pairs from
upstate New York (New York State excluding New York City). The mothers were 15-49 years of age in
2003-2005.185 The study restricted the cohort to mothers with blood lead levels < 10 ng/dL. The mean
and median blood lead levels for the cohort were 2.1 ng/dL and 2 ng/dL, respectively. The mean birth
weight was 3,331 grams.

To assess the relationship between maternal blood lead and the continuous outcomes (e.g., birth weight
in grams), Zhu et al. (2010) used a multiple linear regression with fractional polynomials (Royston et al.

benefits the EPA has used ADHD case values that are derived by discounting at the constant 3 and 7 percent rates.
See Appendix F for ADHD case values and benefit results discounted at 3 and 7 percent.

183	An earlier version of this report describing the choice of Zhu et al. was peer reviewed in 2015 as part of the
External Peer Review of the EPA's Approach for Estimating Exposures and Incremental Health Effects from Lead
due to Renovation, Repair, and Painting Activities in Public and Commercial Buildings,

184	Starting in 1992, New York State began requiring that all lead test results be reported to the HMR. The authors
pulled data on potential confounding factors from the birth certificate files.

185	For any individuals who had more than one blood lead measurement, a single measurement was selected at
random. Additionally, for any mothers who had more than one child between 2003 and 2005, only one birth was
selected, also at random.

Final LCRI Economic Analysis

5-49

October 2024


-------
1999). They explored one or two terms of fractional polynomials in terms of xp where the power of p
was -2, -1, -0.5, 0.5, 1, 2, and 3, and also used a natural logarithmic transformation of lead.186

The authors state that the model that assumed a linear relationship between birth weight and the
square root of blood lead fit the data better than models with all other combinations of fractional
polynomials. The final model developed by Zhu et al. (2010) was adjusted for timing of the lead test,
gestational age, maternal age, race, Hispanic ethnicity, education, smoking, alcohol drinking, drug abuse,
in wedlock, participation in special financial assistance program, parity, and infant sex. The
concentration-response relationship from Zhu et al. is:

BW = b0 + (-^x PbB°-5)	(Equation 12)

Where:

BW = Birth weight in grams

b0 = Birth weight when blood lead is equal to 0 ng/dL187
PbB = Blood lead in ng/dL

The results from the study are presented in Exhibit 5-27, which shows that changes in birth weight
associated with a 1 ng/dL change in blood lead vary based on the starting blood lead concentration. For
example, the reduction in birth weight from a change in blood lead from 0 to 2 ng/dL is approximately
40 grams and from 8 to 10 ng/dL is approximately 10 grams. As Zhu points out, "the model predicts the
strongest estimated effects at the lowest levels of exposure, without a lower threshold of PbB [blood
lead] below which there would be no predicted effect on birth weight" (p. 1473).

186	While 0.5 is not listed in the methods of Zhu et al. (2010), this is stated to be the resulting best fit model;
therefore, it is included our list.

187	The birthweight when blood lead is equal to zero was not provided in the paper however from Figure 1 it
appears to be approximately 3,310 g.

Final LCRI Economic Analysis

5-50

October 2024


-------
Exhibit 5-27: Association between a Change in Blood Lead Concentration and Birth Weight,
Upstate New York, 2003-2005 from Zhu et al. (2010)

Change in Blood Pb
Concentration (|ig/dL)

Estimate (grams)

95% CI (grams)

0

Reference

-

1

-27.4

-17.1 to -37.8

2

-38.8

-24.1 to -53.4

3

-47.5

-29.6 to -65.4

4

-54.8

-34.2 to -75.5

5

-61.3

-38.2 to -84.4

6

-67.2

-41.8 to -92.5

7

-72.5

-45.2 to -99.9

8

-77.6

-48.3 to -106.8

9

-82.3

-51.2 to-113.3

10

-86.7

-54.0 to-119.4

Source: Table 3 from Zhu et al. (2010).

Notes: 1) The model was a linear regression with fractional polynomials after adjustment for timing of Pb test,
gestational age, maternal age, race, Hispanic ethnicity, education, smoking, alcohol and drinking, drug abuse, in
wedlock, participation in special financial assistance programs, parity, and infant sex. Blood Pb concentration was
transformed using a square root. The coefficient was -27.4 with a standard error (SE) of 5.3.

2) In the LCRI analysis, modeled blood lead levels do not exceed 2.35 ng/dL.

5.5.6 Valuation of Avoided Reductions in Birth Weight

The valuation of changes in birth weight is based on an approach further described in Abt Associates
(2022c) which was finalized after undergoing peer review coordinated by the EPA.188Their analysis of
U.S. Department of Health and Human Services, Medical Expenditure Panel Survey (MEPS) data found
that birth weight in the very low birth weight (VLBW)/ low birth weight (LBW) and normal ranges
influences medical expenditures. The report provides simulated cost changes based on inpatient
hospital stays. Since these models were non-linear, Abt Associates (2022c) conducted simulations to
understand the magnitude of the overall effect of birth weight on expenditures.

Using birth weight spline specifications, the authors found the simulated cost changes for increases in
birth weight are negative and significant in the VLBW, LBW, and normal birth weight ranges in models
that do not also control for a preterm birth indicator189 (see Exhibit 5-28). The effects are largest at
lower starting birth weights. For an increase of 0.22 lb, expenditures for inpatient hospital stays

188	Note this methodology was externally peer review, see MDB Inc. (2022).

189	Due to strong negative correlation between birth weight and preterm birth, there are fewer significant results
in the VLBW range when the preterm indicator is included.

Final LCRI Economic Analysis

5-51

October 2024


-------
decrease by $1,652190 at the VLBW threshold of 3.3 lbs, and less than $100 at the normal birth weight
threshold of 5.5 lbs.

Exhibit 5-28: Simulated Cost Changes (2010 USD) on Annual Medical Expenditures for
Inpatient Hospital Stays, using Birth Weight Spline Specifications (N with Positive

Expenditures = 450)

Birth Weight (lbs)

BW Splines (excluding Preterm)

+0.04 lb

+0.11 lb

+0.22 lb

2

-974.24
(573.13)*

-2,375.82
(1,395.14)*

-4,560.19
(2,669.45)*

2.5

-663.98
(376.82)*

-1,618.46
(915.69)*

-3,104.15
(1,747.07)*

3

-449.22
(240.68)*

-1,094.45
(583.64)*

-2,097.43
(1,109.73)*

3.3

-354.03
(180.93)*

-862.28
(438.13)*

-1,651.66
(831.06)**

4

-200.83
(87.76)**

-488.77
(211.65)**

-935.06
(398.76)**

4.5

-132.60
(49.29)***

-322.52
(118.43)***

-616.44
(221.78)***

5

-86.76
(26.01)***

-210.92
(62.23)***

-402.74
(115.69)***

5.5

-16.35
(6.85)**

-40.55
(16.91)**

-79.99 (33.09)**

6

-14.42
(5.61)**

-35.75
(13.83)**

-70.51(27.05)**

7

-11.18
(3.66)***

-27.71
(9.02)***

-54.65 (17.61)***

8

-8.64

(2 29)***

-21.41
(5.64)***

-42.21(10.99)***

9

10.63
(9.96)

26.93
(25.51)

55.03
(53.17)

10

15.47
(22.73)

39.14
(58.63)

79.86
(123.71)

Notes: 1) Results show mean and standard error of the difference between simulated cost for baseline birthweight
(left) and each birth weight increase. Significance estimates for the difference are indicated at the 1% (***), 5%
(**), and 10% (*) levels.

2) Results are based on the log-log model (probability) and a gamma distribution (expenditures), which appear to
fit the data best (see Appendix D). Estimates are averaged over all infants/toddlers (including those with and
without non-zero expenditures) up to age two years.

190 In 2010 United States Dollars.

Final LCRI Economic Analysis

5-52

October 2024


-------
Exhibit 5-29: Simulated Cost Changes (2010 USD) on Annual Medical Expenditures for
Inpatient Hospital Stays, for Birth Weight Indicator and a Pre-term Indicator Only Model (N

with Positive Expenditures = 450)

Change in
Indicator Value

Model with Indicators for LBW and
Preterm Birth

Model with LBW
Indicator (excluding
Preterm)

Model with Preterm
Indicator (Excluding
BW)

Simulated Change:
LBW

Simulated Change:
Preterm

Simulated Change:
LBW

Simulated Change:
Preterm

Oto 1

3,088.13
(1,154.62)***

949.29
(359.67)***

4,203.38
(1,278.27)***

2,316.15
(563.47)***

Notes: 1) Results show mean and standard error of the difference between simulated cost at each indicator
variable value (0 to 1 for either LBW or Preterm indicator variables). Significance is indicated at the 1% (***), 5%
(**), and 10% (*) levels.

2) Results are based on the log-log model (probability) and a gamma distribution (expenditures), which appear to
fit the data best (see Appendix D). Estimates are averaged over all infants/toddlers (including those with and
without non-zero expenditures) up to age two years.

In the SafeWater LCR model, costs are inflated to 2022 dollars in order to be consistent with the
timeframe chosen for the regulatory analysis (using a multiplier based on GDP191).

Applying the cost of illness (COI) value in the benefits calculation is done by following the steps:

Step 1. Calculate the change in birth weight due to the rule. Outputs from Zhu et al. (2010) for each
change in LSL/GRR service line, CCT, POU or pitcher filter use scenario provide this output.

Step 2. Calculate the valuation of the change in birth weight due to the rule based on the proportion
of infants born at each birth weight. Because Abt Associates (2022) estimated COI values for three
discrete changes in birth weight (0.04 lb, 0.11 lb, or 0.22 lb; or 20 grams, 50 grams, or 100 grams), this
results in the assumption that changes in birth weight below 0.04 lb have no value192, changes of 0.04 lb
to below 0.11 lb have a value equal to the COI presented for 0.04 lb changes, changes of 0.11 lb to
below 0.22 lb have a value equal to the COI presented for 0.11 lb changes, and changes of 0.22 lb and
above have a value equal to the COI presented for 0.22 lb changes. We assume that any change in birth
weight resulting from the rule impacts infants with baseline birth weights equal to the distribution of
birth weights in the United States (see Exhibit 5-30. Using this distribution, the EPA calculates the
valuation of the change in birth weight due to the rule using the following equation:

191	The EPA used the U.S. Bureau of Economic Analysis Table 1.1.9 Implicit Price Deflators for Gross Domestic
Product (the May 30, 2024 revision) to adjust dollar values to 2022. See:

https://apps.bea.gov/iTable/?reqid=19&step=3&isuri=l&1921=survev&1903=13#evJhcHBpZCI6MTkslnN0ZXBzlipb
MSwvLDMsM10slmRhdGEiOltblk5JUEFfVGFibGVfTGIzdClsliEzll0sWvJDYXRIZ29vaWVzliwiU3VvdmV5ll0sWvJGaXJzd
F9ZZWFvliwiMiAxNiJdLFsiTGFzdF9ZZWFvliwiMiAvMiJdLFsiU2NhbGUiLCIwll0sWvJTZXJpZXMiLCJBIIldfQ==

192	In reality, there is likely value below this level and therefore this analysis results in an underestimate of benefits.

Final LCR! Economic Analysis

5-53

October 2024


-------
Value of Change in Birth Weight = Y,bw2° (\V H bw,d * Pbw * P°P\ + |2 * VBbw (i * P[,w * pop|) (Equation 13)
where:

Sbw2° = Sum °f "value" equation above for each birth weight listed in Exhibit 5-30 below;

VHbw d = Savings in initial birth-related hospital stay expenditures for the applicable 0.04 lb, 0.11 lb, or

0.22 lb birth weight change (d) for the applicable baseline birth weight (bw);

VBbw,d = Savings in annual hospital stay expenditures in the first two years of life for the applicable 0.04

lb, 0.11 lb, or 0.22 lb birth weight change (d) for the applicable baseline birth weight (bw);

Pbw = Proportion of total births that belong to a particular baseline birth weight (bw); and

pop = Number of children born to number of women of childbearing age in each option scenario (the

annual fertility rate is 62.5 births per 1,000 women aged 15-44 in 2015).

Exhibit 5-30: Distribution of Birth Weights in the United States

Birth Weight (lbs)

Proportion of Total
Births

2

0.7%

2.5

0.3%

3

0.3%

3.3

0.5%

4

0.9%

4.5

1.3%

5

2.4%

5.5

4.1%

6

13.5%

7

33.2%

8

29.4%

9

11.1%

10

2.4%

Source: Distribution based on CDC WONDER data for 2014 (CDC. 2015).

5.5.7 Concentration-Response Function for Lead and Cardiovascular Disease Premature Mortality

In their review of the proposed LCRR, the EPA's SAB stated, "benefits associated with reduced lead
exposure and associated reduction in hypertension/cardiovascular effects have been well documented

Final LCRI Economic Analysis

5-54

October 2024


-------
(Chowdhury et al. 2018) and should be monetized and included in the EA" (USEPA, 2020b, p.15). For the
LCRI, the EPA uses a methodology to estimate avoided cases of CVD premature mortality193 due to
reductions in lead exposures developed in Brown et al. (2020) and Abt Associates (2023).194 In order to
quantify the benefits of avoided cases of CVD premature mortality, Brown et al. (2020) and Abt
Associates (2023) identified four studies providing a total of five concentration-response functions
relating adult BLLs to CVD premature mortality. Because, uncertainty exists regarding the lead exposure
level, timing, frequency, and duration contributing to the associations observed between a single adult
blood lead measurement and CVD premature mortality (see Section 5.7.7), the EPA selected the two
concentration-response functions that produced the highest and lowest estimated reduction in
mortality, or benefits, from the identified functions. Aoki et al. (2016) was used for the low benefits
estimates, and Lanphear et al. (2018) was used in the high benefits estimates. The EPA will evaluate new
and novel data as they become available, and will consider updating the methodology for estimating
cardiovascular premature mortality effects of changes in adult lead exposure as appropriate.

The four evaluated studies - Menke et al. (2006), Aoki et al. (2016), Lanphear et al. (2018), and Ruiz-
Hernandez et al. (2017) - all use regression models to relate log-transformed blood lead levels to CVD
premature mortality. The concentration-response function associated with the relationship between
blood lead and CVD premature mortality modeled in each study is:

Thus, the function necessary to estimate the number of cases associated with a change in blood lead
levels is:

yi = Baseline hazard rate of CVD premature mortality in baseline scenario (i.e., without the rule)

/? = Beta coefficient, which represents the change in CVD premature mortality per unit change in
blood lead

logz = Log transformation to the base z (e.g., logio)
x2 = Blood lead level associated with the rule
x1= Blood lead level without the rule

193ln 2020, the EPA's Science Advisory Board, in its review of the scientific and technical basis of the Lead and
Copper Rule Revisions, recommended that the EPA quantify and monetize CVD premature mortality impacts in
adults from reductions in lead in drinking water, citing "well documented" evidence of an association with
cardiovascular impacts (EPA SAB, 2020).

194 Note the Abt Associates (2023) methodology was externally peer reviewed. See the MDB, Inc. (2019) "Selection
of Concentration-Response Functions between Lead Exposure and Adverse Health Outcomes for Use in Benefits
Analysis: Cardiovascular-Disease Related Mortality" peer review documentation at
https://cfpub.epa.gov/si/si_public_record_report.cfm?Lab=NCEE&dirEntrylD=342855

Final LCRI Economic Analysis	5-55	October 2024

(Equation 14)

(Equation 15)

Where:


-------
pop =

Population for whom the change in blood lead occurs

Equation 16 can be used to estimate the avoided CVD premature mortality from reductions in blood
lead.

The beta coefficient, (B, varies based on the study in question and is calculated by:

„	In (Hazard ratio)

B = 		(Equation 16)

logz{Fold increase in blood lead for hazard ratio)

For example, the beta from Aoki et al. (2016) is based on a hazard ratio of 1.44, which was derived from
a 10-fold increase in blood lead levels. Thus, the beta coefficient is equal to ln(1.44)/logi0(10), which is
0.36. Exhibit 5-31 displays the study-specific inputs for Equation 16 associated with all five
concentration-response functions presented in Brown et al. (2020) and Abt Associates (2023).

Exhibit 5-31: Inputs to the Health Impact Function Based on Selected Studies

Variable

Aoki et al. (2016)

Lanphear et al. (2018)

Blood Pb <5 Hg/dL

Log transformation (logz)

Logio

Logio

Central beta (P) estimate

0.36

0.96

Lower beta (P) estimate (based on
lower bound of 95% CI for HR)

0.05

0.54

Upper beta (P) estimate (based on
upper bound of 95% CI for HR)

0.68

1.37

Sources: Aoki et al. (2016) and Lanphear et al. (2018).

Note: Bolding identifies the parameters used in the LCRI analysis. For full descriptions of these and the functions
not used to quantify CVD premature mortality, see Brown et al. (2020)

5.5.8 Valuation of Avoided Cardiovascular Disease Premature Mortality

In the scientific literature, estimates of willingness to pay for small reductions in mortality risks are often
referred to as the "value of a statistical life." This is because these values are typically reported in units
that match the aggregate dollar amount that a large group of people would be willing to pay for a
reduction in their individual risks of dying in a year, such that the EPA would expect one fewer death
among the group during that year on average. This is best explained by way of an example. Suppose
each person in a sample of 100,000 people were asked how much they would be willing to pay for a
reduction in their individual risk of dying of 1 in 100,000, or 0.001 percent, over the next year. Since this
reduction in risk would mean that the EPA would expect one fewer death among the sample of 100,000
people over the next year on average, this is sometimes described as "one statistical life saved." Now
suppose that the average response to this hypothetical question was $100. Then the total dollar amount
that the group would be willing to pay to save one statistical life in a year would be $100 per person x
100,000 people, or $10 million. This is what is meant by the "value of a statistical life." Importantly, this
is not an estimate of how much money any single individual or group would be willing to pay to prevent
the certain death of any particular person.

Final LCRI Economic Analysis

5-56

October 2024


-------
The EPA uses a value of a statistical life (VSL) of $12.98 million in 2022 dollars, which is estimated using
the EPA's (2014) recommended VSL of $4.8 million in 1990 dollars and the EPA's (2014) recommended
method for adjusting the VSL for income growth and inflation. The $4.8 million value in 1990 dollars is
updated to the $12.98 million in 2022 dollars by adjusting for inflation using the U.S. Bureau of Labor
Statistics' (2019) Consumer Price Index and adjusting for income growth using real GDP per capita and
an income elasticity of 0.4.

5.6 National Level Benefits Estimates

5.6.1 Implementation of Benefit Calculations in the SafeWater LCR model

Benefits are estimated based on LSL/GRR service line replacements, installation of POU devices,
distribution of pitcher filters and installation and re-optimization of CCT that occur over the 35-year
analysis period.

Benefits are captured in the analysis for each endpoint at a specific age, therefore it is necessary to
estimate the number of people of each age who are served by each PWS receiving a benefit from a
change in the lead concentration of their drinking water. This is handled by multiplying the number of
people experience a drinking water change by the proportion of people that age in the U.S. population.
For example, in order to estimate the number of 7-year-olds receiving a benefit in a given year, the
SafeWater LCR model takes the total population experiencing each water lead change and multiplies
that figure by the proportion of the United States population that is 7 years of age. A similar calculation
is done for the applicable ages for the additional endpoints.

Because the SafeWater LCR model follows the population for a period of 35 years, all children who lived
in areas experiencing the water lead concentration change who are younger than 7 years of age would
also accrue benefits in future years of the 35-year period, as well as children born after the change in
lead concentration as long as they reach the age of 7 during the course of the 35-year period. However,
the proportion of the total PWS population experiencing a change in lead concentration that receives an
IQ benefit in a given year remains the same: approximately 1.34 percent (the percentage of 7-year-olds
in the total United States population according to the 2020 United States Census). This is because both
the age distribution and the population served by each PWS are assumed to remain constant over the
analysis period. Children who turn 7 a year after an LSLR will receive a comparatively smaller benefit
than children who are born after the LSLR, due to living a larger proportion of their life without the
higher contribution of lead in their drinking water, and the resulting difference in BLLs between the
with- and without-rule scenarios (without considering discounting). The EPA refers to these
comparatively smaller benefits as "partial benefits." This same procedure is used for cases of ADHD
avoided, and for prevention of lower birthweight. ADHD benefits are captured at age 7 for the low
benefits estimate and age 11 for the high benefits estimate. For birth weight, benefits are captured once
yearly based on the birth rate in women ages 17-45. For CVD premature mortality, benefits are captured
yearly from ages 40-79.

The EPA does not assume that all homes with replaced LSLs have members living in the home eligible to
experience all four health endpoints. Rather, the EPA assumes that the proportion of each age and sex
(for adults) living in homes that are undergoing an LSLR is equal to the proportion of the United States
population that is that age and sex. This assumption takes care of the need to model the movement of

Final LCR! Economic Analysis

5-57

October 2024


-------
children and adults in and out of homes in the community, as the proportion of the population in these
age groups is assumed to remain constant. For example, for IQ, if there are 1,000 households being
served by a PWS that underwent a change in lead concentration, approximately 1.34 percent of the
population (the percent of the U.S. population 7 years of age) in those households would accrue benefit
annually, regardless of which specific home being served by the PWS they lived in. The accrued benefit
for those children who are served by a PWS that has undergone a change is then a function of changes
in the average lifetime BLL of the children due to the change in lead concentration, and the subsequent
avoided IQ loss.

The modeling assumption that the percentage of children and adults are evenly distributed across LSL
and non-LSL households is necessary to estimate the national level impacts of the final LCRI
requirements. At the national level, total benefits calculated using these assumptions can be accurate,
however, please note that the potential geographic variability in the impacted population of children or
adults will not be represented in this national scale model. For example, some geographic areas of the
country may have higher or lower percentages of young children, receiving greater or fewer benefits
from implementing lead concentration reducing actions like CCT and LSL/GRR service line replacement.
This national scale model does not capture the potential local variation in the estimated unit benefits for
a given unit of cost at the local level.

5.6.2 Monetized National Annual Benefits

As described in Section 5.3, the EPA estimated benefits corresponding to the low and high
scenarios used to characterize uncertainty in the estimates Benefits are discounted back to year one of
the analysis and annualized within the SafeWater LCR model. The EPA summed benefits for all years and
all PWSs, and then annualized benefits for both the baseline 2021 LCRR and LCRI using a 2 percent
discount rate.

•	As described in Section 5.5.1 and Section 5.5.2, the EPA applied both a high and a low
concentration-response function in order to estimate the reductions in IQ loss expected under
the rule, and a value of an IQ. Avoided IQ loss was captured at age 7, using a 2 percent discount
rate, benefits are further discounted, at 2 percent, back to year one of the analysis and
annualized within the SafeWater LCR model.

•	As described in Section 5.5.3, the EPA estimated avoided cases of ADHD with high and low
assumptions for the concentration response function. These avoided cases of ADHD were
captured at age 11 for the high function, and at age 7 for the low function, the difference is due
to the timing and methods in the source studies. The dose-response functions measure the
change in probability that an individual develops ADHD in their lifetime. This is a lifetime change
in risk rather than an annual change. In the case of Froehlich et al. (2009), this is because the
study measured prevalence rather than incidence. In this analysis, the EPA uses prevalence as
the baseline rate of ADHD in both concentration-response functions. As described in Section
5.5.4, high and low values, estimated using a 2 percent discount rate but assuming different
rates of ADHD persistence into adulthood, were applied to each avoided case of ADHD for the
high and low scenario respectively. Benefits are further discounted back to year one of the
analysis and annualized within the SafeWater LCR model using the 2 percent rate.

Final LCRI Economic Analysis

5-58

October 2024


-------
•	As described in Section 5.5.5, the EPA used the same concentration-response function for low
birth weight in both the high and low scenarios, as it was determined that only one study met
the criteria for inclusion. In this case, the only differences between the high and low scenarios
for calculating benefits are the estimated number of systems exceeding the action level, and
therefore the number of people who experience benefits, are due to the cost assumptions (see
Chapter 4, Section 4.2). Following the COI approach in Section 5.5.6 and further described in Abt
Associates (2022c), the EPA valued the avoided reductions in birth weight due to exposures to
women of childbearing age.

•	As described in Section 5.5.7, the EPA also estimated a high and low benefit for avoided CVD
premature mortality in adults ages 40-80. For each adult aged 40-80 during the analysis, annual
avoided CVD premature mortality is calculated with Equation 12, using the yearly blood lead
estimates produced by AALM for each age and sex, and the beta estimate from Aoki et al. (2016)
(low estimate) or Lanphear et al (2018) (high estimate). Age- and sex-specific background rates
of CVD premature mortality are used for the baseline rate obtained from CDC's Wonder (CDC.
2015) database. The available studies that link lead exposure to CVD premature mortality risk do
not provide information about latency or cessation lag between exposure and mortality
incidence. In Safewater LCR, the EPA made the assumption that the timing for the age of the
individuals experiencing CVD premature mortality that is caused by lead is the same as the
distribution of CVD premature mortality by age and sex for CVD premature mortality
irrespective of the cause (the cases due to lead follow the same rate by age as all other CVD
premature mortality). The EPA valued each avoided case of CVD premature mortality at $12.98
million (the EPA's value of a statistical life). The EPA then summed benefits for all years and all
PWSs producing a net present value. Benefits were then annualized for baseline 2021 LCRR and
the final LCRI. Benefits are presented at the 2 percent discount rate.

The national annual children's benefits for a 2 percent discount rate over the 35-year period of analysis
are presented in Exhibit 5-32 for IQ, and Exhibit 5-33 for ADHD. The results for prevented reductions in
birth weight in infants due to reduced exposures in women of childbearing age are presented in Exhibit
5-34 at a 2 percent discount rate. Benefits of avoided CVD premature mortality in adults ages 40-80 are
presented in Exhibit 5-35 at a 2 percent discount rate.195

Exhibit 5-36 summarizes the quantified national benefits for all endpoints. Exhibit F-3 and Exhibit F-4 in
Appendix F present the benefits of the final LCRI at the 3 and 7 percent discount rates under both the
high and low scenarios.

195 Because the EPA provided benefit estimates discounted at 3 and 7 percent for the proposed LCRI based on OMB
guidance which was in effect at the time of the proposed rule analysis (OMB Circular A-4, 2003), the agency has
also calculated the benefit impacts at both the 3 and 7 percent discount rates. See Appendix F for results.

Final LCRI Economic Analysis

5-59

October 2024


-------
Exhibit 5-32: Estimated National Annual Children's IQ Benefits, All PWSs, 2 Percent Discount Rate (millions of 2022 USD)

Low Estimate

High Estimate



Baseline

LCRI Incremental

Baseline

LCRI

Incremental

Annual IQ Point Decrement Avoided due to CCT

21,310

16,696

-4,614

59,586

45,371

-14,215

Annual Value of IQ Impacts Avoided due to CCT (millions of 2022 USD)

$824.8

$628.0

-$196.8

$2,306.1

$1,707.5

-$598.6

Annual IQ Point Decrement Avoided due to SLR

9,771

158,602

148,832

24,476

233,404

208,929

Annual Value of IQ Impacts Avoided due to SLR (millions of 2022 USD)

$381.2

$6,108.2

$5,727.0

$963.6

$8,988.7

$8,025.1

Annual IQ Point Decrement Avoided due to POU

61

8

-53

226

52

-173

Annual Value of IQ Impacts Avoided due to POU (millions of 2022 USD)

$2.5

$0.3

-$2.2

$9.3

$2.0

-$7.3

Annual IQ Point Decrement Avoided due to Filters

0

1,870

1,870

0

5,234

5,234

Annual Value of IQ Impacts Avoided due to Filters (millions of 2022 USD)

$0.0

$94.8

$94.8

$0.0

$264.8

$264.8

Total Annual Child Cognitive Development Benefits (millions of 2022
USD)

$1,208.5

$6,831.3

$5,622.8

$3,279.0

$10,963.0

$7,684.0

Acronyms: CCT = corrosion control treatment; IQ= Intelligence quotient; LCRI = Lead and Copper Rule Improvements; SLR = lead service line replacement; POU
= point-of-use; PWSs = public water systems; USD = United States dollar.

Final LCRI Economic Analysis

5-60

October 2024


-------
Exhibit 5-33: Estimated National Annual Benefits of Avoided ADHD Cases, All PWSs, 2 Percent Discount Rate (millions of 2022

USD)

Low Estimate

High
Estimate



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual Number of ADHD Cases Avoided due to CCT

192

151

-41

767

575

-192

Annual Value of ADHD Cases Avoided due to CCT (millions of 2022 USD)

$22.5

$17.1

-$5.4

$126.1

$91.9

-$34.2

Annual Number of ADHD Cases Avoided due to SLR

93

1,516

1,423

318

3,013

2,695

Annual Value of ADHD Cases Avoided due to SLR (millions of 2022 USD)

$11.0

$176.4

$165.4

$53.2

$491.7

$438.5

Annual Number of ADHD Cases Avoided due to POU

1

0

-1

3

1

-2

Annual Value of ADHD Cases Avoided due to POU (millions of 2022 USD)

$0.1

$0.0

-$0.1

$0.6

$0.1

-$0.5

Annual Number of ADHD Cases Avoided due to Filters

0

19

19

0

76

76

Annual Value of ADHD Cases Avoided due to Filters (millions of 2022 USD)

$0.0

$2.8

$2.8

$0.0

$15.8

$15.8

Total Annual Benefit of ADHD Cases Avoided (millions of 2022 USD)

$33.6

$196.3

$162.7

$179.9

$599.5

$419.6

Acronyms: ADHD = Attention-Deficit/Hyperactivity Disorder; CCT = corrosion control treatment; LCRI = Lead and Copper Rule Improvements; SLR= lead service
line replacement; POU = point-of-use; PWSs = public water systems; USD = United States dollar.

Final LCRI Economic Analysis

5-61

October 2024


-------
Exhibit 5-34: Estimated National Annual Benefits of Low-Weight Births, All PWSs, 2 Percent Discount Rate (millions of 2022 USD)

Low Estimate

High Estimate



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual Number of Low-Weight Birth Cases
Avoided due to CCT

146,324

113,761

-32,563

275,867

207,833

-68,034

Annual Value of Low-Weight Birth Cases Avoided

$0.7

$0.5

-$0.2

$1.3

$1.0

-$0.3

due to CCT (millions of 2022 USD)

Annual Number of Low-Weight Birth Cases
Avoided due to SLR

62,321

988,177

925,856

100,988

938,470

837,482

Annual Value of Low-Weight Birth Cases Avoided

$0.3

$4.8

$4.5

$0.5

$4.6

$4.1

due to SLR (millions of 2022 USD)

Annual Number of Low-Weight Birth Cases

397

50

-347

987

235

-752

Avoided due to POU

Annual Value of Low-Weight Birth Cases Avoided

$0.0

$0.0

$0.0

$0.0

$0.0

$0.0

due to POU (millions of 2022 USD)

Annual Number of Low-Weight Birth Cases

0

10,395

10,395

0

18,650

18,650

Avoided due to Filters

Annual Value of Low-Weight Birth Cases Avoided
due to Filters (millions of 2022 USD)

$0.0

1

o

-oo-

1

o

-oo-

$0.0

1

o

-oo-

1

o

-oo-

Total Annual Benefit of Avoided Low Weight
Births (millions of 2022 USD)

$1.0

$5.4

$4.4

$1.8

$5.7

$3.9

Acronyms: CCT = corrosion control treatment; LCRI = Lead and Copper Rule Improvements; SLR = lead service line replacement; POU = point-of-use; PWSs =
public water systems; USD = United States dollar.

Final LCRI Economic Analysis

5-62

October 2024


-------
Exhibit 5-35: Estimated National Annual Benefits of Avoided from Cardiovascular Disease Premature Mortalities, All PWSs,

2 Percent Discount Rate (millions of 2022 USD)

Low Estimate

High Estimate



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual Number of CVD Premature Mortality Cases Avoided due to CCT

106

82

-24

518

388

-130

Annual Value of CVD Premature Mortality Cases Avoided due to CCT
(millions of 2022 USD)

$1,228.4

$920.4

-$308.0

$5,987.9

$4,359.4

-$1,628.5

Annual Number of CVD Premature Mortality Cases Avoided due to SLR

44

731

687

184

1,756

1,572

Annual Value of CVD Premature Mortality Cases Avoided due to SLR
(millions of 2022 USD)

$518.9

$8,393.9

$7,875.0

$2,166.0

$20,214.3

$18,048.3

Annual Number of CVD Premature Mortality Cases Avoided due to
POU

0

0

0

2

0

-2

Annual Value of CVD Premature Mortality Cases Avoided due to POU
(millions of 2022 USD)

$3.4

$0.4

-$3.0

$21.0

$4.4

-$16.6

Annual Number of CVD Premature Mortality Cases Avoided due to
Filters

0

9

9

0

43

43

Annual Value of CVD Premature Mortality Cases Avoided due to Filters
(millions of 2022 USD)

$0.0

$139.6

$139.6

$0.0

$631.9

$631.9

Total Annual Benefit of CVD Premature Mortality Cases Avoided
(millions of 2022 USD)

$1,750.7

$9,454.3

$7,703.6

$8,174.9

$25,210.0

$17,035.1

Acronyms: CCT = corrosion control treatment; CVD = cardiovascular disease; LCRI = Lead and Copper Rule Improvements; SLR = lead service line replacement;
POU = point-of-use; PWSs = public water systems; USD = United States dollar.

Final LCRI Economic Analysis

5-63

October 2024


-------
Exhibit 5-36: Estimated National Annual Benefits - 2 Percent Discount Rate (millions of 2022 USD)

Low Estimate

High Estimate



Baseline

LCRI Incremental

Baseline

LCRI

Incremental

Annual Child Cognitive Development Benefits

$1,208.5

$6,831.3

$5,622.8

$3,279.0

$10,963.0

$7,684.0

Annual Low-Birth Weight Benefits

$1.0

$5.4

$4.4

00

T—1

-oo-

$5.7

$3.9

Annual ADHD Benefits

$33.6

$196.3

$162.7

$179.9

$599.5

$419.6

Annual Adult CVD Premature Mortality Benefits

$1,750.7

$9,454.3

$7,703.6

$8,174.9

$25,210.0

$17,035.1

Total Annual Benefits

$2,993.8

$16,487.3

$13,493.5

$11,635.6

$36,778.2

$25,142.6

Acronyms: ADHD = Attention-Deficit/Hyperactivity Disorder; CVD = cardiovascular disease; LCRI = Lead and Copper Rule Improvements; USD = United States
dollar.

Final LCRI Economic Analysis

5-64

October 2024


-------
5.7 Uncertainty in the Quantified Benefits

The quantified benefits are based on four endpoints. There is uncertainty in the true magnitude of the
benefits of lead reductions, as there are several health risks that are anticipated to be reduced by the
rule, but were not quantified in this analysis, see Section 5.8, Appendix D, and the EPA's Lead ISA
(2024a). This is the large uncertainty in the analysis and will result in an overall underestimation of
benefits even given the other uncertainties discussed in this section.

It should also be noted that all the results displayed in Exhibit 5-32 through Exhibit 5-36 are national
averages. The EPA expects that there will be individuals that are exposed to higher (or lower) water
concentrations than represented by the mean estimates in the exhibits. These individuals will have
greater (or lower) reductions in risk of adverse health effects, and thus higher (or lower) benefits due to
the final rule for those endpoints quantified here and presented as population averages. Additional
uncertainties as they relate to specific components of the benefits analysis are discussed below. General
uncertainties are discussed in Sections 5.7.1-5.7.3 and endpoint specific uncertainties are discussed in
more detail in Sections 5.7.4-5.7.7. Uncertainty in the underlying assumptions in SafeWater LCR and the
estimated costs can be found in Chapter 4, Section 4.2.2. Limitations of the water concentration
modeling were discussed in Section 5.2.5 and limitations in the assignment of the drinking water
concentrations to the modeled population are discussed in Section 5.3. Additionally, these are briefly
summarized in Exhibit 5-37.

Exhibit 5-37: Uncertainties in the Benefits Analysis

Issue

Addressed with
High/Low Scenario?

Direction of Bias

Changing population demographics including fertility and
immigration rates

No

Unclear

Uncertainty related to the estimation of baseline and policy scenario
drinking water lead concentrations

No

Unclear

The presence and degree to which potential lead exposed individuals
engage in averting behavior under the baseline and regulatory
options.

No

Unclear

Effects of changes in CCT in the absence of LSL status changes

No

Underestimate

Effects of CCT in respect to water chemistry and corrosion control
practices, lead sources other than service lines.

No

Unclear

Relative contribution of particulate lead

No

Underestimate

Number, type, and age of residences with LSLs

No

Unclear

Typical water concentrations and exposure patterns in multi-family
residences, workplaces, schools, and other public places.

No

Unclear

Final LCR! Economic Analysis

5-65

October 2024


-------
Issue

Addressed with
High/Low Scenario?

Direction of Bias

Typical exposure patterns based on water usage patterns in homes,
service line length, and the length of pipes between service line and
tap.

No

Unclear

The EPA assumes that the ages and number of people in households
with LSLs are the same as the distribution in general population

No

Unclear

Estimates of the 90th percentile water concentrations, which trigger
water systems to make changes due to AL exceedances (ALEs).

Yes

Unclear

The EPA does not quantify all health effects from lead exposure.

No

Underestimate

Both the SHEDS-PB and AALM models have been peer reviewed,
however some uncertainty remains in the selected parameters and
inputs. Assumptions around percentage of daily water for the home
tap, water ingestion rates, and lead absorption through the gut will
impact the modeled blood lead levels

No

Unclear

CVD premature mortality studies, LBW and ADHD studies use a
single measurement of BLL

No, discussed
qualitatively

Unclear

The EPA assumes that filters result in the lowest drinking water
concentration modeled. This assumes that everyone is properly and
consistently using the filter. Including after a lead or GRR service line
replacement when water lead concentrations may increase for a
short period of time not captured in EPA's water lead concentration
modeling.

No

Overestimate

The EPA assumes that both population and age distribution remain
stable over the study period.

No

Unclear

Uncertainty in the shape of the concentration-response function for
IQ, ADHD, LBW or CVD premature mortality.

Partially, except for
LBW a high and low
function are used.

Unclear

Uncertainty about the extent of the lag between changes in lead
exposure and reductions in risk for CVD premature mortality and the
fact that no cessation lag is used in the benefits modeling.

No

Overestimate (if
there is a cessation
lag see Section
5.7.7)

The EPA estimates avoided CVD premature mortality impacts for
adults ages 40 through 79 only.

No

Underestimate

ADHD and LBW valuations do not capture willingness-to-pay to avoid
the risk or any reductions in quality of life.

No

Underestimate

Valuation of avoided IQ point losses is not updated for future
changes in real wage growth.

No

Unclear

Final LCRI Economic Analysis

5-66

October 2024


-------
Issue

Addressed with
High/Low Scenario?

Direction of Bias

Other uncertainties

Timing of CCT changes and LSL replacements

No

Unclear

Uncertainty if the system changes source water or treatment
technology as a result of circumstances not related to the LCRI

No

Unclear

Acronyms: AALM = All Ages Lead Model; ADHD = Attention-Deficit/Hyperactivity Disorder; BLL = blood lead level;
CCT = corrosion control treatment; CVD = cardiovascular disease; GRR = galvanized requiring replacement; IQ =
intelligence quotient; LCRI = Lead and Copper Rule Improvements; LBW = low birth weight; LSL = lead service line.

