* O \
KWJ
*1 PRO^^
Regulatory Impact Analysis for Proposed
Federal Implementation Plan Addressing
Regional Ozone Transport for the 2015 Ozone
National Ambient Air Quality Standard
-------
ii
-------
EP A-452/D-22-001
February 2022
Regulatory Impact Analysis for Proposed Federal Implementation Plan Addressing Regional
Ozone Transport for the 2015 Ozone National Ambient Air Quality Standard
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Health and Environmental Impacts Division
Research Triangle Park, NC
in
-------
CONTACT INFORMATION
This document has been prepared by staff from the Office of Air Quality Planning and
Standards, U.S. Environmental Protection Agency. Questions related to this document should be
addressed to U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, C439-02, Research Triangle Park, North Carolina 27711 (email:
oaqpseconomics@epa.gov).
ACKNOWLEDGEMENTS
In addition to EPA staff from the Office of Air Quality Planning and Standards, personnel from
the Office of Atmospheric Programs and the Office of Policy's National Center for
Environmental Economics contributed data and analysis to this document.
iv
-------
TABLE OF CONTENTS
LIST OF TABLES x
LIST OF FIGURES xv
EXECUTIVE SUMMARY ES-1
Overview ES-1
ES. 1 Identifying Needed Emissions Reductions ES-2
ES.2 Baseline and Analysis Years ES-4
ES.3 Air Quality Modeling ES-5
ES.4 Control Strategies and Emissions Reductions ES-6
ES.4.1 EGUs ES-7
ES.4.2 Non-EGUs ES-10
ES.5 Cost Impacts ES-14
ES.6 Benefits ES-14
ES.6.1 Benefits Estimates ES-14
ES.6.2 Climate Benefits ES-20
ES.6.3 Additional Unquantified Benefits ES-21
ES.7 Environmental Justice Impacts ES-21
ES.8 Results of Benefit-Cost Analysis ES-23
CHAPTER 1: INTRODUCTION AND BACKGROUND 1-1
Overview 1-1
1.1 Background 1-2
1.1.1 Role of Executive Orders in the Regulatory Impact Analysis 1-4
1.1.2 Alternatives Analyzed 1-4
1.1.3 The Need for Air Quality or Emissions Regulation 1-7
1.2 Overview and Design of the RIA 1-7
1.2.1 Methodology for Identifying Needed Reductions 1-7
1.2.2 States Covered by the Rule 1-9
1.2.3 Regulated Entities 1-10
1.2.4 Baseline and Analysis Years 1-11
1.2.5 Emissions Controls, Emissions, and Cost Analysis Approach 1-12
1.2.6 Benefits Analysis Approach 1-13
1.3 Organization of the Regulatory Impact Analysis 1-13
CHAPTER 2: INDUSTRY SECTOR PROFILES 2-1
Overview 2-1
v
-------
2.1 Background 2-1
2.2 Power Sector Overview 2-1
2.2.1 Generation 2-2
2.2.2 Transmission 2-9
2.2.3 Distribution 2-10
2.3 Sales, Expenses, and Prices 2-10
2.3.1 Electricity Prices 2-11
2.3.2 Prices of Fossil Fuels Used for Generating Electricity 2-15
2.3.3 Changes in Electricity Intensity of the U.S. Economy from 2014 to 2020 2-15
2.4 Deregulation and Restructuring 2-17
2.5 Industrial Sectors Overview 2-20
2.5.1 Cement and Cement Product Manufacturing 2-21
2.5.2 Iron and Steel Mills and Ferroalloy Manufacturing 2-24
2.5.3 Glass and Glass Product Manufacturing 2-26
2.5.4 Pipeline Transportation of Natural Gas 2-27
2.5.5 Tier 2 Industries 2-27
2.6 References 2-29
CHAPTER 3: AIR QUALITY IMPACTS 3-1
Overview 3-1
3.1 Air Quality Modeling Platform 3-2
3.2 Applying Modeling Outputs to Create Spatial Fields 3-4
3.3 Generation of Spatial Fields for the Proposed FIP for the 2015 Ozone NAAQS 3-6
3.4 Spatial Distribution of Air Quality Impacts 3-7
3.5 Uncertainties and Limitations 3-16
3.6 References 3-18
APPENDIX 3A: METHODOLOGY FOR DEVELOPING AIR QUALITY SURFACES
3A-1
3A.1 Applying Source Apportionment Contributions to Create Air Quality Fields 3A-1
3 A. 1.2 Scaling Ratio Applied to Source Apportionment Tags 3A-3
3A.2 References 3A-5
APPENDIX 3B: OZONE IMPACTS OF ALTERNATIVE CONTROL CASES 3B-1
3B.1 Analysis of Emissions Reductions 3B-2
3B.2 Projected Impacts on Ozone Design Values 3B-6
3B.3 Alternative Control Case Projected Ozone Design Values 3B-12
3B.4 Projected Impacts on Downwind Contributions 3B-19
vi
-------
CHAPTER 4: COST, EMISSIONS, AND ENERGY IMPACTS 4-1
Overview 4-1
4.1 Regulatory Control Alternatives 4-1
4.1.1 EGU Regulatory Control Alternatives Analyzed 4-5
4.1.2 Non-EGU Regulatory Control Alternatives Analyzed 4-8
4.2 Power Sector Modeling Framework 4-11
4.3 EPA's Power Sector Modeling of the Baseline run and Three Regulatory Control
Alternatives 4-14
4.3.1 EPA's IPM Baseline run v.6.20 4-14
4.3.2 Methodology for Evaluating the Regulatory Control Alternatives 4-15
4.3.3 Methodology for Estimating Compliance Costs 4-20
4.4 Analytical Framework for Emission Reduction Assessment for Non-EGUs 4-22
4.5 Estimated Impacts of the Regulatory Control Alternatives 4-27
4.5.1 Emission Reduction Assessment for EGUs 4-27
4.5.2 Compliance Cost Assessment for EGUs 4-33
4.5.3 Impacts on Fuel Use, Prices and Generation Mix 4-35
4.5.4 Emission Reduction and Compliance Cost Assessment for Non-EGUs from Screening
Assessment for 2026 4-42
4.5.6 Total Emissions Reductions and Compliance Costs for EGUs and Non-EGUs 4-46
4.5.7 Impact of Emissions Reductions on Maintenance and Nonattainment Monitors 4-47
4.6 Social Costs 4-47
4.7 Limitations 4-51
4.8 References 4-54
CHAPTER 5: BENEFITS 5-1
Overview 5-1
5.1 Estimated Human Health Benefits 5-2
5.1.1 Health Impact Assessment for Ozone and PM2.5 5-4
5.1.1.1 Selecting Air Pollution Health Endpoints to Quantify 5-5
5.1.1.2 Calculating Counts of Air Pollution Effects Using the Health Impact Function. 5-9
5.1.1.3 Quantifying Cases of Ozone-Attributable Premature Death 5-10
5.1.1.4 Quantifying Cases of PM2.5-Attributable Premature Death 5-11
5.1.1.5 Applying PM2.5 Benefit per Ton Values 5-12
5.1.3 Characterizing Uncertainty in the Estimated Benefits 5-15
5.1.4 Estimated Number and Economic Value of Health Benefits 5-18
5.2 Climate Benefits from Reducing CO2 5-26
vii
-------
5.3 Additional Unquantified Benefits 5-31
5.3.1 NO; Health Benefits 5-34
5.3.2 SO; Health Benefits 5-34
5.3.3 Ozone Welfare Benefits 5-35
5.3.4 NO2 and SO2 Welfare Benefits 5-35
5.3.5 Visibility Impairment Benefits 5-36
5.3.6 Water Quality and Availability Benefits 5-37
5.3.7 Hazardous Air Pollutant Impacts 5-42
5.4 References 5-42
CHAPTER 6: ECONOMIC IMPACTS 6-1
Overview 6-1
6.1 Small Entity Analysis 6-1
6.1.1 EGU Small Entity Analysis and Results 6-2
6.1.2 Non-EGU Small Entity Impacts and Results 6-9
6.1.3 Conclusion 6-11
6.2 Labor Impacts 6-12
6.2.1 EGU Labor Impacts 6-13
6.2.2 Overview of Methodology 6-14
6.2.3 Overview of Power Sector Employment 6-15
6.2.4 Projected Sectoral Employment Changes due to the Proposed Rule 6-16
6.2.5 Non-EGU Labor Impacts 6-18
6.2.6 Conclusions 6-20
6.3 References 6-21
CHAPTER 7: ENVIRONMENTAL JUSTICE IMPACTS 7-1
7.1 Introduction 7-1
7.2 Analyzing EJ Impacts in This Proposal 7-3
7.3 Demographic Proximity Analyses 7-4
7.3.1 EGU and Non-EGU Proximity Assessments 7-5
7.3.2 Tribal Lands Proximity Assessment 7-9
7.4 Ozone Exposure Analysis 7-10
7.4.1 Aggregated Results 7-12
7.4.1.1 Baseline Assessment 7-12
7.4.1.2 Regulatory Alternatives Assessment 7-14
7.4.2 Distributional Results 7-20
viii
-------
7.4.2.1 Baseline Assessment 7-21
7.4.2.2 Regulatory Alternatives Assessment 7-23
7.5 Qualitative Assessment of CO2 7-26
7.6 Qualitative Assessment of PM2.5 7-29
7.7 Summary 7-30
7.8 References 7-32
CHAPTER 8: COMPARISON OF BENEFITS AND COSTS 8-1
Overview 8-1
8.1 Results 8-2
ix
-------
LIST OF TABLES
Table ES-1. Regulatory Control Alternatives for EGUs and Non-EGUs ES-6
Table ES-2. EGU Ozone Season NOx Emissions Changes and Annual Emissions Changes for
NOx, SO2, PM2.5, and CO2 for the Regulatory Control Alternatives from 2023 - 2042 ES-9
Table ES-3. Non-EGU Emissions Unit Types, Emissions Limits, and Industries ES-11
Table ES-4. Ozone Season (OS) NOx Emissions and Emissions Reductions (tons) for the
Proposed Rule and the Less and More Stringent Alternatives ES-12
Table ES-5. By Industry, Number and Type of Emissions Units and Total Estimated Emissions
Reductions (ozone season tons) ES-13
Table ES-6. Total National Compliance Cost Estimates (millions of 2016$) for the Proposed
Rule and the Less and More Stringent Alternatives ES-14
Table ES-7. Estimated Avoided Ozone-Related Premature Respiratory Mortalities and Illnesses
for the Proposal and More and Less Stringent Alternatives for 2023 (95% Confidence Interval)
ES-16
Table ES-8. Estimated Avoided Ozone-Related Premature Respiratory Mortalities and Illnesses
for the Proposal and More and Less Stringent Alternatives for 2026 (95% Confidence Interval)
ES-17
Table ES-9. Estimated Discounted Economic Value of Avoided Ozone and PM2.5-Attributable
Premature Mortality and Illness for the Proposed Policy Scenarios in 2023 (95% Confidence
Interval; millions of 2016$) ES-19
Table ES-10. Estimated Discounted Economic Value of Avoided Ozone and PM2.5-Attributable
Premature Mortality and Illness for the Proposed Policy Scenario in 2026 (95% Confidence
Interval; millions of 2016$) ES-20
Table ES-11. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less and
More Stringent Alternatives for 2023 for the U.S. (millions of 2016$) ES-24
Table ES-12. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less and
More Stringent Alternatives for 2026 for the U.S. (millions of 2016$) ES-25
Table ES-13. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less and
More Stringent Alternatives for 2030 for the U.S. (millions of 2016$) ES-25
Table ES-14. Summary of Present Values and Equivalent Annualized Values for the 2023-2042
Timeframe for Estimated Monetized Compliance Costs, Benefits, and Net Benefits for the
Proposed Rule (millions of 2016$, discounted to 2022) ES-27
Table 1-1. Regulatory Control Alternatives for EGUs and Non-EGUs 1-5
x
-------
Table 2-1. Total Net Summer Electricity Generating Capacity by Energy Source, 2014 and 2020
2-3
Table 2-2. Net Generation in 2014 and 2020 (Trillion kWh = TWh) 2-6
Table 2-3. Coal and Natural Gas Generating Units, by Size, Age, Capacity, and Average Heat
Rate in 2020 2-7
Table 2-4. Total U.S. Electric Power Industry Retail Sales, 2014 and 2020 (billion kWh) 2-11
Table 3B-1. Total anthropogenic 2023 base case NOx emissions, emissions deltas, and percent
reductions by state for each alternative control case 3B-3
Table 3B-2. Total 2026 base case NOx emissions, emissions deltas, and percent reductions by
state for each alternative control case 3B-4
Table 3B-3. Impact on projected 2023 design value of the emissions reductions in the proposed
case, the less stringent case and more stringent case (ppb) 3B-6
Table 3B-4. Impact on projected 2026 design value of the emissions reductions in the proposed
case and the less stringent and more stringent cases (ppb) 3B-9
Table 3B-5. Projected average and maximum design values for the 2023 base case, the proposed
case, less stringent case, and more stringent case (ppb) 3B-12
Table 3B-6. Projected average and maximum design values for the 2026 base case, the proposed
case, less stringent case, and more stringent case (ppb) 3B-16
Table 3B-7. Largest reduction in downwind contribution for 19 upwind states for each
alternative control case 3B-19
Table 4-1. Non-EGU Emissions Unit Types, Emissions Limits, and Industries 4-4
Table 4-2. Regulatory Control Alternatives for EGUs and Non-EGUs 4-4
Table 4-3. Illustrative NOx Ozone Season Emission Budgets (Tons) Evaluated by IPM Run Year
4-7
Table 4-4. Ozone Season (OS) NOx Emissions and Emissions Reductions for the Proposed Rule
and the Less and More Stringent Alternatives 4-10
Table 4-5. Summary of Methodology for Calculating Compliance Costs Estimated Outside of
IPM for Proposed I IP for the 2015 Ozone NAAQS, 2023 (2016$) 4-22
Table 4-6. EGU Ozone Season NOx Emissions and Emissions Changes (thousand tons) for the
Baseline run and the Regulatory Control Alternatives from 2023 - 2042 4-28
Table 4-7. EGU Annual Emissions and Emissions Changes for NOx, SO2, PM2.5, and CO2 for
the Regulatory Control Alternatives for 2023-2042 4-30
XI
-------
Table 4-8. National Power Sector Compliance Cost Estimates (millions of 2016$) for the
Regulatory Control Alternatives 4-33
Table 4-9. 2023 and 2025 Projected U.S. Power Sector Coal Use for the Baseline and the
Regulatory Control Alternatives 4-36
Table 4-10. 2023 and 2025 Projected U.S. Power Sector Natural Gas Use for the Baseline and
the Regulatory Control Alternatives 4-37
Table 4-11. 2023 and 2025 Projected Minemouth and Power Sector Delivered Coal Price
(2016$) for the Baseline and the Regulatory Control Alternatives 4-37
Table 4-12. 2023 and 2025 Projected Henry Hub and Power Sector Delivered Natural Gas Price
(2016$) for the Baseline and the Regulatory Control Alternatives 4-37
Table 4-13. 2023 and 2025 Projected U.S. Generation by Fuel Type for the Baseline and the
Regulatory Control Alternatives 4-38
Table 4-14. 2023 and 2025 Projected U.S. Capacity by Fuel Type for the Baseline run and the
Regulatory Control Alternatives 4-39
Table 4-15. Average Retail Electricity Price by Region for the Baseline and the Regulatory
Control Alternatives, 2023 4-40
Table 4-16. Average Retail Electricity Price by Region for the Baseline and the Regulatory
Control Alternatives, 2025 4-41
Table 4-17. Annual Estimated Emissions Reductions for 2026-2042 (ozone season tons) and
Annual Total Costs for the Proposed Rule 4-44
Table 4-18. By Industry, Number and Type of Emissions Units and Total Estimated Emissions
Reductions (ozone season tons) 4-45
Table 4-19. Annual Estimated Emissions Reductions for 2026-2042 (ozone season tons) and
Annual Total Costs for the Less and More Stringent Alternatives 4-45
Table 4-20. Total Annual Estimated NOx Emissions Reductions (ozone season, thousand tons)
and Compliance Costs (million 2016$), 2023-2042 4-46
Table 4-21. Total National Compliance Cost Estimates (millions of 2016$) for the Proposed Rule
and the Less and More Stringent Alternatives 4-47
Table 5-1. Health Effects of Ambient Ozone and PM2.5 5-8
Table 5-2. Estimated Avoided Ozone-Related Premature Respiratory Mortalities and Illnesses
for the Proposal and More and Less Stringent Alternatives for 2023 (95% Confidence Interval)
5-19
Xll
-------
Table 5-3. Estimated Avoided Ozone-Related Premature Respiratory Mortalities and Illnesses
for the Proposal and More and Less Stringent Alternatives for 2026 (95% Confidence Interval)
5-20
Table 5-4. Estimated Discounted Economic Value of Avoided Ozone and PM2.5-Attributable
Premature Mortality and Illness for the Proposed Policy Scenarios in 2023 (95% Confidence
Interval; millions of 2016$) 5-22
Table 5-5. Estimated Discounted Economic Value of Avoided Ozone and PM2.5-Attributable
Premature Mortality and Illness for the Proposed Policy Scenario in 2026 (95% Confidence
Interval; millions of 2016$) 5-23
Table 5-6. Stream of Human Health Benefits from 2023 through 2042: Monetized Benefits
Quantified as Sum of Long-Term Ozone Mortality for EGUs and Non-EGUs and Long-Term
BPT PM2.5 Mortality for EGUs (Discounted at 3%; millions of 2016$) 5-24
Table 5-7. Stream of Human Health Benefits from 2023 through 2042: Monetized Benefits
Quantified as Sum of Short-Term Ozone Mortality for EGUs and Non-EGUS and Long-Term
BPT PM2.5 Mortality for EGUs (Discounted at 7%; millions of 2016$) 5-25
Table 5-8. Illustrative Estimates of PM2.5-Attributable Premature Mortality and Illnesses for the
Proposal for Non-EGUs (millions of 2016$) 5-26
Table 5-9. Unquantified Health and Welfare Benefits Categories 5-31
Table 6-1. SBA Size Standards by NAICS Code 6-4
Table 6-2. Projected Impact of the Proposed FIP for the 2015 Ozone NAAQS on Small Entities
in 2023 6-8
Table 6-3. Projected Impact of the Proposed FIP for the 2015 Ozone NAAQS on Small Entities
in 2026 6-9
Table 6-4. Non-EGU SBA Size Standards by NAICS Code 6-10
Table 6-5. Summary of Sales Test Ratios for 2026 for Firms Affected by Proposed Rule 6-11
Table 6-6. Summary of Small Parent Company Small Business Size Standards 6-11
Table 6-7. Changes in Labor Utilization: Construction-Related (Number of Job-Years of
Employment in a Single Year) 6-17
Table 6-8. Changes in Labor Utilization: Recurring Non-Construction (Number of Job-Years of
Employment in a Single Year) 6-17
Table 6-9. Relevant Industry Employment (2020) 6-19
Table 6-10. Employment per $1 million Output in the Tier 1 Industries 6-20
-------
Table 7-1. Population Demographics for EGU Facilities
7-7
Table 7-2. Population Demographics for Non-EGU Facilities 7-8
Table 7-3. Tribal Proximity Assessment 7-9
Table 7-4. Populations Included in the Ozone Exposure Analysis 7-10
Table 8-1. EGU Ozone Season NOx Emissions Changes and Annual Emissions Changes for
NOx, SO2, PM2.5, and CO2 for the Regulatory Control Alternatives from 2023 - 2042 8-3
Table 8-2. Non-EGU Ozone Season (OS) NOx Emissions and Emissions Reductions for the
Proposed Rule and the Less and More Stringent Alternatives 8-5
Table 8-3. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less and More
Stringent Alternatives for 2023 for the U.S. (millions of 2016$) 8-6
Table 8-4. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less and More
Stringent Alternatives for 2026 for the U.S. (millions of 2016$) 8-7
Table 8-5. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less and More
Stringent Alternatives for 2030 for the U.S. (millions of 2016$) 8-7
Table 8-6. Summary of Present Values and Equivalent Annualized Values for the 2023-2042
Timeframe for Estimated Monetized Compliance Costs, Benefits, and Net Benefits for the
Proposed Rule (millions of 2016$, discounted to 2022) 8-9
xiv
-------
LIST OF FIGURES
Figure 2-1. National New Build and Retired Capacity (MW) by Fuel Type, 2014-2020 2-4
Figure 2-2. Regional Differences in Generating Capacity (MW), 2020 2-5
Figure 2-3. Cumulative Distribution in 2019 of Coal and Natural Gas Electricity Capacity and
Generation, by Age 2-8
Figure 2-4. Fossil Fuel-Fired Electricity Generating Facilities, by Size 2-9
Figure 2-5. Real National Average Electricity Prices (including taxes) for Three Major End-Use
Categories 2-12
Figure 2-6. Relative Increases in Nominal National Average Electricity Prices for Major End-
Use Categories (including taxes), With Inflation Indices 2-13
Figure 2-7. Real National Average Electricity Prices for Three Major End-Use Categories
(including taxes), 1960-2020 (2019$) 2-14
Figure 2-8. Relative Change in Real National Average Electricity Prices (2019$) for Three Major
End-Use Categories (including taxes) 2-14
Figure 2-9. Relative Real Prices of Fossil Fuels for Electricity Generation; Change in National
Average Real Price per MMBtu Delivered to EGU 2-15
Figure 2-10. Relative Growth of Electricity Generation, Population, Real GDP Since 2014...2-16
Figure 2-11. Relative Change of Real GDP, Population and Electricity Generation Intensity
Since 2014 2-17
Figure 2-12. Status of State Electricity Industry Restructuring Activities 2-18
Figures 2-13. and 2-14. Capacity and Generation Mix by Ownership Type, 2014 & 2020 2-20
Figure 2-15. Geographical Distribution ofNon-EGU Ozone Season NOx Reductions and
Summary of Reductions by Industry and by State 2-21
Figure 3-1. Air Quality Modeling Domain 3-3
Figure 3-2. Baseline AS-M03 concentration in 2023 (ppb) 3-8
Figure 3-3. Baseline AS-M03 concentration in 2026 (ppb) 3-9
Figure 3-4. Reduction in AS-M03 (ppb): 2023 baseline - less stringent EGU-only alternative
(scale: + 0.1 ppb) 3-10
Figure 3-5. Reduction in AS-M03 (ppb): 2023 baseline - EGU-only proposed rule alternative
(scale: + 0.1 ppb) 3-11
xv
-------
Figure 3-6. Reduction in AS-M03 (ppb): 2023 baseline - more stringent EGU-only alternative
(scale: + 0.1 ppb) 3-11
Figure 3-7. Reduction in AS-M03 (ppb): 2026 baseline - less stringent EGU-only alternative
(scale: + 1 ppb) 3-12
Figure 3-8. Reduction in AS-M03 (ppb): 2026 baseline - EGU-only proposed rule alternative
(scale: + 1 ppb) 3-12
Figure 3-9. Reduction in AS-M03 (ppb): 2026 baseline - more stringent EGU-only alternative
(scale: + 1 ppb) 3-13
Figure 3-10. Reduction in AS-M03 (ppb): 2026 baseline - less stringent non-EGU-only
alternative (scale: + 1 ppb) 3-13
Figure 3-11. Reduction in AS-M03 (ppb): 2026 baseline - non-EGU-only proposed rule
alternative (scale: + 1 ppb) 3-14
Figure 3-12. Reduction in AS-M03 (ppb): 2026 baseline - more stringent non-EGU-only
alternative (scale: + 1 ppb) 3-14
Figure 3-13. Reduction in AS-M03 (ppb): 2026 baseline - less stringent EGU+non-EGU
alternative (scale: + 1 ppb) 3-15
Figure 3-14. Reduction in AS-M03 (ppb): 2026 baseline - EGU+non-EGU proposed rule
alternative (scale: + 1 ppb) 3-15
Figure 3-15. Reduction in AS-M03 (ppb): 2026 baseline - more stringent EGU+non-EGU
alternative (scale: + 1 ppb) 3-16
Figure 4-1. Ozone Season NOx Reductions and Costs per Ton (CPT) for Tier 1, Tier 2
Industries, and All Industries (2016$) 4-26
Figure 4-2. Electricity Market Module Regions 4-42
Figure 5-1. Stylized Relationship between the PM2.5 Concentrations Considered in Epidemiology
Studies and our Confidence in the Estimated PM-related Premature Deaths 5-17
Figure 7-1. Heat Map of the National Average AS-M03 Ozone Concentrations Across
Demographic Groups in the Baseline Assessment (ppb) 7-14
Figure 7-2. Heat Map of the National Average AS-M03 Ozone Concentration Reductions by
Demographic Group, Regulatory Alternative, and Affected Facilities (ppb) 7-17
Figure 7-3. Heat Map of State Average AS-M03 Ozone Concentration Reductions by
Demographic Group for EGUs and Non-EGUs Under the Proposed Rule (ppb) 7-19
Figure 7-4. Distributions of Baseline Ozone Concentrations Across Populations in 2023 and
2026 7-22
xvi
-------
Figure 7-5. Distributions of Ozone Concentration Reductions from EGUNOx Emission
Reductions Across Regulatory Alternatives and Populations in 2023 7-24
Figure 7-6. Distributions of Ozone Concentration Reductions from NOx Emission Reductions
Across Affected Facilities, Regulatory Alternatives, and Populations in 2026 7-26
xvii
-------
EXECUTIVE SUMMARY
Overview
This regulatory impact analysis (RIA) supports the proposed rule, the Federal
Implementation Plan Addressing Regional Ozone Transport for the 2015 Ozone National
Ambient Air Quality Standards (FIP for the 2015 ozone NAAQS). In the proposal, in accordance
with the Wisconsin decision, EPA proposes implementation mechanisms to achieve enforceable
emissions reductions required to eliminate significant contribution of ozone precursor emissions
prior to the 2023 ozone season. The initial phase of proposed emissions reductions will therefore
be achieved prior to the August 2, 2024, attainment date for areas classified as Moderate
nonattainment for the 2015 ozone NAAQS.1
EPA is proposing to promulgate new or revised FIPs for 25 states that include new NOx
ozone season emission budgets for electric generating units (EGU) sources, with implementation
of these emission budgets beginning in the 2023 ozone season.2 EPA is also proposing to adjust
these states' emission budgets for each ozone season thereafter to maintain the initial stringency
of the emissions budget, accounting for retirements and other changes to the EGU fleet over
time. EPA is also proposing to extend the Cross-State Air Pollution Rule (CSAPR) NOx Ozone
Season Group 3 Trading Program beginning in the 2023 ozone season through the 2025 ozone
season. Further, EPA is proposing to establish new emissions budgets for the CSAPR NOx
Ozone Season Group 3 Trading Program beginning in the 2026 ozone season, as discussed in
Section VII.B.l. of the preamble. EPA is also proposing to retain two states, Iowa, and Kansas,
in the CSAPR NOx Ozone Season Group 2 Trading Program.
1 The 2015 ozone NAAQS is an 8-hour standard that was set at 70 parts per billion (ppb). See 80 FR 65291
(December 28, 2015).
2 In 2023, the 25 states with EGU reduction requirements include AL, AR, DE, IL, IN, KY, LA, MD, MI, MN, MS,
MO, NV, NJ, NY, OH, OK, PA, TN, TX, UT, VA, WV, WI, and WY. There are no EGU reductions being required
from California, and Oregon's SIP was previously approved.
ES-1
-------
For non-electric generating units (non-EGUs), EPA is proposing to promulgate new FIPs
for 23 states that include new NOx emissions limitations, with initial compliance dates for these
emissions limitations beginning in 2026.3
Consistent with OMB Circular A-4 and EPA's Guidelines for Preparing Economic
Analyses (2010), this RIA presents the benefits and costs of the proposed rule from 2023 through
2042. The estimated benefits are those health benefits expected to arise from reduced PM2.5 and
ozone concentrations. The estimated costs for EGUs are the costs of installing and operating
controls and the increased costs of producing electricity. The estimated costs for non-EGUs are
the costs of installing and operating controls to meet the ozone season emissions limits.4 The
estimated costs for non-EGUs do not include monitoring, recordkeeping, reporting, or testing
costs. Unquantified benefits and costs are described qualitatively. The RIA also provides
estimates of other impacts of the proposed rule including its effect on retail electricity prices and
fuel production.
ES.l Identifying Needed Emissions Reductions
To reduce interstate emission transport under the authority provided in CAA section
110(a)(2)(D)(i)(I), the proposed rule further limits ozone season NOx emissions from EGUs and
non-EGUs using the same framework used by EPA in developing the CSAPR. The Interstate
Transport Framework provides a 4-step process to address the requirements of the good neighbor
provision for ground-level ozone and fine particulate matter (PM2.5) NAAQS: (1) identifying
downwind receptors that are expected to have problems attaining or maintaining the NAAQS;
(2) determining which upwind states contribute to these identified problems in amounts
sufficient to "link" them to the downwind air quality problems (i.e., here, an amount of
contribution equal to or greater than 1 percent of the NAAQS); (3) for states linked to downwind
3 In 2026, the 23 states with non-EGU reduction requirements include AR, CA, IL, IN, KY, LA, MD, MI, MN, MS,
MO, NV, NJ, NY, OH, OK, PA, TX, UT, VA, WV, WI, and WY. AL, DE, and TN are not linked in 2026, and
Oregon's SIP was previously approved.
4 We prepared a non-EGU screening assessment (for more details on the screening assessment, see memorandum
titled Screening Assessment of Potential Emissions Reductions, Air Quality Impacts, and Costs from Non-EGU
Emissions Units for 2026 in the docket), which includes estimated emissions reductions and costs. These estimates
are proxies for (1) non-EGU emissions units that have emission reduction potential, (2) potential controls for and
emissions reductions from these emissions units, and (3) control costs from the potential controls on these emissions
units. This screening assessment is not intended to be, nor take the place of, a unit-specific detailed engineering
analysis that fully evaluates the feasibility of retrofits for the emissions units, potential controls, and related costs.
ES-2
-------
air quality problems, identifying upwind emissions that significantly contribute to downwind
nonattainment or interfere with downwind maintenance of the NAAQS; and (4) for states that are
found to have emissions that significantly contribute to nonattainment or interfere with
maintenance of the NAAQS downwind, implementing the necessary emissions reductions
through enforceable measures. In this action, EPA applies this 4-step Interstate Transport
Framework for the proposed FIP for the 2015 ozone NAAQS.
For EGUs, in identifying levels of uniform control stringency EPA assessed the same
NOx emissions controls that the Agency analyzed in the CSAPR Update and the Revised
CSAPR Update, all of which are considered to be widely available for EGUs: (1) fully operating
existing SCR, including both optimizing NOx removal by existing operational SCRs and turning
on and optimizing existing idled SCRs; (2) installing state-of-the-art NOx combustion controls;
(3) fully operating existing SNCRs, including both optimizing NOx removal by existing
operational SNCRs and turning on and optimizing existing idled SNCRs; (4) installing new
SNCRs; (5) installing new SCRs; and (6) generation shifting (i.e., emission reductions
anticipated to occur from generation shifting from higher to lower emitting units). Levels of
uniform control stringency were represented by $1,800 per ton of NOx (2016$) in 2023 and
$11,000 per ton of NOx (2016$) in 2026.5
For non-EGUs, in identifying appropriate control strategies EPA developed an analytical
framework6 to evaluate the air quality impacts of potential emissions reductions from non-EGU
sources located in the linked upwind states. EPA incorporated air quality modeling information,
annual emissions, and information about potential controls to estimate the NOx emissions
reduction potential from non-EGU sources to determine which non-EGU industries, if subject to
further control requirements, would have the greatest impact in providing air quality
improvements at the downwind receptors. The evaluation in the analytical framework was
subject to a marginal cost threshold of up to $7,500 per ton (2016$), which EPA determined
based on information available to the Agency about existing control device efficiency and cost
5 EGU NOx Mitigation Strategies Proposed Rule TSD, in the docket for this proposed rule (Docket ID No. EPA-
HQ-OAR-2021 -0688).
6 Additional information on the analytical framework is presented in the memorandum titled Screening Assessment
of Potential Emissions Reductions, Air Quality Impacts, and Costs from Non-EGU Emissions Units for 2026, which
is available in the docket for this proposed rulemaking.
ES-3
-------
information. In the framework, EPA identified emissions unit types in seven industries that
provide opportunities for NOx emissions reductions that result in meaningful impacts on air
quality at the downwind receptors. Because EPA determined that 2026 was the earliest potential
date by which controls on non-EGU emissions units could be installed, EPA used the analytical
framework with air quality modeling information for 2026 to prepare a non-EGU screening
assessment for 2026 that provided estimates of emissions reductions and compliance costs.
Additional information on the results of the screening assessment for 2026 is presented in the
memorandum titled Screening Assessment of Potential Emissions Reductions, Air Quality
Impacts, and Costs from Non-EGU Emissions Units for 2026, which is available in the docket
for this proposed rulemaking.
ES.2 Baseline and Analysis Years
The proposed rule sets forth the requirements to eliminate states' significant contribution
to downwind nonattainment or interference with maintenance of the 2015 ozone NAAQS. To
develop and evaluate control strategies for addressing these obligations, it is important to first
establish a baseline projection of air quality in the analysis year of 2023, taking into account
currently on-the-books Federal regulations, substantial Federal regulatory proposals,
enforcement actions, state regulations, population, expected electricity demand growth, and
where possible, economic growth. Establishing this baseline for the analysis then allows us to
estimate the incremental costs and benefits of the additional emissions reductions that will be
achieved by the proposed rule.
The analysis in this RIA focuses on benefits, costs, and certain impacts from 2023 through
2042. We focus on 2023 because it is by the 2023 ozone season, corresponding with the 2024
Moderate area attainment date, that significant contribution from upwind states' must be
eliminated to the extent possible. It is also the first year in which state-of-the-art combustion
controls can be installed on some EGUs. In addition, impacts for 2026 are important because it is
in this period that additional NOx control technologies for EGUs and non-EGUs are expected to
be installed where upwind linkage to downwind receptors persists. Costs and benefits from
control installations may persist beyond 2026, and the RIA provides costs and benefits through
2042.
ES-4
-------
ES.3 Air Quality Modeling
The air quality modeling for the proposed FIP for the 2015 ozone NAAQS used a 2016-
based modeling platform that included meteorology and base year emissions from 2016 and
projected emissions for 2023 and 2026. The air quality modeling included photochemical model
simulations for a 2016 base year and 2023 and 2026 future years to provide hourly
concentrations of ozone nationwide. In addition, source apportionment modeling was performed
for 2026 to quantify the contributions to ozone from NOx emissions from EGUs and from point
sources other than EGUs (i.e., non-EGUs) on a state-by-state basis. The modeling results for
2016, 2023, and 2026, in conjunction with emissions data for the 2023 and 2026 baseline, the
proposal, and more and less stringent alternatives (regulatory control alternatives), were used to
construct the air quality spatial fields that reflect the influence of emissions changes between the
baseline and the regulatory control alternatives.
The air quality model simulations (i.e., model runs) were performed using the
Comprehensive Air Quality Model with Extensions (CAMx) version 7.10 (Ramboll Environ,
2021). Our CAMx nationwide modeling domain (i.e., the geographic area included in the
modeling) covers all lower 48 states plus adjacent portions of Canada and Mexico using a
horizontal grid resolution of 12 x 12 km.
Spatial fields provide the air quality inputs to potentially calculate health benefits for the
proposed FIP for the 2015 ozone NAAQS. The spatial fields were constructed based on a method
that utilizes ozone contributions from emissions in individual states and state-level emissions
reductions for each of the regulatory control alternatives coupled with baseline spatial fields of
ozone concentrations. This method, as described in Appendix 3 A, was originally developed to
support the RIA for the Repeal of the Clean Power Plan, and the Emission Guidelines for
Greenhouse Gas Emissions from Existing Electric Utility Generating Units and, most recently,
the RIA for the Revised CSAPR Update final rule.
We generated spatial fields of seasonal ozone concentrations associated with the regulatory
control alternatives. The data for creating spatial fields for each scenario include: (1) EGU and
non-EGU ozone season NOx emissions for the 2023 and 2026 baseline scenarios and the
ES-5
-------
regulatory control alternatives, (2) spatial fields of April through September MDA87 (AS-M03)
average ozone for the 2023 and 2026 baseline scenarios, and (3) the spatial field of mean AS-
M03 ozone contributions for the hours that correspond to the time periods of MDA8
concentrations. To calculate ozone-related benefits in 2023 and 2026 we used the ozone season
EGU and non-EGU NOx emissions for the 2023 and 2026 baseline scenarios along with
emissions for the regulatory control alternatives.
ES.4 Control Strategies and Emissions Reductions
The RIA analyzes emission budgets for EGUs and ozone season emissions limits for non-
EGUs, as well as a more and a less stringent alternative to the proposed rule. The more and less
stringent alternatives differ from the proposed FIP for the 2015 ozone NAAQS in that they set
different EGU NOx ozone season emission budgets and different dates for compliance with unit-
specific emission rate limits for the affected EGUs and cover different industries or emissions
units for non-EGUs. Table ES-1 below presents the less stringent alternatives, proposed rule
requirements, and more stringent alternatives for EGUs and non-EGUs. For the purposes of
summarizing the results of the benefits and costs of these alternatives, the less stringent
alternative for EGUs is presented with the less stringent alternative for non-EGUs. However, the
cost, emissions, and energy impacts for the EGU and non-EGU alternatives are evaluated
separately.
Table ES-1. Regulatory Control Alternatives for EGUs and Non-EGUs
Regulatory Control NOx Controls Implemented for EGUs within IPM
Alternative
1) 2023 onwards: Shift generation to minimize costs
2) 2023 onwards: Fully operate existing SCRs during ozone season
3) 2023 onwards: Fully operate existing SNCRs during ozone season
4) In 2023 install state-of-the-art combustion controls
5) In 2028 model run year, impose backstop emission rate limits on coal units
Less Stringent Alternative greater than 100 MW within the 23-state region that lack SCR controls,
forcing units to retrofit or retire.
6) In 2028 model run year, impose backstop emission rate limits on oil/gas
steam units greater than 100 MW that operated at a greater than 20% capacity
factor historically within the 23-state region that lack SCR controls, forcing
units to retrofit or retire.8
7 MDA8 is defined as maximum daily 8-hour average ozone concentration, and MDA1 is defined as the maximum
daily 1-hour ozone concentration.
8 The 20% capacity factor cutoff applied is representative of the fleet of O/G steam units assumed to have SCR
retrofit potential in its state budgets. In the proposal, EPA defined this segment using 150 tons per season cutoff,
which provides a similar size of the O/G steam fleet as the 20% capacity factor value used in this analysis.
ES-6
-------
Regulatory Control
Alternative
NOx Controls Implemented for EGUs within IPM
(All Controls above and)
In 2026, impose backstop emission rate limits on coal units greater than 100
MW within the 23-state region that lack SCR controls, forcing units to retrofit
or retire.
In 2026, impose backstop emission rate limits on oil/gas steam units greater
than 100 MW that operated at a greater than 20% capacity factor historically
within the 23-state region that lack SCR controls, forcing units to retrofit or
retire.
More Stringent Alternative
9)
(Controls 1 - 4, 7 and 8 above and)
In 2026, impose backstop emission rate limits on all oil/gas steam units
greater than 100 MW within the 23-state region that lack SCR controls,
forcing units to retrofit or retire.
NOx Emissions Limits for Non-EGUs - Emissions Unit Types and Industries
1)
Reciprocating internal combustion engines in Pipeline Transportation of
Natural Gas,
Less Stringent Alternative
2)
3)
4)
Kilns in Cement and Cement Product Manufacturing,
Boilers and furnaces in Iron and Steel Mills and Ferroalloy Manufacturing,
Furnaces in Glass and Glass Product Manufacturing, and
Proposed Rule
5)
(All emissions unit types and industries above and)
Impactful boilers* in Basic Chemical Manufacturing, Petroleum and Coal
Products Manufacturing, and Pulp, Paper, and Paperboard Mills.
More Stringent Alternative
6)
(All emissions unit types and industries above and)
All boilers in Basic Chemical Manufacturing, Petroleum and Coal Products
Manufacturing, and Pulp, Paper, and Paperboard Mills.
impactful boilers are boilers with design capacity of 100 mmBtu/hr or greater.
ES. 4.1 EGUs
The proposal establishes NOx emissions budgets requiring fossil fuel-fired power plants
(EGUs) in 25 states to participate in an allowance-based ozone season (May 1 through
September 30) trading program beginning in 2023. The EGUs covered by the proposed FIPs and
subject to the budget are fossil-fired EGUs with >25 megawatt (MW) capacity. For details on the
derivation of these budgets, please see Section VI.C. of the preamble.
The proposed FIP requirements establish ozone season NOx emissions budgets for EGUs
in 25 states starting in 2023 and require EGUs in these states to participate in a revised version of
the CSAPR NOx Ozone Season Group 3 Trading Program that was previously established in the
Revised CSAPR Update.9 In addition, beginning in the 2027 ozone season, coal facilities greater
than 100 MW lacking SCR controls and certain oil/gas steam facilities greater than 100 MW that
9 As explained in Section VI. C. 1 of the preamble, EPA proposes finding that EGU sources within the State of
California are sufficiently controlled such that no further emission reductions are needed from them to eliminate
significant contribution to downwind states.
7)
Proposed Rule
8)
ES-7
-------
lack existing SCR controls located in these 23 states must meet daily emission rate limits,
effectively forcing affected units to install new SCR controls, find other means of compliance, or
retire. The 36-month timeframe allows for design, permitting, and installation. EPA used a third-
party global engineering consulting firm in the summer of 2021 to further validate its timing
assumptions. While all those stages can occur within 36 months, the point from capital
investment to completion can be well under 36 months. This timeframe has been demonstrated in
prior installations and is consistent with prior EPA rules. The timing is also consistent with
EPA's legal authorities and obligations as discussed in Section VII in the preamble. States that
do not have additional mitigation measures assumed in 2026 continue to remain part of the
revised group 3 Trading Program.
In the proposal, we introduce additional features to the allowance-based trading program
approach for EGUs, including dynamic adjustments of the emissions budgets over time and
backstop daily emission rate limits for most coal-fired units, that were not included in previous
CSAPRNOx ozone season trading programs. These enhancements will help maintain control
stringency over time and improve emissions performance at individual units, offering an extra
measure of assurance that existing pollution controls will be operated during the ozone season.
The analysis incorporates the daily emission rate requirement for units with existing controls by
forcing operation of these controls in the ozone season for affected sources. For affected
uncontrolled units in the 23 states, starting in 2026, the analysis imposes an emission rate
constraint that forces affected units to either install new SCR retrofits, find other means of
compliance, or retire.10 The analysis does not explicitly capture the dynamic budget adjustments
over time, but the forced operation of controls during the ozone season over the forecast period
(even in the absence of binding mass limits) approximates this feature of the program design.
Any new fossil fuel-fired EGU serving a generator with a nameplate capacity exceeding 25 MW
capacity that meets the applicability criteria and is deployed in any of the states covered by this
proposal's EGU ozone-season NOx program would be subject to the same requirements as other
10 The proposed rule assumes SCR retrofit potential starting in 2026 and it is reflected in the 2026 state emissions
budgets. The daily backstop emission rate does not apply until 2027, but the majority of units retrofitting are
anticipated to do so by 2026 to assist with the 2026 state emissions budget compliance. EPA's IPM model run years
are 2026 and 2028. The SCR compliance behavior is generally expected to occur no later than 2027, and in 2026 in
many cases. Therefore, EPA models this daily backstop emission rate in 2026 (when choosing between model run
year 2026 and 2028) to conservatively reflect compliance cost in the first year in which the technology is in place
for some units.
ES-8
-------
covered EGUs. Reported heat input data from any new covered EGUs would be factored into
dynamic budgets through the computational process outlined in the proposal.
For the NOx controls for EGUs identified in Table ES-1, under the proposed rule and the
less stringent and the more stringent alternatives, 248 EGUs not already doing so in 2019 are
projected to fully operate existing SCRs.11 Under the proposed rule and the less stringent and
more stringent alternatives, 27 units are projected to fully operate existing SNCRs. Under the
proposed rule and the less stringent and the more stringent alternatives, 23 units are projected to
install state-of-the-art combustion controls. The book-life of the new combustion controls is
assumed to be 15 years.
Under the proposed rule an incremental 18 GW of coal and 4 GW of oil/gas retirements
are projected by 2030. Under the more stringent alternative 20 GW of coal and 7 GW of oil/gas
retirements are projected by 2030. Under the less stringent alternative 13 GW of coal and 4 GW
of oil/gas retirements are projected by 2030. For additional details, see the EGUNOx Mitigation
Strategies Proposed Rule TSD.
Table ES-2 shows the ozone season NOx emissions reductions expected from the
proposed rule as well as the more and less stringent alternatives analyzed from 2023 through
2030, and for 2035 and 2042. In addition, Table ES-2 shows the annual NOx, SO2, PM2.5, and
CO2 emissions reductions expected from the proposed rule as well as the more and less stringent
alternatives analyzed from 2023 through 2030, and for 2035 and 2042.
Table ES-2. EGU Ozone Season NOx Emissions Changes and Annual Emissions Changes
for NOx, SO2, PM2.5, and CO2 for the Regulatory Control Alternatives from 2023 - 204212
Proposed Rule
Less Stringent
Alternative
More Stringent
Alternative
2023
NOx (ozone season)
6,000
6,000
7,000
NOx (annual)
10,000
10,000
10,000
SO2 (annual)*
--
1,000
2,000
CO2 (annual, thousand metric)
--
--
--
PM2.5 (annual)
2024
NOx (ozone season)
26,000
14,000
29,000
NOx (annual)
42,000
22,000
45,000
S02 (annual)
42,000
20,000
43,000
11 The engineering analysis used to develop the illustrative budgets in the RIA relied on 2019 historical data.
12 This analysis is limited to the geographically contiguous lower 48 states.
ES-9
-------
Proposed Rule
Less Stringent
Alternative
More Stringent
Alternative
CO2 (annual, thousand metric)
18,000
10,000
19,000
PM2.5 (annual)
4,000
1,000
4,000
2025
NOx (ozone season)
46,000
22,000
51,000
NOx (annual)
73,000
33,000
80,000
SO2 (annual)
83,000
39,000
84,000
CO2 (annual, thousand metric)
37,000
19,000
38,000
PM2.5 (annual)
9,000
2,000
9,000
2026
NOx (ozone season)
47,000
32,000
53,000
NOx (annual)
81,000
55,000
87,000
SO2 (annual)
106,000
76,000
108,000
CO2 (annual, thousand metric)
40,000
26,000
42,000
PM2.5 (annual)
9,000
5,000
9,000
2027
NOx (ozone season)
49,000
42,000
54,000
NOx (annual)
88,000
76,000
95,000
S02 (annual)
129,000
113,000
131,000
C02 (annual, thousand metric)
43,000
34,000
46,000
PM2.5 (annual)
10,000
7,000
10,000
2030
NOx (ozone season)
52,000
52,000
57,000
NOx (annual)
96,000
98,000
100,000
SO2 (annual)
104,000
100,000
103,000
CO2 (annual, thousand metric)
50,000
45,000
50,000
PM2.5 (annual)
9,000
9,000
9,000
2035
NOx (ozone season)
49,000
50,000
52,000
NOx (annual)
90,000
93,000
93,000
SO2 (annual)
96,000
93,000
98,000
CO2 (annual, thousand metric)
38,000
36,000
38,000
PM2.5 (annual)
11,000
12,000
10,000
2042
NOx (ozone season)
47,000
47,000
48,000
NOx (annual)
70,000
75,000
71,000
SO2 (annual)
54,000
50,000
54,000
CO2 (annual, thousand metric)
25,000
23,000
24,000
PM2.5 (annual)
8,000
9,000
8,000
Emissions changes for NOx, SO2, and PM2 5 are tons.
* S02 emissions reductions under the proposed rule are 350 tons and rounded to zero. S02 emissions reductions
under the less stringent alternative are 507 tons and rounded to 1000 tons. S02 emissions reductions are 1,699 tons
under the more stringent alternative and rounded to 2,000 tons. Given the rounding, the difference between the
reductions under the proposed rule and the less stringent alternative is approximately 160 tons.
ES.4.2 Non-EGUs
The proposal includes ozone season NOx emissions limitations for non-EGUs with an
ES-10
-------
initial compliance date of 2026 for the 23 states.13 A summary of the non-EGU emissions unit
types, emissions limits, and industries is presented below in Table ES-3. A more detailed
summary of the proposed emissions limits can be found in Section I.B. of the preamble.
Table ES-3. Non-EGU Emissions Unit Types, Emissions Limits, and Industries
Emissions Unit Type
Emissions Limit
Industry
NAICS
Reciprocating internal
combustion engines
g/hp-hr
Pipeline Transportation of Natural Gas
4862
Kilns
lb/ton of clinker
Cement and Concrete Product
Manufacturing
3273
Boilers and furnaces
Depending on equipment type -
lb/mmBtu, lb/ton of steel, lb/ton,
lb/ton coal pushed, lb/ton coal
charged, work practice standards
Iron and Steel Mills and Ferroalloy
Manufacturing
3311
Furnaces
lb/ton glass produced
Glass and Glass Product Manufacturing
3272
Impactful boilers*
lbs NOx/mmBtu
Basic Chemical Manufacturing,
Petroleum and Coal Products
Manufacturing, and Pulp, Paper, and
Paperboard Mills
3251,
3241,
3221
North American Industry Classification System (NAICS)
impactful boilers are boilers with design capacity of 100 mmBtu/hr or greater.
Table ES-4 below provides a summary of the 2019 ozone season emissions for non-EGUs
for the 23 states subject to the proposed FIP in 2026, along with the estimated ozone season
reductions for the proposal and the less and more stringent alternatives for 2026.14 The estimated
emissions reductions by state for the proposed alternative are from the non-EGU screening
assessment, and the estimated reductions by state for the less and more stringent alternatives
were estimated for the RIA using the same methodology. Table ES-5 below shows the industries,
number and type of emissions units expected to install controls, and the total estimated ozone
season emissions reductions, based on the results of the 2026 screening assessment. Additional
results from the screening assessment for 2026 are presented in the memorandum titled
Screening Assessment of Potential Emissions Reductions, Air Quality Impacts, and Costs from
Non-EGU Emissions Units for 2026.
13 If an emissions unit installs SCR or SNCR to meet an emissions limit in response to the proposed FIP that would
be a physical change under new source review (NSR) and lead to an assessment of potential emissions changes. If
the installation of SCR results in an emissions increase that exceeds the thresholds in the NSR regulations for one or
more regulated NSR pollutants, including the netting analysis, the changes would trigger the applicability of NSR.
14 EPA determined that the 2019 inventory was appropriate because it provided a more accurate prediction of
potential near-term emissions reductions. The analysis assumes that the 2019 emissions from the emissions units
will be the same in 2026 and later years.
ES-11
-------
Table ES-4. Ozone Season (OS) NOx Emissions and Emissions Reductions (tons) for the
Proposed Rule and the Less and More Stringent Alternatives*
2019 OS NOx
Emissions
Proposed Rule -
Less Stringent
More Stringent
State
OS NOx
Alternative - OS
Alternative - OS
Reductions
NOx Reductions
NOx Reductions
AR
8,265
1,654
922
1,654
CA
14,579
1,666
1,598
1,777
IL
16,870
2,452
2,452
2,553
IN
19,604
3,175
2,787
3,175
KY
11,934
2,291
2,291
2,291
LA
35,831
6,769
4,121
6,955
MD
2,365
45
45
45
MI
18,996
2,731
2,731
3,093
MN
17,591
673
673
789
MO
9,109
3,103
3,103
3,103
MS
12,284
1,761
1,577
1,761
NJ
2,025
0
0
29
NV
2,418
0
0
0
NY
6,003
500
389
613
OH
19,729
2,790
2,611
2,814
OK
22,146
3,575
3,575
3,871
PA
15,861
3,284
3,132
3,340
TX
47,135
4,440
4,440
6,596
UT
6,276
757
757
757
VA
7,041
1,563
1,465
1,660
WI
6,571
2,150
677
2,234
WV
9,825
982
982
982
WY
10,335
826
826
826
Totals
322,793
47,186
41,153
50,918
* In the non-EGU screening assessment for 2026, EPA estimated emissions reduction potential from the non-EGU
industries and emissions units. In the screening assessment, EPA used CoST to identify emissions units, emissions
reductions, and associated compliance costs to evaluate the effects of potential non-EGU emissions control measures
and technologies. CoST is designed to be used for illustrative control strategy analyses (e.g., NAAQS regulatory
impact analyses) and not for unit-specific, detailed engineering analyses. The estimates from CoST identify proxies
for (1) non-EGU emissions units that have emissions reduction potential, (2) potential controls for and emissions
reductions from these emissions units, and (3) control costs from the potential controls on these emissions units. The
control cost estimates do not include monitoring, recordkeeping, reporting, or testing costs. This screening
assessment is not intended to be, nor take the place of, a unit-specific detailed engineering analysis that fully
evaluates the feasibility of retrofits for the emissions units, potential controls, and related costs. For more
information on CoST, go to the following link: https://www.epa.gov/economic-and-cost-analysis-air-pollution-
regulations/cost-analysis-modelstools-air-pollution.
ES-12
-------
Table ES-5. By Industry, Number and Type of Emissions Units and Total Estimated Emissions Reductions (ozone season tons)
Ozone Season Emission Reductions (tons)
Number of Units by Type by Type Qf Unit
Industry
Region
Boilers
Internal
Industrial
Boilers
Internal
Industrial
Glass and Glass Product Manufacturing
East
-
-
41
-
-
6,367
West
-
-
3
-
-
299
Cement and Concrete Product Manufacturing
East
1
-
39
16
-
5,948
West
-
-
8
-
-
2,128
Iron and Steel Mills and Ferroalloy
East
25
-
15
2,044
-
1,207
Pipeline Transportation of Natural Gas
East
-
296
-
-
22,390
-
West
-
11
-
-
754
-
Basic Chemical Manufacturing
East
17
-
-
1,698
-
-
Petroleum and Coal Products Manufacturing
East
9
-
-
962
-
-
West
1
-
-
68
-
-
Pulp, Paper, and Paperboard Mills
East
25
-
-
3,305
-
-
Blue highlights reflect western state information
ES-13
-------
ES.5 Cost Impacts
Table ES-6 below summarizes the present value (PV) and equivalent annualized value
(EAV) of the total national compliance cost estimates for EGUs and non-EGUs for the proposed
rule and the less and more stringent alternatives. We present the PV of the costs over the twenty-
year period 2023 to 2042. We also present the EAV, which represents a flow of constant annual
values that, had they occurred in each year from 2023 to 2042, would yield a sum equivalent to
the PV. The EAV represents the value of a typical cost for each year of the analysis.
Table ES-6. Total National Compliance Cost Estimates (millions of 2016$) for the Proposed
Rule and the Less and More Stringent Alternatives
Proposed Rule
Less Stringent
Alternative
More Stringent Alternative
3 Percent
7 Percent
3 Percent
7 Percent
3 Percent
7 Percent
Present Value
EGU 2023-2042
$17,000
$11,000
$16,000
$9,400
$23,000
$15,000
Present Value
Non-EGU 2026-2042
$4,800
$3,100
$4,200
$2,700
$5,200
$3,300
Present Value
Total 2023-2042
$22,000
$14,000
$20,000
$12,000
$28,000
$18,000
EGU
Equivalent
Annualized Value
$1,100
$1,000
$1,100
$890
$1,500
$1,400
Non-EGU
Equivalent
Annualized Value
$320
$290
$280
$250
$350
$310
Total
Equivalent
Annualized Value
$1,500
$1,300
$1,300
$1,100
$1,900
$1,700
Note: Values have been rounded to two significant figures
ES.6 Benefits
ES. 6.1 Benefits Estimates
The proposed rule is expected to reduce ozone season and annual NOx emissions. In the
presence of sunlight, NOx, and volatile organic compounds (VOCs) can undergo a chemical
reaction in the atmosphere to form ozone. Reducing NOx emissions generally reduces human
exposure to ozone and the incidence of ozone-related health effects, though the degree to which
ozone is reduced will depend in part on local concentration levels of VOCs. In addition to NOx,
the proposed rule is also expected to reduce emissions of direct PM2.5 and SO2 throughout the
ES-14
-------
year. Because NOx and SO2 are also precursors to secondary formation of ambient PM2.5,
reducing these emissions would reduce human exposure to ambient PM2.5 throughout the year
and would reduce the incidence of PIvfc.s-attributable health effects.
In this RIA for the proposed FIP for the 2015 ozone NAAQS, EPA uses both full-form and
reduced-form techniques to quantify benefits. Both approaches rely on the same methods for
quantifying the number and value of air pollution-attributable premature deaths and illnesses,
which is described in the Technical Support Document for the Final Revised CSAPR Update for
the 2008 Ozone NAAQS titled Estimating PM2.5- and Ozone-Attributable Health Benefits.
Methods used to estimate PM2.5 benefits are described in the TSD titled Estimating the Benefit
per Ton of Reducing Directly-Emitted PM2.5, PM2.5 Precursors and Ozone Precursors from 21
Sectors. Both methods incorporate evidence reported in the most recent completed PM and
Ozone Integrated Science Assessments (ISAs) and accounts for recommendations from the
Science Advisory Board (U.S. EPA 2019a, U.S. EPA 2020b, U.S. EPA-SAB 2019, U.S. EPA-
SAB 2020a). When updating each health endpoint EPA considered: (1) the extent to which there
exists a causal relationship between that pollutant and the adverse effect; (2) whether suitable
epidemiologic studies exist to support quantifying health impacts; (3) and whether robust
economic approaches are available for estimating the value of the impact of reducing human
exposure to the pollutant. Our approach for updating the endpoints and to identify suitable
epidemiologic studies, baseline incidence rates, population demographics, and valuation
estimates is summarized in Chapter 5.
Table ES-7 and Table ES-8 report the estimated number of reduced premature deaths and
illnesses in 2023 and 2026 relative to the baseline along with the 95% confidence interval. The
number of reduced estimated deaths and illnesses from the proposed rule and more and less
stringent alternatives is calculated from the sum of individual reduced mortality and illness risk
across the population. Table ES-9 and Table ES-10 report the estimated economic value of
avoided premature deaths and illness in 2023 and 2026 relative to the baseline along with the
95% confidence interval. In each of these tables, for each discount rate and regulatory control
alternative, multiple benefits estimates are presented reflecting alternative ozone and PM2.5
mortality risk estimates.
ES-15
-------
Table ES-7. Estimated Avoided Ozone-Related Premature Respiratory Mortalities and
Illnesses for the Proposal and More and Less Stringent Alternatives for 2023 (95%
Confidence Interval) a,b
Proposal
More Stringent
Alternative
Less Stringent
Alternative11
Avoided premature respiratory mortalities
Long-
term
Turner et al. (2016)°
44
51
44
exposure
(31 to 57)
(36 to 66)
(31 to 57)
Short-
term
Katsouyanni et al.
(2009)cd and Zanobetti et
2
2.3
2
exposure
al. (2008)dpooled
(0.8 to 3.1)
(0.94 to 3.7)
(0.81 to 3.2)
Morbidity effects
Long-
term
exposure
Asthma onset6
350
(300 to 390)
400
(340 to 450)
350
(300 to 400)
Allergic rhinitis
symptoms®
2,000
(1,000 to 2,900)
2,200
(1,200 to 3,300)
2,000
(1,000 to 2,900)
Hospital admissions—
5.3
6.1
5.3
respiratoryd
(-1.4 to 12)
(-1.6 to 14)
(-1.4 to 12)
Short-
term
ED visits—respiratoryf
110
(30 to 230)
120
(34 to 260)
110
(30 to 230)
Asthma symptoms
62,000
(-7,700 to 130,000)
71,000
(-8,800 to 150,000)
62,000
(-7,700 to 130,000)
exposure
Minor restricted-activity
30,000
34,000
30,000
daysd-f
(12,000 to 47,000)
(14,000 to 54,000)
(12,000 to 48,000)
School absence days
22,000
(-3,100 to 47,000)
26,000
(-3,600 to 54,000)
22,000
(-3,200 to 47,000)
a Values rounded to two significant figures.
b We estimated ozone benefits for changes in NOx for the ozone season and changes in PM2 5 and PM2 5 precursors
for EGUs in 2023. This table does not include benefits from emissions reductions for non-EGUs because emissions
reductions from these sources are not expected prior to 2026 when the proposed standards would become effective.
0 Applied risk estimate derived from April-September exposures to estimates of ozone across the May-September
warm season.
d Converted ozone risk estimate metric from MDA1 to MDA8.
e Applied risk estimate derived from June-August exposures to estimates of ozone across the May-September warm
season.
f Applied risk estimate derived from full year exposures to estimates of ozone across the May-September warm
season.
g Converted ozone risk estimate metric from DA24 to MDA8.
hThe proposed rule imposes unit level emission rate limits on EGUs in the 2026, which are imposed in the 2025
IPM run year, while the less stringent alternative assumes these are imposed in 2028, and in IPM are applied in the
2028 run year. The unit level emission rate limits drive much of the EGU retirement activity, and retirements are
delayed in the less stringent alternative relative to the proposed rule. Since the power sector model is forward
looking, it has an incentive to run units harder before they retire. This incentive is lower in the less stringent
alternative relative to the proposed rule due to delayed retirements. As such, emissions are slightly lower in 2023 in
some states in the less stringent alternative relative to the proposed rule.
ES-16
-------
Table ES-8. Estimated Avoided Ozone-Related Premature Respiratory Mortalities and
Illnesses for the Proposal and More and Less Stringent Alternatives for 2026 (95%
Confidence Interval) a,b,h
Proposal
More Stringent
Alternative
Less Stringent
Alternative
Exposure
Duration
Study
Affected Facility
Avoided premature respiratory mortalities
Long-term
exposure
Turner et al.
(2016)°
EGUs
Non-EGUs
EGUs + Non-
EGUs
450
(310 to 580)
510
(350 to 660)
960
(660 to 1,200)
520
(360 to 670)
550 (380 to 710)
1,100
(740 to 1,400)
210
(140 to 270)
450
(310 to 580)
650
(450 to 850)
Short-term
exposure
Katsouyanni et
al. (2009)cd and
Zanobetti et al.
(2008)^0016(1
EGUs
Non-EGUs
EGUs + Non-
EGUs
20
(8.2 to 32)
23
(9.3 to 36)
43
(18 to 68)
24
(9.5 to 37)
25
(10 to 39)
48
(19 to 76)
9.4
(3.8 to 15)
20
(8.2 to 32)
30
(12 to 47)
Morbidity effects
Long-term
Asthma onset6
EGUs
Non-EGUs
3,300
(2,800 to 3,700)
3,800
(3,300 to 4,400)
3,800
(3,300 to 4,300)
4,200
(3,600 to 4,700)
1,600
(1,300 to 1,800)
3,400
(2,900 to 3,800)
exposure
EGUs + Non-
EGUs
7,100
(6,100 to 8,100)
7,900
(6,800 to 9,000)
4,900
(4,200 to 5,600)
EGUs
19,000
(9,900 to 27,000)
22,000
(11,000 to 32,000)
8,900
(4,700 to 13,000)
Allergic rhinitis
symptoms®
Non-EGUs
22,000
(12,000 to 32,000)
24,000
(13,000 to 35,000)
19,000
(10,000 to 28,000)
EGUs + Non-
EGUs
41,000
(22,000 to 59,000)
46,000
(24,000 to 66,000)
28,000
(15,000 to 41,000)
Hospital
EGUs
55
(-14 to 120)
63
(-17 to 140)
25
(-6.5 to 55)
admissions—
respiratory"1
Non-EGUs
61
(-16 to 140)
66
(-17 to 150)
54
(-14 to 120)
EGUs + Non-
EGUs
120
(-30 to 260)
130
(-34 to 290)
79
(-21 to 170)
Short-term
exposure
EGUs
1,100
(290 to 2,200)
1,200
(340 to 2600)
500
(140 to 1,100)
ED visits—
respiratoryf
Non-EGUs
1,200
(340 to 2,600)
1,300
(360 to 2,800)
1,100
(300 to 2,300)
EGUs + Non-
EGUs
2,300
(630 to 4,800)
2,600
(700 to 5,400)
1,600
(430 to 3,300)
Asthma
symptoms
EGUs
610,000
(-75,000 to
1,300,000)
700,000
(-86,000 to
1,500,000)
290,000
(-35,000 to
590,000)
ES-17
-------
Non-EGUs
710,000
(-87,000 to
1,500,000)
770,000
(-94,000 to
1,600,000)
EGUs + Non-
EGUs
1,300,000
(-160,000 to
2,700,000)
1,500,000
(-180,000 to
3,000,000)
EGUs
280,000
(110,000 to
440,000)
Minor
restricted-
activity daysdf
330,000
(13,000 to
520,000)
Non-EGUs
330,000
(130,000 to
520,000)
360,000
(140,000 to
560,000)
EGUs + Non-
EGUs
610,000
(240,000 to
970,000)
680,000
(270,000 to
1,100,000)
EGUs
220,000
(-30,000 to
450,000)
250,000
(-35,000 to
520,000)
School absence
days
Non-EGUs
250,000
(-35,000 to
530,000)
270,000
(-38,000 to
570,000)
EGUs + Non-
EGUs
470,000
(-66,000 to
980,000)
520,000
(-74,000 to
1,100,000)
620,000
(-77,000 to
1,300,000)
910,000
(-110,000 to
1,900,000)
130,000
(53,000 to
210,000)
290,000
(120,000 to
460,000)
420,000
(170,000 to
670,000)
100,000
(-14,000 to
210,000)
220,000
(-31,000 to
460,000)
320,000
(-46,000 to
670,000)
a Values rounded to two significant figures.
b We estimated ozone benefits for changes in NOx for the ozone season and changes in PM2 5 and PM2 5 precursors
for EGUs in 2026.
0 Applied risk estimate derived from April-September exposures to estimates of ozone across the May-September
warm season.
d Converted ozone risk estimate metric from MDA1 to MDA8.
e Applied risk estimate derived from June-August exposures to estimates of ozone across the May-September warm
season.
f Applied risk estimate derived from full year exposures to estimates of ozone across the May-September warm
season.
g Converted ozone risk estimate metric from DA24 to MDA8.
h Non-EGU benefits estimates are ozone-related only. An illustrative analysis of non-EGU PM benefits estimates is
presented in Chapter 5, Table 5-8.
ES-18
-------
Table ES-9. Estimated Discounted Economic Value of Avoided Ozone and PM2.5-
Attributable Premature Mortality and Illness for the Proposed Policy Scenarios in 2023
(95% Confidence Interval; millions of 2016$)a,b
Disc.