5.7.1	Uncertainty in Blood Lead Modeling

For a discussion of the limitations in the drinking water concentration modeling, see Section 5.2.5. In
order to model the expected blood lead changes due to reductions in drinking water lead exposures, the
EPA used two models, both which have been peer reviewed. For children under seven, the EPA used the
SHEDS-Pb model, and for older children and adults, the AALM. While there is both uncertainty and
variability in the parameters and inputs to the models, both models have demonstrated that they
predict blood levels well. See Zartarian (2017; 2023) and USEPA (2019b). Certain parameters, such as
lead absorption through the gut and drinking water ingestion rate, as well as how much water is home
tap water vs. other sources will have the greatest impact on the resulting modeled blood levels.
However, in estimating the benefits, we are looking at the change between two modeled outputs, which
will minimize the effects of uncertainty in these assumptions.

The EPA models blood lead levels based on the drinking water concentrations in Exhibit 5-12. The EPA
assumes there is no difference in the geometric mean water lead concentration of systems with no LSL,
regardless of the CCT status. In other words, for each of the three scenarios of no LSL - no CCT, no LSL -
partial CCT, and no LSL - representative CCT, the geometric mean water lead concentration is
equivalent. A sensitivity analysis for the 2021 LCRR (USEPA, 2020a) demonstrated that this will result in
an underestimation of benefits, if it is assumed that there are additional water lead reductions with
improved CCT in the absence of an LSL. This is discussed further in Section 5.8.

The lead concentration estimates for soil, air, and food are held constant in the blood lead modeling in
order to represent background lead levels, with the only varying concentration being drinking water. It is
likely that exposure, from soil, air, and food, and resulting BLLs will vary over time, and this uncertainty
propagates across all modeled health benefits. Model simulations for adults also do not account for
higher historical lead exposures and long-term bone accumulation that may have occurred prior to the
baseline or regulatory scenarios. BLLs are reflective of both recent exposures (less than 30 days) and
past exposures (years to decades) that were stored in tissues (e.g., bone) and released endogenously
(Abt Associates, 2023; NTP, 2012).

5.7.2	General Uncertainty in Concentration-Response Relationships and Population

For all endpoints, there is uncertainty in the choice of studies from which to derive the concentration-
response relationship, the best functional form to describe the relationship, and the best method to

Final LCRI Economic Analysis

5-67

October 2024


-------
characterize avoided heath risk at blood lead levels below those observed in the literature, as blood lead
levels in the United States have declined over the years. For endpoint specific discussions, see Sections
5.7.4-5.7.7. For IQ, ADHD and CVD premature mortality, the EPA selected a low and high concentration-
response function from the available literature to estimate of the benefits. For LBW, the EPA
determined that there was only one available high-quality study, Zhu et al. (2010), therefore this was
the only function used for this endpoint. The shape of the concentration-response function impacts the
benefits estimates. For the same absolute change in BLLs (eg. 1 ng/dl), benefits will be smaller at higher
BLLs when using a log or square root function as compared to a linear function.

There will also be uncertainty on the size of the population in the future which will experience benefits.
However, while the fertility rate has decreased over time, there has been a simultaneous increase in the
overall U.S. population, resulting in a fairly constant overall number of children in the U.S. For example,
according to the American Community Survey196, in 2000 there were 19,046,754 children under 5 (total
population = 281 million), and in 2022 there were 19,004,925 children under 5 (total population = 331
million).

5.7.3 General Uncertainty in Valuation

If the EPA used a discount rate lower than 2 percent, it would generally result in an increase in the
estimated dollar amount for benefits above those estimated using a 2 percent discount rate. This
increase in benefits would result from both a higher baseline value of an IQ point, case of ADHD or CVD
premature mortality and lower birth weight due to the decreased discounting of future benefits.
Additionally, the use of the declining discount rate (see Section 5.5.2) when calculating the value of an
IQ point and avoided ADHD case, done to comply with OMB Circular A-4 (OMB, 2023), results in slightly
higher monetized IQand ADHD benefits than if that assumption was not made197. Changes in valuation
assumptions would not impact the overall risk reductions expected due to the rule.

For ADHD and LBW, the EPA uses a cost of illness approach. This approach was necessary as other values
were not available for these endpoints, but this approach may underestimate benefits compared to
other methods such as stated preference studies providing willingness-to-pay estimates (Woodruff,
2015, USEPA, 2010). Additionally, there is uncertainty in discounting benefits, particularly for children,
which occur in the future (Woodruff, 2015). While ADHD may be associated with LBW, there is no
double counting in the monetized benefits. The cost-of-illness for LBW only includes costs before age 2,
and the ADHD cost-of-illness only includes costs after ADHD diagnosis, not in early childhood. The EPA's
cost-of-illness estimate associated with LBW is an underestimate of the total impacts of low birth
weight, as it only includes two years of medical costs and does not include parental productivity loss or
other sequelae of low birth weight. Also, the valuation for ADHD may be underestimated because the
Doshi et al (2012) estimates do not include productivity losses in adulthood related top ADHD after

196 The U.S. Census Bureau's American Community Survey data is available at https://www.census.gov/programs-
survevs/acs/data.html.

197The EPA used a declining discount rate which begins with a 2 percent discount rate for the years 2024-2079, a
1.9 percent discount rate used for the years 2080-2096, and a 1.8 percent discount rate used in years 2095-2102.
This declining rate structure was implemented to comply with updates to OMB Circular A-4 guidance which
indicates that a declining discount rate may be used to capture the uncertainty in the appropriate discount rate
over long time horizons like lifetime labor force participation.

Final LCRI Economic Analysis

5-68

October 2024


-------
adjusting for IQ. Doshi et al. (2012) also does not include estimates of loss of employment or stress
related illness.

5.7.4 Uncertainty in IQ

The relationship between lead and IQ is well documented (see USEPA, 2024a and Appendix D) and the
approach used for the LCRI was reviewed by the SAB (USEPA, 2020b). The USEPA (2024a) ISA found
sufficient evidence to conclude that there is a causal relationship between lead exposure and cognitive
function decrements in children based on several lines of evidence including findings from prospective
studies in diverse populations and coherence with evidence in animals, and evidence identifying
potential modes of action. The NTP Monograph concluded that there is sufficient evidence of
association between blood lead levels <5 ng/dL and decreases in various general and specific measures
of cognitive function in children from 3 months to 16 years of age. This conclusion is based on
prospective and cross-sectional studies using a wide range of tests to assess cognitive function (National
Toxicology Program, 2012, p. 27).

However, there is still uncertainty in the approach to estimate IQ loss at lower blood lead levels, and the
best approach to extrapolate beyond the observed range of blood leads in a given study. This includes
uncertainty in the functional form of the dose response relationship and uncertainty in which study best
describes the relationship (see Appendix D, Section D.7.1.2).

There is a detailed discussion of the methodology and limitations around the value of an IQ point in
Appendix J of the Final 2021 LCRR EA and in USEPA (2019b). In addition, for this analysis, the EPA
included an alternative valuation of an IQ point by Lin et al., (2018) as a sensitivity analysis presented in
Appendix F. Briefly, uncertainties regarding the IQ-earnings relationship underlying the value of an IQ
point include measurement error, lack of controls for non-cognitive skills that also affect test
performance, and potential for bias due to other omitted variables likely to be correlated with both test
performance and earnings, such as a supportive household or extra educational resources. Another
uncertainty is to what extent estimated relationships from the published literature, which are based on
historic data, will apply to future populations. Similarly, uncertainties may be introduced from applying
average estimates based on a representative sample of the U.S. population to smaller subgroups that
may be disproportionately affected by regulation.

Salkever (1995) explicitly modeled the role of education in the IQ-earnings relationship, which sheds
light on the mechanism by which cognitive skills affect earnings and also allows the EPA to account for
educational costs when calculating the change in net lifetime earnings from a change in IQ. The EPA
reanalysis of Salkever (1995) relies on data where respondents range in age from 27 to 32. Extrapolating
the Salkever (1995) IQ-earnings effect at age 30 will generate an estimate of lifetime earnings that is
biased downward if the effect of IQ on earnings grows over the lifecycle, a result found in Barth et al.
(1984), Zax and Rees (2002), Ganzach (2011), and Lin et al. (2018). An advantage of Lin et al. (2018) is
the use of data that extends throughout the lifecycle up to age 50. However, their analysis lacked some
control variables included in Salkever, and their inclusion of non-cognitive traits that are correlated with
IQ and may be affected by lead exposure in the regression may attenuate the estimated effect of IQ on
earnings, leading to a downward bias on the estimate of the total earnings effects of reduced lead
exposure.

Final LCRI Economic Analysis

5-69

October 2024


-------
5.7.5 Uncertainty in ADHD

As described in Appendix D the USEPA (2024a) ISA strengthened the conclusions of the 2013 ISA and
concluded that there was a causal relationship between lead exposure and inattention, impulsivity, and
hyperactivity in children based on recent studies of children with group mean BLLs < 5 |J.g/dL. The 2024
ISA states that "prospective studies of ADHD, including a study of clinical ADHD that controlled for
parental education and SES, although not quality of parental caregiving reported positive associations"
(USEPA, 2024a, p. IS-30). The causes of ADHD are not fully understood, but research suggests a number
of potential causes, including genetics, exposure to environmental toxins, prenatal exposure to cigarette
smoke or alcohol, and brain changes (Tripp et al., 2009; Pliszka et al., 2007). The EPA's 2013 lead ISA
stated that in children, "attention was associated with biomarkers of Pb exposure representing several
different lifestages and time periods. Prospective studies did not examine a detailed Pb biomarker
history, and results do not identify an individual critical lifestage, time period, or duration of Pb
exposure associated with attention decrements in children. Associations in prospective studies for
attention decrements with tooth Pb level, early childhood average and lifetime average blood Pb levels
point to an effect of cumulative Pb exposure." The 2024 ISA addresses the uncertainties presented in
the 2013 ISA by stating that "The largest uncertainty addressed by the recent evidence base is the
previous lack of prospective studies examining ADHD (Appendix 3.5.2.4-3.5.2.5). The bulk of the recent
evidence comprises prospective studies that establish the temporality of the association between Pb
exposure and parent or teacher ratings of ADHD symptoms and clinical ADHD. Across studies,
associations were observed with tooth Pb concentrations, childhood BLLs (<6 ng/dL), and with maternal
or cord BLLs (2-5 ng/dL)." The available studies relating blood lead to ADHD use one-time BLLs, while it
is possible that cumulative exposure is also important. However, one-time and cumulative measures of
BLLs in children are often correlated.

There are several sources of uncertainty in our choice of beta estimates for ADHD. In the 2001-2004
NHANES cycle, children without DSM-IV diagnostic data were younger, of lower SES, and more highly
exposed to both Pb and environmental tobacco smoke than children with DSM-IV diagnostic data
(Froehlich et al., 2009). This may have resulted in a downward bias of the effect size estimate. In Ji et al.
(2018), the study population consisted mostly of a low-income, minority, urban population and is not
representative of the entire US population. However, this population may be relevant as an at risk,
potentially more highly exposed population for regulatory analyses of policies to reduce lead exposure.

An additional source of uncertainty in Froehlich et al. (2009) is the use of one-time, concurrent blood Pb
measures to predict ADHD cases. However, the EPA ISA and NTP Monograph cite several prospective
cohort studies that provide support for the association between Pb and ADHD symptoms, and Ji et al.
(2018) is a prospective study that used early childhood (before age 4) Pb lead measures and found an
association with ADHD measured later in childhood. In addition, the only study on reverse causation
(David et al., 1977) identified in a literature search by Nigg et al. (2008) did not find any evidence for
reverse causation in ADHD and Pb exposure. Instead, David et al. (1977) found that children with
hyperactivity symptoms as a result of a known organic etiology had lower levels of Pb exposure than
children with no known etiology.

It remains unclear whether concurrent blood Pb measure most accurately represents association
between Pb and ADHD or if another blood Pb measure (e.g., lifetime average, early childhood) would be
a better measure. There were no studies that use repeated measures which include early childhood and

Final LCRI Economic Analysis

5-70

October 2024


-------
concurrent measures. However, Froehlich et al. (2009) draw a parallel with studies of IQ, noting that
concurrent blood Pb levels are strongly associated with decreases in cognitive function when compared
with early childhood or peak blood Pb levels. Ji et al., (2018) also found that early childhood lead levels
are associated with ADHD.

For a description of the uncertainties with the valuation used for ADHD which may result in an
underestimation, see Section 5.7.3.

5.7.6	Uncertainty in Reductions in Birth Weight

Unlike the other health endpoints, only one dose response function was used for estimating the
relationship between blood lead and reductions in birth weight. Zhu et al. (2010) was the strongest
study identified, and this choice was supported by peer reviewers (Versar, 2015). A discussion of the
limitations of the Zhu et al. (2010) study can be found in Abt Associates (2022b); however, these should
not prohibit the use of the concentration-response function.

The 2024 Pb ISA expands on the findings of the 2013 Pb ISA, specifically regarding Pb exposure and
effects on preterm birth and low birthweight; thus, the evidence was sufficient to conclude that there is
"likely to be a causal relationship between Pb exposure and effects on pregnancy and birth outcomes"
(USEPA, 2024a, pp. IS-51-52). The 2024 ISA also acknowledges the "uncertainties related to the specific
biomarkers of exposure associated with pregnancy and birth outcomes, the critical window of exposure,
and potential confounding by co-occurring metals" (USEPA, 2024a, pp. IS-52). It is possible that the
timing of the prenatal Pb exposure is key, and that studies that are cross-sectional in nature do not
detect an association because the critical exposure window that would result in decreased birth weight
has passed when blood Pb measurements are taken at birth. This possibility is supported by the fact that
the vast majority of identified cohort studies that measured blood Pb levels prior to birth found an
inverse and statistically significant relationship with birth weight.

There are also uncertainties and limitations related to the valuation estimate for low birth weight, which
are described in detail in Abt Associates (2022c) which was externally peer-reviewed (MDB, 2022).
Overall, these will likely lead to an underestimation in the quantified benefits. The main source of
underestimation is that not all costs are included, and that the MEPs data has been shown to
underestimate costs by about 10% due to both underutilization of healthcare and respondents
underreporting care. Additionally, Due to the use of the pregnancy detail file in the analysis there was a
limited sample size (Abt Associates 2022c).

5.7.7	Uncertainty in Cardiovascular Disease Premature Mortality Benefits

Detailed discussions of the uncertainty and variability associated with quantified CVD premature
mortality benefits are provided in Brown et al., (2020) and Abt Associates (2023).198 This section briefly
summarizes these uncertainties. First, there is uncertainty about the functional form which best
describes the relationship between blood Pb and CVD premature mortality. In order to partially mitigate

198 Note the Abt Associates (2023) methodology was externally peer review. See the MDB, Inc. (2019) "Selection of
Concentration-Response Functions between Lead Exposure and Adverse Health Outcomes for Use in Benefits
Analysis: Cardiovascular-Disease Related Mortality" peer review documentation at
https://cfpub.epa.gov/si/si_public_record_report.cfm?Lab=NCEE&dirEntrylD=342855

Final LCRI Economic Analysis

5-71

October 2024


-------
this, the EPA chose a high and low function based on those identified. There is also uncertainty in the
exposure window that is most relevant to the risk outcome. Brown et al., (2020) and Abt Associates
(2023) hypothesize that four major conceptual models may explain the temporal relationship between
blood Pb and CVD premature mortality:

Model 1. CVD premature mortality risk=/(one-time blood Pb).

Model 2. CVD premature mortality risk=/(average blood Pb over x years).

Model 3. CVD premature mortality risk=/(average blood Pb overxyears)+latency.

Model 4. CVD premature mortality risk=/(peak blood Pb).

However, when considering the relationship between adult Pb exposure and CVD premature mortality,
it is uncertain which conceptual model is best. No studies used repeated measures of the biomarkers in
evaluating the connection between blood Pb and CVD premature mortality. This resulted in uncertainty
because blood Pb is reflective of both recent exposures (less than 30 days) and past exposures (years to
decades) that were stored in tissues (e.g., bone) and released endogenously (Abt Associates, 2023; NTP,
2012). Therefore, as the EPA ISA (2024a) points out, uncertainties remain with respect to the timing,
frequency, and magnitude of Pb exposure that best correlate with CVD premature mortality risk.

Additionally, as described in Brown et al., (2022), given the consistent finding across the literature, it can
be concluded that the one-time measurement from NHANES is a predictor of CVD premature mortality,
either because Model 1 is true or because it is a proxy measure due to its correlation with average blood
Pb models over time (e.g., Model 2). Currently, Model 1 is the only model for which requisite data are
available given that there are no studies, or data sources such as NHANES, published to date evaluating
multiple blood Pb measurements in the same individual in association with CVD premature mortality.
Therefore, the result is interpreted as CVD premature mortality risk being a function of the one-time
blood Pb measurement.

Finally, although there is a lag between the blood Pb sample and death due to CVD premature mortality,
both the Cox proportional hazards model and the Poisson regression analysis assume that the hazard
ratio will be the same regardless of the follow-up time frame. There is no cessation lag or latency
assumed in the model. Therefore, the result is interpreted as CVD premature mortality risk being a
function of the one-time blood Pb measurement. Incorporating a cessation lag would result in a
decrease in monetary benefits, as the valuation of the avoided risk would occur in later years, and be
more heavily discounted than in the current analysis.

In order to reduce the uncertainties associated with estimating the cardiovascular premature mortality
effects of changes in adult lead exposure, studies using novel datasets or approaches will likely be
required, such as recent bone lead measurements, repeated measurements of blood lead from the
same individuals over time, or quasi-experimental variation in adult lead exposure linked to
cardiovascular outcomes. If such studies become available in the future, EPA will consider updating the
methodology for estimating the cardiovascular premature mortality effects of changes in adult lead
exposure as appropriate.

Final LCRI Economic Analysis

5-72

October 2024


-------
5.8 Summary of Non-Quantified and Non-Monetized Benefits

In addition to the benefits monetized in the final LCRI analysis for reductions in lead exposure, there are
several other benefits that are not quantified. The risk of adverse health effects due to lead exposure
that are expected to decrease as a result of the final LCRI are summarized in Appendix D and are
expected to affect both children and adults. The EPA focused its non-quantified impacts assessment on
the endpoints identified using two comprehensive U.S. Government documents summarizing the
literature on lead exposure health impacts. These documents are the EPA's Integrated Science
Assessment for Lead (ISA) (USEPA, 2024a), and the U.S. Department of Health and Human Services' NTP
Monograph on Health Effects of Low-Level Lead (NTP, 2012). Both sources present comprehensive
reviews of the literature as of the time of publication on the risk of adverse health effects associated
with lead exposure. The EPA summarized those endpoints to which either the EPA ISA or the NTP
Monograph assigned one of the top two tiers of confidence in the relationship between lead exposure
and the risk of adverse health effects. These endpoints include cardiovascular morbidity effects, renal
effects, reproductive and developmental effects (apart from ADHD and low birth weight), immunological
effects, neurological effects (apart from children's IQ), and cancer.

There are a number of final LCRI requirements that reduce lead exposure to both children and adults
that the EPA could not quantify. The final rule will require additional lead public education requirements
that target consumers directly, schools and child care facilities, health agencies, and people living in
homes with lead and GRR service lines. Increased education will lead to additional averting behavior on
the part of the exposed public, resulting in reductions in the negative impacts of lead. The rule will also
require the development of service line inventories that include additional information on lead
connectors and make the location of the lead content service lines publicly accessible. This will give
potentially exposed consumers more information and will provide potential home buyers with this
information as well. Homeowners may request LSL/GRR service line removal earlier than a water system
might otherwise plan on replacing the line. The benefits of moving these lead and GRR service line
removals forward in time are not quantified in the analysis of the final LCRI. Because of the lack of
granularity in the lead tap water concentration data available to the EPA for the regulatory analysis, the
benefits of small improvements in CCT to individuals residing in homes with lead content service lines,
like those modeled under Distribution System and Site Assessment are not quantified.

The EPA also did not quantify the CCT benefits of reduced lead exposure from lead-containing plumbing
components (not including from LSL/GRRs) to individuals who reside in both: 1) homes that have
LSL/GRRs but also have other lead-containing plumbing components, and 2) those that do not have
LSL/GRRs but do have lead-containing plumbing components.199 The EPA has determined that the final
LCRI requirements may result in reduced lead exposure to the occupants of both these types of
buildings as a result of improved monitoring and additional actions to optimize CCT. In the analysis of
the LCRI, the number of both LSL/GRR and non-LSL/GRR homes potentially affected by water systems
increasing their corrosion control during the 35-year period of analysis is 5.2 million in the low scenario

199 Although the EPA estimated an average lead concentration for the first 10 liters of drinking water to inform the
water lead concentration estimates used to quantify benefits the EPA could not calculate the CCT benefits
associated with lead containing plumbing components (apart from LSL/GRR service lines), because the EPA used a
pooled estimate for all CCT conditions in residences with no LSL/GRR in place (See Section 5.2.3 for additional
information).

Final LCRI Economic Analysis

5-73

October 2024


-------
and 9.1 million in the high scenario. Some of these households may have leaded plumbing materials
apart from LSL/GRRs, including leaded brass fixtures and lead solder. These households could potentially
see reductions in tap water lead concentrations.

Some researchers have pointed to the potential for CCT cobenefits associated with reduced corrosion,
or material damage, to plumbing pipes, fittings, and fixture, and appliances that use water owned by
both water systems and homeowners (Levin, 2023). The corrosion inhibitors used by systems that are
required to install or re-optimize CCT as a result of the final LCRI are expected to result in additional
benefits associated with the increased useful life of the plumbing components and appliances (e.g.,
water heaters), reduced maintenance costs, reduced treated water loss from the distribution system
due to leaks, and reduced potential liability and damages from broken pipes in buildings that receive
treated water from the system. The replacement of GRR service lines may also lead to reduced treated
water loss from the distribution system due to leaks (AwwaRF and DVGW-Technologiezentrum Wasser,
1996). The EPA did not have sufficient information to estimate these impacts nationally for the final LCRI
rule analysis.

Additionally, the risk of adverse health effects associated with copper that are expected to be reduced
by the final LCRI are summarized in Appendix E. These risks include acute gastrointestinal symptoms,
which are the most common adverse effect observed among adults and children. In sensitive groups,
there may be reductions in chronic hepatic effects, particularly for those with rare conditions such as
Wilson's disease and children predisposed to genetic cirrhosis syndromes. These diseases disrupt copper
homeostasis, leading to excessive accumulation that can be worsened by excessive copper ingestion
(NRC, 2000).

5.9 Disbenefits from Greenhouse Gas Emissions

The EPA is committed to understanding and addressing climate change impacts in carrying out the
agency's mission of protecting human health and the environment. While the EPA is not required by
SDWA 1412(b)(3)(C) to consider climate disbenefits under the HRRCA the agency has estimated the
potential climate disbenefits caused by increased greenhouse gas (GHG) emissions associated with the
operation of CCT at drinking water treatment facilities and the use of construction and transport
vehicles in the replacement of LSLs and GRR service lines. As explained in section VI.A of the preamble,
this disbenefits analysis is presented solely for the purpose of complying with E.O. 12866.

This section is broken into three parts that discuss the steps the EPA took to estimate the climate
disbenefits associated with the final LCRI:

•	Sub-section 5.9.1 describes the estimation of the per unit energy consumed in the operation of
CCT and SLR, and the per unit GHG emissions associated with the energy consumed.

•	Sub-section 5.9.2 discusses the calculation of the total incremental emissions in the SafeWater
LCR model.

•	Sub-section 5.9.3 describes the social cost of GHG estimates used to monetize the climate
disbenefits and presents the results of the EPA analysis.

Final LCRI Economic Analysis

5-74

October 2024


-------
5.9.1 Energy Consumption and Unit Greenhouse Gas Emissions

As part of the estimation of the climate disbenefits associated with the requirements of the final LCRI
the EPA developed:

•	Estimates of annual energy consumed to operate CCT by types of CCT (pH adjustment,
orthophosphate addition, or both pH and orthophosphate use) and system size;

•	Estimates of energy consumed during the replacement of a lead service line which includes the
use of diesel medium/heavy-duty vehicles and backhoe excavation equipment; and

•	Unit greenhouse gas emissions for both electricity and vehicle fuel consumption.

This section describes the steps the EPA used to collect the energy consumption information and unit
emission information.

5.9.1.1 Energy Consumption Estimates

The first step in estimating the incremental annual greenhouse gas emissions for the final LCRI is to
develop unit energy consumption estimates for operating CCT and conducting a service line
replacement.

5.9.1.1.1 Electricity Consumption Operating CCT

The EPA estimated the electricity consumed by a system, in each system size category and by each type
of CCT (pH adjustment, orthophosphate addition, or both pH and orthophosphate use), that would be
required to operate CCT under the final LCRI. Estimates of electricity use have already been calculated as
part of the CCT unit costing effort described in Chapter 4, Section 4.3.3. The EPA uses the electricity
consumption values derived from its CCT Work Breakdown Structure (WBS) cost models, which are
described in Section 4.2.2.3 and detailed in Technologies and Costs for Corrosion Control to Reduce Lead
in Drinking Water (USEPA, 2023b). The estimated WBS O&M cost equations are a function of average
daily flow (ADF) and include electricity consumption for the following:

•	Increased building lighting requirements associated with the incremental operator labor required to
operate the corrosion control processes.

•	For large systems of 1 million gallons per day (MGD) or greater design flow, periodic operation of
transfer pumps that move corrosion control chemicals between bulk storage tanks and day tanks.

The models assume that smaller systems (less than 1 MGD design flow) do not include transfer pumping
because they feed chemicals directly from 55-gallon drums. The models assume electricity consumption
by chemical metering pumps is negligible because these pumps require less than 1 horsepower to
operate. There are no other sources of electricity consumption associated with the corrosion control
processes.

As shown in Exhibit 5-38, the WBS model outputs are specific to the type of corrosion control
technology and available for each of eight model system size categories. These outputs assume
installation and operation of a new chemical addition process.

Final LCRI Economic Analysis

5-75

October 2024


-------
Exhibit 5-38: Corrosion Control Treatment Total Annual Electricity Consumption by System

Size and Type of Chemical Addition

Technology and
Population Served

Design
Flow
(MGD)

Average
Flow
(MGD)

Lighting
Electricity
Consumption
(MWh/year)

Transfer Pump

Electricity
Consumption
(MWh/year)

Total Electricity
Consumption
(MWh/year)

Centralized Orthophosp

late Treatment (a)

25 to 100

0.03

0.007

0.000861875

0

0.000861875

101 to 500

0.124

0.035

0.000861875

0

0.000861875

501 to 1,000

0.305

0.094

0.000861875

0

0.000861875

1,001 to 3,300

0.74

0.251

0.0010452

0

0.0010452

3,301 to 10,000

2.152

0.819

0.0014228

0.017011281

0.018434081

10,001 to 50,000

7.365

3.2

0.0022218

0.017011281

0.019233081

50,001 to 100,000

22.614

11.087

0.005738175

0.017011281

0.022749456

Greater than 100,000

75.072

37.536

0.011041875

0.017011281

0.028053156

Centralized pH Adjustment (b)

25 to 100

0.03

0.007

0.000921875

0

0.000921875

101 to 500

0.124

0.035

0.000921875

0

0.000921875

501 to 1,000

0.305

0.094

0.000921875

0

0.000921875

1,001 to 3,300

0.74

0.251

0.0015188

0

0.0015188

3,301 to 10,000

2.152

0.819

0.0023658

0.017011281

0.019377081

10,001 to 50,000

7.365

3.2

0.0037458

0.017011281

0.020757081

50,001 to 100,000

22.614

11.087

0.010253075

0.017011281

0.027264356

Greater than 100,000

75.072

37.536

0.016692575

0.034022563

0.050715138

Both Centralized pH Adjustment and Orthophosphate Treatment (a, b)

25 to 100

0.03

0.007

0.00178375

0

0.00178375

101 to 500

0.124

0.035

0.00178375

0

0.00178375

501 to 1,000

0.305

0.094

0.00178375

0

0.00178375

1,001 to 3,300

0.74

0.251

0.002564

0

0.002564

3,301 to 10,000

2.152

0.819

0.0037886

0.034022563

0.037811163

10,001 to 50,000

7.365

3.2

0.0059676

0.034022563

0.039990163

50,001 to 100,000

22.614

11.087

0.01599125

0.034022563

0.050013813

Greater than 100,000

75.072

37.536

0.02773445

0.051033844

0.078768294

Notes: (a) assumes phosphate addition of 3.2 milligrams per liter as phosphate; (b) assumes pH increase
from 6.8 to 7.5. MGD = million gallons per day; MWh = megawatt hours

5.9.1.1.2 Service Line Replacement Fuel Consumption

The EPA used the following assumptions in the estimation of the fossil fuel energy consumed as part of
an average service line replacement:

• Diesel medium/heavy duty trucks would be used to transport work supplies and backhoe to the
SLR site. The assumption of a medium/heavy duty truck is necessary given the towing capacity
needed to haul the backhoe to the SLR site.

Final LCRI Economic Analysis

5-76

October 2024


-------
•	The weighted average round trip distance to each SLR site is estimated to be 12.32 miles. This
value comes from the estimated driving distance per system size category as presented in
Section 4.3.2.1.2, Exhibit 4-14, multiplied by two to approximate the round trip distance, and
weighted by the estimated total number of lead (partial and full) and GRR service lines to be
replaced in each size category as presented in Section 3.3.4.1.2, Exhibit 3-19200. Note that the
EPA is assuming that each SLR requires an individual round trip from a central staging area and
that there is no grouping as would likely occur in the case of planned scheduled SLRs, therefore,
the estimated emissions from SLR are likely overestimated.

•	Backhoes would operate on site for one hour consuming two gallons of diesel. The two gallon
per hour value is the central estimate from a 1.5 to 2.5 range provided by the website:
https://cpower.com/2021/ll/16/types-of-gas-for-your-rental-construction-vehicle/.

5.9.1.2 Converting Consumed Energy Estimates into Greenhouse Gas Emissions

As a second step, the EPA developed estimates of greenhouse gas emissions associated with the
electricity used to operate CCT and the fossil fuels used in the replacement of service lines.

5.9.1.2.1 Electricity Emissions

To convert the estimated increase in electricity use associated with the annual operation of CCT into
greenhouse gas emissions per system, the EPA used the latest reference case from the EPA's peer-
reviewed Integrated Planning Model (IPM) (USEPA, 2023d).201 The EPA uses the IPM to analyze the
projected impact of environmental policies on the electric power sector, and it also provides projections
of C02 emissions from the power sector through 2055. The latest reference case, "Post-IRA 2022
reference case" was published in April of 2023 and reflects the impacts of the Inflation Reduction Act
(IRA).

Although the U.S. electricity grid continues to decrease its reliance on coal combustion in favor of
natural gas and renewable alternatives, electricity consumption continues to be associated with GHG
emissions across the entire system of production and delivery. Combustion of fossil fuels releases C02,
CH4, and N20; sulfur hexafluoride (SF6) and perfluorocarbons (PFCs) are used in electricity transmission
and distribution equipment; and additional GHG emissions are associated with the manufacture and
installation of equipment as well the extraction and delivery of fossil fuels (USEPA, 2023c). An exact
accounting of all these emissions categories would yield the most precise estimate of electricity sector
climate-related impacts. However, emissions from fossil fuel combustion comprise the vast majority of
the electricity sector GHG emissions, 90 to 97 percent. Therefore, accounting for GHG combustion
emissions is sufficient for the purposes of estimating the approximate magnitude of the climate-related

200	For detailed calculations, see the derivation file "Service Line Characterization using DWINSA_Final.xlsx,"
worksheet "Mileage Weighted by SLs Replace," available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

201	The IPM is a multi-regional, dynamic, deterministic linear programming model of the U.S. electric power sector.
It provides projections of least-cost capacity expansion, electricity dispatch, and emission control strategies for
meeting energy demand and environmental, transmission, dispatch, and reliability constraints (USEPA,).

Final LCRI Economic Analysis

5-77

October 2024


-------
disbenefits of increased electricity consumption associated with the LCRI requirements.202 Note that the
non-GHG emissions impacts associated with changes in electricity consumption are not accounted for in
this analysis. For a more complete description of non-GHG impacts from the electricity sector, including
ozone- and PM2.5-attributable premature mortality and illness as well as discussion of various
unquantified health and welfare impacts, see recent EPA regulatory impact analyses for air pollution
regulations and the utilities sector in particular (USEPA, 2023e).

From IPM reference case summary outputs, the EPA calculated projections of annual national-average
emissions per MWh of electricity generation over the LCRI period of analysis. The EPA mapped IPM
model C02 data to calendar years, corresponding to the 35 years in the LCRI period of analysis, following
the IPM documentation guidance (USEPA, 2023f).203 Exhibit 5-39 shows the IPM summary outputs and
implied national-average C02 emissions factors for each IPM model year. The EPA also used information
from USEPA (2021) to estimate the amount of CH4 and N20 produced for each unit of C02 produced per
unit of electricity. The EPA used these values to estimate CH4 and N20 produced per MWh/year.

Exhibit 5-39: Emissions per MWh Calculated from Post-IRA 2022 IPM Reference Case



CO2









IPM
Model
Year

Emissions
(Million
Metric
Tons/year)3

Grand Total Electricity
Generated (Billions
MWh/year)a

CO2 Emissions
(mt/MWh/year)

CH4 Emissions
(mt/MWh/year)b

N20 Emissions
(mt/MWh/year)b

2028

1,222

4.409

0.277

0.104

0.015

2030

672

4.545

0.148

0.057

0.008

2035

608

4.891

0.124

0.052

0.008

2040

481

5.265

0.091

0.041

0.006

2045

406

5.628

0.072

0.034

0.005

2050

357

6.071

0.059

0.030

0.004

2055

391

6.454

0.061

0.033

0.005

aSource: Post IRA Reference Case SSR.xIsx available at: https://www.epa.gov/power-sector-modeling/post-ira-
2022-reference-case.

bln order to estimate the CH4 and N2O produced per MWh per year the EPA used data from USEPA, 2021 to
estimate the amount of CH4 and N2O produced for each unit of CO2 produced.

5.9.1.2.2 Fuel Emissions

In order to develop the estimated emissions that result from lead and GRR service line replacement the
EPA utilized the emissions factors found in its 2021 publication Emission Factors for Greenhouse Gas

202 See the EPA's Economic Analysis for the Final Per- and Polyfluoroalkyl Substances National Primary Drinking
Water Regulation (USEPA, 2024d) for additional discussion on the relative percent of fossil fuel combustion to
other sources of GHG in the electricity sector.

203The EPA mapped the calendar years 2027 and 2028 to IPM run year 2028, calendar years 2029-31 to IPM run
year 2030, calendar years 2032-37 to IPM run year 2035, calendar years 2038-42 to IPM run year 2040, calendar
years 2043-47 to IPM run year 2045, calendar years 2048-52 to IPM run year 2050, and calendar years 2053-58 to
IPM run year 2055.

Final LCRI Economic Analysis

5-78

October 2024


-------
Inventories (USEPA, 2021). The C02, CH4, and N20 emission values per mile for medium/heavy duty
trucks and per gallon for backhoes are presented in Exhibit 5-40.

Exhibit 5-40: Greenhouse Gas Emission Values Per Mile or Gallon of Fuel

Assumption

Value

Source and Notes

Medium/Heavy
Duty Truck kg
C02 per vehicle
mile

1.407

USEPA, 2021, Table 8. Vehicle-mile factors are appropriate to use when
the entire vehicle is dedicated to transporting the reporting company's
product.

Medium/Heavy
Duty Truck kg
CH4 per vehicle
mile

0.000013

USEPA, 2021, Table 8, converted from g/vehicle-mile. Vehicle-mile
factors are appropriate to use when the entire vehicle is dedicated to
transporting the reporting company's product.

Medium/Heavy
Duty Truck kg
N20 per vehicle
mile

0.000033

USEPA, 2021, Table 8, converted from g/vehicle-mile. Vehicle-mile
factors are appropriate to use when the entire vehicle is dedicated to
transporting the reporting company's product.

Backhoe kg C02
per gallon

10.21

USEPA, 2021, Table 2

Backhoe kg CH4
per gallon

0.0002

USEPA, 2021, Table 4 for construction/mining equipment, converted
from g/gallon

Backhoe kg N20
per gallon

0.00047

USEPA, 2021, Table 4 for construction/mining equipment, converted
from g/gallon

Source: USEPA. 2021. Emission Factors for Greenhouse Gas Inventories. Retrieved from
epa.aov/sites/default/files/2021 -04/documents/emission-factors apr2021.pdf

5.9.2 Calculating Annual Total Incremental Emissions in SafeWater LCR

The EPA tracks and compiles the GHG emissions for both CCT and SLR in the SafeWater LCR model as
step three in the estimation of climate disbenefits. The SafeWater LCR model applies a unit emission
value each time an activity (CCT installation, and LSL/GRR service line replacement) is triggered in the
model.

In the case of CCT see Chapter 3, Section 3.3.5.1 for a discussion of the estimated percent of systems
both in the baseline 2021 LCRR and the final rule that will exceed the action levels from both rules
(0.015 mg/L for 2021 LCRR and 0.010 mg/L for the final LCRI) and must install new CCT. Also see Chapter
4, Section 4.4.3, which provides estimates for the type of CCT to be install or re-optimized be it pH,
orthophosphate, or both pH and orthophosphate. Note that the emissions tracked by the model are
associated with the annual energy consumed to operate CCT, therefore, like CCT O&M costs, the EPA
applies the emission value in the year in which CCT is installed and in each year after until the end of the
period of analysis. Exhibit 5-41 provides the estimated annual per CCT installation C02, CH4, and N20
emission values.

Final LCRI Economic Analysis

5-79

October 2024


-------
Exhibit 5-41: Estimated Emissions per CCT Installation

Technology and
Population Served

Total
Electricity
Consumption
(MWh/year)

C02 Emissions
(mt/MWh/year)

CH4 Emissions
(mt/MWh/year)

N20 Emissions
(mt/MWh/year)

Centralized Orthophosphate Treatment (a)

25 to 100

0.000861875

2.39E-04

2.02557E-08

2.97282E-09

101 to 500

0.000861875

2.39E-04

2.02557E-08

2.97282E-09

501 to 1,000

0.000861875

2.39E-04

2.02557E-08

2.97282E-09

1,001 to 3,300

0.0010452

2.90E-04

2.45642E-08

3.60516E-09

3,301 to 10,000

0.018434081

5.11E-03

4.33236E-07

6.35838E-08

10,001 to 50,000

0.019233081

5.33E-03

4.52014E-07

6.63397E-08

50,001 to 100,000

0.022749456

6.31E-03

5.34656E-07

7.84686E-08

Greater than 100,000

0.028053156

7.78E-03

6.59303E-07

9.67624E-08

Centralized pH Adjustment

b)

25 to 100

0.000921875

2.56E-04

2.16658E-08

3.17978E-09

101 to 500

0.000921875

2.56E-04

2.16658E-08

3.17978E-09

501 to 1,000

0.000921875

2.56E-04

2.16658E-08

3.17978E-09

1,001 to 3,300

0.0015188

4.21E-04

3.56947E-08

5.23872E-09

3,301 to 10,000

0.019377081

5.37E-03

4.55398E-07

6.68364E-08

10,001 to 50,000

0.020757081

5.75E-03

4.87831E-07

7.15964E-08

50,001 to 100,000

0.027264356

7.56E-03

6.40764E-07

9.40416E-08

Greater than 100,000

0.050715138

1.41E-02

1.1919E-06

1.74929E-07

Both Centralized pH Adjustment and Orthophosphate Treatment (a, b)

25 to 100

0.00178375

4.95E-04

4.19215E-08

6.1526E-09

101 to 500

0.00178375

4.95E-04

4.19215E-08

6.1526E-09

501 to 1,000

0.00178375

4.95E-04

4.19215E-08

6.1526E-09

1,001 to 3,300

0.002564

7.11E-04

6.02589E-08

8.84388E-09

3,301 to 10,000

0.037811163

1.05E-02

8.88634E-07

1.3042E-07

10,001 to 50,000

0.039990163

1.11E-02

9.39845E-07

1.37936E-07

50,001 to 100,000

0.050013813

1.39E-02

1.17542E-06

1.7251E-07

Greater than 100,000

0.078768294

2.18E-02

1.85121E-06

2.71692E-07

Notes: (a) assumes phosphate addition of 3.2 milligrams per liter as phosphate; (b) assumes pH increase from 6.8
to 7.5. MGD = million gallons per day; MWh = megawatt hours

The SafeWater LCR model also applies a unit emission value per LSL/GRR service line replacement in
each period when a replacement occurs. This value is the sum of both the per mile emissions for
transportation to the replacement location using a medium/heavy duty vehicle and the per gallon
emissions for the use of a backhoe for onsite excavation. See Chapter 3, Section 3.3.4 for the estimated

Final LCR! Economic Analysis

5-80

October 2024


-------
number of lead and GRR service lines in the baseline, and see Chapter 1, Section 1.1 for the 2021 LCRR
and final LCRI replacement requirements. Exhibit 5-42 shows the per service line replacement C02, CH4,
N20 total emissions.

Exhibit 5-42: Estimated Emissions Per Service Line Replacement

Type of
Greenhouse
Gas

Medium/heavy
Duty Vehicle
Emissions
(kg/SLR)

Backhoe
Emissions
(kg/SLR)

Total Emissions
(kg/SLR)

o
u

16.41969

20.42

36.83969

ch4

0.00015171

0.0004

0.00055171

n2o

0.00038511

0.00094

0.00132511

As indicated in Chapter 4, Exhibit 4-134, the SafeWater LCR model estimated that over the 35-year
period of analysis the final LCRI results in an incremental increase in lead and GRR service line
replacement of between 6,392,911 and 6,109,511 under the low and high scenarios, respectively.
Incremental installation of CCT is projected to occur at 3,089 systems under the low scenario and 3,994
systems under the high scenario. Exhibit 5-43 shows the estimated total incremental GHG emissions by
year for the final LCRI. As indicated in the exhibit, the primary source of GHG emissions is associated
with SLR as indicated by the relatively high emissions in the years 4 through 13. Also note that for a
number of years at the beginning and end of the period of analysis, projected 2021 LCRR GHG emissions
are higher than final LCRI emissions, resulting in negative incremental GHG emissions for the final LCRI in
those periods. Total incremental emissions over 35 years are positive for the final LCRI with ranges of
241,504 to 230,669 metric tons of C02, 9 to 8 metric tons of CH4, and 4 to 3 metric tons of N20, between
the low and high scenarios.

Exhibit 5-43: Estimated Total Annual Incremental Greenhouse Gas Emissions for Final LCRI

SafeWater
Model
Period

Calander
Year

Low Scenario

High Scenario

Emission Changes (metric tons)

Emission Changes (metric tons)

C02

ch4

N20

C02

ch4

NzO

1

2024

(1593.002)

(0.057)

(0.024)

(2857.419)

(0.102)

(0.042)

2

2025

(1591.932)

(0.057)

(0.024)

(2874.874)

(0.103)

(0.043)

3

2026

(1588.819)

(0.057)

(0.024)

(2901.083)

(0.104)

(0.043)

4

2027

23679.172

0.847

0.351

22433.824

0.802

0.333

5

2028

23675.548

0.846

0.351

22434.117

0.800

0.333

6

2029

24290.219

0.868

0.360

23126.776

0.825

0.343

7

2030

24874.708

0.889

0.369

24702.458

0.881

0.367

8

2031

24867.434

0.890

0.369

24715.734

0.885

0.367

9

2032

24721.220

0.884

0.367

24532.075

0.878

0.364

10

2033

24713.680

0.884

0.367

24528.046

0.878

0.364

11

2034

24703.196

0.884

0.366

24527.649

0.878

0.364

12

2035

24705.405

0.884

0.366

24547.124

0.878

0.364

Final LCRI Economic Analysis

5-81

October 2024


-------
SafeWater
Model
Period

Calander
Year

Low Scenario

High Scenario

Emission Changes (metric tons)

Emission Changes (metric tons)

C02

ch4

N20

C02

CH4

NzO

13

2036

24715.450

0.884

0.367

24539.064

0.878

0.364

14

2037

750.996

0.027

0.011

650.899

0.024

0.010

15

2038

731.454

0.027

0.011

614.957

0.022

0.009

16

2039

745.727

0.027

0.011

647.835

0.024

0.010

17

2040

758.176

0.028

0.011

659.921

0.024

0.010

18

2041

752.179

0.027

0.011

664.430

0.024

0.010

19

2042

756.668

0.027

0.011

659.966

0.024

0.010

20

2043

760.771

0.027

0.011

670.945

0.024

0.010

21

2044

59.186

0.002

0.001

(32.525)

(0.001)

(0.001)

22

2045

(16.962)

(0.000)

(0.000)

(104.318)

(0.003)

(0.002)

23

2046

(303.706)

(0.011)

(0.005)

(402.862)

(0.014)

(0.006)

24

2047

(307.444)

(0.011)

(0.005)

(407.982)

(0.014)

(0.006)

25

2048

(301.742)

(0.011)

(0.004)

(409.150)

(0.014)

(0.006)

26

2049

(301.891)

(0.011)

(0.004)

(410.415)

(0.014)

(0.006)

27

2050

(305.466)

(0.011)

(0.005)

(416.248)

(0.015)

(0.006)

28

2051

(303.461)

(0.011)

(0.005)

(422.171)

(0.015)

(0.006)

29

2052

(312.687)

(0.011)

(0.005)

(423.160)

(0.015)

(0.006)

30

2053

(305.008)

(0.011)

(0.005)

(405.398)

(0.014)

(0.006)

31

2054

(307.399)

(0.011)

(0.005)

(397.235)

(0.014)

(0.006)

32

2055

(307.624)

(0.011)

(0.005)

(386.657)

(0.014)

(0.006)

33

2056

(304.927)

(0.011)

(0.005)

(380.137)

(0.013)

(0.006)

34

2057

(300.724)

(0.011)

(0.004)

(380.607)

(0.013)

(0.006)

35

2058

(304.884)

(0.011)

(0.005)

(374.672)

(0.013)

(0.006)

Total Over 35 Years

241,503.509

8.643

3.582

230,668.909

8.253

3.421

5.9.3 Valuation of GHG Emissions

To monetize the climate disbenefits of the final LCRI, the EPA takes the estimated incremental annual
emissions from Exhibit 5-43and applies the EPA's Social Cost of Greenhouse Gas values for all three
tracked gases, C02, CH4, and N20.

The SC-GHG is the monetary value of the net harm to society from emitting a metric ton of GHGs into
the atmosphere in a given year, or the benefit of avoiding that increase. In principle, SC-GHG is a
comprehensive metric that includes the value of all future climate change impacts (both negative and
positive), including changes in net agricultural productivity, human health effects, property damage
from increased flood risk, changes in the frequency and severity of natural disasters, disruption of
energy systems, risk of conflict, environmental migration, and the value of ecosystem services. The SC-
GHG, therefore, reflects the societal value of reducing GHG emissions by one metric ton and is the
theoretically appropriate value to use in conducting benefit-cost analyses of policies that affect GHG
emissions. In practice, data and modeling limitations restrain the ability of SC-GHG estimates to include

Final LCRI Economic Analysis

5-82

October 2024


-------
all physical, ecological, and economic impacts of climate change, implicitly assigning a value of zero to
the omitted climate damages. The estimates are, therefore, a partial accounting of climate change
impacts and likely underestimate the marginal benefits of abatement (and marginal damages from
emissions).

Since 2008, the EPA has used estimates of the social cost of various greenhouse gases (i.e., social cost of
carbon (SC-C02), social cost of methane (SC-CH4), and social cost of nitrous oxide (SC-N20)), collectively
referred to as the "social cost of greenhouse gases" (SC-GHG), in analyses of actions that affect GHG
emissions. The values used by the EPA from 2009 to 2016, and since 2021 have been consistent with
those developed and recommended by the Interagency Working Group (IWG) on the SC-GHG; and the
values used from 2017 to 2020 were consistent with those required by E.O. 13783, which disbanded the
IWG. During 2015-2017, the National Academies conducted a comprehensive review of the SC-C02 and
issued a final report in 2017 recommending specific criteria for future updates to the SC-C02 estimates
(which are also applicable to SC-CH4 and SC-N20), a modeling framework to satisfy the specified
criteria, and both near-term updates and longer-term research needs pertaining to various components
of the estimation process (National Academies, 2017). The IWG was reconstituted in 2021 and E.O.
13990 directed it to develop a comprehensive update of its SC-GHG estimates, recommendations
regarding areas of decision-making to which SC-GHG should be applied, and a standardized review and
updating process to ensure that the recommended estimates continue to be based on the best available
economics and science going forward.

The EPA is a member of the IWG and is participating in the IWG's work under E.O. 13990. While that
process continues, as noted in previous EPA RIAs, the EPA is continuously reviewing developments in the
scientific literature on the SC-GHG, including more robust methodologies for estimating damages from
emissions, and looking for opportunities to further improve SC-GHG estimation going forward. In the
December 2022 RIAforthe Standards of Performance for New, Reconstructed, and Modified Sources
and Emissions Guidelines for Existing Sources: Oil and Natural Gas Sector Climate Review, the agency
included a sensitivity analysis of the climate benefits of the Supplemental Proposal using a new set of
SC-GHG estimates that incorporates recent research addressing recommendations of the National
Academies (2017) in addition to using the interim SC-GHG estimates presented in the Technical Support
Document: Social Cost of Carbon, Methane, and Nitrous Oxide Interim Estimates under Executive Order
13990 (IWG, 2021) that the IWG recommended for use until updated estimates that address the
National Academies' recommendations are available.

The EPA solicited public comment on the sensitivity analysis and the accompanying draft technical
report, EPA Report on the Social Cost of Greenhouse Gases: Estimates Incorporating Recent Scientific
Advances, which explains the methodology underlying the new set of estimates, in the December 2022
Supplemental Proposal (USEPA, 2023h). The public comments and the response to comments document
can be found in the docket for the Standards of Performance for New, Reconstructed, and Modified
Sources and Emissions Guidelines for Existing Sources: Oil and Natural Gas Sector Climate Review.204

To ensure that the methodological updates adopted in the technical report are consistent with
economic theory and reflect the latest science, the EPA also initiated an external peer review panel to
conduct a high-quality review of the technical report, completed in May 2023. The peer reviewers

204 https://www.regulations.gov/document/EPA-HQ-OAR-2021-0317-4009

Final LCRI Economic Analysis

5-83

October 2024


-------
commended the agency on its development of the draft update, calling it a much needed improvement
in estimating the SC-GHG and a significant step towards addressing the National Academies'
recommendations with defensible modeling choices based on current science. The peer reviewers
provided numerous recommendations for refining the presentation and for future modeling
improvements, especially with respect to climate change impacts and associated damages that are not
currently included in the analysis. Additional discussion of omitted impacts and other updates have been
incorporated in the technical report to address peer reviewer recommendations. Complete information
about the external peer review, including the peer reviewer selection process, the final report with
individual recommendations from peer reviewers, and the EPA's response to each recommendation is
available on the EPA's website.205

For an overview of the methodological updates incorporated into the SC-GHG estimates applied in the
EA for the final LCRI, see Section 3.2 of the RIA for the Standards of Performance for New,

Reconstructed, and Modified Sources and Emissions Guidelines for Existing Sources: Oil and Natural Gas
Sector Climate Review (USEPA, 2023g). A more detailed explanation of each input and the modeling
process is provided in the technical report, Supplementary Material for the RIA: EPA Report on the Social
Cost of Greenhouse Gases: Estimates Incorporating Recent Scientific Advances (USEPA, 2023h), included
in the docket for the Standards of Performance for New, Reconstructed, and Modified Sources and
Emissions Guidelines for Existing Sources: Oil and Natural Gas Sector Climate Review, and included in
the docket for this action.

Exhibit 5-44 summarizes the resulting averaged certainty-equivalent SC-GHG estimates under each near-
term Ramsey discount rate that are used to estimate the climate disbenefits of the changes in GHG
emissions expected to result from the final rule. These estimates are reported in 2022 dollars but are
otherwise identical to those presented in USEPA (2023h). The SC-GHG values increase over time within
the models — i.e., the societal harm from one metric ton emitted in 2030 is higher than the harm
caused by one metric ton emitted in 2025 — because future emissions produce larger incremental
damages as physical and economic systems become more stressed in response to greater climatic
change, and because GDP is growing overtime and many damage categories are modeled as
proportional to GDP. The full results generated from the updated methodology for emissions years 2020
through 2080 are provided in US EPA (2023h).

205 https://www.epa.gov/environmental-economics/scghg-tsd-peer-review

Final LCRI Economic Analysis

5-84

October 2024


-------
Exhibit 5-44: Estimates of the Social Cost of C02, CH4, and N20, 2024-2058 (in 2022 USD)

Gas

C02

C02

COz

ch4

ch4

ch4

NzO

NzO

NzO

Near-term
Ramsey
Discount
Rate

2.50%

2.00%

1.50%

2.50%

2.00%

1.50%

2.50%

2.00%

1.50%

2024

143

233

399

1,706

2,183

2,967

43,687

66,096

104,812

2025

146

237

403

1,780

2,267

3,064

44,748

67,468

106,587

2026

149

241

409

1,855

2,352

3,160

45,810

68,840

108,362

2027

152

245

414

1,930

2,436

3,258

46,871

70,212

110,137

2028

156

250

420

2,005

2,521

3,354

47,932

71,585

111,911

2029

158

253

425

2,079

2,605

3,451

48,993

72,956

113,686

2030

161

257

430

2,154

2,690

3,548

50,055

74,329

115,461

2031

165

262

435

2,241

2,788

3,661

51,153

75,728

117,241

2032

168

265

441

2,329

2,886

3,774

52,251

77,127

119,020

2033

171

270

446

2,415

2,985

3,886

53,349

78,527

120,800

2034

174

274

451

2,502

3,083

3,999

54,448

79,925

122,579

2035

177

278

457

2,589

3,182

4,112

55,546

81,324

124,359

2036

180

282

461

2,677

3,279

4,225

56,644

82,724

126,137

2037

184

287

467

2,763

3,378

4,338

57,741

84,123

127,917

2038

187

290

472

2,850

3,476

4,450

58,839

85,522

129,696

2039

190

294

477

2,938

3,575

4,563

59,938

86,922

131,476

2040

194

299

483

3,025

3,672

4,676

61,036

88,321

133,255

2041

197

303

488

3,119

3,778

4,797

62,279

89,900

135,244

2042

200

308

494

3,214

3,886

4,919

63,524

91,478

137,234

2043

204

312

499

3,308

3,992

5,040

64,768

93,057

139,222

2044

208

317

505

3,403

4,098

5,161

66,012

94,636

141,211

2045

212

321

510

3,497

4,205

5,282

67,257

96,215

143,201

2046

215

326

517

3,592

4,311

5,404

68,500

97,793

145,190

2047

218

331

523

3,686

4,418

5,525

69,745

99,372

147,178

2048

223

336

528

3,782

4,524

5,646

70,989

100,951

149,168

2049

226

340

534

3,876

4,630

5,766

72,233

102,530

151,157

2050

229

345

540

3,971

4,737

5,889

73,478

104,108

153,145

Final LCRI Economic Analysis

5-85

October 2024


-------
Gas

C02

C02

C02

ch4

ch4

ch4

N20

N20

N20

Near-term
Ramsey
Discount
Rate

2.50%

2.00%

1.50%

2.50%

2.00%

1.50%

2.50%

2.00%

1.50%

2051

233

349

545

4,057

4,836

6,004

74,640

105,589

155,026

2052

236

353

550

4,143

4,936

6,119

75,803

107,070

156,906

2053

240

357

555

4,231

5,034

6,234

76,965

108,551

158,786

2054

243

362

560

4,317

5,134

6,350

78,128

110,032

160,666

2055

246

365

565

4,403

5,234

6,464

79,290

111,515

162,546

2056

249

369

571

4,490

5,332

6,579

80,453

112,996

164,426

2057

252

374

575

4,576

5,432

6,695

81,615

114,477

166,306

2058

255

378

581

4,663

5,531

6,810

82,778

115,958

168,187

Data from: EPA Report on the Social Cost of Greenhouse Gases: Estimates Incorporating Recent Scientific
Advances (https://www.epa.gov/system/files/documents/2023-12/epa_scghg_2023_report_final.pdf).

Note: The EPA used the GDP Price Deflator to adjust the 2020 Social Cost of GHG number provided in the report to
2022 dollars.

The methodological updates described in USEPA (2023h) represent a major step forward in bringing SC-
GHG estimation closer to the frontier of climate science and economics and address many of the
National Academies' (2017) near-term recommendations. Nevertheless, the resulting SC-GHG estimates,
including the SC-C02 estimates presented in Exhibit 5-44, still have several limitations, as would be
expected for any modeling exercise that covers such a broad scope of scientific and economic issues
across a complex global landscape. There are still many categories of climate impacts and associated
damages that are only partially or not reflected yet in these estimates and sources of uncertainty that
have not been fully characterized due to data and modeling limitations. Please see Section 3.2 of USEPA
(2023h) for further discussion.

Exhibit 5-45 and Exhibit 5-46 present the monetized climate disbenefits from the GHG emissions
associated with both the operation of CCT and SLR under the final LCRI low and high scenarios. The EPA
multiplied the projected C02, CH4, and N20 emissions each year (shown in Exhibit 5-43) by the social cost
of C02, CH4, and N20 estimates for that year (from Exhibit 5-44) and annualized these results over the
35-year period of analysis.206 Monetized climate effects are presented under a 1.5 percent, 2 percent,

206 Consistent with the approach taken in EPA regulatory analyses from 2009 through 2016 and since 2021, the SC-
GHG estimates used in this analysis reflect a global measure of climate damages from GHG emissions. As discussed
at length in USEPA (2023h), because of the distinctive global nature of climate change in which GHG emissions
contribute to damages around the world regardless of where they are emitted, the assessment of global net
damages of GHG emissions allows the EPA to fully disclose and contextualize the net climate disbenefits of GHG
emission increases expected from this final rule. Some modeling frameworks can also provide a partial
characterization of U.S.-specific damages. For example, the Framework for Evaluating Damages and Impacts
(FrEDI) model reflects the availability of U.S.-specific data and research on climate change effects
(Hartin et al. 2023, EPA 2021). Applying U.S.-specific partial SC-GHG estimates derived from FrEDI to the GHG

Final LCRI Economic Analysis

5-86

October 2024


-------
and 2.5 percent near-term Ramsey discount rate, consistent with the EPA's updated SC-GHG
estimates.207 The EPA estimates climate disbenefits associated with this rule ranging from $3.3 million
dollars per year under the LCRI low scenario (under a 1.5 percent near-term Ramsey discount rate) to
$1.3 million dollars per year under the LCRI high scenario (under a 2.5 percent near-term Ramsey
discount rate).

Exhibit 5-45: Climate Disbenefits of the Final LCRI Low Scenario (millions of 2022 USD)



Near-term Ramsey Discount Rate

2.5%

2%

1.5%

Present and Annualized Values of C02 Emission Changes (millions, 2022$)

Present Value in 2022 (2022$)

31.70

52.72

91.81

Annualized Value (35 Years, 2022$)

1.37

2.11

3.39

Present and Annualized Values of CH4 Emission Changes (millions, 2022$)

Present Value in 2022 (2022$)

0.02

0.02

0.03

Annualized Value (35 Years, 2022$)

0.00

0.00

0.00

Present and Annualized Values of N20 Emission Changes (millions, 2022$)

Present Value in 2022 (2022$)

0.15

0.23

0.37

Annualized Value (35 Years, 2022$)

0.01

0.01

0.01

Total Present and Annualized Values of all GHG Emission Changes (C02, CH4, and N20) (millions,
2022$)

Present Value in 2022 (2022$)

31.86

52.96

92.20

Annualized Value (35 Years, 2022$)

1.38

2.12

3.41

Exhibit 5-46: Climate Disbenefits of the Final LCRI High Scenario (millions of 2022 USD)



Near-term Ramsey Discount Rate

2.5%

2.0%

1.5%

Present and Annualized Values of C02 Emission Changes (millions, 2022$)

Present Value in 2022 (2022$)

30.28

50.36

87.70

Annualized Value (35 Years, 2022$)

1.31

2.01

3.24

Present and Annualized Values of CH4 Emission Changes (millions, 2022$)

Present Value in 2022 (2022$)

0.01

0.02

0.03

Annualized Value (35 Years, 2022$)

0.00

0.00

0.00

Present and Annualized Values of N20 Emission Changes (millions, 2022$)

emission increases expected under the final rule would yield an annualized value of climate disbenefits of $8 to
$8.4 million per year (under a 2 percent near-term Ramsey discount rate).

207 As described in USEPA (2023h), the SC-GHG estimates rely on a dynamic discounting approach that provides
internal consistency within the modeling and a more complete accounting of uncertainty consistent with economic
theory and the National Academies' (2017) recommendation to employ a more structural, Ramsey-like approach
to discounting that explicitly recognizes the relationship between economic growth and discounting uncertainty.
This approach is also consistent with the National Academies' (2017) recommendation to use three sets of Ramsey
parameters that reflect a range of near-term certainty-equivalent discount rates and are consistent with theory
and empirical evidence on consumption rate uncertainty. See USEPA (2023h) for a more detailed discussion of the
entire discounting module and methodology used to value risk aversion in the SC-GHG estimates.

Final LCRI Economic Analysis

5-87

October 2024


-------
Present Value in 2022 (2022$)

0.14

0.22

0.35

Annualized Value (35 Years, 2022$)

0.01

0.01

0.01

Total Present and Annualized Values of all GHG Emission Changes (C02, CH4, and N20) (millions,
2022$)

Present Value in 2022 (2022$)

30.44

50.60

88.08

Annualized Value (35 Years, 2022$)

1.32

2.02

3.25

Exhibit 5-45 and Exhibit 5-46 also show that estimated annualized climate disbenefits range from $2.1
million under the LCRI low scenario to $2.0 million under the LCRI high scenario when discounted at a 2
percent near-term Ramsey discount rate, in 2022 dollars. These disbenefits constitute less than 0.02-
0.01 percent of the monetized benefits of the rule, at a 2 percent near-term discount rate, under the
low and high scenarios, respectively.

Note that the EPA did not quantify the potential emissions changes associated with the production and
delivery of CCT chemicals, and the construction required for the installation of CCT technology, and the
production and transport of copper and plastic replacement piping and plumbing components. The EPA
recognizes that many activities directly and indirectly associated with drinking water treatment produce
GHG emissions; however, the agency determined that it could not accurately quantify all the potential
factors that could increase and decrease greenhouse gas emissions that are not solely attributable to
the direct onsite CCT operations and SLR field operations directly required by the rule. The EPA also
notes that this analysis uses the 2021 LCRR as a baseline in order to calculate the incremental GHG
emissions.

5.10 References

Abt Associates. 2022a. Developing a Concentration-Response Function Relating Childhood Lead
Exposures and ADHD. Prepared for: National Center for Environmental Economics, Office of Policy, U.S.
Environmental Protection Agency.

Abt Associates. 2022b. Assessment of the Literature on the Dose-Response Relationship between Lead
Exposure in Expectant Mothers and Lower Birth Weight in Newborns. Prepared for: National Center for
Environmental Economics, Office of Policy, U.S. Environmental Protection Agency.

Abt Associates. 2022c. Development of Medical Cost Estimates for Adverse Birth Outcomes. Prepared
for: National Center for Environmental Economics, Office of Policy, U.S. Environmental Protection
Agency.

Abt Associates. 2023. Selection of Concentration-Response Functions between Lead Exposure and
Adverse Health Outcomes for Use in Benefits Analysis: Cardiovascular-Disease Related Mortality.
Prepared for: National Center for Environmental Economics, Office of Policy, U.S. Environmental
Protection Agency.

Aguiar, A., P.A. Eubig, and S.L. Schantz. 2010. Attention deficit/hyperactivity disorder: a focused
overview for children's environmental health researchers. Environmental Health Perspectives 118(12):
1646-1653.

Final LCRI Economic Analysis

5-88

October 2024


-------
American Psychiatric Association. 2013. Diagnostic and Statistical Manual of Mental Disorders (5th ed.).
Washington, D.C.

AwwaRF and DVGW-Technologiezentrum Wasser. 1996. Internal Corrosion of Water Distribution
Systems. 2nd edition. AwwaRF Order 90508. Project #725. AWWA Research Foundation (now Water
Research Foundation) and AWWA. Denver, CO.

Bates, D.M. 2010. Ime4: Mixed effects modeling with R. Springer.

Bates, D., M. Machler, B. Bolker, and S. Walker. 2015. Fitting linear mixed effects models using Ime4.
Journal of Statistical Software 67(1): 1-48. doi.org/10.18637/iss.v067. iOl.

Barbaresi, W. J., R. C. Colligan, et al. (2013). "Mortality, ADHD, and psychosocial adversity in adults with
childhood ADHD: a prospective study." Pediatrics 131(4): 637-644.

Bolker, B.M., M.E. Brooks, C.J. Clark, S.W. Geange, J.R. Poulsen, M.H.H. Stevens, and J.S.S. White. 2009.
Generalized linear mixed models: A practical guide for ecology and evolution. Trends in Ecology and
Evolution 24(3): 127-135.

Brown, L., M. Lynch, A. Belova, R. Klein, and A. Chiger. 2020. Developing a Health Impact Model for Adult
Lead Exposure and Cardiovascular Disease Mortality Environmental Health Perspectives 128:9 CID:
097005. https://doi.org/10.1289/EHP6552

Budtz-J0rgensen, E., D. Bellinger, B. Lanphear, and P. Grandjean. 2013. An international pooled analysis
for obtaining a benchmark dose for environmental lead exposure in children. Risk Analysis 33(3): 450-
461. doi:10.1111/i. 1539-6924.2012.01882.x.

Camara, E., K.R. Montreuil, A.K. Knowles, and G.A. Gagnon. 2013. Role of the water main in lead service
line replacement: A utility case study. Journal AWWA 108(8): E423-E431.
doi:10.5942/jawwa.2013.105.0102.

Campbell, A. 2016. Unpublished raw data.

Centers for Disease Control and Prevention (CDC). 2014. National Center for Health Statistics (NCHS).
National Health and Nutrition Examination Survey Questionnaire. Hyattsville, MD: U.S. Department of
Health and Human Services, Centers for Disease Control and Prevention. Available at
https://www.cdc.gov/nchs/nhanes.

CDC. 2015. National Center for Health Statistics (NCHS). Underlying Cause of Death 1999-2014 on CDC
WONDER Online Database, released 2015. Retrieved from: http://wonder.cdc.gov/ucd-icdlO.html.

CDC. 2016. National Center for Health Statistics (NCHS). National Health and Nutrition Examination
Survey Data. Hyattsville, MD: U.S. Department of Health and Human Services, Centers for Disease
Control and Prevention, Available at https://www.cdc.gov/nchs/nhanes.

CDC. 2017. National Vital Statistics Reports. Births: Final Data for 2015.

Chowdhury, R., A. Ramond, L. M. O'Keeffe, S. Shahzad, S. K. Kunutsor, T. Muka,. . . E. Di Angelantonio.
2018. Environmental toxic metal contaminants and risk of cardiovascular disease: systematic review and
meta-analysis. BMJ, 362: k3310. doi:10.1136/bmj.k3310

Final LCRI Economic Analysis

5-89

October 2024


-------
Commons, C. 2011. Effect of Partial Lead Service Line Replacement on Total Lead at the Tap. Rhode
Island Department of Health, Office of Drinking Water Quality.

Commons, C. 2014. A four-year follow-up on total lead at the tap following partial service line
replacement in Cranston, Rl. Addendum to a paper originally published in June 2012 entitled: Effect of
partial lead service line replacement on total lead at the tap in Cranston, Rhode Island. Journal New
England Water Works Association 187-188.

Craik, S. 2016. Unpublished raw data.

Crump, K.S., C. Van Landingham, T.S. Bowers, D. Cahoy, and J.K. Chandalia. 2013. A statistical
reevaluation of the data used in the Lanphear et al. (2005) pooled-analysis that related low levels of
blood lead to intellectual deficits in children. Critical Reviews in Toxicology 43(9): 785-799.
Doi:10.3109/10408444.2013.832726.

Danielson, M. L., R. Bitsko, R. M. Ghandour, J. R. Holbrook, M. D. Kogan, and S. J. Blumberg. 2018.
Prevalence of Parent-Reported ADHD Diagnosis and Associated Treatment Among U.S. Children and
Adolescents, 2016. Journal of Clinical Child & Adolescent Psychology. 47:2, 199-212, DOI:
10.1080/15374416.2017.1417860

DC Water. 2016. Unpublished raw data.

DC Water. 2017. Personal communication with Maureen Schmelling. May 2017.

Del Toral, M.A., A. Porter, and M.R. Schock. 2013. Detection and evaluation of elevated lead release
from service lines: Afield study. Environmental Science and Technology 47: 9300-9307.

Del Toral, M.A. 2016. Unpublished raw data. Flint sequential water sampling data for 2016.

Deshommes, E., A. Bannier, L. Laroche, S. Nour, and M. Prevost. 2016. Monitoring-based framework to
detect and manage lead water service lines. Journal AWWA 108(11): E555-E570.

Desmarais, M.C., B. Trueman, P. Wilson, D. Huggins, J. Swertfeger, E. Deshommes, and M. Prevost. 2015.
Impact of Partial Lead Service Line Replacements on Water Quality: Lead Profiling Sampling Results in 6
North-American Utilities. Presented at the AWWA Water Quality Technology Conference & Expo, Salt
Lake City, UT, November 15-19, 2015.

Doshi, J. A., P. Hodgkins, J. Kahle, V. Sikirica, M.J. Cangelosi, J. Setyawan, M.H. Erder, and P.J. Neumann.
2012. Economic impact of childhood and adult attention-deficit/hyperactivity disorder in the United
States. Journal of the American Academy of Child & Adolescent Psychiatry, 51(10), 990-1002.

EPCOR Water Services. 2008. Lead at Customers Taps. Results of the 2007 Sampling Program. Draft
Report.

Estes-Smargiassi, S., S.J.J. Steinkrauss, A. Sandvig, and T. Young. 2006. Impacts of Lead Service Line
Replacement on Lead Levels at the Tap. Presented at the AWWA Annual Conference, San Antonio, TX,
June 11-15, 2006.

Folkman, S. 2018. Water Main Break Rates In the USA and Canada: A Comprehensive Study. Utah State
University, Buried Structures Laboratory.

Final LCRI Economic Analysis

5-90

October 2024


-------
Froehlich, T. E., B.P. Lanphear, P. Auinger, R. Hornung, J.N. Epstein, J. Braun, & R.S. Kahn. (2009).
Association of tobacco and lead exposures with attention-deficit/hyperactivity disorder. Pediatrics,
124(6), el054-1063. https://doi.org/10.1542/peds.2009-0738

Glen, GV. Zartarian, L. Smith, and J. Xue. 2012. The Stochastic Human Exposure and Dose Simulation
Model for Multimedia, Multipathway Chemicals (SHEDS-Multimedia): Residential Module. SHEDS-
Residential version 4 Technical Manual. Available at www.epa.gov/sites/production/files/2015-
02/documents/shedsresidential techmanual 2012.pdf.

Hartin, et al. 2023. "Advancing the estimation of future climate impacts within the United States." Earth
System Dynamics 14.5 (2023): 1015-1037.

Hayes, C.R., T.N. Croft, A. Campbell, I.P. Douglas, P. Gadoury, and M.R. Schock. 2014. Computational
modelling techniques in the optimization of corrosion control for reducing lead in Canadian drinking
water. Water Quality Research Journal of Canada 49(1): 82.

Henning, C., C. Guider, D. Reilly, G. Glen, J. Cohen, B. Groza, and M. Myer. 2024. Comparison of empirical
and mechanistic blood lead models for children. Indoor Environments. 1(3): 100022.
https://doi.Org/10.1016/j.indenv.2024.100022.

Hubbard, H., H. Ozkaynak, G. Glen, J. Cohen, K. Thomas, L. Phillips, and N. Tulve. Model-based
predictions of soil and dust ingestion rates for U.S. adults using the stochastic human exposure and dose
simulation soil and dust model. Science of the Total Environment. 2022 Nov 10;846:157501. doi:
10.1016/j.scitotenv.2022.157501. Epub 2022 Jul 21. PMID: 35870603.

Interagency Working Group (IWG). 2021. Technical Support Document: Social Cost of Carbon, Methane,
and Nitrous Oxide Interim Estimates under Executive Order 13990. Retrieved from:
https://www.whitehouse.gov/wp-

content/uploads/2021/02/TechnicalSupportDocument_SocialCostofCarbonMethaneNitrousOxide.pdf

Yuelong, Ji, X. Hong, G. Wang, N. Chatterjee, A.W. Riley, L-C. Lee, et al. A Prospective Birth Cohort Study
on Early Childhood Lead Levels and Attention Deficit Hyperactivity Disorder: New Insight on Sex
Differences. The Journal of Pediatrics. 2018;199:124-31.

Jusko, T.A., C.R. Henderson, B.P. Lanphear, D.A. Cory-Slechta, P.J. Parsons, and R.L. Canfield. 2008. Blood
lead concentrations < 10 microg/dL and child intelligence at 6 years of age. Environmental Health
Perspectives 116(2): 243-248. doi:10.1289/ehp,10424.

Kirrane, E.F. and M.M. Patel. 2014. Memorandum from EPA NCEAto ISA for Lead Docket: Identification
and Consideration of Errors in Lanphear et al. (2005) "Low-Level Environmental Lead Exposure and
Children's Intellectual Function: An International Pooled Analysis."

Lanphear, B.P., R. Hornung, J. Khoury, K. Yolton, P.A. Baghurst, D.C. Bellinger, and R. Roberts. 2005.
Low-level environmental lead exposure and children's intellectual function: An international pooled
analysis. Environmental Health Perspectives 113(7): 894-899. doi:10.1289/ehp.7688.

Lanphear, B.P., R. Hornung, J. Khoury, K. Yolton, P.A. Baghurst, D.C. Bellinger, R. Roberts. 2019. Erratum:
Low-level environmental lead exposure and children's intellectual function: An international pooled
analysis. Environmental Health Perspectives 127(9): 099001.

Final LCRI Economic Analysis

5-91

October 2024


-------
Leggett, R.W. 1993. An age-specific kinetic model of lead metabolism in humans. Environmental Health
Perspectives 101: 598-616.

Levin, R., and J. Schwartz. 2023. A better cost:benefit analysis yields better and fairer results: EPA's lead
and copper rule revision. Environmental Research 229: 115738.
https://doi.Org/10.1016/j.envres.2023.115738

Lin, D., R. Lutter, and C.J. Ruhm. 2018. Cognitive performance and labour market outcomes. Labour
Economics 51:121-135.

LSLR Collaborative, n.d. Getting Started on an LSL Inventory. Retrieved November 11, 2023, from
https://www.lslr-collaborative.org/preparing-an-

inventorv.html#:~:text=Lead%20pipe%20was%20tvpically%20installed.as%203%20inches%20in%20dia
meter.

MDB, Inc. (2019) "Selection of Concentration-Response Functions between Lead Exposure and Adverse
Health Outcomes for Use in Benefits Analysis: Cardiovascular-Disease Related Mortality" Peer Review
Combined Documents. Available at

https://cfpub.epa.gov/si/si public record report.cfm?Lab=NCEE&dirEntrylD=342855

MDB, Inc. (2022) Memorandum to NCEE. Peer Review of "Development of Medical Cost Estimates for
Adverse Birth Outcomes"

Menke, A., Muntner, P., Batuman, V., Silbergeld, E. K., and Guallar, E. 2006. Blood lead below 0.48
micromol/L (10 microg/dL) and mortality among US adults. Circulation 114(13): 1388-1394.
doi:10.1161/circulationaha. 106.628321

Min, M.O., L.T. Singer, H.L. Kirchner, S. Minnes, E. Short, Z. Hussain, and S. Nelson. 2009. Cognitive
development and low-level lead exposure in poly-drug exposed children. Neurotoxicology and
Teratology 31(4): 225-231. doi:10.1016/j.ntt.2009.03.002.

Muylwyk, Q. 2016. Unpublished raw data.

National Academies of Sciences, Engineering, and Medicine (National Academies). 2017. Valuing Climate
Damages: Updating Estimation of the Social Cost of Carbon Dioxide. Washington, DC: The National
Academies Press, https://doi.org/10.17226/24651.

National Research Council (NRC). 2000. Copper in Drinking Water. Washington, D.C.: The National
Academies Press.

NSF International. 2019. Certified Product Listings for Lead Reduction. Available at
http://info.nsf.org/Certified/DWTU/listings leadreduction.asp?ProductFunction=0531 Lead+Reduction&
ProductFunction=058| Lead+Reduction&ProductType=&submit2=Search.

National Toxicology Program (NTP). 2012. NTP Monograph: Health Effects of Low-Level Lead. U.S.
Department of Health and Human Services. Office of Health Assessment and Translation. Division of the
National Toxicology Program. Available at

https://ntp.niehs.nih.gov/ntp/ohat/lead/final/monographhealtheffectslowlevellead newissn 508.pdf.

Final LCRI Economic Analysis

5-92

October 2024


-------
O'Brien & Gere. 2015. Lead Service Line Monitoring to Assess Treatment pH Change. O'Brien & Gere
Engineers, Inc.

Ozkaynak, H., J. Xue, V. Zartarian, G. Glen, and L. Smith. 2011. Modeled estimates of soil and dust
ingestion rates for children. Risk Analysis 31(4): 592-608. doi:10.1111/j. 1539-6924.2010.01524x.

Ozkaynak, H., G. Glen, J. Cohen, H. Hubbard, K. Thomas, L. Phillips, and N. Tulve. 2022. Model based
prediction of age-specific soil and dust ingestion rates for children. Journal of Exposure Science &
Environmental Epidemiology 32: 472-480; https://doi.org/10.1038/s41370-021-00406-5

Pinheiro, J., D. Bates, S. DebRoy, D. Sarkar, and R Core Team. 2017. nlme: Linear and Nonlinear Mixed
Effects Models. R package version 3.1-128. Available at http://CRAN.R-proiect.org/package=nlme.

Pliszka, S. 2007. AACAP Work Group on Quality Issues. Practice parameter for the assessment and
treatment of children and adolescents with attention-deficit/hyperactivity disorder. Journal of the
American Academy of Child & Adolescent Psychiatry.Jul;46(7):894-921.

Pocock, S.J., A.G. Shaper, M. Walker, C.J. Wale, B. Clayton, T. Delves, R.F. Lacey, R.F. Packham, and P.
Powell. 1983. Effects of tap water lead, water hardness, alcohol, and cigarettes on blood lead
concentrations. Journal of Epidemiology and Community Health 37(1): 1-7.

R Core Team. 2016. R: A language and environment for statistical computing R Foundation for Statistical
Computing (Version 3.4.1. ed.). Vienna, Austria.

Salkever, D.S. 1995. Updated estimates of earnings benefits from reduced exposure of children to
environmental lead. Environmental Research 70: 1-6. doi:0013-9351/95.

Schock, M. 2016. Unpublished raw data.

Schwartz, J. and D. Otto. 1991. Lead and minor hearing impairment. Archives of Environmental and
Occupational Health 46(5): 300-305. doi:10.1080/00039896.1991.9934391.