Rate
Pollutant
Proposal
More Stringent Alternative
Less Stringent Alternative
3%
Ozone
Benefits
$57($15
to $120)°
and
$460 ($51
to $l,200)d
$65 ($17
to
$140)'
and
$530 ($59
to $1.400)d
$57
($15 to
$120)'
and
$460
($51 to
$1.200)d
PM BPTs
$44
and
$45
$190
and
$190
$59
and
$(()
O/.onc
Benefits
plus PM
BPTs
$100
($59 to
$160)°
and
$500
($96 to
$l,200)d
$250
($200 to
$330)°
and
$720
($250 to
$l,600)d
$120
($74 to
$180)°
and
$520
($110 to
$l,300)d
7%
Ozone
Benefits
$51 ($9.6
to 110)c
and
$410 ($42
to $l,100)d
$58 ($11
to
$130)°
and
$480 ($49
to $l,300)d
$51
($9.6 to
$110)°
and
$410
($42 to
$l,100)d
PM BPTs
$40
and
$41
$170
and
$170
$53
and
$54
Ozone
Benefits
plus PM
BPTs
$90
($49 to
$150)°
and
$450
($83 to
$l,100)d
$230
($180 to
$300)°
and
$650
($220 to
$l,400)d
$100
($63 to
$170)°
and
$470
($97 to
$l,100)d
a Values rounded to two significant figures. The two benefits estimates are separated by the word "and" to signify
that they are two separate estimates. The estimates do not represent lower- and upper-bound estimates and should
not be summed.
b We estimated ozone benefits for changes in NOx for the ozone season and changes in PM2 5 and PM2 5 precursors
for EGUs in 2023. This table does not include benefits from reductions for non-EGUs because reductions from these
sources are not expected prior to 2026 when the proposed standards would become effective.
0 Using the pooled short-term ozone exposure mortality risk estimate.
d Using the long-term ozone exposure mortality risk estimate.
ES-19
-------
Table ES-10. Estimated Discounted Economic Value of Avoided Ozone and PM2.5-
Attributable Premature Mortality and Illness for the Proposed Policy Scenario in 2026
(95% Confidence Interval; millions of 2016$)a,b
Disc
Rate
Pollutant
Proposal
More Stringent Alternative
Less Stringent Alternative
3%
Ozone
$1,200
$10,000
$1,300
$11,000
$830
$6,900
Benefits
($310 to
and
($1,100 to
(340 to
and
($1,200 to
($210 to
and
($760 to
$2.600)c
$26,000)d
$2.900)1
$29,000)d
$1,800)'
$18.000)d
PM BPTs
$8,100
and
$8,300
$7,800
and
$7,900
$3,400
and
$3,500
O/.onc
Benefits
plus PM
BPTs
$9,300
($8,400 to
$11,000)°
and
$18,000
($9,400 to
$35,000)d
$9,100
($8,100 to
$11,000)°
and
$19,000
($9,200 to
$37,000)d
$4,300
($3,700
to
$5,200)°
and
$10,000
($4,300 to
$22,000)d
7%
Ozone
$1,100
$9,000
$1,200
$10,000
$740
$6,200
Benefits
($200 to
$2,400)°
and
($920 to
$24,000)d
($220 to
$2,700)°
and
($1,000 to
$26,000)d
($140 to
$1,700)°
and
($630 to
$ 16,000)d
PM BPTs
$7,300
and
$7,400
$7,000
and
$7,100
$3,100
and
$3,200
Ozone
Benefits
$8,400
$16,000
$8,200
$17,000
$3,800
$9,300
plus PM
($7,500 to
and
($8,300 to
($7,200 to
and
($8,200 to
($3,200
and
($3,800 to
BPTs
$9,700)°
$31,000)d
$9,700)°
$34,000)d
to
$4,800)°
$19,000)d
a Values rounded to two significant figures. The two benefits estimates are separated by the word "and" to signify
that they are two separate estimates. The estimates do not represent lower- and upper-bound estimates and should
not be summed.
b We estimated changes in NOx for the ozone season and changes in PM2 5 and PM2 5 precursors in 2026. This table
represents changes in EGU and non-EGU ozone season and annual controls.
0 Sum of ozone mortality estimated using the pooled short-term ozone exposure risk estimate and the Di et al. (2017)
long-term PM2 5 exposure mortality risk estimate.
d Sum of the Turner et al. (2016) long-term ozone exposure risk estimate and the Di et al. (2017) long-term PM2 5
exposure mortality risk estimate.
ES.6.2 Climate Benefits
Elevated concentrations of greenhouse gases (GHGs) in the atmosphere have been
warming the planet, leading to changes in the Earth's climate including changes in the frequency
and intensity of heat waves, precipitation, and extreme weather events, rising seas, and retreating
snow and ice. The well-documented atmospheric changes due to anthropogenic GHG emissions
are changing the climate at a pace and in a way that threatens human health, society, and the
natural environment. Climate change touches nearly every aspect of public welfare in the U.S.
with resulting economic costs, including: changes in water supply and quality due to changes in
drought and extreme rainfall events; increased risk of storm surge and flooding in coastal areas
and land loss due to inundation; increases in peak electricity demand and risks to electricity
infrastructure; and the potential for significant agricultural disruptions and crop failures (though
offset to some extent by carbon fertilization).
ES-20
-------
There will be important climate benefits associated with the CO2 emissions reductions
expected from this proposed rule. Climate benefits from reducing emissions of CO2 can be
monetized using estimates of the social cost of carbon (SC-CO2). However, due to a court order,
EPA cannot present these monetized estimates in the analysis of this proposed rule at this time.
On February 11, 2022, the U.S. District Court for the Western District of Louisiana issued an
injunction concerning the monetization of benefits of greenhouse gas emission reductions by
EPA and other defendants. See Louisiana v. Biden, No. 21-cv-01074-JDC-KK (W.D. La. Feb.
11, 2022). Accordingly, monetized climate benefits are not presented in the benefit-cost analysis
of this proposal conducted pursuant to E.O. 12866. See Chapter 5, Section 5.2 for more
discussion.
ES. 6.3 Additional Unquantified Benefits
Data, time, and resource limitations prevented EPA from quantifying the estimated health
impacts or monetizing estimated benefits associated with direct exposure to NO2 and SO2
(independent of the role NO2 and SO2 play as precursors to PM2.5 and ozone), as well as
ecosystem effects, and visibility impairment due to the absence of air quality modeling data for
these pollutants in this analysis. While all health benefits and welfare benefits were not able to be
quantified, it does not imply that there are not additional benefits associated with reductions in
exposures to ozone, PM2.5, NO2 or SO2. For a qualitative description of these and water quality
benefits, please see Chapter 5, Section 5.3, Table 5-9.
ES.7 Environmental Justice Impacts
Environmental justice (EJ) concerns for each rulemaking are unique and should be
considered on a case-by-case basis, and EPA's EJ Technical Guidance15 states that "[t]he
analysis of potential EJ concerns for regulatory actions should address three questions:
1. Are there potential EJ concerns associated with environmental stressors affected
by the regulatory action for population groups of concern in the baseline?
15 U.S. Environmental Protection Agency (EPA), 2015. Guidance on Considering Environmental Justice During the
Development of Regulatory Actions.
ES-21
-------
2. Are there potential EJ concerns associated with environmental stressors affected
by the regulatory action for population groups of concern for the regulatory
option(s) under consideration?
3. For the regulatory option(s) under consideration, are potential EJ concerns created
or mitigated compared to the baseline?"
To address these questions, EPA developed an analytical approach that considers the
purpose and specifics of the proposed rulemaking, as well as the nature of known and potential
exposures and impacts. For the proposal, we quantitatively evaluate 1) the proximity of affected
facilities to potentially disadvantaged populations (Chapter 7, Section 7.3.1), 2) the distribution
of total ozone concentrations in the baseline across different demographic groups (Chapter 7,
Sections 7.4.1.1 and 7.4.2.1), and 3) how regulatory alternatives differentially impact the ozone
concentration changes experienced by different demographic populations (Chapter 7, Sections
7.4.1.2 and 7.4.2.2). Each of these analyses depends on mutually exclusive assumptions, was
performed to answer separate questions, and is associated with unique limitations and
uncertainties.
Because the pollution impacts that are the focus of this proposal are substantially
downwind from affected facilities, the proximity analysis cannot be used to demonstrate
disproportionate pollution impacts in the baseline. However, the analysis indicates that certain
demographic subgroups living near affected facilities could potentially experience differential
effects in terms of local environmental stressors such as noise and traffic.
The baseline analysis of the average April-September warm season maximum daily 8-
hour average ozone concentrations (AS-M03) addresses EJ concerns more directly than the
proximity analyses, as it evaluates the environmental stressor (ozone) primarily affected by the
regulatory action. Results of this baseline analysis suggest that there likely are potential EJ
concerns associated with small average differences in ozone exposure across demographic
groups in the baseline. However, when we consider how the regulatory alternatives will affect
ozone concentrations, and the distribution of those concentrations across different demographic
groups, we see that overall, reductions in AS-M03 concentrations under the proposal, as well as
the more stringent and less stringent alternatives, are predicted to result in very similar ozone
reductions across different demographic groups in future years across both EGUs and non-
ES-22
-------
EGUs. Importantly, this proposal is expected to lower ozone in many areas, including residual
ozone nonattainment areas, and thus mitigate some pre-existing health risks of ozone across all
populations evaluated.
ES.8 Results of Benefit-Cost Analysis
Below in Table ES-11 through
Table ES-13, we present the annual costs and benefits estimates for 2023, 2026, and 2030,
respectively. This analysis uses annual compliance costs reported above as a proxy for social
costs. The net benefits of the proposed rule and more and less stringent alternatives reflect the
benefits of implementing EGU and non-EGU emissions reductions strategies for the affected
states via the FIPs minus the costs of those emissions reductions. The estimated social costs to
implement the proposed rule, as described in this document, are approximately -$210 million in
2023 and $1,100 million in 2026 (2016$). Compliance costs are negative because in 2023 the
EGU compliance costs are negative. While seemingly counterintuitive, estimating negative
compliance costs in a single year is possible given IPM's objective function is to minimize the
discounted net present value (NPV) of a stream of annual total cost of generation over a multi-
decadal time period. As such the model may undertake a compliance pathway that pushes higher
costs later into the forecast period, since future costs are discounted more heavily than near term
costs. This can result in a policy scenario showing single year costs that are lower than the
Baseline, but over the entire forecast horizon, the policy scenario shows higher costs.
The estimated monetized benefits associated with reductions in PM2.5 and ozone
concentrations from implementation of the proposed rule are approximately $100 and $500
million in 2023 (2016$, based on a real discount rate of 3 percent). For 2026, the estimated
monetized benefits from implementation of the proposed rule are approximately $9,300 and
$18,000 million (2016$, based on a real discount rate of 3 percent). The monetized benefits
estimates do not include important climate benefits that were not monetized in this RIA. In
addition, there are important water quality benefits and health benefits associated with reductions
in concentrations of air pollutants other than PM2.5 and ozone that are not quantified.
EPA calculates the monetized net benefits of the proposal by subtracting the estimated
monetized compliance costs from the estimated monetized benefits in 2023, 2026, and 2030.
ES-23
-------
The annual monetized net benefits of the proposed rule in 2023 (in 2016$) are approximately
$310 and $710 million using a 3 percent discount rate. The annual monetized net benefits of the
proposal in 2026 are approximately $8,200 and $17,000 million using a 3 percent real discount
rate. The annual monetized net benefits of the proposal in 2030 are approximately $7,700 and
$18,000 million using a 3 percent real discount rate. Table ES-11 presents a summary of the
monetized benefits, costs, and net benefits of the proposed rule and the more and less stringent
alternatives for 2023. Table ES-12 presents a summary of these impacts for the proposed rule
and the more and less stringent alternatives for 2026, and
Table ES-13 presents a summary of these impacts for the proposed rule and the more and
less stringent alternatives for 2030. These results present an incomplete overview of the effects
of the proposal, because important categories of benefits — including benefits from reducing
climate pollution, other types of air pollutants, and water pollution - were not monetized and are
therefore not reflected in the cost-benefit tables. We anticipate that taking non-monetized effects
into account would show the proposal to be more net beneficial than this table reflects.
Table ES-11. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less
and More Stringent Alternatives for 2023 for the U.S. (millions of 2016$)a,b
Proposed Rule
Less Stringent
Alternative
More Stringent
Alternative
Benefits0
Costs'1
$100 and $500
-$210
$120 and $520
-$170
$250 and $720
-$180
Net Benefits
$310 and $710
$290 and $690
$430 and $900
a We focus results to provide a snapshot of costs and benefits in 2023, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Rows may not appear to add correctly due to rounding.
0 Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates and are presented at a real discount
rate of 3 percent. Several categories of benefits remain unmonetized and are thus not reflected in the table. Non-
monetized benefits include important climate benefits from reductions in CO2 emissions. The U.S. District Court for
the Western District of Louisiana has issued an injunction concerning the monetization of the benefits of greenhouse
gas emission reductions by EPA and other defendants. See Louisiana v. Biden, No. 21-cv-01074-JDC-KK (W.D.
La. Feb. 11, 2022). Therefore, such values are not presented in the benefit-cost analysis of this proposal conducted
pursuant to E.O. 12866. Please see Chapter 5, Section 5.2 for more discussion. In addition, there are important
unqualified water quality benefits and benefits associated with reductions in other air pollutants.
d The costs presented in this table are 2023 annual estimates for each alternative analyzed. An NPV of costs was
calculated using a 3.76% real discount rate consistent with the rate used in IPM's objective function for cost-
minimization.
ES-24
-------
Table ES-12. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less
and More Stringent Alternatives for 2026 for the U.S. (millions of 2016$)a,b
Proposed Rule
Less Stringent
Alternative
More Stringent
Alternative
Benefits0
$9,300 and $18,000
$4,300 and $10,000
$9,100 and $19,000
Costs'1
$1,100
-$49
$1,600
Net Benefits
$8,200 and $17,000
$4,300 and $10,000
$7,500 and $17,000
a We focus results to provide a snapshot of costs and benefits in 2026, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Rows may not appear to add correctly due to rounding.
0 Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates and are presented at a real discount
rate of 3 percent. Several categories of benefits remain unmonetized and are thus not reflected in the table. Non-
monetized benefits include important climate benefits from reductions in CO2 emissions. The U.S. District Court for
the Western District of Louisiana has issued an injunction concerning the monetization of the benefits of greenhouse
gas emission reductions by EPA and other defendants. See Louisiana v. Biden, No. 21-cv-01074-JDC-KK (W.D.
La. Feb. 11, 2022). Therefore, such values are not presented in the benefit-cost analysis of this proposal conducted
pursuant to E.O. 12866. Please see Chapter 5, Section 5.2 for more discussion. In addition, there are important
unqualified water quality benefits and benefits associated with reductions in other air pollutants.
d The costs presented in this table are 2026 annual estimates for each alternative analyzed. An NPV of costs was
calculated using a 3.76% real discount rate consistent with the rate used in IPM's objective function for cost-
minimization.
Table ES-13. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less
and More Stringent Alternatives for 2030 for the U.S. (millions of 2016$)a,b
Proposed Rule
Less Stringent
Alternative
More Stringent
Alternative
Benefits0
$9,400 and $20,000
$4,300 and $11,000
$9,200 and $21,000
Costs'1
$1,600
$1,600
$2,200
Net Benefits
$7,700 and $18,000
$2,800 and $9,700
$7,000 and $19,000
a We focus results to provide a snapshot of costs and benefits in 2030, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Rows may not appear to add correctly due to rounding.
0 Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates and are presented at a real discount
rate of 3 percent. Several categories of benefits remain unmonetized and are thus not reflected in the table. Non-
monetized benefits include important climate benefits from reductions in CO2 emissions. The U.S. District Court for
the Western District of Louisiana has issued an injunction concerning the monetization of the benefits of greenhouse
gas emission reductions by EPA and other defendants. See Louisiana v. Biden, No. 21-cv-01074-JDC-KK (W.D.
La. Feb. 11, 2022). Therefore, such values are not presented in the benefit-cost analysis of this proposal conducted
pursuant to E.O. 12866. Please see Chapter 5, Section 5.2 for more discussion. In addition, there are important
unqualified water quality benefits and benefits associated with reductions in other air pollutants.
d The costs presented in this table are 2030 annual estimates for each alternative analyzed. An NPV of costs was
calculated using a 3.76% real discount rate consistent with the rate used in IPM's objective function for cost-
minimization.
As part of fulfilling analytical guidance with respect to E.O. 12866, EPA presents
estimates of the present value (PV) of the monetized benefits and costs over the twenty-year
period 2023 to 2042. To calculate the present value of the social net-benefits of the proposed
ES-25
-------
rule, annual benefits and costs are discounted to 2022 at 3 percent and 7 discount rates as
directed by OMB's Circular A-4. The EPA also presents the equivalent annualized value (EAV),
which represents a flow of constant annual values that, had they occurred in each year from 2023
to 2042, would yield a sum equivalent to the PV. The EAV represents the value of a typical cost
or benefit for each year of the analysis, in contrast to the year-specific estimates mentioned
earlier in the RIA.
For the twenty-year period of 2023 to 2042, the PV of the net benefits, in 2016$ and
discounted to 2022, is $220,000 million when using a 3 percent discount rate and $130,000 when
using a 7 percent discount rate. The EAV is $15,000 million per year when using a 3 percent
discount rate and $12,000 million when using a 7 percent discount rate. The comparison of
benefits and costs in PV and EAV terms for the proposed rule can be found in Table ES-14.
Estimates in the table are presented as rounded values.
ES-26
-------
Table ES-14. Summary of Present Values and Equivalent Annualized Values for the 2023-2042 Timeframe for Estimated
Monetized Compliance Costs, Benefits, and Net Benefits for the Proposed Rule (millions of 2016$, discounted to 2022)a'b
Benefits Cost0 Net Benefits
3%
7%
3%
7%
3%
7%
2023
$500
$450
($210)
$710
$660
2024
$520
$460
$710
-$190
-$240
2025
$530
$470
$710
-$180
-$230
2026
$18,000
$16,000
$1,100
$17,000
$15,000
2027
$19,000
$17,000
$2,000
$17,000
$15,000
2028
$18,000
$16,000
$2,000
$16,000
$14,000
2029
$19,000
$17,000
$2,000
$17,000
$15,000
2030
$20,000
$18,000
$1,600
$18,000
$16,000
2031
$20,000
$18,000
$1,600
$19,000
$16,000
2032
$21,000
$18,000
$2,100
$18,000
$16,000
2033
$20,000
$18,000
$2,100
$18,000
$16,000
2034
$21,000
$18,000
$2,100
$19,000
$16,000
2035
$21,000
$19,000
$2,100
$19,000
$16,000
2036
$21,000
$19,000
$2,100
$19,000
$17,000
2037
$22,000
$19,000
$2,100
$19,000
$17,000
2038
$21,000
$19,000
$1,300
$20,000
$18,000
2039
$22,000
$19,000
$1,300
$20,000
$18,000
2040
$22,000
$19,000
$1,300
$21,000
$18,000
2041
$22,000
$19,000
$1,300
$21,000
$18,000
2042
$22,000
$20,000
$1,300
$21,000
$18,000
PV
$250,000
$150,000
$22,000
$14,000
$220,000
$130,000
2023-2042
EAV
$17,000
$14,000
$1,500
$1,300
$15,000
$12,000
2023-2042
aRows may not appear to add correctly due to rounding.
b The annualized present value of costs and benefits are calculated over a 20-year period from 2023 to 2042. The benefits values use
the larger of the two benefits estimates presented in Table ES-9 and Table ES-10, as well as for all other years. Monetized benefits include
those related to public health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates and are presented at a real discount rate of 3 percent. Several categories of benefits remain unmonetized and are thus not
reflected in the table. Non-monetized benefits include important climate benefits from reductions in C02 emissions. The U.S. District Court
ES-27
-------
for the Western District of Louisiana has issued an injunction concerning the monetization of the benefits of greenhouse gas emission reductions
by EPA and other defendants. See Louisiana v. Biden, No. 21-cv-01074-JDC-KK (W.D. La. Feb. 11, 2022). Therefore, such values are not
presented in the benefit-cost analysis of this proposal conducted pursuant to E.O. 12866. Please see Chapter 5, Section 5.2 for more discussion.
In addition, there are important unqualified water quality benefits and benefits associated with reductions in other air pollutants.
0 The costs presented in this table are consistent with the costs presented in Chapter 4. To estimate these annualized costs, EPA uses a conventional
and widely accepted approach that applies a capital recovery factor (CRF) multiplier to capital investments and adds that to the annual incremental
operating expenses. Costs were calculated using a 3.76% real discount rate consistent with the rate used in IPM's objective function for cost-minimization.
ES-28
-------
CHAPTER 1: INTRODUCTION AND BACKGROUND
Overview
In this proposed rule, the Federal Implementation Plan Addressing Regional Ozone
Transport for the 2015 Ozone National Ambient Air Quality Standards (FIP for the 2015 ozone
NAAQS), in accordance with the Wisconsin decision, EPA proposes implementation
mechanisms to achieve enforceable emissions reductions required to eliminate significant
contribution of ozone precursor emissions prior to the 2023 ozone season. The initial phase of
proposed emissions reductions will therefore be achieved prior to the August 2, 2024, attainment
date for areas classified as Moderate nonattainment for the 2015 ozone NAAQS.1
EPA is proposing to promulgate new or revised FIPs for 25 states that include new NOx
ozone season emission budgets for EGU sources, with implementation of these emission budgets
beginning in the 2023 ozone season.2 EPA is also proposing to adjust these states' emission
budgets for each ozone season thereafter to maintain the initial stringency of the emissions
budget, accounting for retirements and other changes to the fleet over time. EPA is also
proposing to extend the Cross-State Air Pollution Rule (CSAPR) NOx Ozone Season Group 3
Trading Program beginning in the 2023 ozone season through the 2025 ozone season. EPA is
proposing to establish new emissions budgets for the CSAPR NOx Ozone Season Group 3
Trading Program beginning in the 2026 ozone season, as discussed in Section VII.B.l. of the
preamble. EPA is also proposing to retain two states, Iowa and Kansas, in the CSAPR NOx
Ozone Season Group 2 Trading Program.
EPA is proposing to promulgate new FIPs for 23 states that include new NOx emissions
limitations for non-electric generating unit (non-EGU) sources, with initial compliance dates for
these emissions limitations beginning in 2026.3
1 The 2015 ozone NAAQS is an 8-hour standard that was set at 70 parts per billion (ppb). See 80 FR 65291
(December 28, 2015).
2 In 2023, the 25 states with EGU reduction requirements include AL, AR, DE, IL, IN, KY, LA, MD, MI, MN, MS,
MO, NV, NJ, NY, OH, OK, PA, TN, TX, UT, VA, WV, WI, and WY. There are no EGU reductions being required
from California, and Oregon's SIP was previously approved.
3 In 2026, the 23 states with non-EGU reduction requirements include AR, CA, IL, IN, KY, LA, MD, MI, MN, MS,
MO, NV, NJ, NY, OH, OK, PA, TX, UT, VA, WV, WI, and WY. AL, DE, and TN are not linked in 2026, and
Oregon's SIP was previously approved.
1-1
-------
Consistent with OMB Circular A-4 and EPA's Guidelines for Preparing Economic
Analyses (2010), this regulatory impact analysis (RIA) presents the benefits and costs of the
proposed rule from 2023 through 2042. The estimated monetized benefits are those health
benefits expected to arise from reduced PM2.5 and ozone concentrations. The estimated
monetized costs for EGUs are the costs of installing and operating controls and the increased
costs of producing electricity. The estimated monetized costs for non-EGUs are the costs of
installing and operating controls to meet the ozone season emissions limits.4 The estimated costs
for non-EGUs do not include monitoring, recordkeeping, reporting, or testing costs.
Unquantified benefits and costs are described qualitatively. The RIA also provides (i) estimates
of other impacts of the rule including its effect on retail electricity prices and fuel production and
(ii) an assessment of how expected compliance with the proposed rule will affect concentrations
at nonattainment and maintenance receptors. This chapter contains background information
relevant to the proposed rule and an outline of the chapters of this RIA.
1.1 Background
Clean Air Act (CAA or the Act) section 110(a)(2)(D)(i)(I), which is also known as the
"good neighbor provision," requires states to prohibit emissions that will contribute significantly
to nonattainment or interfere with maintenance in any other state with respect to any primary or
secondary NAAQS. The statute vests states with the primary responsibility to address interstate
emission transport through the development of good neighbor State Implementation Plans (SIPs),
which are one component of larger SIP submittals typically required three years after EPA
promulgates a new or revised NAAQS. These larger SIPs are often referred to as "infrastructure"
SIPs or iSIPs. See CAA section 110(a)(1) and (2).
EPA originally published the Cross-State Air Pollution Rule (CSAPR) on August 8, 2011,
to address interstate transport of ozone pollution under the 1997 ozone National Ambient Air
4 We prepared a non-EGU screening assessment (for more details on the screening assessment, see memorandum
titled Screening Assessment of Potential Emissions Reductions, Air Quality Impacts, and Costs from Non-EGU
Emissions Units for 2026 in the docket), which includes estimated emissions reductions and costs. These estimates
are proxies for (1) non-EGU emissions units that have emission reduction potential, (2) potential controls for and
emissions reductions from these emissions units, and (3) control costs from the potential controls on these emissions
units. This screening assessment is not intended to be, nor take the place of, a unit-specific detailed engineering
analysis that fully evaluates the feasibility of retrofits for the emissions units, potential controls, and related costs.
1-2
-------
Quality Standards (NAAQS).5 On October 26, 2016, EPA published the CSAPR Update, which
finalized Federal Implementation Plans (FIPs) for 22 states that EPA found failed to submit a
complete good neighbor State Implementation Plan (SIP) (15 states)6 or for which EPA issued a
final rule disapproving their good neighbor SIP (7 states).7 The FIPs promulgated for these states
included new electric generating unit (EGU) oxides of nitrogen (NOx) ozone season emission
budgets to reduce interstate transport for the 2008 ozone NAAQS.8 These emission budgets took
effect in 2017 in order to assist downwind states with attainment of the 2008 ozone NAAQS by
the 2018 Moderate area attainment date. EPA acknowledged at the time that the FIPs
promulgated for 21 of the 22 states only partially addressed good neighbor obligations under the
2008 ozone NAAQS.9
On March 31, 2021 EPA promulgated the Revised CSAPR Update (RCU) in response to
the United States Court of Appeals for the District of Columbia Circuit's (D.C. Circuit)
September 13, 2019 remand of the CSAPR Update.10 The D.C. Circuit found that the CSAPR
Update was unlawful to the extent it allowed those states to continue their significant
contributions to downwind ozone problems beyond the statutory dates by which downwind states
must demonstrate their attainment of the air quality standards. The RCU resolved 21 states'
outstanding interstate ozone transport obligations with respect to the 2008 ozone NAAQS.
As described in the preamble of the proposed rule, to reduce interstate emission transport
under the authority provided in CAA section 110(a)(2)(D)(i)(I), this rule further limits ozone
season (May 1 through September 30) NOx emissions from EGUs in 25 states beginning in 2023
and non-EGUs in 23 states beginning in 2026 using the Interstate Transport Framework. The
5 CSAPR also addressed interstate transport of fine particulate matter (PM2 5) under the 1997 and 2006 PM2 5
NAAQS.
6 Alabama, Arkansas, Illinois, Iowa, Kansas, Maryland, Michigan, Mississippi, Missouri, New Jersey, Oklahoma,
Pennsylvania, Tennessee, Virginia, and West Virginia.
7 Indiana, Kentucky, Louisiana, New York, Ohio, Texas, and Wisconsin.
8 The 2008 ozone NAAQS is an 8-hour standard that was set at 75 parts per billion (ppb). See 73 FR 16436 (March
27, 2008).
9 In the CSAPR Update, EPA found that the finalized Tennessee emission budget fully addressed Tennessee's good
neighbor obligation with respect to the 2008 ozone NAAQS. As such, the number of states included was reduced
from 22 to 21 states.
10 EPA took the action to address the remand of the CSAPR Update in Wisconsin v. EPA, 938 F.3d 303 (D.C. Cir.
2019). The court remanded but did not vacate the CSAPR Update, finding that vacatur of the rule could cause harm
to public health and the environment or disrupt the trading program EPA had established and that the obligations
imposed by the rule may be appropriate and sustained on remand.
1-3
-------
Interstate Transport Framework, the framework developed by EPA in the original CSAPR,
provides a 4-step process to address the requirements of the good neighbor provision for ground-
level ozone and fine particulate matter (PM2.5) NAAQS: (1) identifying downwind receptors that
are expected to have problems attaining or maintaining the NAAQS; (2) determining which
upwind states contribute to these identified problems in amounts sufficient to "link" them to the
downwind air quality problems {i.e., here, a 1 percent contribution threshold); (3) for states
linked to downwind air quality problems, identifying upwind emissions that significantly
contribute to downwind nonattainment or interfere with downwind maintenance of the NAAQS;
and (4) for states that are found to have emissions that significantly contribute to nonattainment
or interfere with maintenance of the NAAQS downwind, implementing the necessary emissions
reductions through enforceable measures.
1.1.1 Role of Executive Orders in the Regulatory Impact Analysis
Several statutes and executive orders apply to federal rulemakings. The analyses required
by these statutes, along with a brief discussion of several executive orders, are presented in
Chapter 9. Below we briefly discuss the requirements of Executive Orders 12866 and 13563 and
the guidelines of the Office of Management and Budget (OMB) Circular A-4 (U.S. OMB, 2003).
In accordance with Executive Orders 12866 and 13563 and the guidelines of OMB
Circular A-4, the RIA analyzes the benefits and costs associated with emissions reductions for
compliance with the proposed rule. OMB Circular A-4 requires analysis of one potential
regulatory control alternative more stringent than the proposed rule and one less stringent than
the proposed rule. This RIA evaluates the benefits, costs, and certain impacts of a more and a
less stringent alternative to the selected alternative in this proposal.
1.1.2 Alternatives Analyzed
For EGUs, the FIP for the 2015 ozone NAAQS would require power plants in the 25 states
to participate in the CSAPR NOx Ozone Season Group 3 Trading Program created by the
Revised CSAPR Update. For non-EGUs, the FIP for the 2015 ozone NAAQS would require
units subject to the proposal to meet ozone season emissions limits.
1-4
-------
In response to OMB Circular A-4, this RIA analyzes the FIP for the 2015 ozone NAAQS
emission budgets for EGUs and ozone season emissions limits for non-EGUs, as well as a more
and a less stringent alternative to the proposed rule. The more and less stringent alternatives
differ from the FIP for the 2015 ozone NAAQS in that they set different EGU NOx ozone season
emission budgets and different dates for compliance with unit-specific emission rate limits for
the affected EGUs and cover different industries or emissions units for non-EGUs. Table 1-1
below presents the less stringent alternatives, proposed rule requirements, and more stringent
alternatives for EGUs and non-EGUs.
Table 1-1. Regulatory Control Alternatives for EGUs and Non-EGUs
Regulatory Control
Alternative
NOx Controls Implemented for EGUs within IPM
1) 2023 onwards: Shift generation to minimize costs
2) 2023 onwards: Fully operate existing SCRs during ozone season
3) 2023 onwards: Fully operate existing SNCRs during ozone season
4) In 2023 install state-of-the-art combustion controls
5) In 2028 model run year, impose backstop emission rate limits on coal units
greater than 100 MW within the 23-state region that lack SCR controls,
forcing units to retrofit or retire.
6) In 2028 model run year, impose backstop emission rate limits on oil/gas
steam units greater than 100 MW that operated at a greater than 20% capacity
factor historically within the 23-state region that lack SCR controls, forcing
units to retrofit or retire.11
Less Stringent Alternative
Proposed Rule
(All Controls above and)
7) In 2026, impose backstop emission rate limits on coal units greater than 100
MW within the 23-state region that lack SCR controls, forcing units to retrofit
or retire.
8) In 2026, impose backstop emission rate limits on oil/gas steam units greater
than 100 MW that operated at a greater than 20% capacity factor historically
within the 23-state region that lack SCR controls, forcing units to retrofit or
retire.
More Stringent Alternative
9)
(Controls 1 - 4, 7 and 8 above and)
In 2026, impose backstop emission rate limits on all oil/gas steam units
greater than 100 MW within the 23-state region that lack SCR controls,
forcing units to retrofit or retire.
NOx Emissions Limits for Non-EGUs - Emissions Unit Types and Industries
1)
Reciprocating internal combustion engines in Pipeline Transportation of
Natural Gas,
Less Stringent Alternative
2)
3)
4)
Kilns in Cement and Cement Product Manufacturing,
Boilers and furnaces in Iron and Steel Mills and Ferroalloy Manufacturing,
Furnaces in Glass and Glass Product Manufacturing, and
Proposed Rule
5)
(All emissions unit types and industries above and)
Impactful boilers* in Basic Chemical Manufacturing, Petroleum and Coal
Products Manufacturing, and Pulp, Paper, and Paperboard Mills.
11 The 20% capacity factor cutoff applied is representative of the fleet of O/G steam units assumed to have SCR
retrofit potential in its state budgets. In the proposal, EPA defined this segment using 150 tons per season cutoff,
which provides a similar size of the O/G steam fleet as the 20% capacity factor value used in this analysis.
1-5
-------
Regulatory Control NOx Controls Implemented for EGUs within IPM
Alternative
(All emissions unit types and industries above and)
More Stringent Alternative 6) All boilers in Basic Chemical Manufacturing, Petroleum and Coal Products
Manufacturing, and Pulp, Paper, and Paperboard Mills.
impactful boilers are boilers with design capacity of 100 mmBtu/hr or greater.
For the EGUs, all three alternatives use emission budgets that were developed using
uniform control stringency represented by $1,800 per ton of NOx (2016$) in 2023 and $11,000
per ton of NOx (2016$) in 2026. The less-stringent alternative imposes unit-specific emission
rate limits in the 2028 run year, while the proposed rule and more stringent alternative impose
unit-specific emission rate limits in the 2025 run year. For the proposed rule and more stringent
alternative, unit-specific emission rate limits are imposed on all coal units within the linked
states that are greater than 100 MW and lack SCR controls. Emission rate limits are also imposed
on all oil/gas steam units within the linked states that are greater than 100 MW and lack SCR
controls that operated at a greater than 20 percent historical capacity factor. In addition to the
unit-specific rate limits present in the proposed rule and the less stringent alternative, the more
stringent alternative also imposes unit-specific emission rate limits on all oil/gas steam units in
the affected states that are greater than 100 MW, lack SCR controls and have operated at below a
20 percent capacity factor historically. See section VII.B. of the preamble, and the EGU NOx
Mitigation Strategies Proposed Rule TSD, in the docket for this rule12 for further details of these
emission budgets.
For non-EGUs, a less stringent alternative would require the emissions limits for all
emission units from the proposed policy alternative except for impactful boilers in Basic
Chemical Manufacturing, Petroleum and Coal Products Manufacturing, and Pulp, Paper, and
Paperboard Mills. A more stringent alternative would require the emissions limits for all
emission units from the proposed policy alternative and all boilers, not just impactful boilers, in
Basic Chemical Manufacturing, Petroleum and Coal Products Manufacturing, and Pulp, Paper,
and Paperboard Mills. The emissions limits for the emissions units are the same for each
alternative, while the anticipated total number of emissions units to which the limits apply is
different between alternatives. See Section VII.C. of the preamble for details on the proposed
12 Docket ID No. EPA-HQ-OAR-2021-0668
1-6
-------
emissions limits. See Chapter 4, Section 4.1.2 of this RIA for more details on the less-stringent
and more-stringent alternatives for non-EGUs.
1.1.3 The Needfor Air Quality or Emissions Regulation
OMB Circular A-4 indicates that one of the reasons a regulation may be issued is to
address a market failure. The major types of market failure include externalities, market power,
and inadequate or asymmetric information. Correcting market failures is one reason for
regulation; it is not the only reason. Other possible justifications include improving the function
of government, correcting distributional unfairness, or securing privacy or personal freedom.
Environmental problems are classic examples of externalities - uncompensated benefits
or costs imposed on another party as a result of one's actions. For example, the smoke from a
factory may adversely affect the health of local residents and soil the property in nearby
neighborhoods. Pollution emitted in one state may be transported across state lines and affect air
quality in a neighboring state. If bargaining were costless and all property rights were well
defined, people would eliminate externalities through bargaining without the need for
government regulation.
From an economics perspective, achieving emissions reductions (i.e., by establishing the
EGUNOx ozone-season emissions budgets in this proposal) through a market-based mechanism
is a straightforward and cost-effective remedy to address an externality in which firms emit
pollutants, resulting in health and environmental problems without compensation for those
incurring the problems. Capping emissions through allowance allocations incentivizes those who
emit the pollutants to reduce their emissions, which lessens the impact on those who suffer the
health and environmental problems from higher levels of pollution.
1.2 Overview and Design of the RIA
1.2.1 Methodology for Identifying Needed Reductions
In order to apply the first and second steps of the CSAPR 4-step Interstate Transport
Framework to interstate transport for the 2015 ozone NAAQS, EPA performed air quality
modeling to project ozone concentrations at air quality monitoring sites in 2023, 2026, and 2032.
EPA evaluated projected ozone concentrations for the 2023 analytic year at individual
1-7
-------
monitoring sites and considered current ozone monitoring data at these sites to identify receptors
that are anticipated to have problems attaining or maintaining the 2015 ozone NAAQS. This
analysis was then repeated using projected ozone concentrations for 2026 and 2032. In these
analyses, downwind air quality problems are defined by receptors that are projected to be unable
to attain (i.e., nonattainment receptor) or maintain (i.e., maintenance receptor) the 2015 ozone
NAAQS.
To apply the second step of the Interstate Transport Framework, EPA used air quality
modeling to quantify the contributions from upwind states to ozone concentrations in 2023 and
2026 at downwind receptors. Once quantified, EPA then evaluated these contributions relative to
a screening threshold of 1 percent of the NAAQS. States with contributions that equal or exceed
1 percent of the NAAQS are identified as warranting further analysis for significant contribution
to nonattainment or interference with maintenance.13 States with contributions below 1 percent
of the NAAQS are considered to not significantly contribute to nonattainment or interfere with
maintenance of the NAAQS in downwind states.
To apply the third step of the Interstate Transport Framework, EPA applied a multi-factor
test to evaluate cost, available emission reductions, and downwind air quality impacts to
determine the appropriate level of NOx control stringency that addresses the impacts of interstate
transport on downwind nonattainment or maintenance receptors. EPA used this multi-factor
assessment to gauge the extent to which emission reductions are needed, and to ensure any
required reductions do not result in over-control.
For EGUs, in identifying levels of uniform control stringency EPA assessed the same
NOx emissions controls that the Agency analyzed in the CSAPR Update and the Revised
CSAPR Update, all of which are considered to be widely available in for EGUs: (1) fully
operating existing SCR, including both optimizing NOx removal by existing operational SCRs
and turning on and optimizing existing idled SCRs; (2) installing state-of-the-art NOx
combustion controls; (3) fully operating existing SNCRs, including both optimizing NOx
13 EPA assessed the magnitude of the maximum projected design value for 2023 at each receptor in relation to the
2015 ozone NAAQS. Where the value exceeds the NAAQS, EPA determined that receptor to be a maintenance
receptor for purposes of defining interference with maintenance. That is, monitoring sites with a maximum design
value that exceeds the NAAQS are projected to have a maintenance problem in 2023.
1-8
-------
removal by existing operational SNCRs and turning on and optimizing existing idled SNCRs; (4)
installing new SNCRs; (5) installing new SCRs; and (6) generation shifting (i.e., emission
reductions anticipated to occur from generation shifting from higher to lower emitting units).
Levels of uniform control stringency were represented by $1,800 per ton of NOx (2016$) in
2023 and $11,000 per ton of NOx (2016$) in 2026.14
For non-EGUs, in identifying appropriate control strategies EPA developed an analytical
framework15 to evaluate the air quality impacts of potential emissions reductions from non-EGU
sources located in the linked upwind states. EPA incorporated air quality modeling information,
annual emissions, and information about potential controls to estimate the NOx emissions
reduction potential from non-EGU sources to determine which non-EGU industries, if subject to
further control requirements, would have the greatest impact in providing air quality
improvements at the downwind receptors. The evaluation was subject to a marginal cost
threshold of up to $7,500 per ton (2016$), which EPA determined based on information available
to the Agency about existing control device efficiency and cost information. EPA identified
emissions unit types in seven industries (see Chapter 4, Section 4.4 for discussion of the
approach used to identify the industries) that provide opportunities for NOx emissions reductions
that result in meaningful impacts on air quality at the downwind receptors.
1.2.2 States Covered by the Rule
For EGUs, the FIP for the 2015 ozone NAAQS would require power plants in the 25
states to participate in the CSAPR NOx Ozone Season Group 3 Trading Program created by the
Revised CSAPR Update.16
• The following twelve states currently participating in the Group 3 Trading Program
would remain in the program, with revised provisions beginning in the 2023 ozone
season, under this proposed rule: Illinois, Indiana, Kentucky, Louisiana, Maryland,
14 EGU NOx Mitigation Strategies Proposed Rule TSD, in the docket for this proposed rule (Docket ID No. EPA-
HQ-OAR-2021 -0688).
15 Additional information on the analytical framework is presented in the memorandum titled Screening Assessment
of Potential Emissions Reductions, Air Quality Impacts, and Costs from Non-EGU Emissions Units for 2026, which
is available in the docket for this proposed rulemaking.
16 As explained in Section VI. C. 1 of the preamble, EPA proposes finding that EGU sources within the State of
California are sufficiently controlled such that no further emission reductions are needed from them to eliminate
significant contribution to downwind states.
1-9
-------
Michigan, New Jersey, New York, Ohio, Pennsylvania, Virginia, and West
Virginia.
• Affected EGUs in eight states currently covered by the CSAPR NOx Ozone Season
Group 2 Trading Program - Alabama, Arkansas, Mississippi, Missouri, Oklahoma,
Tennessee, Texas, and Wisconsin - would transition from the Group 2 program to
the revised Group 3 trading program beginning with the 2023 control period.
• Affected EGUs in five states not currently covered by any CSAPR trading program
for seasonal NOx emissions - Delaware, Minnesota, Nevada, Utah, and Wyoming
- would enter the Group 3 trading program in the 2023 control period following the
effective date of a final rule.
In addition, EPA is proposing to revise other aspects of the Group 3 trading program to provide
improved environmental outcomes and increase compliance, as described in Section VII of the
preamble. The proposed rule does not revise the budget stringency and geography of the existing
CSAPR NOx Ozone Season Group 1 trading program.
Aside from the eight states moving from the Group 2 trading program to the Group 3
trading program under the proposed rule, this proposal otherwise leaves unchanged the budget
stringency of the existing CSAPR NOx Ozone Season Group 2 trading program.
For non-EGUs, the proposal also includes NOx emissions limitations with an initial
compliance date of 2026 applicable to certain non-EGU stationary sources in 23 states:
Arkansas, California, Illinois, Indiana, Kentucky, Louisiana, Maryland, Michigan, Minnesota,
Mississippi, Missouri, Nevada, New Jersey, New York, Ohio, Oklahoma, Pennsylvania, Texas,
Utah, Virginia, West Virginia, Wisconsin, and Wyoming.
1.2.3 Regulated Entities
The proposal affects EGUs in 26 states and regulates utilities (electric, natural gas, other
systems) classified as code 221112 by the North American Industry Classification System
(NAICS) and have a nameplate capacity of greater than 25 megawatts (MWe). In addition, the
rule affects certain non-EGUs in 23 states in the following industries, as defined by 4-digit
NAICS: Pipeline Transportation of Natural Gas, 4862; Cement and Concrete Product
1-10
-------
Manufacturing, 3273; Iron and Steel Mills and Ferroalloy Manufacturing, 3311; Glass and Glass
Product Manufacturing, 3272; Basic Chemical Manufacturing, 3251; Petroleum and Coal
Products Manufacturing, 3241; Pulp, Paper, and Paperboard Mills, 3221. For additional
discussion of the non-EGUs affected, see Section VII. C. of the preamble.
1.2.4 Baseline and Analysis Years
As described in the preamble, EPA aligns implementation of this proposal with relevant
attainment dates for the 2015 ozone NAAQS. The initial phase of proposed emissions reductions
will therefore be achieved prior to the August 2, 2024 attainment date for areas classified as
Moderate nonattainment for the 2015 ozone NAAQS.
To develop and evaluate control strategies for addressing these obligations, it is important
to first establish a baseline projection of air quality in the analysis year of 2023, taking into
account currently on-the-books Federal regulations, substantial Federal regulatory proposals,
enforcement actions, state regulations, population, and where possible, economic growth.17
Establishing this baseline for the analysis then allows us to estimate the incremental costs and
benefits of the additional emissions reductions that will be achieved by the proposal.
The analysis in this RIA focuses on benefits, costs and certain impacts from 2023 through
2042. We focus on 2023 because it is by the 2023 ozone season, corresponding with the 2024
attainment date for areas classified as Moderate nonattainment, that significant contribution from
upwind states' must be eliminated to the extent possible. In addition, impacts in 2026 are
important because it is in this period that additional NOx control technologies are expected to be
installed where upwind linkage to downwind receptors persists.
17 The technical support document (TSD) for the 2016v2 emissions modeling platform titled Preparation of
Emissions Inventories for the 2016v2 North American Emissions Modeling Platform is included in the docket for
this rule. The TSD includes additional discussion on mobile source rules included in the baseline. For this proposal,
the future-year emissions estimates for onroad mobile sources represent all national control programs known at the
time of modeling including rules newly added in MOVES3: the Greenhouse Gas Emissions and Fuel Efficiency
Standards for Medium- and Heavy-Duty Engines and Vehicles (HDGHG) - Phase 2 and the Safer Affordable Fuel-
Efficient (SAFE) Vehicles Rule. Other finalized rules incorporated into the onroad mobile source emissions
estimates include: Tier 3 Standards (March 2014), the Light-Duty Greenhouse Gas Rule (March 2013), Heavy (and
Medium)-Duty Greenhouse Gas Rule (August 2011), the Renewable Fuel Standard (February 2010), the Light Duty
Greenhouse Gas Rule (April 2010), the Corporate-Average Fuel Economy standards for 2008-2011 (April 2010),
the 2007 Onroad Heavy-Duty Rule (February 2009), and the Final Mobile Source Air Toxics Rule (MSAT2)
(February 2007).
1-11
-------
EPA's analysis for the third step of the Interstate Transport Framework reflects emissions
reductions for 2023 from EGUs based on a control stringency at a representative cost threshold
of $1,800 per ton. Those reductions are commensurate with optimization of existing SCRs and
SNCRs and installation of state-of-the-art combustion controls. For 2026, the selected control
stringency (at a representative cost per ton threshold of $11,000 for EGUs and a marginal cost
threshold of $7,500 for non-EGUs) includes additional EGU controls and estimated non-EGU
emissions reductions. See Section VI.D.4 of the preamble for additional discussion. Additional
benefits and costs are expected to occur after 2026 as EGUs subject to this rule continue to
comply with the tighter allowance budget, which is below their baseline emissions, and non-
EGUs remain subject to ozone season emissions limits.
1.2.5 Emissions Controls, Emissions, and Cost Analysis Approach
EPA estimated the control strategies and compliance costs of the rule using the Integrated
Planning Model (IPM) as well as certain costs that are estimated outside the model but use IPM
inputs for their estimation. These cost estimates reflect costs incurred by the power sector and
include (but are not limited to) the costs of purchasing, installing, and operating NOx control
technology, changes in fuel costs, and changes in the generation mix. A description of the
methodologies used to estimate the costs and economic impacts to the power sector is contained
in Chapter 4 of this RIA. This analysis also provides estimates of NOx emissions changes during
the May through September ozone season and year-round, as well as annual emissions changes
in PM2.5, SO2, and carbon dioxide (CO2) due to changes in power sector operation.
In addition, to identify appropriate control strategies for non-EGU sources to achieve
NOx emissions reductions that would result in meaningful air quality improvements in
downwind areas, EPA developed an analytical framework to evaluate the air quality impacts of
potential emissions reductions from non-EGU sources located in the linked upwind states. EPA
incorporated air quality modeling information, annual emissions, and available information about
potential to determine which industries, if subject to further control requirements, would have the
greatest impact in providing air quality improvements at the downwind receptors. This
evaluation was subject to a marginal cost threshold of up to $7,500 per ton, which EPA
determined based on information available to the Agency about existing control device
1-12
-------
efficiency and cost information. EPA used the control strategy tool (CoST)18, the control
measures database19, and the 2019 emissions inventory with the analytical framework to prepare
a screening assessment for 2026. Additional information on the analytical framework is included
in the memorandum titled Screening Assessment of Potential Emissions Reductions, Air Quality
Impacts, and Costs from Non-EGU Emissions Units for 2026 20 This screening assessment is not
intended to be, nor take the place of, a unit-specific detailed engineering analysis that fully
evaluates the feasibility of retrofits for the emissions units, potential controls, and related costs.
We used CoST to identify emissions units, emissions reductions, and costs to include in a
proposed FIP; however, CoST was designed to be used for illustrative control strategy analyses
(e.g., NAAQS regulatory impact analyses) and not for unit-specific, detailed engineering
analyses. The estimates from CoST identify proxies for (1) non-EGU emissions units that have
emission reduction potential, (2) potential controls for and emissions reductions from these
emissions units, and (3) control costs from the potential controls on these emissions units.
1.2.6 Benefits Analysis Approach
Implementing the FIP for the 2015 ozone NAAQS proposal is expected to reduce
emissions of PM2.5, NOx and SO2 throughout the year. Because NOx and SO2 are also precursors
to formation of ambient PM2.5, reducing these emissions would reduce human exposure to
ambient PM2.5 throughout the year and would reduce the incidence of PIvfc.s-attributable health
effects. For more details on associated estimated benefits, see Chapter 5.
1.3 Organization of the Regulatory Impact Analysis
This RIA is organized into the following remaining chapters:
• Chapter 2: Sector Profiles. This chapter describes the electric power sector in detail, as
well as provides an overview of the other non-EGU industries.
• Chapter 3: Air Quality Impacts. The data, tools, and methodology used for the air quality
modeling are described in this chapter, as well as the post-processing techniques used to
produce air quality metric values for input into the analysis of benefits and costs.
18 Further information on CoST can be found at the following link: https://www.epa.gov/economic-and-cost-
analysis-air-pollution-regulations/cost-analysis-modelstools-air-pollution.
19 The control measures database is available at the following link: https://www.epa.gov/economic-and-cost-
analysis-air-pollution-regulations/cost-analysis-modelstools-air-pollution.
20 The costs did not include monitoring, recordkeeping, report, or testing costs.
1-13
-------
• Chapter 4: Cost, Emissions, and Energy Impacts. The chapter summarizes the data
sources and methodology used to estimate the costs and other impacts incurred by the
power sector. The chapter summarizes the non-EGU screening assessment used to
estimate costs for the non-EGU industries.
• Chapter 5: Benefits. The chapter presents the health-related benefits of the ozone-related
air quality improvements.
• Chapter 6: Economic Impacts. The chapter includes a discussion of small entity,
economic, and labor impacts.
• Chapter 7: Environmental Justice Impacts. This chapter includes an assessment of
downwind ozone impacts across potential environmental justice populations.
• Chapter 8: Comparison of Benefits and Costs. The chapter compares estimates of the
total benefits with total costs and summarizes the net benefits of the three regulatory
control alternatives analyzed.
1-14
-------
CHAPTER 2: INDUSTRY SECTOR PROFILES
Overview
This chapter discusses important aspects of the regulated industries that relate to the
proposed rule with respect to the interstate transport of emissions of nitrogen oxides (NOx) that
contribute significantly to nonattainment or interfere with maintenance of the 2015 ozone
NAAQS in downwind states. This chapter describes types of existing power-sector sources
affected by the proposed regulation and provides background on the power sector and electricity
generating units (EGUs). In addition, this chapter also briefly describes the relevant non-EGU
industries included in the proposal.
2.1 Background
In the past decade there have been significant structural changes in both the mix of
generating capacity and in the share of electricity generation supplied by different types of
generation. These changes are the result of multiple factors in the power sector, including normal
replacements of older generating units with new units, changes in the electricity intensity of the
U.S. economy, growth and regional changes in the U.S. population, technological improvements
in electricity generation from both existing and new units, changes in the prices and availability
of different fuels, and substantial growth in electricity generation by renewable and
unconventional methods. Many of these trends will continue to contribute to the evolution of the
power sector. The evolving economics of the power sector, specifically the increased natural gas
supply and subsequent relatively low natural gas prices, have resulted in more natural gas being
used as base load energy in addition to supplying electricity during peak load. Additionally rapid
growth in the penetration of renewables has led to their now constituting a significant share of
generation. This chapter presents data on the evolution of the power sector from 2014 through
2020. Projections of future power sector behavior and the impact of this proposed rule are
discussed in more detail in Chapter 4 of this RIA.
2.2 Power Sector Overview
The production and delivery of electricity to customers consists of three distinct segments:
generation, transmission, and distribution.
2-1
-------
2.2.1 Generation
Electricity generation is the first process in the delivery of electricity to consumers. There
are two important aspects of electricity generation; capacity and net generation. Generating
Capacity refers to the maximum amount of production an EGU is capable of producing in a
typical hour, typically measured in megawatts (MW) for individual units, or gigawatts (1 GW =
1,000 MW) for multiple EGUs. Electricity Generation refers to the amount of electricity actually
produced by an EGU over some period of time, measured in kilowatt-hours (kWh) or gigawatt-
hours (1 GWh = 1 million kWh). Net Generation is the amount of electricity that is available to
the grid from the EGU (i.e., excluding the amount of electricity generated but used within the
generating station for operations). Electricity generation is most often reported as the total annual
generation (or some other period, such as seasonal). In addition to producing electricity for sale
to the grid, EGUs perform other services important to reliable electricity supply, such as
providing backup generating capacity in the event of unexpected changes in demand or
unexpected changes in the availability of other generators. Other important services provided by
generators include facilitating the regulation of the voltage of supplied generation.
Individual EGUs are not used to generate electricity 100 percent of the time. Individual
EGUs are periodically not needed to meet the regular daily and seasonal fluctuations of
electricity demand. Furthermore, EGUs relying on renewable resources such as wind, sunlight
and surface water to generate electricity are routinely constrained by the availability of adequate
wind, sunlight, or water at different times of the day and season. Units are also unavailable
during routine and unanticipated outages for maintenance. These factors result in the mix of
generating capacity types available (e.g., the share of capacity of each type of EGU) being
substantially different than the mix of the share of total electricity produced by each type of EGU
in a given season or year.
Most of the existing capacity generates electricity by creating heat to create high pressure
steam that is released to rotate turbines which, in turn, create electricity. Natural gas combined
cycle (NGCC) units have two generating components operating from a single source of heat. The
first cycle is a gas-fired turbine, which generates electricity directly from the heat of burning
natural gas. The second cycle reuses the waste heat from the first cycle to generate steam, which
is then used to generate electricity from a steam turbine. Other EGUs generate electricity by
2-2
-------
using water or wind to rotate turbines, and a variety of other methods including direct
photovoltaic generation also make up a small, but growing, share of the overall electricity
supply. The generating capacity includes fossil-fuel-fired units, nuclear units, and hydroelectric
and other renewable sources (see Table 2-1). Table 2-1 also shows the comparison between the
generating capacity in 2014 and 2020.
In 2020 the power sector consisted of over 23,417 generating units with a total capacity1 of
1,116 GW, an increase of 47 GW (or 4 percent) from the capacity in 2014 (1,068 GW). The 47
GW increase consisted primarily of natural gas fired EGUs (54 GW), and wind (54 GW) and
solar generators (38 GW), and the retirement/re-rating of 84 GW of coal capacity. Substantially
smaller net increases and decreases in other types of generating units also occurred.
Table 2-1. Total Net Summer Electricity Generating Capacity by Energy Source,
2014 and 2020
2014
2020
Change Between
2014 and 2020
Energy Source
Net
Summer
Capacity
(MW)
% Total
Capacity
Net
Summer
Capacity
(MW)
% Total
Capacity
%
Increase
Capacity
Change
(MW)
Coal
299,094
28%
215,554
19%
-28%
-83,540
Natural Gas
432,150
40%
485,807
44%
12%
53,657
Nuclear
98,569
9%
96,501
9%
-2.1%
-2,069
Hydro
102,162
9.56%
102,941
9.23%
0.8%
778
Petroleum
41,135
3.85%
27,569
2.47%
-33%
-13,566
Wind
64,232
6.01%
118,379
10.61%
84%
54,147
Solar
10,323
0.97%
48,054
4.31%
365%
37,731
Other Renewable
16,049
2%
15,522
1%
-3%
-527
Misc
4,707
0.44%
5,355
0.48%
14%
648
Total
1,068,422
100%
1,115,681
100%
4%
47,259
Note: This table presents generation capacity. Actual net generation is presented in Table 2-2.
Source: EIA. Electric Power Annual 2014 and 2020, Table 4.3
The 4 percent increase in generating capacity is the net impact of newly built generating
units, retirements of generating units, and a variety of increases and decreases to the nameplate
1 This includes generating capacity at EGUs primarily operated to supply electricity to the grid and combined heat
and power facilities classified as Independent Power Producers (IPP) and excludes generating capacity at
commercial and industrial facilities that does not operate primarily as an EGU. Natural Gas information in this
chapter (unless otherwise stated) reflects data for all generating units using natural gas as the primary fossil heat
source. This includes Combined Cycle Combustion Turbine, Gas Turbine, steam, and miscellaneous (< 1 percent).
2-3
-------
capacity of individual existing units due to changes in operating equipment, changes in emission
controls, etc. During the period 2014 to 2020, a total of 173 GW of new generating capacity was
built and brought online, and 123 GW of existing units were retired. The net effect of the re-
rating of existing units reduced the total capacity by 2.8 GW. The overall net change in capacity
was an increase of 47 GW, as shown in Table 2-1.
The newly built generating capacity was primarily natural gas (67.1 GW), which was
partially offset by gas retirements (23.6 GW of gas steam retirements, 5.2 GW of combined cycle
and 7.3 GW of combustion turbine retirements for a total of 36.1 GW of gas retirements). Wind
capacity was the second largest type of new builds (59 GW), followed by solar (41 GW). The
largest decline was from coal retirements and re-rating, which amounted to 84 GW over this
period. The overall mix of newly built and retired capacity, along with the net effect, is shown in
Figure 2-1. The data for Figure 2-1 is from the EIA Preliminary Monthly Generator Inventory.
Figure 2-1 also shows wind and solar retirements of 1,060 MW.
200,000
160,000
120,000
|
¦ Coal
40,000
¦ Wind & Solar
¦ Gas
New Build Retirement Net Change
(40,000)
(80,000)
(120,000)
Figure 2-1. National New Build and Retired Capacity (MW) by Fuel Type, 2014-2020
The information in Table 2-1 and Figure 2-1 present information about the generating
capacity in the entire U.S. The proposed Federal Implementation Plan (FIP) Addressing Regional
Ozone Transport for the 2015 Ozone National Ambient Air Quality Standards (FIP for the 2015
ozone NAAQS), however, directly affects EGUs in 25 eastern states. The share of generating
capacity from each major type of generation differs between the FIP for the 2015 NAAQS
2-4
-------
Ozone Region and the rest of the U.S. (non-region). Figure 2-2 shows the mix of generating
capacity for each region. In 2020, the overall capacity in the FIP for the 2015 Ozone NAAQS
Region is 60 percent of the national total, reflecting the larger total population in the region. The
mix of capacity is noticeably different in the two regions. In the FIP for the 2015 Ozone NAAQS
Region in 2020, coal makes up a significantly larger share of total capacity (23 percent) than it
does in the rest of the country (13 percent). The share of natural gas in the FIP for the 2015
Ozone NAAQS Region is 46 percent as compared to 39 percent in the rest of the country. The
difference in the share of coal's capacity is primarily balanced by relatively more hydro, wind,
and solar capacity in the rest of country compared to the FIP for the 2015 Ozone NAAQS
Region.
700,000
600,000
500,000
¦ Other
400,000
¦ Wind & Solar
¦ Hydro
300,000
¦ Nuclear
200,000
¦ Gas
100,000
¦ Coal
0
In Region
Non-Region
Figure 2-2. Regional Differences in Generating Capacity (MW), 2020
Source: Form EIA-860. Note: "Other" includes petroleum, geothermal, other renewable, waste materials and
miscellaneous.
In 2020, electric generating sources produced a net 4,049TWh to meet national electricity
demand, which was roughly flat from 2014. As presented in Table 2-2, 60 percent of electricity
in 2020 was produced through the combustion of fossil fuels, primarily coal and natural gas, with
natural gas accounting for the largest single share. Although the share of the total generation
from fossil fuels in 2020 (60 percent) was only modestly smaller than the total fossil share in
2014 (67 percent), the mix of fossil fuel generation changed substantially during that period.
Coal generation declined by 51 percent and petroleum generation by 42 percent, while natural
2-5
-------
gas generation increased by 44 percent. This reflects both the increase in natural gas capacity
during that period as well as an increase in the utilization of new and existing gas EGUs during
that period. The combination of wind and solar generation also grew from 5 percent of the mix in
2014 to 12 percent in 2020.
Table 2-2. Net Generation in 2014 and 2C
)20 (Trillion kWh
= TWh)
Change Between '14 and
2014
2020
'20
Net
Generat
Net
Net
ion
% Change
Generation
Fuel Source
Generation Fuel Source
Change
in Net
(TWh)
Share
(TWh)
Share
(TWh)
Generation
Coal
1,582
39%
773
19%
-808
1496%
Natural Gas
1,127
27%
1,624
40%
498
-907%
Nuclear
797
19%
790
20%
-7
13%
Hydro
253
6%
280
7%
27
-60%
Petroleum
30
1%
17
0%
-13
24%
Wind
182
4%
338
8%
156
-289%
Solar
18
0%
131
3%
102
-136%
Other
Renewable
91
2%
71
2%
-9
-44%
Misc
25
1%
25
1%
-1
3%
Total
4,105
100%
4,049
100%
-56
100%
Source: EI A 2014 and 2020 Electric Power Annual, Tables 3.1
Coal-fired and nuclear generating units have historically supplied "base load" electricity,
the portion of electricity loads that are continually present and typically operate throughout all
hours of the year. Although much of the coal fleet operates as base load, there can be notable
differences across various facilities (see Table 2-3). For example, coal-fired units less than 100
megawatts (MW) in size compose 18 percent of the total number of coal-fired units, but only 2
percent of total coal-fired capacity. Gas-fired generation is better able to vary output and is the
primary option used to meet the variable portion of the electricity load and has historically
supplied "peak" and "intermediate" power, when there is increased demand for electricity (for
example, when businesses operate throughout the day or when people return home from work
and run appliances and heating/air-conditioning), versus late at night or very early in the
morning, when demand for electricity is reduced.
2-6
-------
Table 2-3 also shows comparable data for the capacity and age distribution of natural gas
units. Compared with the fleet of coal EGUs, the natural gas fleet of EGUs is generally smaller
and newer. While 67 percent of the coal EGU fleet capacity is over 500 MW per unit, 75 percent
of the gas fleet is between 50 and 500 MW per unit. Many of the largest gas units are gas-fired
steam-generating EGUs.
Table 2-3. Coal and Natural Gas Generating Units, by Size, Age, Capacity, and Average
Heat Rate in 2020
Avg. Net
Total Net
Unit Size
Summer
Summer
Avg. Heat
Grouping
% of All
Capacity
Capacity
% Total
Rate
(MW) No. Units
Units
Avg. Age
(MW)
(MW)
Capacity
(Btu/kWh)
COAL
0-24
35
7%
46
11
372
0%
11,027
25-49
30
6%
35
37
1,096
1%
11,638
50-99
22
4%
37
75
1,653
1%
11,688
100 - 149
36
7%
49
121
4,362
2%
11,153
150 - 249
59
12%
50
196
11,560
6%
10,908
250 - 499
120
24%
41
373
44,729
23%
10,690
500 - 749
132
27%
40
608
80,256
40%
10,325
750 - 999
49
10%
37
826
40,485
20%
10,125
1000 - 1500
11
2%
42
1,264
13,903
7%
9,834
Total Coal
494
100%
42
402
198,416
100%
10,703
NATURAL GAS
0-24
13,616
69%
29
4
60,851
8%
6,356
25-49
1,713
9%
33
38
65,603
8%
7,000
50-99
1,782
9%
28
71
126,171
16%
7,202
100 - 149
802
4%
25
122
98,217
12%
4,935
150 - 249
1,365
7%
16
181
246,875
31%
6,235
250 - 499
394
2%
19
327
128,773
16%
6,115
500 - 749
57
0%
37
584
33,265
4%
7,985
750 - 1000
42
0%
43
879
36,932
5%
9,825
Total Gas
19,771
100%
28
40
796,687
100%
6,439
Source: National Electric Energy Data System (NEEDS) v.6
Note: The average heat rate reported is the mean of the heat rate of the units in each size category (as opposed to a
generation-weighted or capacity-weighted average heat rate.) A lower heat rate indicates a higher level of fuel
efficiency. Table is limited to coal-steam units in operation in 2018 or earlier and excludes those units in NEEDS
with planned retirements in 2020 or 2021.
In terms of the age of the generating units, almost 50 percent of the total coal generating
capacity has been in service for more than 40 years, while nearly 50 percent of the natural gas
capacity has been in service less than 15 years. Figure 2-3 presents the cumulative age
distributions of the coal and gas fleets, highlighting the pronounced differences in the ages of the
fleets of these two types of fossil-fuel generating capacity. Figure 2-3 also includes the
distribution of generation, which is similar to the distribution of capacity.
2-7
-------
Age of EGU (Years)
Coal Cap — — — Coal Gen Gas Cap — — — Gas Gen
Figure 2-3. Cumulative Distribution in 2019 of Coal and Natural Gas Electricity Capacity
and Generation, by Age
Source: eGRID 2019 (March 2021 release from EPA eGRID website). Figure presents data from generators that
came online between 1949 and 2019 (inclusive); a 71-year period. Full eGrid data includes generators that came
online as far back as 1915. Full data from 1915 onward is used in calculating cumulative distributions; figure
truncation at 70 years is merely to improve visibility of diagram. Figure is limited to coal-steam units in NEEDS v6
in operation in 2019 or earlier.
The locations of existing fossil units in EPA's National Electric Energy Data System
(NEEDS) v.6 are shown in Figure 2-4.
2-8
-------
Figure 2-4. Fossil Fuel-Fired Electricity Generating Facilities, by Size
Source: National Electric Energy Data System (NEEDS) v.6
Note: This map displays fossil capacity at facilities in the NEEDS v.6 IPM frame. NEEDS v.6 reflects generating
capacity expected to be on-line at the end of 2023. This includes planned new builds already under construction and
planned retirements. In areas with a dense concentration of facilities, some facilities may be obscured.
2.2.2 Transmission
Transmission is the term used to describe the bulk transfer of electricity over a network of
high voltage lines, from electric generators to substations where power is stepped down for local
distribution. In the U.S. and Canada, there are three separate interconnected networks of high
voltage transmission lines,2 each operating synchronously. Within each of these transmission
networks, there are multiple areas where the operation of power plants is monitored and
controlled by regional organizations to ensure that electricity generation and load are kept in
balance. In some areas, the operation of the transmission system is under the control of a single
2 These three network interconnections are the Western Interconnection, comprising the western parts of both the US
and Canada (approximately the area to the west of the Rocky Mountains), the Eastern Interconnect ion, comprising
the eastern parts of both the US and Canada (except those part of eastern Canada that are in the Quebec
Interconnection), and the Texas Interconnection (which encompasses the portion of the Texas electricity system
commonly known as the Electric Reliability Council of Texas (ERCOT)). See map of all NERC interconnections at
https://www.nerc.com/AboutNERC/keyplayers/PublisliingImages/NERC%20Interconnections.pdf.