Sherlock, J., G. Smart, G.I. Forbes, M.R. Moore, W.J. Patterson, W.N. Richards, and T.S. Wilson. 1982.
Assessment of lead intakes and dose-response for a population in Ayr exposed to a plumbosolvent
water supply. Human & Experimental Toxicology 1(2): 115-122.

Stanek, L.W., J. Xue, C.R. Lay, E.C. Helm, M. Schock, D.A. Lytle, T.F. Speth, and V.G. Zartarian. 2020.
Modeled impacts of drinking water Pb reduction scenarios on children's exposures and blood lead
levels. Environmental Science and Technology 54(15): 9474-9482. doi:10.1021/acs.est.0c00479.

Sibley, M. H., L. E. Arnold, et al. (2022). "Variable patterns of remission from ADHD in the multimodal
treatment study of ADHD." American Journal of Psychiatry 179(2): 142-151.

Swertfeger, J., D.J. Harman, C. Shrive, D.H. Metz, and J. DeMarco. 2006. Water Quality Effects of Partial
Lead Line Replacement. Presented at the AWWA Annual Conference, San Antonio, TX, June 11-15, 2006.

The Cadmus Group, Inc. 2007. Review of the Interim Optimal Corrosion Control Treatment for
Washington, DC. The Cadmus Group, Inc. [OCCT Review 2007],

Final LCRI Economic Analysis

5-93

October 2024


-------
Triantafyllidou, S., M. Schock, M.K. DeSantis, and C. White. 2015. Low contribution of Pb02-coated lead
service lines to water lead contamination at the tap. Environmental Science & Technology 49(6): 3746-
3754.

Tripp, G., Wickens, J.R. Neurobiology of ADHD. Neuropharmacology. 2009 Dec;57(7-8):579-89.

Urbanic, M., D. Lytle, R. Achtemeier, AND A. Paul. Lead Exposure, Service Line Identification, and Other
Sampling Methods - Comparison of two cities with opposite levels of corrosion control. Presented at
2022 AWWA Water Quality Technology Conference Cincinnati, OH, Cincinnati, OH, November 14 -16,
2022.

United States Census Bureau. 2013. 2010 Census of Population and Housing, Summary Population and
Housing Characteristics. CPH-1-1, U.S. Government Printing Office. Washington, DC.
https://www2.census.gov/librarv/publications/2012/dec/cph-l-l.pdf.

United States Department of Housing and Urban Development (HUD). 2011. American Healthy Homes
Survey: Lead and Arsenic Findings. Office of Healthy Homes and Lead Hazard Control.
https://www.hud.gov/sites/documents/AHHS REPORT.PDF.

United States Department of Housing and Urban Development (HUD). 2021. American Healthy Homes
Survey II: Lead Findings. Office of Healthy Homes and Lead Hazard Control.
https://hud.gov/sites/dfiles/HH/documents/AHHS_ll_Lead_Findings_Report_Final_29oct21.pdf

United States Environmental Protection Agency (USEPA). 1994. Guidance Manual for the Integrated
Exposure Uptake Biokinetic Model for Lead in Children. February 1994. Office of Solid Waste and
Emergency Response. EPA/540/R-93/081. Available at
https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockev=2000WN4R.TXT.

USEPA. 2003. Recommendations of the Technical Review Workgroup for Lead for an Approach to
Assessing Risks Associated with Adult Exposures to Lead in Soil. January 2003. Technical Review
Workgroup for Lead. EPA-540-R03-001. Available at https://semspub.epa.gov/work/06/199244.pdf.

USEPA. 2007. Lead Human Exposure and Health Risk Assessments for Selected Case Studies (Draft
Report): Volume I. Human Exposure and Health Risk Assessments - Full Scale. July 2007. Office of Air
Quality Planning and Standards. EPA-452/D-07-001a. Available at
https://www3.epa.gov/ttn/naaqs/standards/pb/data/Pb-RA-Vol-l-073007.pdf.

USEPA. 2011. Exposure Factors Handbook 2011 Edition (Final Report). U.S. Environmental Protection
Agency, Washington, DC, EPA/600/R-09/052F, 2011.

USEPA. 2012. Guidelines for Preparing Economic Analyses. Chapter 7. National Center for Environmental
Economics, Office of Policy Washington, DC: US Environmental Protection Agency; 2010:7-10

USEPA. 2013. Integrated Science Assessment for Lead. United States Environmental Protection Agency,
Office of Research and Development. Research Triangle Park, NC. EPA/600/R-10/075F.

USEPA. 2016. Stochastic Human Exposure and Dose Simulation (SHEDS) to Estimate Human Exposure to
Chemicals. United States Environmental Protection Agency. Available at https://www.epa.gov/chemical-
research/stochastic-human-exposure-and-dose-simulation-sheds-estimate-human-exposure.

Final LCRI Economic Analysis

5-94

October 2024


-------
USEPA. 2017. Exposure Factors Handbook Chapter 5 (Update): Soil and Dust Ingestion. USEPA Office of
Research and Development, Washington, DC, EPA/600/R-17/384F, 2017.
https://www.epa.gov/sites/default/files/2018-01/documents/efh-chapter05_2017.pdf

USEPA. 2019a. Economic Analysis of the Final Rule to Revise the TSCA Dust-Lead Hazard Standards. June

2019.	Office of Pollution Prevention and Toxics.

USEPA. 2019b All-Ages Lead Model (AALM), Version 2.0 (External Review Draft, 2019). U.S.
Environmental Protection Agency, Washington, DC, 2019.

USEPA. 2020a. Economic Analysis Appendices for the Final Lead and Copper Rule Revisions. December

2020.	Office of Water. EPA 816-R-20-008a.

USEPA. 2020b. Science Advisory Board (SAB) Consideration of the Scientific and Technical Basis of EPA's
Proposed Rule Titled National Primary Drinking Water Regulations: Proposed Lead and Copper Rule
Revisions. EPA-SAB-20-007

sab.epa.gov/ords/sab/r/sab apex/sabl00/advisorvactivitvdetail?pl8 id=2567&clear=18&session=4371
171524473

USEPA. 2021. Emission Factors for Greenhouse Gas Inventories. Retrieved from
epa.gov/sites/default/files/2021-04/documents/emission-factors_apr2021.pdf

USEPA. 2023a. Economic Analysis of Updated Soil Lead Guidance for Sites and Facilities Being Addressed
Under CERCLA and RCRA Authorities

USEPA. 2023b. Technologies and Costs for Corrosion Control to Reduce Lead in Drinking Water. August
2023. Office of Water.

USEPA. 2023c. Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2021. Retrieved from
https://www.epa.gov/system/files/documents/2023-04/US-GHG-lnventory-2023-Main-Text.pdf

USEPA. 2023d. Power Sector Modeling, https://www.epa.gov/power-sector-modeling.

USEPA. 2023e. Regulatory Impact Analyses for Air Pollution Regulations.

https://www.epa.gov/economic-and-cost-analvsis-air-pollution-regulations/regulatorv-impact-analvses-
air-pollution.

USEPA. 2023f. Documentation for Post-IRA 2022 Reference Case, https://www.epa.gov/power-sector-
modeling/post-ira-2022-reference-case.

USEPA. 2023g. Regulatory Impact Analysis for Standards of Performance for New, Reconstructed, and
Modified Sources and Emissions Guidelines for Existing Sources: Oil and Natural Gas Sector Climate
Review. Washington, D.C. Retrieved from https://www.epa.gov/system/files/documents/2023-
12/eol2866_oil-and-gas-nsps-eg-climate-review-2060-avl6-ria-20231130.pdf

USEPA. 2023h. Supplementary Material for the Regulatory Impact Analysis for the Final Rulemaking,
"Standards of Performance for New, Reconstructed, and Modified Sources and Emissions Guidelines for
Existing Sources: Oil and Natural Gas Sector Climate Review": EPA Report on the Social Cost of
Greenhouse Gases: Estimates Incorporating Recent Scientific Advances. Washington, D.C. Retrieved
from https://www.epa.gov/system/files/documents/2023-12/epa_scghg_2023_report_final.pdf

Final LCRI Economic Analysis

5-95

October 2024


-------
USEPA. 2024a. Integrated Science Assessment for Lead (Final Report). U.S. Environmental Protection
Agency, Washington, DC. EPA/600/R-23/375.

USEPA. 2024c. Users Guide for the All Ages Lead Model (AALM) version 3.0-Excel User Interface and
Fortran Model Executable. U.S. Environmental Protection Agency, Washington, DC, 2024.

USEPA. 2024c. EPA's 2024 Economic Assessment Updated Soil Lead Guidance for Sites and Facilities
Being Addressed Under CERCLA and RCRA Authorities.

USEPA. 2024d. Economic Analysis for the Final Per- and Polyfluoroalkyl Substances National Primary
Drinking Water Regulation.

USEPA. 2024b. Technical Support Document for the All Ages Lead Model (AALM) version 3.0-
Parameters, Equations, and Evaluations. EPA 600/R-23/346.

United States Food and Drug Administration (FDA). 2017. Analytical Results of the Total Diet Study:
Individual Year Analytical Results for elements 2007-2014. Available at https://www.fda.gov/food/total-
diet-studv/analvtical-results-total-diet-studv.

Versar. 2015. External Peer Review of EPA's Approach for Estimating Exposures and Incremental Health
Effects from Lead Due to Renovation, Repair, and Painting Activities in Public and Commercial Buildings.
Prepared for EPA under contract EP-C-12-045 Task Order 39.

White, P.D., P. Van Leeuwen, B.D. Davis, M. Maddaloni, K.A. Hogan, A.H. Marcus, and R.W. Elias. 1998.
The conceptual structure of the integrated exposure uptake biokinetic model for lead in children.
Environmental Health Perspectives 106(Suppl 6): 1513-1530.

Woodruff, T.J. 2015. Making it real-the environmental burden of disease. What does it take to make
people pay attention to the environment and health? J Clin Endocrinol Metab. 2015 Apr;100(4): 1241-4.
doi: 10.1210/jc.2015-1622. PMID: 25844765; PMCID: PMC6693198.

Xu, G., L. Strathearn, B. Liu, B. Yang, & W. Bao. (2018). Twenty-year trends in diagnosed attention-
deficit/hyperactivity disorder among US children and adolescents, 1997-2016. JAMA network open, 1(4),
el81471-el81471.

Xue, J., V. Zartarian, J. Liu, and A.M. Geller. 2012a. Methyl mercury exposure from fish consumption in
vulnerable racial/ethnic populations: Probabilistic SHEDS-Dietary model analyses using 1999-2006
NHANES and 1990-2002 TDS data. Science of the Total Environment 414: 373-379.
doi:10.1016/j.scitotenv.2011.10.010.

Xue, J., V. Zartarian, S.W. Wang, S.V. Liu, and P. Georgopoulos. 2010. Probabilistic modeling of dietary
arsenic exposure and dose and evaluation with 2003-2004 NHANES data. Environmental Health
Perspectives 118(3): 345-350. doi:10.1289/ehp.0901205.

Xue, J., V. Zartarian, and S. Nako. 2012b. The Stochastic Human Exposure and Dose Simulation Model for
Multimedia, Multipathway Chemicals (SHEDS-Multimedia):Dietary Module. SHEDS-Dietary Version 1
Technical Manual. Available at https://www.epa.gov/sites/production/files/2015-
02/documents/shedsdietarv techmanual 2012.pdf.

Final LCRI Economic Analysis

5-96

October 2024


-------
Xue, J., J. Liu, V. Zartarian, A.M. Geller, and B.D. Schultz. 2014a. Analysis of NHANES measured blood
PCBs in the general US population and application of SHEDS model to identify key exposure factors.
Journal of Exposure Science and Environmental Epidemiology 24: 615-621. doi:1559-0631/14.

Xue, J., V. Zartarian, R. Tornero-Velez, and N. Tulve. 2014b. EPA's SHEDS-multimedia model: Children's
cumulative pyrethroid exposure estimates and evaluation against NHANES biomarker data.
Environmental International 73: 304-311. doi:10.1016/j.envint.2014.08.008.

Zartarian, V., J. Xue, H. Ozkaynak, W. Dang, G. Glen, L. Smith, and C. Stallings. 2006. A probabilistic
arsenic exposure assessment for children who contact Chromated Copper Arsenate (CCA)-treated
playsets and decks, Part 1: Model methodology, variability results, and model evaluation. Risk Analysis
26(2): 515-531. doi:10.1111/j,1539-6924.2006.00747.x.

Zartarian, V., G. Glen, S. Luther, and J. Xue. 2008. Stochastic Human Exposure and Dose Simulation
Model for Multimedia, Multipathway Chemicals. SHEDS-Multimedia Model Version 3 Technical Manual.

Zartarian, V., J. Xue, G. Glen, L. Smith, N. Tulve, and R. Tornero-Velez. 2012. Quantifying children's
aggregate (dietary and residential) exposure and dose to permethrin: Application and evaluation of
EPA's probabilistic SHEDS-Multimedia model. Journal of Exposure Science and Environmental
Epidemiology 22: 267-273. doi:10.1038/jes.2012.12.

Zartarian, V., J. Xue, R. Tornero-Velez, and J. Brown. 2017. Children's Lead Exposure: A multimedia
Modeling Analysis to Guide Public Health Decision-Making. Environmental Health Perspectives. 125(9).
CID 097009. Available at https://doi.org/10.1289/EHP1605.

Zartarian, V., J. Xue, E. Gibb-Snyder, J.J. Frank, R. Tornero-Velez, and L.W. Stanek. 2023. Children's lead
exposure in the U.S.: Application of a national-scale, probabilistic aggregate model with a focus on
residential soil and dust lead (Pb) scenarios. Science of the Total Environment. 905: 167132. ISSN 0048-
9697. https://doi.Org/10.1016/j.scitotenv.2023.167132.

Zhu, M., E.F. Fitzgerald, K.H. Gelberg, S. Lin, and C.M. Druschel. 2010. Maternal low-level lead exposure
and fetal growth. Environmental Health Perspectives 118(10): 1471-1475. doi:10.1289/ehp.0901561

Final LCRI Economic Analysis

5-97

October 2024


-------
6 Comparison of Costs to Benefits

This chapter compares the incremental costs and benefits of the final Lead and Copper Rule
Improvements (LCRI). As a reminder, the incremental cost is the difference between costs that will be
incurred under the final LCRI and the costs that would have been incurred if the 2021 Lead and Copper
Rule Revisions (LCRR) and other State regulations requiring lead service line replacement (LSLR)that go
beyond the 2021 LCRR LSLR requirements (Illinois, Michigan, New Jersey, and Rhode Island) remained in
place with no changes. The baseline also accounts for resent LCRI compliant tap sampling in schools and
child cares, in 17 states and the District of Columbia, which reduces the estimated incremental burden
for both the 2021 LCRR and LCRI tap sampling requirements. For additional information of baseline
characterization see Chapter 3. In Section 6.1, the United States Environmental Protection Agency (the
EPA) summarizes the incremental costs that were discussed in detail in Chapter 4. In Section 6.2, the
EPA summarizes the incremental benefits that were presented in Chapter 5. Finally, in Section 6.3 the
EPA compares the incremental costs and benefits.

6.1 Summary of the Incremental Costs of the Final LCRI
6.1.1 Monetized Incremental Costs

Exhibit 6-1 provides the estimated incremental costs of the final LCRI, for both the low and high
scenarios, at a 2 percent discount rate in millions of 2022 dollars.

Final LCRI Economic Analysis

6-1

October 2024


-------
Exhibit 6-1: Estimated National Annualized Monetized Incremental Costs of the Final LCRI at 2

Percent Discount Rate (millions of 2022 USD)





Low Estimate





High Estimate





Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS Annual Costs













Sampling

$134.0

$166.0

$32.0

$143.6

$176.2

$32.6

PWS SLR*

$84.6

$1,259.0

$1,174.4

$124.5

$1,763.9

$1,639.4

Corrosion Control
Technology

$552.0

$591.1

$39.1

$647.8

$692.9

$45.1

Point-of Use Installation
and Maintenance

$2.4

1

lo
-oo-

$2.7

$5.9

$9.6

$3.7

Public Education and
Outreach

$69.6

$267.3

$197.7

$72.1

$302.2

$230.1

Rule Implementation and
Administration

1

o

-oo-

$3.4

$3.3

$0.2

$3.4

$3.2

Total Annual PWS Costs

$842.7

$2,291.9

$1,449.2

$994.1

$2,948.2

$1,954.1

Household SLR Costs**

r—1

00

$0.0

T—1

00
-c/>

1

$26.4

$0.0

-$26.4

State Rule Implementation
and Administration

$38.4

$66.1

$27.7

$41.8

$67.6

$25.8

Wastewater Treatment
Plant Costs***

$3.0

$3.0

$0.0

$4.8

1

LO
-oo-

$0.3

Total Annual Rule Costs

$892.2

$2,361.0

$1,468.8

$1,067.1

$3,020.9

$1,953.8

Acronyms: LCRI = Lead and Copper Rule Improvements; SLR = lead service line replacement; PWS = public water
system; USD = United States dollars.

Notes: Previous Baseline costs are projected over the 35-year period of analysis and are affected by the EPA's

assumptions on three uncertain variables which vary between the low and high cost scenarios.

*Service line replacement includes full and partial lead service lines and galvanized requiring replacement service

lines.

**The EPA in the 2021 LCRR economic analysis (USEPA, 2020a) assumed that the cost of customer-side service line
replacements made under the goal-based replacement requirement would be paid for by households. The agency
also assumed that system-side service line replacements under the goal-based replacement requirement and all
service line replacements (both customer-side and systems-side) would be paid by the PWS under the 3 percent
mandatory replacement requirement. The EPA made these modeling assumptions based on the different levels of
regulatory responsibility systems faced operating under a goal-based replacement requirement versus a
mandatory replacement requirement. While systems would not be subject to a potential violation for not meeting
the replacement target under the goal-based replacement requirement, under the 3 percent mandatory
replacement requirement the possibility of a violation could motivate more systems to meet the replacement
target even if they had to adopt customer incentive programs that would shift the cost of replacing customer-side
service lines from customers to the system. To be consistent with these 2021 LCRR modeling assumptions, under
the final LCRI, the EPA assumed that mandatory replacement costs would fall only on systems. Therefore, the
negative incremental values reported for the "Household SLR Costs" category do not represent a net cost savings
to households. They represent an assumed shift of the estimated service line replacement costs from households
to systems. The EPA has insufficient information to estimate the actual service line replacement cost sharing
relationship between customers and systems at the national level of analysis.

Final LCRI Economic Analysis

6-2

October 2024


-------
***Due to many water systems operating both the wastewater and drinking water systems, the EPA is evaluating
the costs of additional phosphate usage for informational purposes. These costs are not "likely to occur solely as a
result of compliance" with the final LCRI, and therefore are not costs considered as part of the Health Risk
Reduction and Cost Analysis (HRRCA) under the Safe Drinking Water Act (SDWA), section 1412(b)(3)(C)(i)(lll).

6.1.2 Non-monetized and Non-quantified Costs

As discussed in Section 4.5 of Chapter 4, the final LCRI is expected to result in additional phosphate
being added to drinking water to reduce the amount of lead leaching into the water in the distribution
system. Although the downstream ecological impacts are not "likely to occur solely as a result of
compliance" with the final LCRI, and therefore are not costs considered as part of the Health Risk
Reduction and Cost Analysis (HRRCA) under the Safe Drinking Water Act (SDWA), section
1412(b)(3)(C)(i)(lll), the EPA for informational purposes has quantified incremental phosphorus loadings
and outlined potential downstream ecological impacts. The SafeWater LCR model estimated that,
nationwide, the final LCRI may result in post wastewater treatment plant (WWTP) total incremental
phosphorus loads to receiving waterbodies increasing over the period of analysis, under the low and
high scenarios, by a range of 225,000 to 272,000 pounds fifteen years after promulgation, and by a
range of 216,000 to 260,000 pounds at Year 35. At the national level, under the high scenario, this
additional phosphorus loading to waterbodies is relatively small, less than 0.03 percent of the total
phosphorous load deposited annually from all other anthropogenic sources.

However, while the percent increase in phosphorus loadings nationally, due to the LCRI corrosion
control treatment (CCT) requirements, is small in relation to all other sources of phosphorous, it is
possible that the additional phosphorous loadings may result in negative localized impacts, such as
eutrophication, in water bodies with elevated phosphorous levels that do not yet have restrictions on
additional phosphate loadings. Exhibit 6-2 shows the location of WWTPs that discharge into waterbodies
that currently have phosphorous limits in place. There is a significant concentration of these plants in
the Great Lakes region as well as in New England and the mid-Atlantic regions. It is reasonable to
assume that other waterbodies in these regions would be at higher risk of experiencing negative
localized ecological impacts associated with the increases in phosphorous loadings resulting from the
LCRI CCT requirements.

Final LCRI Economic Analysis

6-3

October 2024


-------
Exhibit 6-2: Wastewater Treatment Plants with Phosphorous Limits in 2024





•*

V •
& *

• źoo^0#%o 0,VS ź° «. °0<*° .

A * < . - •"

'3% •<.

• • p **$?

• *• •• • °

	 o^° * ;.vs#. 4

Bi»on fifXio*	o

ii

Mmiiw Njuju '•

C>l(UUr

Du'-ingo Mtx.co Clu

Š apersntMive. nan

Source: USEPA. 2024b. Water Pollution Search, https://echo.epa.gov/trends/loading-tool/water-pollution-search

The EPA also notes that there exist unquantified costs associated with service line replacement (SLR).
Costs associated with the disruption of normal traffic patterns in communities implementing SLR
programs are not accounted for in the monetized cost estimates of the rule. This impact to traffic could
be significant in localized areas where lead, galvanized requiring replacement (GRR), and unknown
service lines are co-located with high traffic roads. During SLR worksite activities and characteristics
have the potential to increase car and pedestrian accidents. Also given the necessity to shut off water
service to buildings and residences during SLR the probability of fire damage and negative
health/sanitation impacts may increase. Given that SLR takes a relatively small amount of time (4 hours
on average), the low probability of accidents and fire, the advance notice provided to building
occupants, and alternative local sources of water available in emergencies {e.g., fire hydrants), it is
unlikely that these unquantified cost are nationally significant.

6.2 Summary of the Incremental Benefits of the Final LCRI
6.2.1 Monetized Incremental Benefits

Exhibit 6-3 shows the estimated incremental monetized benefits of the final LCRI at a 2 percent discount
rate under the low and high benefit scenarios. The benefit values are also broken out by the quantified

Final LCRI Economic Analysis

6-4

October 2024


-------
and monetized health endpoints stemming from avoided reductions in intelligence quotient (IQ) and
cases of Attention-Deficit/Hyperactivity Disorder (ADHD) in children, lower birth weights in children of
women of childbearing age, and cases of cardiovascular disease (CVD) premature mortality in adults.

Exhibit 6-3: Estimated National Annualized Monetized Benefits of the Final LCRI at 2 Percent

Discount Rate (millions of 2022 USD)

Low Estimate

High Estimate



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual Child Cognitive
Development Benefits

$1,208.5

$6,831.3

$5,622.8

$3,279.0

$10,963.0

$7,684.0

Annual Low-Birth Weight
Benefits

$1.0

$5.4

$4.4

00

t-H

-oo-

$5.7

$3.9

Annual ADHD Benefits

$33.6

$196.3

$162.7

$179.9

$599.5

$419.6

Annual Adult CVD Premature
Mortality Benefits

$1,750.7

$9,454.3

$7,703.6

$8,174.9

$25,210.0

$17,035.1

Total Annual Benefits

$2,993.8

$16,487.3

$13,493.5

$11,635.6

$36,778.2

$25,142.6

Acronyms: ADHD = Attention-Deficit/Hyperactivity Disorder; CVD = cardiovascular disease; LCRI = Lead and Copper
Rule Improvements; USD = United States dollar.

The EPA is committed to understanding and addressing climate change impacts in carrying out the
agency's mission of protecting human health and the environment. While the EPA is not required by
SDWA 1412(b)(3)(C)(i)(lII) to consider climate disbenefits under the HRRCA, the agency has estimated
the potential climate disbenefits caused by increased greenhouse gas (GHG) emissions associated with
the operation of CCT at drinking water treatment facilities and the use of construction and transport
vehicles in the replacement of lead and GRR service lines. As explained in section VI.A of the preamble,
this disbenefits analysis is presented solely for the purpose of complying with Executive Order 12866.

The EPA analysis found that the climate disbenefits of the final LCRI from C02, CH4, and N20 emissions
associated with increased electricity use in the operation of CCT at drinking water treatment facilities
and the direct combustion of fossil fuels from the use of construction and transport vehicles in the
replacement of lead and GRR service lines resulted in monetized annualized values that range from $2.1
million under the low scenario to $2.0 million under the high scenario discounted at 2 percent, in 2022
dollars. These disbenefit values constitute less than 0.02- 0.01 percent of the monetized benefits of the
rule, at a 2 percent discount rate, under the low and high scenarios, respectively. Note that the EPA did
not quantify the potential emissions changes associated with the production and delivery of CCT
chemicals, the construction required for the installation of CCT technology, and the production and
transport of copper and plastic replacement piping and plumbing components. The EPA recognizes that
many activities directly and indirectly associated with drinking water treatment produce GHG emissions;
however, the agency determined that it could not accurately quantify all the potential factors that could
increase and decrease greenhouse gas emissions that are not solely attributable to the onsite CCT

Final LCRI Economic Analysis

6-5

October 2024


-------
operations and SLR field operations directly required by the rule. The EPA also notes that this analysis
uses the 2021 LCRR as a baseline in order to calculate the incremental GHG emissions.

6.2.2 Non-monetized and Non-quantified Benefits

In addition to the benefits monetized in the final LCRI analysis for reductions in lead exposure, there are
several other benefits that are not quantified. The risk of adverse health effects due to lead exposure
that are expected to decrease as a result of the final LCRI are summarized in Appendix D and are
expected to affect both children and adults. The EPA focused its non-quantified impacts assessment on
the endpoints identified using two comprehensive United States Government documents summarizing
the literature on lead exposure health impacts. These documents are the EPA's Integrated Science
Assessment for Lead (ISA) (USEPA, 2024), and the United States Department of Health and Human
Services' National Toxicology Program (NTP) Monograph on Health Effects of Low-Level Lead (NTP,
2012). Both sources present comprehensive reviews of the literature as of the time of publication on the
risk of adverse health effects associated with lead exposure. The EPA summarized those endpoints to
which either the EPA ISA or the NTP Lead Monograph assigned one of the top two tiers of confidence in
the relationship between lead exposure and the risk of adverse health effects. These endpoints include
cardiovascular morbidity effects, renal effects, reproductive and developmental effects (apart from
ADHD and low birth weight initial hospitalization), immunological effects, neurological effects (apart
from children's IQ), and cancer.

There are a number of final LCRI requirements that reduce lead exposure to both children and adults
that the EPA could not quantify. The final rule will require additional lead public education requirements
that target consumers directly, schools and child care facilities, health agencies, and people living in
homes with lead and GRR service lines. Increased education will lead to additional averting behavior on
the part of the exposed public, resulting in reductions in the negative impacts of lead. The rule will also
require the development of service line inventories that include additional information on lead
connectors and make the location of the lead content service lines publicly accessible. This will give
potentially exposed consumers more information and will provide potential home buyers with this
information as well. Homeowners may request lead service lines (LSL)/GRR service line removal earlier
than a water system might otherwise plan on replacing the line. The benefits of moving these lead and
GRR service line removals forward in time are not quantified in the analysis of the final LCRI. Because of
the lack of granularity in the lead tap water concentration data available to the EPA for the regulatory
analysis, the benefits of small improvements in CCT to individuals residing in homes with lead content
service lines, like those modeled under the Distribution System and Site Assessment requirements, are
not quantified.

The EPA also did not quantify the CCT benefits of reduced lead exposure from lead-containing plumbing
components (not including from lead and/or GRR service lines) to individuals who reside in both: 1)
homes that have lead and/or GRR service lines but also have other lead-containing plumbing
components, and 2) those that do not have lead and/or GRR service lines but do have lead-containing
plumbing components.208 The EPA has determined that the final LCRI requirements may result in

208 Although the EPA estimated an average lead concentration for the first 10 liters of drinking water to inform the
water lead concentration estimates used to quantify benefits the EPA could not calculate the CCT benefits

Final LCRI Economic Analysis

6-6

October 2024


-------
reduced lead exposure to the occupants of both these types of buildings as a result of improved
monitoring and additional actions to optimize CCT. In the analysis of the LCRI, the number of both
homes served by lead and/or GRR service lines and homes not served by lead and/or GRR service lines
potentially affected by water systems increasing their corrosion control during the 35-year period of
analysis is 5.2 million in the low scenario and 9.1 million in the high scenario. Some of these households
may have leaded plumbing materials apart from lead or GRR service lines, including lead connectors,
leaded brass fixtures, and lead solder. These households could potentially see reductions in tap water
lead concentrations.

Some researchers have pointed to the potential for CCT cobenefits associated with reduced corrosion,
or material damage, to plumbing pipes, fittings, and fixture, and appliances that use water owned by
both water systems and homeowners (Levin, 2023). The corrosion inhibitors used by systems that are
required to install or re-optimize CCT as a result of the final LCRI are expected to result in additional
benefits associated with the increased useful life of the plumbing components and appliances (e.g.,
water heaters), reduced maintenance costs, reduced treated water loss from the distribution system
due to leaks, and reduced potential liability and damages from broken pipes in buildings that receive
treated water from the system. The replacement of GRR service lines may also lead to reduced treated
water loss from the distribution system due to leaks (AwwaRF and DVGW-Technologiezentrum Wasser,
1996). The EPA did not have sufficient information to estimate these impacts nationally for the final rule
analysis.

Additionally, the risk of adverse health effects associated with copper that are expected to be reduced
by the final LCRI are summarized in Appendix E. These risks include acute gastrointestinal symptoms,
which are the most common adverse effect observed among adults and children. In sensitive groups,
there may be reductions in chronic hepatic effects, particularly for those with rare conditions such as
Wilson's disease and children predisposed to genetic cirrhosis syndromes. These diseases disrupt copper
homeostasis, leading to excessive accumulation that can be worsened by excessive copper ingestion
(NRC, 2000).

6.3 Comparison of Incremental Costs to Incremental Benefits

Exhibit 6-4 and Exhibit 6-5 compare the yearly undiscounted incremental cost and benefits under the
low and high scenario, respectively. The incremental costs of the rule are highest between 2027 to 2036
as public water systems (PWSs) are replacing SLs with lead content. In year 14 costs drop considerably.
The incremental benefits of the rule generally increase over time as SLs with lead content are replaced.
Yearly incremental net benefits are positive in the first three periods as a result of larger spending under
the baseline 2021 LCRR. Starting in year 2027, incremental net benefits are negative for a period of 5
years. Over the remainder of the period of analysis incremental yearly net benefits are positive. Total
undiscounted net benefits range from $479 billion to $918 billion.

associated with lead containing plumbing components (apart from lead and/or LSL/GRR service lines), because the
EPA used a pooled estimate for all CCT conditions in residences with no lead and/or LSL/GRR service lines in place
(See Chapter 5, section 5.2.3 for additional information).

Final LCRI Economic Analysis

6-7

October 2024


-------
Exhibit 6-4: Comparison of Yearly Monetized National Incremental Costs to Benefits of the

LCRI under Low Scenario (millions 2022 USD)

Year

Yearly Incremental Costs

Yearly Incremental Benefits

Yearly Net Benefits

2024

-$614.3

-$14.0

$600.3

2025

-$211.2

-$41.9

$169.3

2026

-$308.1

-$83.9

$224.2

2027

$4,577.3

$245.8

-$4,331.4

2028

$3,856.1

$784.1

-$3,072.1

2029

$4,021.3

$1,396.5

-$2,624.8

2030

$4,435.3

$2,110.4

-$2,324.9

2031

$4,576.8

$2,948.5

-$1,628.2

2032

$3,885.6

$4,629.2

$743.6

2033

$3,775.9

$6,346.3

$2,570.4

2034

$3,729.4

$8,051.5

$4,322.1

2035

$3,756.7

$9,905.9

$6,149.2

2036

$3,779.8

$11,970.0

$8,190.2

2037

$306.7

$14,234.0

$13,927.3

2038

$199.3

$16,477.2

$16,277.9

2039

$190.0

$18,082.5

$17,892.6

2040

$208.5

$19,518.9

$19,310.4

2041

$210.0

$20,802.9

$20,592.9

2042

$278.9

$21,627.3

$21,348.4

2043

$204.4

$22,314.1

$22,109.7

2044

$159.4

$22,709.9

$22,550.5

2045

$166.7

$22,966.6

$22,799.8

2046

$166.7

$23,075.3

$22,908.6

2047

$188.3

$23,034.7

$22,846.4

2048

$251.2

$22,991.5

$22,740.3

2049

$90.9

$22,945.4

$22,854.6

Final LCRI Economic Analysis

6-8

October 2024


-------
Year

Yearly Incremental Costs

Yearly Incremental Benefits

Yearly Net Benefits

2050

$250.1

$22,895.3

$22,645.3

2051

$291.3

$22,840.4

$22,549.1

2052

$426.5

$22,787.1

$22,360.6

2053

$190.1

$22,730.3

$22,540.2

2054

$186.7

$22,671.9

$22,485.2

2055

$184.7

$22,612.4

$22,427.7

2056

$183.9

$22,550.3

$22,366.4

2057

$258.1

$22,489.6

$22,231.6

2058

-$25.4

$22,427.6

$22,452.9

Acronyms: LCRI = Lead and Copper Rule Improvements; USD = United States dollar.

Exhibit 6-5: Comparison of Yearly Monetized National Incremental Costs to Benefits of the

LCRI under High Scenario (millions 2022 USD)

Year

Yearly Incremental Costs

Yearly Incremental Benefits

Yearly Net Benefits

2024

-$888.7

-$52.3

$836.5

2025

-$435.6

-$155.7

$279.9

2026

-$657.0

-$311.6

$345.4

2027

$5,803.7

$546.9

-$5,256.8

2028

$4,898.9

$1,797.6

-$3,101.3

2029

$5,588.0

$3,113.9

-$2,474.1

2030

$6,320.7

$4,146.0

-$2,174.7

2031

$6,758.6

$5,195.1

-$1,563.5

2032

$5,339.0

$8,039.9

$2,700.9

2033

$5,176.1

$11,080.5

$5,904.4

2034

$5,139.5

$14,184.6

$9,045.2

2035

$5,169.0

$17,237.3

$12,068.3

2036

$5,190.3

$21,067.2

$15,876.9

2037

$306.3

$25,611.1

$25,304.8

2038

$199.3

$30,085.7

$29,886.5

2039

$193.5

$32,917.9

$32,724.4

Final LCRI Economic Analysis

6-9

October 2024


-------
Year

Yearly Incremental Costs

Yearly Incremental Benefits

Yearly Net Benefits

2040

$219.8

$35,508.0

$35,288.2

2041

$222.3

$38,242.8

$38,020.5

2042

$289.4

$40,437.2

$40,147.8

2043

$217.5

$42,346.4

$42,128.8

2044

$185.1

$43,182.3

$42,997.2

2045

$196.6

$43,705.1

$43,508.5

2046

-$44.6

$43,903.8

$43,948.4

2047

$105.8

$43,769.1

$43,663.3

2048

-$155.4

$43,627.0

$43,782.4

2049

$476.8

$43,479.7

$43,002.9

2050

$128.0

$43,324.6

$43,196.6

2051

$777.1

$43,162.7

$42,385.6

2052

$363.0

$42,999.4

$42,636.5

2053

$427.8

$42,829.2

$42,401.5

2054

$263.3

$42,656.7

$42,393.4

2055

$238.4

$42,482.6

$42,244.2

2056

$208.6

$42,305.6

$42,097.0

2057

$271.1

$42,130.3

$41,859.2

2058

-$424.2

$41,952.8

$42,377.0

Acronyms: LCRI = Lead and Copper Rule Improvements; USD = United States dollar.

Exhibit 6-6 compares the estimated annualized monetized incremental costs and the estimated
annualized monetized incremental benefits of the final LCRI at a 2 percent discount rate; the monetized
net annualized incremental benefits range from $12.0 billion to $23.2 billion.

Exhibit 6-6: Comparison of Estimated Monetized National Annualized Incremental Costs to
Benefits of the LCRI - 2 Percent Discount Rate (millions 2022 USD)



Low Scenario

High Scenario

Annualized Incremental Costs

$1,468.8

$1,953.8

Annualized Incremental Benefits

$13,493.5

$25,142.6

Annual Net Benefits

$12,024.7

$23,188.8

Acronyms: LCRI = Lead and Copper Rule Improvements; USD = United States dollar.

Final LCRI Economic Analysis

6-10

October 2024


-------
It is important to note that, as described in Chapter 4, the EPA determined it does not have enough
information to perform a probabilistic uncertainty analysis as part of the SafeWater LCR model analysis
for this rule. Instead, to capture uncertainty, the EPA estimated compliance costs (and benefits) by
running the SafeWater LCR model under low and high bracketing scenarios. For costs, the bracketing
scenarios are defined by the following three cost drivers:

1.	Likelihood a model PWS will exceed the lead action level (AL) and/or trigger level (TL) under
the 2021 LCRR and the AL under the final LCRI.

2.	LSLR unit costs.

3.	CCT unit costs.

The low and high benefits bracketing scenarios are defined by the following benefits variables:

1.	Likelihood a model PWS will exceed the AL and/or TL under the 2021 LCRR and the AL under
the final LCRI (also used to define the low and high cost scenarios in the cost analysis).

2.	The concentration-response functions that characterize how reductions in blood lead levels
(caused by changes in lead exposure) translate into avoided IQ reductions, cases of ADHD,
and CVD premature mortality.

3.	Two alternative low and high valuations for the ADHD cost of illness.

The EPA expects the significant portion of potential uncertainty is captured by this bracketing approach.
However, some uncharacterized uncertainties still exist which may result in cost and benefit estimates
that fall outside of the range of costs and benefits described in the bracketing model results.

Unquantified uncertainty associated with the quantified cost estimates can come from a number of
sources. There may be uncaptured variation in the three variables that define the low and high cost
scenarios. In general the agency estimated bracketing values that captured national average variability,
therefore extreme measures were not included. The EPA used the 25th and 75th percentiles from the 7th
Drinking Water Infrastructure Needs Survey and Assessment (DWINSA) dataset on SLR costs to define
average national cost of SLR but additional information from the dataset shows that lower or higher SLR
costs can be found across some systems. The CCT unit cost range derived in the work breakdown
structure (WBS) models represents reasonable low and high estimated CCT costs based on quality of
treatment system components and average assumptions about the complexity of the CCT system and
on-site installation requirements. These values may not capture all potential variability in site specific
requirements across systems. The estimated 90th percentile range is based on the range in the Safe
Drinking Water Information System (SDWIS) reported data of the period from 2012 to 2020. The degree
to which this is representative of future sample variation is uncertain, although in this case the range is
based on the maximum and minimum 90th percentiles reported for each system which should capture a
significant portion of the variability. Also, there is uncertainty in the adjustments the EPA makes to the
SDWIS data to account for changes in tap sampling requirements which may introduce uncertainty.
Uncertainty, in this value, affects the number of systems required by the rule to install or re-optimize
CCT, and distribute point-of-use (POU) devices, potentially significantly impacting cost estimates. In
addition to the three variables defining the low/high range uncertainty in the baseline information on
the number of lead, GRR, and unknown service lines, and the number of systems with CCT in-place and
the starting values for pH and orthophosphate at PWSs with existing CCT all are potential sources of
uncertainty that could significantly affect computed cost estimates. Lesser drivers of cost estimate
unquantified variability include the uncertainty associated with smaller burden and unit costs estimates,

Final LCRI Economic Analysis

6-11

October 2024


-------
many of which are point estimates, tied to a number of required activities from rule implementation and
recordkeeping to sampling and public education. Generally, the EPA does not have data to determine if
the uncertainty associated with cost inputs would systematically lead to an over or under estimate of
the national level quantified estimated costs.

Regarding the benefits estimates, again there is unquantified uncertainty associated with the variables
that define the estimated range. Uncertainty in the estimation of the 90th percentile values affect
benefits in the same way it affects costs, potentially not capturing the complete variation in systems
being required to conduct CCT installation or re-optimization, and POU distribution impacting both cost
and benefit values. There is uncertainty about which statistical functions best describes the relationship
between blood lead and health effects which may be only partially captured by the estimated functions
from the studies the EPA uses to bracket the quantified benefit estimates. Also, there is uncertainty in
the extrapolation of concentration-response functions between lead and adverse health effects to blood
lead levels lower than those observed in the original studies. There is also uncertainty on the timing of
blood lead measurements in relation to the health effects, as low birth weight (LBW), CVD premature
mortality and ADHD rely on one time-blood lead measurements. For IQ, this has been well studied with
cohorts of children with repeated measures, and relationships were observed with all measures
(Lanphear et al., 2015,2019). However, this level of detail was not available for other endpoints. Also, no
cessation lag is assumed in the modeling. This would impact the monetary estimate for CVD premature
mortality the most, potentially causing an overestimation of the monetary value of this avoided risk.
Additionally, the use of cost of illness estimates for LBW and ADHD does not include an individual's
willingness to pay to avoid health risk, or any reductions in quality of life due to the health condition,
which may lead to an underestimation of benefits. Uncertainties in the value of an IQ point are
described in detail in Appendix J of the LCRR Economic Analysis (USEPA, 2020). There may also be
additional uncertainty associated with the persistence of ADHD into adulthood not accounted for in the
EPA characterization using values from Barbaresi et al. (2013) and Sibley et al. (2022) to produce a range
of low and high cost of illness estimates used in the valuation of ADHD cases.

In addition to the health endpoint valuation uncertainties there exists uncertainty related to the
estimation of baseline and policy scenario drinking water lead concentrations that can affect the
estimated benefits values. The data available to estimate these concentrations was limited, and the EPA
relied on detailed information from 18,571 samples collected from 1,657 homes in 16 cities representing
15 city water systems across the United States and Canada. Modeling was then used to estimate the
concentrations that were assumed to apply nationwide. It is unclear if better characterization of drinking
water concentrations would result in higher or lower benefits estimates. In addition to the drinking
water lead concentrations other background sources of lead from soil, air, and food were held constant
when estimating BLLs. Given the many Federal and State lead reduction initiatives in these other lead
source areas it is likely that over the period of analysis background lead exposure in the population will
decrease. If lead levels in other media besides water are decreasing in the baseline, then this analysis
would underestimate benefits from reductions due to the final LCRI because the dose-response
functions for IQ, low birth weight, and CVD premature mortality are log linear and that the
concentration response functions show no evidence of a threshold below which effects cease. The EPA
also does not account for the presence and degree of averting behavior which would take place in the
baseline 2021 LCRR or the final LCRI. The relative degree of averting behavior that could be present
across the regulatory scenarios could impact the incremental benefits of the rule.

Final LCRI Economic Analysis

6-12

October 2024


-------
The EPA also considered both monetized and non-monetized costs and benefits for the final rule; see
Sections VI.F.l and VI.F.2 of the final rule Federal Register Notice.

The EPA, in April 2024, finalized the PFAS National Primary Drinking Water Regulation. The PFAS rule's
total annualized cost is expected to be $1.5 billion in2022 dollars discounted at 2 percent. The PFAS rule
has estimated total annualized benefits also of $1.5 billion in 2022 dollars discounted at 2 percent. For
additional detailed information on the PFAS rule and its potential costs and benefits see the PFAS final
rule Federal Register Notice (88 FR 18638). Implementation timing associated with this PFAS rule and
the final LCRI has the potential to overlap. To the extent implementation overlaps, some rule start-up,
administrative, and sampling/SL inventory costs associated with both rules could affect a large number
of PWSs and States. The more significant costs of installing and operating PFAS treatment technology in
a similar time frame with installing and operating CCT and/or conducting service line replacement are
expected to fall on some systems. The EPA does not have sufficiently detailed PFAS occurrence, and
LSL/GRR service line and 90th percentile lead tap sample data to explore the potential treatment cost
interactions of the two rules. The EPA further notes that SDWA section 1412(b)(3)(C)(i)(lll) requires that
the EPA include quantifiable and non-quantifiable costs that are likely to occur solely as a result of
compliance with the rule including monitoring, treatment and other costs and excluding costs resulting
from compliance with other proposed or promulgated regulations.

6.4 References

AwwaRF and DVGW-Technologiezentrum Wasser. 1996. Internal Corrosion of Water Distribution
Systems. 2nd edition. AwwaRF Order 90508. Project #725. AWWA Research Foundation (now Water
Research Foundation) and AWWA. Denver, CO.

Lanphear, B.P., R. Hornung, J. Khoury, K. Yolton, P.A. Baghurst, D.C. Bellinger, and R. Roberts. 2005.
Low-level environmental lead exposure and children's intellectual function: An international pooled
analysis. Environmental Health Perspectives 113(7): 894-899. doi:10.1289/ehp.7688.

Lanphear, B.P., R. Hornung, J. Khoury, K. Yolton, P.A. Baghurst, D.C. Bellinger, R. Roberts. 2019. Erratum:
Low-level environmental lead exposure and children's intellectual function: An international pooled
analysis. Environmental Health Perspectives 127(9): 099001.

Levin, R., and J. Schwartz. 2023. A better cost:benefit analysis yields better and fairer results: EPA's lead
and copper rule revision. Environmental Research 229: 115738.
https://doi.Org/10.1016/j.envres.2023.115738

National Research Council (NRC). 2000. Copper in Drinking Water. Washington, D.C.: National Academies
Press.

National Toxicology Program (NTP). 2012. NTP Monograph: Health Effects of Low-Level Lead. U.S.
Department of Health and Human Services. Office of Health Assessment and Translation. Division of the
National Toxicology Program. Available at

https://ntp.niehs.nih.gov/ntp/ohat/lead/final/monographhealtheffectslowlevellead newissn 508.pdf.

United States Census Bureau. 2010. Table AVG1. Average Number of People Per Household, By Race and
Hispanic Origin, Marital Status, Age, And Education of Householder: 2010. Available at
http://www.census.gov/hhes/families/data/cps2010.html.

Final LCRI Economic Analysis

6-13

October 2024


-------
United States Environmental Protection Agency (USEPA). 2013. Integrated Science Assessment for Lead.
July 2013. Office of Research and Development. EPA/600/R-10/075F.

USEPA. 2024. Integrated Science Assessment for Lead (Final Report). U.S. Environmental Protection
Agency, Washington, DC. EPA/600/R-23/375

USEPA. 2020. Economic Analysis Appendices for the Final Lead and Copper Rule Revisions. December
2020. Office of Water. EPA 816-R-20-008a.

Final LCRI Economic Analysis

6-14

October 2024


-------
7 Statutory and Administrative Requirements

7.1	Introduction

As part of the rulemaking process, the United States Environmental Protection Agency (EPA) is required
to address the direct and indirect burden that the final Lead and Copper Rule Improvements (LCRI) may
place on certain types of governments, businesses, and populations. This chapter presents analyses
performed by the EPA in accordance with the following federal mandates and statutory requirements:

•	Executive Order 12866: Regulatory Planning and Review and Executive Order 13563: Improving
Regulation and Regulatory Review.

•	Paperwork Reduction Act (PRA).

•	The Regulatory Flexibility Act (RFA) of 1980, as amended by the Small Business Regulatory
Enforcement Fairness Act (SBREFA) of 1996.

•	Unfunded Mandates Reform Act (UMRA) of 1995.

•	Executive Order 13132: Federalism.

•	Executive Order 13175: Consultation and Coordination with Indian Tribal Governments.

•	Executive Order 13045: Protection of Children from Environmental Health and Safety Risks.

•	Executive Order 13211: Actions That Significantly Affect Energy Supply, Distribution, or Use.

•	National Technology Transfer and Advancement Act of 1995 (NTTAA).

•	Executive Order 12898: Federal Action to Address Environmental Justice in Minority Populations
and Low-Income Populations and Executive Order 14096 (Revitalizing Our Nation's Commitment
to Environmental Justice for All).

•	The Safe Drinking Water Act (SDWA) required consultations with the Science Advisory Board
(SAB), National Drinking Water Advisory Council (NDWAC), and the Secretary of Health and
Human Services.

Many of the statutory requirements and executive orders listed above call for an explanation of why the
final LCRI requirements are necessary, the statutory authority for the requirements, and the primary
objectives that the final requirements are intended to achieve (see Chapter 2 for additional information
regarding the goals of the final LCRI). Others are designed to assess the financial and health effects of
the final regulatory requirements on sensitive, low-income, and tribal populations as well as on small
systems and governments.

7.2	Executive Order 12866: Regulatory Planning and Review and Executive Order 14094:
Modernizing Regulatory Review

Executive Order 12866, Regulatory Planning and Review (58 FR 51735, October 4, 1993), gives the Office
of Management and Budget (OMB) the authority to review regulatory actions that are categorized as

Final LCRI Economic Analysis

7-1

October 2024


-------
"significant" under Section 3(f) of Executive Order 12866 as modified by Section 1 of Executive Order
14094 (88 FR 21879, April 6, 2023). The Order defines a "significant regulatory action" as any regulatory
action that is likely to result in a rule that may:

1.	Have an annual effect on the economy of $200 million or more (adjusted every three years by
the Administrator of the Office of Information and Regulatory Affairs (OIRA) for changes in gross
domestic product;) or adversely affect in a material way the economy, a sector of the economy,
productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal
governments or communities;

2.	Create a serious inconsistency or otherwise interfere with an action taken or planned by another
agency;

3.	Materially alter the budgetary impact of entitlements, grants, user fees, or loan programs or the
rights and obligations of recipients thereof; or

4.	Raise legal or policy issues for which centralized review would meaningfully further the
President's priorities or the principles set forth in this Executive order, as specifically authorized
in a timely manner by the Administrator of OIRA in each case.

This action is significant under section 3(f)(1) of Executive Order 12866 and was submitted to the OMB
for review. Any changes made in response to recommendations arising from OMB's review process have
been documented in the docket. In Chapter 6, Exhibit 6-6, compares the monetized annual estimated
incremental costs and the monetized annual incremental benefits of the final LCRI at a 2 percent
discount rate. The net monetized annual incremental benefits range from $12 to $23 billion. The range
in reported values represent cost-benefit estimation under the low and high scenarios developed by the
agency to characterize uncertainty in the computed estimates.

In addition to the monetized costs and benefits of the final LCRI, a number of non-monetized and non-
quantified impacts exist. See Chapter 6, Sections 6.1.2 and 6.2.2 for a detailed listing of the non-
monetized costs and non-quantified benefits, respectively, associated with the lead exposure reductions
of the final LCRI.

7.3 Paperwork Reduction Act

The information collection requirements for the final LCRI have been submitted for approval to OMB
under the PRA, 44 U.S.C. 3501 etseq. The Information Collection Request (ICR) document that the EPA
prepared has been assigned the EPA ICR number 2788.02 and is available in the docket at EPA-HQ-OW-
2022-0801 at www.regulations.gov.

The PRA requires the EPA to estimate the burden, as defined in 5 CFR 1320.3(b), on systems and States
of complying with the rule. ("State" is used throughout this chapter to describe States, Tribes, and
territories with primary enforcement responsibility.) The information collected as a result of the final
LCRI should allow States and the EPA to determine appropriate requirements for specific systems and
evaluate compliance with the final LCRI. Burden is defined at 5 CFR 1320.3(b) and means the total time,
effort, and financial resources required to generate, maintain, retain, disclose, or provide information to

Final LCRI Economic Analysis

7-2

October 2024


-------
or for a federal agency. The burden includes the time needed to conduct State and system activities
during the first three years after promulgation, as described in Sections 7.3.1 and 7.3.2, respectively.

7.3.1	State Activities

The EPA anticipates States will be involved in the following activities for the first three years after
publication of the final LCRI:

Implementation

•	Adopt the rule and develop implementation program.

•	Modify data management systems.

•	Provide State staff training.

•	Provide water system staff with training and technical assistance for implementation.

•	Review the updated Lead and Copper Rule Revisions (LCRR) initial inventories that will contain
lead connector information and public water system (PWS) demonstrations and written
statements of only non-lead service lines, non-lead connectors, or no connectors present from
systems in lieu of a publicly accessible inventory data.

•	Review water system service line replacement plans, including reviewing information on
deferred deadline and associated replacement rate in the SLR plan and determine fastest
feasible rate.

•	Provide a template for the public education materials on lead, GRR, and unknown service lines
that water systems must deliver annually to customers served by those types of lines.

•	Review the public education materials on lead, GRR, and unknown service lines that water
systems develop to be delivered annually to customers served by those types of lines.

•	Review the updated tap sampling plans.

7.3.2	System Activities

The EPA anticipates systems will be involved in the following activities for the first three years after
publication of the final LCRI:

Implementation

•	Read and understand the rule.

•	Assign personnel and resources for rule implementation.

•	Attends training and receive technical assistance from the State.

•	Update and submit to the State a service line inventory that includes lead connector
information.

Final LCRI Economic Analysis

7-3

October 2024


-------
•	Develop and submit to the State a service line replacement plan, including information on
participation in a deferred replacement plan, if eligible, and identifying funding options for full
service line replacements.

•	Update unknown service lines through normal field operations.

•	Develop and submit to the State public education materials on lead, GRR, and unknown service
lines that must be delivered annually to customers served by those types of lines.

•	Annually distribute public education materials on lead, GRR, and unknown service lines to
customers served by those types of lines.

•	Update and submit to the State a tap sampling plan.

For the first three years after publication of the rule in the Federal Register, the major information
requirements apply to 67,003 respondents annually, including 66,947 PWSs and 56 Primacy Agencies.
The net change in burden associated with moving from the information requirements of the 2021 LCRR
to those in the final LCRI over the three years covered by the ICR is -916,723 hours, for an average of -
305,574 hours per year. The total net change in costs from the most recent ICR approved for the 2021
LCRR209 over the three-year compliance period covered by this ICR are $131.5 million, for an average of
$43.8 million per year (simple average over three years). The net average burden per response (i.e., the
amount of time needed for each activity that requires a collection of information) is -0.11 hours; the net
average cost per response is -$6.65. Because the final LCRI requirements will nullify most of the
requirements of the 2021 LCRR during the three-year implementation period for the LCRI, the burden
for the final LCRI is lower than the anticipated burden that would have occurred under the 2021 LCRR
over the same period, resulting in a negative net burden for the final LCRI. The costs for system activities
under the LCRI, however, are greater than those for the same period under the 2021 LCRR. The
collection requirements are mandatory under the SDWA (42 U.S.C. 300j-4 subsections (a)(1)(A) and
(a)(1)(B)). Details on the calculation of the final LCRI information collection burden and costs can be
found in the ICR for the final LCRI and Chapter 4 of this economic analysis (EA).

A summary of the average annual net burden and costs of the collection is presented in Exhibit 7-1.

Exhibit 7-1: Estimated Change in Average Annual Net Burden and Costs for the Final LCRI ICR

Item

Burden (labor)

Labor Costs
($2022)

Non-Labor
Costs
($2022)

Total Costs ($2022)

Responses

Systems

-79,849

1,546,949

$59,669,117

$61,216,127

37,771,470

States

-225,725

-17,366,566

0

-$17,366,566

-204,381

Total

-305,574

-15,819,617

$59,669,117

$43,849,560

37,567,089

Average per response

-0.11

-5.02

-$1.63

-$6.65

not applicable

Source: ICR Supporting Statement, available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.
Note: Calculated in 2022 dollars. Detail may not add to totals because of independent rounding. Results show the
upper bound estimate for the number of lead lines located in non-transient non-community water systems

209 Office of Management and Budget (OMB) approved the initial "Information Collection Request for Lead and
Copper Rule Revisions (LCRR)" on July 25, 2022.

Final LCRI Economic Analysis

7-4

October 2024


-------
(NTNCWSs), Chapter 4 of this EA documents a difference of approximately five lines between the upper- and
lower-bound estimates.

The total responses, burden, and cost for system and State startup activities, LSL inventory, public
education, and service line replacement plan is provided in Exhibit 7-2.

Exhibit 7-2: Estimated Total Responses, Burden, and Costs for the Final LCRI ICR for Each

Required Activity

Requirement

Responses

Burden
(Hours)

Labor Cost
($2022)

Non-Labor
Cost ($2022)

Total Cost
($2022)

System start-up activities (read
rule, assign staff, attend
training)

200,841

1,338,940

$50,720,720

$0

$50,720,720

Systems review records for
connector material to prepare
the updated initial inventory

200,841

4,469,095

$220,425,702

$0

$220,425,702

Systems submit the updated
initial inventory with connector
information

200,841

381,190

$14,596,761

$0

$14,596,761

Systems conduct normal and
field operations to update
unknown service lines

9,392,500

0

$0

$431,570,015

$431,570,015

Systems develop and submit a
service line replacement plan

25,823

423,876

$17,510,054

$0

$17,510,054

Systems include information on
deferred deadline and
associated replacement rate in
the SLR plan

5

33

$1,773

$0

$1,773

Systems identify funding
options for full SLRs

25,360

1,838,704

$72,886,662

$0

$72,886,662

Systems develop public
education materials for
customers on service lines with
lead or unknown content and
submit to primacy agencies for
review

25,823

180,761

$7,028,788

$0

$7,028,788

Systems distribute public
education materials for
customers on service lines with
lead or unknown content

176,983,809

1,978,966

$87,823,986

$68,788,580

$156,612,566

Systems update and submit tap
sampling plan

66,947

347,828

$13,823,729

$0

$13,823,729

System Subtotal

187,122,790

10,959,392

$484,818,175

$500,358,595

$985,176,770

States start-up activities (read
rule, adopt rule, modify data
systems, provide training)

224

399,168

$23,944,566

$0

$23,944,566

Final LCRI Economic Analysis

7-5

October 2024


-------
Requirement

Responses

Burden
(Hours)

Labor Cost
($2022)

Non-Labor
Cost ($2022)

Total Cost
($2022)

States review updated initial
inventories with connector
information

200,841

200,841

$12,047,686

$0

$12,047,686

States review service line
replacement plan

25,823

205,690

$12,338,559

$0

$12,338,559

State reviews information on
deferred deadline and
associated replacement rate in
the SLR plan and determine
fastest feasible rate

5

16

$979

$0

$979

States provide templates to
systems for public education on
service lines with lead or
unknown content and reviews
developed material

25,823

23,767

$1,425,669

$0

$1,425,669

States review updated tap
sampling plan

66,947

174,784

$10,484,626

$0

$10,484,626

State Subtotal

319,663

1,004,266

$60,242,085

$0

$60,242,085

Combined Systems and State

187,442,453

11,963,658

$545,060,260

$500,358,595

$1,045,418,855

Source: ICR Supporting Statement, available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov
Note: Calculated in 2022 dollars. Detail may not add to totals because of independent rounding. Results show the
upper bound estimate for the number of lead lines located in NTNCWSs.

An agency may not conduct or sponsor, and a person is not required to respond to, a collection of
information unless it displays a currently valid OMB control number. The OMB control numbers for the
EPA's regulations in 40 CFR are listed in 40 CFR part 9.

7.4 The Regulatory Flexibility Act

The RFA of 1980, amended by the SBREFA of 1996, requires regulators to assess the effects of
regulations on small entities including businesses, nonprofit organizations, and governments.
RFA/SBREFA generally requires an agency to prepare a final regulatory flexibility analysis (FRFA) of any
rule subject to notice and comment rulemaking requirements under the Administrative Procedure Act
or any other statute unless the agency certifies that the rule will not have a significant economic impact
on a substantial number of small entities (SISNOSE). Small entities include small businesses, small
organizations, and small governmental jurisdictions.

The RFA provides default definitions for each type of small entity. Small entities are defined as: 1) a
small business as defined by the Small Business Administration's (SBA) regulations at 13 CFR 121.201; 2)
a small governmental jurisdiction that is a government of a city, county, town, school district, or special
district with a population of less than 50,000; and 3) a small organization that is any "not-for-profit
enterprise which is independently owned and operated and is not dominant in its field." However, the
RFA also authorizes an agency to use alternative definitions for each category of small entity, "which are
appropriate to the activities of the agency" after proposing the alternative definition(s) in the Federal

Final LCRI Economic Analysis

7-6

October 2024


-------
Register and taking comment (5 (JSC 601(3)-(5)). In addition, to establish an alternative small business
definition, agencies must consult with SBA's Chief Counsel for Advocacy.

For purposes of assessing the impacts of the final LCRI on small entities, the EPA considered small
entities to be systems serving 10,000 people or fewer. This is the cut-off level specified by Congress in
the 1996 Amendments to SDWAfor small system flexibility provisions. As required by the RFA, the EPA
proposed using this alternative definition in the Federal Register (FR) (63 FR 7620, USEPA, 1998a),
requested public comment, consulted with the SBA, and finalized the alternative definition in the
agency's Consumer Confidence Reports (CCR) regulation (63 FR 44524, USEPA, 1998b). As stated in that
Final Rule, the alternative definition would be applied for all future drinking water regulations.

The materials presented and referenced in this RFA section represent the EPA's regulatory flexibility
analysis. They examine the impacts of the final rule on small entities along with regulatory alternatives
that could minimize the impacts of the rulemaking.

7.4.1	Need for and Objectives of the Rule

The need for the rule, the objectives of the rulemaking, the stakeholder outreach conducted, and the
statutory authority the EPA is utilizing to finalize the rule are described in detail in Chapter 2. See
Section 2.1 for detailed information on the need for the rule and the Lead and Copper Rule's (LCR)
regulatory history, Sections 2.2 through 2.4 for information on stakeholder outreach during the
rulemaking process, and Section 2.5 for additional detail on the statutory authority for the promulgation
of the final LCRI.

7.4.2	Summary of SBAR Comments and Recommendations

The EPA convened a Small Business Advocacy Review (SBAR) Panel to review the planned final LCRI and
consult with small entity representatives (SERs) as required by Section 609(b) of the RFA and amended
by the 1996 SBREFA. Prior to convening the Panel, the EPA conducted outreach with 11 out of 14
potential SERs through a pre-Panel outreach meeting held on September 12, 2022, to solicit input on the
potential small systems implications of the forthcoming final LCRI. On November 15, 2022, the EPA's
Small Business Advocacy Chairperson convened the Panel with the Director of the Standards and Risk
Management Division within the EPA's Office of Ground Water and Drinking Water (OGWDW), the
Administrator of the OIRA within OMB, and the Chief Counsel for Advocacy of the SBA. The Panel met
with 8 out of 14 SERs to hear their comments on the planned final LCRI during the Panel outreach
meeting held on November 29, 2022. Through the pre-Panel and Panel outreach meetings, the SERs
provided feedback on key areas, including achieving 100 percent lead service line replacement (LSLR) in
small systems, complying with a revised tap sampling protocol, complying with a revised action level
(AL) and trigger level (TL) construct, reducing rule complexity, adding protection from sustained lead
levels above the AL, and changing the 2021 LCRR small system flexibilities. SERs also provided feedback
on additional topics, such as corrosion control treatment (CCT), schools, and public education.

The Panel's findings are summarized below.

Number and Types of Entities Affected

The SERs commented that some of the changes in the existing 2021 LCRR and final LCRI might pose
problems for water systems serving fewer than 100 people and water systems that primarily serve

Final LCRI Economic Analysis

7-7

October 2024


-------
schools and child care facilities. The Panel recommended that the EPA evaluate whether it is appropriate
to further differentiate LCRI requirements based on the differences among smaller water systems (e.g.,
flexibilities for very small systems serving fewer than 500 people, small systems serving between 501
and 3,300 people, and small systems serving between 3,301 and 10,000 people). The Panel also
recommended that the EPA consider the costs associated with multiple rule areas of the final LCRI
requirements in the EA and ways to reduce the burden on small systems including the interrelationship
amongst the areas of the rule requirements.

Recordkeeping, Reporting, and Other Compliance Requirements

The 2021 LCRR includes reporting and recordkeeping requirements for service line inventorying and
replacement, monitoring results, public notification, public education, and sampling results. At the same
time, the PRA requires that all reporting and recordkeeping requirements have practical utility and
appropriately balance the needs of the government with the burden on the public. As the EPA proceeds
with the final LCRI, the EPA assessed the need for revisions to 2021 LCRR reporting and recordkeeping
requirements and considered them in the estimation of the burden and benefits of the rule changes.
The EPA is committed to keeping paperwork requirements to the minimum necessary and to fulfill its
statutory obligations as required by the PRA.

Related Federal Rules

There are National Primary Drinking Water Regulations for over 90 contaminants. The EPA's drinking
water rules were developed with careful attention to the interaction between each new rule that
requires treatment changes. The Panel recommended that the EPA continue to ensure that any revisions
to the 2021 LCRR be coordinated with, and do not either duplicate or conflict with, the requirements of
other drinking water regulations, and the EPA should consider other drinking water rule costs for small
systems.

One of the treatment strategies that the pre-2021 LCR and 2021 LCRR identify for controlling lead and
copper corrosion is to add orthophosphate to drinking water, which may impact the phosphorus levels
in the wastewater discharges in communities, including those with numeric discharge criteria for
phosphorous under the Clean Water Act. The Panel recommended that the EPA estimate the impacts of
the addition of phosphate on wastewater treatment plants in the final LCRI. Some water systems are
responsible for both the drinking water system as well as the wastewater treatment system. Under
SDWA, the EPA is required to set regulatory standards that reduce adverse health effects to the extent
feasible; this includes the lead and copper regulations. The EPA has previously determined that CCT is
technologically feasible and affordable.

Regulatory Flexibility Alternatives

Lead Service Line Replacement

The EPA is finalizing improvements to the 2021 LCRR LSLR requirements under the LCRI, including a
requirement to achieve the goal of replacing all lead and GRR service lines in the nation as quickly as
feasible. In addition to regulatory requirements, the EPA has and will continue to take non-regulatory
actions to achieve replacement of all lead and GRR service lines.

The Panel recognized the steps the EPA has taken, and will continue to take, to ensure federal funds are
available to drinking water systems, especially those within disadvantaged communities. These funds
include but are not limited to available funding through the Bipartisan Infrastructure Law (BIL), the

Final LCRI Economic Analysis

7-8

October 2024


-------
Drinking Water State Revolving Fund, and the Water Infrastructure Improvement for the Nation Act.
Despite the many efforts the EPA takes to ensure federal funds are available to water systems, the Panel
recognized that funding streams are not guaranteed to be available to all small systems, that some small
systems may not pursue available funding opportunities for a variety of reasons, and that, in the
absence of this funding, these communities may have difficulty financing LSLR. The Panel recommended
that the EPA evaluate available recent data and LSLR cost information (including the EPA's Drinking
Water Infrastructure Needs Survey and Assessment) to inform the EA for the final LCRI. When evaluating
the cost of compliance, the Panel recommended that the EPA recognize that external funding sources
may not be available to all small systems.

SERs identified factors such as customer engagement and cooperation, contractor availability, and
supply chain issues that will challenge the rate at which they can replace 100 percent of their LSLs.

When developing the LSLR requirements, the Panel recommended that the EPA consider the barriers to
100 percent LSLR that SERs identified that make LSLR challenging. In the 2021 LCRR, the EPA recognized
that customers may refuse to participate in LSLR and required documentation of customer engagement.
The Panel recommended that the EPA include a provision in the final LCRI to account for customer
refusals in the mandatory LSLR provision and increase clarity in terms of what "good faith" attempts
mean when engaging the customer. The Panel recommended that the EPA provide additional time for
small systems to comply with applicable service line replacement requirements from the 2021 LCRR that
are revised by the final LCRI, including a transition period following the effective date to provide time for
small systems to plan replacement-required activities.

SERs expressed the importance of national-level technical assistance for small systems in both the pre-
Panel and Panel meetings. Therefore, the Panel recommended that the EPA respond to SER concerns on
the need for assistance in understanding and complying with the LCRI requirements. The EPA supports
small systems through several different avenues, i.e., developing guidance on the initial service line
inventory, providing technical assistance through the Environmental Finance Centers, holding monthly
webinars focused on issues small systems face, and hosting an annual drinking water workshop to bring
together stakeholders in drinking water systems. Considering the SERs continued concerns and the
degree to which technical assistance is crucial in reducing regulatory compliance costs, the Office of
Advocacy recommended that the EPA continue to consult regularly with small entities and State
regulatory authorities to ensure the efforts to provide technical assistance to small systems to address
regulated and emerging contaminants are effective and remain appropriately targeted.

The EPA is including LCRI requirements that are intended to achieve more equitable human health
protection outcomes, especially for service line replacement such as requiring the replacement plan to
be made available to the public to increase transparency in the process. Due to the cost of replacing the
customer-portion of an LSL, underserved communities could potentially experience disproportionate
exposure to lead from LSLs if measures to ensure equity are not incorporated into the final LCRI. The
EPA specifically asked for SER input about ways to ensure equitable service line replacement in the final
LCRI. Multiple SERs stated that LSLR and other system repairs are generally based on (1) infrastructure
needs and what may fail first rather than who the infrastructure serves and (2) how to complete the
most pressing infrastructure work as efficiently as possible. One SER mentioned that equity should
consider factors outside of finances, such as English as a second language and achieving proper
communication and notice on construction projects.

Final LCRI Economic Analysis

7-9

October 2024


-------
The 2021 LCRR required LSLR plans to include a LSLR prioritization strategy based on factors including
but not limited to the targeting of known LSLs, LSLR for disadvantaged consumers, and populations most
sensitive to the effects of lead. Systems can include additional factors important to their community,
e.g., unknown service lines suspected to be lead, areas with pressing system repairs or infrastructure
needs, areas with older homes, populations with higher blood lead levels based on available data. The
Panel recommended that the EPA consider the range of additional factors raised by SERs in addition to
equity principles when developing the final LCRI service line replacement plan and other service line
replacement requirements (e.g., areas with pressing system repairs, infrastructure needs, and areas
with older homes).

Tap Sampling

In the LCRR review, the EPA concluded that there are opportunities to better identify the communities
that are most at risk of elevated drinking water lead levels. For the final LCRI, the EPA evaluated
alternative tap sampling protocols that may better identify higher lead levels.

The EPA is finalizing a new tap sampling protocol that requires systems to collect both first- and fifth-
liter samples at LSL sites and to use the higher concentration for the 90th percentile lead level
calculation. SERs discussed various factors that may pose challenges for small systems to comply with a
new sampling protocol, including increased costs and burden for systems with LSLs, increased
complexity of the protocol and communicating instructions to customers, and difficulty obtaining
customer participation. SERs also expressed a lack of confidence in relying on homeowners to take
routine samples and suggested ideas like developing training videos on how to take fifth-liter samples.
Under the 2021 LCRR, systems with low 90th percentile lead levels and those without lead sources may
reduce their monitoring frequency. By updating the sampling protocol, among other rule requirements,
there will likely be additional systems that exceed the AL, thus requiring actions to reduce drinking
water lead exposure not otherwise required in order to protect public health. The EPA accounted for the
costs and benefits of these additional actions into consideration in the EA for the final LCRI. The Panel
recommended that the EPA clarify aspects of the sampling protocol in the final LCRI rule language, such
as a definition of a wide-mouth bottle, and provide additional time for small systems to comply with
monitoring and sampling requirements from the 2021 LCRR that are revised by the LCRI.

Reduced Rule Complexity

To provide better health protection and more effective rule implementation, the EPA evaluated options
for utilities to address lead contamination at lower levels and improve sampling methods. Additionally,
the EPA is finalizing revisions to the 2021 LCRR to reduce complexity of the lead AL and TL construct as
well as to ensure the final rule is easily understandable and requires appropriate and feasible corrective
actions.

The 2021 LCRR review identified a possible revision to eliminate the lead TL and lower the AL, which the
EPA is finalizing with the LCRI. Most SERs stated that the lead TL should be removed to reduce rule
complexity; however, one SER advocated for retaining the TL, noting that it could be beneficial to have a
warning prior to an action level exceedance (ALE). The Panel recommended that the EPA consider
removing the TL.

The Panel noted that the EPA has committed to evaluating lower ALs to increase public health
protection and the impacts that such a change will have on smaller systems, even though many of the

Final LCRI Economic Analysis

7-10

October 2024


-------
SERs expressed concern about the impact such a change would have. The Panel recommended that, if
the EPA determines that a lower AL is required, the EPA provide additional time for small systems to
comply with AL requirements from the 2021 LCRR that are revised by the final LCRI, including additional
time for planning for the lower AL. The Panel recommended that the EPA also consider the appropriate
level of public education requirements following an ALE for small systems. The Panel further
recommended that the EPA consider additional flexibilities and compliance assistance for small entities
serving isolated or primary non-English language-speaking communities. The Panel also recommended
that the EPA issue guidance on the LCRI, including sampling, on the same date as the date of publication
of the final rule (or as soon as possible after that date) to ensure the maximum time available for
training and transition.

Small System Flexibility

The EPA is also finalizing additional changes to improve public health protection and improve rule
implementation to ensure that the LCRI prevents adverse health effects of lead to the extent feasible.
Specifically, the EPA stated in the LCRR review that the agency could make improvements to the 2021
LCRR small system flexibility. The SERs discussed the small system flexibility compliance option of
installing, maintaining, and monitoring point-of-use (POU) devices. A SER noted that POU devices are
helpful for non-transient non-community water systems (NTNCWSs) and very small community water
systems (CWSs) due to implementation concerns; in addition, water systems are also experiencing
challenges obtaining certified pitcher filters, and the SER wondered how that might affect
noncompliance. Another SER noted that systems serving between 3,301 and 10,000 people typically
choose optimal corrosion control treatment after an ALE instead of the other available compliance
options. A different SER mentioned a study on the cost of POU filters and bottled water. The Panel
recommended that the EPA request comment on additional flexibilities for small water systems to
effectively reduce drinking water lead exposure and whether the EPA should allow these methods as
compliance alternatives as part of the small systems flexibilities. The Panel recommended that the EPA
should review the costs and availability of compliant POU or point-of-entry devices to ensure that the
flexibility remains available to small systems that want to use it.

For additional information on the recommendations the EPA received, see the SBAR Panel report
available at https://www.epa.gov/reg-flex/potential-sbar-panel-national-primary-drinking-water-
regulation-lead-and-copper-rule and in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov.

7.4.3 Number and Description of Small Entities Affected

The EPA used Safe Drinking Water Information System/Federal Version (SDWIS/Fed) data from the
fourth quarter 2020210 to identify 62,518 small PWSs that may be impacted by the final LCRI. A small
PWS serves 10,000 or fewer people. These water systems include 45,139 CWSs that serve year-round
residents and 17,379 NTNCWSs that serve the same persons over six months per year (e.g., a PWS that
is an office park or school). Additional information on the characteristics of these small drinking water

210 See Chapter 3, Section 3.2.1 of this document for a description of SDWIS/Fed. Section 3.2.1.1 provides
information on how systems are classified in SDWIS/Fed including by size category. Section 3.2.1.2 discusses Lead
and Copper Rule-specific data available in SDWIS/Fed including 90th percentile tap sampling data, violations, and
compliance milestones. Section 3.2.1.3 discusses the CCT treatment information available in SDWIS/Fed and
Section 3.2.1.4 outlines the programmatic review process for SDWIS/Fed data verification.

Final LCRI Economic Analysis

7-11

October 2024


-------
systems along with a discussion of uncertainty in the dataset used to derive the estimated number of
small systems impacted by the final LCRI can be found in Chapter 3, Section 3.3.1. Specifically, Exhibit 3-
2 and Exhibit 3-3 provide information on the inventory of small CWSs and NTNCWSs, respectively, by
source water type and by refined size categories including systems serving: 100 or fewer people, 101-
500 people, 501-1,000 people, 1,001-3,300 people, and 3,301-10,000 people. Of the total number of
small systems serving 10,000 or fewer people, 22,235 CWSs and 434 NTNCWSs are estimated to have
service lines with lead content or unknown/potential lead content. See Exhibit 3-10 and Exhibit 3-22 for
additional detail on the projected number of small CWSs and NTNCWSs, respectively, with lead content
service lines by the refined small system size categories. Also note that the EPA has estimated low and
high scenario percent of systems, including small systems, that will exceed the lead tap sample 90th
percentile final AL of 0.010 mg/L. Exhibit 4-4, in Chapter 4, Section 4.2.2.1, provides the estimated
percent of systems over the final AL. The low scenario estimates for systems exceeding the final lead AL
ranges from 4.4 to 21.0 percent depending on a system's LSL status (i.e., the presence or absence of
LSLs). The high scenario estimated percent of systems projected to be above the final lead AL ranges
from 8.7 to 38.9 percent depending on the system's LSL status.

7.4.4 Description of the Compliance Requirements of the Rule

For a description of the general regulatory requirements under the final LCRI see Chapter 1, Section 1.1
and the Federal Register Notice (FRN) for this final rule (USEPA, 2024).

Of particular importance to small entities is the flexibility for CWSs serving 3,300 or fewer people and all
NTNCWSs provided in the final LCRI to select the compliance options that best protects public health,
recognizing the unique nature of these systems. This flexibility applies to CWSs serving 3,300 or fewer
people and all NTNCWSs that exceed the final lead AL of 0.010 mg/L. Compliance options for these
systems after an ALE include the evaluation of CCT for installation or re-optimization. In lieu of CCT
requirements to address lead, with State approval, systems may also choose: (1) provision and
maintenance of POU devices or (2) replacement all lead-bearing plumbing materials. A CWS serving
3,300 or fewer people or any NTNCWS that exceeds the AL must select a compliance option and submit
a recommendation to the State for approval within six months from the end of the tap sampling
monitoring period in which it exceeded the AL. The State has six months to approve the
recommendation or designate an alternative approach. If the system has a subsequent ALE, it must
implement the compliance option selected and approved by the State.

Reporting and recordkeeping requirements associated with the final LCRI are discussed under the PRA in
Section 7.3, which requires that all reporting and recordkeeping requirements have practical utility and
appropriately balance the needs of the government with the burden on the public. The agency has
assessed the need for revisions to reporting and recordkeeping requirements and has considered them
in the estimation of the burden and benefits of the final rule changes.

The final LCRI includes requirements for: conducting an service line inventory, which include lead
connectors, that is updated annually; requiring mandatory full service line replacement; improving tap
sampling; installing or re-optimizing CCT when water quality declines; enhancing water quality
parameter monitoring; a Distribution System and Site Assessment provision to evaluate and remediate
elevated lead at a site where the tap sample exceeds the lead AL; utilizing pitcher filters and POU
devices; and improving customer education and outreach. These final rule requirements include

Final LCRI Economic Analysis

7-12

October 2024


-------
reporting and recordkeeping requirements. States are required to implement operator certification (and
recertification) programs per SDWA section 1419 to ensure operators of CWSs and NTNCWSs, including
small water system operators, have the appropriate level of certification to complete the required task,
including the recordkeeping requirements, for the final LCRI.

7.4.5 Costs and Benefits of the Final LCRI by Small System Size Category

The EPA estimated the incremental costs and benefits, as well as the incremental net benefits of the
final LCRI by PWS size category for small systems. As shown in Exhibit 7-3, the incremental benefits
exceed the incremental costs of the final LCRI for most PWS size categories. There is one exception for
the smallest PWSs serving 100 or fewer people: the incremental costs of the rule exceed the benefits.