2-9
-------
regional operator;3 in others, individual utilities4 coordinate the operations of their generation,
transmission, and distribution systems to balance the system across their respective service
territories.
2.2.3 Distribution
Distribution of electricity involves networks of lower voltage lines and substations that
take the higher voltage power from the transmission system and step it down to lower voltage
levels to match the needs of customers. The transmission and distribution system is the classic
example of a natural monopoly, in part because it is not practical to have more than one set of
lines running from the electricity generating sources to substations or from substations to
residences and businesses.
Over the last few decades, several jurisdictions in the United States began restructuring the
power industry to separate transmission and distribution from generation, ownership, and
operation. Historically, vertically integrated utilities established much of the existing
transmission infrastructure. However, as parts of the country have restructured the industry,
transmission infrastructure has also been developed by transmission utilities, electric
cooperatives, and merchant transmission companies, among others. Distribution, also historically
developed by vertically integrated utilities, is now often managed by a number of utilities that
purchase and sell electricity, but do not generate it. As discussed below, electricity restructuring
has focused primarily on efforts to reorganize the industry to encourage competition in the
generation segment of the industry, including ensuring open access of generation to the
transmission and distribution services needed to deliver power to consumers. In many states,
such efforts have also included separating generation assets from transmission and distribution
assets to form distinct economic entities. Transmission and distribution remain price-regulated
throughout the country based on the cost of service.
2.3 Sales, Expenses, and Prices
These electric generating sources provide electricity for ultimate commercial, industrial
and residential customers. Each of the three major ultimate categories consume roughly a quarter
3 For example, PMJ Interconnection, LLC, Western Area Power Administration (which comprises 4 sub-regions).
4 For example, Los Angeles Department of Power and Water, Florida Power and Light.
2-10
-------
to a third of the total electricity produced5 (see Table 2-4). Some of these uses are highly
variable, such as heating and air conditioning in residential and commercial buildings, while
others are relatively constant, such as industrial processes that operate 24 hours a day. The
distribution between the end use categories changed very little between 2014 and 2020.
Table 2-4. Total U.S. Electric Power Industry Retail Sales, 2014 and 2020 (billion kWh)
2014
2020
Sales/Direct
Sales/Direct
Use (Billion
Share of Total
Use (Billion
Share of Total
kWh)
End Use
kWh)
End Use
Residential
1,407
36%
1,465
38%
Sales
Commercial
1,352
35%
1,287
34%
959
Industrial
998
26%
24%
Transportation
8
0%
7
0%
Total
3,765
96%
3,718
96%
Direct Use
139
4%
139
4%
Total End Use
3,903
100%
3,856
100%
Source: Table 2.2, EIA Electric Power Annual, 2014 and 2020
Notes: Retail sales are not equal to net generation (Table 2-2) because net generation includes net imported
electricity and loss of electricity that occurs through transmission and distribution, along with data collection frame
differences and non-sampling error. Direct Use represents commercial and industrial facility use of onsite net
electricity generation; electricity sales or transfers to adjacent or co-located facilities; and barter transactions.
2.3.1 Electricity Prices
Electricity prices vary substantially across the United States, differing both between the
ultimate customer categories and by state and region of the country. Electricity prices are
typically highest for residential and commercial customers because of the relatively high costs of
distributing electricity to individual homes and commercial establishments. The higher prices for
residential and commercial customers are the result both of the necessary extensive distribution
network reaching to virtually every part of the country and every building, and also the fact that
generating stations are increasingly located relatively far from population centers (which
increases transmission costs). Industrial customers generally pay the lowest average prices,
reflecting both their proximity to generating stations and the fact that industrial customers
receive electricity at higher voltages (which makes transmission more efficient and less
5 Transportation (primarily urban and regional electrical trains) is a fourth ultimate customer category which
accounts less than one percent of electricity consumption.
2-11
-------
expensive). Industrial customers frequently pay variable prices for electricity, varying by the
season and time of day, while residential and commercial prices historically have been less
variable. Overall industrial customer prices are usually considerably closer to the wholesale
marginal cost of generating electricity than residential and commercial prices.
On a state-by-state basis, all retail electricity prices vary considerably. In 2020, the national
average retail electricity price (all sectors) was 10.59 cents/KWh, with a range from 7.51 cents
(Louisiana) to 27.55 cents (Hawaii).6
Average national retail electricity prices decreased between 2014 and 2020 by 8 percent
in real terms (2019$).7 The amount of decrease differed for the three major end use categories
(residential, commercial and industrial). National average industrial prices decreased the most
(14 percent), and residential prices decreased the least (4 percent). The real year prices for 2014
through 2020 are shown in Figure 2-5.
_ 14
•
¦E 2
0
2014 2015 2016 2017 2018 2019 2020
Residential Commercial Industrial — —Total
Figure 2-5. Real National Average Electricity Prices (including taxes) for Three Major
End-Use Categories
Source: EIA Monthly Energy Review (October 2021), Table 9.8.
6 EIA State Electricity Profiles with Data for 2020 (http://www.eia.gov/electricity/state/)
7 All prices in this section are estimated as real 2019 prices adjusted using the GDP implicit price deflator unless
otherwise indicated.
2-12
-------
Most of these electricity price decreases occurred between 2014 and 2015, when nominal
residential electricity prices followed inflation trends, while nominal commercial and industrial
electricity prices declined. The years 2016 and 2017 saw an increase in nominal commercial and
industrial electricity prices, while 2018 and 2019 saw flattening of this growth. Industrial
electricity prices declined in 2019 and 2020. The increase in nominal electricity prices for the
major end use categories, as well as increases in the GDP price and CPI-U indices for
comparison, are shown in Figure 2-6.
12%
-8%
Residential Commercial Industrial
_ —CPI-U GDP Price
Figure 2-6. Relative Increases in Nominal National Average Electricity Prices for Major
End-Use Categories (including taxes), With Inflation Indices
Source: EIA Monthly Energy Review (October 2021), Table 9.8.
For a longer-term perspective, Figure 2-7 shows real (2019$) electricity prices for the
three major customer categories since 1960, and Figure 2-8 shows the relative change in real
electricity prices relative to the prices since 1960. As can be seen in the figures, the price for
industrial customers has always been lower than for either residential or commercial customers,
but the industrial price has been more volatile. While the industrial real price of electricity in
2020 was 11 percent lower than in 1960, residential and commercial real prices are 26 percent
and 35 percent lower respectively than in 1960.
2-13
-------
Figure 2-7. Real National Average Electricity Prices for Three Major End-Use Categories
(including taxes), 1960-2020 (2019$)
Source: EIA Monthly Energy Review. October 2021, Table 9.8
o
KD
tH
o
c
-------
2.3.2 Prices of Fossil Fuels Usedfor Generating Electricity
Another important factor in the changes in electricity prices are the changes in delivered
fuel prices8 for the three major fossil fuels used in electricity generation: coal, natural gas and
residual fuel oil. Relative to real prices in 2014, the national average real price (in 2019$) of coal
delivered to EGUs in 2020 had decreased by 26 percent, while the real price of natural gas
decreased by 56 percent. The real price of delivered residual fuel oil also decreased by 55
percent, and petroleum products declined as an EGU fuel (in 2020 petroleum products generated
0.4% percent of electricity). The combined real delivered price of all fossil fuels (weighted by
heat input) in 2020 decreased by 39 percent over 2014 prices. Figure 2-9 shows the relative
changes in real price of all 3 fossil fuels between 2000 and 2020.
300%
250%
O
O 200%
QJ
c 150%
Coal ¦ Residual Fuel Oil Gas — —Average
Figure 2-9. Relative Real Prices of Fossil Fuels for Electricity Generation; Change in
National Average Real Price per MMBtu Delivered to EGU
Source: EIA Monthly Energy Review, October 2021, Table 9.9.
2.3.3 Changes in Electricity Intensity of the U.S. Economy fi'om 2014 to 2020
An important aspect of the changes in electricity generation (i.e., electricity demand)
between 2014 and 2020 is that while total net generation decreased by 1.4 percent over that
period, the demand growth for generation was lower than both the population growth (4 percent)
8 Fuel prices in this section are all presented in terms of price per MMBtu to make the prices comparable.
2-15
-------
and real GDP growth (10 percent). Figure 2-10 shows the growth of electricity generation,
population and real GDP during this period.
Figure 2-10. Relative Growth of Electricity Generation, Population and Real GDP Since
2014
Sources: Generation: U.S. EIA Monthly Energy Review, October 2021. Table 7.2a Electricity Net Generation: Total
(All Sectors). Population: U.S. Census. Real GDP: 2021 Economic Report of the President, Table B-3.
Because demand for electricity generation grew more slowly than both the population
and GDP, the relative electric intensity of the U.S. economy improved (i.e., less electricity used
per person and per real 2019 dollar of output) during 2014 to 2020. On a per capita basis, real
GDP per capita grew by 6 percent between 2014 and 2020. At the same time electricity
generation per capita decreased by 5 percent. The combined effect of these two changes
improved the overall electricity generation efficiency in the U.S. market economy. Electricity
generation per dollar of real GDP decreased 11 percent. These relative changes are shown in
Figure 2-11.
2-16
-------
Figure 2-11. Relative Change of Real GDP, Population and Electricity Generation Intensity
Since 2014
Sources: Generation: EIA Monthly Energy Review, October 2021. Table 7.2a Electricity Net Generation: Total (All
Sectors). Population: U.S. Census. Real GDP: 2021 Economic Report of the President, Table B-3.
2.4 Deregulation and Restructuring
The process of restructuring and deregulation of wholesale and retail electricity markets
has changed the structure of the electric power industry. In addition to reorganizing asset
management between companies, restructuring sought a functional unbundling of the generation,
transmission, distribution, and ancillary services the power sector has historically provided, with
the aim of enhancing competition in the generation segment of the industry.
Beginning in the 1970s, government policy shifted against traditional regulatory
approaches and in favor of deregulation for many important industries, including transportation
(notably commercial airlines), communications, and energy, which were all thought to be natural
monopolies (prior to 1970) that warranted governmental control of pricing. However,
deregulation efforts in the power sector were most active during the 1990s. Some of the primary
drivers for deregulation of electric power included the desire for more efficient investment
choices, the economic incentive to provide least-cost electric rates through market competition,
reduced costs of combustion turbine technology that opened the door for more companies to sell
power with smaller investments, and complexity of monitoring utilities' cost of service and
2-17
-------
establishing cost-based rates for various customer classes. Deregulation and market restructuring
in the power sector involved the divestiture of generation from utilities, the formation of
organized wholesale spot energy markets with economic mechanisms for the rationing of scarce
transmission resources during periods of peak demand, the introduction of retail choice
programs, and the establishment of new forms of market oversight and coordination.
The pace of restructuring in the electric power industry slowed significantly in response to
market volatility in California and financial turmoil associated with bankruptcy filings of key
energy companies. By the end of 2001, restructuring had either been delayed or suspended in
eight states that previously enacted legislation or issued regulatory orders for its implementation
(shown as "Suspended" in Figure 2-12). Eighteen other states that had seriously explored the
possibility of deregulation in 2000 reported no legislative or regulatory activity in 2001 (EIA,
2003) ("Not Active" in Figure 2-12). Currently, there are 15 states plus the District of Columbia
where price deregulation of generation (restructuring) has occurred ("Active" in Figure 2-12).
Power sector restructuring is more or less at a standstill; by 2010 there were no active proposals
under review by the Federal Energy Regulatory Commission (FERC) for actions aimed at wider
restructuring, and no additional states have begun retail deregulation activity since that time.
Figure 2-12. Status of State Electricity Industry Restructuring Activities
Source: EIA 2010. "Status of Electricity Restructuring by State." Available online at:
.
2-18
-------
One major effect of the restructuring and deregulation of the power sector was a
significant change in type of ownership of electricity generating units in the states that
deregulated prices. Throughout most of the 20th century electricity was supplied by vertically
integrated regulated utilities. The traditional integrated utilities provided generation, transmission
and distribution in their designated areas, and prices were set by cost-of-service regulations set
by state government agencies (e.g., Public Utility Commissions). Deregulation and restructuring
resulted in unbundling of the vertical integration structure. Transmission and distribution
continued to operate as monopolies with cost-of-service regulation, while generation shifted to a
mix of ownership affiliates of traditional utility ownership and some generation owned and
operated by competitive companies known as Independent Power Producers (IPPs). The
resulting generating sector differed by state or region, as the power sector adapted to the
restructuring and deregulation requirements in each state.
By the year 2000, the major impacts of adapting to changes brought about by
deregulation and restructuring during the 1990s were nearing completion. In 2014, traditional
utilities owned 58 percent of U.S. generating capacity (MW) while IPPs9 owned 39 percent of
U.S. generating capacity, respectively. The mix of electricity generated (MWh) was more
heavily weighted towards the utilities, with a distribution in 2014 of 61 percent, and 39 percent
for IPPs. In 2020, the share of capacity (54 percent utility, 43 percent IPPs) and generation (54
percent utility, 42 percent IPP) has remained relatively stable relative to 2014 levels.
The mix of capacity and generation for each of the ownership types is shown in Figures
2-13 (capacity) and 2-14 (generation). A portion of the shift of capacity and generation is due to
sales and transfers of generation assets from traditional utilities to IPPs, rather than strictly the
result of newly built units.
9 IPP data presented in this section include both combined and non-combined heat and power plants.
2-19
-------
Capacity Mix, 2014 & 2020 Generation Mix, 2014 & 2020
700
600
500
2,500
Solar
JZ.
2,000
¦ Other
5
1,500
¦ Wind
c
o
¦ Hydro
¦ Nuclear
¦ Gas
ra
l
0J
c
HI
1,000
500
¦ Coal
0
I
Solar
_ u _ J _ ¦ Other
- 400 ¦ ¦ - ¦ -1'500 ¦ ¦
£- | ¦ ¦ Wind | 9 ¦ ¦ Wind
| 300 I I I 9 - | Hydro 100Q I I I ¦ Hydro
° 200 ¦ ¦ Nuclear $ H ¦ Nuclear
l l II ¦ Gas 500 I I I I' ¦ Gas
® ® ® ® B B
2014 2020 2014 2020 2014 2020 2014 2020
Utility I PR Utility IPP
Figures 2-13. and 2-14. Capacity and Generation Mix by Ownership Type, 2014 & 2020
Source: Table 3.2, EIA Electric Power Annual, 2014 and 2020
2.5 Industrial Sectors Overview
The proposed regulation establishes various ozone season NOx emission limits beginning
in 2026, including emissions limits for reciprocating internal combustion engines in Pipeline
Transportation of Natural Gas; for kilns in Cement and Cement Product Manufacturing; for
boilers and furnaces in Iron and Steel Mills and Ferroalloy Manufacturing; for furnaces in Glass
and Glass Product Manufacturing; and for impactful boilers in Basic Chemical Manufacturing,
Petroleum and Coal Products Manufacturing, and Pulp, Paper, and Paperboard Mills.10 Figure 2-
15 shows the locations11 of the estimated non-EGU emissions reductions by industry. A
description of the Tier 1 and Tier 2 industries, as well as a discussion of how the reductions were
estimated are in Chapter 4, Section 4.4. For additional discussion of the emissions limits, see
Section I.B. of the preamble. The following sections provide overviews of these industries. For
additional information on these non-EGU industries please see the non-EGU Sectors TSD in the
docket.
111 Impactful boilers are boilers with design capacity of 100 minBtu/lir or greater.
11 Facility location information is based on the 2019 inventory, which is discussed in Chapter 4, Section 4.5.4.
2-20
-------
Non-EGU Ozone Season NOx Reductions
Cement and
Concrete
Product
Manufacturing
Glass and
Glass
Product
Manufacturing
LA
TX
OK
PA
IN
MO
OH
Ml
IL
KY
W1
MS
CA
AR
VA
WV
WY
UT
MN
NY
MD
0
1,234
586
888
468
1.296
116
371
234
0
0
0
1.162
0
398
230
446
520
0
142
0
206
1.470
190
1.379
338
227
451
50
901
0
677
0
299
47
174
0
0
0
115
141
0
Iron and
Steel Mills
and
Ferroalloy
Manufacturing
Pipeline
Transportation
of Natural
Gas
High
Emitting
Equipment
from Tier 2
industries
Total
0
3,915
2,649
6,769
0
1,736
0
4,440
0
2,799
0
3.575
438
427
152
3.284
1,829
152
388
3.175
0
1,581
0
3.103
847
1,198
179
2.790
38
2,272
0
2.731
0
1,316
0
2.452
0
2,291
0
2.291
0
0
1.472
2,150
0
1,577
184
1.761
0
137
68
1.666
6
868
732
1.654
92
801
98
1.563
0
751
0
982
0
380
0
826
0
237
0
757
0
558
0
673
0
106
111
500
0
45
0
45
• Cement and Concrete Product Manufacturing O >1000 tons
• Glass and Glass Product Manufacturing O 500-1000 tons
© Iron and Steel Mills and Ferroalloy Manufacturing o 100-500 tons
O Pipeline Transportation of Natural Gas ° Under 100 tons
• High Emitting Equipment from Tier 2 industries
Figure 2-15. Geographical Distribution of Non-EGU Ozone Season NOx Reductions and
Summary of Reductions by Industry and by State
2.5.1 Cement and Cement Product Manufacturing
Hydraulic cement (primarily Portland cement) is a key component of an important
construction material: concrete. Concrete is used in a wide variety of applications (e.g.,
residential and commercial buildings, public works projects), and cement demand is influenced
by national and regional trends in these sectors.
Portland cement is a fine powder, gray or white in color, that consists of a mixture of
hydraulic cement materials comprising primarily calcium silicates, aluminates and alumino-
ferrites. More than 30 raw materials are known to be used in the manufacture of portland cement,
and these materials can be divided into four distinct categories: calcareous, siliceous,
argillaceous, and ferrifrous (containing iron). These materials are chemically combined through
pyroprocessing (heat) and subjected to subsequent mechanical processing operations to form
gray and white portland cement. Gray portland cement is used for structural applications and is
the more common type of cement produced. White portland cement has lower iron and
manganese contents than gray portland cement and is used primarily for decorative purposes.
2-21
-------
There are two processes for manufacturing cement: the wet process and the dry process.
In the wet process, water is added to the raw materials during the blending process and before
feeding the mixture into the rotary kiln. In contrast, the dry process feeds the blended material
directly into the rotary kiln in a dry state. Newer dry process plants also use preheater and
precalciner technologies that partially heat and calcine the blended raw materials before they
enter the rotary kiln. These technologies can increase the overall energy efficiency of the cement
plant and reduce production costs. The fuel efficiency differences between the wet and dry
processes have led to a substantial decline in clinker capacity provided by the wet process over
the last 3 decades. (Van Oss and Padovani, 2002). The number of wet process plants fell from 32
in 2000 to 7 in 2017 (DOI, USGS, 2020).
Cement kilns are used by the cement industry in the production of cement. Portland
cement, used in almost all construction applications, is the industry's primary product.
Essentially all of the NOx emissions associated with cement manufacturing are generated in the
kilns because of high process temperatures. To manufacture cement, raw materials such as
limestone, cement rock, sand, iron ore, clay and shale are crushed, blended, and fed into a kiln.
These materials are then heated in the kiln to temperatures above 2900°F to induce a chemical
reaction (called "fusion") that produces cement "clinker," a round, marble-sized, glass-hard
material. The clinker is then cooled, mixed with gypsum and ground to produce cement. Clinker
is also defined as the product of a portland cement kiln from which finished cement is
manufactured by milling and grinding.
Nearly all cement clinker is produced in large rotary kiln systems. The rotary kiln is a
refractory brick lined cylindrical steel shell equipped with an electrical drive to rotate it at 1-3
revolutions per minute, through which hot combustion gases flow counter-currently to the feed
materials. The kiln can be fired with coal, oil, natural gas, waste (e.g., solvents) or a combination
of these fuels. There are various types of kilns in use, including long wet kilns, long dry kilns,
kilns with a preheater and kilns with a precalciner. The long wet and dry kilns and most
preheater kilns have only one fuel combustion zone, whereas the newer precalciner kilns and
preheater kilns with a riser duct have two fuel combustion zones.
2-22
-------
In a wet kiln, the ground raw materials are suspended in water to form a slurry and
introduced into the inlet feed. This kiln type employs no preheating of the dry feed. In a long dry
kiln, the raw materials are dried to a powder and introduced into the inlet feed in a dry form, but
this kiln type employs no preheating of the dry feed. Currently more cement plants use the dry
process because of its lower energy requirement. In a precalciner kiln, the feed to the kiln system
is preheated in cyclone chambers; the kiln uses a second burner to calcine material in a separate
vessel attached to the preheater before the final fusion in a kiln that forms clinker.
Because the typical operating temperatures of these kilns differ, the NOx formation
mechanisms also differ among these kiln types. In a primary combustion zone at the hot end of a
kiln, the high temperatures lead to predominantly thermal NOx formation. In the secondary
combustion zone, however, lower gas-phase temperatures suppress thermal NOx formation. The
temperatures at which these kilns operate influence what NOx control technologies can be
applied. For instance, SNCR can operate effectively at typical cement kiln temperatures (above
1500°F), while SCR typically operates effectively at lower temperatures (550-800°F). Energy
efficiency is also important in reducing NOx emissions; for example, a high thermal efficiency
equates to less heat and fuel being consumed and, therefore, less NOx is produced.
Portland cement is produced using a combination of variable inputs such as raw materials,
labor, electricity, and fuel. U.S. Census data for the cement industry (North American Industry
Classification System [NAICS] 32731: cement manufacturing) provides an initial overview of
aggregated industry expenditures on these inputs (Department of Commerce [DOC], Bureau of
the Census, 2021). In 2019, the total value of shipments was $9 billion, and the industry spent
approximately $1.5 billion on materials, parts, and packaging, or 16.6% of the value of
shipments. Total compensation for all employees (includes payroll and fringe benefits) amounted
to $1.4 billion (15.6%) and included 15,590 employees.
A review and description of market characteristics (i.e., degree of concentration, entry
barriers, and product differentiation) can enhance our understanding of how U.S. cement markets
operate. These characteristics provide indicators of a firm's ability to influence market prices by
varying the quantity of cement it sells. For example, in markets with large numbers of sellers and
identical products, firms are unlikely to be able to influence market prices via their production
2-23
-------
decisions (i.e., they are "price takers"). However, in markets with few firms, significant barriers
to entry (e.g., licenses, legal restrictions, or high fixed costs), or products that are similar but can
be differentiated, the firm may have some degree of market power (i.e., set or significantly
influence market prices).
Cement sales are often concentrated locally among a small number of firms for two
reasons: high transportation costs and production economies of scale. Transportation costs
significantly influence where cement is ultimately sold; high transportation costs relative to unit
value provide incentives to produce and sell cement locally in regional markets (USITC, 2006).
To support this claim, the empirical literature has typically pointed to Census of Transportation
data showing over 80% of cement shipments were made within a 200-mile radius (Jans and
Rosenbaum, 1997) and reported evidence of high transportation costs per dollar of product value
from case studies (Ryan, 2006). The cement industry is also very capital intensive and entry
requires substantial investments. In additional, large plants are typically more economical
because they can produce cement at lower unit costs; this reduces entry incentives for small sized
cement plants and firms. EPA has recognized these aspects of the cement industry and its market
structure in its economic impact analyses of rules on this industry in previous reports, such as the
RIA prepared in 2010 for the portland cement NESHAP and NSPS (EPA, 2010).
2.5.2 Iron and Steel Mills and Ferroalloy Manufacturing
Iron is produced from iron ore, and steel is produced by progressively removing impurities
from iron ore or ferrous scrap. The first step is iron making. Primary inputs to the iron making
process are iron ore or other sources of iron, coke or coal, and flux. Pig iron is the primary output
of iron making and the primary input to the next step in the process, steel making. Metal scrap
and flux are also used in steel making. The steel making process produces molten steel that is
shaped into solid forms at forming mills. Finishing mills then shape, harden, and treat the semi-
finished steel to yield its final marketable condition.
Blast furnaces are the primary site of iron making at integrated facilities where iron ore is
converted into more pure and uniform iron. Blast furnaces are tall steel vessels lined with heat-
resistant brick (AISI, 1989). They range in size from 23 to 45 feet in diameter and are over 100
2-24
-------
feet tall (Hogan and Koelble, 1996; Lankford et al., 1985). Conveyor systems of carts and ladles
carry inputs and outputs to and from the blast furnace.
Steel making is carried out in basic oxygen furnaces or electric arc furnaces (EAFs), while
iron making is only carried out in blast furnaces. Basic oxygen furnaces are the standard steel
making furnace used at integrated mills. EAFs are the standard furnace at mini-mills since they
use scrap metal efficiently on a small scale. Open hearth furnaces were used to produce steel
prior to 1991 but have not been used in the United States since that time.
Steel often undergoes additional, referred to as secondary, metallurgical processes after it
is removed from the steel making furnace. Secondary steel making takes place in vessels, smaller
furnaces, or the ladle. These sites do not have to be as strong as the primary refining furnaces
because they are not required to contain the powerful primary processes. Secondary steel making
can have many purposes, such as removal of oxygen, sulfur, hydrogen, and other gases by
exposing the steel to a low-pressure environment; removal of carbon monoxide through the use
of deoxidizers such as aluminum, titanium, and silicon; and changing of the composition of
unremovable substances such as oxides to further improve mechanical properties.
In 2019, the United States produced 87.8 million metric tons of steel (USGS, 2019). Steel
is primarily used as a major input to consumer products such as automobiles and appliances.
Therefore, the demand for steel is a derived demand that depends on a diverse base of consumer
products. In addition, the Infrastructure Investment and Jobs Act, signed into law in 2021, will
likely increase demand in both the iron and steel industry as well as the concrete and cement
industry. The historic investment in roads, bridges, airports, and other physical infrastructure
around the country will require large inputs from these industries.
U.S. Census data for the iron and steel industry (North American Industry Classification
System [NAICS] 331110: Iron and steel mills and ferroalloy manufacturing) provides an initial
overview of aggregated industry expenditures on these inputs (Census Bureau, 2021). In 2019,
the total value of shipments was $93.7 billion, and the industry spent approximately $56.4 billion
on materials, parts, and packaging, or 60% of the value of shipments. Total compensation for all
employees (includes payroll and fringe benefits) amounted to $10.1 billion (10.8%) and included
85,707 employees.
2-25
-------
2.5.3 Glass and Glass Product Manufacturing
Commercially produced glass can be classified as soda-lime, lead, fused silica,
borosilicate, or 96 percent silica. Soda-lime glass consists of sand, limestone, soda ash, and cullet
(broken glass). The manufacturing of such glass occurs in four phases: (1) preparation of raw
material, (2) melting in the furnace, (3) forming and (4) finishing. The products of the glass
manufacturing industry are flat glass, container glass, and pressed and blown glass. The
procedures for manufacturing glass are the same for all products except forming and finishing.
Container glass and pressed and blown glass use pressing, blowing, or pressing and blowing to
form the desired product. Flat glass, which is the remainder, is formed by float, drawing, or
rolling processes.
As the sand, limestone, and soda ash raw materials are received, they are crushed and
stored in separate elevated bins. These materials are then transferred through a gravity feed
system to a weigher and mixer, where the material is mixed with cullet to ensure homogeneous
melting. The mixture is conveyed to a batch storage bin where it is held until dropped into the
feeder to the glass melting furnace. All equipment used in handling and preparing the raw
material is housed separately from the furnace and is usually referred to as a batch plant.
The glass melting furnaces contribute to most of the total emissions from the glass plant.
Essentially all of the NOx emissions associated with glass manufacturing are generated in the
melting furnaces due to the high process temperatures. These materials are then heated in the
furnace to temperatures around 3000°F in order to induce fusion that produces molten glass.
After molten glass is produced, it then goes to be shaped by pressing, blowing, pressing and
blowing, drawing, rolling, or floating to produce the desired product. The end products undergo
finishing (decorating or coating) and annealing (removing unwanted stress area in the glass) as
required. During the inspection process, any damaged or undesirable glass is transferred back to
the batch plant to be used as cullet.
Glass manufacturing furnaces can vary between the various categories of glass produced
(container, flat, or pressed/blown). This is because the different types of glass vary in
composition and quality specifications. Therefore, each type of glass produced requires different
energy inputs to fuse the raw materials. As a result, the emissions from similar furnaces
2-26
-------
producing different types of glass can vary significantly. Furnaces can also be fired with gaseous
or liquid fuels.
U.S. Census data for the glass manufacturing industry (North American Industry
Classification System [NAICS] 32721) provides an initial overview of aggregated industry
expenditures on these inputs (Census Bureau, 2021). In 2019, the total value of shipments was
$27.6 billion, and the industry spent approximately $10.9 billion on materials, parts, and
packaging, or 40% of the value of shipments. Total compensation for all employees (includes
payroll and fringe benefits) amounted to $5.3 billion and included 91,988 employees.
2.5.4 Pipeline Transportation of Natural Gas
This industry comprises establishments primarily engaged in the pipeline transportation of
natural gas from processing plants to local distribution systems. This industry includes the
storage of natural gas because the storage is usually done by the pipeline establishment and
because a pipeline is inherently a network in which all the nodes are interdependent.
U.S. Census data for the pipeline transportation of natural gas industry (North American
Industry Classification System [NAICS] 486210) provides an initial overview of aggregated
industry expenditures on these inputs (Census Bureau, 2021). In 2019, the total value of
shipments was $27.6 billion, annual payroll totaled $3.3 billion, and the industry included 27,294
employees.
2.5.5 Tier 2 Industries
This proposed rulemaking includes NOx emission limits on the most impactful boilers
from an additional three industries. The first is the pulp, paper, and paperboard mills industry.
Manufacturing of paper and paper products is a complex process that is carried out in two
distinct phases: the pulping of wood and the manufacture of paper. Pulping is the conversion of
fibrous wood into a "pulp" material suitable for use in paper, paperboard, and building materials.
Pulping and papermaking may be integrated at the same production facility, or facilities may
produce either pulp or paper alone. In addition to facilities that produce pulp and/or paper, there
are numerous establishments that do not manufacture paper, but convert paper into secondary
products.
2-27
-------
Steam boilers are pivotal in the paper industry for the process of drying the paper, energy
requirement, and the cooking of wood chips in the digester. The steam is used for cooking wood
chips, dryer cans, and to produce power for the plant. Power can be produced through the
combustion of bark, black liquor, and fuel oil to reduce the cost with large electric demand and
increase reliability versus outside power sources. Firms engaged in pulp and paper
manufacturing under the North American Industry Classification System (NAICS) code 3221. In
2019, the pulp and paper industry shipped products valued at over $76 billion and included
92,283 employees (U.S. Census Bureau, 2021). This industry has declined in the United States
with a 22% decrease in the number of establishments and a 42% decrease in the number of
employees from 2000 to 2019.
The next industry is the petroleum and coal products manufacturing industry. The
impacted boilers in this industry come from petroleum refineries. Petroleum pumped directly out
of the ground, or crude oil, is a complex mixture of hydrocarbons (chemical compounds that
consist solely of hydrogen and carbon) and various impurities, such as salt. To manufacture the
variety of petroleum products recognized in everyday life, this complex mixture must be refined
and processed over several stages. Boilers are used for several functions in a petroleum refining
facility. The steam generated from the boiler can be used to power turbines and pumps or for
heating of facilities and processes. Large refineries use lots of steam to heat crude oil during the
distillation process.
The process of refining crude oil into useful petroleum products can be separated into two
phases and a number of supporting operations. In the first phase, crude oil is desalted and then
separated into its various hydrocarbon components (known as "fractions"). These fractions
include gasoline, kerosene, naphtha, and other products. In the second phase, the distilled
fractions are converted into petroleum products (such as gasoline and kerosene) using three
different types of downstream processes: combining, breaking, and reshaping (EPA, 1995).
The petroleum refining industry is comprised of establishments primarily engaged in
refining crude petroleum into finished petroleum products. Examples of these products include
gasoline, jet fuel, kerosene, asphalt, lubricants, and solvents. Firms engaged in petroleum
refining are categorized under the North American Industry Classification System (NAICS) code
2-28
-------
324110. In 2019, the petroleum refining industry shipped products valued at over $547 billion
and included 63,659 employees (U.S. Census Bureau, 2021).
The third industry is basic chemical manufacturing, which includes establishments
primarily engaged in manufacturing chemicals using basic processes, such as thermal cracking
and distillation. Chemicals manufactured in this industry group are usually separate chemical
elements or separate chemically-defined compounds.
The chemicals industry is one of the most complex and diverse industries in the U.S., and
simple characterizations are impossible. While the EIA Manufacturing Energy Consumption
Survey (MECS) identifies 10 significant steam-consuming product categories within the
chemical industry, it identifies only nine for the food, paper, refining and primary metals
industries, combined. The major steam consuming processes in the chemical industry include
stripping, fractionalization, power generation, mechanical drive, quenching and dilution.
U.S. Census data for the basic chemical manufacturing industry (North American Industry
Classification System [NAICS] 3251) provides an initial overview of aggregated industry
expenditures. In 2019, the value of shipments for the industry was $206 billion and included
143,000 employees (U.S. Census Bureau, 2021).
2.6 References
American Iron and Steel Institute (AISI). 1989. Steelmaking Flowlines. Washington, DC: AISI
Hogan, William T., and Frank T. Koelble. 1996. "Steel's Coke Deficit: 5.6 Million Tons and
Growing." New Steel 12(12):50-59.
Jans, I. and D. I. Rosenbaum. 1997. "Multimarket Contact and Pricing: Evidence from the U.S.
Cement Industry." International Journal of Industrial Organization 15:391-412.
Ryan, S. 2006. "The Cost of Environmental Regulation in a Concentrated Industry." Available at
.
U.S. Census Bureau. 2021. "Annual Survey of Manufactures: Summary Statistics for Industry
Groups and Industries in the U.S.: 2019 and 2018" Available at:
https ://data. census, gov/cedsci/
2-29
-------
U.S. Department of the Interior, U.S. Geological Survey. 2020. 2017 Minerals Yearbook,
Cement. Washington, DC: U.S. Department of the Interior. Tables 11 and 15.
http://minerals.er.usgs.gov/minerals/pubs/commodity/cement/.
U.S. Environmental Protection Agency. 1995. Economic Impact Analysis for Petroleum
Refineries NESHAP. EPA-452/R-95-003, Final Report. Washington DC: Government
Printing Office.
U.S. Environmental Protection Agency. 2010. Regulatory Impact Analysis: Amendments to
the NESHAP and NSPS for the Portland Cement Manufacturing Industry. Final Report.
Available at Regulatory Impact Analysis: Amendments to the National Emission
Standards for Hazardous Air Pollutants and New Source Performance Standards (NSPS)
for the Portland Cement Manufacturing Industry - Final Report (epa.gov).
U.S. International Trade Commission (USITC). 2006. Gray Portland Cement and Cement
Clinker from Japan Investigation No. 73 l-TA-461 (Second Review). Publication 3856.
Available at
USGS. 2019. "2019 Minerals Yearbook: Iron and Steel" Available at:
https://www.usgs.gov/centers/national-minerals-information-center/iron-and-steel-
statistics-and-information
Van Oss, H.G., and A.C. Padovani. 2002. "Cement Manufacture and the Environment Part I:
Chemistry and Technology." Journal of Industrial Ecology 6(1):89-105.
2-30
-------
CHAPTER 3: AIR QUALITY IMPACTS
Overview
This Chapter describes the impacts on ozone concentrations in 2023 and 2026 of the three
alternative control cases (i.e., proposal case, less stringent case, and more stringent case)
analyzed in this RIA. First, we describe the methods for developing spatial fields of air quality
concentrations for the baseline and regulatory control alternatives in 2023 and 2026. These
spatial fields provide the air quality inputs to potentially calculate health benefits from reduced
concentrations of PM2.5 and ozone for the proposed Federal Implementation Plan (FIP)
Addressing Regional Ozone Transport for the 2015 Ozone National Ambient Air Quality
Standards (FIP for the 2015 ozone NAAQS). In brief, the spatial fields were constructed based
on a method that utilizes ozone contributions from emissions in individual states and state-level
emissions reductions for each of the regulatory control alternatives coupled with baseline spatial
fields of ozone concentrations. This method, as described in Appendix 3 A, was originally
developed to support the RIA for the Repeal of the Clean Power Plan, and the Emission
Guidelines for Greenhouse Gas Emissions from Existing Electric Utility Generating Units (U.S.
EPA 2019) and, most recently, the RIA for the Revised CSAPR Update final rule.
Second, we provide the estimated impacts on projected 2023 and 2026 ozone design
values expected to result from the EGU and non-EGU regulatory control alternatives analyzed in
this RIA. Because of timing constraints, we were not able to perform full-scale photochemical air
quality modeling for these cases to quantify the ozone impacts. Rather, we applied the Air
Quality Assessment Tool (AQAT) that was used to inform the air quality analyses in Step 3 of
the 4-step transport framework as the method for estimating the impacts of the three control
cases.1 The methodology for estimating ozone impacts and the resulting impacts on ozone design
values at individual receptors are provided in Appendix 3B. In Section 3.1 we describe the air
quality modeling platform used for this proposed FIP; in Section 3.2 we describe the method for
processing air quality modeling outputs to create spatial fields for estimating benefits; in Section
3.3 we describe how the method was applied in the proposed FIP for the 2015 ozone NAAQS; in
Section 3.4 we present maps showing the impacts on ozone concentrations of each of the
1 See the Ozone Policy Analysis Proposed Rule TSD which can be found in the docket for this proposed rule for
details on the construction of the ozone AQAT.
3-1
-------
regulatory control alternatives compared to the corresponding baseline; and in Section 3.5 we
identify uncertainties and limitations in the application of the method for generating spatial fields
of pollutant concentrations.
3.1 Air Quality Modeling Platform
The air quality modeling for the proposed FIP utilized a 2016-based modeling platform
which included meteorology and base year emissions from 2016 and projected emissions for
2023 and 2026. The air quality modeling included photochemical model simulations for a 2016
base year and 2023 and 2026 future years to provide hourly concentrations of ozone nationwide.
In addition, source apportionment modeling was performed for 2026 to quantify the
contributions to ozone from NOx emissions from electric generating units (EGUs) and from
point sources other than EGUs (i.e., non-EGUs) on a state-by-state basis. As described below,
the modeling results for 2016, 2023, and 2026, in conjunction with emissions data for the 2023
and 2026 baseline and regulatory control alternatives, were used to construct the air quality
spatial fields that reflect the influence of emissions changes between the baseline and the
regulatory control alternatives.
The air quality model simulations {i.e., model runs) were performed using the
Comprehensive Air Quality Model with Extensions (CAMx) version 7.10 (Ramboll Environ,
2021). Our CAMx nationwide modeling domain {i.e., the geographic area included in the
modeling) covers all lower 48 states plus adjacent portions of Canada and Mexico using a
horizontal grid resolution of 12 x 12 km shown in Figure 3-1.
3-2
-------
Figure 3-1. Air Quality Modeling Domain
The contributions to ozone from EGU and, separately, from non-EGU emissions in
individual states were modeled using a tool called "source apportionment." In general, source
apportionment modeling quantifies the air quality concentrations formed from individual, user-
defined groups of emissions sources or "tags". These source tags are tracked through the
transport, dispersion, chemical transformation, and deposition processes within the model to
obtain hourly gridded2 contributions from the emissions in each individual tag to hourly modeled
concentrations. Thus, the source apportionment method can be used to provide an estimate of the
effect of changes in emissions from each group of emissions sources (i.e., each tag) to changes in
ozone concentrations. For this analysis we applied outputs from source apportionment modeling
for ozone using the 2026 modeled case to obtain the contributions from EGUs and non-EGUs
NOx emissions in each state to ozone concentrations in each 12 x 12 km model grid cell
nationwide. Ozone contributions were modeled using the Ozone Source Apportionment
Technique/Anthropogenic Precursor Culpability Assessment (OSAT/APCA) tool (Ramboll,
2021). The source apportionment modeling was performed for the period April through
September to provide data for developing spatial fields for the April through September
2 Hourly contribution information is provided for each grid cell to provide spatial patterns of the contributions from
each tag.
3-3
-------
maximum daily eight-hour (MDA8) (i.e., AS-M03) average ozone concentration exposure
metric.3'4
3.2 Applying Modeling Outputs to Create Spatial Fields
In this section we describe the method for creating spatial fields of AS-M03 based on the
2016, 2023, and 2026 modeling. The foundational data include (1) ozone concentrations in each
model grid cell from the 2023 and 2026 baseline modeling, (2) contributions in 2026 from EGUs
and non-EGUs emissions from each state in each model grid cell5, (3) 2023 and 2026 emissions
for EGUs and non-EGUs that were input to the contribution modeling, and (4) the EGU and non-
EGU emissions for each of the regulatory scenarios. The method to create spatial fields is based
on scaling ratios that apply emissions changes between the baseline and the control case to the
baseline contributions, described below.
To create the spatial fields for each future emissions scenario the 2026 state-sector source
apportionment modeling outputs are used in combination with the 2023 and 2026 EGU and non-
EGUNOx emissions for each scenario. Contributions from each state-sector contribution "tag"
were scaled based on the ratio of emissions in the year/scenario being evaluated to the emissions
in the modeled 2023 or 2026 baseline scenario. Contributions from tags representing sources
other than EGUs and non-EGUs are held constant at baseline levels for each of the regulatory
alternative scenarios. For each control scenario analyzed, the scaled contributions from all
sources were summed together to create a gridded surface of total modeled ozone. Finally,
spatial fields of ozone were created based on "fusing" modeled data with measured
concentrations at air quality monitoring locations. The process is described in a step-by-step
manner below.
(1) The enhanced Voroni Neighbor Average (eVNA) technique was applied to ozone model
predictions in conjunction measured data to create modeled/measured fused surfaces (i.e.,
spatial fields) of AS-M03 for the 2016 base year.
3 Information on the emissions inventories used for the modeling described in Preparation of Emissions Inventories
for the 2016v2 North American Emissions Modeling Platform
4 The air quality modeling performed to support the analyses in this proposed RIA can be found in the Air Quality
Modeling Technical Support Document Federal Implementation Plan Addressing Regional Ozone Transport for het
2015 Ozone National Ambient Air Quality Standards Proposed Rulemaking
5 Contributions from EGUs and non-EGUs were modeled using baseline emissions for 2026. The resulting
contributions were used to construct spatial fields in both 2023 and 2026.
3-4
-------
(2) The model-predicted spatial fields (i.e., not the eVNA fields) of AS-M03 in 2016 were
paired with the corresponding model-predicted spatial fields in 2023 and 2026 to
calculate the ratio of AS-M03 between 2016 and each of these future year baselines in
each model grid cell.
(3) The ratios for 2016/2023 and 2016/2026 were applied to the eVNA spatial field for 2016
created in step (1) to produce eVNA spatial fields for the 2023 and 2026 baseline
scenarios.
(4) The EGU and non-EGU ozone season NOx emissions for the alternative control
scenarios in 2023 and 2026 and the corresponding 2023 and 2026 baseline NOx
emissions were used to calculate the ratio of control scenario emissions to 2023 and 2026
baseline emissions for each EGU and non-EGU state contribution tag (i.e., an ozone-
season scaling factor for each tag).
(5) The source apportionment modeling provided separate ozone contributions for ozone
formed in VOC-limited chemical regimes (O3V) and ozone formed in NOx-limited
chemical regimes (O3N).6 The EGU and non-EGU NOx emissions for the control
scenarios and the corresponding baseline emissions are used to calculate the ratio of the
control scenario emissions to the baseline emission to create scaling ratios for EGUs and
for non-EGUs. The emissions scaling ratios are multiplied by the corresponding O3N
gridded contributions to MDA8 concentrations. This step results in adjusted gridded
MDA8 contributions due to NOx changes for individual state EGU and non-EGU tags
that reflects the emissions in a specific control scenario.
(6) For MDA8, the adjusted contributions for each EGU and non-EGU state tag from step (3)
are added together to produce adjusted EGU and non-EGU tag totals. Since there are no
predicted changes in VOC emissions in the control scenarios, the O3V contributions
remain unchanged. The contributions from the unaltered O3V tags are added to the
summed adjusted O3N EGU and non-EGU tags.
6 Information on the treatment of ozone contributions under NOx-limited and VOC-limited chemical regimes in the
CAMx APCA source apportionment technique can be found in the CAMx v7.10 User's Guide (Ramboll, 2021).
3-5
-------
(7) The EGU MDA8 contributions from step (6) are then combined with the contributions to
MDA8 from all other sources. This step results in MDA8 concentrations for each of the
control scenario in each model grid cell, nationwide for each day in the ozone season.
(8) We then average the daily MDA8 concentrations across all days in the period April
through September.
(9) The seasonal mean concentrations from step (8) are divided by the corresponding
seasonal mean concentrations from the 2016 base year air quality model run. This step
provides a Relative Response Factor (i.e., RRF) between the base period and control
scenario for MDA8 ozone in each model grid cell.
(10) The RRFs for the AS-M03 metric from step (9) are then multiplied by the
corresponding eVNA 2016 base year from step (1) to produce the eVNA AS-M03
spatial fields for the control scenario that are input to BenMAP-CE.
3.3 Generation of Spatial Fields for the Proposed FIP for the 2015 Ozone NAAQS
In this section we describe how we generated spatial fields of seasonal ozone
concentrations associated with the regulatory control alternatives (i.e., the proposed policy case
and the less stringent and more stringent alternatives). The data for creating spatial fields for
each scenario include (1) EGU and non-EGU ozone season NOx emissions for the 2023 and
2026 baseline scenarios and the regulatory control alternatives, (2) spatial fields of AS-M03 for
the 2023 and 2026 baseline scenarios, and (3) the spatial field of mean AS-M03 ozone
contributions for the hours that correspond to the time periods of MDA8 concentrations.
To calculate ozone-related benefits in 2023 and 2026 we used the ozone season EGU and
non-EGU NOx emissions for the 2023 and 2026 baseline scenarios along with emissions for the
regulatory control alternatives. These emissions were applied using the method described in the
previous section to produce spatial fields of the AS-M03 for the three regulatory cases for EGU
controls in 2023 and the EGU-only, non-EGU-only, and EGU plus non-EGU regulatory cases
analyzed in this RIA.
3-6
-------
3.4 Spatial Distribution of Air Quality Impacts
The spatial fields of baseline AS-M03 in 2023 and 2026 are presented in Figure 3-2 and
Figure 3-3, respectively. It is important to recognize that ozone is a secondary pollutant, meaning
that it is formed through chemical reactions of precursor emissions in the atmosphere. As a result
of the time necessary for precursors to mix in the atmosphere and for these reactions to occur,
ozone is typically not highest at the location of the precursor emissions but rather peaks at some
distance downwind of those emissions sources. The spatial gradients of ozone depend on a
multitude of factors including the spatial patterns of NOx and VOC emissions and the
meteorological conditions on a particular day. Thus, on any individual day, high ozone
concentrations may be found in narrow plumes downwind of specific point sources, may appear
as urban outflow with large concentrations downwind of urban source locations or may have a
more regional signal. However, in general, because the AS-M03 metric is based on the average
of concentrations over more than 180 days in the spring and summer, the resulting spatial fields
are rather smooth without sharp gradients, compared to what might be expected when looking at
the spatial patterns of MDA8 ozone concentrations on specific high ozone episode days.
The distribution of AS-M03 baseline concentrations in 2023 and 2026 are similar, but
the concentrations are somewhat lower in 2026, as is expected due to emissions reductions
resulting from continued implementation of existing "on-the-books" rules and regulations. The
figures show that, from a regional perspective, the highest AS-M03 concentrations are in the
inter-mountain and southwest portions of the western U.S. where contributions from background
sources are dominant outside of urban areas, and in southern and central California where there
are high emissions of ozone precursor pollutants. Within the eastern U.S. the highest
concentrations are seen in the Ohio Valley and portions of the Midwest, as well as along the
Northeast Corridor and near urban areas such as Atlanta and Houston.
3-7
-------
Figure 3-2. Baseline AS-M03 concentration in 2023 (ppb).
2023 Baseline: April-Sept MDA8 03
6aselin«_2023
159 239
Mln* 24.370 at (396.11). Max = 71.308 at (48.99)
'
I
65 O
60 O
55 O
50.0
45.0
40.0
35.0
30 O
25.0
3-8
-------
20.0
Figure 3-3. Baseline AS-M03 concentration in 2026 (ppb).
The estimated impacts on AS-M03 between the baseline and each of the regulatory
control alternatives for 2023 and 2026 are presented in Figure 3-4 through Figure 3-15. Note that
the impacts of the control alternatives in 2026 are much larger than the impacts of the control
alternatives in 2023. In this regard, the scale used to display the impacts is different for the 2023
cases compared to the 2026 cases. Note that the scale ranges from 0 to 0.1 ppb on the plots for
2023, whereas the scale ranges from 0 to 1.0 ppb on the plots for 2026 because the impacts in
2026 are much greater than in 2023.
The data shown in Figures 3-4 through 3-15 are calculated as the baseline minus the
regulatory control alternative concentrations (i.e., positive values indicate reductions in pollutant
concentrations). The spatial patterns of the impacts of emissions reductions are a result of (1) the
spatial distribution of EG I • and non-EGU sources with changes in emissions between the
3-9
-------
baseline and the individual regulatory control alternatives and (2) the physical or chemical
processing that the model simulates in the atmosphere.
2023 Baseline - EGU Less Stringent Case
Baseltne_2023 - 2023 Pr EGU Less
Mln = 0 OOE+O at (1,1). Max = 0 299 at (246,124)
Figure 3-4. Reduction in AS-M03 (ppb):
2023 baseline - less stringent EGU-only alternative (scale: + 0.1 ppb).
3-10
-------
2023 Baseline - EGU Policy Case
Baseline_2023 - 2023 Pr EGU Policy
h.
i
—
i m
r4J
W\ J
I
V \
> \
j "V--t*^*"
Vjf
J ^
V .v'
vr
yK» 1
r L
I
159 239
Min = 0 00E+0 at (1,1). Max = 0 299 at (246.124)
Figure 3-5. Reduction in AS-M03 (ppb): 2023 baseline - EGU-only proposed rule
alternative (scale: + 0.1 ppb).
2023 Baseline - More Stringent Case
Basellrie_2023 - 2023 Pr EGU More
A_
/ /
J X
! W
F
/
*
j~\
i
159 239
Min = -0 012 at (345,126). Max = 0 303 at (246,124)
Figure 3-6. Reduction in AS-M03 (ppb):
2023 baseline - more stringent EGU-only alternative (scale: + 0.1 ppb).
3-11
-------
2026 Baseline - EGU Less Stringent Case
Basel ine_2026 - 2026 Pr EGU Less
159 239
Min = -0.042 at (34,111), Max = 0.889 at (235,57)
Figure 3-7. Reduction in AS-M03 (ppb):
2026 baseline - less stringent EGU-only alternative (scale: + 1 ppb).
2026 Baseline - EGU Policy Case
Baseline_2026- 2026 Pr EGU Policy
159 239 318
Min = -9.90E-3 at (33,111), Max = 1.797 at (238,87)
Figure 3-8. Reduction in AS-M03 (ppb):
2026 baseline - EGU-only proposed rule alternative (scale: + 1 ppb).
3-12
-------
2026 Baseline - EGU More Stringent Case
Baseline_2026 - 2026 Pr EGU More
Figure 3-9. Reduction in AS-M03 (ppb):
2026 baseline - more stringent EGU-only alternative (scale: + 1 ppb).
159 239
Min = -0.011 at (33,111), Max = 1.828 at (238,87)
2026 Baseline - Non-EGU Less Stringent Case
Baseline_2026 - 2026 Pr NONEGU Less
159 239
Min = 0.00E+0 at (1,1), Max = 0.884 at (252,119)
n
Figure 3-10. Reduction in AS-M03 (ppb):
2026 baseline - less stringent noil-EGlJ-only alternative (scale: + 1 ppb).
3-13
-------
2026 Baseline - Non-EGU Policy Case
Baseline_2026- 2026 Pr NONEGU Policy
Figure 3-11. Reduction in AS-M03 (ppb):
2026 baseline - non-EGU-only proposed rule alternative (scale: + 1 ppb)
2026 Baseline - Non-EGU More Stringent case
Baseline_2026 - 2026 Pr NONEGU More
159 239
Min = 0.00E+0 at (1,1), Max = 0.941 at (252,119)
Figure 3-12. Reduction in AS-M03 (ppb):
2026 baseline - more stringent non-EGU-only alternative (scale: + 1 ppb)
3-14
-------
2026 Baseirie - EGU & Non-EGU Less Stringent Case
Baseline_2026 - 2026 Pr EGU+NONEGU Less
159 239 318
Min = -8.95E-3 at (47,235), Max = 1.231 at (235,57)
Figure 3-13. Reduction in AS-M03 (ppb):
2026 baseline - less stringent EGU+non-EGU alternative (scale: + 1 ppb).
2026 Baseline - EGU & Non-EGU Policy Case
Baseline_2026 - 2026 Pr EGU+NONEGU Policy
Min = -6.61E-3 at (47,235), Max = 2.177 at (238,87)
Figure 3-14. Reduction in AS-M03 (ppb):
2026 baseline - EGU+non-EGU proposed rule alternative (scale: + 1 ppb).
3-15
-------
2026 Baseline - EGU & Non-EGU More Stringent Case
Basel i ne_2026 - 2026 Pr EGU+NONEGU More
1 80 159 239 318 397
Min = -6.54E-3 at (47,235), Max = 2.236 at (238,87)
Figure 3-15. Reduction in AS-M03 (ppb):
2026 baseline - more stringent EGU+non-EGU alternative (scale: + X ppb).
3.5 Uncertainties and Limitations
One limitation of the scaling methodology for creating ozone surfaces associated with the
baseline and regulatory alternatives described above is that it treats air quality changes from the
tagged sources as linear and additive. It therefore does not account for nonlinear atmospheric
chemistry and does not account for interactions between emissions of different pollutants and
between emissions from different tagged sources. This is consistent with how air quality
estimations have been treated in past regulatory analyses (U.S. EPA 2012; 2019; 2020b). We
note that air quality is calculated in the same manner for the baseline and the regulatory
alternatives, so any uncertainty associated with these assumptions is carried through both sets of
scenarios in the same manner and is thus not expected to impact the air quality differences
between scenarios. In addition, emissions changes between baseline and the regulatory
alternatives are relatively small compared to modeled 2023 emissions that form the basis of the
source apportionment approach described in Appendix 3 A. Previous studies have shown that air
pollutant concentrations generally respond linearly to small emissions changes of up to 30
percent (Dunker et al., 2002; Cohan et al., 2005; Napelenok et al., 2006; Koo et al., 2007; Zavala
3-16
-------
et al., 2009; Cohan and Napelenok, 2011) and that linear scaling from source apportionment can
do a reasonable job of representing impacts of 100 percent of emissions from individual sources
(Baker and Kelly 2014). Therefore, while simplistic, it is reasonable to expect that the emissions
concentration differences between the baseline and regulatory control alternatives can be
adequately represented using this methodology and any uncertainty should be weighed against
the speed in which this method may be used to account for spatial differences in the effect of
EGU emissions on ozone concentrations.
A second limitation is that the source apportionment contributions represent the spatial
and temporal distribution of the emissions from each source tag as they occur in the 2026
modeled case. Thus, the contribution modeling results do not allow us to represent any changes
to "within tag" spatial distributions. As a result, the method does not account for any changes of
spatial patterns that would result from changes in the relative magnitude of sources within a
source tag in the scenarios investigated here.
In addition, the 2023 and 2026 CAMx-modeled concentrations themselves have some
uncertainty. While all models have some level of inherent uncertainty in their formulation and
inputs, the base-year 2016 model outputs have been evaluated against ambient measurements
and have been shown to adequately reproduce spatially and temporally varying ozone
concentrations (EPA, 2022).
The regulatory alternatives lead to decreased concentrations of ozone, the extent to which
varies by location, relative to the baseline. However, the analysis does not account for how
interaction with NAAQS compliance would affect the benefits and costs of the regulatory
alternatives, which introduces uncertainty in the benefits and costs of the alternatives. To the
extent the proposed FIP for the 2015 ozone NAAQS will decrease NOx and consequentially
ozone concentrations, these changes may affect compliance with existing NAAQS standards and
subsequently affect the actual benefits and costs of the rule. In areas not projected to attain the
2015 ozone NAAQS without further emissions reductions from the baseline, states may be able
avoid applying some emissions control measures to reduce emissions from local sources as a
result of this rule. If compliance behavior with the 2015 ozone NAAQS were accounted for in
the baseline in this RIA there may be additional financial and economic benefits from reduced
3-17
-------
compliance costs, while the level and spatial pattern of changes in ozone concentrations, and
their associated health and ecological benefits, would differ. The directional effect on the
benefits, costs, and net-benefits of this source of uncertainty is ambiguous.
Similarly, the regulatory alternatives may project decreases in ozone concentrations in
areas attaining the NAAQS in the baseline. In practice, these potential changes in concentrations
may influence NAAQS compliance plans in these areas, which in turn would further influence
concentrations and the cost of complying with the NAAQS. However, such behavior will be
mitigated by NAAQS requirements such as Prevention of Significant Deterioration (PSD)
requirements. This RIA does not account for how interaction with NAAQS compliance would
affect the benefits and costs of the regulatory alternatives.
3.6 References
Baker, Kirk R., and James T. Kelly. 2014. "Single Source Impacts Estimated with
Photochemical Model Source Sensitivity and Apportionment Approaches." Atmospheric
Environment 96 (October): 266-74. https://doi.Org/10.1016/j.atmosenv.2014.07.042.
Cohan Daniel S., Amir Hakami, Yongtao Hu, Armistead G. Russell. 2005. "Nonlinear response
of ozone to emissions: Source apportionment and sensitivity analysis." Environmental
Science & Technology 39:6739-6748
Cohan, Daniel, and Sergey Napelenok. 2011. "Air Quality Response Modeling for Decision
Support." Atmosphere 2 (December): 407-25. https://doi.org/10.3390/atmos2030407.
Ding, Dian, Yun Zhu, Carey Jang, Che-Jen Lin, Shuxiao Wang, Joshua Fu, Jian Gao, Shuang
Deng, Junping Xie, and Xuezhen Qiu. 2016. "Evaluation of Health Benefit Using
BenMAP-CE with an Integrated Scheme of Model and Monitor Data during Guangzhou
Asian Games." Journal of Environmental Sciences 42 (April): 9-18.
https://doi.Org/10.1016/j.jes.2015.06.003.
Dunker, Alan M., Greg Yarwood, Jerome P. Ortmann, and Gary M. Wilson. 2002. "The
Decoupled Direct Method for Sensitivity Analysis in a Three-Dimensional Air Quality
Model Implementation, Accuracy, and Efficiency." Environmental Science &
Technology 36 (13): 2965-76. https://doi.org/10.1021/es0112691.
Hand, J. L., B.A. Schichtel, M. Pitchford, W.C. Malm, andN.H. Frank. 2012. "Seasonal
Composition of Remote and Urban Fine Particulate Matter in the United States. Journal
of Geophysical Research, 117, D05209, doi: 10.1029/2011JD017122.
Koo, Bonyoung, Alan M. Dunker, and Greg Yarwood. 2007. "Implementing the Decoupled
Direct Method for Sensitivity Analysis in a Particulate Matter Air Quality Model."
3-18
-------
Environmental Science & Technology 41 (8): 2847-54.
https://doi.org/10.1021/es0619962.
Napelenok, Sergey L., Daniel S. Cohan, Yongtao Hu, and Armistead G. Russell. 2006.
"Decoupled Direct 3D Sensitivity Analysis for Particulate Matter (DDM-3D/PM)."
Atmospheric Environment A0 (32): 6112-21.
https://doi.Org/10.1016/j.atmosenv.2006.05.039.
Ramboll Environ. 2021. "Comprehensive Air Quality Model with Extensions Version 7.10."
User's Guide. Novato, CA: Ramboll Environ International Corporation.
http://www.camx.com/files/camxusersguide_v7-10.pdf.
US EPA, 2012. "Regulatory Impact Analysis for the Final Revisions to the National Ambient Air
Quality Standards for Particulate Matter." EPA-452/R-12-005. Research Triangle Park,
NC: U.S. Environmental Protection Agency.
https://www3.epa.gov/ttnecasl/regdata/RIAs/finalria.pdf.
US EPA, 2017. Documentation for the EPA's Preliminary 2028 Regional Haze Modeling.
Research Triangle Park, NC
(https://www3.epa.gov/ttn/scram/reports/2028_Regional_Haze_Modeling-TSD.pdf).
US EPA, 2019. "Regulatory Impact Analysis for the Repeal of the Clean Power Plan, and the
Emission Guidelines for Greenhouse Gas Emissions from Existing Electric Utility
Generating Units." EPA-452/R-19-003. Research Triangle Park, NC: U.S. Environmental
Protection Agency, https://www.epa.gov/sites/production/files/2019-
06/documents/utilities_ria_final_cpp_repeal_and_ace_2019-06.pdf.
US EPA, 2020a. "Regulatory Impact Analysis for Revisions to the Effluent Limitations
Guidelines and Standards for the Steam Electric Power Generating Point Source
Category". EPA-821-R-20-004. Washington, DC: U.S. Environmental Protection
Agency, https://www.epa.gov/sites/production/files/2020-
08/documents/steam_electric_elg_2020_final_reconsideration_rule_regulatory_impact_a
nalysis.pdf.
US EPA, 2020b. "Analysis of Potential Costs and Benefits for the "National Emission Standards
for Hazardous Air Pollutants: Coal- and Oil-Fired Electric Utility Steam Generating
Units - Subcategory of Certain Existing Electric Utility Steam Generating Units Firing
Easter". Memo to Docket for rulemaking: "National Emission Standards for Hazardous
Air Pollutants: Coal- and Oil-Fired Electric Utility Steam Generating Units -
Subcategory of Certain Existing Electric Utility Steam Generating Units Firing Eastern
Bituminous Coal Refuse for Emissions of Acid Gas Hazardous Air Pollutants" (EPA-
HQ-OAR-2018-0794), April 8, 2020. Available at:
https://www.epa.gOv/sites/production/files/2020-04/documents/mats_coal_refuse_cost-
benefit_memo.pdf
US EPA, 2022. "Air Quality Modeling Technical Support Document: Federal Implementation
Plan Addressing Regional Ozone Transport for the 2015 Ozone National Ambient Air
Quality Standards Proposed Rulemaking.
3-19
-------
Zavala, M., Lei, W., Molina, M.J., Molina, L.T., 2009. Modeled and observed ozone sensitivity
to mobile-source emissions in Mexico City. Atmos. Chem. Phys. 9, 39-55.
3-20
-------
APPENDIX 3A: METHODOLOGY FOR DEVELOPING AIR QUALITY SURFACES
In this appendix we describe the methodology that was used to prepare the air quality
surfaces that could inform the calculation of health benefits of the proposed Federal
Implementation Plan (FIP) Addressing Regional Ozone Transport for the 2015 Ozone National
Ambient Air Quality Standards (FIP for the 2015 ozone NAAQS). As described in chapter 3, the
foundational data include (1) spatial fields of April through September MDA8 concentrations for
the 2023 and 2026 baselines, (2) ozone season EGU and non-EGU emissions for the baseline
scenarios and each of the regulatory control alternatives in 2023 and 2026, and (3) the 2026 EGU
and non-EGU April through September MDA8 ozone contribution data.
3A.1 Applying Source Apportionment Contributions to Create Air Quality Fields
Air quality surfaces for the 2023 and 2026 baseline and regulatory control alternatives
were created by scaling the EGU and non-EGU sector tagged contributions from the 2026
modeling based on relative changes in EGU and/or non-EGU emissions associated with each
tagged category between the modeled scenario and the 2023 and 2026 baseline and regulatory
control alternatives. Below, we provide equations used to apply these scaling ratios.
3A. 1.1 Creating Fused Fields Based on Observations and Model Surfaces
In this section we describe steps taken to create ozone gridded surfaces that combine
modeled and monitor data to estimate ozone concentrations in 2023 and 2026 that serve as the
starting point for estimating ozone under the baseline and regulatory control in 2023 and 2026
respectively. Ozone MDA8 concentrations were processed into April through September average
surfaces which combine observed values with model predictions using the enhanced Veronoi
Neighbor Average (eVNA) method (Gold et al., 1997; US EPA, 2007; Ding et al., 2015). First,
we create a 2016 eVNA surface for MDA8 ozone using EPA's software package, Software for
the Modeled Attainment Test - Community Edition (SMAT-CE)1 . SMAT-CE calculates April
through September MDA8 average values (i.e., AS-M03) at each monitoring site with available
1 Software download and documentation available at https://www.epa.gov/scram/photochemical-modeling-tools.
Software has been previously documented both in the user's guide for the predecessor software (Abt, 2014) and in
EPA's modeling guidance document (U.S. EPA, 2014b).
3A-1
-------
measured data. For this calculation we used 3 years of monitoring data (2015-2017)2. SMAT-CE
then creates an interpolated field of the AS-M03 using inverse distance weighting resulting in a
separate 3-year average interpolated observed field for this metric. The interpolated observed
fields are then adjusted to match the spatial gradients from the 2016 modeled data. These two
steps can be calculated using Equation (1):
eVNAgi2016 = 1 WeightxMonitorXi2Qls_2Q17M0^2016 (Eq-1)
Moa(SlX,2016
Where:
• eVNAg 2016 is the gradient adjusted AS-M03 eVNA value at grid-cell, g in 2016
• Weightx is the inverse distance weight for monitor x at the location of grid-cell,
g;
• Monitor^ 2015-2017 is the 3-year (2015-2017) AS-M03, at monitor, x;
• Modelg 2016 is the 2016 modeled AS-M03 concentrations at grid cell, g; and
• Modelx 2016 is the 2016 modeled AS-M03 concentration at the location of
monitor, x.
The 2016 eVNA field serves as the starting point for future-year eVNA surfaces for the 2023 and
2026 modeled cases. To create a gridded 2023 and 2026 eVNA surfaces, we take the ratio of the
modeled future year3 AS-M03 concentration to the modeled 2016 AS-M03 concentration in
each grid cell then and multiply that ratio by the corresponding 2016 eVNA AS-M03
concentration in that grid cell (Equation 2).
• eVNAgJuture = (eVNAg2016) x Mode^2oig (Eq-2)
2 Three years of ambient data is used to provide a more representative picture of air pollution concentrations.
3 In this analysis the "future year" represents either the 2023 or 2026 modeled case.
3A-2
-------
3A. 1.2 Scaling Ratio Applied to Source Apportionment Tags
The relative contributions from each source in the 2026 source apportionment modeling
is applied to adjust the 2023 and 2026 eVNA surfaces to estimate AS-M03 associated with the
2023 and 2026 baseline and regulatory control scenarios. Source apportionment contributions
from EGU and NONEGU source are scaled to represent emissions in the baseline and regulatory
control scenarios for each year. Scaling ratios for ozone formed in NOx-limited regimes4
("03N") were based on relative changes in ozone season (May-September) NOx emissions.
Scaling ratios for ozone formed in VOC-limited regimes ("03 V") were set to 1 in all cases
because no changes in VOC emissions were simulated as part of this rule. The scaling ratios
were determined based on emissions provided for each scenario. Relative contributions from all
other sources remain the same as the relative contributions from the 2026 source apportionment
modeling. The final AS-M03 for each scenario is calculated using equation (3):
Ozonegiy = eVNAgy
• Ozoneg i y is the estimated fused model-obs AS-M03 for grid-cell, "g", scenario,
"i"5, and year, "y"6;
• eVNAg y is the eVNA future year AS-M03 for grid-cell "g" and year "y" calculated
using Eq-13.
4 The CAMx model internally determines whether the ozone formation regime is NOx-limited or VOC-limited
depending on predicted ratios of indicator chemical species.
5 Scenario "i" can represent either baseline or regulatory control scenario.
6 Year "y" can represent either 2023 or 2026.
£NONEGUVOC,g,t
(Eq-3)
Cg,Tot
where:
3A-3
-------
Cgjot is the total modeled AS-M03 for grid-cell "g" from all source in the 2026
source apportionment modeling
Cg Bc is the 2026 AS-M03 modeled contribution from the modeled boundary inflow;
Cg int is the 2026 AS-M03 modeled contribution from international emissions within
the modeling domain;
Cg.bio is the 2026 AS-M03 modeled contribution from biogenic emissions;
Cg,f^s is the 2026 AS-M03 modeled contribution from fires;
Cg usanthro is the total 2026 AS-M03 modeled contribution from U.S. anthropogenic
sources other than EGUs and non-EGUs;
CEGuvoc,g,t is the 2026 AS-M03 modeled contribution from EGU emissions of VOCs
from state, "t";
£EGUNOx.g.d.t is the 2026 AS-M03 modeled contribution from EGU emissions of
NOx from state, "t";
£NONEGUvoc,g,d,t's the 2026 AS-M03 modeled contribution from EGU emissions of
VOCs from state, "t";
£NONEGUNOx,g,d,t's the 2026 AS-M03 modeled contribution from EGU emissions of
NOx from state, "t"; and
Se,t,i,y is the EGU NOx scaling ratio for state, "t", scenario "i", and year, "y".
Sn,t,i,y is the non-EGU NOx scaling ratio for state, "t", scenario "i", and year, "y".
3A-4
-------
3A.2 References
82 FR 1733. Notice of Availability of the Environmental Protection Agency's Preliminary
Interstate Ozone Transport Modeling Data for the 2015 Ozone National Ambient Air
Quality Standard (NAAQS), (January 6, 2017).
Abt Associates, 2014. User's Guide: Modeled Attainment Test Software.
http://www.epa.gov/scram001/modelingapps_mats.htm
Cohan, D.S., Hakami, A., Hu, Y.T., Russell, A.G., 2005. Nonlinear response of ozone to
emissions: Source apportionment and sensitivity analysis. Environ. Sci. Technol. 39,
6739-6748.
Cohan, D.S., Napelenok, S.L., 2011. Air Quality Response Modeling for Decision Support.
Atmosphere. 2, 407-425.
Ding, D., Zhu, Y., Jang, C., Lin, C., Wang, S., Fu, J., Gao, J., Deng, S., Xie, J., Qui, X., 2015.
Evaluation of heath benefit using BenMAP-CE with an integrated scheme of model and
monitor data during Guangzhou Asian Games. Journal of Environmental Science. 29,
178-188.
Dunker, A.M., Yarwood, G., Ortmann, J.P., Wilson, G.M., 2002. The decoupled direct method
for sensitivity analysis in a three-dimensional air quality model—Implementation,
accuracy, and efficiency. Environ. Sci. Technol. 36, 2965-2976.
Gold C, Remmele P.R., Roos T., 1997. In: Algorithmic Foundation of Geographic Information
Systems. In: Lecture Notes in Computer Science, Vol. 1340 (van Kereveld M, Nievergelt
J, Roos T, Widmayer P, eds) Berlin, Germany: Springer-Verlag. Voronoi methods in
GIS. pp. 21-35.
Henderson, B.H., Akhtar, F., Pye, H.O.T., Napelenok, S.T., Hutzell, W.T., 2014. A Database and
Tool for Boundary Conditions for Regional Air Quality Modeling: Description and
Evaluations, Geoscientific Model Development. 7, 339-360.
Koo, B., Dunker, A.M., Yarwood, G., 2007. Implementing the decoupled direct method for
sensitivity analysis in a particulate matter air quality model. Environ. Sci. Technol. 41,
2847-2854.
Napelenok, S.L., Cohan, D.S., Hu, Y., Russell, A.G., 2006. Decoupled direct 3D sensitivity
analysis for particulate matter (DDM-3D/PM). Atmospheric Environment. 40, 6112-
6121.
Ramboll Environ, 2021. User's Guide Comprehensive Air Quality Model with Extensions
version 7.10. Ramboll Environ International Corporation, Novato, CA.
Simon, H., Baker, K. R., Phillips, S., 2012. Compilation and interpretation of photochemical
model performance statistics published between 2006 and 2012, Atmos. Environ. 61,
124-139.
3A-5
-------
US EPA, 2007, Technical Report on Ozone Exposure, Risk, and Impact Assessments for
Vegetation. EPA 452/R-07-002. Prepared by Abt Associates Inc. for U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Health and
Environmental Impacts Division. Research Triangle Park, NC.
(https://www3.epa.gov/ttn/naaqs/standards/ozone/data/2007_01_environmental_tsd.pdf).
US EPA, 2014b, Modeling Guidance for Demonstrating Attainment of Air Quality Goals for
Ozone, PM2.5, and Regional Haze- December 2014 DRAFT, Research Triangle Park,
NC. (https://www3.epa.gov/ttn/scram/guidance/guide/Draft_03-PM-
RH_Modeling_Guidance-2014.pdf).
US EPA, 2015, Regulatory Impact Analysis of the Final Revisions to the National Ambient Air
Quality Standards for Ground-Level Ozone, EPA-452/R-15-07, Research Triangle Park,
NC. (https://www.epa.gOv/sites/production/files/2016-02/documents/20151001ria.pdf).
US EPA, 2017. Air Quality Modeling Technical Support Document Federal Implementation
Plan Addressing Regional Ozone Transport for the 2015 Ozone National Ambient Air.
Zavala, M., Lei, W., Molina, M.J., Molina, L.T., 2009. Modeled and observed ozone
sensitivity to mobile-source emissions in Mexico City. Atmos. Chem. Phys. 9, 39-55.
3A-6
-------
APPENDIX 3B: OZONE IMPACTS OF ALTERNATIVE CONTROL CASES
In this appendix we provide the estimated impacts on projected 2023 and 2026 ozone
design values that are expected to result from the EGU and non-EGU control alternatives
analyzed in this RIA. As described in Chapter 4, the alternative scenarios include the proposed
rule control case along with scenarios that reflect less stringent and more stringent controls on
EGU and non-EGUs. Because of timing constraints, we were not able to perform full-scale
photochemical air quality modeling for these cases to quantify the ozone impacts. Rather, we
applied the Air Quality Assessment Tool (AQAT) that was used to inform the air quality
analyses in Step 3 of the 4-step transport framework as the method for estimating the impacts of
the three control cases.1 In the application of AQAT for the analysis presented here, we started
with the model-projected average and maximum design values and state-to-receptor air quality
contributions for the 2023 and 2026 base case scenarios at individual receptors along with the
emissions changes in 2023 and in 2026 that are expected to result from the implementation of
emissions controls for the alternative cases analyzed in this RIA. Using the emissions data, we
calculated emissions reduction fractions compared to the 2026 base case and then we applied
these fractions to the 2026 state-to-receptor contribution data to modulate the contributions at
each receptor. Next, the change in contributions were adjusted using "calibration factors" to
reflect the effects of the nonlinear response of ozone to changes in NOx emissions. The
"calibrated" change in contributions were then subtracted from the corresponding 2023 or 2026
base case contributions to reflect how the base case contributions to that receptor are expected to
change as a result of emissions reductions. The adjusted state-to-receptor contributions are then
summed to estimate design values for each control case. Finally, the control case design values
are compared to the corresponding base case values to determine the "ppb" impacts at individual
receptors. In the application of AQAT to estimate ozone impacts at individual receptors, we
included the combined effects of emissions reductions in each linked state at each receptor. As
part of this approach, the impacts at an individual receptor reflect the effects of emissions
reductions in all upwind states, not just those upwind states linked to that particular receptor. In
1 See the Ozone Policy Analysis Proposed Rule TSD which can be found in the docket for this proposed rule for
details on the construction of the ozone AQAT.
3B-1
-------
addition, the ozone reductions at a receptor also reflect the impact from emissions within that
receptor's state, if that state is linked to a receptor in another state.2
3B.1 Analysis of Emissions Reductions
In Tables 3B-1 and 3B-2, respectively, we provide the ozone season state total NOx
emissions (tons) for the 2023 and 2026 base case scenarios along with the changes in emissions
by state expressed in terms of tons reduced (i.e., emissions delta) and percent reduction from the
corresponding base case.3 Details on the factors which drive these emissions changes can be
found in Chapter 4. In 2023 the magnitude of emissions reductions expected from the proposed
case and the less stringent case are very similar in most states. In the more stringent case,
emissions reductions are notably greater than the proposed case in Illinois, Kentucky, and
Pennsylvania. The controls included in the three alternative cases are expected to reduce state
total ozone season NOx emissions from 1 to 2 percent in Kentucky, Minnesota, Missouri, and
Utah with lesser percent reductions in other state covered by this proposed rule. In 2026, the
magnitude and geographic extent of emissions reductions are both much greater than in 2023 due
to the additional control opportunities by 2026 for reducing emissions from EGUs and the
availability of controls for reducing emissions from non-EGUs. In contrast to 2023, the emission
reductions from the proposal case and more stringent case are similar, but notably exceed the
amount of reduction in the less stringent case. Under the proposal and more stringent cases, 21
states are expected to see reductions in total NOx emissions of greater than 5 percent. Moreover,
in the 2026 proposal case, NOx reductions of 10 percent or more are expected in Arkansas,
Kentucky, Louisiana, Mississippi, and Wyoming.
2 For example, the impacts on ozone at receptors in Texas reflect the effects of emissions reductions in Texas
combined with emissions reductions in all states that are upwind of Texas.
3 The explanation for the change in emissions between each alternative emissions control scenario and the
corresponding base case is provided in Chapter 4.
3B-2
-------
Table 3B-1. Total anthropogenic 2023 base case NOx emissions, emissions deltas, and
percent reductions by st
ate for eac
i alternative control case.
State
2023
Base
NOx
Emission
Delta
Proposed
Rule
Emission
Delta
Less
Stringent
Emission
Delta
More
Stringent
Percent
Reduction
Proposed
Rule
Percent
Reduction
Less
Stringent
Percent
Reduction
More
Stringent
Alabama
66,312
0
0
3
0.0%
0.0%
0.0%
Arizona
38,612
0
0
0
0.0%
0.0%
0.0%
Arkansas
43,202
206
207
111
0.5%
0.5%
0.3%
California
139,593
3
3
3
0.0%
0.0%
0.0%
Colorado
53,121
0
0
0
0.0%
0.0%
0.0%
Connecticut
11,820
0
0
-2
0.0%
0.0%
0.0%
Delaware
6,878
5
7
5
0.1%
0.1%
0.1%
District of
Columbia
1,390
0
0
0
0.0%
0.0%
0.0%
Florida
100,080
5
5
5
0.0%
0.0%
0.0%
Georgia
67,589
-2
-2
-2
0.0%
0.0%
0.0%
Idaho
19,622
0
0
0
0.0%
0.0%
0.0%
Illinois
97,086
24
25
155
0.0%
0.0%
0.2%
Indiana
73,491
241
239
270
0.3%
0.3%
0.4%
Iowa
46,836
-12
-16
2
0.0%
0.0%
0.0%
Kansas
62,587
15
15
32
0.0%
0.0%
0.1%
Kentucky
54,506
635
588
1,034
1.2%
1.1%
1.9%
Louisiana
103,038
245
245
231
0.2%
0.2%
0.2%
Maine
14,097
-4
-2
-4
0.0%
0.0%
0.0%
Maryland
25,735
4
4
-3
0.0%
0.0%
0.0%
Massachusetts
28,105
-1
-1
3
0.0%
0.0%
0.0%
Michigan
80,760
-16
-15
6
0.0%
0.0%
0.0%
Minnesota
62,656
971
972
970
1.5%
1.6%
1.5%
Mississippi
34,435
37
37
37
0.1%
0.1%
0.1%
Missouri
76,251
1,448
1,447
1,445
1.9%
1.9%
1.9%
Montana
28,408
0
0
0
0.0%
0.0%
0.0%
Nebraska
43,826
-11
-11
-17
0.0%
0.0%
0.0%
Nevada
18,286
3
4
4
0.0%
0.0%
0.0%
New
Hampshire
7,287
-1
-1
-1
0.0%
0.0%
0.0%
New Jersey
34,476
53
51
51
0.2%
0.1%
0.1%
New Mexico
65,186
48
48
44
0.1%
0.1%
0.1%
New York
69,960
68
68
92
0.1%
0.1%
0.1%
North
Carolina
58,908
3
2
0
0.0%
0.0%
0.0%
4 Note that positive values indicate reductions and negative values indicate increases in emissions.
3B-3
-------
Emission
Emission
Emission
Percent
Percent
Percent
2023
Delta
Delta
Delta
Reduction
Reduction
Reduction
Base
Proposed
Less
More
Proposed
Less
More
State
NOx
Rule
Stringent
Stringent
Rule
Stringent
Stringent
North Dakota
59,167
12
12
12
0.0%
0.0%
0.0%
Ohio
85,480
62
45
97
0.1%
0.1%
0.1%
Oklahoma
90,114
51
51
58
0.1%
0.1%
0.1%
Oregon
33,155
0
0
0
0.0%
0.0%
0.0%
Pennsylvania
107,022
58
117
231
0.1%
0.1%
0.2%
Rhode Island
4,559
0
0
1
0.0%
0.0%
0.0%
South
Carolina
43,650
0
0
-5
0.0%
0.0%
0.0%
South Dakota
12,972
0
0
0
0.0%
0.0%
0.0%
Tennessee
52,389
0
1
-4
0.0%
0.0%
0.0%
Texas
305,019
826
828
823
0.3%
0.3%
0.3%
Utah
35,692
688
688
689
1.9%
1.9%
1.9%
Vermont
3,853
0
0
0
0.0%
0.0%
0.0%
Virginia
50,590
57
56
-107
0.1%
0.1%
-0.2%
Washington
53,412
0
0
0
0.0%
0.0%
0.0%
West Virginia
43,830
0
0
0
0.0%
0.0%
0.0%
Wisconsin
45,503
-2
0
0
0.0%
0.0%
0.0%
Wyoming
34,211
99
99
98
0.3%
0.3%
0.3%
Tribal Data
4,057
0
0
0
0.0%
0.0%
0.0%
Table 3B-2. Total 2026 base case NOx emissions, emissions deltas, and percent reductions
jy state for each alternative control case.
Emission
Emission
Emission
Percent
Percent
Percent
2026
Delta
Delta
Delta
Reduction
Reduction
Reduction
Base
Proposed
Less
More
Proposed
Less
More
State
NOx
Rule
Stringent
Stringent
Rule
Stringent
Stringent
Alabama
61,759
162
67
183
0%
0%
0%
Arizona
33,463
-2
-4
-24
0%
0%
0%
Arkansas
39,488
5,542
1,406
5,569
14%
4%
14%
California
133,629
1,636
1,500
1,746
1%
1%
1%
Colorado
49,825
-172
-239
-174
0%
0%
0%
Connecticut
10,887
0
1
9
0%
0%
0%
Delaware
6,447
5
5
6
0%
0%
0%
District of
Columbia
1,302
0
0
0
0%
0%
0%
Florida
92,166
249
239
282
0%
0%
0%
Georgia
60,266
83
8
25
0%
0%
0%
Idaho
17,321
-3
-17
-3
0%
0%
0%
3B-4
-------
Emission
Emission
Emission
Percent
Percent
Percent
2026
Delta
Delta
Delta
Reduction
Reduction
Reduction
Base
Proposed
Less
More
Proposed
Less
More
State
NOx
Rule
Stringent
Stringent
Rule
Stringent
Stringent
Illinois
91,069
4,135
3,157
4,184
5%
3%
5%
Indiana
68,291
6,047
3,554
6,173
9%
5%
9%
Iowa
41,049
77
-436
89
0%
-1%
0%
Kansas
59,107
402
389
403
1%
1%
1%
Kentucky
50,887
8,980
5,446
9,285
18%
11%
18%
Louisiana
100,361
11,402
8,225
13,012
11%
8%
13%
Maine
12,918
0
0
1
0%
0%
0%
Maryland
23,671
49
50
1
0%
0%
0%
Massachusetts
26,353
2
1
9
0%
0%
0%
Michigan
75,940
7,156
5,348
7,642
9%
7%
10%
Minnesota
55,972
1,747
726
1,858
3%
1%
3%
Mississippi
33,156
3,904
1,546
3,901
12%
5%
12%
Missouri
67,664
6,391
4,541
6,397
9%
7%
9%
Montana
25,642
-117
-118
-117
0%
0%
0%
Nebraska
38,322
-8
-17
-1
0%
0%
0%
Nevada
16,178
-8
-7
-9
0%
0%
0%
New
Hampshire
6,719
1
1
2
0%
0%
0%
New Jersey
31,805
56
61
88
0%
0%
0%
New Mexico
62,210
93
86
91
0%
0%
0%
New York
65,642
800
515
1,545
1%
1%
2%
North
Carolina
51,986
24
26
15
0%
0%
0%
North Dakota
55,294
746
1,300
729
1%
2%
1%
Ohio
78,681
4,006
3,429
4,289
5%
4%
5%
Oklahoma
83,411
4,223
3,574
4,500
5%
4%
5%
Oregon
29,345
11
8
12
0%
0%
0%
Pennsylvania
103,565
3,440
3,185
5,998
3%
3%
6%
Rhode Island
4,187
0
0
0
0%
0%
0%
South
Carolina
38,939
94
91
128
0%
0%
0%
South Dakota
11,084
-8
-7
-8
0%
0%
0%
Tennessee
47,475
-33
-146
-15
0%
0%
0%
Texas
280,717
10,438
9,288
12,576
4%
3%
4%
Utah
29,762
2,774
1,681
2,801
9%
6%
9%
Vermont
3,378
0
0
0
0%
0%
0%
Virginia
46,496
1,680
1,537
1,724
4%
3%
4%
Washington
47,754
-27
-34
-27
0%
0%
0%
3B-5
-------
Emission
Emission
Emission
Percent
Percent
Percent
2026
Delta
Delta
Delta
Reduction
Reduction
Reduction
Base
Proposed
Less
More
Proposed
Less
More
State
NOx
Rule
Stringent
Stringent
Rule
Stringent
Stringent
West Virginia
39,500
923
673
994
2%
2%
3%
Wisconsin
41,032
2,023
689
2,101
5%
2%
5%
Wyoming
32,928
3,336
1,299
3,338
10%
4%
10%
Tribal Data
4,052
0
0
0
0%
0%
0%
3B.2 Projected Impacts on Ozone Design Values
The expected impacts on ozone design value in 2023 and 2026 for the proposed, less
stringent, and more stringent cases are provided in Tables 3B-3 and 3B-4 respectively. In 2023,
there is little difference in the amount of ozone reduction across the three cases at individual
receptors, which is consistent with the expected changes in NOx emissions, as shown in Table
3B-1, above. Overall, in 2023 the estimated ozone reductions from all three of the alternative
cases are projected to be less than 0.1 ppb at most receptors. The exceptions are at certain
receptors in Connecticut, Illinois, Texas, and Utah where impacts are between 0.1 and 0.2 ppb.
In the 2026 the largest impacts in the proposed case are estimated at the two receptors in Texas
(i.e., Brazoria County and Harris County, where the average reduction is 1.3 ppb. Elsewhere, the
average reductions for the proposed case are on the order of 0.5 ppb at receptors in Connecticut,
Illinois, and Wisconsin. The average reduction for the four receptors in Utah is approximately
0.3 ppb, while the average reduction at receptors in Colorado and California reductions are
approximately 0.2 ppb. The data in Table 3B-4 indicates that the less stringent case provides
approximately 0.1 to 0.3 ppb less ppb reduction (i.e., 30 to 40 percent less reduction), on
average, compared to the proposed case at receptors in the East and in Colorado and Utah. The
more stringent case does not appear to provide any notable additional ozone reductions
compared to the proposed case in all receptor areas, except at receptors in Connecticut and Texas
where the average reduction is 0.1 ppb and 0.2 ppb with the more stringent case, respectively.
Table 3B-3. Impact on projected 2023 design value of the emissions reductions in the
Site ID
State
County
Proposed
Case
Less
Stringent
More
Stringent
40278011
AZ
Yuma
0.09
0.09
0.09
60070007
CA
Butte
0.03
0.03
0.03
3B-6
-------
Site ID
State
County
Proposed
Case
Less
Stringent
More
Stringent
60090001
CA
Calaveras
0.09
0.09
0.09
60170010
CA
El Dorado
0.00
0.00
0.00
60170020
CA
El Dorado
0.09
0.09
0.09
60190007
CA
Fresno
0.04
0.04
0.04
60190011
CA
Fresno
0.03
0.03
0.03
60190242
CA
Fresno
0.04
0.04
0.04
60194001
CA
Fresno
0.07
0.07
0.07
60195001
CA
Fresno
0.03
0.03
0.03
60250005
CA
Imperial
0.02
0.02
0.02
60251003
CA
Imperial
0.06
0.06
0.06
60290007
CA
Kern
0.03
0.03
0.03
60290008
CA
Kern
0.02
0.02
0.02
60290011
CA
Kern
0.06
0.06
0.06
60290014
CA
Kern
0.03
0.03
0.03
60290232
CA
Kern
0.09
0.09
0.09
60292012
CA
Kern
0.03
0.03
0.03
60295002
CA
Kern
0.01
0.01
0.01
60311004
CA
Kings
0.06
0.06
0.06
60370002
CA
Los Angeles
0.09
0.09
0.09
60370016
CA
Los Angeles
0.02
0.02
0.02
60371103
CA
Los Angeles
0.06
0.06
0.06
60371201
CA
Los Angeles
0.10
0.10
0.10
60371602
CA
Los Angeles
0.04
0.04
0.04
60371701
CA
Los Angeles
0.01
0.01
0.01
60372005
CA
Los Angeles
0.02
0.02
0.02
60376012
CA
Los Angeles
0.00
0.00
0.00
60379033
CA
Los Angeles
0.01
0.01
0.01
60390004
CA
Madera
0.07
0.07
0.07
60392010
CA
Madera
0.02
0.02
0.02
60430003
CA
Mariposa
0.04
0.04
0.04
60470003
CA
Merced
0.06
0.06
0.06
60570005
CA
Nevada
0.05
0.05
0.05
60592022
CA
Orange
0.09
0.09
0.09
60595001
CA
Orange
0.06
0.06
0.06
60610003
CA
Placer
0.01
0.01
0.01
60610004
CA
Placer
0.02
0.02
0.02
60610006
CA
Placer
0.00
0.00
0.00
60650008
CA
Riverside
0.06
0.06
0.06
60650012
CA
Riverside
0.03
0.03
0.03
3B-7
-------
Site ID
State
County
Proposed
Case
Less
Stringent
More
Stringent
60650016
CA
Riverside
0.00
0.00
0.00
60651016
CA
Riverside
0.02
0.02
0.02
60652002
CA
Riverside
0.02
0.02
0.02
60655001
CA
Riverside
0.05
0.05
0.05
60656001
CA
Riverside
0.05
0.05
0.05
60658001
CA
Riverside
0.00
0.00
0.00
60658005
CA
Riverside
0.03
0.03
0.03
60659001
CA
Riverside
0.04
0.04
0.04
60670002
CA
Sacramento
0.09
0.09
0.09
60670012
CA
Sacramento
0.00
0.00
0.00
60675003
CA
Sacramento
0.00
0.00
0.00
60710001
CA
San Bernardino
0.03
0.03
0.03
60710005
CA
San Bernardino
0.03
0.03
0.03
60710012
CA
San Bernardino
0.08
0.08
0.08
60710306
CA
San Bernardino
0.02
0.02
0.02
60711004
CA
San Bernardino
0.04
0.04
0.04
60711234
CA
San Bernardino
0.04
0.04
0.04
60712002
CA
San Bernardino
0.04
0.04
0.04
60714001
CA
San Bernardino
0.06
0.06
0.06
60714003
CA
San Bernardino
0.08
0.08
0.08
60719002
CA
San Bernardino
0.00
0.00
0.00
60719004
CA
San Bernardino
0.09
0.09
0.09
60731006
CA
San Diego
0.08
0.08
0.08
60773005
CA
San Joaquin
0.00
0.00
0.00
60990005
CA
Stanislaus
0.01
0.01
0.01
60990006
CA
Stanislaus
0.03
0.03
0.03
61070006
CA
Tulare
0.04
0.04
0.04
61070009
CA
Tulare
0.08
0.08
0.08
61072002
CA
Tulare
0.01
0.01
0.01
61072010
CA
Tulare
0.04
0.04
0.04
61090005
CA
Tuolumne
0.06
0.06
0.06
61112002
CA
Ventura
0.08
0.08
0.08
80350004
CO
Douglas
0.03
0.03
0.03
80590006
CO
Jefferson
0.11
0.11
0.11
80590011
CO
Jefferson
0.04
0.04
0.04
90010017
CT
Fairfield
0.09
0.09
0.10
90013007
CT
Fairfield
0.11
0.11
0.12
90019003
CT
Fairfield
0.10
0.10
0.11
90099002
CT
New Haven
0.12
0.12
0.13
3B-8
-------
Site ID
State
County
Proposed
Case
Less
Stringent
More
Stringent
170310001
IL
Cook
0.04
0.04
0.04
170310032
IL
Cook
0.07
0.07
0.07
170310076
IL
Cook
0.03
0.03
0.03
170314201
IL
Cook
0.07
0.07
0.07
170317002
IL
Cook
0.11
0.10
0.11
320030075
NV
Clark
0.07
0.07
0.07
420170012
PA
Bucks
0.07
0.07
0.08
480391004
TX
Brazoria
0.15
0.15
0.15
481210034
TX
Denton
0.12
0.12
0.12
482010024
TX
Harris
0.14
0.14
0.14
482010055
TX
Harris
0.16
0.16
0.15
482011034
TX
Harris
0.16
0.16
0.16
482011035
TX
Harris
0.17
0.17
0.16
490110004
UT
Davis
0.07
0.07
0.07
490353006
UT
Salt Lake
0.13
0.13
0.13
490353013
UT
Salt Lake
0.12
0.12
0.12
490570002
UT
Weber
0.12
0.12
0.12
490571003
UT
Weber
0.12
0.12
0.12
550590019
WI
Kenosha
0.06
0.06
0.06
550590025
WI
Kenosha
0.04
0.04
0.04
551010020
WI
Racine
0.06
0.06
0.07
Table 3B-4. Impact on projected 2026 design value of the emissions reductions in the
proposed case and the less stringent and more stringent cases (PPb)- .
4. Impact on projected 2026 design value of the emissions reductions in
case and the less stringent and more stringent cases (PPb)-
Site ID
State
County
Proposed
Case
Less
Stringent
More
Stringent
40278011
AZ
Yuma
0.07
0.07
0.07
60090001
CA
Calaveras
0.18
0.17
0.18
60170010
CA
El Dorado
0.18
0.17
0.19
60170020
CA
El Dorado
0.20
0.19
0.21
60190007
CA
Fresno
0.14
0.13
0.14
60190011
CA
Fresno
0.19
0.18
0.19
60190242
CA
Fresno
0.20
0.19
0.21
60194001
CA
Fresno
0.16
0.15
0.17
60195001
CA
Fresno
0.16
0.15
0.17
60250005
CA
Imperial
0.05
0.05
0.05
60251003
CA
Imperial
0.03
0.03
0.03
60290007
CA
Kern
0.23
0.21
0.24
60290008
CA
Kern
0.24
0.22
0.25
3B-9
-------
Site ID
State
County
Proposed
Case
Less
Stringent
More
Stringent
60290011
CA
Kern
0.12
0.11
0.13
60290014
CA
Kern
0.22
0.20
0.23
60290232
CA
Kern
0.20
0.19
0.21
60292012
CA
Kern
0.26
0.25
0.27
60295002
CA
Kern
0.28
0.27
0.30
60311004
CA
Kings
0.16
0.15
0.17
60370002
CA
Los Angeles
0.26
0.24
0.27
60370016
CA
Los Angeles
0.25
0.24
0.27
60371201
CA
Los Angeles
0.19
0.17
0.19
60371602
CA
Los Angeles
0.12
0.11
0.13
60371701
CA
Los Angeles
0.27
0.25
0.28
60372005
CA
Los Angeles
0.18
0.17
0.19
60376012
CA
Los Angeles
0.17
0.15
0.18
60379033
CA
Los Angeles
0.21
0.20
0.22
60390004
CA
Madera
0.17
0.15
0.17
60392010
CA
Madera
0.18
0.17
0.19
60430003
CA
Mariposa
0.07
0.06
0.07
60470003
CA
Merced
0.18
0.17
0.19
60570005
CA
Nevada
0.16
0.14
0.17
60592022
CA
Orange
0.16
0.15
0.17
60595001
CA
Orange
0.15
0.14
0.16
60610003
CA
Placer
0.21
0.20
0.22
60610004
CA
Placer
0.19
0.18
0.20
60610006
CA
Placer
0.16
0.15
0.17
60650008
CA
Riverside
0.19
0.18
0.20
60650012
CA
Riverside
0.23
0.21
0.24
60650016
CA
Riverside
0.15
0.14
0.16
60651016
CA
Riverside
0.20
0.18
0.21
60652002
CA
Riverside
0.13
0.11
0.13
60655001
CA
Riverside
0.20
0.19
0.21
60656001
CA
Riverside
0.24
0.23
0.25
60658001
CA
Riverside
0.20
0.18
0.21
60658005
CA
Riverside
0.22
0.20
0.23
60659001
CA
Riverside
0.20
0.18
0.21
60670012
CA
Sacramento
0.21
0.20
0.22
60710001
CA
San Bernardino
0.14
0.13
0.15
60710005
CA
San Bernardino
0.25
0.23
0.27
60710012
CA
San Bernardino
0.24
0.23
0.25
60710306
CA
San Bernardino
0.17
0.15
0.18
3B-10
-------
Site ID
State
County
Proposed
Case
Less
Stringent
More
Stringent
60711004
CA
San Bernardino
0.26
0.24
0.27
60711234
CA
San Bernardino
0.17
0.16
0.17
60712002
CA
San Bernardino
0.21
0.20
0.23
60714001
CA
San Bernardino
0.23
0.21
0.24
60714003
CA
San Bernardino
0.28
0.26
0.29
60719002
CA
San Bernardino
0.15
0.14
0.16
60719004
CA
San Bernardino
0.27
0.25
0.28
60731006
CA
San Diego
0.10
0.09
0.10
60773005
CA
San Joaquin
0.18
0.17
0.19
60990005
CA
Stanislaus
0.16
0.14
0.17
60990006
CA
Stanislaus
0.18
0.17
0.19
61070006
CA
Tulare
0.15
0.14
0.15
61070009
CA
Tulare
0.18
0.18
0.19
61072002
CA
Tulare
0.18
0.17
0.19
61072010
CA
Tulare
0.14
0.13
0.15
61090005
CA
Tuolumne
0.12
0.11
0.13
80350004
CO
Douglas
0.21
0.11
0.22
80590006
CO
Jefferson
0.17
0.09
0.17
80590011
CO
Jefferson
0.18
0.10
0.18
90010017
CT
Fairfield
0.48
0.39
0.60
90013007
CT
Fairfield
0.57
0.43
0.73
90019003
CT
Fairfield
0.46
0.35
0.58
90099002
CT
New Haven
0.54
0.41
0.67
170310001
IL
Cook
0.54
0.34
0.57
170310032
IL
Cook
0.32
0.19
0.35
170310076
IL
Cook
0.44
0.26
0.48
170314201
IL
Cook
0.49
0.32
0.52
170317002
IL
Cook
0.55
0.38
0.58
480391004
TX
Brazoria
1.37
0.97
1.55
482010024
TX
Harris
1.27
0.96
1.47
490110004
UT
Davis
0.38
0.27
0.38
490353006
UT
Salt Lake
0.31
0.21
0.32
490353013
UT
Salt Lake
0.40
0.27
0.41
490570002
UT
Weber
0.31
0.19
0.31
550590019
WI
Kenosha
0.51
0.32
0.54
550590025
WI
Kenosha
0.58
0.37
0.61
551010020
WI
Racine
0.59
0.38
0.62
3B-11
-------
3B.3 Alternative Control Case Projected Ozone Design Values
The projected average and maximum design values in 2023 at individual receptors are
provided in Table 3B-5 for the proposed, less stringent and more stringent cases. Comparing the
magnitude of the design values relative to the level of the NAAQS indicates that three of 101
receptors in 2023 are projected to change attainment status as a result of this proposed rule.
Specifically, receptors in Clark County, Nevada, Butte County, California, and Riverside County
California (Monitor ID: 060650008) are projected to switch from maintenance-only in the 2023
base case to attainment and the receptor in Harris County, Texas is projected to switch from
nonattainment to maintenance-only under any of the alternative cases in 2023. In 2026, six of 89
receptors are projected to change attainment status as a result of the proposed rule. Specifically,
receptors in Calaveras County, California, Brazoria County, Texas, and in Kenosha County,
Wisconsin (Monitor ID: 550590025) are projected to switch from maintenance-only to
attainment in 2026 and a receptor in Riverside County, California (Monitor ID: 060650016) is
projected to switch from nonattainment to maintenance under any of the alternative cases. The
receptor in Douglas County, Colorado and one of the receptors in Cook County, Illinois (Monitor
ID: 170310076) are projected to switch from maintenance-only to attainment under the proposed
and more stringent cases, but these receptors are projected to remain as maintenance-only in the
less stringent case.
Table 3B-5. Projected average and maximum design values for the 2023 base case, the
Site ID
State
County
2023
Avg
9 """ "
2023
Max
Proposed
Avg
Proposed
Max
Less
Stringent
Avg
Less
Stringent
Max
More
Stringent
Avg
More
Stringent
Max
40278011
AZ
Yuma
70.5
72.2
70.4
72.2
70.4
72.2
70.4
72.2
60070007
CA
Butte
68.9
71.0
68.8
70.9
68.8
70.9
68.8
70.9
60090001
CA
Calaveras
70.9
71.9
70.8
71.8
70.8
71.8
70.8
71.8
60170010
CA
El Dorado
76.3
78.7
76.2
78.7
76.2
78.7
76.2
78.7
60170020
CA
El Dorado
74.3
76.2
74.2
76.1
74.2
76.1
74.2
76.1
60190007
CA
Fresno
80.4
82.2
80.3
82.2
80.3
82.2
80.3
82.2
60190011
CA
Fresno
82.9
83.8
82.8
83.8
82.8
83.8
82.8
83.8
60190242
CA
Fresno
79.5
81.1
79.4
81.1
79.4
81.1
79.4
81.1
60194001
CA
Fresno
82.8
84.4
82.7
84.3
82.7
84.3
82.7
84.3
60195001
CA
Fresno
83.7
86.4
83.6
86.4
83.6
86.4
83.6
86.4
60250005
CA
Imperial
76.3
76.6
76.2
76.6
76.2
76.6
76.2
76.6
3B-12
-------
Site ID
State
County
2023
Avg
2023
Max
Proposed
Avg
Proposed
Max
Less
Stringent
Avg
Less
Stringent
Max
More
Stringent
Avg
More
Stringent
Max
60251003
CA
Imperial
75.4
75.4
75.3
75.3
75.3
75.3
75.3
75.3
60290007
CA
Kern
82.8
84.0
82.7
84.0
82.7
84.0
82.7
84.0
60290008
CA
Kern
79.1
81.0
79.0
81.0
79.0
81.0
79.0
81.0
60290011
CA
Kern
78.8
80.4
78.7
80.4
78.7
80.4
78.7
80.4
60290014
CA
Kern
81.3
83.2
81.2
83.2
81.2
83.2
81.2
83.2
60290232
CA
Kern
74.9
77.5
74.8
77.4
74.8
77.4
74.8
77.4
60292012
CA
Kern
84.1
84.7
84.0
84.8
84.0
84.8
84.0
84.8
60295002
CA
Kern
82.4
84.0
82.3
84.0
82.3
84.0
82.3
84.0
60311004
CA
Kings
76.9
77.6
76.8
77.5
76.8
77.5
76.8
77.5
60370002
CA
Los
Angeles
88.0
92.4
87.9
92.4
87.9
92.4
87.9
92.4
60370016
CA
Los
Angeles
93.4
96.2
93.3
96.2
93.3
96.2
93.3
96.2
60371103
CA
Los
Angeles
70.5
71.5
70.4
71.5
70.4
71.5
70.4
71.5
60371201
CA
Los
Angeles
82.7
85.3
82.6
85.2
82.6
85.2
82.6
85.2
60371602
CA
Los
Angeles
73.6
73.9
73.5
73.9
73.5
73.9
73.5
73.9
60371701
CA
Los
Angeles
85.6
88.4
85.5
88.4
85.5
88.4
85.5
88.4
60372005
CA
Los
Angeles
80.7
81.9
80.6
81.9
80.6
81.9
80.6
81.9
60376012
CA
Los
Angeles
91.6
93.4
91.5
93.4
91.5
93.4
91.5
93.4
60379033
CA
Los
Angeles
80.7
82.2
80.6
82.3
80.6
82.3
80.6
82.3
60390004
CA
Madera
75.7
78.3
75.6
78.2
75.6
78.2
75.6
78.2
60392010
CA
Madera
77.0
78.2
76.9
78.2
76.9
78.2
76.9
78.2
60430003
CA
Mariposa
74.2
77.1
74.1
77.1
74.1
77.1
74.1
77.1
60470003
CA
Merced
74.7
75.9
74.6
75.9
74.6
75.9
74.6
75.9
60570005
CA
Nevada
78.1
81.5
78.0
81.4
78.0
81.4
78.0
81.4
60592022
CA
Orange
72.5
72.8
72.4
72.8
72.4
72.8
72.4
72.8
60595001
CA
Orange
72.3
73.0
72.2
73.0
72.2
73.0
72.2
73.0
60610003
CA
Placer
77.1
79.8
77.0
79.8
77.0
79.8
77.0
79.8
60610004
CA
Placer
71.9
77.0
71.8
77.0
71.8
77.0
71.8
77.0
60610006
CA
Placer
72.8
73.7
72.7
73.7
72.7
73.7
72.7
73.7
60650008
CA
Riverside
71.0
73.3
70.9
73.3
70.9
73.3
70.9
73.3
60650012
CA
Riverside
85.9
88.3
85.8
88.3
85.8
88.3
85.8
88.3
60650016
CA
Riverside
72.0
72.9
71.9
72.9
71.9
72.9
71.9
72.9
3B-13
-------
Site ID
State
County
2023
Avg
2023
Max
Proposed
Avg
Proposed
Max
Less
Stringent
Avg
Less
Stringent
Max
More
Stringent
Avg
More
Stringent
Max
60651016
CA
Riverside
89.8
90.9
89.7
90.9
89.7
90.9
89.7
90.9
60652002
CA
Riverside
76.4
78.5
76.3
78.5
76.3
78.5
76.3
78.5
60655001
CA
Riverside
80.5
82.6
80.4
82.6
80.4
82.6
80.4
82.6
60656001
CA
Riverside
83.5
84.1
83.4
84.1
83.4
84.1
83.4
84.1
60658001
CA
Riverside
89.5
90.7
89.4
90.6
89.4
90.6
89.4
90.6
60658005
CA
Riverside
87.9
90.7
87.8
90.6
87.8
90.6
87.8
90.6
60659001
CA
Riverside
80.8
82.9
80.7
82.9
80.7
82.9
80.7
82.9
60670002
CA
Sacramento
71.4
71.7
71.3
71.7
71.3
71.7
71.3
71.7
60670012
CA
Sacramento
74.8
75.4
74.7
75.5
74.7
75.5
74.7
75.5
60675003
CA
Sacramento
70.2
71.7
70.1
71.8
70.1
71.8
70.1
71.8
60710001
CA
San
Bernardino
74.5
75.4
74.4
75.4
74.4
75.4
74.4
75.4
60710005
CA
San
Bernardino
100.3
101.8
100.2
101.8
100.2
101.8
100.2
101.8
60710012
CA
San
Bernardino
87.3
90.1
87.2
90.1
87.2
90.1
87.2
90.1
60710306
CA
San
Bernardino
76.8
78.6
76.7
78.6
76.7
78.6
76.7
78.6
60711004
CA
San
Bernardino
97.2
100.2
97.1
100.2
97.1
100.2
97.1
100.2
60711234
CA
San
Bernardino
70.6
74.2
70.5
74.3
70.5
74.3
70.5
74.3
60712002
CA
San
Bernardino
90.1
91.3
90.0
91.3
90.0
91.3
90.0
91.3
60714001
CA
San
Bernardino
82.6
83.3
82.5
83.3
82.5
83.3
82.5
83.3
60714003
CA
San
Bernardino
95.2
98.0
95.1
98.0
95.1
98.0
95.1
98.0
60719002
CA
San
Bernardino
80.1
81.6
80.1
81.7
80.1
81.7
80.1
81.7
60719004
CA
San
Bernardino
99.5
101.6
99.4
101.6
99.4
101.6
99.4
101.6
60731006
CA
San Diego
76.9
77.9
76.8
77.8
76.8
77.8
76.8
77.8
60773005
CA
San
Joaquin
71.3
72.8
71.2
72.9
71.2
72.9
71.2
72.9
60990005
CA
Stanislaus
75.4
76.3
75.3
76.3
75.3
76.3
75.3
76.3
60990006
CA
Stanislaus
77.5
77.8
77.4
77.8
77.4
77.8
77.4
77.8
61070006
CA
Tulare
79.1
80.3
79.0
80.3
79.0
80.3
79.0
80.3
3B-14
-------
Site ID
State
County
2023
Avg
2023
Max
Proposed
Avg
Proposed
Max
Less
Stringent
Avg
Less
Stringent
Max
More
Stringent
Avg
More
Stringent
Max
61070009
CA
Tulare
82.6
82.6
82.5
82.5
82.5
82.5
82.5
82.5
61072002
CA
Tulare
75.5
77.6
75.4
77.6
75.4
77.6
75.4
77.6
61072010
CA
Tulare
77.0
78.8
76.9
78.8
76.9
78.8
76.9
78.8
61090005
CA
Tuolumne
75.6
77.8
75.5
77.7
75.5
77.7
75.5
77.7
61112002
CA
Ventura
70.9
71.6
70.8
71.5
70.8
71.5
70.8
71.5
80350004
CO
Douglas
71.7
72.3
71.6
72.2
71.6
72.2
71.6
72.2
80590006
CO
Jefferson
72.6
73.3
72.5
73.1
72.5
73.1
72.5
73.1
80590011
CO
Jefferson
73.8
74.4
73.7
74.4
73.7
74.4
73.7
74.4
90010017
CT
Fairfield
73.0
73.7
72.9
73.6
72.9
73.6
72.9
73.6
90013007
CT
Fairfield
74.2
75.1
74.1
75.0
74.1
75.0
74.1
75.0
90019003
CT
Fairfield
76.1
76.4
76.0
76.2
76.0
76.2
76.0
76.2
90099002
CT
New
Haven
71.8
73.9
71.7
73.8
71.7
73.8
71.7
73.8
170310001
IL
Cook
69.6
73.4
69.5
73.4
69.5
73.4
69.5
73.4
170310032
IL
Cook
69.8
72.4
69.7
72.4
69.7
72.4
69.7
72.4
170310076
IL
Cook
69.3
72.1
69.2
72.1
69.2
72.1
69.2
72.1
170314201
IL
Cook
69.9
73.4
69.8
73.4
69.8
73.4
69.8
73.4
170317002
IL
Cook
70.1
73.0
70.0
73.0
70.0
73.0
70.0
73.0
320030075
NV
Clark
70.0
71.0
69.9
70.9
69.9
70.9
69.9
70.9
420170012
PA
Bucks
70.7
72.2
70.6
72.2
70.6
72.2
70.6
72.1
480391004
TX
Brazoria
70.1
72.3
70.0
72.1
70.0
72.1
70.0
72.1
481210034
TX
Denton
70.4
72.2
70.3
72.2
70.3
72.2
70.3
72.1
482010024
TX
Harris
75.2
76.8
75.1
76.6
75.1
76.6
75.1
76.6
482010055
TX
Harris
71.0
72.0
70.9
71.9
70.9
71.9
70.9
71.9
482011034
TX
Harris
70.3
71.6
70.2
71.4
70.2
71.4
70.2
71.4
482011035
TX
Harris
68.0
71.6
67.9
71.4
67.9
71.4
67.9
71.4
490110004
UT
Davis
72.9
75.1
72.8
75.0
72.8
75.0
72.8
75.0
490353006
UT
Salt Lake
73.6
75.3
73.5
75.1
73.5
75.1
73.5
75.1
490353013
UT
Salt Lake
74.4
74.9
74.3
74.8
74.3
74.8
74.3
74.8
490570002
UT
Weber
70.6
72.5
70.5
72.4
70.5
72.4
70.5
72.4
490571003
UT
Weber
70.5
71.5
70.4
71.3
70.4
71.3
70.4
71.3
550590019
WI
Kenosha
72.8
73.7
72.7
73.6
72.7
73.6
72.7
73.6
550590025
WI
Kenosha
69.2
72.3
69.1
72.2
69.1
72.2
69.1
72.2
551010020
WI
Racine
71.3
73.2
71.2
73.1
71.2
73.1
71.2
73.1
3B-15
-------
Table 3B-6. Projected average and maximum design values for the 2026 base case, the
proposed case, less stringent case, and more stringent case (ppb).
Site ID
State
County
2026
Avg
2026
Max
Proposed
Avg
Proposed
Max
Less
Stringent
Avg
Less
Stringent
Max
More
Stringent
Avg
More
Stringent
Max
40278011
AZ
Yuma
70.1
71.8
70.0
71.7
70.0
71.7
70.0
71.7
60090001
CA
Calaveras
70.2
71.1
70.0
70.9
70.0
70.9
70.0
70.9
60170010
CA
El Dorado
75.0
77.4
74.8
77.2
74.8
77.2
74.8
77.2
60170020
CA
El Dorado
73.2
75.0
73.0
74.8
73.0
74.8
73.0
74.8
60190007
CA
Fresno
79.5
81.3
79.3
81.1
79.3
81.1
79.3
81.1
60190011
CA
Fresno
81.9
82.8
81.7
82.6
81.7
82.6
81.7
82.6
60190242
CA
Fresno
78.7
80.3
78.5
80.1
78.5
80.1
78.5
80.1
60194001
CA
Fresno
81.8
83.3
81.6
83.1
81.6
83.1
81.6
83.1
60195001
CA
Fresno
82.7
85.4
82.5
85.2
82.5
85.2
82.5
85.2
60250005
CA
Imperial
76.2
76.5
76.1
76.4
76.1
76.4
76.1
76.4
60251003
CA
Imperial
75.3
75.3
75.2
75.2
75.2
75.2
75.2
75.2
60290007
CA
Kern
82.2
83.4
81.9
83.1
81.9
83.1
81.9
83.1
60290008
CA
Kern
78.6
80.5
78.3
80.2
78.3
80.2
78.3
80.2
60290011
CA
Kern
78.3
79.9
78.1
79.7
78.1
79.7
78.1
79.7
60290014
CA
Kern
80.7
82.6
80.5
82.4
80.5
82.4
80.4
82.3
60290232
CA
Kern
74.4
76.9
74.2
76.7
74.2
76.7
74.2
76.7
60292012
CA
Kern
83.4
84.1
83.2
83.9
83.2
83.9
83.2
83.9
60295002
CA
Kern
81.7
83.3
81.5
83.0
81.5
83.1
81.4
83.0
60311004
CA
Kings
76.0
76.6
75.8
76.4
75.8
76.4
75.8
76.4
60370002
CA
Los
Angeles
87.1
91.5
86.9
91.3
86.9
91.3
86.9
91.3
60370016
CA
Los
Angeles
92.4
95.2
92.2
94.9
92.2
95.0
92.1
94.9
60371201
CA
Los
Angeles
81.8
84.3
81.6
84.1
81.6
84.1
81.6
84.1
60371602
CA
Los
Angeles
73.0
73.3
72.9
73.2
72.9
73.2
72.9
73.2
60371701
CA
Los
Angeles
84.6
87.4
84.4
87.2
84.4
87.2
84.4
87.2
60372005
CA
Los
Angeles
79.9
81.1
79.7
80.9
79.7
80.9
79.7
80.9
60376012
CA
Los
Angeles
90.6
92.4
90.4
92.2
90.4
92.2
90.4
92.2
60379033
CA
Los
Angeles
79.8
81.4
79.6
81.2
79.6
81.2
79.6
81.2
3B-16
-------
Site ID
State
County
2026
Avg
2026
Max
Proposed
Avg
Proposed
Max
Less
Stringent
Avg
Less
Stringent
Max
More
Stringent
Avg
More
Stringent
Max
60390004
CA
Madera
75.0
77.5
74.8
77.3
74.8
77.3
74.8
77.3
60392010
CA
Madera
76.1
77.3
75.9
77.1
75.9
77.1
75.9
77.1
60430003
CA
Mariposa
74.0
76.9
73.9
76.8
73.9
76.8
73.9
76.8
60470003
CA
Merced
73.9
75.1
73.7
74.9
73.7
74.9
73.7
74.9
60570005
CA
Nevada
77.2
80.5
77.0
80.3
77.0
80.3
77.0
80.3
60592022
CA
Orange
71.8
72.1
71.6
71.9
71.6
71.9
71.6
71.9
60595001
CA
Orange
71.7
72.4
71.5
72.2
71.5
72.2
71.5
72.2
60610003
CA
Placer
75.9
78.6
75.7
78.4
75.7
78.4
75.7
78.4
60610004
CA
Placer
70.9
76.0
70.7
75.8
70.7
75.8
70.7
75.8
60610006
CA
Placer
71.7
72.6
71.5
72.4
71.5
72.4
71.5
72.4
60650008
CA
Riverside
70.4
72.7
70.2
72.5
70.3
72.6
70.2
72.5
60650012
CA
Riverside
84.9
87.3
84.6
87.0
84.7
87.1
84.6
87.0
60650016
CA
Riverside
71.1
72.0
70.9
71.8
70.9
71.8
70.9
71.8
60651016
CA
Riverside
88.8
89.9
88.5
89.6
88.6
89.7
88.5
89.6
60652002
CA
Riverside
75.7
77.8
75.5
77.6
75.6
77.7
75.5
77.6
60655001
CA
Riverside
79.6
81.7
79.4
81.5
79.4
81.5
79.4
81.5
60656001
CA
Riverside
82.5
83.1
82.3
82.9
82.3
82.9
82.3
82.9
60658001
CA
Riverside
88.6
89.7
88.3
89.4
88.4
89.5
88.3
89.4
60658005
CA
Riverside
87.0
89.7
86.8
89.5
86.8
89.5
86.8
89.5
60659001
CA
Riverside
79.9
82.0
79.7
81.8
79.7
81.8
79.7
81.8
60670012
CA
Sacramento
73.6
74.3
73.4
74.1
73.4
74.1
73.4
74.1
60710001
CA
San
Bernardino
74.0
74.9
73.8
74.7
73.9
74.8
73.8
74.7
60710005
CA
San
Bernardino
99.2
100.7
98.9
100.4
98.9
100.4
98.9
100.4
60710012
CA
San
Bernardino
86.4
89.2
86.2
89.0
86.2
89.0
86.2
89.0
60710306
CA
San
Bernardino
76.0
77.8
75.8
77.6
75.8
77.6
75.8
77.6
60711004
CA
San
Bernardino
96.1
99.1
95.8
98.8
95.8
98.8
95.8
98.8
60711234
CA
San
Bernardino
70.3
74.0
70.2
73.9
70.2
73.9
70.2
73.9
60712002
CA
San
Bernardino
89.2
90.4
88.9
90.1
89.0
90.2
88.9
90.1
60714001
CA
San
Bernardino
81.7
82.4
81.5
82.2
81.5
82.2
81.5
82.2
3B-17
-------
Site ID
State
County
2026
Avg
2026
Max
Proposed
Avg
Proposed
Max
Less
Stringent
Avg
Less
Stringent
Max
More
Stringent
Avg
More
Stringent
Max
60714003
CA
San
Bernardino
94.2
97.0
93.9
96.7
94.0
96.8
93.9
96.7
60719002
CA
San
Bernardino
79.3
80.9
79.1
80.7
79.2
80.8
79.1
80.7
60719004
CA
San
Bernardino
98.5
100.6
98.2
100.3
98.2
100.3
98.2
100.3
60731006
CA
San Diego
76.1
77.0
76.0
76.9
76.0
76.9
75.9
76.8
60773005
CA
San
Joaquin
70.8
72.4
70.6
72.2
70.6
72.2
70.6
72.2
60990005
CA
Stanislaus
74.7
75.6
74.5
75.4
74.5
75.4
74.5
75.4
60990006
CA
Stanislaus
76.7
77.0
76.5
76.8
76.5
76.8
76.5
76.8
61070006
CA
Tulare
78.2
79.4
78.1
79.3
78.1
79.3
78.0
79.2
61070009
CA
Tulare
81.6
81.6
81.5
81.5
81.5
81.5
81.4
81.4
61072002
CA
Tulare
74.3
76.4
74.2
76.3
74.2
76.3
74.1
76.2
61072010
CA
Tulare
75.9
77.7
75.7
77.5
75.7
77.5
75.7
77.5
61090005
CA
Tuolumne
75.0
77.1
74.8
76.9
74.8
76.9
74.8
76.9
80350004
CO
Douglas
70.5
71.1
70.3
70.9
70.4
71.0
70.3
70.9
80590006
CO
Jefferson
71.7
72.3
71.5
72.1
71.6
72.2
71.5
72.1
80590011
CO
Jefferson
72.6
73.3
72.4
73.1
72.5
73.2
72.4
73.1
90010017
CT
Fairfield
71.5
72.2
71.1
71.8
71.2
71.9
70.9
71.6
90013007
CT
Fairfield
72.8
73.7
72.2
73.1
72.3
73.2
72.0
72.9
90019003
CT
Fairfield
74.6
74.8
74.1
74.3
74.2
74.4
74.0
74.2
90099002
CT
New
Haven
70.4
72.4
69.8
71.8
69.9
71.9
69.7
71.7
170310001
IL
Cook
68.7
72.5
68.2
72.0
68.4
72.2
68.2
71.9
170310032
IL
Cook
69.1
71.7
68.8
71.4
68.9
71.5
68.8
71.4
170310076
IL
Cook
68.5
71.3
68.0
70.8
68.2
71.0
68.0
70.8
170314201
IL
Cook
68.9
72.4
68.5
71.9
68.6
72.1
68.4
71.9
170317002
IL
Cook
69.1
72.0
68.6
71.5
68.8
71.6
68.6
71.4
480391004
TX
Brazoria
69.1
71.2
67.7
69.7
68.1
70.2
67.5
69.6
482010024
TX
Harris
74.2
75.7
72.9
74.4
73.2
74.7
72.7
74.2
490110004
UT
Davis
71.7
73.9
71.4
73.5
71.5
73.7
71.3
73.5
490353006
UT
Salt Lake
72.5
74.1
72.2
73.7
72.3
73.9
72.2
73.7
490353013
UT
Salt Lake
73.5
74.0
73.1
73.6
73.2
73.7
73.1
73.6
490570002
UT
Weber
69.8
71.7
69.4
71.3
69.6
71.4
69.4
71.3
550590019
WI
Kenosha
71.7
72.6
71.1
72.0
71.3
72.2
71.1
72.0
550590025
WI
Kenosha
68.1
71.1
67.5
70.5
67.7
70.7
67.5
70.4
551010020
WI
Racine
70.2
72.1
69.6
71.5
69.8
71.7
69.6
71.5
3B-18
-------
3B.4 Projected Impacts on Downwind Contributions
As noted above, the method for projecting design values in 2023 and 2026 for each
alternative control case involved estimating the change in contribution from each state to each
receptor for each case. We compared the contributions from each alternative control case to the
corresponding contributions in the 2023 and 2026 base cases. For each of these years, we
evaluated the reductions in contributions for each linkage to identify the largest reduction from
each upwind state to a linked downwind receptor. In 2023, the largest reduction between the base
case contributions and each of the three alternative cases were 0.01 ppb or less. In 2026, impacts
on upwind state contributions are greater than in 2023, which is consistent with the overall larger
amount of NOx reduction in 2026 compared to 2023. In 2026 we found that 19 of the 24 linked
upwind states are projected to have their downwind contribution reduced by 0.01 ppb or more to
at least one receptor. In 12 of these 19 states, the largest reduction in downwind contribution is at
least 0.05 ppb. In half of these 12 states, the largest reduction in downwind contribution is 0.10
ppb or more. In Table 3B-7 we provide the largest impact on downwind contributions to a linked
receptor for those 19 upwind states that are projected to have a reduction in contribution of 0.01
ppb or more to at least one downwind receptor. A review of the impact on contributions of the
emissions reductions in the alternative control cases indicates that each state that is linked in the
2023 base case and in the 2026 base case is still linked to at least one downwind receptor in the
alternative control cases.
Table 3B-7. Largest reduction in downwind contribution for 19 upwind states for each
alternative control case
State
Proposed
vs Base
Less vs
Base
More vs
Base
AR
0.16
0.04
0.16
CA
0.03
0.02
0.03
IN
0.31
0.18
0.32
KY
0.10
0.06
0.11
LA
0.58
0.42
0.66
MI
0.08
0.06
0.08
MN
0.01
0.00
0.01
MO
0.10
0.07
0.10
MS
0.07
0.03
0.07
NY
0.03
0.02
0.07
OH
0.06
0.05
0.06
3B-19
-------
OK
0.02
0.02
0.02
PA
0.15
0.14
0.27
TX
0.04
0.03
0.05
UT
0.08
0.05
0.08
VA
0.03
0.03
0.03
WI
0.08
0.02
0.09
wv
0.02
0.01
0.02
WY
0.06
0.02
0.06
3B-20
-------
CHAPTER 4: COST, EMISSIONS, AND ENERGY IMPACTS
Overview
This chapter reports the compliance costs, emissions, and energy analyses performed for
the proposed Federal Implementation Plan (FIP) Addressing Regional Ozone Transport for the
2015 Ozone National Ambient Air Quality Standards (FIP for the 2015 ozone NAAQS). EPA
used the Integrated Planning Model (IPM)1 to conduct the electric generating units (EGU)
analysis discussed in this chapter and the Control Strategy Tool (CoST)2, the Control Measures
Database (CMDB)3, and the 2019 emissions inventory to conduct a screening assessment for
non-electric generating units (non-EGUs) for 2026. As explained in detail below, this chapter
presents analysis for three regulatory control alternatives that differ in the level of EGU nitrogen
oxides (NOx) ozone season emissions budgets in the 25 states subject to this action. These
regulatory control alternatives impose different budget levels. The different budget levels are
calculated assuming the application of different NOx mitigation technologies. The chapter also
presents three regulatory control alternatives for non-EGUs that differ in the number of sources
subject to emission limits.
The chapter is organized as follows: following a summary of the regulatory control
alternatives analyzed and a summary of EPA's methodologies, we present estimates of
compliance costs for EGUs, as well as estimated impacts on emissions, generation, capacity, fuel
use, fuel price, and retail electricity price. We then present a summary of the results of the non-
EGU screening assessment for 2026. Section 4.6 of this chapter describes the relationship
between the compliance cost estimates and social costs.
4.1 Regulatory Control Alternatives
The proposal establishes NOx emissions budgets requiring fossil fuel-fired power plants
(EGUs) in 25 states to participate in an allowance-based ozone season (May 1 through
September 30) trading program beginning in 2023. The EGUs covered by the proposed FIPs and
1 Information on IPM can be found at the following link: https://www.epa.gov/airmarkets/power-sector-modeling.
2 Further information on CoST can be found at the following link: https://www.epa.gov/economic-and-cost-analysis-
air-pollution-regulations/cost-analysis-modelstools-air-pollution.
3 The CMDB is available at the following link: https://www.epa.gov/economic-and-cost-analysis-air-pollution-
regulations/cost-analysis-modelstools-air-pollution.
4-1
-------
subject to the budget are fossil-fired EGUs with >25 megawatt (MW) capacity. For details on the
derivation of these budgets, please see Section VI.C. of the preamble.
The proposed FIP requirements establish ozone season NOx emissions budgets for EGUs
in 25 states starting in 2023 (Alabama, Arkansas, Delaware, Illinois, Indiana, Kentucky,
Louisiana, Maryland, Michigan, Minnesota, Mississippi, Missouri, Nevada, New Jersey, New
York, Ohio, Oklahoma, Pennsylvania, Tennessee, Texas, Utah, Virginia, West Virginia,
Wisconsin, and Wyoming) and require EGUs in these states to participate in a revised version of
the Cross-State Air Pollution Rule (CSAPR) NOx Ozone Season Group 3 Trading Program that
was previously established in the Revised CSAPR Update.4 EPA is proposing to amend existing
FIPs for 12 states currently participating in the CSAPR NOx Ozone Season Group 3 Trading
Program (Illinois, Indiana, Kentucky, Louisiana, Maryland, Michigan, New Jersey, New York,
Ohio, Pennsylvania, Virginia, and West Virginia) to replace their existing emissions budgets
established in the Revised CSAPR Update with new emissions budgets. For eight states currently
covered by the CSAPR NOx Ozone Season Group 2 Trading Program under State
Implementation Plans (SIPs) or FIPs, EPA is proposing to issue new FIPs for two states
(Alabama and Missouri) and amend existing FIPs for six states (Arkansas, Mississippi,
Oklahoma, Tennessee, Texas, and Wisconsin) to transition EGU sources in these states to the
revised Group 3 Trading Program beginning with the 2023 ozone season. EPA proposes to issue
new FIPs for five states not currently covered by any CSAPR NOx ozone season trading
program: Delaware, Minnesota, Nevada, Utah, and Wyoming. In 2026 the seasonal NOx
emissions budgets are reduced further, in particular for 23 of these states (Arkansas, California,
Illinois, Indiana, Kentucky, Louisiana, Maryland, Michigan, Minnesota, Mississippi, Missouri,
Nevada, New Jersey, New York, Ohio, Oklahoma, Pennsylvania, Texas, Utah, Virginia, West
Virginia, Wisconsin, and Wyoming). In addition, beginning in the 2027 ozone season, coal
facilities greater than 100 MW lacking SCR controls and certain oil/gas steam facilities greater
than 100 MW that lack existing SCR controls located in these 23 states must meet daily emission
rate limits, effectively forcing affected units to install new SCR controls, find other means of
compliance, or retire. States that do not have additional control measures assumed in 2026
4 As explained in Section VI.C.l of the preamble, EPA proposes finding that EGU sources within the State of
California are sufficiently controlled such that no further emission reductions are needed from them to eliminate
significant contribution to downwind states.
4-2
-------
continue to remain part of the revised group 3 Trading Program.
In the proposal, we introduce additional features to the allowance-based trading program
approach for EGUs, including dynamic adjustments of the emissions budgets over time and
backstop daily emission rate limits for most coal-fired units, that were not included in previous
CSAPRNOx ozone season trading programs. These enhancements will help maintain control
stringency over time and improve emissions performance at individual units, offering an extra
measure of assurance that existing pollution controls will be operated during the ozone season.
This analysis incorporates the daily emission rate requirement for units with existing controls by
forcing operation of these controls in the ozone season for affected sources. For affected
uncontrolled units in the 23 state group, starting in 2026, this analysis imposes an emission rate
constraint that forces affected units to either install new SCR retrofits, find other means of
compliance, or retire.5 The analysis does not explicitly capture the dynamic budget adjustments
over time, but the forced operation of controls during the ozone season over the forecast period
(even in the absence of binding mass limits) approximates this feature of the program design. For
details of the controls modeled for each of the regulatory control alternatives please see Table
4-2 below.
The proposal also includes NOx emissions limitations with an initial compliance date of
2026 applicable to certain non-EGU stationary sources in 23 states: Arkansas, California,
Illinois, Indiana, Kentucky, Louisiana, Maryland, Michigan, Minnesota, Mississippi, Missouri,
Nevada, New Jersey, New York, Ohio, Oklahoma, Pennsylvania, Texas, Utah, Virginia, West
Virginia, Wisconsin, and Wyoming. The proposed rule establishes NOx emissions limitations
during the ozone season for the unit types listed in Table 4-1 below. A more detailed summary of
the proposed emissions limits can be found in Section I.B. of the preamble.
5 The proposed rule assumes SCR retrofit potential starting in 2026 and it is reflected in the 2026 state emission
budgets. The daily backstop emission rate does not apply until 2027, but the majority of units retrofitting are
anticipated to do so by 2026 to assist with the 2026 state emissions budget compliance. EPA's IPM model run years
are 2026 and 2028. The SCR compliance behavior is generally expected to occur no later than 2027, and in 2026 in
many cases. Therefore, EPA models this daily backstop emission rate in 2026 (when choosing between model run
year 2026 and 2028) to conservatively reflect compliance cost in the first year in which the technology is in place
for some units.
4-3
-------
Table 4-1. Non-EGU Emissions Unit Types, Emissions Limits, and Industries
Emissions Unit Type
Emissions Limit
Industry
NAICS
Reciprocating internal
combustion engines
g/hp-hr
Pipeline Transportation of Natural Gas
4862
Kilns
lb/ton of clinker
Cement and Concrete Product
Manufacturing
3273
Boilers and furnaces
Depending on equipment type -
lb/mmBtu, lb/ton of steel, lb/ton,
lb/ton coal pushed, lb/ton coal
charged, work practice standards
Iron and Steel Mills and Ferroalloy
Manufacturing
3311
Furnaces
lb/ton glass produced
Glass and Glass Product Manufacturing
3272
Impactful boilers*
lbs NOx/mmBtu
Basic Chemical Manufacturing,
Petroleum and Coal Products
Manufacturing, and
Pulp, Paper, and Paperboard Mills
3251,
3241,
3221
North American Industry Classification System (NAICS)
impactful boilers are boilers with design capacity of 100 mmBtu/hr or greater.
This regulatory impact analysis (RIA) evaluates the benefits, costs and certain impacts of
compliance with three regulatory control alternatives: the proposed FIP for the 2015 ozone
NAAQS, a less-stringent alternative, and a more-stringent alternative. Table 4-2 below presents
the less stringent alternatives, proposed rule requirements, and more stringent alternatives for
EGUs and non-EGUs. For the purposes of summarizing the results of the benefits and costs of
these alternatives, the less stringent alternative for EGUs is presented with the less stringent
alternative for non-EGUs. However, the cost, emissions, and energy impacts for the EGU and
non-EGU alternatives are evaluated separately.