Exhibit 7-3: Estimated Incremental Costs and Benefits of the Final LCRI by Small System Size
Category - 2 Percent Discount Rate (millions of 2022 USD)

PWS Population ^

Metric Low Scenario High Scenario

Served

25 -100

Total Annual Costs $41.9 $43.2
Total Annual Benefits $18.8 $28.4
Net Benefits -$23.1 -$14.8

101-500

Total Annual Costs $55.6 $58.1
Total Annual Benefits $76.3 $135.3
Net Benefits $20.7 $77.3

501-1,000

Total Annual Costs $23.7 $25.3
Total Annual Benefits $65.1 $113.7
Net Benefits $41.3 $88.5

1,001-3,300

Total Annual Costs $45.6 $50.4
Total Annual Benefits $200.4 $354.5
Net Benefits $154.8 $304.1

3,301-10,000

Total Annual Costs $110.0 $135.8
Total Annual Benefits $1,063.1 $1,845.3
Net Benefits $953.1 $1,709.4

Acronyms: PWS = public water system.

7.4.6 Analysis of Alternative Small System Rule Requirements

The EPA considered two options that would mitigate the economic impact of the final LCRI on small
entities. The options differed by the size threshold at which CWSs could take advantage of the
compliance flexibilities. The selected option, in the final LCRI, includes flexibility for CWSs that serve
3,300 or fewer people, and all NTNCWSs. If these water systems have a lead 90th percentile above the
AL, they can choose from the following three options, following State approval, to reduce the
concentration of lead in their water:

1. Optimize existing CCT or install new CCT.

Final LCRI Economic Analysis

7-13

October 2024


-------
2.	Install and maintain POU devices at all locations being served.

3.	Replace all lead-bearing plumbing materials in the system.211

To estimate the economic impact on small entities, the EPA's cost model applies the least-cost
compliance option to all model PWSs that exceed the AL. To determine the least-cost compliance
option, the cost of each alternative is computed across each representative model PWS in the cost
model based on its assigned characteristics including: the presence of CCT, the cost and effectiveness of
CCT, the starting water quality parameters, monitoring requirements, the number of entry points, the
unit cost of POU, and the number of households served. For an expanded discussion on the assignment
of system characteristics, see Chapter 4. These characteristics are the primary drivers in determining the
costs once a water system has been triggered into CCT installation or re-optimization or POU
requirements. The model estimates the net present value for implementing each compliance option and
selects the least-cost alternative to retain in the summarized final rule costs.

The EPA estimated low- and high-cost scenarios to characterize uncertainty in the cost model results.
These scenarios are functions of assigning different (low and high) input values to a number of variables
that affect the relative cost of the small system compliance options (see Chapter 4, Section 4.2.2 for
additional information on uncertain variable value assignment). Therefore, the selection of a compliance
option is different across the low- and high-cost scenarios.

The number of systems serving under 3,300 people that choose to install and maintain POUs under the
final LCRI range from 2,406 to 4,066. These PWS serve a total of between 250,048 and 474,266 people.

A second form of flexibility provided to all PWSs impacted by the rule, but that is most likely to benefit
small PWSs, is the ability for systems with ALEs to choose to replace all of their lead and GRR service
lines in five years or less and avoid the expense of having to conduct a pipe loop study prior to installing
or reoptimizing CCT. Systems choosing this option must replace at least 20 percent of lead and GRR
service lines annually and at the end of the five years, have no lead, GRR, or unknown service lines
remaining in their inventory. For systems with approximately 50 LSL or fewer, most or all the lines can
be replaced for the cost of the pipe rig study. These systems instead would be able to conduct a much
less expensive coupon study if needed after the mandatory service line replacement program has been
completed within five years or less.

In the case of the regulatory flexibility analysis, the EPA limited the assessment to small CWSs since
small NTNCWSs operate in numerous industries and the EPA does not have information on NTNCWSs
revenue. The EPA's decision to limit its regulatory flexibility analysis to CWSs is supported by the EPA's
Assessment of the Vulnerability of Noncommunity Water Systems to SDWA Cost Increases (2008). In this
study, the EPA examined the burden of SDWA rule costs in comparison to the average revenues of
various categories of NTNCWSs. All of the NTNCWS categories reviewed were found to be less
vulnerable to SDWA-related increases than a typical CWS. The report notes that, in some categories of
businesses, costs are more easily passed on to the customer base than in others. However, in each
NTNCWS category, expenditures on water were found to be a relatively small percentage of total

211 The EPA could not evaluate the cost of removing lead-bearing plumbing components from small systems, but
the agency notes that, if a system should select this option, it would likely be considered the lowest cost
alternative of the compliance options. Therefore, since the EPA has not included this option in its cost modeling,
the agency's small system compliance costs may be overestimated.

Final LCRI Economic Analysis

7-14

October 2024


-------
revenues. Water expenditures (including expenditures for sewer service and miscellaneous other
utilities) totaled less than 1 percent of total revenues in nearly all cases and were not more than 1.3
percent of total revenues for any category. Several caveats were put forth in this report, including one
that considered the potential for underestimating the impact to golf courses, which were grouped in
with other recreational entities whose use of water was less significant to the core business than the
golf courses. Despite the significant caveats listed, the report strongly suggests that NTNCWSs should
not be considered particularly vulnerable to operating cost increases resulting from SDWA rulemakings.

The EPA calculated the annual revenue per CWS by using each PWS's average daily flow and the average
revenue per thousand gallons delivered from the Community Water System Survey (CWSS) (USEPA,
2009, Table 61). These revenue estimates were then inflated to 2022 dollars using the consumer price
index (CPI) for utilities.

Exhibit 7-4 and Exhibit 7-5 provide the estimated total number of small CWSs, by system size and source
water type, which have incremental annual costs that exceed the 1 percent and 3 percent of annual
revenue threshold values under the low- and high-cost scenarios. Under the final LCRI, the number of
small CWSs that will experience incremental annual costs of more than 1 percent of revenues ranges
from 35,895 to 37,069 (80 percent to 82 percent of all small CWSs) and the number of small CWSs that
will have annual incremental costs exceeding 3 percent of revenues ranges from 26,993 to 27,568 (60
percent to 61 percent of small CWSs).

Final LCRI Economic Analysis

7-15

October 2024


-------
Exhibit 7-4: Estimated Incremental Costs vs. Revenue for Small CWSs - Low Scenario*

Category

Source Water

Size Category

Number of CWSs Number of CWSs Percent of CWSs
Number of CWSswith Cost Revenue with Cost Revenue with Cost Revenue
Ratio > 1% Ratio > 3% Ratio > 1%

Percent of CWSs
with Cost Revenue
Ratio > 3%

Private

Ground

<100

9,400

9,341

9,309

99%

99%

Private

Ground

101 to 500

8,190

7,842

6,699

96%

82%

Private

Ground

501 to 1,000

1,299

826

318

64%

24%

Private

Ground

1,001 to 3,300

1014

493

147

49%

14%

Private

Ground

3,301 to 10,000

336

179

120

53%

36%

Private

Surface

<100

404

400

398

99%

99%

Private

Surface

101 to 500

769

737

544

96%

71%

Private

Surface

501 to 1,000

232

133

41

57%

18%

Private

Surface

1,001 to 3,300

272

100

25

37%

9%

Private

Surface

3,301 to 10,000

182

100

64

55%

35%

Public

Ground

<100

1,409

1,404

1392

100%

99%

Public

Ground

101 to 500

4,838

4,705

3452

97%

71%

Public

Ground

501 to 1,000

2,869

2216

813

11%

28%

Public

Ground

1,001 to 3,300

4,488

2536

775

57%

17%

Public

Ground

3,301 to 10,000

2,459

1434

1011

58%

41%

Public

Surface

<100

518

516

510

100%

98%

Public

Surface

101 to 500

1,287

1245

776

97%

60%

Public

Surface

501 to 1,000

930

673

192

72%

21%

Public

Surface

1,001 to 3,300

2,193

1019

253

46%

12%

Public

Surface

3,301 to 10,000

2,049

1170

729

57%

36%

Total





45,138

37,069

27,568

82%

61%

Acronyms: CWS = community water system.

Notes

* When evaluating the economic impacts on PWSs, the EPA uses the estimated PWS cost of capital to discount future costs (not the 2 percent discount rate
used to evaluate social costs and benefit), as this best represents the actual costs of compliance that water systems would incur over time. For more
information on cost of capital see Chapter 4, Section 4.2.3.3.

Final LCRI Economic Analysis

7-16

October 2024


-------
Exhibit 7-5: Estimated Incremental Costs vs. Revenue for Small CWSs - High Scenario*

Category

Source Water

Size Category

Number of
CWSs

Number of CWSs Number of CWSs Percent of CWSs Percent of CWSs
with Cost Revenue with Cost Revenue with Cost Revenue with Cost Revenue
Ratio > 1% Ratio > 3% Ratio > 1% Ratio > 3%

Private

Ground

<100

9,400

9,281

9,222

99%

98%

Private

Ground

101 to 500

8,190

7,387

6,192

90%

76%

Private

Ground

501 to 1,000

1,299

806

333

62%

26%

Private

Ground

1,001 to 3,300

1014

478

154

47%

15%

Private

Ground

3,301 to 10,000

336

177

131

53%

39%

Private

Surface

<100

404

397

391

98%

97%

Private

Surface

101 to 500

769

692

490

90%

64%

Private

Surface

501 to 1,000

232

124

44

53%

19%

Private

Surface

1,001 to 3,300

272

105

30

39%

11%

Private

Surface

3,301 to 10,000

182

102

69

56%

38%

Public

Ground

<100

1,409

1,393

1376

99%

98%

Public

Ground

101 to 500

4,838

4,544

3226

94%

67%

Public

Ground

501 to 1,000

2,869

2171

862

76%

30%

Public

Ground

1,001 to 3,300

4,488

2396

866

53%

19%

Public

Ground

3,301 to 10,000

2,459

1377

1051

56%

43%

Public

Surface

<100

518

512

499

99%

96%

Public

Surface

101 to 500

1,287

1190

727

92%

56%

Public

Surface

501 to 1,000

930

630

209

68%

22%

Public

Surface

1,001 to 3,300

2,193

976

278

45%

13%

Public

Surface

3,301 to 10,000

2,049

1157

843

56%

41%

Total





45,138

35,895

26,993

80%

60%

Acronyms: CWS = community water system.

Notes

* When evaluating the economic impacts on PWSs, the EPA uses the estimated PWS cost of capital to discount future costs (not the 2 percent discount rate used
to evaluate social costs and benefit), as this best represents the actual costs of compliance that water systems would incur over time. For more information on
cost of capital see Chapter 4, Section 4.2.3.3.

Final LCRI Economic Analysis

7-17

October 2024


-------
7.4.6.1 Alternative Small System Flexibility Option

The EPA assessed, but did not select, a second small system flexibility option. Like the selected final LCRI
small system flexibility option this alternative option would have mitigated the economic impact of an
ALE on small entities by allowing the PWSs to choose between (1) optimizing existing CCT or installing
new CCT or (2) installing and maintaining POU devices at all locations being served.212 This second small
system flexibility option would be available to all NTNCWSs and CWSs serving up to 10,000 people. This
option differs from the final LCRI requirements, which allow CWSs serving up to 3,300 people the choice
between the two compliance alternatives, by increasing the CWS size threshold so that systems serving
up to 10,000 people would have the ability to choose between the two regulatory compliance
alternatives. Under this alternative option, the EPA estimates that no additional CWSs serving between
3,301 to 10,000 people will elect to install and maintain POU devices.213

See Chapter 8, Section 8.8, for the estimated monetized annualized total costs and benefits of this
alternative size threshold (CWSs serving 10,000 or fewer people) compared with the final LCRI with the
small CWS threshold of 3,300 or fewer people.

7.5 Unfunded Mandates Reform Act

The UMRA (1995) seeks to protect State, local, and tribal governments from the imposition of unfunded
federal mandates. In addition, the Act seeks to strengthen the partnership among the federal
government and State, local, and Tribal governments.

Title II of UMRA establishes requirements for Federal agencies to assess the effects of their regulatory
actions on State, local, and Tribal governments and the private sector. Under section 202 of UMRA, the
EPA must prepare a written statement, including a cost-benefit analysis, for rules with "federal
mandates" that may result in expenditures by State, local, and tribal governments, in the aggregate, or
by the private sector, of $100 million or more in any one year, adjusted for inflation. The EPA has
calculated the cost of the rule in 2022 dollars, therefore, the UMRA requirements are triggered if
expenditures exceed $174 million in any one year.

Section 205 of UMRA requires the EPA to identify and consider a reasonable number of regulatory
alternatives and adopt the least costly, most cost-effective, or least burdensome option that achieves
the objectives of the rule. The provisions of section 205 do not apply when they are inconsistent with
applicable law. Moreover, section 205 allows the EPA to adopt an alternative other than the least costly,
most cost-effective, or least burdensome alternative if the Administrator publishes with the rule an
explanation why that alternative was not adopted. The EPA's analysis of alternative regulatory options,
presented in Chapter 8, is provided in Exhibit 7-6.

212	Refer to footnote 211 regarding the lead-bearing compliance alternative.

213	Note that actual model estimates provide for fractional system implementation. The model predicts that fewer
than 0.5 systems implement POU treatment, based on the modeling assumption of cost minimization, resulting in
a rounded number of zero systems implementing POU in the 3,301 to 10,000 system size category.

Final LCRI Economic Analysis

7-18

October 2024


-------
Exhibit 7-6: Summary of Alternative Other Options Considered for the Final LCRI

Alternative Option Considered

Difference in
Annual Net
Benefits of
Alternative vs.
Final Rule
(High Scenario, 2%
Discount Rate,
million 2022
Dollars)

EPA Reason(s) for Not Adopting
Alternative

Lead Action Level:





Lead AL of <0.015 mg/L

-$1,289.6

Lower net benefits driven mostly by
lower health benefits.

Lead AL of <0.005 mg/L

$2,325.3

The EPA is concerned with the technical
feasibility of PWS achieving an AL below
0.010 mg/L

Service Line Replacement Rate:





Service lines are replaced at an annual rate of

7%.

-$3,190.8

Lower net benefits driven mostly by
lower health benefits.

Definition of Lead Content to be Replaced:





Systems must replace lead service lines and
galvanized lines previously downstream of
lead lines or unknown lead content lines, and
lead connectors and galvanized lines
previously downstream of lead connectors.

$5,331.9

The EPA concerned about how these
activities might pull resources away from
the removal of lead and GRR service lines
that pose a greater exposure risk. Also,
due to very limited data on the reduction
in lead concentration associated with
removing lead connectors, the EPA used
the same concentration reductions seen
with partial SLR. This likely significantly
overestimates the benefits of replacing
lead connectors and makes the benefits
associated with lead connectors highly
uncertain.

SLR Deferred Deadline:





Systems may be given a deferred deadline for
finishing all LSL and GRR replacements
resulting in a maximum rate which is the
lower of 10,000 lines per year or 39
replacements per 1000 connections per year
(proposed rule-with change to connections
per year from households per year).

-$8.3

Lower net benefits driven mostly by
delay in providing health benefits.

Systems may be given a deferred deadline for
finishing all LSL and GRR replacements
resulting in a maximum rate which is the
lower of 8,000 lines per year or 39
replacements per 1000 connections per year.

-$65.0

Lower net benefits driven mostly by
delay in providing health benefits.

Lead Tap Sampling:





All systems return to standard 6-month

Very little

The EPA's monetized cost and benefit

Final LCRI Economic Analysis

7-19

October 2024


-------
Alternative Option Considered

Difference in
Annual Net
Benefits of
Alternative vs.
Final Rule
(High Scenario, 2%
Discount Rate,
million 2022
Dollars)

EPA Reason(s) for Not Adopting
Alternative

monitoring with an ALE. Systems with lead,
GRR, and/or unknown service lines at the
compliance date conduct standard 6-month
monitoring at the compliance date and non-
lead service line systems remain on LCR
monitoring schedule until new LCRI protocol
sampling may change P90. When (& if) a non-
lead system finds an LSL/GRR they return to
6-month monitoring, (proposed rule).

Systems that sampled using the new protocol
and are below the LCRI AL prior to the
compliance date may qualify to retain their
current schedule.

difference in
modelled costs and
benefits

estimates were too close to conclusively
determine if the alternative option or the
final LCRI has greater net benefits. The
EPA is concerned about the potentially
high volume of systems required to start
standard monitoring (especially small
systems), and the States' ability to
handle the increased demands. The EPA
considered a phased approach but
decided that the complexity of a phased
approach was not commensurate with
the benefits.

Multiple ALE Filter Programs:





Systems with at least 2 lead ALEs in a rolling
5-year period must prepare and submit a
filter plan to State. Systems with at least 3
lead ALEs in a rolling 5-year period must
make filters available to all customers with
lead, GRR, and unknown lead content service
lines.

The annual cost of
this option is $27.5
million lower than
the final rule.
Benefits are likely
to be the same as
the rule.

The EPA selected the final LCRI multiple
ALE option because it protects individuals
in systems with multiple ALEs that do not
have lead, GRR, or unknown service lines.
The monetized costs and benefit could
not be used in this case given a number
of known uncertainties. The EPA notes
that the estimated benefits of the LCRI in
this case are underestimated (given the
model does not account for benefits at
non-lead and GRR service line locations).
Costs are also likely overestimated for
both the alternative option and the final
LCRI, given an assumption of a 100%
filter pick-up rate. Because more
households are covered under the LCRI
costs the cost overestimate is greater for
the final rule.

Systems with at least 2 lead ALEs in a rolling
5-year period must prepare and submit a
filter plan to State. Systems with at least 3
lead ALEs in a rolling 5-year period must
deliver temporary filters directly to all
customers.

-$5.8

The EPA has feasibility concerns with this
option given the possible economic and
logistical challenges for systems. Also
note that lower net benefits are driven
by higher costs.

Small System Flexibility:





Final LCRI Economic Analysis

7-20

October 2024


-------
Alternative Option Considered

Difference in
Annual Net
Benefits of
Alternative vs.
Final Rule
(High Scenario, 2%
Discount Rate,
million 2022
Dollars)

EPA Reason(s) for Not Adopting
Alternative

CWSs that serve 10,000 or fewer people, and
all NTNCWSs, are provided compliance
flexibility when they exceed the AL.

$1.1

The EPA finds that the complexity of
implementing point-of-use filtration at all
residences in a system serving 3,300 to
10,000 individuals, or potentially 1,300 to
4,000 separate locations, cannot be
correctly captured in the estimated cost
structure within the economic model and
makes this option infeasible.

Acronyms: AL = action level; ALE = action level exceedance; CCT = corrosion control treatment; CWS = community
water system; GRR = galvanized requiring replacement; LCRI = Lead and Copper Rule Improvements; LSL = lead
service line; NTNCWS = non-transient, non-community water system; P90 = 90th percentile lead level; POU = point-
of-use; PWS = public water system; SLR = service line replacement.

Before the EPA establishes any regulatory requirements that may significantly or uniquely affect small
governments, including tribal governments, it must have developed under section 203 of UMRA a small
government agency plan. The plan must provide for notifying potentially affected small governments,
enabling officials of affected small governments to have meaningful and timely input in the
development of the EPA regulatory proposals with significant Federal intergovernmental mandates, and
informing, educating, and advising small governments on compliance with the regulatory requirements.

The final LCRI does contain a federal mandate that may result in expenditures to State, local, and Tribal
governments, in the aggregate, or to the private sector, of $174 million or more in any one year. Under
the low scenario, the highest annual incremental cost over the 35-year period of analysis is estimated to
happen in 2027. In 2027, publicly owned PWSs are expected to have undiscounted incremental costs of
$3.8 billion, privately owned PWSs are expected to have undiscounted incremental costs of $700
million, and States will have undiscounted incremental costs of $119 million. Under the high scenario,
the highest annual incremental cost over the 35-year period of analysis is estimated to happen in 2031.
In 2031, publicly owned PWSs are expected to have undiscounted incremental costs of $5.9 billion,
privately owned PWSs are expected to have undiscounted incremental costs of $875 million, and States
will have undiscounted incremental costs of $40 million. Therefore, the final LCRI is subject to the
requirements of sections 202 and 205 of UMRA. The EPA notes that the Federal government is providing
potential sources of funding to offset some of those direct compliance costs of the LCRI, including $15
billion as part of the BIL. However, the final rule's costs still exceed $174 million for a given year even
when considering currently available Federal funds.

Final LCRI Economic Analysis

7-21

October 2024


-------
The annualized incremental costs and benefits of the final LCRI, that are borne by public, private and
Tribal PWSs under the low and high scenarios are provided in Exhibit 7-7, and Exhibit 7-8 provides the
same information for small PWSs (10,000 or fewer people).214

As these exhibits show, public entities bare the vast majority of the costs, and their customers accrue
most of the benefits, of the final LCRI. In addition to these PWS costs, as discussed in Chapter 4 under
the final LCRI, States will incur annualized incremental administrative costs of $25.8 million to $27.7
million (2 percent discount rate). Finally, wastewater treatment plants, most of which are publicly
owned, will incur an incremental annualized cost of between $0.1 million and $0.3 million (2 percent
discount rate).

Exhibit 7-7: Estimated Total Annualized Incremental Costs and Benefits at 2 Percent Discount

Rate (millions of 2022 Dollars)

Type of System

Low Scenario

High Scenario

Public PWS Incremental Annualized Costs

$1,239.2

$1,690.2

Private PWS Incremental Annualized Costs

$206.3

$260.6

Tribal PWS Incremental Annualized Costs

$5.9

$7.2

Public PWS Incremental Annualized Benefits

$11,997.2

$22,386.3

Private PWS Incremental Annualized Benefits

$1,454.3

$2,680.0

Tribal PWS Incremental Annualized Benefits

$42.0

$76.4

Acronyms: PWS = public water system.

Note: Public systems include public-private partnerships. In addition, for the UMRA analysis, Federally
owned systems are excluded from the public costs.

Exhibit 7-8: Estimated Total Annualized Incremental Costs and Benefits
for Small PWSs (< 10,000 people) at 2 Percent Discount Rate (millions of 2022 Dollars)

Type of System

Low Scenario

High Scenario

Small Public PWS Incremental Annualized Costs

$169.7

$202.2

Small Private PWS Incremental Annualized Costs

$81.1

$88.2

Small Tribal PWS Incremental Annualized Costs

$3.8

$4.3

Small Public PWS Incremental Annualized Benefits

$1,173.0

$2,045.9

Small Private PWS Incremental Annualized Benefits

$231.2

$397.5

Small Tribal PWS Incremental Annualized Benefits

$19.5

$33.9

Acronyms: PWS = public water system.

Note: Public systems include public-private partnerships. In addition, for the UMRA analysis, Federally owned
systems are excluded from the public costs.

7.6 Executive Order 13132: Federalism

Executive Order 13132, Federalism (64 FR 43255, August 10, 1999), requires the EPA to develop an
accountable process to ensure "meaningful and timely input by state and local officials in the
development of regulatory policies that have federalism implications." "Policies that have federalism

214 For the UMRA analysis, a small PWS is defined as one that serves 10,000 or fewer people.

Final LCRI Economic Analysis

7-22

October 2024


-------
implications" are defined in the Executive Order to include regulations that have "substantial direct
effects on the states, on the relationship between the national government and the states, or on the
distribution of power and responsibilities among the various levels of government."

This action has federalism implications due to the substantial direct compliance costs on State or local
governments. The net change in regulatory implementation and oversite related cost to State, local, and
tribal governments in the aggregate is estimated to be between $25.8 and $27.7 million, in 2022 dollars,
at a 2 percent discount rate. However, the EPA notes that the federal government is providing a
potential source of funds to offset some of those direct compliance costs through the BIL.

To fulfill requirements of Executive Order 13132 section 6 (and UMRA), the EPA held a Federalism
consultation on October 13, 2022, with 15 organizations. These organizations representing State and
local governments had significant experience with intergovernmental relationships as well as expertise
in drinking water.215 During the meeting, the EPA presented background information and questions for
feedback on key areas of the final rule. The EPA specifically requested input on the following key rule
areas: achieving 100 percent LSLR, tap sampling and compliance, reducing rule complexity, and small
system flexibility. During the 60-day public comment period which followed the October 13, 2022
meeting, the EPA provided the members of the organizations present at the meeting, and those
contacted through the EPA Federalism consultation notification letter sent directly to State and local
government officials, the opportunity to provide input at requested follow-up Federalism meetings
and/or to the EPA docket. The EPA received requests for additional meetings and held meetings with the
Association of State Drinking Water Administrators and member States on October 5, 2022 and
November 2, 2022. A summary report of the views expressed during the Federalism consultation
meeting and in written submissions is available in the docket at EPA-HQ-OW-2022-0813 at
www.regulations.gov.

7.7 Executive Order 13175: Consultation and Coordination with Indian Tribal Governments

Executive Order 13175, Consultation and Coordination with Indian Tribal Governments (65 FR 67249,
November 9, 2000), requires the EPA to develop an accountable process to ensure "meaningful and
timely input by tribal officials in the development of regulatory policies that have tribal implications."
The Executive Order defines "policies that have tribal implications to include regulations that have
"substantial direct effects on one or more Indian tribes, on the relationship between the federal

215 Specifically, the EPA invited the following national organizations to the Federalism meeting: the National
Governor's Association, the National Conference of State Legislatures, the Council of State Governments, the
National League of Cities, the U.S. Conference of Mayors, the National Association of Counties, the International
City/County Management Association, the National Association of Towns and Townships, the Council of State
Governments, County Executives of America, and the Environmental Council of the States. The EPA also invited the
Association of State Drinking Water Administrators, the Association of Metropolitan Water Agencies, the National
Rural Water Association, the American Water Works Association, the Association of State and Territorial Health
Officials, the National Association of County and City Health Officials, the American Public Works Association, the
Association of Clean Water Administrators, the Western States Water Council, the African American Mayors
Association, the National Association of State Attorneys General, the Western Governors' Association, the National
School Board Association, the American Association of School Administrators, and the Council of the Great City
Schools to participate in the meeting.

Final LCRI Economic Analysis

7-23

October 2024


-------
government and the Indian tribes, or on the distribution of power and responsibilities between the
federal government and Indian tribes."

The final LCRI has tribal implications since it may impose substantial direct compliance costs on Tribal
governments, and the federal government will not provide the funds necessary to pay those costs.

There are 996 PWSs serving tribal communities, 87 of which are federally owned. This EA of the final
LCRI requirements estimated that the total annualized incremental costs placed on all systems serving
tribal communities ranges from $5.9 - $7.2million. The EPA notes that these estimated impacts will not
fall evenly across all Tribal systems. The final LCRI does offer regulatory relief by providing flexibility for
CWSs serving 3,300 or fewer people and all NTNCWSs to choose CCT, POU devices, and replacement of
lead-bearing materials to address lead in drinking water. This flexibility may result in LCRI
implementation cost savings for many tribal systems since 89 percent of tribal CWSs serve 3,300 or
fewer people and 16 percent of all tribal systems are NTNCWSs.

Consistent with the EPA Policy on Consultation and Coordination with Indian Tribes (May 4, 2011), the
EPA consulted with Tribal officials during the development of this action to gain an understanding of
Tribal views about potential revisions to key areas of the 2021 LCRR. Between October 6, 2022 and
December 9, 2022, the EPA consulted with tribal officials from federally recognized Indian Tribes
through the EPA's American Indian Environmental Office. The consultation included two webinars with
interested Tribes on October 27, 2022 and November 9, 2022, where the EPA provided an overview of
rulemaking information and requested input. The EPA requested input on four specific areas: achieving
100 percent LSLR, tap sampling and compliance, reducing rule complexity, and small system flexibility. A
total of 11 Tribal representatives participated in the two webinars. Webinar participants provided verbal
comments, but the EPA did not receive any written consultation comments from Tribal organizations
during the 60 -day comment period that followed the webinars. A summary report of the views
expressed during Tribal consultations is available in the docket at EPA-HQ-OW-2022-0801 at
www.regulations.gov.

7.8 Executive Order 13045: Protection of Children from Environmental Health and Safety
Risks

Executive Order 13045, Protection of Children from Environmental Health and Safety Risks (62 FR 19885,
April 23, 1997), applies to any rule initiated after April 21, 1998, that 1) is determined to be
"economically significant" as defined under Executive Order 12866; and 2) concerns an environmental,
health, or safety risk that the EPA has reason to believe may have a disproportionate effect on children.
If the regulatory action meets both criteria, the EPA must evaluate the environmental, health, or safety
effects of the planned rule on children, and explain why the planned regulation is preferable to other
potentially effective and reasonably feasible options considered by the EPA.

The final LCRI is subject to Executive Order 13045 because it is economically significant as defined in
Executive Order 12866, and based on the record, the EPA finds that the environmental health or safety
risk addressed by this final action would have a disproportionate effect on children. Additionally, the
agency's 2021 Policy on Children's Health (https://www.epa.gov/children/epas-policy-evaluating-risk-
children) is to protect children from environmental exposures by consistently and explicitly considering
early life exposures (from conception, infancy, early childhood, and through adolescence) and lifelong

Final LCRI Economic Analysis

7-24

October 2024


-------
health in all human health decisions through identifying and integrating data when conducting risk
assessments of children's health.

This action's health and risk assessments are contained in Chapter 5 and the associated Appendices D, E,
and F. The EPA expects that the final LCRI will provide additional protection to both children and adults
who consume drinking water supplied by systems. The EPA also finds that the benefits of the final LCRI,
including reduced health risk, will provide significant benefits to infants and young children due to
reducing exposure to lead in drinking water. This is due to the fact that developing fetuses, infants, and
young children are at higher risk for the adverse neurodevelopmental effects of lead than adolescents or
adults. These effects include, but are not limited to, decreases in cognitive function, as summarized in
Appendix D. This increased susceptibility is due to several factors, related to both physiology and levels
of exposure to lead during childhood. Physiological differences in neurodevelopment suggest that
infants and young children are at higher risk due to the susceptibility of the developing brain.
Additionally, there are physiological differences in lead absorption: given the same level of lead
exposure, infants, and young children will absorb more lead from the gastrointestinal tract than older
children or adults. Finally, there is also epidemiological evidence demonstrating that there are higher
lead exposures in infants and young children relative to older children or adults, which are attributable
to differences in behavior and diet.

It is important to note that the greater susceptibility in infants and young children does not minimize the
risks of lead exposures in adolescents or adults. Lead is associated with numerous adverse health effects
in these populations as well, including cardiovascular effects, immune system effects, and reproductive
and developmental effects which are also summarized in Appendix D. In addition, lead stored in the
bones of women from prior exposures can be mobilized from bone during pregnancy, leading to
subsequent increases in prenatal and postnatal lead exposures in children (via transfer from the
placenta and from breastmilk, respectively) (USEPA, 2013). It follows then that reductions in exposure to
women even prior to pregnancy will result in further protections for infants and children due to
decreases in exposure during pregnancy. For these reasons, lead exposures throughout the lifespan are
of concern to human health, and the developing fetus, infant and young children are the most
susceptible. Reducing lead exposures in drinking water will protect children from this increased risk.

See Chapter 6, where the EPA evaluated the environmental health or safety effects of lead found in
drinking water on children and estimated the risk reduction and health endpoint impacts to children
associated with the final LCRI requirements that reduce lead in drinking water including the installation
and re-optimization of CCT and the replacement of lead and GRR service lines.

7.9 Executive Order 13211: Actions That Significantly Affect Energy Supply, Distribution, or
Use

Executive Order 13211, Actions Concerning Regulations That Significantly Affect Energy Supply
Distribution, or Use (66 FR 28355, May 22, 2001), provides that agencies shall prepare and submit to the
Administrator of the OIRA, OMB, a Statement of Energy Effects for certain actions identified as
"significant energy actions." Section 4(b) of Executive Order 13211 defines "significant energy actions"
as "any action by an agency (normally published in the Federal Register) that promulgates or is expected
to lead to the promulgation of a final rule or regulation, including notices of inquiry, advance notices of
final rulemaking, and notices of final rulemaking: (l)(i) that is a significant regulatory action under

Final LCRI Economic Analysis

7-25

October 2024


-------
Executive Order 12866 or any successor order, and (ii) is likely to have a significant adverse effect on the
supply, distribution, or use of energy; or (2) that is designated by the Administrator of the Office of
Information and Regulatory Affairs as a significant energy action."

The final LCRI is not a "significant energy action" as defined in Executive Order 13211. This rule is a
significant regulatory action under Executive Order 12866 (see Section 7.2); however, it is not likely to
have a significant adverse effect on the supply, distribution, or use of energy, for the reasons described
as follows.

7.9.1	Energy Supply

The final LCRI does not regulate power generation, either directly or indirectly, and public and private
drinking water systems subject to the final LCRI do not, as a general rule, generate power. Further, the
energy cost increases borne by customers of systems as a result of the final LCRI is a low percentage of
the total cost of water. Therefore, power generation utilities that purchase water as part of their
operations are unlikely to face any significant effects as a result of the final LCRI.

7.9.2	Energy Distribution

The final LCRI does not regulate any aspect of energy distribution and drinking water systems that are
regulated by the final LCRI already have electrical service. The rule is not expected to increase peak
electricity demand at systems. Therefore, the EPA assumes that the existing connections are adequate
and that the final LCRI has no discernible adverse effect on energy distribution.

7.9.3	Energy Use

The EPA has determined that the incremental energy used to implement CCT at drinking water systems
in response to the final regulatory requirements is minimal. Therefore, the EPA does not expect any
noticeable effect at the national level on power generation in terms of average and peak loads.

7.10 National Technology Transfer and Advancement Act (NTTAA)

Section 12(d) of the NTTAA of 1995 directs the EPA to use voluntary consensus standards in its
regulatory activities unless to do so would be inconsistent with applicable law or otherwise impractical.
Voluntary consensus standards are technical standards (e.g., materials specifications, test methods,
sampling procedures, and business practices) that are developed or adopted by voluntary consensus
standards bodies. NTTAA directs the EPA to provide Congress, through OMB, explanations when the EPA
decides not to use available and applicable voluntary consensus standards.

The final LCRI may involve existing voluntary consensus standards in that it requires additional
monitoring for lead and copper. Monitoring and sample analysis methodologies are often based on
voluntary consensus standards. However, the final LCRI does not change any methodological
requirements for monitoring or sample analysis. The EPA's approved monitoring and sampling protocols
generally include voluntary consensus standards developed by agencies such as the American National
Standards Institute (ANSI) and other such bodies wherever the EPA deems these methodologies
appropriate for compliance monitoring. The EPA notes that in some cases, the final LCRI revises the
required frequency and number of samples taken.

Final LCRI Economic Analysis

7-26

October 2024


-------
7.11 Executive Order 12898: Federal Actions to Address Environmental Justice in Minority
Populations and Low-Income Populations and Executive Order 14096 (Revitalizing Our
Nation's Commitment to Environmental Justice for All)

The EPA anticipates the final LCRI will not create disproportionate and adverse human health or
environmental effects on communities with environmental justice (EJ) concerns under Executive Order
14096 (88 FR 25251, April 21, 2023); see also Executive Order 12898 (59 FR 7629, February 16, 1994).
The documentation for this finding, including additional details on the methodology, results, and
conclusions, are included in the EPA's Environmental Justice Analysis for the Proposed Lead and Copper
Rule Improvements Report (USEPA, 2023) and is available in the public docket for this action (EPA-HQ-
OW-2022-0801).

Executive Order 12898 established Federal executive policy on EJ. The main provision of Executive Order
12898 directs Federal agencies, to the greatest extent practicable and permitted by law, to make
achieving EJ part of their mission. Executive Order 12898 states "each Federal agency shall make
achieving environmental justice part of its mission by identifying and addressing as appropriate,
disproportionately high and adverse human health or environmental effects of their programs, policies,
and activities on minority populations and low-income populations in the United States and its
territories and possessions."

Executive Order 14096 directs the Federal government to build upon and strengthen its commitment to
deliver EJ to all communities across America through an approach that is informed by scientific research,
high-quality data, and meaningful Federal engagement with communities with EJ concerns.

Consistent with the agency's Technical Guidance for Assessing Environmental Justice in Regulatory
Analysis (USEPA, 2016), the EPA conducted an EJ analysis for the proposed LCRI to assess impacts
anticipated to result from the proposed LCRI (USEPA, 2023). The analysis builds on and advances the
analysis conducted under the 2021 LCRR, which evaluated baseline exposure to lead in drinking water.
The proposed LCRI's EJ analysis evaluated potential EJ concerns associated with lead in drinking water in
a baseline identified in the EJ analysis and the proposed LCRI, including consideration of whether
potential EJ concerns are created or mitigated by the proposed LCRI relative to the baseline of the
analysis. The EPA compiled recent peer-reviewed research on the relationship between lead exposure
and socioeconomic status and found that Black, Indigenous, and People of Color (BIPOC) and/or low-
income populations are at higher risk of lead exposure and associated health risks. The EPA also
conducted an analysis of seven case study cities further described below. The EPA selected the case
studies because they represented a range of system sizes and geographic regions.

Because updated service line inventories were not available for the EJ analysis conducted under the
2021 LCRR, the EPA used housing age as a proxy indicator for LSL presence in the EJ analysis for the
proposed 2021 LCRR216. In that EJ analysis, the EPA identified some trends indicating disproportionate

216 Housing vintage is an indicator for risk of LSLs, lead solder, and leaded brass fixtures (Rabin, 2008).
LSLs were installed through the 1980s, with decreases in the number of installations in the decades
following 1930. The EPA used Integrated Public Use Microdata Series (IPUMS) dataset to link individuals

Final LCRI Economic Analysis

7-27

October 2024


-------
and adverse human health risk for exposure to lead in drinking water based on LSL presence in minority
populations and low-income populations, and also that populations of children in minority households
and/or low-income households are disproportionately at risk of exposure to lead in drinking water
because they are more likely to live in housing built when LSLs were more commonly used.

Updated inventories were similarly not widely available yet; however, some water systems have
published updated inventories online. In the EJ analysis for the LCRI, the EPA evaluated service line
inventories from seven water systems with published inventories to estimate baseline exposure to lead
in drinking water using LSL presence as a proxy for lead exposure (USEPA, 2023). The EPA found a range
of outcomes with respect to the sociodemographic and housing unit variables in areas served by LSLs in
the cities investigated. While the EPA found that block groups with LSLs often had higher percentages of
low-income residents, renters, and people of color (specifically, Black, Hispanic, or linguistically isolated
individuals) compared to block groups without LSLs, there was little evidence that the number of LSLs
per capita was positively correlated with block group demographic characteristics across all seven case
studies. However, block groups with the highest number of LSLs per capita (top quartile) had a notably
larger percentage of Black residents than the service area as a whole for six case studies. Two other
measures (traffic density and pre-1960 housing) were included to capture the possibility of other
sources of lead. The analysis results showed that pre-1960 housing is notably higher in block groups with
LSLs compared to those without. The percent of housing built prior to 1960, which corresponds to a
higher likelihood of containing lead-based paint and LSL presence, was also positively correlated with
the number of LSLs per capita for every case study and was also elevated in the top quartile compared
to the service area as a whole. A separate EPA analysis also revealed that LSL prevalence in Cincinnati,
OH and Grand Rapids, Ml was a stronger predictor of the prevalence of elevated blood lead levels
compared with the EPA's EJScreen 2017 Lead Paint EJ Index or the U.S. Department of Housing and
Urban Development's Deteriorated Paint Index (Tornero-Velez et al., 2023).