Table 4-2. Regulatory Control Alternatives for EGUs and Non-EGUs
Regulatory Control NOx Controls Implemented for EGUs within IPM
Alternative
1) 2023 onwards: Shift generation to minimize costs
2) 2023 onwards: Fully operate existing SCRs during ozone season
3) 2023 onwards: Fully operate existing SNCRs during ozone season
4) In 2023 install state-of-the-art combustion controls
5) In 2028 model run year, impose backstop emission rate limits on coal units
Less Stringent Alternative greater than 100 MW within the 23-state region that lack SCR controls,
forcing units to retrofit or retire.
6) In 2028 model run year, impose backstop emission rate limits on oil/gas
steam units greater than 100 MW that operated at a greater than 20% capacity
factor historically within the 23-state region that lack SCR controls, forcing
units to retrofit or retire.6
Proposed Rule (All Controls above and)
6 The 20% capacity factor cutoff applied is representative of the fleet of O/G steam units assumed to have SCR
retrofit potential in its state budgets. In the proposal, EPA defined this segment using 150 tons per season cutoff,
which provides a similar size of the O/G steam fleet as the 20% capacity factor value used in this analysis.
4-4
-------
Regulatory Control
Alternative
NOx Controls Implemented for EGUs within IPM
7) In 2026, impose backstop emission rate limits on coal units greater than 100
MW within the 23-state region that lack SCR controls, forcing units to retrofit
or retire.
8) In 2026, impose backstop emission rate limits on oil/gas steam units greater
than 100 MW that operated at a greater than 20% capacity factor historically
within the 23-state region that lack SCR controls, forcing units to retrofit or
retire.
More Stringent Alternative
9)
(Controls 1 - 4, 7 and 8 above and)
In 2026, impose backstop emission rate limits on all oil/gas steam units
greater than 100 MW within the 23-state region that lack SCR controls,
forcing units to retrofit or retire.
NOx Emissions Limits for Non-EGUs - Emissions Unit Types and Industries
1)
Reciprocating internal combustion engines in Pipeline Transportation of
Natural Gas,
Less Stringent Alternative
2)
3)
4)
Kilns in Cement and Cement Product Manufacturing,
Boilers and furnaces in Iron and Steel Mills and Ferroalloy Manufacturing,
Furnaces in Glass and Glass Product Manufacturing, and
Proposed Rule
5)
(All emissions unit types and industries above and)
Impactful boilers* in Basic Chemical Manufacturing, Petroleum and Coal
Products Manufacturing, and Pulp, Paper, and Paperboard Mills.
More Stringent Alternative
6)
(All emissions unit types and industries above and)
All boilers in Basic Chemical Manufacturing, Petroleum and Coal Products
Manufacturing, and Pulp, Paper, and Paperboard Mills.
impactful boilers are boilers with design capacity of 100 mmBtu/hr or greater.
4.1.1 EGURegulatory Control Alternatives Analyzed
The illustrative emission budgets in this RIA represent EGU NOx ozone season emission
budgets for each state in 2023 and in 2026.7 This RIA analyzes the proposed FIP for the 2015
ozone NAAQS emission budgets, as well as a more and a less stringent alternative to the
proposed FIP for the 2015 ozone NAAQS. The more and less stringent alternatives differ from
the proposed rule in that they set different NOx ozone season emission budgets for the affected
EGUs and different dates for compliance with backstop emission rate limits. All three scenarios
use emission budgets that were developed using uniform control stringency represented by
$1,800 per ton of NOx (2016$) in 2023 and $11,000 per ton of NOx (2016$) in 2026. The less-
stringent alternative imposes backstop emission rate limits in the 2028 run year8 (reflective of
7 The budget setting process is described in section VII.B. of the preamble and in detail in the Ozone Transport
Policy Analysis Proposed Rule Technical Support Document (TSD).
8 IPM uses model years to represent the full planning horizon being modeled. By mapping multiple calendar years to
a run year, the model size is kept manageable. For this analysis, IPM maps the calendar year 2023 to run year 2023,
calendar years 2024-2026 to run year 2025 and calendar years 2027-2029 to run year 2028. For model details, please
see Chapter 2 of the IPM documentation, available at:
4-5
-------
imposition in the 2027 calendar year), while the proposed rule and more stringent alternative
impose backstop emission rate limits in the 2025 run year (reflective of imposition in the 2026
calendar year) that force uncontrolled units to either install controls or retire. The backstop
emission rate limits are imposed on all coal units within the 23 state region that are greater than
100 MW and lack SCR controls (excepting circulating fluidized bed (CFB) units). The emission
rate limits are also imposed on all oil/gas steam units within the linked states that are greater than
100 MW and lack SCR controls that operated at a greater than 20 percent historical capacity
factor. In addition to the backstop rate limits present in the proposed rule and the less stringent
alternative, the more stringent alternative also imposes backstop emission rate limits on all
oil/gas steam units in the affected states that are greater than 100 MW, lack SCR controls, and
have operated at below a 20 percent capacity factor historically.
All three alternatives are illustrative in nature in part because the budgets included in the
proposed FIP for the 2015 ozone NAAQS alternative differ slightly from the budgets imposed in
the modeling of these RIA alternatives. Furthermore, the proposed alternative analyzed herein
assumes oil/gas steam units are subject to backstop emission rate limits, whereas the proposed
rule does not impose those limits. That is because subsequent to completion of the analysis of
these three alternatives, EPA made updates to budgets in the proposal itself. In particular, the
budgets proposed in the rule account for emission reductions commensurate with the installation
of SCR at oil/gas steam units greater than 100 MW without an SCR and a three-year (2019-21)
average of ozone season emissions of at least 150 tons beginning in 2026. The proposed rule
scenario assumes emission reductions commensurate with installation of SCRs at oil/gas steam
units greater than 100 MW without an SCR and a three-year average capacity factor of greater
than 20% beginning in 2026. Additionally, backstop emission rates are not applicable to this
oil/gas steam capacity in the proposed rule, while the proposed rule scenario assumes these units
are subject to backstop emission rate limits. In the proposed rule, for the 12 Revised CSAPR
Update states the 2023 budgets assume that state-of-the-art combustion controls are installed in
2023, while combustion controls in the remaining 13 states are not assumed to be installed until
2024. Under the modeling for the proposed rule, the illustrative budget assumes that all 25 states
https://www.epa.gOv/system/files/documents/2021-09/epa-platform-v6-summer-2021-reference-case-09-ll-21-
v6.pdf
4-6
-------
install combustion controls in 2023. Installation of state-of-the-art combustion controls is an
exogenous input to the model, and state-of-the-art combustion control installations are imposed
in all 25 states in 2023. Finally, the Engineering Analysis used to develop the illustrative budgets
relied on 2019 historical data, while the Engineering Analysis used to develop the proposed
budgets relied on 2021 historical data.
EPA finds that the three illustrative regulatory control alternatives presented in this RIA
provide a reasonable approximation of the impacts of the proposed rule, as well as an evaluation
of the relative impacts of two regulatory alternatives. This finding is supported by a side analysis
of the costs and impacts (but not the benefits) of the emission budgets included in the proposed
FIP for the 2015 Ozone NAAQS, which is provided in the docket for this proposed rulemaking.
Table 4-3. reports the illustrative EGUNOx ozone season emission budgets that are
evaluated in this RIA for the 2023 and 2025 IPM run years. As described above, starting in 2023,
emissions from affected EGUs in the 25 states cannot exceed the sum of emissions budgets but
for the ability to use banked allowances from previous years for compliance. For individual
states, emissions cannot exceed 121% of the state emission budget (the assurance levels). In
these RIA scenarios, no further reductions in budgets occur after 2026, and budgets remain in
place for future years. These budgets are imposed in addition to the control measures outlined in
Table 4-2.
Table 4-3. Illustrative NOx Ozone Season Emission Budgets (Tons) Evaluated by IPM Run
Year
Region
Proposed Rule
2023 2025
Less Stringent
Alternative
2023 2025
More Stringent
Alternative
2023 2025
Alabama
7,444 7,444
7,444 7,444
7,445 7,445
Arkansas
8,848 4,019
8,848 4,019
8,848 3,837
Delaware
333 333
333 333
204 204
Illinois
6,985 5,396
6,985 5,396
6,984 5,354
Indiana
11,315 7,798
11,315 7,798
11,315 7,797
Kentucky
11,410 6,897
11,410 6,897
11,410 6,821
Louisiana
13,698 4,988
13,698 4,988
13,698 4,255
Maryland
1,245 1,325
1,245 1,325
1,245 1,226
Michigan
10,896 7,779
10,896 7,779
10,897 7,732
Minnesota
4,072 2,371
4,072 2,371
4,393 2,614
Mississippi
6,431 2,709
6,431 2,709
6,431 2,048
Missouri
10,211 7,467
10,211 7,467
10,211 7,467
Nevada
2,392 1,028
2,392 1,028
940 745
4-7
-------
Region
New Jersey
Proposed Rule
2023 2025
1,099 1,099
Less Stringent
Alternative
2023 2025
1,099 1,099
More Stringent
Alternative
2023 2025
1,098 1,098
New York
3,283 2,762
3,283 2,762
3,283 2,331
Ohio
8,612 8,437
8,612 8,437
8,612 8,395
Oklahoma
8,765 4,229
8,765 4,229
8,764 3,782
Pennsylvania
8,340 8,008
8,340 8,008
8,340 7,274
Tennessee
4,394 4,394
4,394 4,394
4,393 4,393
Texas
41,169 23,898
41,169 23,898
41,169 22,188
Utah
9,526 1,760
9,526 1,760
9,360 1,610
Ute
2,144 409
2,144 409
2,144 409
Virginia
3,856 3,172
3,856 3,172
3,856 2,955
West Virginia
12,015 9,125
12,015 9,125
12,015 9,125
Wisconsin
4,892 2,752
4,892 2,752
4,892 2,733
Wyoming
8,684 4,215
8,684 4,215
8,639 4,158
Aggregated State
Emission Budgets
212,059 133,814
212,059 133,814
210,584 127,995
Note that EGUs have flexibility in determining how they will comply with the allowance
trading program. As discussed below, the way that they comply may differ from the methods
forecast in the modeling for this RIA. See Section 4.3 for further discussion of the modeling
approach used in the analysis presented below.
4.1.2 Non-EGU Regulatory Control Alternatives Analyzed
As discussed in Section I.B. of the preamble and Sections 4.4 and 4.5 below, EPA
developed an analytical framework and using that framework prepared a screening assessment
for 2026 to estimate emissions reduction potential from impactful non-EGU industries and non-
EGU emissions units. Impactful industries are those that have large, meaningful air quality
impacts downwind from potentially controllable NOx emissions. For additional discussion of
impactful industries, see the memorandum titled Screening Assessment of Potential Emissions
Reductions, Air Quality Impacts, and Costs from Non-EGU Emissions Units for 2026 (non-EGU
screening assessment). Impactful emissions units are emissions units with greater than 100 tons
per year (tpy) of NOx emissions.
First, EPA developed an analytical framework using data from 2023. In the analytical
framework, EPA identified potential NOx emissions reductions for non-EGU sources that would
result in meaningful air quality improvements in downwind areas. EPA incorporated air quality
modeling information, annual emissions, and available information about potential controls to
4-8
-------
determine which industries, if subject to further control requirements, would have the greatest
impact in providing air quality improvements at the downwind receptors. This evaluation in the
analytical framework was subject to a marginal cost threshold of up to $7,500 per ton, which
EPA determined based on information available to the Agency about existing control device
efficiency and cost information. In the framework, EPA identified emissions unit types in seven
industries that provide opportunities for NOx emissions reductions and resulting impacts on air
quality at the downwind receptors. Because EPA determined that 2026 was the potential earliest
date by which controls on non-EGU emissions units could be installed, EPA used the analytical
framework with air quality modeling information for 2026 to prepare a screening assessment for
2026. Results of the screening assessment for 2026 are discussed in Section 4.5. EPA did not
estimate emissions reductions of SO2, PM2.5, CO2 and other pollutants that may be associated
with controls on non-EGU emissions units.
As described in Section 4.1, the proposed rule imposes emissions limits on each of the
emissions unit types identified in Table 4-1. For non-EGUs, the less stringent alternative
assumes there are emissions limits for all emission units from the proposed rule alternative
except for impactful boilers in Basic Chemical Manufacturing, Petroleum and Coal Products
Manufacturing, and Pulp, Paper, and Paperboard Mills. The more stringent alternative assumes
emissions limits for all emission units from the proposed rule alternative and all boilers, not just
impactful boilers, in Basic Chemical Manufacturing, Petroleum and Coal Products
Manufacturing, and Pulp, Paper, and Paperboard Mills.
After the non-EGU emissions units are identified in the non-EGU screening assessment,
the proposed rule includes a separate evaluation of the emissions limits that are to be applied to
each of the non-EGU emissions unit types. The emissions limits are not based on the assumed
emission reductions modeled for each emissions unit in the non-EGU screening assessment.
Rather, for each emissions unit type, EPA considered the range of emissions limits that currently
apply to these sources under other Clean Air Act programs, such as reasonably available control
technology (RACT), new source performance standards (NSPS), national emissions standards
for hazardous air pollutants (NESHAP), and Ozone Transport Commission (OTC) model rules,
to develop an emissions limit that should be achievable by all sources after installing the controls
identified in the non-EGU screening assessment.
4-9
-------
Table 4-4 below provides a summary of the 2019 ozone season emissions for non-EGUs
for the 23 states subject to the proposed FIP in 2026, along with the estimated ozone season
reductions for the proposal and the less and more stringent alternatives. The estimated emissions
reductions by state for the proposed alternative are from the non-EGU screening assessment, and
the estimated reductions by state for the less and more stringent alternatives were estimated for
the RIA using the same methodology.
Table 4-4. Ozone Season (OS) NOx Emissions and Emissions Reductions for the Proposed
Rule and the Less and More Stringent Alternatives*
State
2019 OS NOx
Emissions
Proposed Rule - OS
NOx Reductions
Less Stringent
Alternative - OS NOx
Reductions
More Stringent
Alternative - OS NOx
Reductions
AR
8,265
1,654
922
1,654
CA
14,579
1,666
1,598
1,777
IL
16,870
2,452
2,452
2,553
IN
19,604
3,175
2,787
3,175
KY
11,934
2,291
2,291
2,291
LA
35,831
6,769
4,121
6,955
MD
2,365
45
45
45
MI
18,996
2,731
2,731
3,093
MN
17,591
673
673
789
MO
9,109
3,103
3,103
3,103
MS
12,284
1,761
1,577
1,761
NJ
2,025
0
0
29
NV
2,418
0
0
0
NY
6,003
500
389
613
OH
19,729
2,790
2,611
2,814
OK
22,146
3,575
3,575
3,871
PA
15,861
3,284
3,132
3,340
TX
47,135
4,440
4,440
6,596
UT
6,276
757
757
757
VA
7,041
1,563
1,465
1,660
WI
6,571
2,150
677
2,234
wv
9,825
982
982
982
WY
10,335
826
826
826
Totals
322,793
47,186
41,153
50,918
* In the non-EGU screening assessment for 2026, EPA estimated emissions reduction potential from the non-EGU
industries and emissions units. In the screening assessment, EPA used CoST to identify emissions units, emissions
reductions, and associated compliance costs to evaluate the effects of potential non-EGU emissions control measures
and technologies. CoST is designed to be used for illustrative control strategy analyses (e.g., NAAQS regulatory
impact analyses) and not for unit-specific, detailed engineering analyses. The estimates from CoST identify proxies
for (1) non-EGU emissions units that have emissions reduction potential, (2) potential controls for and emissions
4-10
-------
reductions from these emissions units, and (3) control costs from the potential controls on these emissions units. The
control cost estimates do not include monitoring, recordkeeping, reporting, or testing costs. This screening
assessment is not intended to be, nor take the place of, a unit-specific detailed engineering analysis that fully
evaluates the feasibility of retrofits for the emissions units, potential controls, and related costs. For more
information on CoST, go to the following link: https://www.epa.gov/economic-and-cost-analysis-air-pollution-
regulations/cost-analysis-modelstools-air-pollution.
4.2 Power Sector Modeling Framework
IPM is a state-of-the-art, peer-reviewed, dynamic linear programming model that can be
used to project power sector behavior under future business-as-usual conditions and to examine
prospective air pollution control policies throughout the contiguous United States for the entire
electric power system. EPA used IPM to project likely future electricity market conditions with
and without the proposed FIP for the 2015 ozone NAAQS.
IPM, developed by ICF, is a multi-regional, dynamic, deterministic linear programming
model of the contiguous U.S. electric power sector. It provides estimates of least cost capacity
expansion, electricity dispatch, and emissions control strategies while meeting energy demand
and environmental, transmission, dispatch, and reliability constraints. Due to lack of lead time,
EPA does not allow IPM to build certain new capital investments such as new, unplanned natural
gas or renewable capacity or new SCR or SNCR through the 2023 run year in response to the
state emission budgets (i.e., retrofits, retirements or builds additional to those selected in the
baseline are not allowed in 2023). The compliance analysis of the proposed rule and alternatives
allows for new combustion controls in the 2023 analysis year.
EPA has used IPM for almost three decades to better understand power sector behavior
under future business-as-usual conditions and to evaluate the economic and emissions impacts of
prospective environmental policies. The model is designed to reflect electricity markets as
accurately as possible. EPA uses the best available information from utilities, industry experts,
gas and coal market experts, financial institutions, and government statistics as the basis for the
detailed power sector modeling in IPM. The model documentation provides additional
4-11
-------
information on the assumptions discussed here as well as all other model assumptions and
inputs.9
The model incorporates a detailed representation of the fossil-fuel supply system that is
used to estimate equilibrium fuel prices. The model uses natural gas fuel supply curves and
regional gas delivery costs (basis differentials) to simulate the fuel price associated with a given
level of gas consumption within the system. These inputs are derived using ICF's Gas Market
Model (GMM), a supply/demand equilibrium model of the North American gas market.10
IPM also endogenously models the partial equilibrium of coal supply and EGU coal
demand levels throughout the contiguous U.S., taking into account assumed non-power sector
demand and imports/exports. IPM reflects 36 coal supply regions, 14 coal grades, and the coal
transport network, which consists of over four thousand linkages representing rail, barge, and
truck and conveyer linkages. The coal supply curves in IPM were developed during a thorough
bottom-up, mine-by-mine approach that depicts the coal choices and associated supply costs that
power plants would face if selecting that coal over the modeling time horizon. The IPM
documentation outlines the methods and data used to quantify the economically recoverable coal
reserves, characterize their cost, and build the 36 coal regions' supply curves.11
To estimate the annualized costs of additional capital investments in the power sector, EPA
uses a conventional and widely accepted approach that applies a capital recovery factor (CRF)
multiplier to capital investments and adds that to the annual incremental operating expenses. The
CRF is derived from estimates of the power sector's cost of capital (i.e., private discount rate),
the amount of insurance coverage required, local property taxes, and the life of capital.12 It is
9 Detailed information and documentation of EPA's Baseline run using IPM (v6), including all the underlying
assumptions, data sources, and architecture parameters can be found on EPA's website at:
https://www.epa.gov/airmarkets/documentation-epas-power-sector-modeling-platform-v6-summer-2021-reference-
case.
10 See Chapter 8 of EPA's Baseline run using IPM v6 documentation, available at:
https://www.epa.gov/airmarkets/documentation-epas-power-sector-modeling-platform-v6-summer-2021-reference-
case.
11 See Chapter 7 of the IPM v6 documentation. The documentation for EPA's power sector modeling platform v6 -
summer 2021 reference case consists of a comprehensive document for the Summer 2021 release of IPM v. 6.20 and
is available at: https://www.epa.gov/airmarkets/documentation-epas-power-sector-modeling-platform-v6-summer-
2021 -reference-case.
12 See Chapter 10 of the documentation for EPA's power sector modeling platform v6 - summer 2021 reference case,
available at: https://www.epa.gov/airmarkets/documentation-epas-power-sector-modeling-platform-v6-summer-
2021 -reference-case
4-12
-------
important to note that there is no single CRF factor applied in the model; rather, the CRF varies
across technologies, book life of the capital investments, and regions in the model in order to
better simulate power sector decision-making.
EPA has used IPM extensively over the past three decades to analyze options for reducing
power sector emissions. Previously, the model has been used to estimate the costs, emission
changes, and power sector impacts for the Clean Air Interstate Rule (U.S. EPA, 2005), the Cross-
State Air Pollution Rule (U.S. EPA, 2011), the Mercury and Air Toxics Standards (U.S. EPA,
201 la), the Clean Power Plan for Existing Power Plants (U.S. EPA, 2015), the Carbon Pollution
Standards for New Power Plants (U.S. EPA, 2015), the Cross-State Air Pollution Update Rule
(U.S. EPA 2016), the Affordable Clean Energy Rule (U.S. EPA, 2019), the Clean Power Plan
Repeal (U.S. EPA, 2019), and the Revised Cross-State Air Pollution Update Rule (U.S. EPA
2021). EPA has also used IPM to estimate the air pollution reductions and power sector impacts
of water and waste regulations affecting EGUs, including Cooling Water Intakes (316(b)) Rule
(U.S. EPA, 2014), Disposal of Coal Combustion Residuals from Electric Utilities (U.S. EPA,
2015b), Steam Electric Effluent Limitation Guidelines (ELG) (U.S. EPA, 2015c), and Steam
Electric Reconsideration Rule (U.S. EPA 2020).
The model and EPA's input assumptions undergo periodic formal peer review. The
rulemaking process also provides opportunity for expert review and comment by a variety of
stakeholders, including owners and operators of capacity in the electricity sector that is
represented by the model, public interest groups, and other developers of U.S. electricity sector
models. The feedback that the Agency receives provides a highly detailed review of key input
assumptions, model representation, and modeling results. IPM has received extensive review by
energy and environmental modeling experts in a variety of contexts. For example, in October
2014 U.S. EPA commissioned a peer review13 of EPA Baseline run version 5.13 using the
Integrated Planning Model. Additionally, and in the late 1990s, the Science Advisory Board
reviewed IPM as part of the CAA Amendments Section 812 prospective studies14 that are
periodically conducted. The Agency has also used the model in a number of comparative
13 See Response and Peer Review Report EPA Baseline run Version 5.13 Using IPM, available at:
https://www.epa.gov/airmarkets/response-and-peer-review-report-epa-base-case-version-513-using-ipm.
14 http://www2.epa.gov/clean-air-act-overview/benefits-and-costs-clean-air-act
4-13
-------
modeling exercises sponsored by Stanford University's Energy Modeling Forum over the past 20
years. IPM has also been employed by states (e.g., for the Regional Greenhouse Gas Initiative,
the Western Regional Air Partnership, Ozone Transport Assessment Group), other Federal and
state agencies, environmental groups, and industry.
4.3 EPA's Power Sector Modeling of the Baseline run and Three Regulatory Control
Alternatives
The IPM "baseline run" for any regulatory impact analysis is a business-as-usual scenario
that represents expected behavior in the electricity sector under market and regulatory conditions
in the absence of a regulatory action. As such, an IPM baseline run represents an element of the
baseline for this RIA.15 EPA frequently updates the IPM baseline run to reflect the latest
available electricity demand forecasts from the U.S. Energy Information Administration (EIA) as
well as expected costs and availability of new and existing generating resources, fuels, emission
control technologies, and regulatory requirements.
4.3.1 EPA 's IPM Baseline run v. 6.20
For our analysis of the proposed FIP for the 2015 ozone NAAQS, EPA used the Summer
2021 release of IPM version 6.20 to provide power sector emissions data for air quality
modeling, as well as a companion updated database of EGU units (the National Electricity
Energy Data System, or NEEDS, v.6.20 Summer 2021 Reference Case16) that is used in EPA's
modeling applications of IPM. The IPM Baseline run includes the CSAPR, CSAPR Update, and
the Revised CSAPR Update, as well as the Mercury and Air Toxics Standards. The Baseline run
also includes the 2015 Effluent Limitation Guidelines (ELG) and the 2015 Coal Combustion
Residuals (CCR), and the recently finalized 2020 ELG and CCR rules.17 The impacts of the
Later Model Year Light-Duty Vehicle GHG Emissions Standards are not captured in the
baseline, nor is the impact of the Proposed Standards of Performance for New, Reconstructed,
15 As described in Chapter 5 of EPA's Guidelines for Preparing Economic Analyses, the baseline "should
incorporate assumptions about exogenous changes in the economy that may affect relevant benefits and costs (e.g.,
changes in demographics, economic activity, consumer preferences, and technology), industry compliance rates,
other regulations promulgated by EPA or other government entities, and behavioral responses to the proposed rule
by firms and the public." (USEPA, 2010).
16 https://www.epa.gov/airmarkets/national-electric-energy-data-system-needs-v6.20
17 For a full list of modeled policy parameters, please see:
https://www.epa.gov/airmarkets/documentation-epas-power-sector-modeling-platform-v6-summer-2021-reference-
case
4-14
-------
and Modified Sources and Emissions Guidelines for Existing Sources: Oil and Natural Gas
Sector Climate Review.18 The analysis of power sector cost and impacts presented in this chapter
is based on a single IPM baseline run, and represents incremental impacts projected solely as a
result of compliance with the emissions budgets presented in Table 4-3 above.
4.3.2 Methodology for Evaluating the Regulatory Control Alternatives
To estimate the costs, benefits, and economic and energy market impacts of the proposed
FIP for the 2015 ozone NAAQS, EPA conducted quantitative analysis of the three regulatory
control alternatives: the proposed FIP for the 2015 ozone NAAQS emission budgets and a more
and a less stringent alternative. Details about these regulatory control alternatives, including
state-specific EGUNOx ozone-season emissions budgets for each alternative as analyzed in this
RIA, are provided above in Section 4.1.
Before undertaking power sector analysis to evaluate compliance with the regulatory
control alternatives, EPA first considered available EGU NOx mitigation strategies that could be
implemented for the 2023 ozone season. EPA considered all widely-used EGU NOx control
strategies: optimizing19 NOx removal by existing operational selective catalytic reduction
(SCRs) and turning on and optimizing existing idled SCRs; optimizing existing idled selective
non-catalytic reduction (SNCRs); installation of (or upgrading to) state-of-the-art NOx
combustion controls; shifting generation to units with lower NOx emission rates; and installing
new SCRs and SNCRs. EPA determined that affected EGUs within the 25 states could
implement all of these NOx mitigation strategies, except installation of new SCRs or SNCRs and
state-of-the-art combustion controls, for the 2023 ozone season.20 After assessing the available
NOx mitigation methods, this RIA projects the system-wide least-cost strategies for complying
with the annual budgets and backstop emission rate limits. Least-cost compliance may lead to the
application of different control strategies at a given source compared to the particular control
18 Available at: https://www.federalregister.gov/documents/2021/ll/15/2021-24202/standards-of-performance-for-
new-reconstructed-and-modified-sources-and-emissions-guidelines-for
19 Optimization of controls refers to the process of fully operating controls in order to meet the "widely achievable
emission rate" as outlined in the EGU NOx Mitigation Strategies Proposed Rule TSD.
20 The analysis assumes that SNCR and SCR optimization is available starting in 2023 and is adopted by all units
that do not currently optimize these controls. This compliance choice is an exogenous input into IPM.
4-15
-------
measure assumed for that source in the analysis used to calculate the budgets, which is in
keeping with the cost-saving compliance flexibility afforded by this allowance trading program.
Within IPM, units are assigned NOx emission rates based on historical data. To account for
changes in emission rates based on the seasonal operation of controls, each unit is assigned four
modes of operation. When the model is run, IPM selects the appropriate mode for each season
based on historical data (i.e., how the unit operated in the past), whether or not the unit is subject
to any seasonal or annual NOx reduction requirements, and whether the unit installs any
additional controls.21 The proposal's emission control requirements for EGUs only apply during
the program's ozone season (May 1 through September 30). Historically, some EGUs have
either reduced performance or idled their SCRs during the ozone season to reduce costs of
catalyst and ammonia injection in the SCR. This behavior has been observed more frequently
during periods when the prevailing allowance price has fallen to very low levels that do not
provide an adequate economic incentive to operate the SCR. We would not expect this behavior
to occur going forward with the dynamic budgets and backstop emission rate requirements
proposed in this policy.
Many of these mitigation strategies are captured within IPM. However, due to limitations
on model size, IPMv.6.20 does not have the ability to endogenously determine whether or not to
operate existing EGU post-combustion NOx controls (i.e., SCR or SNCR), optimize existing
SCRs and SNCRs, and install combustion controls in response to a regulatory emissions
requirement.22 The treatment of these controls in the analyses are described in turn. The
operating status of existing post-combustion NOx controls at a particular EGU in a model
scenario is determined by the model user. In order to evaluate compliance with the regulatory
alternatives, EPA determined outside of IPM the operation of existing controls that are idle in the
baseline that would be expected for compliance with each of the evaluated regulatory
alternatives and for which model years they can feasibly be applied. EPA considers a unit to
have optimized use of an SCR if emissions rates are equal to (or below) the "widely achievable"
21 For details on the emission rate assumptions within the model, please refer to chapter 3 of the IPM documentation
available at: https://www.epa.gov/system/files/documents/2021-09/epa-platform-v6-summer-2021-reference-case-
09-ll-21-v6.pdf.
22 EGUs with idled SCR or SNCR in the Baseline run represent a small percentage (less than 10 percent) of the EGU
fleet that is equipped with NOx post-combustion controls.
4-16
-------
rate of 0.08 lbs/MMBtu for coal steam units, 0.03 lbs/MMBtu for oil/gas and combustion turbine
units, and 0.012 lb/MMBtu for Combined Cycle units.23 Within IPM, units with partially
operating or idled SCRs are defined as SCR-equipped units with ozone season NOx emission
rates exceeding the optimized rates in the baseline run. These units had their emission rates
lowered to the applicable "widely achievable" optimized emissions rate. These control options
(optimizing partially operating SCR controls or turning on idled SCR controls) are achievable in
2023 and have a uniform control cost of $1,800 per ton (2016$) for coal units and $900 per ton
(2016$) for the other identified sources. As explained below in Section 4.3.3, the costs associated
with this measure are accounted for outside of the model, and no further adjustments were made
inside the model to the variable and fixed operating cost of these units or to their modeled heat
rates. Under the proposed rule, 248 units are projected to fully run existing SCR controls in 2023
and in each year thereafter until the year the unit retires or at the end of the model period.
EPA considers a unit to have optimized use of an SNCR if NOx emissions rates are equal
to or less than the mode 2 rate from the NEEDS database (Summer 2021). As described in
Chapter 3 of EPA's power sector IPM Modeling Documentation, these backstop NOx mode rates
are calculated from historical data and reflect operation of existing post-combustion controls.
Mode 2 for SNCR-controlled coal units is intended to reflect the operation of that unit's post-
combustion control based on prior years when that unit operated its control. Hence any units with
existing SNCRs with NOx emission rates greater than their mode 2 rates in the 25-state region
had their rates lowered to their mode 2 rates. These control options are achievable in 2023 and
have a uniform control cost of $2,000 per ton (2016$). As explained below in Section 4.3.3, the
costs associated with this measure are accounted for outside of the model, and no further
adjustments were made inside the model to the variable and fixed operating cost of these units.
Under the proposed rule, 27 units are projected to fully run existing SNCR controls in 2023 and
in each year thereafter until the year the unit retires or at the end of the model period.
Finally, unit combustion control configurations listed in NEEDS were compared against
Table 3-14 in the documentation for EPA Power Sector Modeling Platform v.6.20 Summer 2021
Reference Case, which lists state-of-the-art combustion control configurations based on unit
23 For details on the derivation of this standard, please see preamble Section VII.B. 1.
4-17
-------
firing type. This allowed EPA to identify units that would receive state-of-the-art combustion
control upgrades in IPM. EPA then followed the procedure in the EGU NOx Mitigation
Strategies Proposed Rule TSD to calculate each of these unit's new NOx emission rate. These
upgrades were assumed to occur in 2023 and have a uniform control cost of $1,800 per ton
(2016$). As explained below in Section 4.3.3, the costs associated with this measure are
accounted for outside of the model, and no further adjustments were made inside the model to
the variable and fixed operating cost of these units. Under the proposed rule, 23 units are
projected to install state-of-the-art combustion controls in 2023 and operate them in each year
thereafter until the year the unit retires or at the end of the model period. The book-life of the
new combustion controls is assumed to be 15 years, hence the stream of costs from 2023-42 fully
captures the cost of any incremental controls under the proposed rule.
The EGU NOx mitigation strategies that are assumed to operate or are available to reduce
NOx in response to each of the regulatory control alternatives are shown in Table 4-2 above;
more information about the estimated costs of these controls can be found in the EGU NOx
Mitigation Strategies Proposed Rule TSD.
Under the proposed rule 32 GW of SCR installations are projected. Under the more
stringent alternative 54 GW of SCR installations are projected. Under the less stringent
alternative 31 GW of new SCR installations are projected. The book-life of the new SCRs is
assumed to be 15 years, hence the stream of costs from 2023-42 fully captures the cost of any
incremental controls under the proposed rule. Under the proposed rule an incremental 18 GW of
coal (63 units) and 4 GW of oil/gas (13 units) retirements are projected by 2030. Under the more
stringent alternative 20 GW of coal and 7 GW of oil/gas retirements are projected by 2030.
Under the less stringent alternative 13 GW of coal and 4 GW of oil/gas retirements are projected
by 2030. The associated costs of retirement are fully captured within the total costs of the
proposed rule presented in the RIA.
In addition to the limitation on ozone season NOx emissions required by the EGU
emissions budgets for the 25 states and the backstop rate limits, there are four important features
of the allowance trading program represented in the model that may influence the level and
location of NOx emissions from affected EGUs, including: the ability of affected EGUs to buy
and sell NOx ozone season allowances from one another for compliance purposes; the ability of
4-18
-------
affected EGUs to bank NOx ozone season allowances for future use; the effect of limits on the
total ozone season NOx emissions from affected EGUs in each state required by the assurance
provisions; and the treatment of banked pre-2023 vintage NOx ozone season allowances issued
under the Revised CSAPR Update now being revised under this proposed rule. Each of these
features of the ozone season allowance trading program is described below. The analysis does
not explicitly capture the dynamic budget adjustments over time, but the forced operation of
controls during the ozone season over the forecast period (even in the absence of binding mass
limits) approximates this feature of the program design.
Affected EGUs are expected to choose the least-cost method of complying with the
requirements of the allowance trading program, and the distribution of ozone season NOx
emissions across affected EGUs is generally governed by this cost-minimizing behavior in the
analysis. The total ozone season NOx emissions from affected EGUs in this analysis are limited
to the amount allowed by the sum of the NOx budgets across the 25 states, the starting bank of
allowances, and any additional allowances that are banked for future use. The number of banked
allowances is influenced by the determination of whether (i) existing controls that are idle in the
baseline run are turned on, (ii) it is less costly to abate ozone season NOx emissions in a current
ozone season than to abate emissions in a later ozone season, and (iii) the restriction on the total
size of the bank which is 10.5 percent of the sum of the state emissions budgets for the current
control period. Affected EGUs are expected to bank NOx ozone season allowances in the 2023
ozone season for use in a later ozone season. The model starts with an assumed bank level in
2023 (described below) and endogenously determines the bank in each subsequent year.
The proposed rule allows pre-2023 vintage NOx ozone season allowances to be used for
compliance with this proposed rule. The sources that would be participants in a revised Group 3
Trading Program under this proposal are transitioning from several different starting points -
with some sources already in the Group 3 Trading Program under its current regulations, some
sources coming from the Group 2 Trading Program, and some sources not currently participating
in any seasonal NOx trading program. As described in Section VII.B.l 1 of the preamble, EPA is
proposing transitional provisions that differ across the sets of potentially affected sources based
on the sources' different starting points. Based on EPA's expectation of the size of the NOx
allowance bank after the one-time conversion carried out pursuant to the terms of this proposed
4-19
-------
rule, the treatment of these banked allowances is represented in the modeling as an additional
22,226 tons of NOx allowances, the equivalent of half of one year of the variability limit
associated with the emission budgets, that may be used by affected EGUs during the 2023 ozone
season or in later ozone seasons under the proposed FIP for the 2015 ozone NAAQS and the
more and less stringent alternatives.
While there are no explicit limits on the exchange of allowances between affected EGUs
and on the banking of 2023 and future-year vintage NOx ozone season allowances, the assurance
provisions limit the amount of seasonal NOx emissions by affected EGUs in each of the 25
states. The assurance level limits affected EGU emissions over an ozone season to the state's
NOx ozone season emissions budget plus an increment equal to 21 percent of each state's
emissions budget. This increment is called the variability limit. See Section VII.B.5 of the
preamble for a discussion of the purpose of the assurance provision and further detail about how
the variability limits and assurance levels are determined. If a state exceeds its assurance level in
a given year, sources within that state are assessed a 3-to-l allowance surrender penalty on the
excess tons. Section VII.B.5 of the preamble also explains how EPA then determines which
EGUs are subject to this surrender requirement. In the modeling, the assurance provisions are
represented by a limit on the total ozone season NOx emissions that may be emitted by affected
EGUs in each state, and thus the modeling does not permit affected EGUs to emit beyond the
assurance levels and thus incur penalties.
4.3.3 Methodology for Estimating Compliance Costs
This section describes EPA's approach to quantify estimated compliance costs associated
with the three illustrative regulatory control alternatives. These compliance costs include
estimates projected directly by the model as well as calculations performed outside of the model
that use IPM model inputs and methods. The model projections capture the costs associated with
shifting generation to lower-NOx emitting EGUs. As discussed in the previous subsection, the
costs of increasing the use and optimizing the performance of existing and operating SCRs and
SNCRs,24 and for installing or upgrading NOx combustion controls, were estimated outside of
the model. The costs for these three NOx mitigation strategies are calculated based on IPM
24 This includes optimizing the performance of SCRs that were not operating.
4-20
-------
emissions projections and use the same NOx control cost equations used in IPM. Therefore, this
estimate is consistent with modeled projections and provides the best available quantification of
the costs of these NOx mitigation strategies.
The following steps summarize EPA's methodology for estimating the component of
compliance costs that are calculated outside of the model for the proposed rule alternative in
2023. Similar calculations are performed for every year in the forecast horizon25:
(1) In the model projections, identify all EGUs in the 25 states that can adopt the following
NOx mitigation strategies (described in previous subsection):
• Fully operating existing SCRs
• Fully operating existing SNCRs
• Installing state-of-the-art combustion controls
(2) Estimate the total NOx reductions that are attributable to each of these strategies:
• Fully operating existing SCRs at coal steam units: 2,090 tons
• Fully operating existing SCRs at oil/gas steam, combined cycle, and
combustion turbine units: 1,526 tons
• Fully operating existing SNCRs: 341 tons
• Installing state-of-the-art combustion controls: 2,056 tons
(3) Estimate the average cost (in 2016$) associated with each of these strategies:26
• Fully operating existing SCRs at coal steam units: $l,800/ton
• Fully operating existing SCRs at oil/gas steam, combined cycle, and
combustion turbine units: $900/ton
• Fully operating existing SNCRs: $2,000/ton
• Installing state-of-the-art combustion controls: $l,600/ton
25 For more information on the derivation of costs and useful life of combustion controls, please see EGU NOx
Mitigation Strategies Proposed Rule TSD.
26 See EGU NOx Mitigation Strategies Proposed Rule TSD for derivation of cost-per-ton estimates for fully
operating SCRs and upgrading to state-of-the-art combustion controls.
4-21
-------
(4) Multiply (2) by (3) to estimate the total cost associated with each of these strategies.
Table 4-5 summarizes the results of this methodology for the proposed rule alternative in
2023.
Table 4-5. Summary of Methodology for Calculating Compliance Costs Estimated Outside
of IPM for Proposed FIP for the 2015 Ozone NAAQS, 2023 (2016$)
NOx Mitigation Strategy
NOx Ozone
Season
Emissions
(tons)
Average Cost
($/ton)
Total Cost
($MM)
Optimize existing SCRs at coal steam units
2,090
1,800
3.8
Optimize existing SCRs at oil/gas, combined
cycle, and combustion turbine units
1,526
900
1.4
Optimize existing SNCRs
361
2,000
0.72
Installing state- of-the-art combustion controls
2,056
1,600
3.3
EPA exogenously updated the emissions rates for the identified EGUs within the 25 states
consistent with the set of controls determined for 2023-2025 within IPM. The model was
updated to incorporate the emissions budgets identified for each case, and the first-year bank
adjustment as outlined in Section 4.3.2. Unit level emission rate constraints were also imposed
on affected uncontrolled units as outlined in Table 4-2 which forced units to choose to either
retrofit or retire in a given year.
The change in the reported power system production cost between the proposed rule
alternative model run and the baseline run was used to capture the cost of generation shifting and
the cost of new SCR installations. The total costs of compliance with the regulatory control
alternatives are estimated as the sum of the costs that are modeled within IPM and the costs that
are calculated outside the model.
4.4 Analytical Framework for Emission Reduction Assessment for Non-EGUs
The number of different industries and emissions unit categories and types, as well as the
total number of emissions units that comprise the non-EGUs makes it challenging to define a
method to identify appropriate control technologies, measures, or strategies and resulting
impactful emissions reductions. The Agency incorporated air quality information as a first step in
4-22
-------
an analytical framework to help determine potentially impactful industries to focus on for further
assessing potential controls, emission reduction potential, air quality improvements, and costs.
Given the timing of this proposal, we developed the analytical framework using inputs from the
air quality modeling for the Revised CSAPR Update (RCU) for 2023, as well as the projected
2023 annual emissions inventory that was used for the air quality modeling for this FIP proposal.
Additional information on the analytical framework is presented in the non-EGU screening
assessment available in the docket.
Using the RCU modeling for 2023, we identified upwind states linked to downwind
nonattainment or maintenance receptors using the 1% of the NAAQS threshold criterion, which
is 0.7 ppb (1% of a 70 ppb NAAQS). In 2023 there were 27 linked states for the 2015 NAAQS:
Alabama, Arkansas, California, Delaware, Iowa, Illinois, Indiana, Kentucky, Louisiana,
Maryland, Michigan, Minnesota, Missouri, Mississippi, New Jersey, New York, Nevada, Ohio,
Oklahoma, Pennsylvania, Tennessee, Texas, Utah, Virginia, Wisconsin, West Virginia, and
Wyoming.
To analyze non-EGU emissions units, we aggregated the underlying projected 2023
emissions inventory data into industries defined by 4-digit NAICS. Then for linked states, we
followed the 2-step process below:
1. Step 1 — We identified industries whose potentially controllable emissions are estimated,
by applying the analytical framework, to have the greatest ppb impact on downwind air
quality, and
2. Step 2 - We determined which of the most impactful industries and emissions units had
the most emissions reductions that would make meaningful air quality improvements at
the downwind receptors at a marginal cost threshold we determined using underlying
control device efficiency and cost information.
To estimate the contributions by industry, defined by 4-digit NAICS, at each downwind
receptor we used the 2023 state-receptor specific RCU ppb/ton values and the RCU calibration
4-23
-------
factors used in the air quality assessment tool (AQAT) for control analyses in 2023.27 We
focused on assessing emissions units that emit >100 tpy of NOx. By limiting the focus to
potentially controllable emissions, well-controlled sources that still emit >100 tpy are excluded.
Instead, the focus is on uncontrolled sources or sources that could be better controlled at a
reasonable cost. As a result, reductions from any industry identified by this process are more
likely to be achievable and to lead to air quality improvements.
Based on the industry contribution data, we prepared a summary with the estimated total,
maximum, and average contributions from each industry and the number of receptors with
contributions >= 0.01 ppb. We evaluated this information to identify breakpoints in the data.
These breakpoints were then used to determine which industries to identify the most impactful
industries to focus on for the next steps in the analysis.
A review of the contribution data indicated that we should focus the assessment of NOx
reduction potential and cost primarily on four industries. These industries each (1) have a
maximum contribution to any one receptor of >0.10 ppb AND (2) contribute >= 0.01 ppb to at
least 10 receptors. We refer to these four industries identified below as comprising "Tier 1".
• Pipeline Transportation of Natural Gas
• Cement and Concrete Product Manufacturing
• Iron and Steel Mills and Ferroalloy Manufacturing
• Glass and Glass Product Manufacturing
In addition, the contribution data suggests that we should include five additional industries
as a second tier in the assessment. These industries each either have (1) a maximum contribution
to any one receptor >=0.10 ppb but contribute >=0.01 ppb to fewer than 10 receptors, or (2) a
maximum contribution <0.10 ppb but contribute >=0.01 ppb to at least 10 receptors. We refer to
these five industries identified below as comprising "Tier 2".
27 The calibration factors are receptor-specific factors. For the RCU, the calibration factors were generated using
2016 base case and 2023 base case air quality model runs. These receptor-level ppb/ton factors are discussed in the
Ozone Transport Policy Analysis Final Rule TSD found here: https://www.epa.gov/sites/default/files/2021-
03/documents/ozone_transport_policy _analysis_final_rule_tsd_0.pdf.
4-24
-------
• Basic Chemical Manufacturing
• Petroleum and Coal Products Manufacturing
• Metal Ore Mining
• Lime and Gypsum Product Manufacturing
• Pulp, Paper, and Paperboard Mills
For additional discussion of the contribution information, see Appendix A of the non-EGU
screening assessment.
Next, to identify an annual cost threshold for evaluating potential emissions reductions in
the Tier 1 and Tier 2 industries, the EPA used CoST, the CMDB, and the projected 2023
emissions inventory to prepare a listing of potential control measures, and their costs, applied to
non-EGU emissions units in the projected 2023 emissions inventory. Using this data, we plotted
curves for Tier 1 industries, Tier 2 industries, Tier 1 and 2 industries, and all industries at $500
per ton increments in the cost per ton threshold using known controls.28 Figure 4-1 shows that
there is a "knee in the curve" at approximately $7,500 per ton (2016$). We used this marginal
cost threshold to further assess potential control strategies, estimated emissions reductions, air
quality improvements, and costs from the potentially impactful industries. Note that controls and
related emissions reductions are available at several estimated cost levels up to the $7,500 per
ton threshold. The costs do not include monitoring, recordkeeping, reporting, or testing costs.
28 Known controls are well-demonstrated control devices and methods that are currently used in practice in many
industries. Known controls do not include cutting edge or emerging pollution control technologies. They also do not
include reductions in operations, changes in processes, or changing inputs such as fuels. Costs reflect capital and
variable costs of installing and operating controls. The costs reflect annual costs of operating controls.
4-25
-------
120,000-
Tjj 100.000-
c
2
o
Q.
c
o
0
3
¦o
01
a:
X
O
z
c
o
-------
2. We used the 2023 state-receptor specific RCU ppb/ton values and the RCU calibration
factors used in the AQAT for control analyses in 2023. We multiplied the estimated non-
EGU reductions by the ppb/ton values and by the receptor-specific calibration factor to
estimate the ppb impacts from these emissions reductions.
Next, because boilers represent the majority emissions unit in the Tier 2 industries for
which there were controls that cost up to $7,500 per ton, we further targeted emissions
reductions and air quality improvements in Tier 2 industries by identifying potentially impactful
industrial, commercial, and institutional (ICI) boilers. To identify potentially impactful boilers,
using the projected 2023 emissions inventory in the linked upwind states we identified a universe
of boilers with >100 tpy NOx emissions that had any contributions at downwind receptors.30'31'32
We refined the universe of boilers to a subset of impactful boilers by sequentially applying the
three criteria below to each boiler. This approach is similar to the overall analytical framework
and was tailored for application to individual boilers.33
• Criterion 1 — Estimated maximum air quality contribution at an individual receptor of
>=0.0025 ppb or estimated total contribution across downwind receptors of >=0.01 ppb.
• Criterion 2 — Controls that cost up to $7,500 per ton.
• Criterion 3 — Estimated maximum air quality improvement at an individual receptor of
>=0.001 ppb.
4.5 Estimated Impacts of the Regulatory Control Alternatives
4.5.1 Emission Reduction Assessment for EGUs
As indicated in Chapter 1, the NOx emissions reductions are presented in this RIA from
2023 through 2042 and are based on IPM projections. As outlined in Section 4.3.2 IPM is
30 We used the 2023fj non-EGU point source inventory files.
31 MD, MO, NV, and WY did not have boilers with >100 tpy NOx emissions.
32 Some coal, oil, and gas-fired industrial boilers may have already installed combustion or post-combustion control
equipment, such as SCR or SNCR, to meet the emission limits contained within EPA's NSPS located at 40 CFR 60
Subpart Db, which requires that some fossil fuel-fired units that commenced construction, modification, or
reconstruction after June 19, 1984 meet various NOx emission limits based on factors such as unit type or heat rate.
Additionally, industrial boilers located in ozone nonattainment areas or within the ozone transport region may have
installed controls to meet emission limits adopted by states to meet NOx RACT requirements.
33 For the impactful boiler assessment, the estimated air quality contributions and improvements were not based on
modeling of individual emissions units or emissions source sectors. The air quality estimates were derived by using
the 2023 state/receptor specific RCU ppb/ton values and the RCU calibration factors used in AQAT. The results
indicate a level of precision not supported by the underlying air quality modeling. The results were intended to
provide an indication of the relative impact across sources.
4-27
-------
operating existing and newly installed controls seasonally based on historical operation patterns
and seasonal and annual emission constraints within the model. Table 4-6 presents the estimated
reduction in power sector NOx emissions resulting from compliance with the evaluated
regulatory control alternatives (i.e., emissions budgets) in the 25 states, as well as the impact on
other states. The emission reductions follow an expected pattern: the less stringent alternative
produces smaller emissions reductions than the proposed rule emissions budgets, and the more
stringent alternative results in more NOx emissions reductions.
Table 4-6. EGU Ozone Season NOx Emissions and Emissions Changes (thousand tons) for
the Baseline run and the Regulatory Control Alternatives from 2023 - 204234
Ozone Season NOX
Total Emissions
(thousand tons)
Baseline
run
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
25 States
176
170
170
169
-6
-6
-6
2023
Other States
88
88
88
88
0
0
0
Total
264
258
258
257
-6
-6
-7
25 States
167
142
154
139
-25
-13
-28
2024
Other States
85
84
84
84
-1
-1
-1
Total
252
226
238
223
-26
-14
-29
25 States
158
114
137
109
-44
-20
-49
2025
Other States
82
80
81
80
-2
-1
-2
Total
240
194
218
189
-46
-22
-51
25 States
163
116
132
111
-47
-32
-53
2026
Other States
85
85
85
85
0
0
0
Total
248
201
216
195
-47
-32
-53
25 States
169
118
126
113
-51
-43
-56
2027
Other States
87
89
89
89
2
1
2
Total
256
207
215
202
-49
-42
-54
25 States
172
119
118
114
-53
-53
-57
2030
Other States
93
94
94
94
1
1
0
Total
265
213
213
208
-52
-52
-57
25 States
169
118
118
115
-50
-51
-54
2035
Other States
90
91
91
92
1
1
2
Total
259
210
209
206
-49
-50
-52
25 States
158
111
111
109
-47
-48
-49
2042
Other States
83
83
83
84
0
0
1
Total
241
194
194
193
-47
-47
-48
Change from Baseline run
34 This analysis is limited to the geographically contiguous lower 48 states.
4-28
-------
Within the compliance modeling, in addition to compliance with the mass budgets,
emissions reductions are also driven by the assumption that units fully operate their controls
during the ozone season. For units with existing controls, this is reflected in the achievement of
the "widely achievable" rate as outlined above. For units that lack existing SCR controls, this is
reflected in the decision to install new controls (which must be operated in the ozone season) or
retire. The proposed rule and less stringent alternative feature identical budgets, but the less
stringent alternative assumes units that lack SCRs must retrofit or retire in the 2028 run year as
compared to the 2025 run year in the proposed rule. Hence emissions reductions under the less
stringent alternative are lower in the 2025 run year than the proposed rule but are similar
thereafter. Similarly, the more stringent alternative features a larger universe of oil/gas steam
units that must choose to retrofit or retire in the 2025 run year, driving higher abatement than the
proposed rule. For details on the emission rate limits assumed in each of the regulatory control
alternatives, please see Table 4-2.
The results of EPA's analysis show that, with respect to compliance with the EGU NOx
emission budgets in 2023, maximizing the use of existing operating SCRs provides the largest
amount of ozone season NOx emission reductions (58 percent, affecting 248 units), installing
state-of-the-art combustion controls provides the next highest levels of ozone season reductions
(33 percent, affecting 23 units), while optimizing existing SNCRs (6 percent, affecting 27 units)
and generation shifting (4 percent) make up the remaining ozone season NOx reductions. Based
on this analysis of how EGUs are expected to comply with the proposed FIP for the 2015 ozone
NAAQS, none of the Group 3 states are projected to hit their variability limits, nor withdraw a
substantial additional number of allowances above the starting bank during the 2023-2042
period.35
In addition to the ozone season NOx reductions, there will also be reductions of other air
emissions associated with EGUs burning fossil fuels (i.e., co-pollutants) that result from
compliance strategies to reduce seasonal NOx emissions, such as generation shifting. These other
35 As shown in Table 4-6, in 2023 and 2025 seasonal NOx emissions from affected EGUs in the Group 3 states are
projected to emit at levels equal to or below the aggregated state budgets, and therefore (i) will not bank additional
allowances, or (ii) on net, not use any banked allowances available at the end of the previous year or, in the case of
2023, from the starting bank.
4-29
-------
emissions include the annual total changes in emissions of NOx, SO2, CO2, and direct PM2.5
emissions changes. The emissions reductions are presented in Table 4-7.
Table 4-7. EGU Annual Emissions and Emissions Changes for NOx, SO2, PM2.5, and CO2
for the Regulatory Control Alternatives for 2023-2042
Annual NOx
Total Emissions
Change from Baseline
(thousand tons)
Baseline
run
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
25 States
401
393
393
392
-8
-9
-9
2023
Other States
193
192
192
192
-1
-1
-1
Total
594
584
584
584
-10
-10
-10
25 States
373
333
354
330
-40
-20
-43
2024
Other States
186
184
184
184
-2
-2
-2
Total
559
518
538
514
-42
-22
-45
25 States
346
274
315
268
-71
-31
-78
2025
Other States
179
177
176
177
-2
-2
-2
Total
524
451
491
444
-73
-33
-80
25 States
366
285
311
278
-81
-55
-88
2026
Other States
186
186
186
186
0
0
0
Total
552
471
497
464
-81
-55
-87
25 States
386
295
307
288
-91
-78
-98
2027
Other States
193
196
195
196
3
2
3
Total
579
491
503
484
-88
-76
-95
25 States
399
301
299
297
-97
-100
-102
2030
Other States
205
207
207
207
2
2
2
Total
604
508
506
504
-96
-98
-100
25 States
394
302
299
298
-92
-95
-96
2035
Other States
199
201
201
201
2
2
3
Total
592
502
499
499
-90
-93
-93
25 States
340
268
264
267
-72
-76
-73
2042
Other States
175
177
176
177
1
1
2
Total
515
445
440
444
-70
-75
-71
Annual SO2
Total Emissions
Change from Baseline
(thousand tons)
Baseline
run
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
25 States
483
483
483
482
1
0
-1
2023
Other States
149
148
148
148
-1
-1
-1
4-30
-------
Total
631
631
631
630
0
-1
-2
25 States
439
399
420
397
-40
-19
-42
2024
Other States
138
137
137
137
-1
-1
-1
Total
578
536
558
535
-42
-20
-43
25 States
395
314
358
313
-81
-38
-83
2025
Other States
128
127
126
127
-2
-2
-1
Total
524
441
484
440
-83
-39
-84
25 States
446
339
369
337
-106
-77
-108
2026
Other States
137
137
137
137
0
0
1
Total
583
476
506
475
-106
-76
-108
25 States
496
365
380
362
-131
-116
-134
2027
Other States
145
147
148
147
2
2
2
Total
641
512
528
510
-129
-113
-131
25 States
551
446
449
446
-105
-101
-105
2030
Other States
159
160
160
160
1
1
1
Total
710
605
609
606
-104
-100
-103
25 States
555
467
470
465
-89
-85
-91
2035
Other States
164
156
156
157
-8
-7
-7
Total
719
623
626
621
-96
-93
-98
25 States
515
460
464
461
-56
-51
-55
2042
Other States
149
150
151
150
1
2
1
Total
664
610
615
611
-54
-50
-54
Annual CO2
Total Emissions
Change from Baseline
(million metric tonnes)
Baseline
run
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
25 States
902
902
902
902
0
0
0
2023
Other States
417
416
416
416
-1
-1
-1
Total
1319
1319
1319
1318
0
0
0
25 States
856
838
847
837
-18
-9
-19
2024
Other States
411
411
410
411
0
-1
0
Total
1267
1248
1257
1247
-18
-10
-19
25 States
810
773
791
771
-36
-18
-38
2025
Other States
405
405
404
405
0
-1
0
Total
1215
1178
1196
1176
-37
-19
-38
25 States
844
801
816
799
-43
-29
-45
2026
Other States
416
419
418
419
3
2
3
Total
1260
1220
1234
1218
-40
-26
-42
25 States
879
830
840
828
-49
-39
-52
2027
Other States
427
433
432
432
6
5
6
4-31
-------
Total
1306
1263
1272
1260
-43
-34
-46
25 States
910
857
861
856
-53
-49
-54
2030
Other States
438
442
441
443
3
3
4
Total
1348
1298
1302
1298
-50
-45
-50
25 States
926
886
887
885
-40
-39
-41
2035
Other States
443
445
446
446
2
3
3
Total
1369
1331
1333
1331
-38
-36
-38
25 States
886
862
864
862
-25
-23
-25
2042
Other States
422
422
422
423
0
0
1
Total
1308
1284
1285
1284
-25
-23
-24
Annual PM2.5
Total Emissions
Change from Baseline
(thousand tons)
Baseline
run
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
25 States
66
66
66
66
0
0
0
2023
Other States
32
32
32
32
0
0
0
Total
98
98
98
98
0
0
0
25 States
63
59
62
59
-4
-1
-4
2024
Other States
32
31
31
31
0
0
0
Total
95
91
94
91
-4
-1
-4
25 States
61
52
59
52
-9
-2
-9
2025
Other States
31
31
31
31
0
0
0
Total
92
83
90
83
-9
-2
-9
25 States
63
54
58
54
-9
-5
-9
2026
Other States
32
32
32
32
0
0
0
Total
95
86
90
86
-9
-5
-9
25 States
65
55
58
55
-10
-8
-10
2027
Other States
33
34
34
34
0
0
0
Total
99
89
91
89
-10
-7
-10
25 States
66
56
56
57
-9
-10
-9
2030
Other States
35
35
35
35
0
0
1
Total
100
91
91
92
-9
-9
-9
25 States
69
58
57
59
-11
-12
-11
2035
Other States
35
35
36
36
0
0
0
Total
104
94
93
94
-11
-12
-10
25 States
67
59
58
59
-8
-9
-8
2042
Other States
34
34
34
34
0
0
0
Total
101
93
92
93
-8
-9
-8
4-32
-------
4.5.2 Compliance Cost Assessment for EGUs
The estimates of the changes in the cost of supplying electricity for the regulatory control
alternatives are presented in Table 4-8. Since the proposed rule does not result in any additional
recordkeeping, monitoring or reporting requirements, the costs associated with compliance,
monitoring, recordkeeping, and reporting requirements are not included within the estimates in
this table.
Table 4-8. National Power Sector Compliance Cost Estimates (millions of 2016$) for the
Regulatory Control Alternatives
Proposed
Rule
More-
Stringent
Alternative
Less-
Stringent
Alternative
2023-2027 (Annualized)
$690
$1,076
$5
2023-2042 (Annualized)
$1,204
$1,624
$1,104
2023 (Annual)
($209)
($178)
($173)
2024 (Annual)
$707
$1,180
($406)
2025 (Annual)
$707
$1,180
($406)
2026 (Annual)
$707
$1,180
($406)
2027 (Annual)
$1,544
$1,983
$1,540
2030 (Annual)
$1,235
$1,740
$1,200
2035 (Annual)
$1,729
$2,335
$1,596
2042 (Annual)
$910
$1,001
$1,757
"2023-2027 (Annualized)" reflects total estimated annual compliance costs levelized over the period 2023 through
2027 and discounted using a 3.76 real discount rate.36 This does not include compliance costs beyond 2027. "2023-
2042 (Annualized)" reflects total estimated annual compliance costs levelized over the period 2023 through 2042
and discounted using a 3.76 real discount rate. This does not include compliance costs beyond 2042. "2023
(Annual)" through "2042 (Annual)" costs reflect annual estimates in each of those years.37
There are several notable aspects of the results presented in Table 4-8. The most notable
result in Table 4-8 is that the estimated annual compliance costs for the less stringent alternative
is negative (i.e., a cost reduction) in 2023 through 2026, although this regulatory control
alternative reduces NOx emissions by 40 thousand tons as shown in Table 4-7. While seemingly
counterintuitive, estimating negative compliance costs in a single year is possible given the
36 This table reports compliance costs consistent with expected electricity sector economic conditions. An NPV of
costs was calculated using a 3.76% real discount rate consistent with the rate used in IPM's objective function for
cost-minimization. The NPV of costs was then used to calculate the levelized annual value over a 5-year period
(2023-2027) and a 20-year period (2023-2042) using the 3.76% rate as well. Tables ES-19 and 8-4 report the NPV
of the annual stream of costs from 2023-2042 using 3% and 7% consistent with OMB guidance.
37 Cost estimates include financing charges on capital expenditures that would reflect a transfer and would not
typically be considered part of total social costs.
4-33
-------
assumption of perfect foresight. IPM's objective function is to minimize the discounted net
present value (NPV) of a stream of annual total cost of generation over a multi-decadal time
period.38 For example, with the assumption of perfect foresight it is possible that on a national
basis within the model the least-cost compliance strategy may be to delay a new investment or
retirement that was projected to occur sooner in the baseline run. Such a delay could result in a
lowering of annual cost in an early time period and increase it in later time periods.39 The less-
stringent alternative is designed to impose unit-level emission rates in the 2028 run year as
compared to the 2025 run year as under the proposed rule and the more stringent alternative.
This results in delayed retrofit and retirement at facilities covered by those rate limits, which in
turn leads to negative total cost point estimates in 2023 through 2026. Under the proposed rule,
operating existing SCR and SNCR controls and upgrading to state-of-the-art combustion controls
provides a large share of the total emissions reductions in 2023. Generation shifting costs are
negative in 2023, but positive thereafter. The result is that the costs in 2023 are lower than costs
thereafter. Projected costs for the illustrative proposed rule peak in 2035 at $1.7 billion (2016$)
and annualized costs for the 2023-2042 period are $1.2 billion (2016$). To put these costs into
context, the incremental 2035 projected cost constitutes 1.2 percent of total projected baseline
system production costs.
Under the more stringent alternative, while 2023 includes the same set of controls as under
the proposed FIP for the 2015 ozone NAAQS, a larger number of non-SCR controlled Oil/Gas
steam units are subject to backstop emission rates that force units to either retrofit or retire. This,
combined with more stringent state budgets driving generation shifting costs positive in every
year, results in costs that grow over the 2024 - 2035 period.
In addition to evaluating annual compliance cost impacts, EPA believes that a full
understanding of these three regulatory control alternatives benefits from an evaluation of
annualized costs over the 2023-2027 timeframe. Starting with the estimated annual cost time
series, it is possible to estimate the net present value of that stream, and then estimate a levelized
38 For more information, please see Chapter 2 of the IPM documentation.
39 As a sensitivity, EPA re-calculated costs assuming annual costs cannot be negative. This resulted in annualized
2023-42 costs under the proposed rule increasing from $1,204 million to $1,219 million (1.3%), and did not change
the conclusions of this RIA.
4-34
-------
annual cost associated with compliance with each regulatory control alternative.40 For this
analysis we first calculated the NPV of the stream of costs from 2023 through 202741 using a
3.76 percent discount rate. EPA typically uses a 3 and a 7 percent discount rate to discount future
year social benefits and social costs in regulatory impact analyses (USEPA, 2010). In this cost
annualization we use a 3.76 percent discount rate, which is consistent with the rate used in IPM's
objective function for minimizing the NPV of the stream of total costs of electricity generation.
This discount rate is meant to capture the observed equilibrium market rate at which investors
are willing to sacrifice present consumption for future consumption and is based on a Weighted
Average Cost of Capital (WACC).42 After calculating the NPV of the cost streams, the same 3.76
percent discount rate and 2023-2027 time period are used to calculate the levelized annual (i.e.,
annualized) cost estimates shown in Table 4-8.43 The same approach was used to develop the
annualized cost estimates for the 2023-2042 timeframe. Additionally, note that the 2023-2027
and 2023-2042 equivalent annualized compliance cost estimates have the expected relationship
to each other; the annualized costs are lowest for the less stringent alternative, and highest for the
more stringent alternative.
4.5.3 Impacts on Fuel Use, Prices and Generation Mix
The proposed FIP for the 2015 ozone NAAQS is expected to result in significant NOx
emissions reductions and impacts to the power sector. While these impacts are relatively small in
percentage terms, consideration of these potential impacts is an important component of
assessing the relative impact of the regulatory control alternatives. In this section we discuss the
estimated changes in fuel use, fuel prices, generation by fuel type, capacity by fuel type, and
retail electricity prices for the 2023 and 2025 IPM model run years.
40 The XNPV() function in Microsoft Excel 2013 was used to calculate the NPV of the variable stream of costs, and
the PMT() function in Microsoft Excel 2013 is used to calculate the level annualized cost from the estimated NPV.
41 Consistent with the relationship between IPM run years and calendar years, EPA assigned 2023 compliance cost
estimates to both 2022 and 2023 in the calculation of NPV, and 2025 compliance cost to 2024 and 2025. For more
information, see Chapter 7 of the IPM Documentation.
42 The IPM Baseline run documentation (Section 10.4.1 Introduction to Discount Rate Calculations) states "The real
discount rate for all expenditures (capital, fuel, variable operations and maintenance, and fixed operations and
maintenance costs) in the EPA Platform v6 is 3.76%."
43 The PMT() function in Microsoft Excel 2013 is used to calculate the level annualized cost from the estimated
NPV.
4-35
-------
Table 4-9 and Table 4-10 present the percentage changes in national coal and natural gas
usage by EGUs in the 2023 and 2025 run years. These fuel use estimates reflect a modest shift to
natural gas and renewables from coal in 2023 as a result of tightening budgets. In the 2025 run
year, coal consumption reductions under the proposed rule and the more stringent scenario are
driven by increasing coal EGU retirements and reduced coal dispatch as a result of tightening
budgets and the need to install SCR controls or retire uncontrolled units as shown in Table 4-11.