Taken together, these findings support the concern that adverse health effects associated with baseline
lead exposure from LSLs may be inequitably distributed based on analysis of LSL presence. While the
limited number of water systems included in the analysis do not permit conclusions to be made about EJ
and LSL presence outside of the context of these individual systems, the analysis does point to several
findings. The analysis demonstrated significant differences in socioeconomic and housing characteristics
and the prevalence of LSLs across these systems. It also demonstrated the importance of considering
characteristics within the individual system context. Taken together, these findings support the concern
that adverse health effects associated with lead exposure from LSLs may be inequitably distributed with
respect to LSL presence.

Statistical analysis did not identify strong associations between LSLR and the characteristics of the
Census block group in which they occurred (e.g., socioeconomic and housing characteristics) in any of
the case studies. This is because, in general, at the time of the analysis either no LSLs or relatively few
LSLs have been removed in the locations of the case studies, which affects the EPA's ability to quantify a
relationship with LSLR. Conversely, in the case study of the water system in Newark, New Jersey, almost
all LSLs were removed in a short period of time, similarly obscuring the relationship between removals

to housing units by decennial age group beginning with housing built before 1939 and ending with
housing bult after 1980.

Final LCRI Economic Analysis

7-28

October 2024


-------
and the socioeconomic and housing unit variables. Nevertheless, the EPA recognizes the potential that
even in a water system where there are no EJ concerns with respect to LSL presence, the sequence and
timing in which LSLs and GRR service lines are replaced by a system's service line replacement program
can potentially create a concern. Section III.H of the preamble highlights the final LCRI provisions
intended to facilitate water system planning to prevent or minimize EJ concerns from being created
within the replacement program (, as well as other requirements that can make full replacements and
information more accessible to all customers. In Sections III.G and III.H of the preamble, the EPA also
highlights external funding available to support full service line replacement, as well as water systems'
obligations under Federal Civil Rights law.

On October 25, 2022, and November 1, 2022, the EPA held public meetings related to EJ and the
development of the proposed LCRI. The meetings provided an opportunity for the EPA to share
information and for individuals to offer input on EJ considerations related to the development of the
proposed LCRI and how to more equitably address lead in drinking water issues in their communities.

During the meetings and in subsequent written comments, the EPA received public comment on topics
including disproportionate exposure to lead and its health effects among BIPOC and low-income
communities; LSLR funding; methods to prioritize LSLR; access to LSLR for renters; filter distribution and
use during LSLR; lowering the lead AL; establishing a maximum contaminant level (MCL) for lead;
updating the lead health effects language required for public education, public notification, and the CCR;
ensuring that public education and public notification reaches communities that are most at risk; first -
and fifth-liter sampling; remediating lead identified through sampling in schools and child care facilities;
EJ concerns with corrosion control studies; community engagement; and regulatory enforcement and
oversight. For more information on the public meetings, please refer to the Public Meeting on
Environmental Justice (EJ) Considerations for the Development of the Proposed Lead and Copper Rule
Improvements (LCRI) Meeting Summary for each of the meeting dates in the public docket at
https://www.regulations.gov/docket/EPA-HQ-OW-2022-0801. Written public comments can also be
found in the docket. The EPA's Environmental Justice analysis is available in the public docket associated
with this rulemaking Docket ID Number: EPA-HQ-OW-2022-0801 available at
https://www.regulations.gov/document/EPA-HQ-OW-2022-0801-0689.

7.12 Consultations with the Science Advisory Board, National Drinking Water Advisory
Council, and the Secretary of Health and Human Services

7.12.1 Consultation with Science Advisory Board

As required by section 1412(e) of SDWA, in 2022, the EPA consulted with the SAB on the key areas being
considered for the proposed LCRI and tools, indicators, and measures for use in future analyses to
determine EJ impacts of LSL presence and replacement in drinking water systems. The EPA provided the
SAB with charge questions and shared the agency's preliminary analyses and draft results on case
studies for three cities to help inform the agency's EJ analysis for the proposed LCRI (USEPA, 2022). The
EPA charged the SAB with the following three questions:

• Are there potential environmental justice concerns associated with environmental stressors
affected by the regulatory action for population groups of concern in the baseline?

Final LCRI Economic Analysis

7-29

October 2024


-------
•	Are there potential environmental justice concerns associated with environmental stressors
affected by the regulatory action for population groups of concern for each regulatory option
under consideration?

•	For each regulatory option under consideration, are potential environmental justice concerns
created or mitigated compared to the baseline?

These questions asking the SAB to evaluate the potential EJ impacts of the proposed LCRI are in accord
with Executive Order 12898, which directs agencies to "identify and address the disproportionately high
and adverse human health or environmental effects of their actions on minority and low-income
populations" (59 FR 7629, February 16, 1994).

The EPA also sought an evaluation of the three of the agency's draft case studies for the proposed LCRI
EJ analysis. The EPA asked the SAB to evaluate the following four EJ issues: 1) the tools, indicators, and
metrics the EPA should consider when developing LSLR case studies; 2) whether a sub-set of variables
within these indices that should be given higher weights in the LCRI EJ assessment; 3) the
indicator/measure that is most suitable for studying the EJ impacts associated with LSLs and their
replacement; and 4) whether any of the tools or indicators under consideration for use in the LCRI EJ
assessment could help to better assess lead impacts from other co-located exposure pathways to inform
the EPA's understanding of lead exposures from non-drinking water sources.

The SAB deliberated and sought input from public meetings held on November 3 and 4, 2022. The SAB
provided initial verbal advice and comments on the proposed rule and case studies, as well as written
comments through November 21, 2022. The SAB provided its final report to the EPA Administrator on
December 20, 2022, regarding the agency's EJ analysis for LCRI (USEPA, 2022).

SAB members recommended using indicators from multiple tools (e.g., EJScreen, Center for Disease
Control and Prevention's (CDC's) Environmental Justice Index (EJ I), CDC/Agency for Toxic Substances and
Disease registry (ATSDR) Social Vulnerability Index (SVI), Area Deprivation Index (ADI)) in order to more
effectively identify communities that are disproportionately burdened by lead exposure and evaluate EJ
impacts of LSLs and LSLR. Recommended indicators for studying LSL and LSLR EJ impacts included
minority populations, low-income population, population under age 5, pre-1960 housing, pre-1980
housing, people with disabilities, single-parent households, occupied housing units without complete
plumbing, proximity to lead mines, hazardous waste proximity, superfund proximity, and particulate
matter (PM) 2.5. Some members also suggested that the EPA focus on indicators most relevant to
children, such as children under age 5, maternal education, birth weight, and quality of home
environment, because children are most sensitive to the effects of lead. Some members also suggested
giving higher weights to indicators that address populations disproportionately vulnerable to lead
exposure and its adverse health effects, such as population under 5 years old and low-income
communities, because they are more likely to consume tap water. Additional indicators suggested for
weighting were location based, including residential areas near legacy pollution sites. Recommended
tools to consider lead from other pathways included EJScreen, SVI, ADI, and EJI. Some SAB members also
recommended using proximity to traffic and pre-1960s housing, as these could indicate compound lead
exposure from pathways other than drinking water.

As a result of the consultation, the EPA incorporated the suggestions from the SAB in a study of the EJ
implications of the LCRI (USEPA, 2023). The EPA evaluated correlations between per capita LSLs (in a

Final LCRI Economic Analysis

7-30

October 2024


-------
Census block group) and different ethnic groups including American Indian or Alaska Native, Asian or
Pacific Islander, Other or Two Races, Hispanic, Non-Hispanic Black, and Non-Hispanic white. The EPA
also evaluated the relationship between the presence of LSLs and indicators representing the
populations most at risk of lead exposure, such as low income and children under age 5. Indicators
addressing characteristics that are associated with exposure to other lead sources were also
incorporated in the study including structures built prior to 1960 and proximity to traffic. Additional
information on SAB recommendations is included in the SAB report (USEPA, 2022) available in the
docket EPA-HQ-OW-2022-0801 at www.regulations.gov.

7.12.2	Consultation with National Drinking Water Advisory Council (NDWAC)

The NDWAC is a Federal Advisory Committee that supports the EPA in performing its duties and
responsibilities related to the national drinking water program and was created through a provision in
SDWA in 1974. In accordance with section 1412(d) of SDWA, the EPA consulted with the NDWAC on
both the proposed and final LCRI. These consultations are further described in this section.

On November 30, 2022, the EPA held a public teleconference with NDWAC during which the EPA
presented the proposed LCRI and solicited input from the NDWAC. The EPA provided background on
lead in drinking water and the LCR, an overview of the 2021 LCRR published in January 2021, annualized
cost estimates from the 2021 LCRR EA, and a summary of the outcome of the EPA's review of the 2021
LCRR. The NDWAC provided key input on four key areas: service line replacement, tap sampling and
compliance, reducing rule complexity, and small system flexibility. The public was also given an
opportunity to provide their comments to the NDWAC. A summary of the NDWAC consultation is
available in the NDWAC, Fall 2022 Meeting Summary Report (NDWAC, 2022) and in the docket EPA-HQ-
OW-2022-0801 at www.regulations.gov.

On January 31, 2024, the EPA held a public teleconference to consult with the NDWAC on the final LCRI.
The EPA provided an overview of the proposed LCRI as well as key revisions in the proposed rule. The
public was also given an opportunity to provide their comments to the NDWAC. A summary of the
NDWAC meeting, the public comments to the NDWAC, and the EPA's presentation are available in the
NDWAC Summary Report (NDWAC, 2024) and is also available in the docket.

7.12.3	Consultation with Health and Human Services

In accordance with section 1412(d) of SDWA, on August 18, 2023, the EPA consulted with the
Department of Health and Human Services (HHS) on the proposed LCRI and on July 15, 2024, the EPA
consulted with the HHS on the final rule. The EPA provided information to HHS officials on both the
proposed LCRI and the draft final LCR. The EPA received and considered comments from the HHS for
both the proposal and final rule through the interagency review process under Executive Order 12866.

7.13 References

Executive Order 12866. 1993. Regulatory Planning and Review. Federal Register. 58 FR 51735. October
4, 1993. Available at https://www.reginfo.gov/public/isp/Utilities/EO 12866.pdf.

Final LCRI Economic Analysis

7-31

October 2024


-------
Executive Order 12898. 1994. Federal Actions to Address Environmental Justice in Minority Populations
and Low-Income Populations. Federal Register. 59 FR7629. February 16, 1994. Available at
https://www.gpo.gov/fdsvs/pkg/FR-1994-02-16/html/94-3685.htm.

Executive Order 13045. 1997. Protection of Children from Environmental Health Risks and Safety Risks.
Federal Register. 62 FR 19885. April 23, 1997. Available at https://www.gpo.gov/fdsvs/pkg/FR-1997-04-
23/pdf/97-10695.pdf.

Executive Order 13132. 1999. Federalism. Federal Register. 64 FR 43255. August 10, 1999. Available at
https://www.gpo.gov/fdsys/pkg/FR-1999-08-10/pdf/99-20729.pdf.

Executive Order 13175. 2000. Consultation and Coordination with Indian Tribal Governments. Federal
Register. 65 FR 67249. November 9, 2000. Available at https://www.gpo.gov/fdsys/pkg/FR-2000-ll-
09/pdf/00-29003.pdf.

Executive Order 13211. 2001. Actions Concerning Regulations That Significantly Affect Energy Supply,
Distribution, or Use. Federal Register. 66 FR 28355. May 22, 2001. Available at
https://www.gpo.gov/fdsvs/pkg/FR-2001-05-22/pdf/01-13116.pdf.

Executive Order 13563. 2011. Improving Regulation and Regulatory Review. Federal Register. 76 FR
3821. January 21, 2011. Available at https://www.gpo.gov/fdsys/pkg/FR-2011-01-21/pdf/2011-
1385.pdf.

Executive Order 14094. 2023. Modernizing Regulatory Review. Federal Register. 88 FR 21879. April 11,
2023. Available at https://www.govinfo.gov/content/pkg/FR-2023-04-ll/pdf/2023-0776Q.pdf

Executive Order 14096. 2023. Revitalizing Our Nation's Commitment to Environmental Justice for All.
Federal Register. 88 FR25251. April 26, 2023. Available at

https://www.energv.gov/sites/default/files/2023-04/eo-14Q96-revitalizing-commitment-to-
environmental-justice.pdf.

National Drinking Water Advisory Council (NDWAC). 2022. U.S. Environmental Protection Agency's
National Drinking Water Advisory Council Public Meeting November 30, 2022: Meeting Summary
https://www.epa.gov/ndwac/national-drinking-water-advisorv-council-meeting-summarv-november-
30-2022-0

NDWAC. (2024). National Drinking Water Advisory Council Meeting Summary, January 31, 2024.
Retrieved from https://www.epa.gov/system/files/documents/2024-06/ndwac-meeting-summarv-
ianuarv-2024-508 l.pdf

National Technology Transfer and Advancement Act of 1995. Public Law 104-113. 104th Congress. 110
Stat. 783. March 7, 1996. Available at https://www.gpo.gov/fdsys/pkg/PLAW-104publll3/pdf/PLAW-
104publll3.pdf.

Paperwork Reduction Act, as amended. 2015. 44 (JSC 3501-44 (JSC 3521. Available at

https://www.gpo.gov/fdsys/pkg/USCODE-2015-title44/pdf/USCODE-2015-title44-chap35-subchapl-

sec3501.pdf.

Rabin, R. 2008. The lead industry and lead water pipes "A modest campaign." American Journal of Public
Health 98(9): 15840-1592. doi: 10.2105/AJPH.2007.113555.

Final LCRI Economic Analysis

7-32

October 2024


-------
Regulatory Flexibility Act of 1980, as amended. Available at 5 (JSC 601-et seq. Available at
https://www.sba.gov/advocacv/regulatorv-flexibility-act.

Safe Drinking Water Act. 42 U.S.C. 300g et seq. 2015. Available at

https://www.gpo.gov/fdsys/pkg/USCODE-2015-title42/pdf/USCODE-2015-title42-chap6A-
subchapXll.pdf.

Safe Drinking Water Act Amendments of 1996. Public Law 104-182. 104th Congress. 110 Stat. 1613.
August 6, 1996. Available at https://www.congress.gov/104/plaws/publl82/PLAW-104publl82.pdf.

Tornero-Velez, R., Christian, M., V. Zartarian, and K. R. Simoneau (2023). Tapping into Lead Service Line
Information: Two City Case Study: Lead Data Mapping: Methods and Tools for Lead Prioritization,
Prevention, and Mitigation. Presented at the National Environmental Health Association Annual
Education Conference & Exhibition, July 31-August 3, New Orleans.

Unfunded Mandates Reform Act of 1995. Public Law 104-4. 104th Congress. 110 Stat. 48. March 22,
1995. Available at https://www.gpo.gov/fdsys/pkg/PLAW-104publ4/pdf/PLAW-104publ4.pdf.

United States Environmental Protection Agency (USEPA). 1998a. National Primary Drinking Water
Regulations: Consumer Confidence Report. Proposed Rule. Federal Register. 63 FR 7606. February 13,
1998. Available at https://www.govinfo.gov/content/pkg/FR-1998-02-13/pdf/98-3752.pdf.

USEPA. 1998b. National Primary Drinking Water Regulations: Consumer Confidence Reports; Final Rule.
Federal Register. 63 FR 44512. August 19, 1998. Available at https://www.govinfo.gov/content/pkg/FR-
1998-08-19/pdf/98-22056.pdf.

USEPA. 2008. Assessment of the Vulnerability of Noncommunity Water Systems to SDWA Cost Increases.

USEPA. 2009. Community Water System Survey Volume II: Detailed Tables and Survey Methodology. May
2009. Office of Water. EPA 815-R-09-002. Available at

https://nepis.epa.gov/Exe/ZvPDF.cgi/P1009USA.PDF?Dockev=P1009USA.PDF.

USEPA. 2011. Re-Energizing the Capacity Development Program: Findings and Best Practices from the
Capacity Development Re-Energizing Workgroup. April 2011. Office of Water. EPA 816-R-11-004.
https://nepis.epa.gov/Exe/ZvPDF.cgi/P100M EY5.PDF?Dockev=P100MEY5. PDF.

USEPA. 2013. Integrated Science Assessment for Lead. June 2013. Office of Research and Development.
EPA/600/R-10/075F. Available at https://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=255721.

USEPA. 2016. Technical Guidance for Assessing Environmental Justice in Regulatory Analysis. June 2016.
Available at https://www.epa.gov/sites/default/files/2016-06/documents/eitg 5 6 16 v5.1.pdf.

USEPA. 2021. National Primary Drinking Water Regulations: Lead and Copper Rule Revisions. Final Rule.
Federal Register. 86 FR 4198. January 15, 2021. Available at https://www.govinfo.gov/content/pkg/FR-
2021-01-15/pdf/2020-28691.pdf.

USEPA. 2022. Consultation on Environmental Justice Analysis for EPA's Lead and Copper Rule
Improvements. From Alison CCullen, Sc. D Chair to EPA Administrator Michael S. Regan. Science
Advisory Board. EPA-SAB-23_003. December 20, 2022.

Final LCRI Economic Analysis

7-33

October 2024


-------
https://sab.epa.g0v/0rds/sab/r/sab apex/sab/advisorvactivitvdetail?pl8 id=2628&clear=RP.18&sessio
n=11133043673738#report

USEPA. 2023. Environmental Justice Analysis for the Proposed Lead and Copper Rule Improvements
Report. Office of Water. EPA 815-R-23-004. November 2023.

USEPA. 2024. National Primary Drinking Water Regulations: Lead and Copper Rule Improvements. Final
Rule.

Final LCRI Economic Analysis

7-34

October 2024


-------
8 Other Options Considered

8.1 Introduction

This chapter presents alternative options the United States Environmental Protection Agency (EPA)
considered when developing the final Lead and Copper Rule Improvements (LCRI) related to: the
required lead action level (AL); the service line replacement (SLR) rate; the definition of lead content to
be replaced as part of the SLR program; the potential for deferred deadlines under the SLR program;
changes to the lead tap sampling schedule; the temporary filter requirements under a multiple lead
action level exceedance (ALE) program; and the size threshold of the small system compliance flexibility.
Due to the large number of alternative options considered, this analysis uses the high scenario
assumptions to illustrate how their monetized benefits and costs compare to those of the final LCRI.

Also note that EPA has feasibility concerns with the implementation of some of the alternative options
analyzed which raises the level of uncertainty associated with the estimated cost and benefit values for
those alternatives. The agency has noted in the following subsections the alternative options impacted
by feasibility concerns. Exhibit 8-1 provides a detailed summary of the final LCRI requirements and the
alternative options considered.

Exhibit 8-1: Summary of Alternative Other Options Considered for the Final LCRI

Area

Alternative Option Considered

Final LCRI

Lead Action Level

1.	Lead AL of <0.015 mg/L

2.	Lead AL of <0.005 mg/L

Lead AL of <0.010 mg/L (proposed rule)

Service Line
Replacement Rate

Service lines are replaced at an annual
rate of 7%

Service lines are replaced at an annual
rate of 10% (proposed rule)

Definition of Lead
Content to be Replaced

Systems must replace lead service lines
and galvanized lines previously
downstream of lead lines or unknown
lead content lines, and lead connectors
and galvanized lines previously
downstream of lead connectors

Systems must replace lead service lines
and galvanized lines previously
downstream of lead lines or unknown
lead content lines. Lead connectors are
replaced when encountered (proposed
rule)

SLR Deferred Deadline

1.	Systems may be given a deferred
deadline for finishing all LSL and GRR
replacements resulting in a maximum
rate which is the lower of 10,000 lines
per year or 39 replacements per 1000
connections per year (proposed rule-
with change to connections per year
from households per year)

2.	Systems may be given a deferred
deadline for finishing all LSL and GRR
replacements resulting in a maximum
rate which is the lower of 8,000 lines
per year or 39 replacements per 1000
connections per year

Systems may be given a deferred
deadline for finishing all LSL and GRR
replacements resulting in a maximum
rate of 39 replacements per 1000
connections per year

Final LCRI Economic Analysis

8-1

October 2024


-------
Area

Alternative Option Considered

Final LCRI

Lead Tap Sampling

All systems return to standard 6-month
monitoring with an ALE. Systems with
lead, GRR, and/or unknown service lines
at the compliance date conduct
standard 6-month monitoring at the
compliance date and non-lead service
line systems remain on LCR monitoring
schedule until new LCRI protocol
sampling may change P90. When (& if) a
non-lead system finds an LSL/GRR they
return to 6-month monitoring,
(proposed rule). Systems that sampled
using the new protocol and are below
the LCRI AL prior to the compliance date
may qualify to retain their current
schedule.

All systems return to standard 6-month
monitoring with an ALE. Systems with
lead and GRR service lines return to
standard 6-month monitoring at
compliance date. Unknown and non-lead
systems remain on LCR monitoring
schedule until new LCRI protocol
sampling may change P90. When (& if) a
non-lead/all unknown system finds an
LSL/GRR they return to 6-month
monitoring. Systems with lead and GRR
service lines that sampled using the new
protocol and are below the LCRI AL prior
to the compliance date may qualify to
retain their current schedule.

Multiple ALE Filter
Programs

Systems with at least 2 lead ALEs in a
rolling 5-year period must prepare and
submit a filter plan to State. Systems
with at least 3 lead ALEs in a rolling 5-
year period must:

1.	Make filters available to all customers
with lead, GRR, and unknown lead
content service lines

2.	Deliver temporary filters directly to
all customers

Systems with at least 2 lead ALEs in a
rolling 5-year period must prepare and
submit a filter plan to State. Systems
with at least 3 lead ALEs in a rolling 5-
year period must make filters available to
all customers (proposed rule-with filter
plan being required after 2 ALEs instead
of 3 ALEs for the final rule)

Small System Flexibility

CWSs that serve 10,000 or fewer
people, and all NTNCWSs, are provided
compliance flexibility when they exceed
the AL

CWSs that serve 3,300 or fewer people,
and all NTNCWSs, are provided
compliance flexibility when they exceed
the AL (proposed rule)

Acronyms: AL = action level; ALE = action level exceedance; CWS = community water system; GRR = galvanized
requiring replacement; LCR= Lead and Copper Rule; LCRI = Lead and Copper Rule Improvements; LSL= lead service
line; NTNCWS = non-transient non-community water system; P90 = lead 90th percentile level; SLR = service line
replacement.

Note: (Proposed Rule) indicates if a final rule component or alternative option were originally considered as part of
the proposed LCRI.

8.2 Alternative Lead Action Levels

The EPA's final LCRI set the AL at 0.010 mg/L. The agency, as part of the final rule development process,
also considered two alternative lead ALs, <0.005 mg/L or <0.015 mg/L.

Final LCRI Economic Analysis

8-2

October 2024


-------
Exhibit 8-2 and Exhibit 8-3 compare the quantified costs and benefits of the final LCRI to the quantified
costs and benefits at an AL of 0.015 mg/L holding all other final LCRI rule requirements constant. Results
in these tables are provided for the high scenario at a 2 percent discount rate.217

217 Note the following for all cost results in this chapter: All cost tables provide estimates for household cost of
LSLR under the 2021 Lead and Copper Rule Revisions (LCRR) baseline but this value is not computed for the LCRI.
The EPA in the 2021 LCRR economic analysis (USEPA, 2020) assumed that the cost of customer-side SLRs made
under the goal-based replacement requirement would be paid for by households. The agency also assumed that
system-side SLRs under the goal-based replacement requirement and all SLRs (both customer-side and systems-
side) would be paid by the public water system (PWS) under the 3 percent mandatory replacement requirement.
The EPA made these modeling assumptions based on the different levels of regulatory responsibility systems faced
operating under a goal-based replacement requirement versus a mandatory replacement requirement. While
systems would not be subject to a potential violation for not meeting the replacement target under the goal-based
replacement requirement, under the 3 percent mandatory replacement requirement the possibility of a violation
could motivate more systems to meet the replacement target even if they had to adopt customer incentive
programs that would shift the cost of replacing customer-side service lines from customers to the system. The EPA
cannot require such incentive programs within a National Primary Drinking Water Regulation (NPDWR). To be
consistent with these 2021 LCRR modeling assumptions, under the final LCRI, the EPA assumed that mandatory
replacement costs would fall only on systems. Therefore, the negative incremental values reported for the
"Household SLR Costs" category do not represent a net cost savings to households. They represent an assumed
shift of the estimated SLR costs from households to systems. Note, however, that systems might pass along these
costs to rate payers. The EPA has insufficient information to estimate the actual SLR cost sharing relationship
between customers and systems at the national level of analysis.

Final LCRI Economic Analysis

8-3

October 2024


-------
Exhibit 8-2: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and
Alternative Lead Action Level Option (AL < 0.015 mg/L) - High Scenario - 2 Percent Discount

Rate (millions of 2022 USD)





Final Rule



Alternative Option (AL <

D.015 mg/L)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS Annual Costs













Sampling

$143.6

$176.2

$32.6

$143.6

$168.1

$24.5

PWS SLR

$124.5

$1,763.9

$1,639.4

$124.5

$1,765.2

$1,640.7

Corrosion Control
Technology
Point-of Use
Installation and
Maintenance
Public Education and
Outreach

$647.8
$5.9
$72.1

$692.9
$9.6
$302.2

$45.1
$3.7
$230.1

$647.8
$5.9
$72.1

$621.1
$5.6
$274.7

-$26.7
-$0.3
$202.6

Rule Implementation
and Administration

$0.2

$3.4

$3.2

$0.2

$3.4

$3.2

Total Annual PWS
Costs

$994.1

$2,948.2

$1,954.1

$994.1

$2,838.1

$1,844.0

Household SLR Costs

$26.4

$0.0

-$26.4

$26.4

$0.0

-$26.4

State Rule
Implementation and
Administration
Wastewater
Treatment Plant
Costs

$41.8
$4.8

$67.6
$5.1

$25.8
$0.3

$41.8
$4.8

$66.2
$3.3

$24.4
-$1.5

Total Annual Rule
Costs

$1,067.1

$3,020.9

$1,953.8

$1,067.1

$2,907.6

$1,840.5

Acronyms: AL = action level; LCRI = Lead and Copper Rule Improvements; PWS = public water system; SLR = service
line replacement; USD = United States dollar.

Notes: (1) Previous Baseline costs are projected over the 35-year period of analysis and are affected by the EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

(2) Very small differences in results between the final rule and the regulatory option are due to inter-run variability
in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences between the costs
and benefits of the final rule and the alternative option.

Final LCRI Economic Analysis

8-4

October 2024


-------
Exhibit 8-3: Estimated National Annual Benefit Comparison Between the Final LCRI and
Alternative Lead Action Level Option (AL < 0.015 mg/L) - High Scenario - 2 Percent Discount

Rate (millions of 2022 USD)





Final Rule



Alternative Option (AL <

0.015 mg/L)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual IQ Benefits

$3,279.0

$10,963.0

$7,684.0

$3,279.0

$10,586.0

$7,307.0

Annual Low-Birth Weight
Benefits

00

t-H

-oo-

$5.7

$3.9

00

t-H

-oo-

$5.5

$3.7

Annual ADHD Benefits

$179.9

$599.5

$419.6

$179.9

$580.4

$400.5

Annual Adult CVD Premature
Mortality Benefits

$8,174.9

$25,210.0

$17,035.1

$8,174.9

$24,203.4

$16,028.5

Total Annual Benefits

$11,635.6

$36,778.2

$25,142.6

$11,635.6

$35,375.3

$23,739.7

Acronyms: ADHD = attention-deficit/hyperactivity disorder; AL = action level; CVD = cardiovascular disease; IQ =
intelligence quotient; LCRI = Lead and Copper Rule Improvements; USD = United States dollar.

Note: Very small differences in results between the final rule and the regulatory option are due to inter-run
variability in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences
between the costs and benefits of the final rule and the alternative option.

Exhibit 8-4 and Exhibit 8-5 compare the quantified costs and benefits of the final LCRI to the quantified
costs and benefits at an AL of 0.005 mg/L holding all other final LCRI rule requirements constant. Results
in these tables are provided for the high scenario at a two percent discount rate. Note that the
estimated results for the alternative option, which assumes water systems can achieve lead levels below
a lead AL of < 0.005 mg/L is feasible, must be viewed as having a higher degree of uncertainty. Although
the EPA has adjusted ALE data that allows for the calculation of the cost and benefits of this alternative,
the agency has concerns about the feasibility of implementing this option. See section IV.F.4 of the
Federal Register Notice (USEPA, 2024) for a detailed discussion of the lead AL and its function to support
the feasibility of the CCT treatment technique.

Final LCRI Economic Analysis

8-5

October 2024


-------
Exhibit 8-4: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and
Alternative Lead Action Level Option (AL < 0.005 mg/L) - High Scenario - 2 Percent Discount

Rate (millions of 2022 USD)





Final Rule



Alternative Option (AL < 0.005
mg/L)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS Annual Costs













Sampling

$143.6

$176.2

$32.6

$143.6

$198.7

$55.1

PWS SLR

$124.5

$1,763.9

$1,639.4

$124.5

$1,762.4

$1,637.9

Corrosion Control Technology

$647.8

$692.9

$45.1

$647.8

$819.4

$171.6

Point-of Use Installation and
Maintenance

$5.9

$9.6

$3.7

$5.9

$15.7

$9.8

Public Education and Outreach

$72.1

$302.2

$230.1

$72.1

$374.2

$302.1

Rule Implementation and
Administration

$0.2

$3.4

$3.2

$0.2

$3.6

$3.4

Total Annual PWS Costs

$994.1

$2,948.2

$1,954.1

$994.1

$3,174.0

$2,179.9

Household SLR Costs

$26.4

$0.0

-$26.4

$26.4

$0.0

-$26.4

State Rule Implementation and
Administration

$41.8

$67.6

$25.8

$41.8

$71.7

$29.9

Wastewater Treatment Plant
Costs

$4.8

1

LO

$0.3

$4.8

$8.2

$3.4

Total Annual Rule Costs

$1,067.1

$3,020.9

$1,953.8

$1,067.1

$3,253.9

$2,186.8

Acronyms: AL = action level; LCRI = Lead and Copper Rule Improvements; PWS = public water system; SLR = service
line replacement; USD = United States dollar.

Notes: (1) Previous Baseline costs are projected over the 35-year period of analysis and are affected by the EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

(2) Very small differences in results between the final rule and the regulatory option are due to inter-run variability
in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences between the costs
and benefits of the final rule and the alternative option.

Final LCRI Economic Analysis

8-6

October 2024


-------
Exhibit 8-5: Estimated National Annual Benefit Comparison Between the Final LCRI and
Alternative Lead Action Level Option (AL < 0.005 mg/L) - High Scenario - 2 Percent Discount

Rate (millions of 2022 USD)



Final Rule



Alternative Option (AL < 0.005 mg/L)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual IQ
Benefits

$3,279.0

$10,963.0

$7,684.0

$3,279.0

$11,651.2

$8,372.2

Annual Low-Birth
Weight Benefits

00
Šuy

$5.7

$3.9

00
Šuy

$6.0

$4.2

Annual ADHD
Benefits

$179.9

$599.5

$419.6

$179.9

$634.9

$455.0

Annual Adult
CVD Premature
Mortality
Benefits

$8,174.9

$25,210.0

$17,035.1

$8,174.9

$27,044.4

$18,869.5

Total Annual
Benefits

$11,635.6

$36,778.2

$25,142.6

$11,635.6

$39,336.5

$27,700.9

Acronyms: ADHD = attention-deficit/hyperactivity disorder; AL = action level; CVD = cardiovascular disease; IQ =
intelligence quotient; LCRI = Lead and Copper Rule Improvements; USD = United States dollar.

Note: Very small differences in results between the final rule and the regulatory option are due to inter-run
variability in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences
between the costs and benefits of the final rule and the alternative option.

The EPA identified 0.010 mg/L as being generally representative of OCCT based on updated data and
over 30 years of LCR implementation experience (see section IV.F.4 of the Federal Register notice for a
discussion on the action level analysis). In selecting this action level, the EPA considered what is
technically possible for small and medium systems in light of the identified challenges that still exist,
including their fewer resources and more limited technical capacity compared to large systems and a
limited number of CCT experts available nationally. Therefore, the EPA has determined that an action
level of 0.010 mg/L would support the treatment technique for CCT overall, in addition to other
elements of this treatment technique, and is the most health protective level technically possible; it thus
meets the feasibility standard at SDWA section 1412(b)(7)(A).

Given the concerns over feasibility and therefore the uncertainty associated with the estimated costs
and benefits of this alternative option, the EPA is discounting the fact that estimated net benefits for
this alternative option are greater than the estimated net benefits for the final LCRI. The LCRI maintains
the lead action level at < 0.010 mg/L.

8.3 Alternative Service Line Replacement Rate

The final LCRI sets the required SLR rate at 10 percent per year, unless subject to a shortened or
deferred deadline. The agency as part of the proposal development process also considered an
alternative SLR rate, 7 percent per year.

Final LCRI Economic Analysis

8-7

October 2024


-------
Exhibit 8-6 and Exhibit 8-7 compare the quantified costs and benefits of the final LCRI to the quantified
costs and benefits of the rule with an alternative SLR rate of 7 percent, holding all other rule
requirements constant. Results are provided for the high scenario at a 2 percent discount rate.

Exhibit 8-6: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and
Alternative Service Line Replacement Option (SLR Rate = 7%) - High Scenario - 2 Percent

Discount Rate (millions of 2022 USD)





Final Rule



Alternative Option (SLR Rate = 7%)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS Annual Costs













Sampling

$143.6

$176.2

$32.6

$143.6

$176.1

$32.5

PWS SLR

$124.5

$1,763.9

$1,639.4

$124.5

$1,672.2

$1,547.7

Corrosion Control Technology

$647.8

$692.9

$45.1

$647.8

$696.0

$48.2

Point-of Use Installation and
Maintenance

$5.9

$9.6

$3.7

$5.9

$10.2

$4.3

Public Education and Outreach

$72.1

$302.2

$230.1

$72.1

$341.0

$268.9

Rule Implementation and
Administration

$0.2

$3.4

$3.2

$0.2

$3.4

$3.2

Total Annual PWS Costs

$994.1

$2,948.2

$1,954.1

$994.1

$2,898.9

$1,904.8

Household SLR Costs

$26.4

$0.0

-$26.4

$26.4

$0.0

-$26.4

State Rule Implementation and
Administration

$41.8

$67.6

$25.8

$41.8

$67.7

$25.9

Wastewater Treatment Plant
Costs

$4.8

$5.1

$0.3

$4.8

$5.2

$0.4

Total Annual Rule Costs

$1,067.1

$3,020.9

$1,953.8

$1,067.1

$2,971.8

$1,904.7

Acronyms: LCRI = Lead and Copper Rule Improvements; PWS = public water system; SLR = service line
replacement; USD = United States dollar.

Notes: (1) Previous Baseline costs are projected over the 35-year period of analysis and are affected by EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

(2) Very small differences in results between the final rule and the regulatory option are due to inter-run variability
in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences between the costs
and benefits of the final rule and the alternative option.

Final LCRI Economic Analysis

8-8

October 2024


-------
Exhibit 8-7: Estimated National Annual Benefit Comparison Between the Final LCRI and
Alternative Service Line Replacement Option (SLR Rate = 7%) - High Scenario - 2 Percent

Discount Rate (millions of 2022 USD)





Final Rule



Alternative Option (SLR Rate = 7%)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual IQ Benefits

$3,279.0

$10,963.0

$7,684.0

$3,279.0

$9,994.8

$6,715.8

Annual Low-Birth Weight
Benefits

00
Šuy

$5.7

$3.9

00
Šuy

$5.2

$3.4

Annual ADHD Benefits

$179.9

$599.5

$419.6

$179.9

$540.5

$360.6

Annual Adult CVD Premature
Mortality Benefits

$8,174.9

$25,210.0

$17,035.1

$8,174.9

$22,997.8

$14,822.
9

Total Annual Benefits

$11,635.6

$36,778.2

$25,142.6

$11,635.6

$33,538.3

$21,902.
7

Acronyms: ADHD = attention-deficit/hyperactivity disorder; CVD = cardiovascular disease; IQ = intelligence

quotient; LCRI = Lead and Copper Rule Improvements; SLR = service line replacement USD = United States dollar.
Note: Very small differences in results between the final rule and the regulatory option are due to inter-run
variability in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences
between the costs and benefits of the final rule and the alternative option.

8.4 Alternative Definition of Lead Content Service Lines to Be Replaced

The final LCRI requires that systems replace lead connectors as they are encountered but does not
include these lead connectors or galvanized lines previously downstream of lead connectors as part of
the 10 percent of lead content lines that must be replaced annually. The EPA as part of the final rule
development process also considered an alternative definition of lead content service lines that are
required to be replaced. This alternative definition included lead service lines (LSLs) and galvanized
service lines downstream from a LSL or unknown lead content service line but also required the
replacement of lead connectors and galvanized lines downstream from lead connectors. The unit cost
for replacing a galvanized line downstream of a lead connector is the same as replacing a galvanized
requiring replacement (GRR) service line. For the unit cost of replacing a lead connector, the EPA used a
low and a high estimate to represent uncertainty. The low estimate is based on experiences of Portland
Oregon, where they found the average cost to replace a lead connector to be $1,891 (in 2020 dollars).
The high estimate is based on partial SLR cost. Although the lead gooseneck is only 2 feet and the
system-owned service line length is an average of 30 feet (Sandvig et al., 2008), replacing a gooseneck
still involves digging in the street, mobilization, and traffic coordination. As the high estimate, the EPA
approximated the cost to replace a connector as the average cost of a partial SLR divided by 2 ($3803/2
= $1902 in 2020 dollars). For more details, see the derivation file "LSLR Unit Costs.xlsx", worksheet "LSLR
for Connectors", available in the docket at EPA-HQ-OW-2022-0801 at www.regulations.gov. For the
benefits analysis, the EPA estimated the benefits of these two types of replacements to be the same as a
partial replacement or GRR, in the absence of detailed lead concentration data for these scenarios.

Final LCRI Economic Analysis

8-9

October 2024


-------
Exhibit 8-8 and Exhibit 8-9 compare the quantified costs and benefits of the final LCRI to the quantified
costs and benefits of requiring all lead connectors and all galvanized lines downstream and previously
downstream from lead connectors be replaced along with LSLs and galvanized downstream of a lead line
or unknown lead content service line at the 10 percent annual replacement rate. Results are provided
for the high scenario at a 2 percent discount rate. As discussed in sections IV.B.2 and IV.0.3 of the
Federal Register Notice (USEPA, 2024) both the complete inventorying and mandatory removal of lead
connectors and galvanized service lines downstream and previously downstream of lead connectors is
not feasible without significantly delaying the replacement of lead and GRR service lines. Therefore,
note that although the EPA was able to estimate costs and benefits for this alternative option, using the
7th Drinking Water Infrastructure Needs Survey and Assessment survey data on lead content service
lines, the estimated results are uncertain and likely overestimate both costs and benefits since full lead
and GRR SLR is assumed to still occur within the required 10 year window (except for those systems on
deferred deadlines) when in fact these replacement may be delayed as a result of implementing the
requirements of this option. Given the concerns over feasibility and therefore the uncertainty associated
with the estimated costs and benefits of this alternative option (note benefits estimates would be
overestimated to a larger extent than costs), the EPA is discounting the fact that estimated net benefits
for this alternative option are greater than the estimated net benefits for the final LCRI. The final LCRI
maintains the final rules requirement to replace all LSLs and galvanized service lines downstream of LSLs
or unknown lead content service lines at the 10 percents annual replacement rate (except for those
systems on deferred deadlines).