To put these reductions into context, under the Baseline, coal consumption is projected to
decrease from 64 million tons in 2023 to 48 million tons in 2025 (13 percent annually), whereas
under the proposed rule coal consumption is projected to decrease from 64 million tons in 2023
to 44 million tons in 2025 (16 percent annually). Between 2015 and 2020, annual coal
consumption in the electric power sector fell between 8 and 19 percent annually.44
Table 4-9. 2023 and 2025 Projected U.S. Power Sector Coal Use for the Baseline and the
Regulatory Control Alternatives
Million Tons
Percent Change from Baseline
Year
Baseline
Proposed
Rule
Less-
Stringent
Alt.
More-
Stringent
Alt.
Proposed
Rule
Less-
Stringent
Alt.
More-
Stringent
Alt.
Appalachia
64
64
64
64
0.5%
0.7%
0.7%
Interior
60
59
59
59
-0.9%
-0.8%
-1.2%
Waste Coal
2023
4
4
4
4
0.0%
0.0%
0.0%
West
180
180
179
180
0.0%
0.0%
0.0%
Total
308
308
308
308
0.0%
-0.1%
-0.1%
Appalachia
48
44
47
45
-8.3%
-4.0%
-7.5%
Interior
44
45
45
43
1.0%
0.7%
-1.8%
Waste Coal
2025
4
4
4
4
0.0%
0.0%
0.0%
West
151
136
144
136
-10.2%
-4.3%
-10.1%
Total
248
229
240
228
-7.7%
-3.3%
-7.9%
44 US EIA Monthly Energy Review, Table 6.2, January 2022.
4-36
-------
Table 4-10. 2023 and 2025 Projected U.S. Power Sector Natural Gas Use for the Baseline
and the Regulatory Control Alternatives
Trillion Cubic Feet
Percent Change from Baseline
Year
Baseline
Proposed
Rule
Less-
Stringent
Alt.
More-
Stringent
Alt.
Proposed
Rule
Less-
Stringent
Alt.
More-Stringent
Alt.
2023
12
12
12
12
0.05%
0.04%
0.04%
2025
12
12
12
12
0.20%
-0.77%
0.25%
Table 4-11 and Table 4-12 present the projected coal and natural gas prices in 2023 and
2025, as well as the percent change from the baseline run projected due to the regulatory control
alternatives. These minor impacts in 2023 are consistent with the small changes in fuel use
summarized above. The projected impacts in 2025 are larger in absolute value and consistent
with tightening budgets.
Table 4-11. 2023 and 2025 Projected Minemouth and Power Sector Delivered Coal Price
(2016$) for the Baseline and the Regulatory Control Alternatives
$/MMBtu
Percent Change from Baseline
Baseline
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Minemouth
2023
Delivered
1.13
1.59
1.13
1.58
1.13
1.58
1.13
1.58
-0.72%
-0.52%
-0.63%
-0.47%
-0.79%
-0.58%
Minemouth
2025
Delivered
1.17
1.57
1.17
1.53
1.16
1.55
1.17
1.53
0.30%
-2.51%
-0.68%
-1.33%
0.10%
-2.67%
Table 4-12. 2023 and 2025 Projected Henry Hub and Power Sector Delivered Natural Gas
Price (2016$) for the Baseline and the Regulatory Control Alternatives
$/MMBtu
Percent Change from Baseline
Baseline
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Henry Hub
2023
Delivered
2.40
2.50
2.39
2.49
2.39
2.49
2.39
2.49
-0.55%
-0.53%
-0.46%
-0.43%
-0.45%
-0.41%
Henry Hub
2025
Delivered
2.22
2.31
2.22
2.31
2.22
2.31
2.22
2.31
0.00%
0.00%
-0.01%
0.03%
0.00%
0.04%
4-37
-------
Table 4-13 presents the projected percentage changes in the amount of electricity
generation in 2023 and 2025 by fuel type. Consistent with the fuel use projections and emissions
trends above, EPA projects an overall shift from coal to gas and renewables. The projected
impact in 2025 larger, reflecting the tightening budgets.
Table 4-13. 2023 and 2025 Projected U.S. Generation by Fuel Type for the Baseline and the
Regulatory Control Alternatives
Generation (TWh)
Percent Change from Baseline
Year
Baseline
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
„ , Less- More-
'"''ruIc Stringent Stringent
Alternative Alternative
Coal
596
595
595
595
-0.10%
-0.09%
-0.12%
Natural Gas
1,657
1,658
1,658
1,659
0.03%
0.03%
0.09%
Nuclear
741
741
741
741
0.00%
0.00%
0.00%
Hydro
294
294
294
294
-0.01%
0.00%
-0.02%
2023
Non-Hydro RE
745
745
745
745
0.00%
0.00%
0.00%
Oil/Gas Steam
51
51
51
51
0.47%
0.22%
-1.05%
Other
36
36
36
36
0.00%
0.00%
0.00%
Grand Total
4,135
4,135
4,135
4,135
0.00%
0.00%
0.00%
Coal
485
447
469
445
-7.77%
-3.25%
-8.22%
Natural Gas
1,660
1,663
1,655
1,665
0.15%
-0.32%
0.30%
Nuclear
689
689
689
689
0.00%
0.00%
0.00%
Hydro
293
293
293
293
0.19%
0.08%
0.17%
2025
Non-Hydro RE
949
983
975
983
3.64%
2.77%
3.64%
Oil/Gas Steam
58
58
52
58
0.31%
-9.36%
-0.18%
Other
36
36
36
36
0.00%
0.00%
0.00%
Grand Total
4,185
4,186
4,186
4,185
0.02%
0.01%
0.01%
Note: In this table, "Non-Hydro RE" includes biomass, geothermal, landfill gas, solar, and wind.
Table 4-14 presents the projected percentage changes in the amount of generating capacity
in 2023 and 2025 by primary fuel type. As explained above, none of the regulatory control
alternatives are expected to have a net impact on overall capacity by primary fuel type in 2023,
and the model was specified accordingly. By 2030 the proposed rule is projected to result in an
additional 18 GW of coal and 4 GW of oil/gas steam retirements nationwide relative to the
Baseline run, constituting a reduction of 13 percent of national coal capacity and 2 percent of
4-38
-------
oil/gas steam capacity, partially reflecting some earlier retirement under the proposed rule
relative to the Baseline run. This is compared to an average recent historical retirement rate of
11 GW per year from 2015 -2020 (https://www.eia.gov/todayinenergy/detail.php?id=50838).
Additionally, the proposed rule is projected to incentivize an incremental 18 GW of SCR
retrofit at coal plants and 14 GW of SCR retrofit at oil/gas steam plants. The proposed rule is
also projected to result in an incremental 14 GW of renewable capacity additions in 2025
(consistent primarily of solar capacity builds). These builds reflect early action, i.e., builds that
would otherwise have occurred later in the forecast period. By 2035-40 total solar capacity
equilibrates between the baseline and proposed rule scenarios.
Table 4-14. 2023 and 2025 Projected U.S. Capacity by Fuel Type for the Baseline run and
the Regulatory Control Alternatives
Capacity (GW)
Percent Change from Baseline run
Year
Baseline
run
Proposed
Rule
Less-
Stringent
Alt
More-
Stringent
Alt
Proposed
Rule
Less-
Stringent
Alt
More-
Stringent
Alt
Coal
164
164
164
164
0%
0%
0%
Natural Gas
429
429
429
429
0%
0%
0%
Nuclear
93
93
93
93
0%
0%
0%
Hydro
102
102
102
102
0%
0%
0%
2023
Non-Hydro RE
234
234
234
234
0%
0%
0%
Oil/Gas Steam
63
63
63
63
0%
0%
0%
Other
7
7
7
7
0%
0%
0%
Grand Total
1,104
1,104
1,104
1,104
0%
0%
0%
Coal
157
134
144
132
-15%
-8%
-16%
Natural Gas
430
432
431
433
0%
0%
1%
Nuclear
86
86
86
86
0%
0%
0%
Hydro
102
102
102
102
0%
0%
0%
2025
Non-Hydro RE
281
295
291
295
5%
4%
5%
Oil/Gas Steam
64
65
60
63
2%
-6%
-2%
Other
7
7
7
7
0%
0%
0%
Grand Total
1,140
1,134
1,135
1,131
-1%
0%
-1%
Note: In this table, "Non-Hydro RE" includes biomass, geothermal, landfill gas, solar, and wind
4-39
-------
EPA estimated the change in the retail price of electricity (2016$) using the Retail Price
Model (RPM) 45 The RPM was developed by ICF for EPA and uses the IPM estimates of
changes in the cost of generating electricity to estimate the changes in average retail electricity
prices. The prices are average prices over consumer classes (i.e., consumer, commercial, and
industrial) and regions, weighted by the amount of electricity used by each class and in each
region. The RPM combines the IPM annual cost estimates in each of the 64 IPM regions with
EIA electricity market data for each of the 25 electricity supply regions (shown in Figure 4-2) in
the electricity market module of the National Energy Modeling System (NEMS).46
Table 4-15 and Table 4-16 present the projected percentage changes in the retail price of
electricity for the three regulatory control alternatives in 2023 and 2025, respectively. Consistent
with other projected impacts presented above, average retail electricity prices at both the national
and regional level are projected to be small in 2023. In 2025, EPA estimates that this proposed
rule will result in a 1 percent increase in national average retail electricity price, or by about 1.02
mills/kWh.
Table 4-15. Average Retail Electricity Price by Region for the Baseline and the Regulatory
Control Alternatives, 2023
All Sector
2023 Average Retail Electricity Price
(2016 mills/kWh)
Percent Change from Baseline
Region
Baseline
Proposed
Rule
Less-
Stringent
Alt.
More-
Stringent
Alt.
Proposed
Rule
Less-
Stringent
Alt.
More-
Stringent
Alt.
TRE
71
69
69
69
-3%
-3%
-3%
FRCC
91
91
91
91
0%
0%
0%
MISW
95
95
95
95
0%
0%
0%
MISC
85
84
84
84
-1%
-1%
-1%
MISE
92
92
91
92
0%
0%
0%
MISS
73
73
73
73
0%
0%
0%
ISNE
130
130
130
130
0%
0%
0%
NYCW
533
536
537
501
1%
1%
-6%
NYUP
114
114
114
114
0%
0%
0%
PJME
154
157
157
153
2%
1%
-1%
PJMW
89
88
88
88
-1%
-1%
-1%
PJMC
85
83
83
83
-2%
-2%
-2%
PJMD
69
67
67
67
-2%
-2%
-2%
45 See documentation available at: https://www.epa.gov/airmarkets/retail-price-model
46 See documentation available at:
https://www.eia.gov/outlooks/aeo/nems/documentation/electricity /pdf/m068(2020).pdf
4-40
-------
SRCA
91
91
91
91
0%
0%
0%
SRSE
91
91
91
91
0%
0%
0%
SRCE
67
67
67
67
0%
0%
0%
SPPS
74
73
73
73
0%
0%
0%
SPPC
100
100
100
100
0%
0%
0%
SPPN
65
65
65
65
0%
0%
0%
SRSG
96
96
96
96
0%
0%
0%
CANO
140
140
140
140
0%
0%
0%
CASO
171
171
171
171
0%
0%
0%
NWPP
70
70
70
70
0%
0%
0%
RMRG
90
90
90
90
0%
0%
0%
BASN
84
84
84
84
0%
0%
0%
NATIONAL
103
103
103
102
0%
0%
-1%
Table 4-16. Average Retail Electricity Price by Region for the Baseline and the Regulatory
Control Alternatives, 2025
All Sector
2025 Average Retail Electricity Price
(2016 mills/kWh)
Percent Change from Baseline
Region
Baseline
Proposed
Rule
Less-
Stringent
Alt.
More-
Stringent
Alt.
Proposed
Rule
Less-
Stringent
Alt.
More-
Stringent
Alt.
TRE
66
68
66
68
3%
0%
3%
FRCC
89
89
89
89
0%
0%
0%
MISW
93
94
93
94
1%
0%
1%
MISC
83
86
84
86
3%
1%
3%
MISE
80
80
80
83
-1%
0%
3%
MISS
73
74
73
74
1%
0%
1%
ISNE
135
135
135
135
0%
0%
0%
NYCW
173
173
173
173
0%
0%
0%
NYUP
108
109
109
109
1%
1%
0%
PJME
94
94
93
94
0%
0%
1%
PJMW
83
86
85
89
5%
3%
8%
PJMC
74
81
78
86
9%
5%
15%
PJMD
62
66
65
69
6%
3%
10%
SRCA
89
89
89
89
0%
0%
0%
SRSE
89
89
89
89
-1%
-1%
-1%
SRCE
66
66
66
66
0%
0%
1%
SPPS
75
75
74
75
0%
-1%
1%
SPPC
99
100
99
100
1%
0%
1%
SPPN
64
63
63
63
-2%
-1%
-2%
SRSG
95
95
95
95
0%
0%
0%
CANO
147
147
147
147
0%
0%
0%
4-41
-------
CASO
179
179
179
179
0%
0%
0%
NWPP
71
71
71
71
0%
0%
0%
RMRG
88
88
88
88
0%
0%
0%
BASN
84
85
83
85
1%
0%
1%
NATIONAL
90
91
91
92
1%
0%
2%
NWPP
MISWj
19
SPPN
1^8
I NYCW
>12
'PJWC
f21H
[cano:
S1MWj
BASN
W2*W
RMRG1
13
PJMD
K ¦
[SRCfl
Figure 4-2. Electricity Market Module Regions
Source: EM (http:/Avww. eia.gov/forecasts/aeo/pdf/nerc_map.pdf
4.5.4 Emission Reduction and Compliance Cost Assessment for Non-EGUs from Screening
Assessment for 2026
EPA determined that 2026 was the potential earliest date by which controls on non-EGU
emissions units could be installed. EPA updated its analytical framework to the analytic year of
2026 by which EPA is proposing non-EGU controls be installed across the Tier 1 and Tier 2
industries and various emissions unit types. As such, we prepared a screening assessment for the
year 2026 by generally applying the analytical framework detailed above. The screening
assessment for 2026 is not intended to be, nor take the place of, a unit-specific detailed
engineering analysis that fully accounts for retrofit difficulty for the emissions units, potential
controls, and related costs.
4-42
-------
Specifically in the screening assessment for 2026, we:
• Retained the impactful industries identified in Tier 1 and Tier 2, the $7,500 cost per ton
threshold, and the methodology for identifying impactful boilers,
• Modified the framework to address challenges associated with using the projected 2023
emissions inventory by using the 2019 emissions inventory47, and
• Updated the air quality modeling data by using data for 2026.
We used CoST to identify emissions units, emissions reductions, and costs to include in a
proposed FIP; however, CoST is designed to be used for illustrative control strategy analyses
(e.g., NAAQS regulatory impact analyses) and not for unit-specific, detailed engineering
analyses. These estimates from CoST identify proxies for (1) non-EGU emissions units that have
emission reduction potential, (2) potential controls for and emissions reductions from these
emissions units, and (3) control costs from the potential controls on these emissions units.
To prepare the screening assessment for 2026, we applied the analytical framework
detailed above. The assessment includes emissions units from the Tier 1 industries and impactful
boilers from the Tier 2 industries. Using the latest air quality modeling for 2026, we identified
upwind states linked to downwind nonattainment or maintenance receptors using the 1% of the
NAAQS threshold criterion, or 0.7 ppb. In 2026 there are 23 linked states for the 2015 NAAQS:
Arkansas, California, Illinois, Indiana, Kentucky, Louisiana, Maryland, Michigan, Minnesota,
Missouri, Mississippi, New Jersey, New York, Nevada, Ohio, Oklahoma, Pennsylvania, Texas,
Utah, Virginia, Wisconsin, West Virginia, and Wyoming.
We re-ran CoST with known controls, the CMDB, and the 2019 emissions inventory. The
analysis assumes that the 2019 emissions from the emissions units will be the same in 2026 and
later years. We specified CoST to allow replacing an existing control if a replacement control is
estimated to be >10 percent more effective than the existing control. We did not replace an
existing control if the 2019 emissions inventory indicated the presence of that control, even if the
CMDB reflects a greater control efficiency for that control. Also, we removed six facilities from
consideration because they are subject to an existing consent decree, are shut down, or will shut
47 EPA determined that the 2019 inventory was appropriate because it provided a more accurate prediction of
potential near-term emissions reductions. Also, see memorandum titled Screening Assessment of Potential
Emissions Reductions, Air Quality Impacts, and Costs from Non-EGU Emissions Units for 2026, available in the
docket, for a discussion of the challenges associated with using the projected 2023 emissions inventory.
4-43
-------
down by 2026. For additional detail on the six facilities removed, see Appendix B in the non-
EGU screening assessment. Table 4-17 below summarizes the estimated reductions and annual
total and average annual costs (2016$) for the proposal. The cost estimates do not include
monitoring, recordkeeping, reporting, or testing costs.48 The proposed rule alternative includes
489 non-EGU emissions units.
Table 4-18 below summarizes, by industry, the number of emissions units, reductions, and
costs for the proposal. Table 4-19 below summarizes the estimated reductions and annual total
and average annual costs (2016$) for the less and more stringent alternatives.
Because the proposed FIP for the 2015 ozone NAAQS includes ozone season emissions
limits for the non-EGU emissions units and because we do not know if all affected sources will
run controls year-round or only during ozone season, we include estimates of ozone season NOx
emissions reductions and not annual estimates in Table 4-17 and Table 4-19. Note that some of
the EGU controls are assumed to run year-round. Also, because the proposed FIP for the 2015
ozone NAAQS includes emissions limits, and the non-EGU screening assessment does not
account for growth in the affected industries and capital turnover over time, the reductions are
the same each year over the period from 2026 to 2042.
For additional 2026 screening assessment results — including by industry and by state,
estimated emissions reductions and costs — see the non-EGU screening assessment.
Table 4-17. Annual Estimated Emissions Reductions for 2026-2042 (ozone season tons) and
Annual Total Costs for the Proposed Rule
48 EPA submitted an information collection request (ICR) to OMB associated with the proposed monitoring,
calibrating, recordkeeping, reporting and testing activities required for non-EGU emissions units ~ ICR for the
Proposed Rule, Federal Implementation Plan Addressing Regional Ozone Transport for the 2015 Primary Ozone
National Ambient Air Quality Standard: Transport Obligations for non-Electric Generating Units, EPA ICR No.
2705.01. The ICR is summarized in Section XI.B.2 of the proposed rule preamble. The ICR includes estimated
monitoring, recordkeeping, reporting, and testing costs of approximately $11.45 million per year for the first three
years. These costs are not reflected in the cost estimates in Table 4-17 and Table 4-19.
Proposed Alternative
Tier 1 Industries with Known
Controls that Cost up to
$7,500/ton
Ozone Season NOx Annual Total Cost (million 2016$)
Emissions Reductions (Average Annual Cost/Ton)
41,153
$356.6 ($3,610)
4-44
-------
Ozone Season NOx Annual Total Cost (million 2016$)
Proposed Alternative Emissions Reductions (Average Annual Cost/Ton)
Impactful Boilers in Tier 2
Industries with Known Controls 6,033 $54.2 ($3,744)
that Cost up to $7,500/ton
Totals 47,186 $410.8
Table 4-18. By Industry, Number and Type of Emissions Units and Total Estimated
Emissions Reductions (ozone season tons)
Number of Units by Type
Ozone Season Emission Reductions
Industry
Region
Boilers
Internal
Combustion
Engines
Industrial
Processes
Boilers
Internal
Combustion
Engines
Industrial
Processes
Glass and Glass Product
Manufacturing
East
-
-
41
-
-
6,367
West
.
.
3
.
.
299
Cement and Concrete
Product Manufacturing
East
1
-
39
16
-
5,948
West
.
.
8
.
.
2,128
Iron and Steel Mills and
Ferroalloy
Manufacturing
East
25
15
2,044
1,207
Pipeline Transportation
of Natural Gas
East
-
296
-
-
22,390
-
West
.
11
.
.
754
.
Basic Chemical
Manufacturing
East
17
-
-
1,698
-
-
Petroleum and Coal
Products Matinfacluri rig
East
9
-
-
962
-
-
West
1
.
.
68
.
.
Pulp, Paper, and
Paperboard Mills
East
25
~
~
3,305
~
~
Blue highlights reflect western
Orange highlights reflect. Tier 2 industries
Table 4-19. Annual Estimated Emissions Reductions for 2026-2042 (ozone season tons) and
Annual Total Costs for the Less and More Stringent Alternatives
Ozone Season NOx
Annual Total Cost (million 2016$)
Alternative
Emissions Reductions
(Average Annual Cost/Ton)
Less Stringent Alternative -
Tier 1 Industries with Known
41,153
$356.6 ($3,610)
Controls that Cost up to
$7,500/ton
More Stringent Alternative
Tier 1 Industries with Known
Controls that Cost up to
$7,500/ton and
50,918
$445.1 ($3,642)
All Boilers in Tier 2 Industries
with Known Controls that Cost
4-45
-------
Alternative
up to $7,500/ton
Ozone Season NOx Annual Total Cost (million 2016$)
Emissions Reductions (Average Annual Cost/Ton)
4.5.6 Total Emissions Reductions and Compliance Costs for EGUs and Non-EGUs
For years between 2023 and 2042, Table 4-20 below summarizes the total annual
estimated emissions reductions and compliance costs for EGUs and non-EGUs for the proposed
rule and the less and more stringent alternatives. Table 4-21 below summarizes the present value
(PV) and equivalent annualized value (EAV) of the total national compliance cost estimates for
EGUs and non-EGUs for the proposed rule and the less and more stringent alternatives. We
present the PV of the costs over the twenty-year period 2023 to 2042. We also present the EAV,
which represents a flow of constant annual values that, had they occurred in each year from 2023
to 2042, would yield a sum equivalent to the PV. The EAV represents the value of a typical cost
for each year of the analysis.
Table 4-20. Total Annual Estimated NOx Emissions Reductions (ozone season, thousand
tons) and Compliance Costs (million 2016$), 2023-2042
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Proposed
Rule
Less-
Stringent
Alternative
More-
Stringent
Alternative
Emissions Reductions
(ozone season, thousand tons)
Compliance Costs
(million 2016$)
2023
EGUs
Non-EGUs
6
6
7
(209)
(173)
(178)
Total
6
6
7
(209)
(173)
(178)
2026
EGUs
47
32
53
707
(406)
1,180
Non-EGUs
47
41
51
411
357
445
Total
95
73
103
1,117
(49)
1,625
2027
EGUs
49
42
54
1,544
1,540
1,983
Non-EGUs
47
41
51
411
357
445
Total
96
83
105
1,955
1,896
2,428
2030
EGUs
52
52
57
1,235
1,200
1,740
Non-EGUs
47
41
51
411
357
445
Total
99
93
108
1,646
1,557
2,185
2035
EGUs
49
50
52
1,729
1,596
2,335
Non-EGUs
47
41
51
411
357
445
Total
96
91
103
2,139
1,953
2,780
2042
EGUs
47
47
48
910
1,757
1,001
Non-EGUs
47
41
51
411
357
445
Total
94
88
99
1,321
2,114
1,446
4-46
-------
Table 4-21. Total National Compliance Cost Estimates (millions of 2016$) for the Proposed
Rule and the Less and More Stringent Alternatives
Proposed Rule
Less Stringent
Alternative
More Stringent Alternative
3 Percent
7 Percent
3 Percent
7 Percent
3 Percent
7 Percent
Present Value
EGU 2023-2042
$17,000
$11,000
$16,000
$9,400
$23,000
$15,000
Present Value
Non-EGU 2026-2042
$4,800
$3,100
$4,200
$2,700
$5,200
$3,300
Present Value
Total 2023-2042
$22,000
$14,000
$20,000
$12,000
$28,000
$18,000
EGU
Equivalent Annualized
Value
$1,100
$1,000
$1,100
$890
$1,500
$1,400
Non-EGU
Equivalent Annualized
Value
$320
$290
$280
$250
$350
$310
Total
Equivalent Annualized
Value
$1,500
$1,300
$1,300
$1,100
$1,900
$1,700
Note: Values have been rounded to two significant figures
4.5.7 Impact of Emissions Reductions on Maintenance and Nonattainment Monitors
EPA evaluated whether reductions resulting from the selected control stringencies for
EGUs in 2023 and 2026 combined with the emissions reductions expected for non-EGUs in 2026
can be anticipated to resolve any downwind nonattainment or maintenance problems. See
Appendix 3B for additional discussion of the estimated improvements in downwind air quality
for each of the regulatory control alternatives analyzed in this RIA, as well as data on average
and maximum design value changes at downwind receptors.
4.6 Social Costs
As discussed in EPA's Guidelines for Preparing Economic Analyses, social costs are the
total economic burden of a regulatory action (USEPA, 2010). This burden is the sum of all
opportunity costs incurred due to the regulatory action, where an opportunity cost is the value
lost to society of any goods and services that will not be produced and consumed as a result of
reallocating some resources towards pollution mitigation. Estimates of social costs may be
compared to the social benefits expected as a result of a regulation to assess its net impact on
society.
4-47
-------
The social costs of this regulatory action will not necessarily be equal to the expenditures
by the electricity sector and other affected industries to comply with the proposed rule.
Nonetheless, here we use total national compliance costs for EGUs and non-EGUs as a proxy for
social costs. Table 4-20 above presents the total annual estimated compliance costs for EGUs for
2023 and EGUs and non-EGUs for 2026-2042.
The compliance cost estimates for EGUs in the proposed and more or less stringent
regulatory control alternatives presented above are the change in expenditures by the electricity
generating sector required by the power sector for compliance under each alternative. The
change in the expenditures required by the power sector to achieve and maintain compliance
reflect the changes in electricity production costs resulting from application of NOx control
strategies necessary to comply with the emissions budgets and the backstop emissions limits. The
production cost changes included changes in fuel expenditures.
Ultimately, depending on the market structure and the demand and supply price
elasticities for electricity, some compliance costs may be borne by electricity consumers through
higher electricity prices. Furthermore, the share of compliance costs ultimately borne by owners
of electricity generating capacity and other capital may be borne unevenly, with some firms
becoming more profitable as a result of the regulation. These asset owners and electricity
consumers include U.S. citizens and residents as well as non-residents (e.g., foreign owners of
electricity-consuming commercial enterprises). For additional discussion of impacts on fuel use
and electricity prices, see Section 4.5.3 above.
The compliance cost estimates for non-EGUs in the proposed and more or less stringent
regulatory control alternatives are the change in expenditures by the industries required for
compliance under each alternative. The change in the expenditures required by the industries to
maintain compliance reflect the changes in production costs resulting from application of NOx
control technologies or measures. As in the power sector, ultimately, depending on market
structure and the demand and supply price elasticities for these industrial products, some part of
the compliance costs may be borne by consumers through higher prices, and these costs are
distributed among U.S. citizens and residents and foreign asset owners.
4-48
-------
For non-EGUs the estimated compliance costs in Table 4-20 are derived from
engineering cost estimates, and for EGUs the estimated compliance costs are generated using the
Integrated Planning Model (IPM). IPM solves for the least-cost approach to meet new regulatory
requirements in the electricity sector with highly detailed information on electricity generation
and air pollution control technologies and primary energy sector market conditions (coal and
natural gas) while meeting fixed electricity demands, regulatory requirements, and other
constraints. However, potential effects outside of the electricity, coal and natural gas sectors are
not evaluated within IPM.
Changes in production in a directly regulated sector may have indirect effects on a
myriad of other markets when output from that sector - for this proposal electricity and certain
industrial products - is used as an input in the production of many other goods. It may also affect
upstream industries that supply goods and services to the sector, along with labor and capital
markets, as these suppliers alter production processes in response to changes in factor prices. In
addition, households may change their demand for particular goods and services due to changes
in the price of electricity and other final goods prices, such as cement.
When new regulatory requirements are expected to result in effects outside of regulated
and closely related sectors, a key challenge is determining whether they are of sufficient
magnitude to warrant explicit evaluation (Hahn and Hird 1990). It is not possible to estimate the
magnitude and direction of these potential effects outside of the regulated sector(s) without an
economy-wide modeling approach. For example, studies of air pollution regulations for the
power sector have found that the social costs and benefits may be greater or lower than when
secondary market impacts are taken into account, and that the direction of the estimates may
depend on the form of the regulation (e.g., Goulder et al. 1999, Williams 2002, Goulder et al.
2016).
Economy-wide models - and, more specifically, computable general equilibrium (CGE)
models - are analytical tools that can be used to evaluate the broad impacts of a regulatory action.
A CGE-based approach to cost estimation concurrently considers the effect of a regulation across
all sectors in the economy. It is structured around the assumption that, for some discrete period
of time, an economy can be characterized by a set of equilibrium conditions in which supply
4-49
-------
equals demand in all markets. When the imposition of a regulation alters conditions in one
market, a general equilibrium approach will determine a new set of prices for all markets that
will return the economy to equilibrium. These prices in turn determine the outputs and
consumption of goods and services in the new equilibrium. In addition, a new set of prices and
demands for the factors of production (labor, capital, and land), the returns to which compose the
income of businesses and households, will be determined in general equilibrium. The social cost
of the regulation can then be estimated by comparing the value of variables in the pre-regulation
"baseline" equilibrium with those in the post-regulation, simulated equilibrium.
In 2015, the EPA established a Science Advisory Board (SAB) panel to consider the
technical merits and challenges of using economy-wide models to evaluate costs, benefits, and
economic impacts in regulatory development. In its final report (U.S. EPA 2017), the SAB
recommended that the EPA begin to integrate CGE modeling into regulatory analysis to offer a
more comprehensive assessment of the effects of air regulations. The SAB noted that CGE
models can provide insight into the likely social costs of a regulation even when they do not
include a characterization of the likely social benefits of the regulation. CGE models may also
offer insights into the ways costs are distributed across regions, sectors, or households.
The SAB also noted that the case for using CGE models to evaluate a regulation's effects
is strongest when the costs of compliance are expected to be large in magnitude and the sector
has strong linkages to the rest of the economy. The report also noted that the extent to which
CGE models add value to the analysis depends on data availability. CGE models provide
aggregated representations of the entire economy and are designed to capture substitution
possibilities between production, consumption, and trade; interactions between economic
sectors; and interactions between a policy shock and pre-existing distortions, such as taxes.
However, one also needs to adequately represent a regulation in the model to estimate its effects.
In response to the SAB's recommendations, the EPA built a new CGE model called
SAGE. A second SAB panel performed a peer review of SAGE, and the reviewed concluded in
2020.49 While EPA now has a peer reviewed CGE model for analyzing the potential economy-
49 See U.S. EPA (2020). The model peer review and other SAB reports can be downloaded at:
https://sab.epa.gov/ords/sab/f?p=100:12:15036376991605::: 12::
4-50
-------
wide effects of regulations, we have not used the model in the RIA for this proposal due to the
expedited proposed rulemaking timeline. However, EPA continues to be committed to the use of
CGE models to evaluate the economy-wide effects of its regulations.
4.7 Limitations
EPA's modeling is based on expert judgment of various input assumptions for variables
whose outcomes are uncertain. As a general matter, the Agency reviews the best available
information from engineering studies of air pollution controls and new capacity construction
costs to support a reasonable modeling framework for analyzing the cost, emission changes, and
other impacts of regulatory actions for EGUs. The annualized cost of the proposed rule for
EGUs, as quantified here, is EPA's best assessment of the cost of implementing the proposal on
the power sector. These costs are generated from rigorous economic modeling of changes in the
power sector due to implementation of the proposed FIP for the 2015 ozone NAAQS.
The IPM-projected annualized cost estimates of private compliance costs provided in this
analysis are meant to show the increase in production (generating) costs to the power sector in
response to the proposal. To estimate these annualized costs, as discussed earlier in this chapter,
EPA uses a conventional and widely accepted approach that applies a capital recovery factor
(CRF) multiplier to capital investments and adds that to the annual incremental operating
expenses to calculate annual costs. The CRF is derived from estimates of the cost of capital
(private discount rate), the amount of insurance coverage required, local property taxes, and the
life of capital. The private compliance costs presented earlier are EPA's best estimate of the
direct private compliance costs of the proposal.
In addition, there are several key areas of uncertainty related to the electric power sector
that are worth noting, including:
• Electric demand: The analysis includes an assumption for future electric demand. To the
extent electric demand is higher and lower, it may increase/decrease the projected future
composition of the fleet.
• Natural gas supply and demand: Large increases in supply over the last few years, and
relatively low prices, are represented in the analysis. To the extent prices are higher or lower, it
4-51
-------
would influence the use of natural gas for electricity generation and overall competitiveness of
other EGUs (e.g., coal and nuclear units).
• Longer-term planning by utilities: Many utilities have announced long-term clean energy
and/or climate commitments, with a phasing out of large amounts of coal capacity by 2030 and
continuing through- 2050. These announcements, some of which are not legally binding, are not
necessarily reflected in the baseline, and may alter the amount of coal capacity projected in the
baseline that would be covered under this proposed rule.
These are key uncertainties that may affect the overall composition of electric power
generation fleet and could thus have an effect on the estimated costs and impacts of this action.
However, these uncertainties would affect the modeling of the baseline and illustrative policy
scenarios similarly, and therefore the impact on the incremental projections (reflecting the
potential costs/benefits of the illustrative policy scenario) would be more limited and are not
likely to result in notable changes to the assessment of the proposed FIP for the 2015 ozone
NAAQS found in this chapter. While it is important to recognize these key areas of uncertainty,
they do not change the EPA's overall confidence in the estimated impacts of the illustrative
policy scenario presented in this chapter. The EPA continues to monitor industry developments
and makes appropriate updates to the modeling platforms in order to reflect the best and most
current data available.
While the baseline includes modeling to capture the recently finalized 2020 effluent
Limitation Guidelines (ELG), it does not incorporate information provided by owners of affected
facilities to state permitting authorities in October 2021 that indicate their likely compliance
pathway, including retirement by 2028. Potential future incorporation of this information may
result in additional coal plant retirements relative to the baseline scenario, which would - all else
equal - reduce the modeled costs and benefits of the proposed rule depending on the extent that
these retirements occur before compliance deadlines for this action. Similarly, the baseline
accounts for the effect of expected compliance methods for the 2020 CCR Rule. However, plants
may adopt compliance methods that are different than those represented in the baseline.
As discussed in section 4.3.2, IPM v.6.20 does not have the capacity to endogenously
determine whether or not to maximize the use of existing EGU post-combustion NOx controls
4-52
-------
(i.e., SCR), or install/upgrade combustion controls in response to a regulatory control
requirement. These decisions were imposed exogenously on the model, as documented in section
4.3.2. While the emissions projections reflect operation of these controls, the projected
compliance costs were supplemented with exogenously estimated costs of optimizing SCR
operation, optimizing SNCR operation, and installing/upgrading combustion controls (see
section 4.3.3). As a result of this modeling approach, the dispatch decisions made within the
model do not take into consideration the additional operating costs associated with these three
types of compliance strategies (the operating costs of the units on which these strategies are
imposed do not reflect the additional costs of these strategies). The effect of changes in facility
and system-wide emissions from these changes in operating costs are also not accounted for in
the air quality modeling for the regulatory alternatives described in Chapter 3.
The impacts of the Later Model Year Light-Duty Vehicle GHG Emissions Standards50 is
not captured in the baseline. This rule is projected to increase the total demand for electricity by
0.5% in 2030 and 1% in 2040 relative to 2020 levels.51 This translates into a 0.4% increase in
electricity demand in 2030 and a 0.8% increase in electricity demand in 2040 relative to the
baseline electricity demand projections assumed in this analysis. The impact of the Proposed
Standards of Performance for New, Reconstructed, and Modified Sources and Emissions
Guidelines for Existing Sources: Oil and Natural Gas Sector Climate Review52 are also not
included in this analysis. Inclusion of these standards would likely increase the price of natural
gas modestly as a result of limitations on the usage of reciprocating internal combustion engines
in the pipeline transportation of natural gas. All else equal inclusion of these two programs
would likely result in a modest increase in the total cost of compliance for this rule.
Lastly, EPA used estimated emissions reductions and costs from the non-EGU screening
assessment in this RIA as a proxy for the least-cost compliance strategy for complying with the
emissions limits proposed for the non-EGU industries. In the screening assessment, which is
available in the docket for this proposed rulemaking, EPA used CoST to identify emissions units,
50 Available at: https://www.federalregister.gov/documents/2021/08/10/2021-16582/revised-2023-and-later-model-
year-light-duty-vehicle-greenhouse-gas-emissions-standards
51 Regulatory Impact Analysis available at: https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P1012ONB.pdf
52 Available at: https://www.federalregister.gov/documents/2021/ll/15/2021-24202/standards-of-performance-for-
new-reconstructed-and-modified-sources-and-emissions-guidelines-for
4-53
-------
emissions reductions, and associated compliance costs; CoST is designed to be used for
illustrative control strategy analyses (e.g., NAAQS regulatory impact analyses) and not for unit-
specific, detailed engineering analyses. The estimates from CoST identify proxy values for (1)
non-EGU emissions units that have emissions reduction potential, (2) potential controls for and
emissions reductions from these emissions units, and (3) control costs from the potential controls
on these emissions units. The control cost estimates assume an average level of retrofit difficulty
for control applications, and do not include monitoring, recordkeeping, reporting, or testing
costs. This screening assessment is not intended to be, nor take the place of, a unit-specific
detailed engineering analysis that fully evaluates the feasibility of retrofits for the emissions
units, potential controls, and related costs. It is not possible to determine whether this approach
leads to an over or underestimate of the costs, and consequent NOx and other pollutant emissions
changes, benefits, and other impacts, including the effect on downwind receptors, of the
proposed rule and the analyzed alternatives. This is because we did not directly evaluate the
emissions reductions that would be achieved at the emissions units included in the proposal using
their baseline emissions and emissions rates, their emission limits, and their likely compliance
strategy. Also, we did not project the potential changes in the number of existing and new units
resulting from industry growth or capital turnover, nor whether the emissions limitations would
require further NOx emissions reductions at new units relative to what is required of them in the
baseline.
4.8 References
Goulder, L., Parry, I., Williams, R., and Burtraw, D. 1999. "The Cost-Effectiveness of
Alternative Instruments for Environmental Protection in a Second-Best Setting." Journal
of Public Economics, 72(3): 329-360. 144
Goulder, L., M. Hafstead, and R. Williams III 2016. "General Equilibrium Impacts of a Federal
Clean Energy Standard." American Economic Journal - Economic Policy 8(2): 186-218.
Hahn, R., and J. Hird. 1990. "The Costs and Benefits of Regulation: Review and Synthesis."
Yale Journal of Regulation 8: 233-278.
U.S. Energy Information Administration (EIA). 2020. The Electricity Market Module of the
National Energy Modeling System: Model Documentation 2020. Available at: <
https://www. eia.gov/outlooks/aeo/nems/documentati on/electricity/pdf/m068(2020).pdf^.
Accessed 2/10/2021.
4-54
-------
U.S. EPA. 2020. Technical Review of EPA's Computable General Equilibrium Model, SAGE.
EPA-SAB-20-010.
U.S. EPA. 2017. SAB Advice on the Use of Economy-Wide Models in Evaluating the Social
Costs, Benefits, and Economic Impacts of Air Regulations. EPA-SAB-17-012.
U.S. EPA, 2015. Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric
Utility Generating Units (FinalRule), http://www2.epa.gov/cleanpowerplan/clean-
power-plan-existing-power-plants.
U.S. EPA, 2015 a. Standards of Performance for Greenhouse Gas Emissions from New,
Modified, and Reconstructed Stationary Sources: Electric Utility Generating Units (Final
Rule), http://www2.epa.gov/cleanpowerplan/carbon-pollution-standards-new-modified-
and-reconstructed-power-plants.
U.S. EPA, 2015b. Disposal of Coal Combustion Residuals from Electric Utilities (Final Rule),
http://www2.epa.gov/coalash/coal-ash-rule.
U.S. EPA, 2015c. Steam Electric Power Generating Effluent Guidelines (Final Rule),
http://www2.epa.gov/eg/steam-electric-power-generating-effluent-guidelines-2015-final-
rule.
U.S. EPA, 2014. Final Rule for Existing Power Plants and Factories,
http://www2.epa.gov/cooling-water-intakes.
U.S. EPA, 2011. Cross-State Air Pollution Rule,
http://www3.epa.gov/airtransport/CSAPR/index.html
U.S. EPA, 201 la. Mercury and Air Toxics Standards (MATS), http://www3.epa.gov/mats/.
U.S. EPA. 2010. EPA Guidelines for Preparing Economic Analyses. Available at:
. Accessed 9/21/2015.
U.S. EPA. 2010a. Regulatory Impact Analysis for the Proposed Federal Transport Rule
Analyses. Available at: < http://www3.epa.gov/ttnecasl/ria.html>. Accessed 9/21/2015.
U.S. EPA, 2005. Clean Air Interstate Rule,
http://archive.epa.gov/airmarkets/programs/cair/web/html/index.html.
Williams III, R. 2002. "Environmental tax interactions when pollution affects health or
productivity." Journal of Environmental Economics and Management 44(2): 261-270.
4-55
-------
CHAPTER 5: BENEFITS
Overview
This proposed Federal Implementation Plan (FIP) Addressing Regional Ozone Transport
for the 2015 Ozone National Ambient Air Quality Standards (FIP for the 2015 ozone NAAQS) is
expected to reduce emissions of nitrogen oxides (NOx) transported from states that contribute
significantly to nonattainment or interfere with maintenance of the 2015 ozone National Ambient
Air Quality Standards (NAAQS) in downwind states. Implementing the proposed FIP for the
2015 ozone NAAQS_is expected to reduce emissions of NOx, which will in turn reduce
concentrations of ground-level ozone and fine particles (PM2.5); the proposed rule is also
projected to reduce sulfur dioxide (SO2), direct PM2.5 emissions, carbon dioxide (CO2) emissions
as well as water effluents, and potentially reduce mercury (Hg) emissions. This chapter reports
the estimated monetized health benefits from reducing concentrations of ozone and PM2.5 for
each of three regulatory control alternatives described in prior chapters.1 Though the proposed
rule is likely to also yield positive benefits associated with reducing pollutants other than ozone
and PM2.5, limited time, resource and data limitations prevented us from characterizing the value
of those reductions.
This chapter describes the methods used to estimate the benefits to human health of
reducing concentrations of ozone from affected EGUs (electrical generating units) and non-
EGUs (non-electric generating units, or other stationary source emissions sources) and PM2.5
from affected EGUs. This analysis uses both full-form and reduced-form techniques to quantify
benefits. Both approaches rely on the same methods for quantifying the number and value of air
pollution-attributable premature deaths and illnesses, which is described in the TSD for the Final
Revised CSAPR Update for the 2008 Ozone NAAQS titled Estimating PM2.5- and Ozone-
Attributable Health Benefits. Methods used to estimate PM2.5 benefits are described in the TSD
1 A comprehensive approach to benefit-cost analysis (BCA) is required to assess whether it is conceivable for those
who experience a net gain from a regulatory action to potentially compensate those who experience a net loss. As
such, a BCA should aim to evaluate all benefits and costs resulting from the regulation, which includes welfare
effects from all changes in externalities due to changes in environmental contaminants as well as any other
externalities. This requires evaluating changes in pollutant concentrations induced beyond the contaminant(s)
targeted by the action.
5-1
-------
titled Estimating the Benefit per Ton of Reducing Directly-Emitted PM2.5, PM2.5 Precursors
and Ozone Precursors from 21 Sectors.
When estimating the value of improved air quality over a multi-year time horizon, the
ozone analysis applies population growth and income growth projections for each future year
through 2042 and estimates of baseline mortality incidence rates at five-year increments.
However, due to additional uncertainties associated with baseline air quality projections beyond
2026, annual health benefits beyond 2026 presented in Tables 5-6 and 5-7 are based on 2026 air
quality changes. Additionally, within each 12 km grid cell we assumed the 2023 ozone
concentration change until 2025 and the 2026 ozone concentration change until 2042. As we do
not account fully for changes in the size or distribution of the population beyond the year 2026,
and the changes in the level and location of NOx emissions attributable to this proposal; this
artifact may introduce uncertainty to the ozone analysis and is described below in Section 5.1.3.
When estimating the value of improved air quality over a multi-year time horizon, the PM2.5
analysis applies benefit per ton estimates from 2025 for 2023-2029 and 2030 for 2030-2042,
which also introduces uncertainty.
Data, resource, and methodological limitations prevent EPA from monetizing health
benefits of reducing direct exposure to NO2 and SO2, ecosystem effects and visibility impairment
associated with these pollutants, ozone and PM2.5, as well as benefits from reductions in other
pollutants, such as water effluents. We qualitatively discuss these unquantified benefits in this
chapter.
5.1 Estimated Human Health Benefits
The proposed rule is expected to reduce ozone season and annual NOx emissions. In the
presence of sunlight, NOx and volatile organic compounds (VOCs) can undergo a chemical
reaction in the atmosphere to form ozone. Reducing NOx emissions generally reduces human
exposure to ozone and the incidence of ozone-related health effects, though the degree to which
ozone is reduced will depend in part on local concentration levels of VOCs. In addition to NOx,
the proposed rule is also expected to reduce emissions of direct PM2.5 and SO2 throughout the
year. Because NOx and SO2 are also precursors to secondary formation of ambient PM2.5,
reducing these emissions would reduce human exposure to ambient PM2.5 throughout the year
and would reduce the incidence of PIvfc.s-attributable health effects.
5-2
-------
In this proposed FIP for the 2015 ozone NAAQS regulatory impact analysis (RIA), as
discussed above, EPA uses both full-form and reduced-form techniques to quantify benefits of
changes in PM2.5 and ozone concentrations. In particular, both methods incorporate evidence
reported in the most recent completed PM and Ozone Integrated Science Assessments (ISAs)
and accounts for recommendations from the Science Advisory Board (U.S. EPA 2019a, U.S.
EPA 2020b, U.S. EPA-SAB 2019, U.S. EPA-SAB 2020a). When updating each health endpoint
EPA considered: (1) the extent to which there exists a causal relationship between that pollutant
and the adverse effect; (2) whether suitable epidemiologic studies exist to support quantifying
health impacts; (3) and whether robust economic approaches are available for estimating the
value of the impact of reducing human exposure to the pollutant. Our approach for updating the
endpoints and to identify suitable epidemiologic studies, baseline incidence rates, population
demographics, and valuation estimates is summarized below. Detailed descriptions of these
updates are available in the TSD for the Final Revised Cross-State Air Pollution Rule Update for
the 2008 Ozone NAAQS titled Estimating PM2.5- and Ozone-Attributable Health Benefits.
The Estimating PM2.5- and Ozone-Attributable Health Benefits TSD describes fully the
Agency's approach for quantifying the number and value of estimated air pollution-related
impacts. In this document the reader can find the rationale for selecting health endpoints to
quantify; the demographic, health and economic data used; modeling assumptions; and our
techniques for quantifying uncertainty.2
As structured, the proposed rule would affect the distribution of ozone and PM2.5
concentrations in much of the U.S.; this includes locations both meeting and exceeding the
NAAQS for ozone and particulate matter (PM). This RIA estimates avoided ozone- and PM2.5-
related health impacts that are distinct from those reported in the RIAs for both ozone and PM
NAAQS (U.S. EPA 2012, 2015e). The ozone and PM NAAQS RIAs illustrate, but do not
predict, the benefits and costs of strategies that States may choose to enact when implementing a
revised NAAQS; these costs and benefits are illustrative and cannot be added to the costs and
benefits of policies that prescribe specific emission control measures. This RIA estimates the
benefits (and costs) of specific emissions control measures. As shown and described in Chapter
2 The analysis was completed using BenMAP-CE version 1.5.8, which is a variant of the current publicly available
version.
5-3
-------
3, we project most levels of ozone to decrease, primarily in and around the 26 affected states.3
The ozone-related benefit estimates are based on these modeled changes in summer season
average ozone concentrations We also estimate benefits from EGU PM2.5 emissions changes
using a benefit per ton approach, which is described more fully in Sections 5.1.1.4 and 5.1.1.5.
For non-EGU NOx emissions changes., since the proposed FIP for the 2015 ozone NAAQS
includes ozone season emissions limits for the non-EGU emissions units. As we do not know if
all affected sources will run controls year-round or only during ozone season, we also provide an
illustration of potential PM2.5 benefits that could accrue from non-EGUs if the proposed controls
are run year-round. These illustrative PM2.5 benefits are not added to the total benefits for this
proposal.
5.1.1 Health Impact Assessment for Ozone and PM2.5
The benefits analysis presented in this chapter incorporates science-policy and technical
changes that the Agency adopted and documented in the benefits chapter of the RIA
accompanying the final Revised CSAPR Update, based on the 2019 and 2020 PM and ozone
ISAs (U.S. EPA, 2020c).
Estimating the health benefits of reductions in PM2.5 and ozone3 exposure begins with
estimating the change in exposure for each individual and then estimating the change in each
individual's risks for those health outcomes affected by exposure. The benefit of the reduction in
each health risk is based on the exposed individual's willingness to pay (WTP) for the risk
change, assuming that each outcome is independent of one another. The greater the magnitude of
the risk reduction from a given change in concentration, the greater the individual's WTP, all
else equal. The social benefit of the change in health risks equals the sum of the individual WTP
estimates across all of the affected individuals residing in the U.S.4 We conduct this analysis by
adapting primary research—specifically, air pollution epidemiology studies and economic value
studies—from similar contexts. This approach is sometimes referred to as "benefits transfer."
Below we describe the procedure we follow for: (1) selecting air pollution health endpoints to
3 In a small number of areas in the northwest, we project ozone to increase slightly compared to the baseline.
4 This RIA also reports the change in the sum of the risk, or the change in the total incidence, of a health outcome
across the population. If the benefit per unit of risk is invariant across individuals, the total expected change in the
incidence of the health outcome across the population can be multiplied by the benefit per unit of risk to estimate the
social benefit of the total expected change in the incidence of the health outcome.
5-4
-------
quantify; (2) calculating counts of air pollution effects using a health impact function; (3)
specifying the health impact function with concentration-response parameters drawn from the
epidemiological literature.
5.1.1.1 Selecting Air Pollution Health Endpoints to Quantify
As a first step in quantifying ozone and PM2.5-related human health impacts, the Agency
consults the Integrated Science Assessment for Ozone and Related Photochemical Oxidants
(Ozone ISA) (U.S. EPA 2020b) and the Integrated Science Assessment for Particulate Matter
(PM ISA) (U.S. EPA 2019a). These two documents synthesize the toxicological, clinical and
epidemiological evidence to determine whether each pollutant is causally related to an array of
adverse human health outcomes associated with either acute (i.e., hours or days-long) or chronic
(i.e., years-long) exposure; for each outcome, the ISA reports this relationship to be causal, likely
to be causal, suggestive of a causal relationship, inadequate to infer a causal relationship or not
likely to be a causal relationship. The Agency estimates the incidence of air pollution effects for
those health endpoints above where the ISA classified as either causal or likely-to-be-causal.
In brief, the ISA for ozone found short-term (less than one month) exposures to ozone to
be causally related to respiratory effects, a "likely to be causal" relationship with metabolic
effects and a "suggestive of, but not sufficient to infer, a causal relationship" for central nervous
system effects, cardiovascular effects, and total mortality. The ISA reported that long-term
exposures (one month or longer) to ozone are "likely to be causal" for respiratory effects
including respiratory mortality, and a "suggestive of, but not sufficient to infer, a causal
relationship" for cardiovascular effects, reproductive effects, central nervous system effects,
metabolic effects, and total mortality. The PM ISA found short-term exposure to PM2.5 to be
causally related to cardiovascular effects and mortality (i.e., premature death), respiratory effects
as likely-to-be-causally related, and a suggestive relationship for metabolic effects and nervous
system effects. The ISA identified cardiovascular effects and total mortality as being causally
related to long-term exposure to PM2.5. A likely-to-be-causal relationship was determined
between long-term PM2.5 exposures and respiratory effects, nervous system effects, and cancer
effects; and the evidence was suggestive of a causal relationship for male and female
reproduction and fertility effects, pregnancy and birth outcomes, and metabolic effects.
5-5
-------
Table 5-1 reports the ozone and PIVfo.s-related human health impacts effects we quantified
and those we did not quantify in this RIA. The list of benefit categories not quantified is not
exhaustive. And, among the effects quantified, it might not have been possible to quantify
completely either the full range of human health impacts or economic values. Section 5.3 and
Table 5-9 below report other omitted health and environmental benefits expected from the
emissions and effluent changes as a result of this proposal, such as health effects associated with
NO2 and SO2, and any welfare effects such as acidification and nutrient enrichment.
Specifically, for ozone-related benefits, for EGUs and non-EGUs we conducted a full health
benefits analysis that includes premature deaths and illnesses attributable to photochemical
modeled changes in summer season average ozone concentrations for the years 2023 and 2026.
For PM-related benefits for EGUs, benefit per ton values were used to estimate the benefits from
changes in PM2.5 concentrations from changes in NOx, SO2 and PM2.5 emissions. For PM-related
benefits for non-EGUs, due to uncertainty in whether affected sources will run controls year-
round or only during ozone season, benefit per ton values were used to estimate the benefits from
changes in NOx emissions to illustrate the potential PM2.5 benefits from non-EGUs if the
proposed controls are run year-round. The illustrative non-EGU PM benefits estimates are not
added to the total benefits for this proposal.
Consistent with economic theory, the WTP for reductions in exposure to environmental
hazard will depend on the expected impact of those reductions on human health and other
outcomes. All else equal, WTP is expected to be higher when there is stronger evidence of a
causal relationship between exposure to the contaminant and changes in a health outcome
(McGartland et al., 2017). For example, in the case where there is no evidence of a potential
relationship the WTP would be expected to be zero and the effect should be excluded from the
analysis. Alternatively, when there is some evidence of a relationship between exposure and the
health outcome, but that evidence is insufficient to definitively conclude that there is a causal
relationship, individuals may have a positive WTP for a reduction in exposure to that hazard
(U.S. EPA-SAB 2020b, Kivi and Shogren, 2010). Lastly, the WTP for reductions in exposure to
pollutants with strong evidence of a relationship between exposure and effect are likely positive
and larger than for endpoints where evidence is weak, all else equal. Unfortunately, the
economic literature currently lacks a settled approach for accounting for how WTP may vary
with uncertainty about causal relationships.
5-6
-------
Given this challenge, the Agency draws its assessment of the strength of evidence on the
relationship between exposure to PM2.5 or ozone and potential health endpoints from the ISAs
that are developed for the NAAQS process as discussed above. The focus on categories
identified as having a "causal" or "likely to be causal" relationship with the pollutant of interest
is to estimate the pollutant-attributable human health benefits in which we are most confident.5
All else equal, this approach may underestimate the benefits of PM2.5 and ozone exposure
reductions as individuals may be WTP to avoid specific risks where the evidence is insufficient
to conclude they are "likely to be caus[ed]" by exposure to these pollutants.6 At the same time,
WTP may be lower for those health outcomes for which causality has not been definitively
established. This approach treats relationships with ISA causality determinations of "likely to be
causal" as if they were known to be causal, and therefore benefits could be overestimated.
5 This decision criterion for selecting health effects to quantify and monetize PM2 5 and ozone is only applicable to
estimating the benefits of exposure of these two pollutants. This is also the approach used for identifying the
unqualified benefit categories for criteria pollutants. This decision criterion may not be applicable or suitable for
quantifying and monetizing health and ecological effects of other pollutants. The approach used to determine
whether there is sufficient evidence of a relationship between an endpoint affected by non-criteria pollutants, and
consequently a positive WTP for reductions in those pollutants, for other unqualified benefits described in this
chapter can be found in the source documentation for each of these pollutants (see relevant sections below). The
conceptual framework for estimating benefits when there is uncertainty in the causal relationship between a hazard
and the endpoints it potentially affects described here applies to these other pollutants.
6 EPA includes risk estimates for an example health endpoint with a causality determination of "suggestive, but not
sufficient to infer" that is associated with a potentially substantial economic value in the quantitative uncertainty
characterization (Estimating PM2.5- and Ozone-Attributable Health Benefits TSD section 6.2.3).
5-7
-------
Table 5-1. Health Effects of Ambient Ozone and PM2.5
Category
Effect
Effect
Quantified
Effect
Monetized
More
Information
Premature
mortality from
Adult premature mortality based on cohort study
estimates and expert elicitation estimates (age 65-99
or age 30-99)
~
PM ISA
exposure to PM2 5
Infant mortality (age <1)
~
~
PM ISA
Heart attacks (age >18)
~
¦Z1
PM ISA
Hospital admissions—cardiovascular (ages 65-99)
~
~
PM ISA
Emergency department visits— cardiovascular (age
0-99)
V
~
PM ISA
Hospital admissions—respiratory (ages 0-18 and 65-
99)
Emergency room visits—respiratory (all ages)
Cardiac arrest (ages 0-99; excludes initial hospital
and/or emergency department visits)
Stroke (ages 65-99)
Asthma onset (ages 0-17)
S
PM ISA
~
~
~
7
~
¦Z1
7'J
~
PM ISA
PM ISA
PM ISA
" PM ISA
Asthma symptoms/exacerbation (6-17)
s
~
PM ISA
Nonfatal
morbidity from
Lung cancer (ages 30-99)
Allergic rhinitis (hay fever) symptoms (ages 3-17)
Lost work days (age 18-65)
s
s
7
~
~
~
PM ISA
PM ISA
PM ISA
exposure to PM2 5
Minor restricted-activity days (age 18-65)
s
~
PM ISA
Hospital admissions—Alzheimer's disease (ages 65-
99)
s
PM ISA
Hospital admissions—Parkinson's disease (ages 65-
99)
~
PM ISA
Other cardiovascular effects (e.g., other ages)
—
—
PM ISA2
Other respiratory effects (e.g., pulmonary function,
non-asthma ER visits, non-bronchitis chronic
diseases, other ages and populations)
—
—
PM ISA2
Other nervous system effects (e.g., autism, cognitive
decline, dementia)
—
—
PM ISA2
Metabolic effects (e.g., diabetes)
—
—
PM ISA2
Reproductive and developmental effects (e.g., low
birth weight, pre-term births, etc.)
—
—
PM ISA2
Cancer, mutagenicity, and genotoxicity effects
—
—
PM ISA2
Mortality from
Premature respiratory mortality based on short-term
study estimates (0-99)
~
~
Ozone ISA
exposure to ozone
Premature respiratory mortality based on long-term
study estimates (age 30-99)
~
~
Ozone ISA
Hospital admissions—respiratory (ages 0-99)
~
s
Ozone ISA
Emergency department visits—respiratory (ages 0-
99)
~
s
Ozone ISA
Nonfatal
morbidity from
exposure to ozone
Asthma onset (0-17)
Asthma symptoms/exacerbation (asthmatics age 2-
17]
Allergic rhinitis (hay fever) symptoms (ages 3-17)
Minor restricted-activity days (age 18-65)
~
~
~
~
¦/
~
~
V
O/.onc ISA
Ozone ISA
O/.onc ISA
Ozone ISA
School absence days (age 5-17)
~
V
Ozone ISA
Decreased outdoor worker productivity (age 18-65)
Metabolic effects (e.g., diabetes)
Other respiratory effects (e.g., premature aging of
lungs)
—
—
Ozone ISA2
O/.onc ISA2
Ozone ISA2
5-8
-------
Cardiovascular and nervous system effects — — Ozone ISA2
Reproductive and developmental effects — — Ozone ISA2
'Valuation estimate excludes initial hospital and/or emergency department visits.
2 Not quantified due to data availability limitations and/or because current evidence is only suggestive of causality.
5.1.1.2 Calculating Counts of Air Pollution Effects Using the Health Impact Function
We use EPA's Benefits Mapping and Analysis Program - Community Edition (BenMAP-
CE) to quantify counts of premature deaths and illnesses attributable to photochemical modeled
changes in summer season average ozone concentrations for the years 2023 and 2026 using
health impact functions. The program is also used to estimate the benefit per ton values that are
used to estimate the benefits from changes in PM2.5 concentrations from changes in NOx, SO2
and PM2.5 emissions.
BenMAP quantifies counts of attributable effects using a health impact function, which
combines information regarding the: concentration-response relationship between air quality
changes and the risk of a given adverse outcome; population exposed to the air quality change;
baseline rate of death or disease in that population; and air pollution concentration to which the
population is exposed.
The following provides an example of a health impact function, in this case for PM2.5
mortality risk. We estimate counts of PIVfo.s-related total deaths (yij) during each year i (i=1,... ,1
where I is the total number of years analyzed) among adults aged 30 and older (a) in each county
in the contiguous U.S. j (j=l,.. ,,J where J is the total number of counties) as
yij 2a yija
yija = moija x(ep ACij-l) x pijaj Eq[l]
where moija is the baseline all-cause mortality rate for adults aged a=30-99 in county j in year i
stratified in 10-year age groups, P is the risk coefficient for all-cause mortality for adults
associated with annual average PM2.5 exposure, Cij is the annual mean PM2.5 concentration in
5-9
-------
county j in year i, and Pija is the number of county adult residents aged a=30-99 in county j in
year i stratified into 5-year age groups.7
The BenMAP-CE tool is pre-loaded with projected population from the Woods & Poole
company; cause-specific and age-stratified death rates from the Centers for Disease Control and
Prevention, projected to future years; recent-year baseline rates of hospital admissions,
emergency department visits and other morbidity outcomes from the Healthcare Cost and
Utilization Program and other sources; concentration-response parameters from the published
epidemiologic literature cited in the Integrated Science Assessments for fine particles and
ground-level ozone; and, cost of illness or willingness to pay economic unit values for each
endpoint. Changes in ozone concentrations are taken from the air pollution spatial surfaces for
the analytic years 2023 and 2026 described in Chapter 3. The air pollution spatial surfaces used
to estimate the PM2.5 benefit-per-ton values are described below in Section 5.1.1.4.
5.1.1.3 Quantifying Cases of Ozone-Attributable Premature Death
Mortality risk reductions account for the majority of monetized ozone-related and PM2.5-
related benefits. For this reason, this subsection and the following provide a brief background of
the scientific assessments that underly the quantification of these mortality risks and identifies
the risk studies used to quantify them in this RIA, for ozone and PM2.5 respectively. As noted
above, the Estimating PM2.5- and Ozone-Attributable Health Benefits TSD describes fully the
Agency's approach for quantifying the number and value of ozone and PM2.5 air pollution-
related impacts, including additional discussion of how the Agency selected the risk studies used
to quantify them in this RIA. The TSD also includes additional discussion of the assessments that
support quantification of these mortality risk than provide here.
In 2008, the National Academies of Science (NRC 2008) issued a series of
recommendations to EPA regarding the procedure for quantifying and valuing ozone-related
mortality due to short-term exposures. Chief among these was that "... short-term exposure to
ambient ozone is likely to contribute to premature deaths" and the committee recommended that
"ozone-related mortality be included in future estimates of the health benefits of reducing ozone
7 In this illustrative example, the air quality is resolved at the county level. For this RIA, we simulate air quality
concentrations at 12km by 12km grids. The BenMAP-CE tool assigns the rates of baseline death and disease stored
at the county level to the 12km by 12km grid cells using an area-weighted algorithm. This approach is described in
greater detail in the appendices to the BenMAP-CE user manual.
5-10
-------
exposures..The NAS also recommended that. .the greatest emphasis be placed on the
multicity and [National Mortality and Morbidity Air Pollution Studies (NMMAPS)] ... studies
without exclusion of the meta-analyses" (NRC 2008). Prior to the 2015 Ozone NAAQS RIA, the
Agency estimated ozone-attributable premature deaths using an NMMAPS-based analysis of
total mortality (Bell et al. 2004), two multi-city studies of cardiopulmonary and total mortality
(Huang et al. 2004; Schwartz 2005) and effect estimates from three meta-analyses of non-
accidental mortality (Bell et al. 2005; Ito et al. 2005; Levy et al. 2005). Beginning with the 2015
Ozone NAAQS RIA, the Agency began quantifying ozone-attributable premature deaths using
two newer multi-city studies of non-accidental mortality (Smith et al. 2009; Zanobetti and
Schwartz 2008) and one long-term cohort study of respiratory mortality (Jerrett et al. 2009). The
2020 Ozone ISA included changes to the causality relationship determinations between short-
term exposures and total mortality, as well as including more recent epidemiologic analyses of
long-term exposure effects on respiratory mortality (U.E. EPA, 2020b). Consistent with the RCU
analysis we use two estimates of ozone-attributable respiratory deaths from short-term exposures
are estimated using the risk estimate parameters from Zanobetti et al. (2008) and Katsouyanni et
al. (2009). Ozone-attributable respiratory deaths from long-term exposures are estimated using
Turner et al. (2016). Due to time and resource limitations, we were unable to reflect the warm
season defined by Zanobetti et al. (2008) as June-August. Instead, we apply this risk estimate to
our standard warm season of May-September.
5.1.1.4 Quantifying Cases of PM2.s-Attributable Premature Death
When quantifying PM-attributable cases of adult mortality, we use the effect coefficients
from two epidemiology studies examining two large population cohorts: the American Cancer
Society cohort (Turner et al. 2016) and the Medicare cohort (Di et al. 2017). The Integrated
Science Assessment for Particulate Matter (PM ISA) (U.S. EPA 2019) concluded that the
analyses of the ACS and Medicare cohorts provide strong evidence of an association between
long-term PM2.5 exposure and premature mortality with support from additional cohort studies.
There are distinct attributes of both the ACS and Medicare cohort studies that make them well-
suited to being used in a PM benefits assessment and so here we present PM2.5 related effects
derived using relative risk estimates from both cohorts.
5-11
-------
The PM ISA, which was reviewed by the Clean Air Scientific Advisory Committee of
EPA's Science Advisory Board (SAB-CASAC) (EPA-SAB 2020a), concluded that there is a
causal relationship between mortality and both long-term and short-term exposure to PM2.5 based
on the entire body of scientific evidence. The PM ISA also concluded that the scientific literature
supports the use of a no-threshold log-linear model to portray the PM-mortality concentration-
response relationship while recognizing potential uncertainty about the exact shape of the
concentration-response relationship. The 2019 PM ISA, which informed the setting of the 2020
PM NAAQS, reviewed available studies that examined the potential for a population-level
threshold to exist in the concentration-response relationship. Based on such studies, the ISA
concluded that the evidence supports the use of a "no-threshold" model and that "little evidence
was observed to suggest that a threshold exists" (U.S. EPA 2009) (pp. 2-25 to 2-26). Consistent
with this evidence, the Agency historically has estimated health impacts above and below the
prevailing NAAQS (U.S. EPA 2010c, 2010d, 2011c, 201 Id, 2012, 2013b, 2014a, 2014b, 2014c,
2015a, 2015b, 2015c, 2015d, 2015e, 2016b).
5.1.1.5 Applying PM2.5 Benefit per Ton Values
Implementing the proposal is expected to reduce emissions of NOx during the May
through September ozone season as well as annually from the power sector due to year-round
operation of control measures. The proposal is also expected to reduce annual emissions in NOx,
PM2.5 and SO2 due to changes in power sector operation. Direct PM2.5 and SO2 reductions
reduce ambient PM2.5 concentrations year-round, while NOx emission reductions reduce PM2.5
concentrations in the winter months. To estimate the benefits from these changes, we performed
a benefit per ton analysis. For details on how these benefit per ton values are estimated, see
EPA's updated Technical Support Document Estimating the Benefit per Ton of Reducing
Directly-Emitted PM2.5, PM2.5 Precursors and Ozone Precursors from 21 Sectors (BPT TSD)
(U.S. EPA, 2021b). The procedure for calculating benefit per ton PM2.5 coefficients follows three
steps:
1. Using source apportionment photochemical modeling, predict annual average
ambient concentrations of primary PM2.5, nitrate and sulfate attributable to each of
21 emission sectors located throughout the Continental U.S. The source
apportionment modeling for the power sector uses the 2017 NEI.
5-12
-------
2. For each sector, estimate the health impacts, and the economic value of these
impacts, associated with the attributable ambient concentrations of primary PM2.5,
sulfate and nitrate PM2.5, and Ozone from NOx and Ozone from VOC using the
environmental Benefits Mapping and Analysis Program-Community Edition
(BenMAP-CE vl.5.8) and the risk and valuation estimates documented in the
Estimating PM2.5- and Ozone-Attributable Health Benefits TSD.
3. For each sector, divide the PIvfc.s-related health impacts attributable to each type of
PM2.5, and the monetary value of these impacts, by the level of associated precursor
emissions. That is, primary PM2.5 benefits are divided by direct PM2.5 emissions,
sulfate benefits are divided by SO2 emissions, and nitrate benefits are divided by
NOx emissions.
For this analysis, we modeled expected annual NOx, annual SO2, annual direct-PM2.5,
and warm season NOx emissions reductions by state that reflect the effects of generation shifting
or other EGU controls that are expected to operate year-round. Changes in power sector
emissions are derived from the IPM analysis of the proposed rule relative to the baseline scenario
(for details, please refer to Chapter 4). Depending on the sector, either state or regional benefit
per ton values from the BPT TSD were multiplied by the modeled changes in PM2.5 and PM2.5
precursors for each state. The values were summed across pollutants then summed for each
policy scenario. Benefit per ton values from 2025 were applied for years 2023-2029 and 2030
values were applied for years 2030-2042.
5.1.2 Economic Valuation Methodology for Health Benefits
We next quantify the economic value of the ozone and PIvfc.s-related deaths and illnesses
estimated above. Changes in ambient concentrations of air pollution generally yield small
changes in the risk of future adverse health effects for a large number of people. The appropriate
economic measure of the value of these small changes in risk of a health effect for the purposes
of a benefit-cost analysis is WTP. For some health effects, such as hospital admissions, WTP
estimates are not generally available, so we use the cost of treating or mitigating the effect. These
cost-of-illness (COI) estimates are typically a lower bound estimate of the true value of reducing
the risk of a health effect because they reflect the direct expenditures related to treatment, but not
5-13
-------
the value of avoided pain and suffering. The unit values applied in this analysis are provided in
Table 21 of the Estimating PM2.5- and Ozone-Attributable Health Benefits TSD.
The value of avoided premature deaths generally account for over 95 percent of
monetized ozone-related benefits and over 98 percent of monetized PIvfc.s-related benefits. The
economics literature concerning the appropriate method for valuing reductions in premature
mortality risk is still developing. The value for the projected reduction in the risk of premature
mortality is the subject of continuing discussion within the economics and public policy analysis
community. Following the advice of the SAB's Environmental Economics Advisory Committee
(SAB-EEAC), EPA currently uses the value of statistical life (VSL) approach in calculating
estimates of mortality benefits, because we believe this calculation provides the most reasonable
single estimate of an individual's willingness to trade off money for changes in the risk of death
(U.S. EPA-SAB 2000a). The VSL approach is a summary measure for the value of small
changes in the risk of death experienced by a large number of people.
EPA continues work to update its guidance on valuing mortality risk reductions, and the
Agency consulted several times with the SAB-EEAC on this issue. Until updated guidance is
available, the Agency determined that a single, peer-reviewed estimate applied consistently, best
reflects the SAB-EEAC advice it has received. Therefore, EPA applies the VSL that was vetted
and endorsed by the SAB in the Guidelines for Preparing Economic Analyses (U.S. EPA 2016a)
while the Agency continues its efforts to update its guidance on this issue. This approach
calculates a mean value across VSL estimates derived from 26 labor market and contingent
valuation studies published between 1974 and 1991. The mean VSL across these studies is $4.8
million (1990$). We then adjust this VSL to account for the currency year and to account for
income growth from 1990 to the analysis year. Specifically, the VSL applied in this analysis in
2016$ after adjusting for income growth is $10.7 million for 2025.
The Agency is committed to using scientifically sound, appropriately reviewed evidence
in valuing changes in the risk of premature death and continues to engage with the SAB to
identify scientifically sound approaches to update its mortality risk valuation estimates. In 2016,
the Agency proposed new meta-analytic approaches for updating its estimates (U.S. EPA-SAB
5-14
-------
2017), which were subsequently reviewed by the SAB-EEAC. EPA is reviewing the SAB's
formal recommendations.
In valuing PIvfc.s-related premature mortality, we discount the value of premature
mortality occurring in future years using rates of 3 percent and 7 percent (U.S. Office of
Management and Budget 2003). We assume that there is a multi-year "cessation" lag between
changes in PM exposures and the total realization of changes in health effects. Although the
structure of the lag is uncertain, EPA follows the advice of the SAB-HES to use a segmented lag
structure that assumes 30 percent of premature deaths are reduced in the first year, 50 percent
over years 2 to 5, and 20 percent over the years 6 to 20 after the reduction in PM2.5 (U.S. EPA-
SAB 2004). Changes in the cessation lag assumptions do not change the total number of
estimated deaths but rather the timing of those deaths.
Because estimated counts of short-term ozone-related premature mortality occur within
each analysis year, these estimated ozone-related benefits are identical for all discount rates.
When valuing changes in long-term ozone-attributable respiratory deaths using the Turner et al.
(2015) study, we follow advice provided by the Health Effects Subcommittee of the SAB, which
found that".. .there is no evidence in the literature to support a different cessation lag between
ozone and particulate matter. The HES therefore recommends using the same cessation lag
structure and assumptions as for particulate matter when utilizing cohort mortality evidence for
ozone" (U.S. EPA-SAB 2010).
These estimated health benefits do not account for the influence of future changes in the
climate on ambient concentrations of pollutants (USGCRP 2016). For example, recent research
suggests that future changes to climate may create conditions more conducive to forming ozone;
the influence of changes in the climate on PM2.5 concentrations are less clear (Fann et al. 2015).
The estimated health benefits also do not consider the potential for climate-induced changes in
temperature to modify the relationship between ozone and the risk of premature death (Fann et
al. 2021, Jhun et al. 2014; Ren et al. 2008a, 2008b).
5.1.3 Characterizing Uncertainty in the Estimated Benefits
This analysis includes many data sources as inputs that are each subject to uncertainty.
Input parameters include projected emission inventories, projected emissions and emissions
5-15
-------
changes from the electricity planning model, projected baseline emission and emission
reductions from non-EGUs, air quality data from models (with their associated parameters and
inputs), population data, population estimates, health effect estimates from epidemiology studies,
economic data, and assumptions regarding the future state of the world (i.e., regulations,
technology, and human behavior). When compounded, even small uncertainties can greatly
influence the size of the total quantified benefits.
Our estimate of the total monetized ozone and PIVfo.s-attributable benefits is based on
EPA's interpretation of the best available scientific literature and methods and supported by the
SAB-HES and the National Academies of Science (NRC 2002). Below are key assumptions
underlying the estimates for ozone-related premature deaths, followed by key uncertainties
associated with estimating the number and value of PIvfc.s-related premature mortality.
The estimated number and value of avoided ozone-attributable deaths are subject to
uncertainty. When estimating the economic value of avoided premature mortality from long-term
exposure to ozone, we use a 20-year segment lag (as used for PM2.5) as there is no alternative
empirical estimate of the cessation lag for long-term exposure to ozone. The 20-year segmented
lag accounts for the onset of cardiovascular related mortality, an outcome which is not relevant
to the long-term respiratory mortality estimated here. We use a log-linear impact function
without a threshold in modeling short-term ozone-related mortality. Thus, the estimates include
health benefits from reducing ozone in areas with varied concentrations of ozone down to the
lowest modeled concentrations. However, we acknowledge reduced confidence in specifying the
shape of the concentration-response relationship in the range of < 40ppb and below (2020 Ozone
ISA, section 6.2.6).
We assume that all fine particles, regardless of their chemical composition, are equally
potent in causing premature mortality. This is an important assumption, the PM ISA concluded
that "many constituents of PM2.5 can be linked with multiple health effects, and the evidence is
not yet sufficient to allow differentiation of those constituents or sources that are more closely
related to specific outcomes" (U.S. EPA 2009).
As noted above, we quantify health impacts of fine particles using a log-linear no-
threshold threshold model. Thus, some portion of the air quality and health benefits from the
5-16
-------
regulatory control alternatives will occur in areas not attaining the ozone or PM NAAQS.
Expected changes in the ambient concentrations of both ozone and PM2.5 pollutants may lead to
states changing their NAAQS compliance approaches. However, we do not simulate how states
would account for this proposed rule when complying with the NAAQS, which introduces
uncertainty in the estimated benefits (and costs).
Also, as noted above, we assume that there is a "cessation" lag between the change in PM
exposures and the total realization of changes in mortality effects. Specifically, we assume that
some of the incidences of premature mortality related to PM2.5 exposures occur in a distributed
fashion over the 20 years following exposure based on the advice of the SAB-HES (U.S. EPA-
SAB 2004), which affects the valuation of mortality benefits at different discount rates. The
above assumptions are subject to uncertainty.
In general, we are more confident in the magnitude of the risks we estimate from
simulated PM2.5 concentrations that coincide with the bulk of the observed PM concentrations in
the epidemiological studies that are used to estimate the benefits. Likewise, we are less confident
in the risk we estimate from simulated PM2.5 concentrations that fall below the bulk of the
observed data in these studies. There are uncertainties inherent in identifying any particular point
at which our confidence in reported associations decreases appreciably, and the scientific
evidence provides no clear dividing line. This relationship between the air quality data and our
confidence in the estimated risk is represented below in Figure 5-1.
Less confident More
Figure 5-1. Stylized Relationship between the PM2.5 Concentrations Considered in
Epidemiology Studies and our Confidence in the Estimated PM-related Premature Deaths
confident
Below LRL of PM2.5 data
in epidemiology study
(extrapolation)
I standard deviation below the
mean PM2.5 observed in
epidemiology study
Mean of PM2.5 data in
epidemiology study
5-17
-------
For Turner et al. 2016, the LRL is 2.8 |ig/m3 and for Di et al. 2017, the LRL is 0.02
|ig/m3. Additional information on low concentration exposures in Turner et al. 2016 and Di et al.
2017 can be found in section 6.1.2.1 of the Estimating PM2.5- and Ozone-Attributable Health
Benefits TSD. These results are sensitive to the annual mean PM2.5 concentration the air quality
model predicted in each 12 km by 12 km grid cell. The air quality modeling predicts PM2.5
concentrations to be at or below the annual mean PM2.5 NAAQS (12 |ig/m3) in nearly all
locations. The photochemical modeling we employ accounts for the suite of local, state and
federal policies expected to reduce PM2.5 and PM2.5 precursor emissions in future years, such that
we project a very small number of locations exceeding the annual standard. The results should be
viewed in the context of the air quality modeling technique we used to estimate PM2.5
concentrations. We are more confident in our ability to use the air quality modeling techniques
described above to estimate changes in annual mean PM2.5 concentrations than we are in our
ability to estimate absolute PM2.5 concentrations.
5.1.4 Estimated Number and Economic Value of Health Benefits
Below we report the estimated number of reduced premature deaths and illnesses in each
year relative to the baseline along with the 95% confidence interval (Table 5-2 and Table 5-3).
The number of reduced estimated deaths and illnesses from the proposed rule and more and less
stringent alternatives are calculated from the sum of individual reduced mortality and illness risk
across the population. Table 5-4 and Table 5-5 report the estimated economic value of avoided
premature deaths and illness in each year relative to the baseline along with the 95% confidence
interval. We also report the stream of benefits from 2023 through 2042 for the proposal, more-
and less- stringent alternatives, using the monetized sums of long-term ozone and PM2.5
mortality and morbidity impacts (Table 5-6 and Table 5-7).8 We also provide illustrative PM2.5
benefits for non-EGUs below in Table 5-8.
8 EPA continues to refine its approach for estimating and reporting PM-related effects at lower concentrations. The
Agency acknowledges the additional uncertainty associated with effects estimated at these lower levels and seeks to
develop quantitative approaches for reflecting this uncertainty in the estimated PM benefits.
5-18
-------
Table 5-2. Estimated Avoided Ozone-Related Premature Respiratory Mortalities and
Illnesses for the Proposal and More and Less Stringent Alternatives for 2023 (95%
Confidence Interval) a,b
Proposal
More Stringent
Alternative
Less Stringent
Alternative11
Avoided premature respiratory mortalities
Long-
term
Turner et al. (2016)°
44
51
44
exposure
(31 to 57)
(36 to 66)
(31 to 57)
Short-
term
Katsouyanni et al.
(2009)cd and Zanobetti et
2
2.3
2
exposure
al. (2008)d pooled
(0.8 to 3.1)
(0.94 to 3.7)
(0.81 to 3.2)
Morbidity effects
Long-
term
exposure
Asthma onset6
350
(300 to 390)
400
(340 to 450)
350
(300 to 400)
Allergic rhinitis
symptoms®
2,000
(1,000 to 2,900)
2,200
(1,200 to 3,300)
2,000
(1,000 to 2,900)
Hospital admissions—
5.3
6.1
5.3
respiratoryd
(-1.4 to 12)
(-1.6 to 14)
(-1.4 to 12)
Short-
term
ED visits—respiratoryf
110
(30 to 230)
120
(34 to 260)
110
(30 to 230)
Asthma symptoms
62,000
(-7,700 to 130,000)
71,000
(-8,800 to 150,000)
62,000
(-7,700 to 130,000)
exposure
Minor restricted-activity
30,000
34,000
30,000
daysd-f
(12,000 to 47,000)
(14,000 to 54,000)
(12,000 to 48,000)
School absence days
22,000
(-3,100 to 47,000)
26,000
(-3,600 to 54,000)
22,000
(-3,200 to 47,000)
a Values rounded to two significant figures.
b We estimated ozone benefits for changes in NOx for the ozone season and changes in PM2 5 and PM2 5 precursors
for EGUs in 2023. This table does not include benefits from emissions reductions for non-EGUs because emissions
reductions from these sources are not expected prior to 2026 when the proposed standards would become effective.
0 Applied risk estimate derived from April-September exposures to estimates of ozone across the May-September
warm season.
d Converted ozone risk estimate metric from MDA1 to MDA8.
e Applied risk estimate derived from June-August exposures to estimates of ozone across the May-September warm
season.
f Applied risk estimate derived from full year exposures to estimates of ozone across the May-September warm
season.
g Converted ozone risk estimate metric from DA24 to MDA8.
hThe proposed rule imposes unit level emission rate limits on EGUs in the 2026, which are imposed in the 2025
IPM run year, while the less stringent alternative assumes these are imposed in 2028, and in IPM are applied in the
2028 run year. The unit level emission rate limits drive much of the EGU retirement activity, and retirements are
delayed in the less stringent alternative relative to the proposed rule. Consistent with the power sector analysis in
Chapter 4, the power sector model is forward looking and has an incentive to run units harder before they retire.
This incentive is lower in the less stringent alternative relative to the proposed rule due to delayed retirements. As
such, emissions are slightly lower in 2023 in some states in the less stringent alternative relative to the proposed
rule.
5-19
-------
Table 5-3. Estimated Avoided Ozone-Related Premature Respiratory Mortalities and
Illnesses for the Proposal and More and Less Stringent Alternatives for 2026 (95%
Confidence Interval) a,b,h
More Stringent Less Stringent
Proposal Alternative _ Alternative
Duration0 Study Affected Facility Avoided premature respiratory mortalities
Long-
Turner et ctl
EGUs
450
(310 to 580)
520
(360 to 670)
210
(140 to 270)
term
exposure
(2016)c
Non-EGUs
510 (350 to
660)
550 (380 to 710)
450 (310 to 580)
EGUs + Non-
EGUs
960 (660 to
1,200)
1,100 (740 to
1,400)
650 (450 to 850)
Short-
term
Katsouyanni el
al. (2009)c d and
EGUs
20
8.2 to 32)
24
(9.5 to 37)
9.4
(3.8 to 15)
Zanobetti et al.
Non-EGUs
23 (9.3 to 36)
25 (10 to 39)
20 (8.2 to 32)
exposure
(2008)Voled
EGUs + Non-
EGUs
43 (18 to 68)
48 (19 to 76)
30 (12 to 47)
Morbidity effects
EGUs
3,300
(2,800 to
3,700)
3,800
(3,300 to 4,300)
1,600
(1,300 to 1,800)
Long-
term
exposure
Asthma onset6
Non-EGUs
3,800 (3,300 to
4,400)
4,200 (3,600 to
4,700)
3,400 (2,900 to
3,800)
EGUs + Non-
EGUs
7,100 (6,100 to
8,100)
7,900 (6,800 to
9,000)
4,900 (4,200 to
5,600)
EGUs
19,000
(9,900 to
27,000)
22,000
(11,000 to
32,000)
8,900
(4,700 to 13,000)
Allergic rhinitis
symptoms®
Non-EGUs
22,000 (12,000
to 32,000)
24,000 (13,000
to 35,000)
19,000 (10,000 to
28,000)
EGUs + Non-
EGUs
41,000 (22,000
to 59,000)
46,000 (24,000
to 66,000)
28,000 (15,000 to
41,000)
Short-
term
exposure
Hospital
admissions—
respiratoryd
55
63
25
ED visits—
respiratoryf
Asthma
symptoms
EGUs
(-14 to 120)
(-17 to 140)
(-6.5 to 55)
Non-EGUs
61 (-16 to 140)
66 (-17 to 150)
54 (-14 to 120)
EGUs + Non-
120 (-30 to
EGUs
260)
130 (-34 to 290)
79 (-21 to 170)
1,100
1,200
500
EGUs
(290 to 2,200)
(340 to 2600)
(140 to 1,100)
Non-EGUs
1,200 (340 to
1,300 (360 to
1,100 (300 to
2,600)
2,800)
2,300)
EGUs + Non-
2,300 (630 to
2,600 (700 to
1,600 (430 to
EGUs
4,800)
5,400)
3,300)
610,000
700,000
290,000
(-75,000 to
(-86,000 to
(-35,000 to
EGUs
1,300,000)
1,500,000)
590,000)
Non-EGUs
710,000
770,000
620,000
(-87,000 to
(-94,000 to
(-77,000 to
1,500,000)
1,600,000)
1,300,000)
5-20
-------
EGUs + Non-
EGUs
1,300,000
(-160,000 to
2,700,000)
1,500,000
(-180,000 to
3,000,000)
910,000
(-110,000 to
1,900,000)
EGUs
280,000
(110,000 to
440,000)
330,000
(13,000 to
520,000)
130,000
(53,000 to
210,000)
Minor restricted-
activity daysd-f
Non-EGUs
330,000
(130,000 to
520,000)
360,000
(140,000 to
560,000)
290,000
(120,000 to
460,000)
EGUs + Non-
EGUs
610,000
(240,000 to
970,000)
680,000
(270,000 to
1,100,000)
420,000
(170,000 to
670,000)
EGUs
220,000
(-30,000 to
450,000)
250,000
(-35,000 to
520,000)
100,000
(-14,000 to
210,000)
School absence
days
Non-EGUs
250,000 (-
35,000 to
530,000)
270,000
(-38,000 to
570,000)
220,000
(-31,000 to
460,000)
EGUs + Non-
EGUs
470,000
(-66,000 to
980,000)
520,000
(-74,000 to
1,100,000)
320,000
(-46,000 to
670,000)
a Values rounded to two significant figures.
b We estimated ozone benefits for changes in NOx for the ozone season and changes in PM2 5 and PM2 5 precursors
for EGUs in 2026.
0 Applied risk estimate derived from April-September exposures to estimates of ozone across the May-September
warm season.
d Converted ozone risk estimate metric from MDA1 to MDA8.
e Applied risk estimate derived from June-August exposures to estimates of ozone across the May-September warm
season.
f Applied risk estimate derived from full year exposures to estimates of ozone across the May-September warm
season.
g Converted ozone risk estimate metric from DA24 to MDA8.
h Non-EGU benefits estimates are ozone-related only. An illustrative analysis of non-EGU PM benefits estimates is
presented in Table 5-8.
5-21
-------
Table 5-4. Estimated Discounted Economic Value of Avoided Ozone and PM2.5-
Attributable Premature Mortality and Illness for the Proposed Policy Scenarios in 2023
(95% Confidence Interval; millions of 2016$)a,b
Disc.
Rate
Pollutant
Proposal
More Stringent Alternative
Less Stringent Alternative
3%
Ozone
Benefits
$57 ($15
to $120)°
and
$460 ($51
to $l,200)d
$65 ($17
to
$140)°
and
$530 ($59
to $l,400)d
$57
($15 to
$120)°
and
$460
($51 to
$l,200)d
PM
Benefits
$44
and
$45
$190
and
$190
$59
and
$60
Ozone
plus PM
Benefits
$100
($59 to
$160)°
and
$500
($96 to
$l,200)d
$250
($200 to
$330)°
and
$720
($250 to
$l,600)d
$120
($74 to
$180)°
and
$520
($110 to
$l,300)d
7%
Ozone
Benefits
$51 ($9.6
to 110)°
and
$410 ($42
to $l,100)d
$58 ($11
to
$130)°
and
$480 ($49
to $l,300)d
$51
($9.6 to
$110)°
and
$410
($42 to
$l,100)d
PM
Benefits
$40
and
$41
$170
and
$170
$53
and
$54
Ozone
plus PM
Benefits
$90
($49 to
$150)°
and
$450
($83 to
$l,100)d
$230
($180 to
$300)°
and
$650
($220 to
$l,400)d
$100
($63 to
$170)°
and
$470
($97 to
$l,100)d
a Values rounded to two significant figures. The two benefits estimates are separated by the word "and" to signify
that they are two separate estimates. The estimates do not represent lower- and upper-bound estimates and should
not be summed.
b We estimated ozone benefits for changes in NOx for the ozone season and changes in PM2 5 and PM2 5 precursors
for EGUs in 2023. This table does not include benefits from reductions for non-EGUs because reductions from these
sources are not expected prior to 2026 when the proposed standards would become effective.
0 Using the pooled short-term ozone exposure mortality risk estimate.
d Using the long-term ozone exposure mortality risk estimate.
5-22
-------
Table 5-5. Estimated Discounted Economic Value of Avoided Ozone and PM2.5-
Attributable Premature Mortality and Illness for the Proposed Policy Scenario in 2026
(95% Confidence Interval; millions of 2016$)a,b
Disc
Rate
Pollutant
Proposal
More Stringent
Alternative
Less Stringent Alternative
3%
Ozone
Benefits
$10,000
$1,200 ($1,100
($310 to and to
$2,600)° $26,000)
d
niot0 $11,000
and ($1,200 to
3)2,yuu) $29,000)d
$830 Qnn
($210 to °
«i snm an" ($760 to
3>t,8UU) $18,000)d
PM
Benefits
$8,100 and $8,300
$7,800 and $7,900
$3,400 and $3,500
Ozone
plus PM
Benefits
$18,000
$9,300 ($9,400
($8,400 to and to
$11,000)C $35,000)
d
$9,100
($8,100 $19,000
to and ($9,200 to
$11,000 $37,000)d
)c
$4'300 $10 000
($3 700 aM ($4 3'00 tQ
$5,200)° $22'00°)d
7%
Ozone
Benefits
$1 100 $9'000
ll," , ($920 to
($200 to and '
$2,400)° uuu;
$1'200 $10 000
($220 to n'nn t
ct and ($1,000 to
W'0/UU) $26,000)d
,|740 $6,200
ci and ($63°to
M'C/UU) $16,000)d
PM
Benefits
$7,300 and $7,400
$7,000 and $7,100
$3,100 and $3,200
Ozone
plus PM
Benefits
$16,000
$8,400 ($8,300
($7,500 to and to
$9,700)° $31,000)
d
$8,200
($7 200 aM ($8 2'00 tQ
$9,700)° $34'00°)d
$3,800 $9,300
($3,200 and ($3,800 to
to $19,000)d
$4,800)°
a Values rounded to two significant figures. The two benefits estimates are separated by the word "and" to signify
that they are two separate estimates. The estimates do not represent lower- and upper-bound estimates and should
not be summed.
b We estimated changes in NOx for the ozone season and changes in PM2 5 and PM2 5 precursors in 2026. This table
represents changes in EGU and non-EGU ozone season and annual controls.
0 Sum of ozone mortality estimated using the pooled short-term ozone exposure risk estimate and the Di et al. (2017)
long-term PM2 5 exposure mortality risk estimate.
d Sum of the Turner et al. (2016) long-term ozone exposure risk estimate and the Di et al. (2017) long-term PM2 5
exposure mortality risk estimate.
5-23
-------
Table 5-6. Stream of Human Health Benefits from 2023 through 2042: Monetized
Benefits Quantified as Sum of Long-Term Ozone Mortality for EGUs and Non-EGUs
and Long-Term BPT PM2.5 Mortality for EGUs (Discounted at 3%; millions of 2016$)"
Proposal
More Stringent Alternative
Less Stringent Alternative
2023*
$500
$720
$520
2024
$520
$740
$530
2025
$530
$750
$550
2026*
$18,000
$19,000
$10,000
2027
$19,000
$19,000
$11,000
2028
$18,000
$19,000
$10,000
2029
$19,000
$20,000
$11,000
2030
$20,000
$21,000
$11.000
2031
$20,000
$21,000
$11,000
2032
$21,000
$22,000
$12,000
2033
$20,000
$21,000
$12,000
2034
$21,000
$22,000
$12,000
2035
$21,000
$22,000
$12,000
2036
$21,000
$22,000
$12,000
2037
$22,000
$23,000
$12,000
2038
$21,000
$22,000
$12,000
2039
$22,000
$23,000
$12,000
2040
$22,000
$23,000
$13,000
2041
$22,000
$23,000
$13,000
2042
$22,000
$23,000
$13,000
Net Present Value
$250,000
$270,000
$150,000
*Year in which air quality models were run. Benefits for all other years were extrapolated from years with model-
based air quality estimates. Benefits calculated as value of avoided: PM2 5-attributable deaths (quantified using a
concentration-response relationship from the Di et al. 2017 study); Ozone-attributable deaths (quantified using a
concentration-response relationship from the Turner et al. 2017 study); and PM2 5 and ozone-related morbidity
effects.
aFor the years 2023-2025, benefits associated with non-EGU emissions reductions are not included as
implementation of proposed control technologies will not be complete until 2026.
5-24
-------
Table 5-7. Stream of Human Health Benefits from 2023 through 2042: Monetized
Benefits Quantified as Sum of Short-Term Ozone Mortality for EGUs and Non-EGUS
and Long-Term BPT PM2.5 Mortality for EGUs (Discounted at 7%; millions of 2016$)"
Proposal
More Stringent Alternative
Less Stringent Alternative
2023*
$450
$650
$470
2024
$460
$660
$480
2025
$470
$670
$490
2026*
$16,000
$17,000
$9,300
2027
$17,000
$17,000
$9,400
2028
$16,000
$17,000
$9,300
2029
$17,000
$17,000
$9,500
2030
$18,000
$19,000
$10,000
2031
$18,000
$19,000
$10,000
2032
$18,000
$19,000
$10,000
2033
$18,000
$19,000
$10,000
2034
$18,000
$19,000
$10,000
2035
$19,000
$20,000
$11.000
2036
$19,000
$20,000
$1 1.000
2037
$19,000
$20,000
$11,000
2038
$19,000
$20,000
$11,000
2039
$19,000
$20,000
$11,000
2040
$19,000
$21,000
$11,000
2041
$19,000
$21,000
$1 1.000
2042
$20,000
$21,000
$11,000
Net Present Value
$150,000
$160,000
$88,000
*Year in which air quality models were run. Benefits for all other years were extrapolated from years with model-
based air quality estimates. Benefits calculated as value of avoided: PM2 5-attributable deaths (quantified using a
concentration-response relationship from the Di et al. 2017 study); Ozone-attributable deaths (quantified using a
pooled estimate of results quantified using concentration-response relationships two short-term exposure mortality
studies); and PM2 5 and ozone-related morbidity effects.
aFor the years 2023-2025, benefits associated with non-EGU emissions reductions are not included as
implementation of proposed control technologies will not be complete until 2026.
Since the proposed FIP for the 2015 ozone NAAQS includes ozone season emissions
limits for the non-EGU emissions units and because we do not know if all affected sources will
run controls year-round or only during ozone season, the benefits estimates in Table 5-8 provide
an illustration of potential PM2.5 benefits from non-EGUs if the proposed controls are run year-
round. For this proposal, we are taking comment on whether any of these emissions sources
would run controls year-round. These illustrative benefits estimates are not added to the total
health benefits for this proposal.
5-25
-------
Table 5-8. Illustrative Estimates of PM2.5-Attributable Premature Mortality and Illnesses
for the Proposal for Non-EGUs (millions of 2016$)"
Economic value of long-term mortality and morbidity health effects
Sector
Benefit Per Ton Valuation of
Reducing NOx (discounted at
3%)
Benefit Per Ton Valuation of
Reducing NOx (discounted at 7%)
Cement Kilns
External Combustion Boilers
Integrated Iron and Steel
Oil & Natural Gas Transmission
Refineries
Synthetic Organic Chemical Industry
290
280
59
770
21
19
260
220
52
680
19
17
a To estimate these benefits, we multiplied annual NOx emissions reductions for the non-EGU sources by the
relevant benefit per ton value. The ozone season NOx emissions reductions estimates are found in the Screening
Assessment of Potential Emissions Reductions, Air Quality Impacts, and Costs from Non-EGU Emissions Units for
2026 discussed further in Chapter 4. To estimate annual NOx emissions reductions, the ozone season estimates are
divided by 5/12. The benefit per ton values are from the BPT TSD (U.S. EPA, 2021b). We matched the industries
and NOx emissions reductions from the Screening Assessment of Potential Emissions Reductions, Air Quality
Impacts, and Costs from Non-EGU Emissions Units for 2026 to the industries and sources in the BPT TSD -
approximately 80 percent of the estimated NOx emissions reductions had an applicable benefit per ton value and are
reflected in the estimates in this table.
5.2 Climate Benefits from Reducing CO2
Elevated concentrations of greenhouse gases (GHGs) in the atmosphere have been
warming the planet, leading to changes in the Earth's climate including changes in the frequency
and intensity of heat waves, precipitation, and extreme weather events, rising seas, and retreating
snow and ice. The well-documented atmospheric changes due to anthropogenic GHG emissions
are changing the climate at a pace and in a way that threatens human health, society, and the
natural environment.
Extensive information on climate change is available in the scientific assessments and
EPA documents that are briefly described in this section, as well as in the technical and scientific
information supporting them. One of those documents is EPA's 2009 Endangerment and Cause
or Contribute Findings for Greenhouse Gases Under section 202(a) of the CAA (74 FR 66496,
December 15, 2009). In the 2009 Endangerment Finding, the Administrator found under section
202(a) of the CAA that elevated atmospheric concentrations of six key well-mixed GHGs - CO2,
methane (CH4), nitrous oxide (N20), HFCs, perfluorocarbons (PFCs), and sulfur hexafluoride
(SF6) - "may reasonably be anticipated to endanger the public health and welfare of current and
future generations" (74 FR 66523). The 2009 Endangerment Finding, together with the extensive
scientific and technical evidence in the supporting record, documented that climate change
caused by human emissions of GHGs threatens the public health of the U.S. population. It
5-26
-------
explained that by raising average temperatures, climate change increases the likelihood of heat
waves, which are associated with increased deaths and illnesses (74 FR 66497). While climate
change also increases the likelihood of reductions in cold-related mortality, evidence indicates
that the increases in heat mortality will be larger than the decreases in cold mortality in the U.S.
(74 FR 66525). The 2009 Endangerment Finding further explained that compared with a future
without climate change, climate change is expected to increase tropospheric ozone pollution over
broad areas of the U.S., including in the largest metropolitan areas with the worst tropospheric
ozone problems, and thereby increase the risk of adverse effects on public health (74 FR 66525).
Climate change is also expected to cause more intense hurricanes and more frequent and intense
storms of other types and heavy precipitation, with impacts on other areas of public health, such
as the potential for increased deaths, injuries, infectious and waterborne diseases, and stress-
related disorders (74 FR 66525). Children, the elderly, and the poor are among the most
vulnerable to these climate-related health effects (74 FR 66498).
The 2009 Endangerment Finding also documented, together with the extensive scientific
and technical evidence in the supporting record, that climate change touches nearly every aspect
of public welfare in the U.S. with resulting economic costs, including: changes in water supply
and quality due to changes in drought and extreme rainfall events; increased risk of storm surge
and flooding in coastal areas and land loss due to inundation; increases in peak electricity
demand and risks to electricity infrastructure; and the potential for significant agricultural
disruptions and crop failures (though offset to some extent by carbon fertilization). These
impacts are also global and the effects of climate change occurring outside the U.S. are
reasonably expected to impact the U.S. population. (74 FR 66530).
In 2016, the Administrator issued a similar finding for GHG emissions from aircraft
under section 231(a)(2)(A) of the CAA. In the 2016 Endangerment Finding, the Administrator
found that the body of scientific evidence amassed in the record for the 2009 Endangerment
Finding compellingly supported a similar endangerment finding under CAA section
231(a)(2)(A), and also found that the science assessments released between the 2009 and the
2016 Findings "strengthen and further support the judgment that GHGs in the atmosphere may
reasonably be anticipated to endanger the public health and welfare of current and future
generations" (81 FR 54424).
5-27
-------
Since the 2016 Endangerment Finding, the climate change impacts have continued to
intensify, with new observational records being set for several climate indicators such as global
average surface temperatures, GHG concentrations, and sea level rise. Moreover, heavy
precipitation events have increased in the eastern United States while agricultural and ecological
drought has increased in the western United States along with more intense and larger wildfires.9
Climate impacts that occur outside U.S. borders also increasingly impact the welfare of
individuals and firms that reside in the United States because of their connection to the global
economy. This will occur through the effect of climate change on international markets, trade,
tourism, and other activities. For example, supply chain disruptions are a prominent pathway
through which U.S. business and consumers are, and will continue to be, affected by climate
change impacts abroad (USGCRP 2018, U.S. DOD 2021). Additional climate change induced
international spillovers can occur through pathways such as damages across transboundary
resources, economic and political destabilization, and global migration that can lead to adverse
impacts on U.S. national security, public health, and humanitarian concerns (U.S. DOD 2014,
CCS 2018). These and other trends highlight the increased risk already being experienced due to
climate change as detailed in the 2009 and 2016 Endangerment Findings. Additionally, new
major scientific assessments continue to advance our understanding of the climate system and
the impacts that GHGs have on public health and welfare both for current and future generations.
These assessments include:
• U.S. Global Change Research Program's (USGCRP) 2016 Climate and Health
Assessment and 2017-2018 Fourth National Climate Assessment (NCA4) (USGCRP
2016, 2017, 2018).
• IPCC's 2018 Global Warming of 1.5 °C, 2019 Climate Change and Land, and the 2019
Ocean and Cryosphere in a Changing Climate assessments, as well as the 2021 IPCC
Sixth Assessment Report (AR6) (IPCC 2018, 2019a, 2019b, 2021).
• The National Academies of Sciences, Engineering, and Medicine's 2016 Attribution of
Extreme Weather Events in the Context of Climate Change, 2017 Valuing Climate
Damages: Updating Estimation of the Social Cost of Carbon Dioxide, and 2019 Climate
Change and Ecosystems assessments (NAS 2016, 2017, 2019).
9 See EPA's November 2021 Proposed Standards of Performance for New, Reconstructed, and Modified Sources
and Emissions Guidelines for Existing Sources: Oil and Natural Gas Sector Climate Review
(https://www.govinfo.gOv/content/pkg/FR-2021-ll-15/pdf/2021-24202.pdf) for more discussion of specific
examples. An additional resource for indicators can be found at https://www.epa.gov/climate-indicators.
5-28
-------
• National Oceanic and Atmospheric Administration's (NOAA) annual State of the
Climate reports published by the Bulletin of the American Meteorological Society, most
recently in August of 2020 (Blunden and Arndt 2020).
• EPA Climate Change and Social Vulnerability in the United States: A Focus on Six
Impacts (2021) (EPA 2021c).
Net climate benefits from reducing emissions of CO2 can be monetized using estimates of
the social cost of carbon (SC-CO2). However, as explained below, due to a court order, EPA
cannot present these monetized estimates in the analysis of this proposed rule at this time. The
SC-CO2 is the monetary value of the net harm to society associated with a marginal increase in
CO2 emissions in a given year, or the benefit of avoiding that increase. In principle, SC-CO2
includes the value of all climate change impacts (both negative and positive), including (but not
limited to) changes in net agricultural productivity, human health effects, property damage from
increased flood risk, natural disasters, disruption of energy systems, risk of conflict,
environmental migration, and the value of ecosystem services. The SC-CO2, therefore, should
reflect the societal value of reducing emissions of the gas in question by one metric ton. The SC-
CO2 is therefore, an estimate of the marginal benefit of CO2 abatement along the baseline and the
theoretically appropriate value to use in conducting benefit-cost analyses of policies that affect
CO2 emissions. In practice, data and modeling limitations naturally restrain the ability of SC-
CO2 estimates to include all of the important physical, ecological, and economic impacts of
climate change, such that the estimates are a partial accounting of climate change impacts and
will therefore, tend to be underestimates of the marginal benefits of abatement.
EPA and other federal agencies began regularly incorporating SC-CO2 estimates in
benefit-cost analyses conducted under Executive Order (E.O.) 1286610 in 2008, following a court
ruling in which an agency was ordered to consider the value of reducing CO2 emissions in a
rulemaking process. Specifically, the U.S. Ninth Circuit Court of Appeals remanded a fuel
economy rule to DOT for failing to monetize CO2 emission reductions, stating that "while the
record shows that there is a range of values, the value of carbon emissions reduction is certainly
10 Under E.O. 12866, agencies are required, to the extent permitted by law and where applicable, "to assess both the
costs and the benefits of the intended regulation and, recognizing that some costs and benefits are difficult to
quantify, propose or adopt a regulation only upon a reasoned determination that the benefits of the intended
regulation justify its costs." Some statutes also require agencies to conduct at least some of the same analyses
required under E.O. 12866, such as the Energy Policy and Conservation Act, which mandates the setting of fuel
economy regulations.
5-29
-------
not zero."11 In 2009, the U.S. Government (USG) launched an interagency process, under the
leadership of the Office of Management and Budget (OMB) and the Council of Economic
Advisers (CEA), to ensure that Federal agencies had access to the best available information
when quantifying the benefits of reducing CO2 emissions in regulatory impact analyses and to
promote consistency in the estimated values. This included the establishment of an interagency
working group (IWG) which represented perspectives and technical expertise from many federal
agencies and a commitment to following the peer-reviewed literature. In 2010, the IWG finalized
a set of four SC-CO2 values recommended for use in regulatory analyses and presented them in a
technical support document (TSD) that also provided guidance for agencies on how to use the
estimates (IWG 2010). The SC-CO2 estimates recommended in 2010 were developed from an
ensemble of three widely cited integrated assessment models (IAMs) that estimate global climate
damages using highly aggregated representations of climate processes and the global economy
combined into a single modeling framework. The three IAMs were run using a common set of
input assumptions in each model for future population, economic, and GHG emissions growth,
as well as equilibrium climate sensitivity (ECS) - a measure of the globally averaged
temperature response to increased atmospheric CO2 concentrations. These estimates were
updated in 2013 based on new versions of each IAM. In August 2016 the IWG published
estimates of the social cost of methane (SC-CH4) and nitrous oxide (SC-N20) using
methodologies that are consistent with the methodology underlying the SC-CO2 estimates. In
January 2017, the National Academies of Sciences, Engineering, and Medicine issued
recommendations for an updating process to ensure the estimates continue to reflect the best
available science (National Academies 2017). In March 2017, Executive Order 13783 disbanded
the IWG and instructed agencies when monetizing the value of changes in greenhouse gas
emissions resulting from regulations to follow the Office of Management and Budget's (OMB)
Circular A-4.
On January 20, 2021, President Biden issued E.O. 13990 which re-established the IWG
and asked it to update the estimates of SC-CO2, SC-CH4, and SC-N2O (collectively referred to as
social cost of greenhouse gases (SC-GHG)) used by the U.S. Government (USG) to reflect the
best available science and the recommendations of the National Academies (2017). On February
11 Ctr. for Biological Diversity v. Nat'l Highway Traffic Safety Admin., 538 F.3d 1172, 1200 (9th Cir. 2008).
5-30
-------
26, 2021, the IWG recommended as interim SC-GHG estimates the most recent estimates
developed by the IWG prior to the group being disbanded in 2017. The February 2021 TSD
stated that the interim estimates reflected the best available scientific estimates available for
agencies to use in regulatory benefit-cost analyses and other applications while the more
comprehensive review was underway.
On February 11, 2022, the U.S. District Court for the Western District of Louisiana
issued an injunction concerning the monetization of benefits of greenhouse gas emission
reductions by EPA and other defendants. See Louisiana v. Biden, No. 21-cv-01074-JDC-KK
(W.D. La. Feb. 11, 2022). Accordingly, monetized climate benefits are not presented in the
benefit-cost analysis of this proposal conducted pursuant to E.O. 12866. We note that the
absence of monetized climate benefits from the analysis of benefits and net benefits in this RIA
has no bearing on the legal or technical basis for the proposed action itself. The estimated total
reductions in greenhouse gas emissions projected to result from this proposed action will have
climate benefits by mitigating the impacts of climate change discussed above. Those benefits can
be understood as part of the unquantified benefits of this proposal that are described in
qualitative terms.
5.3 Additional Unquantified Benefits
Data, time, and resource limitations prevented EPA from quantifying the estimated health
impacts or monetizing estimated benefits associated with direct exposure to NO2 and SO2
(independent of the role NO2 and SO2 play as precursors to PM2.5 and ozone), as well as
ecosystem effects, and visibility impairment due to the absence of air quality modeling data for
these pollutants in this analysis. While all health benefits and welfare benefits were not able to be
quantified, it does not imply that there are not additional benefits associated with reductions in
exposures to ozone, PM2.5, NO2 or SO2. In this section, we provide a qualitative description of
these and water quality benefits, which are listed in Table 5-9.
Table 5-9. Unquantified Health and Welfare Benefits Categories
Category
Effect
Effect
Quantified
Effect
Monetized
More
Information
Improved Human Health
Reduced incidence of
Asthma hospital admissions
—
—
NO2 ISA1
morbidity from exposure
Chronic lung disease hospital admissions
—
—
NO2 ISA1
to NO2
Respiratory emergency department visits
—
—
NO2 ISA1
5-31
-------
Asthma exacerbation
NO2 ISA1
Acute respiratory symptoms
—
—
NO2 ISA1
Premature mortality
—
—
NO2 ISA1'2'3
Other respiratory effects (e.g., airway
hyperresponsiveness and inflammation, lung
function, other ages and populations)
—
—
NO2 ISA2-3
Reduced incidence of
mortality and morbidity
through drinking water
from reduced effluent
discharges.
Bladder, colon, and rectal cancer from
halogenated disinfection byproducts
exposure.
—
—
SE ELG BCA4
Reproductive and developmental effects
from halogenated disinfection byproducts
exposure.
—
—
SE ELG BCA4
Neurological and cognitive effects to
children from lead exposure from fish
consumption (including need for specialized
education).
—
—
SE ELG BCA4
Possible cardiovascular disease from lead
exposure
—
—
SE ELG BCA4
Reduced incidence of
morbidity and mortality
from toxics through fish
consumption from reduced
effluent discharges.
Neurological and cognitive effects from in
in-utero mercury exposure from maternal
fish consumption
—
—
SE ELG BCA4
Skin and gastrointestinal cancer incidence
from arsenic exposure
—
—
SE ELG BCA4
Cancer and non-cancer incidence from
exposure to toxic pollutants (lead, cadmium,
thallium, hexavalent chromium etc.
Neurological, alopecia, gastrointestinal
effects, reproductive and developmental
damage from short-term thallium exposure.
—
—
SE ELG BCA4
Reduced incidence of
morbidity and mortality
from recreational water
exposure from reduced
effluent discharges.
Cancer and Non-Cancer incidence from
exposure to toxic pollutants (methyl-
mercury, selenium, and thallium.)
—
—
SE ELG BCA4
Improved Environment
Reduced visibility
Visibility in Class 1 areas
—
—
PM ISA1
impairment
Visibility in residential areas
—
—
PM ISA1
Reduced effects on
materials
Household soiling
—
—
PM ISA1-2
Materials damage (e.g., corrosion, increased
wear)
—
—
PM ISA2
Reduced effects from PM
deposition (metals and
organics)
Effects on individual organisms and
ecosystems
—
—
PM ISA2
Visible foliar injury on vegetation
—
—
Ozone ISA1
Reduced vegetation growth and reproduction
—
—
Ozone ISA1
Reduced vegetation and
ecosystem effects from
exposure to ozone
Yield and quality of commercial forest
products and crops
—
—
Ozone ISA1
Damage to urban ornamental plants
—
—
Ozone ISA2
Carbon sequestration in terrestrial
ecosystems
—
—
Ozone ISA1
Recreational demand associated with forest
aesthetics
—
—
Ozone ISA2
Other non-use effects
Ozone ISA2
5-32
-------
Ecosystem functions (e.g., water cycling,
biogeochemical cycles, net primary
productivity, leaf-gas exchange, community
composition)
—
—
Ozone ISA2
Recreational fishing
—
—
NOxSOxISA1
Tree mortality and decline
—
—
NOxSOxISA2
Reduced effects from acid
deposition
Commercial fishing and forestry effects
—
—
NOxSOxISA2
Recreational demand in terrestrial and
aquatic ecosystems
—
—
NOxSOxISA2
Other non-use effects
NOxSOxISA2
Ecosystem functions (e.g., biogeochemical
cycles)
—
—
NOxSOxISA2
Species composition and biodiversity in
terrestrial and estuarine ecosystems
—
—
NOxSOxISA2
Reduced effects from
Coastal eutrophication
—
—
NOxSOxISA2
Recreational demand in terrestrial and
estuarine ecosystems
nutrient enrichment from
deposition.
—
—
NOxSOxISA2
Other non-use effects
NOxSOxISA2
Ecosystem functions (e.g., biogeochemical
cycles, fire regulation)
—
—
NOxSOxISA2
Reduced vegetation effects
from ambient exposure to
SO2 and NOx
Injury to vegetation from SO2 exposure
—
—
NOxSOxISA2
Injury to vegetation from NOx exposure
—
—
NOxSOxISA2
Improved water aesthetics
from reduced effluent
Improvements in water clarity, color, odor in
residential, commercial and recreational
SE ELG BCA4
discharges.
settings.
Protection of Threatened and Endangered
(T&E) species from changes in habitat and
potential population effects.
—
—
SE ELG BCA4
Effects on aquatic
organisms and other
wildlife from reduced
effluent discharges
Other non-use effects
—
—
SE ELG BCA4
Changes in sediment contamination on
benthic communities and potential for re-
entrainment.
—
—
SE ELG BCA4
Quality of recreational fishing and other
recreational use values.
—
—
SE ELG BCA4
Commercial fishing yields and harvest
quality.
—
—
SE ELG BCA4
Reduced water treatment
costs from reduced
effluent discharges
Reduced drinking, irrigation, and other
agricultural use water treatment costs.
—
—
SE ELG BCA4
Increased storage availability in reservoirs
—
—
SE ELG BCA4
Reduced sedimentation
from effluent discharges
Improved functionality of navigable
waterways
—
—
SE ELG BCA4
Decreased cost of dredging
—
—
SE ELG BCA4
Benefits of reduced water
withdrawal
Benefits from effects aquatic and riparian
species from additional water availability.
—
—
SE ELG BCA4
Increased water availability in reservoirs
increasing hydropower supply, recreation,
and other services.
—
—
SE ELG BCA4
Climate effects
Climate impacts from carbon dioxide (CO2)
—
—
Section 5.2
discussion
Other climate impacts (e.g., ozone, black
carbon, aerosols, other impacts)
IPCC,
Ozone ISA,
PMISA
5-33
-------
1 We assess these benefits qualitatively due to data and resource limitations for this RIA.
2 We assess these benefits qualitatively because we do not have sufficient confidence in available data or methods.
3 We assess these benefits qualitatively because current evidence is only suggestive of causality or there are other significant
concerns over the strength of the association.
4 Benefit and Cost Analysis (BCA) for Revisions to the Effluent Limitations Guidelines (ELG) and Standards for the Steam
Electric (SE) Power Generating Point Source Category.
5.3.1 NO2 Health Benefits
In addition to being a precursor to PM2.5 and ozone, NOx emissions are also linked to a
variety of adverse health effects associated with direct exposure. We were unable to estimate the
health benefits associated with reduced NO2 exposure in this analysis. Following a
comprehensive review of health evidence from epidemiologic and laboratory studies, the
Integrated Science Assessment for Oxides of Nitrogen —Health Criteria (NOx ISA) (U.S. EPA,
2016c) concluded that there is a likely causal relationship between respiratory health effects and
short-term exposure to NO2. These epidemiologic and experimental studies encompass a number
of endpoints including emergency department visits and hospitalizations, respiratory symptoms,
airway hyperresponsiveness, airway inflammation, and lung function. The NOx ISA also
concluded that the relationship between short-term NO2 exposure and premature mortality was
"suggestive but not sufficient to infer a causal relationship," because it is difficult to attribute the
mortality risk effects to NO2 alone. Although the NOx ISA stated that studies consistently
reported a relationship between NO2 exposure and mortality, the effect was generally smaller
than that for other pollutants such as PM.
5.3.2 SO 2 Health Benefits
In addition to being a precursor to PM2.5, SO2 emissions are also linked to a variety of
adverse health effects associated with direct exposure. We were unable to estimate the health
benefits associated with reduced SO2 in this analysis. Therefore, this analysis only quantifies and
monetizes the PM2.5 benefits associated with the reductions in SO2 emissions. Following an
extensive evaluation of health evidence from epidemiologic and laboratory studies, the
Integrated Science Assessment for Oxides of Sulfur —Health Criteria (SO2 ISA) concluded that
there is a causal relationship between respiratory health effects and short-term exposure to SO2
(U.S. EPA 2017). The immediate effect of SO2 on the respiratory system in humans is
bronchoconstriction. Asthmatics are more sensitive to the effects of SO2 likely resulting from
preexisting inflammation associated with this disease. A clear concentration-response
5-34
-------
relationship has been demonstrated in laboratory studies following exposures to SO2 at
concentrations between 20 and 100 ppb, both in terms of increasing severity of effect and
percentage of asthmatics adversely affected. Based on our review of this information, we
identified three short-term morbidity endpoints that the SO2 ISA identified as a "causal
relationship": asthma exacerbation, respiratory-related emergency department visits, and
respiratory-related hospitalizations. The differing evidence and associated strength of the
evidence for these different effects is described in detail in the SO2 ISA. The SO2 ISA also
concluded that the relationship between short-term SO2 exposure and premature mortality was
"suggestive of a causal relationship" because it is difficult to attribute the mortality risk effects to
SO2 alone. Although the SO2ISA stated that studies are generally consistent in reporting a
relationship between SO2 exposure and mortality, there was a lack of robustness of the observed
associations to adjustment for other pollutants.
5.3.3 Ozone Welfare Benefits
Exposure to ozone has been associated with a wide array of vegetation and ecosystem
effects in the published literature (U.S. EPA, 2020b). Sensitivity to ozone is highly variable
across species, with over 65 plant species identified as "ozone-sensitive", many of which occur
in state and national parks and forests. These effects include those that damage or impair the
intended use of the plant or ecosystem. Such effects can include reduced growth and/or biomass
production in sensitive plant species, including forest trees, reduced yield and quality of crops,
visible foliar injury, species composition shift, and changes in ecosystems and associated
ecosystem services. See Section F of the Technical Support Document (TSD) for the Proposed
Federal Implementation Plan Addressing Regional Ozone Transport for the 2015 Ozone
National Ambient Air Quality Standard, Ozone Transport Policy Analysis Proposed Rule TSD
for a summary of an assessment of risk of ozone-related growth impacts on selected forest tree
species.
5.3.4 NO2 and SO2 Welfare Benefits
As described in the Integrated Science Assessment (ISA) for Oxides of Nitrogen, Oxides
of Sulfur and Particulate Matter Ecological Criteria (NOx/SOx/PM ISA) (U.S. EPA, 2020d),
NOx and SO2 emissions also contribute to a variety of adverse welfare effects, including those
5-35
-------
associated with acidic deposition, visibility impairment, and nutrient enrichment. Deposition of
nitrogen and sulfur causes acidification, which can cause a loss of biodiversity of fishes,
zooplankton, and macro invertebrates in aquatic ecosystems, as well as a decline in sensitive tree
species, such as red spruce (Picea rubens) and sugar maple (Acer saccharum) in terrestrial
ecosystems. In the northeastern U.S., the surface waters affected by acidification are a source of
food for some recreational and subsistence fishermen and for other consumers and support
several cultural services, including aesthetic and educational services and recreational fishing.
Biological effects of acidification in terrestrial ecosystems are generally linked to aluminum
toxicity, which can cause reduced root growth, restricting the ability of the plant to take up water
and nutrients. These direct effects can, in turn, increase the sensitivity of these plants to stresses,
such as droughts, cold temperatures, insect pests, and disease leading to increased mortality of
canopy trees. Terrestrial acidification affects several important ecological services, including
declines in habitat for threatened and endangered species (cultural), declines in forest aesthetics
(cultural), declines in forest productivity (provisioning), and increases in forest soil erosion and
reductions in water retention (cultural and regulating). (U.S. EPA, 2008b)
Deposition of nitrogen is also associated with aquatic and terrestrial nutrient enrichment.
In estuarine waters, excess nutrient enrichment can lead to eutrophication. Eutrophication of
estuaries can disrupt an important source of food production, particularly fish and shellfish
production, and a variety of cultural ecosystem services, including water-based recreational and
aesthetic services. Terrestrial nutrient enrichment is associated with changes in the types and
number of species and biodiversity in terrestrial systems. Excessive nitrogen deposition upsets
the balance between native and nonnative plants, changing the ability of an area to support
biodiversity. When the composition of species changes, then fire frequency and intensity can
also change, as nonnative grasses fuel more frequent and more intense wildfires. (U.S. EPA,
2008b)
5.3.5 Visibility Impairment Benefits
Reducing secondary formation of PM2.5 would improve levels of visibility in the U.S.
because suspended particles and gases degrade visibility by scattering and absorbing light (U.S.
EPA, 2009). Fine particles with significant light-extinction efficiencies include sulfates, nitrates,
organic carbon, elemental carbon, and soil (Sisler, 1996). Visibility has direct significance to
5-36
-------
people's enjoyment of daily activities and their overall sense of wellbeing. Good visibility
increases the quality of life where individuals live and work, and where they engage in
recreational activities. Particulate sulfate is the dominant source of regional haze in the eastern
U.S. and particulate nitrate is an important contributor to light extinction in California and the
upper Midwestern U.S., particularly during winter (U.S. EPA, 2009). Previous analyses (U.S.
EPA, 201 la) show that visibility benefits can be a significant welfare benefit category. Without
air quality modeling, we are unable to estimate visibility-related benefits, and we are also unable
to determine whether the emission reductions associated with the final emission guidelines
would be likely to have a significant impact on visibility in urban areas or Class I areas.
Reductions in emissions of NO2 will improve the level of visibility throughout the United
States because these gases (and the particles of nitrate and sulfate formed from these gases)
impair visibility by scattering and absorbing light (U.S. EPA, 2009). Visibility is also referred to
as visual air quality (VAQ), and it directly affects people's enjoyment of a variety of daily
activities (U.S. EPA, 2009). Good visibility increases quality of life where individuals live and
work, and where they travel for recreational activities, including sites of unique public value,
such as the Great Smoky Mountains National Park (U. S. EPA, 2009).
5.3.6 Water Quality and Availability Benefits
As described in Chapter 4, this proposed rule is expected to lead to shifts in electricity
production away from fossil-fired steam generation towards renewable and natural gas
generation. There are several negative health, ecological, and productivity effects associated with
water effluent and intake from coal generation that will be avoided, and the benefits are
qualitatively described below.12 For additional discussion of these effects and their consequent
effect on welfare, see the Benefit and Cost Analysis for Revisions to the Effluent Limitations
Guidelines and Standards for the Steam Electric Power Generating Point Source Category (U.S.
EPA 2020a).
12 While natural gas combined cycle units also emit wastewater effluents and withdrawal demands, which offset
some of the benefits of reduced fossil steam generation, the scale of these waste streams is much smaller than for
other fossil steam generator types.
5-37
-------
Potential Water Quality Benefits of Reducing Coal-Fired Power Generation
Discharges of wastewater from coal-fired power plants can contain toxic and bio-
accumulative pollutants (e.g., selenium, mercury, arsenic, nickel), halogen compounds
(containing bromide, chloride, or iodide), nutrients, and total dissolved solids (TDS), which can
cause human health and environmental harm through surface water and fish tissue
contamination. Pollutants in coal combustion wastewater are of particular concern because they
can occur in large quantities (i.e., total pounds) and at high concentrations in discharges and
leachate to groundwater and surface waters. These potential beneficial effects follow directly
from reductions in pollutant loadings to receiving waters, and indirectly from other changes in
plant operations. The potential benefits come in the form of reduced morbidity, mortality, and
on environmental quality and economic activities; reduction in water use, which provides
benefits in the form of increased availability of surface water and groundwater; and reductions in
the use of surface impoundments to manage Coal Combustion Residual wastes, with benefits in
the form of avoided cleanup and other costs associated with impoundment releases.
Reducing coal-fired power generation affects human health risk by changing exposure to
pollutants in water via two principal exposure pathways: (1) treated water sourced from surface
waters affected by coal-fired power plant discharges and (2) fish and shellfish taken from
waterways affected by coal-fired power plant discharges. The human health benefits from
surface water quality improvements may include drinking water benefits, fish consumption
benefits, and other complimentary measures.
In addition, reducing coal-fired power generation can affect the ecological condition and
recreation use effects from surface water quality changes. EPA expects the ecological impacts
from reducing coal-fired power plant discharges could include habitat changes for fresh- and
saltwater plants, invertebrates, fish, and amphibians, as well as terrestrial wildlife and birds that
prey on aquatic organisms exposed to pollutants from coal combustion. The change in pollutant
loadings has the potential to result in changes in ecosystem productivity in waterways and the
health of resident species, including threatened and endangered (T&E) species. Loadings from
coal-fired power generation have the potential to impact the general health of fish and
invertebrate populations, their propagation to waters, and fisheries for both commercial and
5-38
-------
recreational purposes. Changes in water quality also have the potential to impact recreational
activities such as swimming, boating, fishing, and water skiing.
Potential economic productivity effects may stem from changes in the quality of public
drinking water supplies and irrigation water; changes in sediment deposition in reservoirs and
navigational waterways; and changes in tourism, commercial fish harvests, and property values.
Drinking Water
Pollutants discharged by coal-fired power plants to surface waters may affect the quality of
water used for public drinking supplies. In turn these impacts to public water supplies have the
potential to affect the costs of drinking water treatment (e.g., filtration and chemical treatment)
by changing eutrophication levels and pollutant concentrations in source waters. Eutrophication
is one of the main causes of taste and odor impairment in drinking water, which has a major
negative impact on public perceptions of drinking water safety. Additional treatment to address
foul tastes and odors to bring the finished water into compliance with EPA's National Secondary
Drinking Water Treatment Standards can significantly increase the cost of public water supply.
Likewise, public drinking water supplies are subject to National Primary Drinking Water
Standards that have set legally enforceable maximum contaminant levels (MCLs), for a number
of pollutants, like metals, discharged from coal-fired power plants. Drinking water systems
downstream from these power plants may be required to treat source water to remove the
contaminants to levels below the MCL in the finished water. This treatment will also increase
costs at drinking water treatment plants. Episodic releases from coal fired power plants, may be
detected only after the completion of a several month round of compliance monitoring at
drinking water treatment plants and there could also by a lag between detection of changes in
source water contaminants and the system implementing treatment to address the issue. This lag
may result in consumers being exposed to these contaminants through ingestion, inhalation, and
skin absorption. The constituents found in the power plant discharge may also interact with
drinking water treatment processes and contribute to the formation of disinfection byproducts
that can have adverse human health impacts.
5-39
-------
Fish Consumption
Recreational and subsistence fishers (and their household members) who consume fish
caught in the reaches downstream of coal-fired power plants may be affected by changes in
pollutant concentrations in fish tissue. See the Benefit and Cost Analysis for Revisions to the
Effluent Limitations Guidelines and Standards for the Steam Electric Power Generating Point
Source Category (U.S. EPA 2020a) for a demonstration of the changes in risk to human health
from exposure to contaminated fish tissue. This document describes the neurological effects to
children ages 0 to 7 from exposure to lead; the neurological effects to infants from in-utero
exposure to mercury; the incidence of skin cancer from exposure to arsenic; and the reduced risk
of other cancer and non-cancer toxic effects.
Changes in Surface Water Quality
Reducing coal-fired power plant discharges may affect the value of ecosystem services
provided by surface waters through changes in the habitats or ecosystems (aquatic and
terrestrial). Society values changes in ecosystem services by a number of mechanisms, including
increased frequency of use and improved quality of the habitat for recreational activities (e.g.,
fishing, swimming, and boating). Individuals also value the protection of habitats and species
that may reside in waters that receive water discharges from coal-plants, even when those
individuals do not use or anticipate future use of such waters for recreational or other purposes,
resulting in nonuse values.
Impacts on Threatened and Endangered Species
For T&E species, even minor changes to reproductive rates and mortality levels may
represent a substantial portion of annual population variation. Therefore, changing the discharge
of coal-fired power plant pollutants to aquatic habitats has the potential to impact the
survivability of some T&E species living in these habitats. The economic value for these T&E
species primarily comes from the nonuse values people hold for the survivorship of both
individual organisms and species survival.
5-40
-------
Changes in Sediment Contamination
Water effluent discharges from coal-fired power plants can also contaminate waterbody
sediments. For example, sediment adsorption of arsenic, selenium, and other pollutants found in
water discharges can result in accumulation of contaminated sediment on stream and lake beds,
posing a particular threat to benthic (i.e., bottom-dwelling) organisms. These pollutants can later
be re-released into the water column and enter organisms at different trophic levels.
Concentrations of selenium and other pollutants in fish tissue of organisms of lower trophic
levels can bio-magnify through higher trophic levels, posing a threat to the food chain at large
(Ruhletal., 2012).
Reservoir Capacity and Sedimentation Changes in Navigational Waterways
Reservoirs serve many functions, including storage of drinking and irrigation water
supplies, flood control, hydropower supply, and recreation. Streams can carry sediment into
reservoirs, where it can settle and cause buildup of sediment layers over time, reducing reservoir
capacity (Graf et al., 2010, 2011) and the useful life of reservoirs unless measures such as
dredging are taken to reclaim capacity (Hargrove et al., 2010; Miranda, 2017). Likewise,
navigable waterways, including rivers, lakes, bays, shipping channels and harbors, are prone to
reduced functionality due to sediment build-up, which can reduce the navigable depth and width
of the waterway (Clark et al., 1985; Ribaudo and Johansson, 2006). For many navigable waters,
periodic dredging is necessary to remove sediment and keep them passable. Dredging of
reservoirs and navigable waterways can be costly. EPA expects that changes in suspended solids
effluent discharge from coal-fired power plants could reduce sediment loadings to surface waters
decreasing reservoir and navigable waterway maintenance costs by changing the frequency or
volume of dredging activity.
Changes in Water Consumption and Withdrawals
A reduction in water consumption from coal fired power plants may benefit aquatic and
riparian species downstream of the power plant intake through the provision of additional water
resources in the face of drying conditions and increased rainfall variability. In a study completed,
in 2011, by the U.S. Department of Energy's National Renewable Energy Laboratory (U.S. DOE
5-41
-------
2011), water consumption, which is defined as water removed from the immediate water
environment and can include cooling water evaporation, cleaning, and process related water use
including flue gas desulfurization, was found to range from 100 - 1,100 gal/MWh at generic coal
power plants. This study also found that water withdraws, defined as the amount of water
removed from the ground or diverted from a water source for use, ranged from 300 - 50,000
gal/MWh at a generic coal power plant. Reductions in water consumption and withdraws will
lower the number of aquatic organisms impinged and entrained by the power plant's water
filtration and cooling systems.
5.3.7 Hazardous Air Pollutant Impacts
The proposed rule is expected to reduce fossil-fired EGU generation by up to 8 percent
per year and consequentially is expected to lead to reduced HAP emissions. HAP emissions from
EGUs create risks of premature mortality from heart attacks, cancer, and neurodevelopmental
delays in children, and detrimentally affect economically vital ecosystems used for recreational
and commercial purposes. Further, these public health effects are particularly pronounced for
certain segments of the American population that are especially vulnerable (e.g., subsistence
fishers and their children) to impacts from EGU HAP emissions.
5.4 References
Amorim MIM, Mergler D, Bahia MO, Dubeau H, Miranda D, Lebel J, et al. 2000. Cytogenetic
damage related to low levels of methyl mercury contamination in the Brazilian Amazon.
An Acad Bras Cienc 72:497-507; doi:10.1590/S0001- 37652000000400004.
Anthoff, D. and Tol, R.S.J. 2013a. The uncertainty about the social cost of carbon: a
decomposition analysis using FUND. Climatic Change 117: 515-530.
Anthoff, D. and Tol, R.S.J. 2013b. Erratum to: The uncertainty about the social cost of carbon: A
decomposition analysis using FUND. Climatic Change 121(2): 413.
ATSDR. 1999. ATSDR - Toxicological Profile: Mercury.
Bell, M.L., A. McDermott, S.L. Zeger, J.M. Sarnet, and F. Dominici. 2004. "Ozone and Short-
Term Mortality in 95 U.S. Urban Communities, 1987-2000." Journal of the American
Medical Association. 292(19):2372-8.
Bell, M.L., F. Dominici, and J.M. Samet. 2005. "A Meta-Analysis of Time-Series Studies of
Ozone and Mortality with Comparison to the National Morbidity, Mortality, and Air
Pollution Study." Epidemiology. 16(4):436-45.
5-42
-------
Blunden, J., and D.S. Arndt, Eds. 2020. State of the Climate in 2019. Bull. Amer. Meteor. Soc,
S1-S429, https://doi.org/10.1175/2020BAMSStateof theClimate.l.
Center for Climate and Security (CCS), The. 2018. Military Expert Panel Report Sea Level Rise
and the U.S. Military's Mission. Available at:
https://climateandsecurity.files.wordpress.com/2018/02/militaryexpert-panel-report_sea-
level-rise-and-the-us-militarys-mission_2nd-edition_02_2018.pdf (accessed February 5,
2021)
Clark, E. H., Haverkamp, J. A., and Chapman, W. 1985. Eroding soils. The off-farm impacts:
Conservation Foundation.