Exhibit 8-8: Estimated National Annualized Rule Cost Comparison Between the Final LCRI and
Alternative Option Including Lead Connectors in Definition of Service Lines to be Replaced -
High Scenario - 2 Percent Discount Rate (millions of 2022 USD)





Final Rule



Alternative Option (Lead
Connectors and Galvanized Lines
Previously Downstream of Lead
Connectors Must be Replaced)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS Annual Costs













Sampling

$143.6

$176.2

$32.6

$143.6

$176.4

$32.8

PWS SLR

$124.5

$1,763.9

$1,639.4

$124.5

$1,921.7

$1,797.2

Corrosion Control Technology

$647.8

$692.9

$45.1

$647.8

$701.3

$53.5

Point-of Use Installation and
Maintenance

$5.9

$9.6

$3.7

$5.9

$9.7

$3.8

Public Education and Outreach

$72.1

$302.2

$230.1

$72.1

$306.6

$234.5

Rule Implementation and
Administration

$0.2

$3.4

$3.2

$0.2

$3.4

$3.2

Total Annual PWS Costs

$994.1

$2,948.2

$1,954.1

$994.1

$3,119.1

$2,125.0

Household SLR Costs

$26.4

$0.0

-$26.4

$26.4

$0.0

-$26.4

Final LCRI Economic Analysis

8-10

October 2024


-------




Final Rule



Alternative Option (Lead
Connectors and Galvanized Lines
Previously Downstream of Lead
Connectors Must be Replaced)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

State Rule Implementation and
Administration
Wastewater Treatment Plant
Costs

$41.8
$4.8

$67.6
$5.1

$25.8
$0.3

$41.8
$4.8

$67.9
$5.3

$26.1
$0.5

Total Annual Rule Costs

$1,067.1

$3,020.9

$1,953.8

$1,067.1

$3,192.3

$2,125.2

Acronyms: LCRI = Lead and Copper Rule Improvements; SLR = service line replacement; PWS = public water
system; USD = United States dollar.

Notes: (1) Previous Baseline costs are projected over the 35-year period of analysis and are affected by the EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

(2) Very small differences in results between the final rule and the regulatory option are due to inter-run variability
in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences between the costs
and benefits of the final rule and the alternative option.

Exhibit 8-9: Estimated National Annual Benefit Comparison Between the Final LCRI and
Alternative Option Including Lead Connectors in Definition of Service Lines to be Replaced -
High Scenario - 2 Percent Discount Rate (millions of 2022 USD)





Final Rule



Alternative Option (Lead Connectors
and Galvanized Lines Previously
Downstream of Lead Connectors
Must be Replaced)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual IQ Benefits

$3,279.0

$10,963.0

$7,684.0

$3,279.0

$12,646.8

$9,367.8

Annual Low-Birth Weight
Benefits

00

t-H

-oo-

$5.7

$3.9

00

t-H

-oo-

$6.4

$4.6

Annual ADHD Benefits

$179.9

$599.5

$419.6

$179.9

$684.8

$504.9

Annual Adult CVD Premature
Mortality Benefits

$8,174.9

$25,210.0

$17,035.1

$8,174.9

$28,943.5

$20,768.6

Total Annual Benefits

$11,635.6

$36,778.
2

$25,142.6

$11,635.6

$42,281.5

$30,645.9

Acronyms: ADHD = attention-deficit/hyperactivity disorder; CVD = cardiovascular disease; IQ = intelligence
quotient; LCRI = Lead and Copper Rule Improvements; USD = United States dollar.

Note: Very small differences in results between the final rule and the regulatory option are due to inter-run
variability in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences
between the costs and benefits of the final rule and the alternative option.

Final LCRI Economic Analysis

8-11

October 2024


-------
8.5 Alternative Service Line Replacement Deferral Deadline

Under the final LCRI, systems are eligible for a deferred deadline (or rate) for mandatory SLR if replacing
10 percent of the system's known lead and GRR service lines from the replacement pool (the total
number of lead and GRR service lines) would require the replacement of more than 39 annual
replacements per 1,000 service connections. Effectively, the criteria for a system to be eligible for a
deferred deadline is to exceed 39 annual replacements per 1,000 service connections. The EPA
considered, as part of the development of the final rule, two deferred deadline criteria in addition to the
final rule's per-connection rate: 1) systems may also be able to take advantage of a reduced maximum
replacement rate set at 10,000 lines per year if the rate is lower than the required 39 replacements per
1000 connections metric; and 2) systems may also be able to take advantage of a reduced maximum
replacement rate set at 8,000 lines per year if the rate is lower than the required 39 replacements per
1000 connections metric.

Exhibit 8-10 and Exhibit 8-11 compare the quantified costs and benefits of the final LCRI to the
quantified costs and benefits under an alternative SLR deferred deadline which would allow systems to
replace lead and GRR service lines at a maximum rate equal to the lower of two alternatives: 1) 10,000
lines per year; or 2) 39 replacements per 1000 connections per year, holding all other rule requirements
constant. Results are provided for the high scenario at a two percent discount rate.

Exhibit 8-10: Estimated National Annualized Rule Cost Comparison Between the Final LCRI
and Alternative Deferred Deadline Option (Adding Max Rate of 10,000 SL Per Year) - High
Scenario - 2 Percent Discount Rate (millions of 2022 USD)









Alternative Option (SL Replacement





Final Rule



Deferred Deadline with Additional







Potential Maximum Rate of 10,000 SL











Per Year)





Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS Annual Costs













Sampling

$143.6

$176.2

$32.6

$143.6

$176.0

$32.4

PWS SLR*

$124.5

$1,763.9

$1,639.4

$124.5

$1,763.1

$1,638.6

Corrosion Control
Technology

$647.8

$692.9

$45.1

$647.8

$692.8

$45.0

Point-of Use Installation and
Maintenance

$5.9

$9.6

$3.7

$5.9

$9.7

$3.8

Public Education and
Outreach

$72.1

$302.2

$230.1

$72.1

$302.4

$230.3

Rule Implementation and
Administration

$0.2

$3.4

$3.2

$0.2

$3.4

$3.2

Total Annual PWS Costs

$994.1

$2,948.2

$1,954.1

$994.1

$2,947.4

$1,953.3

Final LCRI Economic Analysis

8-12

October 2024


-------
Household SLR Costs**

$26.4

$0.0

-$26.4

$26.4

$0.0

-$26.4

State Rule Implementation
and Administration

$41.8

$67.6

$25.8

$41.8

$67.6

$25.8

Wastewater Treatment Plant

Costs***

$4.8

1

lo
-oo-

$0.3

$4.8

$5.0

$0.2

Total Annual Rule Costs

$1,067.1

$3,020.9

$1,953.8

$1,067.1

$3,020.0

$1,952.9

Acronyms: LCRI = Lead and Copper Rule Improvements; PWS = public water system; SL = service line; SLR = service
line replacement; USD = United Stated dollar.

Notes: (1) Previous Baseline costs are projected over the 35-year period of analysis and are affected by the EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

(2) Very small differences in results between the final rule and the regulatory option are due to inter-run variability
in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences between the costs
and benefits of the final rule and the alternative option.

Exhibit 8-11: Estimated National Annual Benefit Comparison Between the Final LCRI and
Alternative Deferred Deadline Option (Adding Max Rate of 10,000 SL Per Year) - High Scenario

- 2 Percent Discount Rate (millions of 2022 USD)









Alternative Option (SL Replacement Deferred





Final Rule



Deadline with Additional Potential Maximum
Rate of 10,000 SL Per Year)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual IQ
Benefits

$3,279.0

$10,963.0

$7,684.0

$3,279.0

$10,960.3

$7,681.3

Annual Low-













Birth Weight

00

t-H

$5.7

$3.9

00

t-H

$5.7

$3.9

Benefits













Annual ADHD
Benefits

$179.9

$599.5

$419.6

$179.9

$599.3

$419.4

Annual Adult













CVD Premature
Mortality

$8,174.9

$25,210.0

$17,035.1

$8,174.9

$25,203.7

$17,028.8

Benefits













Total Annual
Benefits

$11,635.6

$36,778.2

$25,142.6

$11,635.6

$36,769.0

$25,133.4

Acronyms: ADHD = attention-deficit/hyperactivity disorder; CVD = cardiovascular disease; IQ = intelligence
quotient; LCRI = Lead and Copper Rule Improvements; SL = service line; USD = United States dollar.

Note: Very small differences in results between the final rule and the regulatory option are due to inter-run
variability in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences
between the costs and benefits of the final rule and the alternative option.

Final LCRI Economic Analysis

8-13

October 2024


-------
Exhibit 8-12 and Exhibit 8-13 compare the quantified costs and benefits of the final LCRI to the
quantified costs and benefits under an alternative SLR deferred deadline which would allow systems to
replace lead and GRR service lines at a maximum rate equal to the lower of two alternatives: 1) 8,000
lines per year; or 2) 39 replacements per 1000 connections per year, holding all other rule requirements
constant. Results are provided for the high scenario at a two percent discount rate.

Exhibit 8-12: Estimated National Annualized Rule Cost Comparison Between the Final LCRI
and Alternative Deferred Deadline Option (Adding Max Rate of 8,000 SL Per Year) - High
Scenario - 2 Percent Discount Rate (millions of 2022 USD)





Final Rule



Alternative Option (SL
Replacement Deferred Deadline
with Additional Potential Maximum
Rate of 8,000 SL Per Year)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS Annual Costs













Sampling

$143.6

$176.2

$32.6

$143.6

$176.0

$32.4

PWS SLR

$124.5

$1,763.9

$1,639.4

$124.5

$1,761.8

$1,637.3

Corrosion Control Technology

$647.8

$692.9

$45.1

$647.8

$692.8

$45.0

Point-of Use Installation and
Maintenance

$5.9

$9.6

$3.7

$5.9

$9.7

$3.8

Public Education and Outreach

$72.1

$302.2

$230.1

$72.1

$302.7

$230.6

Rule Implementation and
Administration

$0.2

$3.4

$3.2

$0.2

$3.4

$3.2

Total Annual PWS Costs

$994.1

$2,948.2

$1,954.1

$994.1

$2,946.4

$1,952.3

Household SLR Costs

$26.4

$0.0

-$26.4

$26.4

$0.0

-$26.4

State Rule Implementation and
Administration

Wastewater Treatment Plant
Costs

$41.8
$4.8

$67.6
$5.1

$25.8
$0.3

$41.8
$4.8

$67.6
$5.0

$25.8
$0.2

Total Annual Rule Costs

$1,067.1

$3,020.9

$1,953.8

$1,067.1

$3,019.0

$1,951.9

Acronyms: LCRI = Lead and Copper Rule Improvements; PWS = public water system; SL = service line; SLR = service
line replacement; USD = United Stated dollar.

Notes: (1) Previous Baseline costs are projected over the 35-year period of analysis and are affected by the EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

(2) Very small differences in results between the final rule and the regulatory option are due to inter-run variability
in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences between the costs
and benefits of the final rule and the alternative option.

Final LCRI Economic Analysis

8-14

October 2024


-------
Exhibit 8-13: Estimated National Annual Benefit Comparison Between the Final LCRI and
Alternative Deferred Deadline Option (Adding Max Rate of 8,000 SL Per Year) - High Scenario -

2 Percent Discount Rate (millions of 2022 USD)





Final Rule



Alternative Option (SL Replacement
Deferred Deadline with Additional
Potential Maximum Rate of 8,000
SL Per Year)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

Annual IQ Benefits

$3,279.0

$10,963.0

$7,684.0

$3,279.0

$10,943.3

$7,664.3

Annual Low-Birth Weight
Benefits

$1.8

$5.7

$3.9

$1.8

$5.7

$3.9

Annual ADHD Benefits

$179.9

$599.5

$419.6

$179.9

$598.3

$418.4

Annual Adult CVD
Premature Mortality
Benefits

$8,174.9

$25,210.0

$17,035.1

$8,174.9

$25,164.0

$16,989.
1

Total Annual Benefits

$11,635.6

$36,778.2

$25,142.6

$11,635.
6

$36,711.3

$25,075.
7

Acronyms: ADHD = attention-deficit/hyperactivity disorder; CVD = cardiovascular disease; IQ = intelligence
quotient; LCRI = Lead and Copper Rule Improvements; SL = service line; USD = United States dollar.

Note: Very small differences in results between the final rule and the regulatory option are due to inter-run
variability in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences
between the costs and benefits of the final rule and the alternative option.

8.6 Alternative Tap Sampling Requirements

Under the final LCRI there are a number of criteria that can result in a system starting standard six-
month lead tap sample monitoring. Systems are required to conduct six-month lead tap sample
monitoring if the system: has an ALE; has known lead and/or GRR service lines at LCRI compliance date;
or discovers any LSLs and/or GRR service lines after the compliance date (unless the system replaces all
the discovered service lines prior to the next tap monitoring period); in addition to other criteria
unchanged from the Lead and Copper Rule Revisions (LCRR). Note that under the final LCRI
requirements non-lead and non-lead/unknown service line systems remain on their existing Lead and
Copper Rule (LCR) monitoring schedule at the rule compliance date. They remain on their previous tap
sampling schedule until new sampling, which is compliant with the LCRI sampling protocols, may change
the system's calculated P90 to exceed the AL. Also, systems with lead and GRR service lines that
sampled using the new LCRI protocol (i.e., correct priority tiering sites, correct sample volume, and
either first-liter sample (at non-lead service line sites) or first- and fifth-liter samples (at sites with LSLs))
and are below the LCRI AL prior to the compliance date may qualify to retain their current tap sampling
schedule. As part of the development of the final rule, the EPA considered an alternative option that
would also require systems with unknown lead content service lines (even when no lead and/or GRR
service lines are known to be present in the system) to conduct standard six-month monitoring.

Final LCRI Economic Analysis

8-15

October 2024


-------
The EPA's analysis of this alternative option found that the expected increase in sampling cost and
potential increase in benefits associated with systems (non-lead/unknown and 100 percent unknown)
taking earlier corrective action as a result of ALEs were small and did not affect estimated national
annualized cost and benefits at the $100,000 significant digit level. Therefore, the EPA is not presenting
exhibits characterizing the differences between the estimated costs and benefit of the final rule and the
lead tap sampling alternative option. However, it is important to note that the EPA has feasibility
concerns associated with the alternative option. The additional cost and burden to public water systems
(PWSs) and States would draw resources away from the implementation of other LCRI rule components
such as corrosion control treatment (CCT) and public education, and the implementation of tap sampling
in higher risk locations. See section IV.E for further discussion. Because of these concerns it is likely that
the estimated cost and benefit of the alternative option are less certain than those of the final rule.

8.7 Alternative Temporary Filter Programs for Systems with Multiple ALEs

The final LCRI includes a requirement that systems with at least two lead ALEs in a rolling year-year
period must prepare and submit a filter plan to the State. In addition, if a system has three or more ALEs
in a rolling five- year period, it must make filters available to all consumers in the distribution system.
The EPA assessed two additional alternative filter programs while developing the final rule. Under both
alternatives systems with at least two ALEs in a rolling five-year period will follow the final rule
requirements to develop and submit to the State a filter plan. For systems with at least three ALEs in a
rolling five-year window, alternative one would require systems to make temporary filters available to
all customers having lead, GRR, and unknown lead content service lines. Alternative two would require
systems to directly deliver temporary filters to all customers in the distribution system.

Exhibit 8-14 compares the quantified costs of the final LCRI to the quantified costs of requiring systems
with at least three ALEs in a rolling five-year window to make filters available to households with lead,
GRR, or unknown lead content service lines. Under this alternative temporary filter option all other final
LCRI rule requirements have been held constant. Cost results are provided for the high scenario at the 2
percent discount rate.

Exhibit 8-14: Estimated National Annualized Rule Cost Comparison Between the Final LCRI
and Alternative Temporary Filters Program for Multiple ALE Systems Option (Filters Made
Available to Lead, GRR, and Unknown Service Line Customers Only) - High Scenario - 2
Percent Discount Rate (millions of 2022 USD)

Final Rule

Alternative Option (Temporary
Filters Made Available to LSL,
GRR, and Unknown Lead Content

Service Line Customers in

Systems Meeting Multiple ALE

Criteria)

Baseline

LCRI Incremental Baseline

LCRI Incremental

PWS Annual Costs

Sampling

$143.6	$176.2

$32.6 $143.6 $176.1

$32.5

Final LCRI Economic Analysis

8-16

October 2024


-------








Alternative Option (Temporary
Filters Made Available to LSL,





Final Rule



GRR, and Unknown Lead Content







Service Line Customers in
Systems Meeting Multiple ALE
Criteria)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS SLR

$124.5

$1,763.9

$1,639.4

$124.5

$1,763.9

$1,639.4

Corrosion Control Technology

$647.8

$692.9

$45.1

$647.8

$692.9

$45.1

Point-of Use Installation and
Maintenance

$5.9

$9.6

$3.7

$5.9

$9.6

$3.7

Public Education and Outreach

$72.1

$302.2

$230.1

$72.1

$274.8

$202.7

Rule Implementation and
Administration

$0.2

$3.4

$3.2

$0.2

$3.4

$3.2

Total Annual PWS Costs

$994.1

$2,948.2

$1,954.1

$994.1

$2,920.7

$1,926.6

Household SLR Costs

$26.4

$0.0

-$26.4

$26.4

$0.0

-$26.4

State Rule Implementation and
Administration

$41.8

$67.6

$25.8

$41.8

$67.6

$25.8

Wastewater Treatment Plant
Costs

$4.8

1

LO

-oo-

$0.3

$4.8

$5.1

$0.3

Total Annual Rule Costs

$1,067.1

$3,020.9

$1,953.8

$1,067.1

$2,993.4

$1,926.3

Acronyms: ALE = action level exceedance; GRR = galvanized requiring replacement; LCRI = Lead and Copper Rule
Improvements; LSL = lead service line; PWS = public water system; SLR = service line replacement; USD = United
States dollar.

Notes: (1) Previous Baseline costs are projected over the 35-year period of analysis and are affected by the EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

(2) Very small differences in results between the final rule and the regulatory option are due to inter-run variability
in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences between the costs
and benefits of the final rule and the alternative option.

Because the EPA's benefit analysis cannot quantify benefits from reducing lead exposures at residences
that do not initially have lead or GRR service lines, the estimated benefits for this option are equal to
those estimated for the final rule and therefore are not repeated. See Exhibit 6-3 for the estimated
benefits of both the final LCRI and this alternative option. A discussion of the EPA's lead concentration
data can be found in Chapter 5, Section 5.2 The quantified benefits of the final rule are in fact a more
accurate representation of the alternative option where filters would not be made available to non-lead,
GRR, and unknown service line customers. The analysis for the final LCRI was not able to quantify the
potential benefits of filter use at non-lead and GRR service line households, resulting in an
underestimate of benefits. Therefore, although not shown in the estimated values, the benefits of the
final LCRI are likely larger than those of the alternative option.

Exhibit 8-15 compares the quantified costs of the final LCRI to the quantified costs s of requiring systems
with at least three ALEs in a rolling five-year window to directly deliver filters to all customers in the
distribution system. Results are provided for the high scenario at a two percent discount rate. Again, the

Final LCRI Economic Analysis

8-17

October 2024


-------
EPA does not present benefit values for this option. The monetized benefits are equivalent to those of
the final LCRI, see Exhibit 6-3. Given concerns over the potential to underestimate the cost impact of the
final LCRI multiple ALE filter program, which is dependent on the number of customers in a system that
chose to obtain a filter from the PWS, the EPA assumed a 100 percent customer filter pick-up rate. This
assumption, made to ensure a conservative assessment of the cost impacts of the program could lead to
a potential overestimate of the benefits of such a program. However, this potential to overestimate
benefits is tempered by the fact that, as discussed above, the EPA can only calculate benefits accruing to
households that initially have lead or GRR service lines. Therefore, although benefits accruing to this
household group may be overestimated, the increased assumed pick-up rate among the non-lead and
GRR households does not affect estimated benefits. So, given that both the final LCRI and the direct
delivery of filters option assume 100 percent filter use rates in the estimation of benefits, the estimated
benefits are equal and likely overestimated. It seems reasonable to postulate that the filter use rate may
be higher for the direct delivery option, given the increased level of effort required of consumers to
pick-up a filter from a PWS designated location under the LCRI (although the EPA has no documented
information to indicated this is true) and therefore this option would result in greater benefits. Note,
however, that the EPA has feasibility concerns, discussed in section IV.K.2 of the Federal Register Notice
(USEPA, 2024), with the required direct delivery of temporary filters to all customers. Therefore, the
alternative option costs and benefits are more uncertain and may be overestimated because the values
assume timely implementation of the requirement.

Because the EPA is unable to quantify benefits from reducing lead exposures at residences that do not
initially have lead or GRR service lines and given the concerns over the feasibility of requiring direct
delivery of temporary filters to all customers, the EPA cannot wholly rely on estimates of net benefits to
determine the optimal temporary filter program regulatory requirements when systems have multiple
ALEs. Although the estimated net benefits for the "only make filters available to customers with lead,
GRR, or unknown lead content service lines" are greater than those estimated for the final rule the EPA
has determined that the additional non-quantifiable potential benefits associated with lead reductions
at households that did not initially have lead or GRR service lines outweighs the additional cost of the
final rule program. Also as stated above the EPA has feasibility concerns with the option requiring direct
delivery to all customers, to the LCRI maintains the final rule requirement that, if a system has three or
more ALEs in a rolling five-year period, it must make filters available to all consumers in the distribution
system.

Exhibit 8-15: Estimated National Annualized Rule Cost Comparison Between the Final LCRI
and Alternative Temporary Filters Program for Multiple ALE Systems Option (Deliver Filters to
All Customers) - High Scenario - 2 Percent Discount Rate (millions of 2022 USD)





Final Rule



Alternative Option (Deliver Temporary

Filters Directly to All Customers in
Systems Meeting Multiple ALE Criteria)



Baseline

LCRI

Incremental

Baseline LCRI Incremental

PWS Annual Costs









Sampling

$143.6

$176.2

$32.6

$143.6 $176.1 $32.5

Final LCRI Economic Analysis

8-18

October 2024


-------








Alternative Option (Deliver Temporary





Final Rule



Filters Directly to All Customers in
Systems Meeting Multiple ALE Criteria)



Baseline

LCRI

Incremental

Baseline

LCRI

Incremental

PWS SLR

$124.5

$1,763.9

$1,639.4

$124.5

$1,763.9

$1,639.4

Corrosion Control Technology

$647.8

$692.9

$45.1

$647.8

$692.9

$45.1

Point-of Use Installation and
Maintenance

$5.9

$9.6

$3.7

$5.9

$9.6

$3.7

Public Education and Outreach

$72.1

$302.2

$230.1

$72.1

$308.1

$236.0

Rule Implementation and
Administration

$0.2

$3.4

$3.2

$0.2

$3.4

$3.2

Total Annual PWS Costs

$994.1

$2,948.2

$1,954.1

$994.1

$2,954.0

$1,959.9

Household SLR Costs

$26.4

$0.0

-$26.4

$26.4

$0.0

-$26.4

State Rule Implementation and
Administration

$41.8

$67.6

$25.8

$41.8

$67.6

$25.8

Wastewater Treatment Plant
Costs

$4.8

1

LO

Šuy

$0.3

$4.8

$5.1

$0.3

Total Annual Rule Costs

$1,067.1

$3,020.9

$1,953.8

$1,067.1

$3,026.7

$1,959.6

Acronyms: ALE = action level exceedance; LCRI = Lead and Copper Rule Improvements; PWS = public water system;
SLR = service line replacement; USD = United States dollar.

Notes: (1) Previous Baseline costs are projected over the 35-year period of analysis and are affected by the EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

(2) Very small differences in results between the final rule and the regulatory option are due to inter-run variability
in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences between the costs
and benefits of the final rule and the alternative option.

8.8 Small System Flexibility

The final LCRI includes compliance flexibility for community water systems (CWSs) that serve 3,300 or
fewer people, and all non-transient non-community water system (NTNCWSs). If these water systems
have a lead 90th percentile above the AL, or an ALE, the system can choose to install or reoptimizing CCT
or choose the alternative of maintaining point-of-use (POU) devices or replacing all lead-bearing
plumbing.218 As part of the rule development process the EPA also considered the alternative CWS size
threshold of serving 10,000 or fewer people. So, CWSs serving up to 10,000 would have the ability to
selects between CCT and the alternative compliance options if they exceed the AL. Note that under both
alternatives NTNCWSs of all sizes qualify for the compliance flexibility. Because providing and

218 The EPA could not evaluate the cost of removing lead-bearing plumbing components from small systems, but
the agency notes that, if a system should select this option, it would likely be considered the lowest cost
alternative of the compliance options. Therefore, since the EPA has not included this option in its cost modeling,
the agency's small system compliance costs may be overestimated.

Final LCRI Economic Analysis

8-19

October 2024


-------
maintaining POU devices is almost always more costly than installing or re-optimizing CCT, for CWSs
serving greater than 3,300, the EPA does not expect systems in the 3,300 to 10,000 persons served size
category to choose POU as a compliance strategy if provided the option. In fact, the EPA's modelling
results predict only 11 PWSs of this size would choose POU over CCT. Therefore, both the expected
incremental costs and benefits, between the final and alternative small system flexibility options, are
similar.

Exhibit 8-16 and Exhibit 8-17, compare the quantified costs and benefits of the final LCRI to the
quantified costs and benefits for the alternative option where the CWS compliance flexibility size
threshold is equal to systems serving 10,000 or fewer persons. The final LCRI sets the CWS compliance
flexibility threshold at systems serving 3,300 or fewer persons. Note under the final rule and the assess
alternative NTNCWSs are allowed compliance flexibility. Results are provided for the high scenario at the
2 percent discount rate. The estimated costs and benefits under the alternative small system
compliance flexibility threshold, of systems serving up to 10,000 persons, assumes the effective
implementation of POU in place of system wide CCT. As discussed in section IV.I of the Federal Register
Notice (USEPA, 2024) the agency finds that in CWSs serving greater than 3,300 persons it is highly
unlikely that POU programs, given their complexity, will be implemented effectively and could not make
a determination that a POU program is as effective as CCT at minimizing exposure to lead in water for
systems serving more than 3,300 persons. For example, in the LCRI proposal, the EPA described a
scenario in which a system that serves 3,301 consumers would have to provide and maintain
approximately 1,000 POU devices (88 FR 84878, USEPA, 2023a). Every year, at least 300 POU devices
would have to be monitored by the water system, which would require a significant coordination effort
and over 300 household visits by the water system. Systems would also need to insure all POU devises
are working correctly, and the filter media is replaced to insure lead removal. This could easily result in
an additional 1,000 or more home visits per year. The burden required to undertake this compliance
alternative and implement it correctly would be difficult for a water system serving more than 3,300
persons to carry out given financial, administrative, and technical limitations. Therefore, under the
alternative threshold option the estimated costs and, to a larger degree, the estimated benefits are
uncertain. Given the concerns over feasibility and therefore the uncertainty associated with the
estimated costs and benefits of this alternative option, the EPA is discounting the fact that estimated net
benefits for this alternative option are greater than the estimated net benefits for the final LCRI. The
final LCRI maintains the small system compliance flexibility threshold at systems serving 3,300 or fewer
persons.

Exhibit 8-16: Estimated National Annualized Rule Cost Comparison Between the Final LCRI
and Alternative Small System Flexibility Option (Flexibility for CWSs Serving up to 10,000
Persons) - High Scenario - 2 Percent Discount Rate (millions of 2022 USD)

Final Rule

Alternative Option (Small System
Flexibility for CWSs Serving up to
	10,000 Persons)	

Baseline

LCRI

Incremental Baseline LCRI Incremental

PWS Annual Costs

Sampling

$143.6

$176.2

$32.6 $143.6 $176.0

$32.4

Final LCRI Economic Analysis

8-20

October 2024


-------




Final Rule



Alternative Option (Small System
Flexibility for CWSs Serving up to
10,000 Persons)



Baseline

LCRI

Incremental

Baseline

LCRI Incremental

PWS SLR

$124.5

$1,763.9

$1,639.4

$124.5

$1,763.9

$1,639.4

Corrosion Control Technology

$647.8

$692.9

$45.1

$647.8

$692.7

$44.9

Point-of Use Installation and
Maintenance

$5.9

$9.6

$3.7

$5.9

$9.6

$3.7

Public Education and Outreach

$72.1

$302.2

$230.1

$72.1

$302.0

$229.9

Rule Implementation and
Administration

$0.2

$3.4

$3.2

$0.2

$3.4

$3.2

Total Annual PWS Costs

$994.1

$2,948.2

$1,954.1

$994.1

$2,947.6

$1,953.5

Household SLR Costs

$26.4

$0.0

-$26.4

$26.4

$0.0

-$26.4

State Rule Implementation and
Administration

$41.8

$67.6

$25.8

$41.8

$67.6

$25.8

Wastewater Treatment Plant
Costs

$4.8

$5.1

$0.3

$4.8

$5.2

$0.4

Total Annual Rule Costs

$1,067.1

$3,020.9

$1,953.8

$1,067.1

$3,020.4

$1,953.3

Acronyms: CWS = community water system; LCRI = Lead and Copper Rule Improvements; SLR = service line
replacement; PWS = public water system; United States dollar.

Notes: (1) Previous Baseline costs are projected over the 35-year period of analysis and are affected by the EPA's
assumptions on three uncertain variables which vary between the low and high cost scenarios.

(2) Very small differences in results between the final rule and the regulatory option are due to inter-run variability
in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences between the costs
and benefits of the final rule and the alternative option.

Exhibit 8-17: Estimated National Annual Benefit Comparison Between the Final LCRI and
Alternative Small System Flexibility Option (Flexibility for CWSs Serving up to 10,000 Persons)
- High Scenario - 2 Percent Discount Rate (millions of 2022 USD)





Final Rule



Alternative Option (Small System
Flexibility for CWSs Serving up to
10,000 Persons)



Baseline

LCRI

Incremental

Baseline LCRI Incremental

Annual IQ Benefits

$3,279.0

$10,963.0

$7,684.0

$3,279.0 $10,963.1

$7,684.1

Annual Low-Birth Weight
Benefits

$1.8

$5.7

$3.9

$1.8 $5.7

$3.9

Annual ADHD Benefits

$179.9

$599.5

$419.6

$179.9 $599.5

$419.6

Annual Adult CVD Premature
Mortality Benefits

$8,174.9

$25,210.0

$17,035.1

$8,174.9 $25,210.5

$17,035.6

Final LCRI Economic Analysis

8-21

October 2024


-------




Final Rule



Alternative Option (Small System
Flexibility for CWSs Serving up to
10,000 Persons)



Baseline

LCRI

Incremental

Baseline LCRI Incremental

Total Annual Benefits

$11,635.6

$36,778.2

$25,142.6

$11,635.6 $36,778.8 $25,143.2

Acronyms: ADHD = attention-deficit/hyperactivity disorder; CVD = cardiovascular disease; CWS = community water

system; IQ = intelligence quotient; LCRI = Lead and Copper Rule Improvements; USD = United States dollar.
Note: Very small differences in results between the final rule and the regulatory option are due to inter-run
variability in the SafeWater LCR model, and/or rounding, and should not be interpreted at true differences
between the costs and benefits of the final rule and the alternative option.

8.9 Summary of Alternative Options Considerations

Exhibit 8-18 provides a summary of the estimated annualized monetized costs, benefits, and net
benefits for the final LCRI and the alternative options considered in this chapter.

Exhibit 8-18: Estimated National Annualized Rule Cost, Benefit, and Net Benefit Comparison
Between the Final LCRI and Alternative Options Considered - High Scenario - 2 Percent

Discount Rate (millions of 2022 USD)



Total

Total







Annualized

Annualized

Net Benefit



Cost

Benefit





Final LCRI

$ 1,953.8

$ 25,142.6

$

23,188.8

Alternative Options Considered

Action Level < 0.015 mg/L

$ 1,840.5

$ 23,739.7

$

21,899.2

Action Level < 0.005 mg/L

$ 2,186.8

$ 27,700.9

$

25,514.1

Service Line Replacement Rate = 7%

$ 1,904.7

$ 21,902.7

$

19,998.0

Lead Connectors and Galvanized Lines Previously









Downstream of Lead Connectors Must be Replaced

$ 2,125.2

$ 30,645.9

$

28,520.7

Service Line Replacement Deferred Deadline with









Additional Potential Maximum Rate of 10,000 SL Per Year

$ 1,952.9

$ 25,133.4

$

23,180.5

Service Line Replacement Deferred Deadline with









Additional Potential Maximum Rate of 8,000 SL Per Year

$ 1,951.9

$ 25,075.7

$

23,123.8

Temporary Filters Made Available to Lead, GRR, and









Unknown Lead Content Service Line Customers Only in









Systems Meeting Multiple ALE Criteria

$ 1,926.3

$ 25,142.6

$

23,216.3

Deliver Temporary Filters Directly to All Customers in









Systems Meeting Multiple ALE Criteria

$ 1,959.6

$ 25,142.6

$

23,183.0

Small System Flexibility for CWSs Serving up to 10,000









Persons

$ 1,953.3

$ 25,143.2

$

23,189.9

Note: The EPA considered an alternative to the final LCRI's lead tap sampling standard monitoring requirements.
The EPA's analysis of this alternative option found that the expected increase in sampling cost and potential
increase in benefits associated with systems (non-lead/unknown and 100 percent unknown) taking earlier

Final LCRI Economic Analysis

8-22

October 2024


-------
corrective action as a result of ALEs were small and did not affect estimated nation annualized costs and benefits
at the EPA $100,000 significant digit level. Therefore, the EPA did not present the estimated cost, benefit, and net
benefit for this lead tap sampling alternative option.

The EPA's analysis of the alternative regulatory options found that the following options had estimated
annual net benefits greater than the final LCRI: (1) setting the AL to 0.005 mg/L; (2) including lead
connectors and galvanized service lines previously downstream of lead connectors in the definition of
lead content requiring replacement; (3) requiring systems with multiple ALEs to make temporary filters
available to households with lead, GRR, or unknown lead content service lines; and (4) allowing systems
serving up to 10,000 persons the ability to utilize the small system compliance flexibility options. From a
purely economic standpoint that would mean these four options are preferable to the final LCRI.
However, three of these options were not selected, in place of the final rule, because of questionable
technical feasibility. SDWA section 1412(b)(4)(D) says the term "feasible" means feasible with the use of
the best technology, treatment techniques and other means which the Administrator finds, after
examination for efficacy under field conditions and not solely under laboratory conditions, are available.
The EPA has discussed the agency's feasibility concerns with regard to: setting the action level to 0.005
mg/L; including lead connectors and galvanized service lines previously downstream of lead connectors
in the definition of lead content requiring replacement; and allowing systems serving up to 10,000
persons the ability to utilize the small system compliance flexibility options, in preceding sections of this
preamble. Regarding setting the AL at a level below 0.010 mg/L, the EPA has expressed concern
associated with feasibility. See section IV.F.4, of the final LCRI Federal Register notice (USEPA, 2024) for
information on feasibility. When considering the inclusion of lead connectors and galvanized service
lines previously downstream of lead connectors in the set of service lines that must be actively replaced,
the EPA was concerned about how these activities might pull resources away from the removal of lead
and GRR service lines that pose a greater exposure risk. See sections IV.B.2 and IV.O.3, of the final LCRI
Federal Register notice (USEPA, 2024) for a detailed discussion. In the case of setting the threshold for
the small system flexibility option to include systems serving up to 10,000 persons or fewer, despite the
modeling results showing an increase net benefits under this option, the EPA finds that the complexity
of implementing POU filtration at all residences in a system serving 3,300 to 10,000 individuals, or
potentially 1,300 to 4,000 separate locations, cannot be correctly captured in the estimated cost
structure within the economic model and makes this option infeasible. See section IV.I, of the final LCRI
Federal Register notice (USEPA, 2024) for additional information on point-of-use feasibility. In addition,
the monetized benefits associated with the implementation of CCT are known to be underestimated
given the potential reductions in lead exposure at homes without lead and GRR service lines in a system
implementing CCT which is not captured in the EPA benefit estimates. The CCT benefits also do not
capture reduced water loss, plumbing repair cost, and water damage costs associated with the
increased use of corrosion control. See section VI.F.2, of the final LCRI Federal Register notice (USEPA,
2024) for more information on the unquantified impacts. See section IV.F, of the final LCRI Federal
Register notice (USEPA, 2024) for additional information on CCT. With regard to estimated annual net
benefits being greater for the alternative option where systems with multiple ALEs would be required to
only make temporary filters available to households with lead, GRR, or unknown lead content service
lines, the EPA has highlighted the inability of the benefits analysis to monetize positive health impact
from reduced lead exposure at non-lead and GRR service line locations which leads to an underestimate
of final LCRI benefits relative to the benefits estimated for this alternative option. Note also that the EPA
made a conservative costing assumption that 100 percent of households that are eligible to receive a

Final LCRI Economic Analysis

8-23

October 2024


-------
filter would pick-up a filter when made available. The EPA has very little information on what the actual
pick-up rate may be but it is possible that the rate could be significantly less than 100 percent and
therefore the costs for both the final LCRI and this alternative multiple ALE temporary filters program
are overestimated, and given the fact that the final LCRI is making filters available to all households in a
system its estimated costs are likely overestimated to a greater extent than the alternative option.
Because of the similar annual estimated net benefits between the two alternatives, only $27.5 million in
2022 dollars, and the benefit and cost estimation uncertainties outlined above the EPA cannot rely on
the relative size of the estimated net benefits in selecting between these options. Therefore, the EPA
selected the final LCRI multiple ALE option because it protects individuals in systems with multiple ALEs
that do not have lead, GRR, or unknown service lines, were as the alternative option while addressing
most of the exposure issues in lead, GRR service line systems today does not cover systems with
multiple ALEs and no lead, GRR, or unknown service lines. The alternative option will also effectively
sunset as all unknowns are identified and lead and GRR service lines are replaced (13 years except for
systems on approved differed deadlines) leaving consumers in systems with chronic ALEs and no lead or
GRR service lines to be exposed to potentially high levels of lead coming from premise plumbing. The
final rule addresses this issue into the future by requiring filters be made available to all customers in
systems with multiple ALEs.

In the case of the alternative lead tap sample monitoring requirements that would have systems with
unknown lead content service lines start standard six-month lead tap sampling at the LCRI compliance
date, the EPA's monetized cost and benefit estimates were too close to conclusively determine if this
alternative option or the final LCRI has greater net benefits. Due to the potentially high volume of
systems required to start standard monitoring, the EPA did not select to move forward with this
alternative lead tap sampling option. One concern is the ability of the States to handle the increased
demands of overseeing the potentially large number of systems requiring sampling assistance during the
compressed time period immediately following the rule compliance date. Another concern is that
requiring systems with unknowns to start standard six-month lead tap sampling would affect a large
number of small systems, as the EPA estimates that 45 percent of small systems, or 20,200 systems,
have an inventory with unknown material service lines and no lead or GRR service lines. Lastly, the EPA
considered a phased approach to include systems with unknowns in the standard monitoring
requirements but decided that the complexity of a phased approach was not commensurate with the
benefits, as nearly all systems will conduct monitoring within three years of the rule promulgation based
on their LCR sampling schedule. See section IV.E, of the final LCRI Federal Register notice (USEPA, 2024)
for additional information on lead tap sampling.

8.10 References

Sandvig, A., P. Kwan, G. Kirmeyer, B. Maynard, D. Mast, R.R. Trussell, S. Trussell, A. Cantor, and A.
Prescott. 2008. Contribution of Service Line and Plumbing Fixtures to Lead and Copper Rule Compliance
Issues. Denver, CO: AWWA Research Foundation.

USEPA. 2020. Economic Analysis for the Final Lead and Copper Rule Revisions. December 2020. Office of
Water. EPA 816-R-20-008.

USEPA. 2024. National Primary Drinking Water Regulations: Lead and Copper Rule Improvements. Final
Rule.

Final LCRI Economic Analysis

8-24

October 2024


-------
Final LCRI Economic Analysis	8-25	October 2024


-------