Di, Q, Wang, Y, Zanobetti, A, Wang, Y, Koutrakis, P, Choirat, C, Dominici, F and Schwartz, JD
(2017). Air pollution and mortality in the Medicare population. New Engl J Med 376(26):
2513-2522.
Executive Order (E.O.) 13783. Promoting Energy Independence and Economic Growth March
28, 2017. Available at: https://www.federalregister.gov/documents/2017/03/31/2017-
06576/promoting-energy-independence-and-economic-growth (accessed February 5,
2021).
Executive Order (E.O.) 13990. Protecting Public Health and the Environment and Restoring
Science to Tackle the Climate Crisis. January 20, 2021. Available at:
https://www.federalregister.gOv/documents/2021/01/25/2021-01765/protecting-public-
health-and-the-environment-and-restoring-science-to-tackle-the-climate-crisis (accessed
February 5, 2021).
Fann, N.; Coffman, E.; Jackson, M.; Jhun, I.; Lamichhane, A. P.; Nolte, C. G.; Roman, H.;
Sacks, J. D. The Role of Temperature in Modifying the Risk of Ozone-Attributable
Mortality under Future Changes in Climate: A Proof-of-Concept Analysis. Environ. Sci.
Technol. 2021. https://doi.org/10.1021/acs.est.lc05975.
Fann N, Nolte CG, Dolwick P, Spero TL, Brown AC, Phillips S, et al. 2015. The geographic
distribution and economic value of climate change-related ozone health impacts in the
United States in 2030. J Air Waste Manag Assoc 65;
doi: 10.1080/10962247.2014.996270.
Graf, W. L., Wohl, E., Sinha, T., & Sabo, J. L. 2010. Sedimentation and sustainability of western
American reservoirs. Water Resources Research, 46(12).
Graf, W. L., Wohl, E., Sinha, T., & Sabo, J. L. 2011. Correction to "Sedimentation and
sustainability of western American reservoirs." Water Resources Research, 47(8).
Gwinn, M.R., J. Craig, D.A. Axelrad, R. Cook, C. Dockins, N. Fann, R. Fegley, D.E. Guinnup,
G. Helfand, B. Hubbell, S.L. Mazur, T. Palma, R.L. Smith, J. Vandenberg, and
B. Sonawane. 2011. "Meeting report: Estimating the benefits of reducing hazardous air
pollutants—summary of 2009 workshop and future considerations." Environmental
Health Perspectives. 119(1): 125-30.
5-43
-------
Hargrove, W. L., Johnson, D., Snethen, D., & Middendorf, J. 2010. From dust bowl to mud
bowl: sedimentation, conservation measures, and the future of reservoirs. Journal of Soil
and Water Conservation, 65(1):14A-17A.
Hope, C. 2013. Critical issues for the calculation of the social cost of CO2: why the estimates
from PAGE09 are higher than those from PAGE2002. Climatic Change, 117: 531-543.
Huang Y, Dominici F, Bell M. 2004. Bayesian Hierarchical Distributed Lag Models for Summer
Ozone Exposure and Cardio-Respiratory Mortality. Johns Hopkins Univ Dept Biostat
Work Pap Ser.
Intergovernmental Panel on Climate Change (IPCC). 2018. Global Warming of 1.5 °C. An IPCC
Special Report on the impacts of global warming of 1.5 °C above pre-industrial levels
and related global greenhouse gas emission pathways, in the context of strengthening the
global response to the threat of climate change, sustainable development, and efforts to
eradicate poverty [Masson-Delmotte, V., P. Zhai, H.-O. Po' rtner, D. Roberts, J. Skea,
P.R. Shukla, A. Pirani, W. Moufouma-Okia, C. Pe'an, R. Pidcock, S. Connors, J.B.R.
Matthews, Y. Chen, X. Zhou, M.I. Gomis, E. Lonnoy, T. Maycock, M. Tignor, and T.
Waterfield (eds.)].
Intergovernmental Panel on Climate Change (IPCC). 2019a. Climate Change and Land: an IPCC
special report on climate change, desertification, land degradation, sustainable land
management, food security, and greenhouse gas fluxes in terrestrial ecosystems [P.R.
Shukla, J. Skea, E. Calvo Buendia, V. Masson-Delmotte, H.-O. Po' rtner, D.C. Roberts, P.
Zhai, R. Slade, S. Connors, R. van Diemen, M. Ferrat, E. Haughey, S. Luz, S. Neogi, M.
Pathak, J. Petzold, J. Portugal Pereira, P. Vyas, E. Huntley, K. Kissick, M. Belkacemi, J.
Malley, (eds.)]
Intergovernmental Panel on Climate Change (IPCC). 2019b. IPCC Special Report on the Ocean
and Cryosphere in a Changing Climate [H.-O. Po "rtner, D.C. Roberts, V. Masson-
Delmotte, P. Zhai M. Tignor, E. Poloczanska, K. Mintenbeck, A. Alegn'a, M. Nicolai, A.
Okem, J. Petzold, B. Rama, N.M. Weyer (eds.)].
Intergovernmental Panel on Climate Change (IPCC). 2021. Summary for Policymakers. In:
Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to
the Sixth Assessment Report of the Intergovernmental Panel on Climate Change
[Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. Pe'an, S. Berger, N. Caud, Y.
Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews,
T.K. Maycock, T. Waterfield, O. Yelekc^, R. Yu and B. Zhou (eds.)]. Cambridge
University Press. In Press.
Ito, K., S.F. De Leon, and M. Lippmann. 2005. "Associations Between Ozone and Daily
Mortality: Analysis and Meta-Analysis." Epidemiology. 16(4):446-57.
Jerrett M, Burnett RT, Pope CA, Ito K, Thurston G, Krewski D, et al. 2009. Long-term ozone
exposure and mortality. N Engl J Med 360:1085-95; doi:10.1056/NEJMoa0803894.
5-44
-------
Jhun I, Fann N, Zanobetti A, Hubbell B. 2014. Effect modification of ozone-related mortality
risks by temperature in 97 US cities. Environment International. 73:128-34.
Katsouyanni, K, Samet, JM, Anderson, HR, Atkinson, R, Le Tertre, A, Medina, S, Samoli, E,
Touloumi, G, Burnett, RT, Krewski, D, Ramsay, T, Dominici, F, Peng, RD, Schwartz, J,
Zanobetti, A and Committee, HEIHR (2009). Air pollution and health: a European and
North American approach (APHENA). Res Rep Health Eff Inst(142): 5-90.
Kivi, PA and Shogren, JF (2010). Second-order ambiguity in very low probability risks: food
safety valuation. J Agric Resour Econ: 443-456.
Krewski D., M. Jerrett, R.T. Burnett, R. Ma, E. Hughes, Y. Shi, et al. 2009. Extended Follow-Up
and Spatial Analysis of the American Cancer Society Study Linking Particulate Air
Pollution and Mortality. HEI Research Report, 140, Health Effects Institute, Boston, MA.
Lepeule, J., F. Laden, D. Dockery, and J. Schwartz. 2012. "Chronic Exposure to Fine Particles
and Mortality: An Extended Follow-Up of the Harvard Six Cities Study from 1974 to
2009." Environmental Health Perspectives. 120(7):965-70.
Levy, J.I., S.M. Chemerynski, and J.A. Sarnat. 2005. "Ozone Exposure and Mortality: An
Empiric Bayes Metaregression Analysis." Epidemiology. 16(4):458-68.
McGartland, A, Revesz, R, Axelrad, DA, Dockins, C, Sutton, P and Woodruff, TJ. 2017.
Estimating the health benefits of environmental regulations. Science 357(6350): 457-458.
Miranda, L. E. (2017). Section 3: Sedimentation. In: Reservoir Fish Habitat Management.
Totowa, New Jersey: Lightning Press.
National Academies of Sciences, Engineering, and Medicine. 2016. Attribution of Extreme
Weather Events in the Context of Climate Change. Washington, DC: The National
Academies Press, https://dio.org/10.17226/21852.
National Academies of Sciences, Engineering, and Medicine (National Academies). 2017.
Valuing Climate Damages: Updating Estimation of the Social Cost of Carbon Dioxide.
Washington, D.C.: National Academies Press.
National Academies of Sciences, Engineering, and Medicine. 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 Academies of Sciences, Engineering, and Medicine. 2019. Climate Change and
Ecosystems. Washington, DC: The National Academies Press.
https://doi.org/10.17226/25504.
National Research Council (NRC). 2000. Toxicological Effects of Methylmercury: Committee
on the Toxicological Effects of Methylmercury." Board on Environmental Studies and
Toxicology. National Academies Press. Washington, DC.
5-45
-------
National Research Council (NRC). 2002. Estimating the Public Health Benefits of Proposed Air
Pollution Regulations. National Academies Press. Washington, DC.
National Research Council (NRC). 2008. Estimating Mortality Risk Reduction and Economic
Benefits from Controlling Ozone Air Pollution. National Academies Press. Washington,
DC.
Nordhaus, W. 2010. Economic aspects of global warming in a post-Copenhagen environment.
Proceedings of the National Academy of Sciences, 107(26): 11721-11726.
Office of Management and Budget (OMB). 2003. Circular A-4: Regulatory Analysis.
Washington, DC. Available at: < http://www.whitehouse.gov/omb/circulars/a004/a-
4.html>.
Pizer, W., M. Adler, J. Aldy, D. Anthoff, M. Cropper, K. Gillingham, M. Greenstone, B. Murray,
R. Newell, R. Richels, A. Rowell, S. Waldhoff, J. Wiener. 2014. "Using and improving
the social cost of carbon." Science, Vol. 346, No. 6214, 12/05/14, pp 1189-1190.
Ren, C., G.M. Williams, K. Mengersen, L. Morawska, and S. Tong. 2008a. "Does Temperature
Modify Short-Term Effects of Ozone on Total Mortality in 60 Large Eastern U.S.
Communities? An Assessment Using the NMMAPS Data." Environment International.
34:451-458.
Ren, C., G.M. William, L. Morawska, K. Mengensen, and S. Tong. 2008b. "Ozone Modifies
Associations between Temperature and Cardiovascular Mortality: Analysis of the
NMMAPS Data." Occupational and Environmental Medicine. 65:255-260.
Ribaudo, M., and Johansson, R. 2006. Water Quality: Impacts on Agriculture. In Keith Wiebe &
Noel Gollehon (Eds.), Agricultural Resources and Environmental Indicators, 2006 Edition
(EIB-16). Washington DC: Economic Research Service, U.S. Department of Agriculture.
Ruhl, L., Vengosh, A., Dwyer, G. S., Hsu-Kim, H., Schwartz, G., Romanski, A., and Smith, S.
D. 2012. The impact of coal combustion residue effluent on water resources: a North
Carolina example. Environmental Science & Technology, 46(21), 12226-12233.
Sisler, J.F. 1996. Spatial and Seasonal Patterns and Long-Term Variability of the Composition of
the Haze in the United States: An analysis of data from the IMPROVE network. CIRA
Report, ISSN 0737-5352-32, Colorado State University.
Schwartz, J. 2005. "How Sensitive is the Association between Ozone and Daily Deaths to
Control for Temperature?" American Journal of Respiratory and Critical Care Medicine.
171(6): 627-31.
Smith RL, Xu B, Switzer P. 2009. Reassessing the relationship between ozone and short-term
mortality in U.S. urban communities. Inhal Toxicol 21 Suppl 2:37-61;
doi: 10.1080/08958370903161612.
5-46
-------
Turner, MC, Jerrett, M, Pope, A, III, Krewski, D, Gapstur, SM, Diver, WR, Beckerman, BS,
Marshall, JD, Su, J, Crouse, DL and Burnett, RT (2016). Long-term ozone exposure and
mortality in a large prospective study. Am J Respir Crit Care Med 193(10): 1134-1142.
U.S. Department of Defense (DOD), Office of the Undersecretary for Policy (Strategy, Plans,
and Capabilities). 2021. Department of Defense Climate Risk Analysis. Report Submitted
to National Security Council, https://media.defense.gov/2021/0ct/21/2002877353/-l/-
1/O/DOD-CLIMATE-RISK-AN ALYSIS-FINAL.PDF.
U.S. Department of Defense (DOD). 2014. Climate Change Adaptation Roadmap. Available at:
https://www.acq.osd.mil/eie/downloads/CCARprint_wForward_e.pdf (accessed February
5, 2021).
U.S. Department of Energy (U.S. DOE) National Renewable Energy Laboratory. 2011. A
Review of Operational Water Consumption and Withdrawal Factors for Electricity
Generating Technologies. Technical Report NREL/TP-6A20-50900.
https://www.nrel.gov/docs/fyllosti/50900.pdf
U.S. Environmental Protection Agency (U.S. EPA). 2002. Methylmercury (MeHg) CASRN
22967-92-6 | IRIS | US EPA, ORD
U.S. Environmental Protection Agency (U.S. EPA). 2008b. Integrated Science Assessment for
Oxides of Nitrogen and Sulfur-Ecological Criteria National (Final Report). National
Center for Environmental Assessment - RTP Division, Research Triangle Park, NC.
EPA/600/R-08/139. December. Available at:
.
U.S. Environmental Protection Agency (U.S. EPA). 2009. Integrated Science Assessment for
Particulate Matter (Final Report). EPA-600-R-08-139F. National Center for
Environmental Assessment - RTP Division, Research Triangle Park, NC. December.
Available at: .
U.S. Environmental Protection Agency (U.S. EPA). 2010a. Integrated Science Assessment for
Carbon Monoxide. National Center for Environmental Assessment, Research Triangle
Park, NC. EPA/600/R-09/019F. January. Available at:
.
U.S. Environmental Protection Agency (U.S. EPA). 2010b. Technical Support Document:
Summary of Expert Opinions on the Existence of a Threshold in the Concentration-
Response Function for PM2.5-related Mortality. Research Triangle Park, NC. June.
Available at: .
U.S. Environmental Protection Agency (U.S. EPA). 2010c. Regulatory Impact Analysis (RIA)
for Existing Stationary Compression Ignition Engines NESHAP Final Draft.
U.S. Environmental Protection Agency (U.S. EPA). 2010d. Regulatory Impact Analysis for the
Proposed Federal Transport Rule.
5-47
-------
U.S. Environmental Protection Agency (U.S. EPA). 201 la. Policy Assessment for the Review of
the Particulate Matter National Ambient Air Quality Standards. EPA-452/D-11-003.
Office of Air Quality Planning and Standards, Health and Environmental Impacts
Division. April. Available at:
.
U.S. Environmental Protection Agency (U.S. EPA). 201 lb. The Benefits and Costs of the Clean
Air Act from 1990 to 2020. Office of Air and Radiation, Washington, DC. March.
Available at: .
U.S. Environmental Protection Agency (U.S. EPA). 201 lc. Regulatory Impact Analysis for the
Federal Implementation Plans to Reduce Interstate Transport of Fine Particulate Matter
and Ozone in 27 States; Correction of SIP Approvals for 22 States.
U.S. Environmental Protection Agency (U.S. EPA). 201 Id. Regulatory Impact Analysis for the
Final Mercury and Air Toxics Standards.
U.S. Environmental Protection Agency (U.S. EPA). 2012. Regulatory Impact Analysis for the
Final Revisions to the National Ambient Air Quality Standards for Particulate Matter.
EPA-452/R-12-003. Office of Air Quality Planning and Standards, Health and
Environmental Impacts Division, Research Triangle Park, NC. Available at: <
http://www.epa.gov/ttnecasl/regdata/RIAs/finalria.pdf>.
U.S. Environmental Protection Agency (U.S. EPA). 2013a. Integrated Science Assessment of
Ozone and Related Photochemical Oxidants (Final Report). EPA/600/R-10/076F.
National Center for Environmental Assessment - RTP Division, Research Triangle Park.
Available at: .
U.S. Environmental Protection Agency (U.S. EPA). 2013b. Regulatory Impact Analysis for the
Final Revisions to the National Ambient Air Quality Standards for Particulate Matter.
U.S. Environmental Protection Agency (U.S. EPA). 2014a. Regulatory Impact Analysis (RIA)
for Proposed Residential Wood Heaters NSPS Revision.
U.S. Environmental Protection Agency (U.S. EPA). 2014b. Regulatory Impact Analysis for the
Proposed Carbon Pollution Guidelines for Existing Power Plants and Emission Standards
for Modified and Reconstructed Power Plants.
U.S. Environmental Protection Agency (U.S. EPA). 2014c. Regulatory Impact Analysis of the
Proposed Revisions to the National Ambient Air Quality Standards for Ground-Level
Ozone.
U.S. Environmental Protection Agency (U.S. EPA). 2015a. Regulatory Impact Analysis for
Residential Wood Heaters NSPS Revision: Final Report.
U.S. Environmental Protection Agency (U.S. EPA). 2015b. Regulatory Impact Analysis for the
Clean Power Plan Final Rule.
5-48
-------
U.S. Environmental Protection Agency (U.S. EPA). 2015c. Regulatory Impact Analysis for the
Proposed Cross-State Air Pollution Rule (CSAPR) Update for the 2008 Ozone National
Ambient Air Quality Standards (NAAQS).
U.S. Environmental Protection Agency (U.S. EPA). 2015d. Regulatory Impact Analysis for the
Proposed Federal Plan Requirements for Greenhouse Gas Emissions from Electric Utility
Generating Units Constructed on or Before January 8, 2014; Model Trading Rules;
Amendments to Framework Regulations. 4-52
U.S. Environmental Protection Agency (U.S. EPA). 2015e. Regulatory Impact Analysis of the
Final Revisions to the National Ambient Air Quality Standards for Ground-Level Ozone,
EPA-452/R-15-07. EPA-452/R-12-003. Office of Air Quality Planning and Standards,
Health and Environmental Impacts Division, Research Triangle Park, NC.
(https://www.epa.gOv/sites/production/files/2016-02/documents/20151001ria.pdf).
U.S. Environmental Protection Agency (U.S. EPA). 2016a. Guidelines for Preparing Economic
Analyses.
U.S. Environmental Protection Agency (U.S. EPA). 2016b. Regulatory Impact Analysis of the
Cross-State Air Pollution Rule (CSAPR) Update for the 2008 National Ambient Air
Quality Standards for Ground-Level Ozone.
U.S. Environmental Protection Agency (U.S. EPA). 2016c. Integrated Science Assessment for
Oxides of Nitrogen - Health Criteria (Final Report). National Center for Environmental
Assessment, Research Triangle Park, NC. July. Available at: <
http s: //cfpub. epa. gov/ncea/i sa/ recordi spl ay. cfm? dei d=310879>.
U.S. Environmental Protection Agency (U.S. EPA). 2017. Integrated Science Assessment for
Sulfur Oxides—Health Criteria (Final Report). National Center for Environmental
Assessment - RTP Division, Research Triangle Park, NC. September. Available at:
.
U.S. Environmental Protection Agency (U.S. EPA). 2018. Environmental Benefits Mapping and
Analysis Program - Community Edition. User's Manual. Office of Air Quality Planning
and Standards, Health and Environmental Impacts Division, Research Triangle Park, NC.
Available at: < https://www.epa.gov/sites/production/files/2015-04/documents/benmap-
ce_user_manual_march_2015 ,pdf>
U.S. Environmental Protection Agency (U.S. EPA). 2019a. Integrated Science Assessment (ISA)
for Particulate Matter (Final Report, 2019). U.S. Environmental Protection Agency,
Washington, DC, EPA/600/R-19/188, 2019.
U.S. Environmental Protection Agency (U.S. EPA). 2019b. Regulatory Impact Analysis for
Repeal of the Clean Power Plan; Emission Guidelines for Greenhouse Gas Emissions
from Existing Electric Utility Generating Units; Revisions to Emission Guidelines
Implementing Regulations. EPA-452/R19-003. June 2019.
U.S. Environmental Protection Agency (U.S. EPA). 2020a. Benefit and Cost Analysis for
5-49
-------
Revisions to the Effluent Limitations Guidelines and Standards for the Steam Electric
Power Generating Point Source Category. U.S. Environmental Protection Agency,
Washington, DC, EPA-821-R-20-003. Available at:
.
U.S. Environmental Protection Agency (U.S. EPA). 2020b. Integrated Science Assessment (ISA)
for Ozone and Related Photochemical Oxidants (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-20/012, 2020.
U.S. Environmental Protection Agency (U.S. EPA). 2020c. Regulatory Impact Analysis for the
Proposed Cross-State Air Pollution Rule (CSAPR) Revised Update for the 2008 Ozone
National Ambient Air Quality Standards (NAAQS).
U.S. Environmental Protection Agency (U.S. EPA). 2020d. Integrated Science Assessment (ISA)
for Oxides of Nitrogen, Oxides of Sulfur and Particulate Matter Ecological Criteria (Final
Report). U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-20/278,
2020.
U.S. Environmental Protection Agency (U.S. EPA). 2021a. Regulatory Impact Analysis for
the Final Revised Cross-State Air Pollution Rule (CSAPR) Update for the 2008 Ozone
National Ambient Air Quality Standards (NAAQS). U.S. Environmental Protection
Agency, Research Triangle Park, NC, EPA-452/R-21-002, March 2021.
U.S. Environmental Protection Agency (U.S. EPA). 2021b. Technical Support Document:
Estimating the Benefit per Ton of Reducing Directly-Emitted PM2.5, PM2.5 Precursors
and Ozone Precursors from 21 Sectors. U.S. Environmental Protection Agency, Research
Triangle Park, NC, December 2021. Available at:
https://www.epa.gov/benmap/estimating-benefit-ton-reducing-pm25-precursors-21-
sectors
U.S. Environmental Protection Agency (U.S. EPA). 2021c. Climate Change and Social
Vulnerability in the United States: A Focus on Six Impacts. U.S. Environmental
Protection Agency, EPA 430-R-21-003.
U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2000. An
SAB Report on EPA's White Paper Valuing the Benefits of Fatal Cancer Risk Reduction.
EPA-SAB-EEAC-00-013. July. Available at:
.
U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2004.
Advisory Council on Clean Air Compliance Analysis Response to Agency Request on
Cessation Lag. EPA-COUNCIL-LTR-05-001. December. Available at:
-------
U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2008.
Subject: Benefits of Reducing Benzene Emissions in Houston, 1990- 2020.
U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2009a.
Review of EPA's Integrated Science Assessment for Particulate Matter (First External
Review Draft, December 2008). EPA-COUNCIL-09-008. May. Available at:
.
U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2009b.
Review of Integrated Science Assessment for Particulate Matter (Second External
Review Draft, July 2009). EPA-CASAC-10-001. November. Available at:
.
U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2010.
Review of EPA's Draft Health Benefits of the Second Section 812 Prospective Study of
the CAA
U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2017. Letter
from Peter S. Thorne, Chair, Science Advisory Board, and Madhu Khanna, Chair, SAB
Environmental Economics Advisory Committee, to Administrator Scott Pruitt. Re: SAB
Review of EPA's Proposed Methodology for Updating Mortality Risk Valuation
Estimates for Policy Analysis. February 23, 2017. EPA-SAB-17-005. Office of the
Administrator, Science Advisory Board U.S. EPA HQ, Washington DC. Available at:
https://yosemite.epa.gov/sab/sabproduct.nsf/436C15984A487ElF852580D0004429CF/$
File/EP A-S AB+2017-005+Unsigned.pdf.
U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2019. Letter
from Louis Anthony Cox, Jr., Chair, Clean Air Scientific Advisory Committee, to
Administrator Andrew R. Wheeler. Re: CASAC Review of the EPA's Integrated
Science Assessment for Particulate Matter (External Review Draft - October 2018).
April 11, 2019. EPA-CASAC-19-002. Office of the Administrator, Science Advisory
Board U.S. EPA HQ, Washington DC. Available at:
https://yosemite.epa.gov/sab/sabproduct.nsf/LookupWebReportsLastMonthCASAC/932
D1DF8C2A9043F852581000048170D?OpenDocument&TableRow=2.3#2.
U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2020a.
Letter from Louis Anthony Cox, Jr., Chair, Clean Air Scientific Advisory Committee, to
Administrator Andrew R. Wheeler. Re:CASAC Review of the EPA's Integrated Science
Assessment for Ozone and Related Photochemical Oxidants (External Review Draft -
September 2019). February 19, 2020. EPA-CASAC-20-002. Office of the Adminstrator,
Science Advisory Board Washington, DC Availbale at:
https://yosemite.epa.gov/sab/sabproduct.nsf/264cbl227d55e02c85257402007446a4/F228
E5D4D848BBED852585150063 54D0/$File/EP A-CASAC-20-002.pdf.
5-51
-------
U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2020b.
Letter from Michael Honeycutt Chair, Scientific Advisory, to Administrator Lisa
Jackson. Re: Science Advisory Board (SAB) Consideration of the Scientific and
Technical Basis of EPA's Proposed Rule titled "Increasing Consistency and
Transparency in Considering Benefits and Costs in the Clean Air Act Rulemaking
Process.". Septemeber 30, 2020. EPA-SAB-20-012. Office of the Administrator, Science
Advisory Board U.S. EPA HQ, Washington DC. Available at:
https://yosemite.epa.gov/sab/sabproduct.nsf/LookupWebReportsLastMonthBOARD/0A3
12659C8 AC 185D852585F80049803C/$File/EP A-SAB-20-012.pdf.
U.S. Global Change Research Program (USGCRP). 2016. The Impacts of Climate Change on
Human Health in the United States: A Scientific Assessment. Crimmins, A., J. Balbus,
J.L. Gamble, C.B. Beard, J.E. Bell, D. Dodgen, R.J. Eisen, N. Fann, M.D. Hawkins, S.C.
Herring, L. Jantarasami, D.M. Mills, S. Saha, M.C. Sarofim, J. Trtanj, and L. Ziska, Eds.
U.S. Global Change Research Program, Washington, DC, 312 pp.
U.S. Global Change Research Program (USGCRP). 2017. Climate Science Special Report:
Fourth National Climate Assessment, Volume I [Wuebbles, D.J., D.W. Fahey, K.A.
Hibbard, D.J. Dokken, B.C. Stewart, and T.K. Maycock (eds.)]. U.S. Global Change
Research Program, Washington, DC, USA, 470 pp, doi: 10.7930/J0J964J6.
U.S. Global Change Research Program (USGCRP). 2018. Impacts, Risks, and Adaptation in the
United States: Fourth National Climate Assessment, Volume II [Reidmiller, D.R., C.W.
Avery, D.R. Easterling, K.E. Kunkel, K.L.M. Lewis, T.K. Maycock, and B.C. Stewart
(eds.)]. U.S. Global Change Research Program, Washington, DC, USA, 1515 pp. doi:
10.7930/NCA4.2018
Woodruff, T.J., J. Grillo, and K.C. Schoendorf. 2008. "Air pollution and postneonatal infant
mortality in the United States, 1999-2002." Environmental Health Perspectives. 116(1):
110-115
Woods & Pool (2015). Complete Demographic Database.
Zanobetti A, Schwartz J. 2008. Mortality displacement in the association of ozone with
mortality: an analysis of 48 cities in the United States. Am J Respir Crit Care Med
177:184-9; doi: 10.1164/rccm.200706-8230C.F
5-52
-------
CHAPTER 6: ECONOMIC IMPACTS
Overview
Economic impact analyses focus on changes in market prices and output levels. If
changes in market prices and output levels in the primary markets are significant enough,
impacts on other markets may also be examined. Both the magnitude of costs needed to comply
with a proposed rule and the distribution of these costs among affected facilities can have a role
in determining how the market will change in response to a rule. This chapter analyzes the
potential impacts on small entities and the potential labor impacts associated with this proposed
rulemaking. For additional discussion of impacts on fuel use and electricity prices, see Chapter 4,
Section 4.5.3.
6.1 Small Entity Analysis
For the proposed rule, the EPA performed a small entity screening analysis for impacts
on all affected EGUs and non-EGU facilities1 by comparing compliance costs to historic
revenues at the ultimate parent company level. This is known as the cost-to-revenue or cost-to-
sales test, or the "sales test." The sales test is an impact methodology the EPA employs in
analyzing entity impacts as opposed to a "profits test," in which annualized compliance costs are
calculated as a share of profits. The sales test is frequently used because revenues or sales data
are commonly available for entities impacted by the EPA regulations, and profits data normally
made available are often not the true profit earned by firms because of accounting and tax
considerations. Also, the use of a sales test for estimating small business impacts for a
rulemaking is consistent with guidance offered by the EPA on compliance with the Regulatory
Flexibility Act (RFA)2 and is consistent with guidance published by the U.S. Small Business
Administration's (SB A) Office of Advocacy that suggests that cost as a percentage of total
revenues is a metric for evaluating cost increases on small entities in relation to increases on
large entities (SBA, 2017).
1 The facilities were identified in the Screening Assessment of Potential Emissions Reductions, Air Quality Impacts,
and Costs from Non-EGU Emissions Units for 2026, or non-EGU screening assessment, available in the docket.
2 The RFA compliance guidance to the EPA rule writers can be found at
6-1
-------
6.1.1 EGU Small Entity Analysis and Results
This section presents the methodology and results for estimating the impact of the
proposal on small EGU entities in 2023 and in 2026 based on the following endpoints:
• annual economic impacts of the proposal on small entities, and
• ratio of small entity impacts to revenues from electricity generation.
In this analysis, EPA considered EGUs that are subject to the proposed FIP and meet the
following five criteria: 1) EGU is represented in NEEDS v6; 2) EGU is fossil fuel-fired; 3) EGU
is located in a state covered by this proposal; 4) EGU is neither a cogeneration unit nor solid
waste incineration unit; and 5) EGU capacity is 25 Megawatt (MW) or larger. EPA next refined
this list of EGUs, narrowing it to those that exhibit at least one of the following changes, in
comparison to the baseline. Please see Chapter 4, Section 4.3 for more discussion of the power
sector modeling.
• Summer fuel use (BTUs) changes by +/- 1 percent or more
• Summer generation (GWh) changes by +/- 1 percent or more
• NOx summer emissions (tons) changes by +/- 1 percent or more
Based on these criteria, EPA identified a total of 130 potentially affected EGUs
warranting examination in 2023 and 481 potentially affected EGUs warranting examination in
2026 in this RFA analysis. Next, we determined power plant ownership information, including
the name of associated owning entities, ownership shares, and each entity's type of ownership.
We primarily used data from Ventyx, supplemented by limited research using publicly available
data.3 Majority owners of power plants with affected EGUs were categorized as one of the seven
ownership types.4 These ownership types are:
1. Investor-Owned Utility (IOU): Investor-owned assets (e.g., a marketer, independent
power producer, financial entity) and electric companies owned by stockholders, etc.
3 The Ventyx Energy Velocity Suite database consists of detailed ownership and corporate affiliation information at
the EGU level. For more information, see: www.ventyx.com.
4 Throughout this analysis, EPA refers to the owner with the largest ownership share as the "majority owner" even
when the ownership share is less than 51 percent.
6-2
-------
2. Cooperative (Co-Op): Non-profit, customer-owned electric companies that generate
and/or distribute electric power.
3. Municipal: A municipal utility, responsible for power supply and distribution in a small
region, such as a city.
4. Sub-division: Political subdivision utility is a county, municipality, school district,
hospital district, or any other political subdivision that is not classified as a municipality
under state law.
5. Private: Similar to an investor-owned utility, however, ownership shares are not openly
traded on the stock markets.
6. State: Utility owned by the state.
7. Federal: Utility owned by the federal government.
Next, EPA used both the D&B Hoover's online database and the Ventyx database to
identify the ultimate owners of power plant owners identified in the Ventyx database. This was
necessary, as many majority owners of power plants (listed in Ventyx) are themselves owned by
other ultimate parent entities (listed in D&B Hoover's).5 In these cases, the ultimate parent entity
was identified via D&B Hoover's, whether domestically or internationally owned.
EPA followed SBA size standards to determine which non-government ultimate parent
entities should be considered small entities in this analysis. These SBA size standards are
specific to each industry, each having a threshold level of either employees, revenue, or assets
below which an entity is considered small.6 SBA guidelines list all industries, along with their
associated North American Industry Classification System (NAICS) code7 and SBA size
standard. Therefore, it was necessary to identify the specific NAICS code associated with each
ultimate parent entity in order to understand the appropriate size standard to apply. Data from
D&B Hoover's was used to identify the NAICS codes for most of the ultimate parent entities. In
5 The D&B Hoover's online platform includes company records that can contain NAICS codes, number of
employees, revenues, and assets. For more information, see: https://www.dnb.com/products/marketing-sales/dnb-
hoovers.html.
6 SBA's table of size standards can be located here: https://www.sba.gov/document/support--table-size-standards.
7 North American Industry Classification System can be accessed at the following link:
https://www.census.gov/naics/
6-3
-------
many cases, an entity that is a majority owner of a power plant is itself owned by an ultimate
parent entity with a primary business other than electric power generation. Therefore, it was
necessary to consider SBA entity size guidelines for the range of NAICS codes listed in Table
4-3. This table represents the range of NAICS codes and areas of primary business of ultimate
parent entities that are majority owners of potentially affected EGUs in EPA's IPM base case.
Table 6-1. SBA Size Standards by NAICS Code
NAICS Codes
NAICS U.S. Industry Title
Size
Standards (millions of
dollars)
Size
Standards (number of
employees)
221111
Hydroelectric Power
Generation
500
221112
Fossil Fuel Electric Power
Generation
750
221113
Nuclear Electric Power
Generation
750
221114
Solar Electric Power
Generation
250
221115
Wind Electric Power
Generation
250
221116
Geothermal Electric Power
Generation
250
221117
Biomass Electric Power
Generation
250
221118
Other Electric Power
Generation
250
221121
Electric Bulk Power
Transmission and
Control
500
221122
Electric Power Distribution
1000
221210
Natural Gas Distribution
1000
221310
Water Supply and Irrigation
Systems
$30
221320
Sewage Treatment Facilities
$22
221330
Steam and Air-Conditioning
Supply
$16
Note: Based on size standards effective at the time EPA conducted this analysis (SBA size standards, effective
August 19, 2019. Available at the following link: https://www.sba.gov/document/support--table-size-standards).
Source: SBA, 2019
EPA compared the relevant entity size criterion for each ultimate parent entity to the SBA
size standard noted in Table 6-1. We used the following data sources and methodology to
estimate the relevant size criterion values for each ultimate parent entity:
6-4
-------
1. Employment, Revenue, and Assets: EPA used the D&B Hoover's database as the
primary source for information on ultimate parent entity employee numbers, revenue, and
assets.8 In parallel, EPA also considered estimated revenues from affected EGUs based
on analysis of IPM parsed-file9 estimates for the baseline run for 2023 and 2026. EPA
assumed that the ultimate parent entity revenue was the larger of the two revenue
estimates. In limited instances, supplemental research was also conducted to estimate an
ultimate parent entity's number of employees, revenue, or assets.
2. Population: Municipal entities are defined as small if they serve populations of less than
50,000.10 EPA primarily relied on data from the Ventyx database and the U.S. Census
Bureau to inform this determination.
Ultimate parent entities for which the relevant measure is less than the SBA size standard were
identified as small entities and carried forward in this analysis.
In 2023 EPA identified 130 potentially affected EGUs, owned by 68 entities. Of these,
EPA identified 15 potentially affected EGUs owned by 9 small entities included in EPA's power
sector baseline. In 2026 total EPA identified 481 potentially affected EGUs, owned by 157
entities. Of these, EPA identified 56 potentially affected EGUs owned by 34 small entities
included in the power sector baseline.
In 2023, an entity can comply with the proposed Federal Implementation Plan
Addressing Regional Ozone Transport for the 2015 Ozone National Ambient Air Quality
Standards (FIP for the 2015 ozone NAAQS) through some combination of the following:
optimizing existing SCRs, optimizing existing SNCR controls, installing state-of-the-art
combustion controls, using allocated allowances, purchasing allowances, or reducing emissions
through a reduction in generation. Additionally, units with more allowances than needed can sell
8 Estimates of sales were used in lieu of revenue estimates when revenue data was unavailable.
9 IPM output files report aggregated results for "model" plants (i.e., aggregates of generating units with similar
operating characteristics). Parsed files approximate the IPM results at the generating unit level.
10 The Regulatory Flexibility Act defines a small government jurisdiction as the government of a city, county,
town, township, village, school district, or special district with a population of less than 50,000
(5 U.S.C. section 601(5)). For the purposes of the RFA, States and tribal governments are not
considered small governments. EPA's Final Guidance for EPA Rulewriters: Regulatory Flexibility Act is located
here: https://www.epa.gov/sites/default/files/2015-06/documents/guidance-regflexact.pdf.
6-5
-------
these allowances in the market. The chosen compliance strategy will be primarily a function of
the unit's marginal control costs and its position relative to the marginal control costs of other
units. In addition to the 2023 compliance options, in 2026 an entity can comply with the
proposed FIP for the 2015 ozone NAAQS by installing SCR or SNCR retrofits.
To attempt to account for each potential control strategy, EPA estimates compliance costs
as follows:
Ccompliance A Coperating+Retrofit A CFuel A CAllowances A (^Transaction "I~ A R
where C represents a component of cost as labeled11, and A R represents the change in revenues,
calculated as the difference in value of electricity generation between the baseline case and the
proposed rule in 2023 or in 2026.
Realistically, compliance choices and market conditions can combine such that an entity
may actually experience a reduction in any of the individual components of cost. Under the
proposed rule, some units will forgo some level of electricity generation (and thus revenues) to
comply and this impact will be lessened on these entities by the projected increase in electricity
prices under the proposed rule. On the other hand, those units increasing generation levels will
see an increase in electricity revenues and as a result, lower net compliance costs. If entities are
able to increase revenue more than an increase in fuel cost and other operating costs, ultimately,
they will have negative net compliance costs (or increased profit). Overall, small entities are not
projected to install relatively costly emissions control retrofits but may choose to do so in some
instances. Because this analysis evaluates the total costs along each of the compliance strategies
laid out above for each entity, it inevitably captures gains such as those described. As a result,
what we describe as cost is actually a measure of the net economic impact of the proposal on
small entities.
For this analysis, EPA used IPM-parsed output to estimate costs based on the parameters
above, at the unit level. These impacts were then summed for each small entity, adjusting for
ownership share. Net impact estimates were based on the following: operating and retrofit costs,
sale or purchase of allowances, and the change in fuel costs or electricity generation revenues
11 Retrofit costs include the costs of fully operating existing controls, as well as the installation of state-of-the-art
combustion controls, SCRs and SNCRs.
6-6
-------
under the proposed FIP for the 2015 ozone NAAQS relative to the base case. These individual
components of compliance costs were estimated as follows:
(1) Operating and retrofit costs (A CoPerating+Retrof,t): Using engineering analytics,
EPA identified which compliance option would be selected by each EGU in 2023
(i.e., SCR/SNCR optimization and/or installing state-of-the-art combustion
controls) and applied the appropriate cost to this choice (for details, please see
Chapter 4 of this RIA). For 2026, IPM projected retrofit costs were also included
in the calculation.
(2) Sale or purchase of allowances (A CAllowances) : To estimate the value of
allowance holdings, allocated allowances were subtracted from projected
emissions, and the difference was then multiplied by $1,800 (2016$) per ton for
2023 and $10,000 (2016$) per ton in 2026, which is the marginal cost of NOx
reductions used to set the modeled budgets in the proposed FIP for the 2015
ozone NAAQS. While this is a reasonable approximation, the analysis of the
proposal which is the source of other costs and revenues used in this calculation,
shows a lower projected allowance price. Units were assumed to purchase or sell
allowances to exactly cover their projected emissions under the proposed FIP for
the 2015 ozone NAAQS.
(3) Fuel costs (A CFuei): The change in fuel expenditures under the proposed FIP for
the 2015 ozone NAAQS was estimated by taking the difference in projected fuel
expenditures between the IPM estimates for the proposed FIP for the 2015 ozone
NAAQS and the baseline.
(4) Value of electricity generated: To estimate the value of electricity generated, the
projected level of electricity generation is multiplied by the regional-adjusted
retail electricity price ($/MWh) estimate, for all entities except those categorized
as private in Ventyx. See Chapter 4, Section 4.5.3 for a discussion of the Retail
Price Model, which was used to estimate the change in the retail price of
electricity. For private entities, EPA used the wholesale electricity price instead of
the retail electricity price because most of the private entities are independent
power producers (IPP). IPPs sell their electricity to wholesale purchasers and do
not own transmission facilities. Thus, their revenue was estimated with wholesale
electricity prices.
6-7
-------
(5) Administrative costs (A Crramactwn): Because most affected units are already
monitored as a result of other regulatory requirements, EPA considered the
primary administrative cost to be transaction costs related to purchasing or selling
allowances. EPA assumed that transaction costs were equal to 1.5 percent of the
total absolute value of the difference between a unit's allocation and projected
NOx emissions. This assumption is based on market research by ICF.
As indicated above, the use of a sales test for estimating small business impacts for a
rulemaking is consistent with guidance offered by the EPA on compliance with the RFA and is
consistent with guidance published by the SBA's Office of Advocacy that suggests that cost as a
percentage of total revenues is a metric for evaluating cost increases on small entities in relation
to increases on large entities. The potential impacts, including compliance costs, of the proposed
FIP for the 2015 ozone NAAQS on small entities are summarized in Table 6-2 and Table 6-3.
All costs are presented in 2016$. EPA estimated the annual net compliance cost to small entities
to be approximately $1.7 million in 2023 and $31 million in 2026.
Table 6-2. Projected Impact of the Proposed FIP for the 2015 Ozone NAAQS on Small
Entities in 2023
Total Net
EGU
Number of
Compliance
Number of Small Entities
Number of Small Entities
Ownership
Potentially
Cost
with Compliance Costs
with Compliance Costs
Type
Affected Entities
($2016
>1% of Generation
>3% of Generation
millions)
Revenues
Revenues
Municipal
2
1.1
0
0
IOU
7
0.6
0
0
Total
9
1.7
0
0
Source: IPM analysis
6-8
-------
Table 6-3. Projected Impact of the Proposed FIP for the 2015 Ozone NAAQS on Small
Entities in 2026
Total Net
EGU
Number
Compliance
Number of
Number of
Ownership
of
Cost
Small
Small
Type
Potentially
($2016
Entities
Entities
Affected
millions)
with
with
Entities
Compliance
Compliance
Costs >1%
Costs >3%
of
of
Generation
Generation
Revenues
Revenues
Municipal
13
3
3
3
IOU
7
36
3
1
Private
10
-9.3
0
0
Co-op
4
1.8
0
0
Total
34
31
6
4
Source: IPM analysis
EPA assessed the economic and financial impacts of the proposed rule using the ratio of
compliance costs to the value of revenues from electricity generation, focusing in particular on
entities for which this measure is greater than 1 percent. Although this metric is commonly used
in EPA impact analyses, it makes the most sense when as a general matter an analysis is looking
at small businesses that operate in competitive environments.12 However, small businesses in the
electric power industry often operate in a price-regulated environment where they are able to
recover expenses through rate increases. Of the 9 small entities considered in this analysis, none
are projected to experience compliance costs greater than 1 percent of generation revenues in
2023. Of the 34 entities considered in this analysis, 6 are projected to experience compliance
costs greater than 1% of generation revenues in 2026, and 4 are projected to experience
compliance costs greater than 3% of generation revenues in 2026.
6.1.2 Non-EGU Small Entity Impacts and Results
We identified 250 facilities, using the non-EGU screening assessment for 2026 discussed
in Chapter 4, owned by 85 parent companies, using information from D&B Hoover's13, that
12 U.S. EPA. EPA's Action Development Process. Final Guidance for EPA Rulewriters: Regulatory Flexibility Act
as Amended by the Small Business Regulatory Enforcement Fairness Act. September 2006. Available at
https://www.epa.gov/sites/production/files/2015-06/documents/guidance-regflexact.pdf.
13 D&B Hoovers is a subscription-based database that compiles publicly available information and can be found at
https://www.dnb.com/products/marketing-sales/dnb-hoovers.html.
6-9
-------
could be affected by the proposed rule. Of the parent companies, five companies, or two percent,
are small entities. We also used information from D&B Hoover's for the parent company
revenues. We identified the NAICS code for all parent companies and applied the SBA's table of
size standards to determine which of the companies were small entities. Table 6-4 below
includes the ranges NAICS codes and SBA entity size guidelines for small entity parent
companies.
Table 6-4. Non-EGU SBA Size Standards by NAICS Code
NAICS
Codes
NAICS U.S. Industry Title
Size
Standards
(million$)
Size
Standards
(number of employees)
327211
Flat Glass Manufacturing
1,000
327212
Other Pressed and Blown Glass and Glassware
Manufacturing
1,250
327213
Glass Container Manufacturing
1,250
327310
Cement Manufacturing
1,000
331110
Iron and Steel Mills and Ferroalloy Manufacturing
1,500
486210
Pipeline Transportation of Natural Gas
$30
322110
Pulp Mills
750
322121
Paper (except Newsprint) Mills
1,250
322130
Paperboard Mills
1,250
324110
Petroleum Refineries
1,500
324199
All Other Petroleum and Coal Products
Manufacturing
500
325110
Petrochemical Manufacturing
1,000
325180
Other Basic Inorganic Chemical Manufacturing
1,000
325199
All Other Basic Organic Chemical Manufacturing
1,250
Also, we calculated the cost-to-sales ratios for all of the affected entities to determine (i)
the magnitude of the costs of the proposal, and (ii) whether there would be a significant impact
on small entities compared to large entities. Non-EGUs do not operate in a price-regulated
environment, like EGUs, where they are able to recover expenses through rate increases. As
presented in Table 6-5 for all firms the average cost-to-sales ratio is approximately 0.1 percent;
the median cost-to-sales ratio is less than 0.01 percent; and the maximum cost-to-sales ratio is
approximately 1.4 percent. For large firms, the average cost-to-sales ratio is approximately 0.1
percent; the median cost-to-sales ratio is less than 0.1 percent; and the maximum cost-to-sales
ratio is approximately 1.4 percent. For small firms, the average cost-to-sales ratio is
6-10
-------
approximately 0.7 percent, the median cost-to-sales ratio is 0.5 percent, and the maximum cost-
to-sales ratio is 1.3 percent.
Table 6-5. Summary of Sales Test Ratios for 2026 for Firms Affected by Proposed Rule
Firm Size
No. of Known
Affected Firms
% of Total
Known
Affected Firms
Mean Cost-
to-Sales
Ratio
Median Cost-
to-Sales Ratio
Min. Cost-to-
Sales Ratio
Max. Cost-
to-Sales
Ratio
Small
5
2.0%
0.7%
0.5%
0.3%
1.3%
Large
245
98.0%
0.1%
<0.0%
<0.0%
1.4%
All
250
100.0%
0.1%
<0.0%
<0.0%
1.4%
As mentioned above, we compare annual compliance costs to annual revenues at the
ultimate parent company level. For the small entities, the small parent companies are the small
facilities; in other words, the small parent companies each own one small facility. Table 6-6
below includes the small parent companies and their projected cost-to-sales ratio, NAICS code,
and small business size standards. The facility-specific costs for the small parent companies
ranged from $227 thousand to $1.8 million annually (2016$).
Table 6-6. Summary of Small Parent Company Small Business Size Standards
Cost to
Number of
SBA Size Standard:
Sales Ratio
Employees
Number of
Small Parent
NAICS
Employees
Company
Cstn Holdings, Inc.
325199
1.3%
600
1,250
Angus Chemical
325199
0.7%
500
1,250
Company
Futurefuel Corp.
325199
0.5%
548
1,250
Capital Aggregates
327310
0.4%
525
1,000
Glass Energy
327213
0.3%
353
1,250
Company, Inc.
6.1.3 Conclusion
Making a no SISNOSE (significant economic impacts on a substantial number of small
entities) determination reflects an assessment of whether an estimated economic impact is
significant and whether that impact affects a substantial number of small entities. We prepared
an analysis of small entity impacts for EGUs in 2023 and in 2026 and for non-EGUs in 2026
separately and combined the 2026 results for a SISNOSE determination for the proposed rule.
For EGUs, estimates indicate that there are nine small entities that see a +/- 1 percent
change in either summer NOx emissions, summer generation, or summer fuel use in 2023, and
6-11
-------
none are projected to have a cost-to-sales impact greater than 1 percent of their revenues in 2023.
In 2026, the analysis indicates that 34 small entities see a +/- 1 percent change in either summer
NOx emissions, summer generation or summer fuel use, and 6 of these are projected to have a
cost impact of greater than 1 percent of their revenues in 2026.
In 2026, EPA identified 157 possibly affected EGU entities. Of these, EPA identified 34
small entities affected by the proposal, and of these 6 small entities may experience costs of
greater than 1 percent of revenues. Of the 6 small entities projected to have costs greater than 1
percent of revenues, two operate in cost-of-service regions and would generally be able to pass
any increased costs along to ratepayers. In EPA's modeling, most of the cost impacts for these
small entities and their associated units are driven by lower electricity generation relative to the
baseline. Specifically, four units reduce their generation by significant amounts, driving the bulk
of the costs for all small entities. Finally, EPA's decision to exclude units smaller than 25 MW
capacity from the proposed FIP, and exclusion of uncontrolled units smaller than 100 MW from
backstop emission rate limits has already significantly reduced the burden on small entities by
reducing the number of affected small entity-owned units. Further, in 2026 for non-EGUs, there
are five small entities, and one small entity is estimated to have a cost-to-sales impact of 1.3
percent of their revenues.
Based on this analysis, for this proposal overall we conclude that the estimated costs for
the proposed rule will not have a significant economic impact on a substantial number of small
entities (SISNOSE).
6.2 Labor Impacts
This section discusses potential employment impacts of this proposed regulation. As
economic activity shifts in response to a regulation, typically there will be a mix of declines and
gains in employment in different parts of the economy over time and across regions. To present a
complete picture, an employment impact analysis will describe the potential positive and
negative changes in employment levels. There are significant challenges when trying to evaluate
the employment effects due to an environmental regulation from employment effects due to a
wide variety of other economic changes, including the impact of the coronavirus pandemic on
labor markets and the state of the macroeconomy generally. Considering these challenges, we
6-12
-------
look to the economics literature to provide a constructive framework and empirical evidence. To
simplify, we focus on impacts on labor demand related to compliance behavior. Environmental
regulation may also affect labor supply through changes in worker health and productivity
(Graff, Zivin and Neidell, 2018).
Economic theory of labor demand indicates that employers affected by environmental
regulation may increase their demand for some types of labor, decrease demand for other types,
or for still other types, not change it at all (Morgenstern et al. 2002, Deschenes 2018, Berman
and Bui 2001). To study labor demand impacts empirically, a growing literature has compared
employment levels at facilities subject to an environmental regulation to employment levels at
similar facilities not subject to that environmental regulation; some studies find no employment
effects, and others find significant differences. For example, see Berman and Bui (2001),
Greenstone (2002), Ferris, Shadbegian and Wolverton (2014), and Curtis (2018, 2020).
A variety of conditions can affect employment impacts of environmental regulation,
including baseline labor market conditions and employer and worker characteristics such as
occupation and industry. We focus our labor impacts analysis primarily on the directly regulated
facilities and other EGUs and related fuel markets and in the different non-EGU industry sectors.
6.2.1 EGULabor Impacts
This section discusses and projects potential employment impacts for the utility power,
coal and natural gas production sectors that may result from the proposed rule. EPA has a long
history of analyzing the potential impacts of air pollution regulations on changes in the amount
of labor needed in the power generation sector and directly related sectors. The analysis
conducted for this RIA builds upon the approaches used in the past and takes advantage of newly
available data to improve the assumptions and methodology.14
The results presented in this section are based on a methodology that estimates the impact
on employment based on the differences in projections between two modeling scenarios: the
baseline scenario, and a scenario that represents the implementation of the proposed rule. The
estimated employment difference between these scenarios can be interpreted as the incremental
effect of the proposed rule on employment in this sector. As discussed in Chapter 4, there is
14 For a detailed overview of this methodology, including all underlying assumptions, see the U.S. EPA
Methodology for Power Sector-Specific Employment Analysis, available in the docket.
6-13
-------
uncertainty related to the future baseline projections. Because the incremental employment
estimates presented in this section are based on projections discussed in Chapter 4, it is important
to highlight the relevance of the Chapter 4 uncertainty discussion to the analysis presented in this
section.
Like previous analyses, this analysis represents an evaluation of "first-order employment
impacts" using a partial equilibrium modeling approach. It includes some of the potential ripple
effects of these impacts on the broader economy. These ripple effects include the secondary job
impacts in both upstream and downstream sectors. The analysis includes impacts on upstream
sectors including coal, natural gas, and uranium. However, the approach does not analyze
impacts on other fuel sectors, nor does it analyze potential impacts related to transmission,
distribution, or storage. This approach also excludes the economy-wide effects of changes to
energy markets (such as higher or lower forecasted electricity prices). At the same time, this
approach excludes labor impacts that are usually included in a benefits analysis for an
environmental policy, such as increased productivity from a healthier workforce and reduced
absenteeism due to fewer sick days of employees and dependent family members (e.g., children).
6.2.2 Overview of Methodology
The methodology includes the following two general approaches, based on the available
data. The first approach utilizes the rich employment data that is available for several types of
generation technologies in the 2020 U.S. Energy and Employment Report.15 For employment
related to other electric power sector generating and pollution control technologies, the second
approach utilizes information available in the U.S. Economic Census.
Detailed employment inventory data is available regarding recent employment related to
coal, hydro, natural gas, geothermal, wind, and solar generation technologies. The data enables
the creation of technology-specific factors that can be applied to model projections of capacity
(reported in megawatts, or MW) and generation (reported in megawatt-hours, or MWh) in order
to estimate impacts on employment. Since employment data is only available in aggregate by
fuel type, it is necessary to disaggregate by labor type in order to differentiate between types of
jobs or tasks for categories of workers. For example, some types of employment remain constant
15 https://www.usenergyjobs.org/
6-14
-------
throughout the year and are largely a function of the size of a generator, e.g., fixed operation and
maintenance activities, while others are variable and are related to the amount of electricity
produced by the generator, e.g., variable operation and maintenance activities.
The approach can be summarized in three basic steps:
• Quantify the total number of employees by fuel type in a given year;
• Estimate total fixed operating & maintenance (FOM), variable operating &
maintenance (VOM), and capital expenditures by fuel type in that year; and
• Disaggregate total employees into three expenditure-based groups and develop factors
for each group (FTE/MWh, FTE/MW-year, FTE/MW new capacity).
Where detailed employment data is unavailable, it is possible to estimate labor impacts
using labor intensity ratios. These factors provide a relationship between employment and
economic output and are used to estimate employment impacts related to construction and
operation of pollution control retrofits, as well as some types of electric generation technologies.
For a detailed overview of this methodology, including all underlying assumptions and the
types of employment represented by this analysis, see the U.S. EPA Methodology for Power
Sector-Specific Employment Analysis, available in the docket.
6.2.3 Overview of Power Sector Employment
In this section we focus on employment related to electric power generation, as well as
coal and natural gas extraction because these are the segments of the power sector that are most
relevant to the projected impacts of the proposed rule. Other segments not discussed here include
other fuels, energy efficiency, and transmission, distribution, and storage. The statistics presented
here are based on the 2020 USEER, which reports data from 2019.16
In 2019, the electric power generation sector employed nearly 900,000 people. Relative
to 2018, this sector grew by over 2 percent, despite job losses related to nuclear and coal
generation. These losses were offset by increases in employment related to other generating
technologies, including natural gas, solar, and wind. The largest component of total 2019
16 While 2020 data is available in the 2021 version of this report, this section of the RIA utilizes 2019 data because
this year does not reflect any short-term trends related to the COVID-19 pandemic. The annual report is available
at: https://www.usenergyjobs.org/.
6-15
-------
employment in this sector is construction (33%). Other components of the electric power
generation workforce include: utility workers (20%), professional and business service
employees (20%), manufacturing (13%), wholesale trade (8%), and other (5%). In 2019, jobs
related to solar and wind generation represent 31% and 14% of total jobs, respectively, and jobs
related to coal generation represent 10% of total employment.
In additional to generation-related employment we also look at employment related to
coal and natural gas use in the electric power sector. In 2019, the coal industry employed about
75,000 workers. Mining and extraction jobs represent the vast majority of total coal-related
employment in 2019 (74%). The natural gas fuel sector employed about 276,000 employees in
2019. About 60% of those jobs were related to mining and extraction.
6.2.4 Projected Sectoral Employment Changes due to the Proposed Rule
Affected EGUs may respond to the proposed rule through a number of means including
optimizing existing controls, upgrading to state-of-the-art combustion controls, shifting
generation from higher emitting to lower emitting sources, and installing new SCRs and SNCRs.
Under the modeling of the proposed rule, 32 GW of SCR installations are projected by the 2025
run year, and an incremental 18 GW of coal and 4 GW of oil/gas retirements are projected by
2030. Additionally, an incremental 14 GW of non-hydro renewable additions are also projected
under the proposed rule by the 2025 run year. These are primarily comprised of solar builds that
occur earlier in the forecast period relative to the baseline projections as a result of the increased
fossil thermal retirements.
Based on these power sector modeling projections, we estimate a sizable increase in
construction-related jobs related to the installation of new pollution controls under the proposed
option, as well as the construction of new generating capacity (largely solar PV). In 2025, we
estimate an increase of over 150,000 construction-related job-years. Some of this capacity is
projected to be built earlier under the policy case than the baseline (which explains the estimated
decrease in 2030). Construction-related job-year changes are one-time impacts, occurring during
each year of the multi-year periods during which construction of new capacity is completed.
Construction-related figures in Table 6-7 represent a point estimate of incremental changes in
construction jobs for each year (for a three-year construction projection, this table presents one-
third of the total jobs for that project). Negative construction job-year estimates occur when
6-16
-------
additional generating capacity is projected to be built in the baseline, but not projected to be built
under the proposed rule.
Table 6-7. Changes in Labor Utilization: Construction-Related (Number of Job-Years of
Employment in a Single Year)
2023
2025
2028
2030
New Pollution Controls
600
11,400
<100
<100
New Capacity
<100
139,600
9,700
-43,900
vfote: "<100" denotes an increase or decrease of less than 100 job-years
We also estimate changes in the number of job-years related to recurring non-
construction employment. Recurring employment changes are job-years associated with annual
recurring jobs including operating and maintenance activities and fuel extraction jobs. Newly
built generating capacity creates a recurring stream of positive job-years, while retiring
generating capacity, as well as avoided new built capacity, create a stream of negative job-years.
The proposed rule is projected to result, generally, in a replacement of relatively labor-intensive
coal capacity with less labor-intensive capacity (primarily solar), which results in an overall
decrease of non-construction jobs. The proposed rule is also projected to result in a small
reduction in recurring employment related to fuel extraction. The total net estimated decrease in
recurring employment is less than 7,500 job years in 2025, which is a small percentage of total
2019 power sector employment reported in the 2020 USEER (approximately 900,000
generation-related jobs, 75,000 coal-related jobs, and 276,000 natural gas-related jobs). Note
that the projected decreases related to operation of existing pollution controls is consistent with
the projected retirements of existing capacity. Table 6-8 provide detailed estimates of recurring
non-construction employment changes.
Table 6-8. Changes in Labor Utilization: Recurring Non-Construction (Number of Job-
Years of Employment in a Single Year)
2023
2025
2028
2030
Pollution Controls
<100
<100
-100
-100
Existing Capacity
<100
-11,200
-11,800
-9,700
New Capacity
<100
4,300
5,200
3,600
Fuels (Coal, Natural
Gas, Uranium)
<100
-600
-900
-500
Coal
<100
-600
-1,000
-1,000
6-17
-------
Natural Gas <100 <100 <100 200
Uranium <100 <100 100 200
Note: "<100" denotes an increase or decrease of less than 100 job-years; Numbers may not sum due to
rounding
6.2.5 Non-EGULabor Impacts
This section begins with a description of baseline conditions in non-EGU industries
affected by the proposed rule, focusing on the directly regulated industries and groups of affected
workers. The directly regulated firms in non-EGU industries fall into two tiers of industries17
(Table 6-9):
• Tier 1 industries that have a maximum contribution to any one receptor of >0.10
ppb and (2) contribute >= 0.01 ppb to at least 10 receptors, and
• Tier 2 industries that either have (1) a maximum contribution to any one receptor
>=0.10 ppb but contribute >=0.01 ppb to fewer than 10 receptors, or (2) a
maximum contribution <0.10 ppb but contribute >=0.01 ppb to at least 10
receptors.
The proposed rule only covers specific boilers in the Tier 2 industries and not every emissions
unit in those industries. Table 6-9 shows the industry definitions and the NAICS codes used to
categorizes the data for the relevant industries.
The cement and concrete product manufacturing industry (NAICS 3273) by far is the
largest of the Tier 1 industries in terms of the number of people employed. BLS Current
Employment Statistics show that the industry employs 186,000 people nationally. The iron and
steel mills and ferroalloy manufacturing industry (NAICS 3311) and glass and glass product
manufacturing industry (NAICS 3772) are similarly sized with 81,400 and 79,900 people
employed, respectively. Each of the non-EGU industries has seen different trends in employment
over the past decade. Both the pipeline transportation of natural gas (NAICS 4862) and cement
and concrete product manufacturing industries saw sizable increases in employment over the past
decade but contracted in 2020 from the COVID-19 pandemic. The iron and steel mills and
17 See Chapter 4, Section 4.4 for further discussion of the industry tiers.
6-18
-------
ferroalloy manufacturing industry has seen steady decline in total employment, while the glass
and glass product manufacturing industry has remained relatively constant over the last decade.18
Table 6-9. Relevant Industry Employment (2020)
N. T„„ Employment Percent Change
(Thousands) 2011 - 2020
Tier 1 Industries
Pipeline Transportation of Natural . 0
^ 4oo L
Gas
Cement and Concrete Product
Manufacturing
Iron and Steel Mills and Ferroalloy
Manufacturing
Glass and Glass Product
Manufacturing
Tier 2 Industries
Basic Chemical Manufacturing 3251 150.1 5%
49.1 19%
3273 186.4 17%
3311 81.4 -10%
3772 79.9 -1%
3241 106.5 -5%
Petroleum and Coal Products
Manufacturing
Pulp, Paper, and Paperboard Mills 3221 92.6 -15%
Source: BLS
These industries are capital intensive. We rely on three public sources to get a range of
estimates of employment per output by sector: the Economic Census (EC), and the Annual
Survey of Manufacturers (ASM), both provided by the U.S. Census Bureau, and employment
and output by industry provided by the U.S. Bureau of Labor Statistics (BLS). The EC is
conducted every 5 years, most recently in 2017. The ASM is an annual subset of the EC and is
based on a sample of establishments. The latest set of data from the ASM is from 2019. Both sets
of U.S. Census Bureau data provide detailed industry data, providing estimates at the 4-digit
NAICS level. They provide separate estimates of the number of employees and the value of
shipments at the 4-digit NAICS, which we convert to a ratio in this employment analysis.
Table 6-10 provides estimates of employment per $1 million of products sold by the
industry for each data source in 2017$. While the ratios are not the same, they are similar across
time for both surveys. Glass and glass product manufacturing seems to be the most labor-
intensive industry followed by cement and concrete product manufacturing.
18 Bureau of Labor Statistics. BLS Employment, Hours, and Earnings from the Current Employment Statistics
survey (National), All-employees, May 2021
6-19
-------
Table 6-10. Employment per $1 million Output in the Tier 1 Industries
Economic
Sector
Census
ASM 2019
Pipeline Transportation of Natural Gas
1.21
N/A
Cement and Concrete Product Manufacturing
2.80
3.05
Iron and Steel Mills and Ferroalloy Manufacturing
0.97
0.91
Glass and Glass Product Manufacturing
3.34
3.35
6.2.6 Conclusions
Generally, there are significant challenges when trying to evaluate the employment
effects due to an environmental regulation from employment effects due to a wide variety of
other economic changes, including the impact of the coronavirus pandemic on labor markets and
the state of the macroeconomy generally. For EGUs, the proposed FIP for the 2015 ozone
NAAQS may result in a sizable increase in construction-related jobs related to the installation of
new pollution controls, as well as the construction of new generating capacity (largely solar PV).
The proposed rule is also projected to result, generally, in a replacement of relatively labor-
intensive coal capacity with less labor-intensive capacity (primarily solar), which results in an
overall decrease of non-construction jobs. For the non-EGU industries, the employment trends
over the last decade vary by industry. Without more detailed information on the labor required
for installing pollution controls in these specific industries and other potential compliance
approaches, we are not able to determine the potential effect of employment changes in the non-
EGU industries.
6-20
-------
6.3 References
Berman, E. and L. T. M. Bui. 2001. "Environmental Regulation and Labor Demand: Evidence
from the South Coast Air Basin." Journal of Public Economics. 79(2): 265-295;
Curtis, E. M. 2018. "Who loses under cap-and-trade programs? The labor market effects of the
NOx budget trading program," Review of Economics and Statistics 100 (1): 151-66;
Curtis, E.M. 2020. "Reevaluating the ozone nonattainment standards: Evidence from the 2004
expansion," Journal of Environmental Economics and Management, 99: 102261;
Deschenes, O. 2018. "Environmental regulations and labor markets," IZA World of Labor: 22:
1-10;
Ferris, A. E., R. Shadbegian, A. Wolverton. 2014 "The Effect of Environmental Regulation on
Power Sector Employment: Phase I of the Title IV S02 Trading Program," Journal of the
Association of Environmental and Resource Economics 1(4): 521-553
Graff Zivin, J. and M. Neidell. 2018. "Air pollution's hidden impacts". Science. 359(6371). 39-
40.
Greenstone, Michael, "The Impacts of Environmental Regulations on Industrial Activity:
Evidence from the 1970 and 1977 Clean Air Act Amendments and the Census of
Manufactures," Journal of Political Economy 110, no. 6 (2002): 1175-1219
Morgenstern, R.D., W.A. Pizer, and J. Shih. 2002. Jobs Versus the Environment: An Industry-
Level Perspective. Journal of Environmental Economics and Management 43: 412-436.
6-21
-------
CHAPTER 7: ENVIRONMENTAL JUSTICE IMPACTS
7.1 Introduction
Executive Order 12898 directs EPA to identify the populations of concern who are most
likely to experience unequal burdens from environmental harms; specifically, minority
populations, low-income populations, and indigenous peoples (59 FR 7629, February 16, 1994).
Additionally, Executive Order 13985 was signed to advance racial equity and support
underserved communities through Federal government actions (86 FR 7009, January 20, 2021).
EPA defines environmental justice (EJ) as the fair treatment and meaningful involvement of all
people regardless of race, color, national origin, or income with respect to the development,
implementation, and enforcement of environmental laws, regulations, and policies. EPA further
defines the term fair treatment to mean that "no group of people should bear a disproportionate
burden of environmental harms and risks, including those resulting from the negative
environmental consequences of industrial, governmental, and commercial operations or
programs and policies".1 Meaningful involvement means that: (1) potentially affected
populations have an appropriate opportunity to participate in decisions about a proposed activity
that will affect their environment and/or health; (2) the public's contribution can influence the
regulatory Agency's decision; (3) the concerns of all participants involved will be considered in
the decision-making process; and (4) the rule-writers and decision-makers seek out and facilitate
the involvement of those potentially affected.
The term "disproportionate impacts" refers to differences in impacts or risks that are
extensive enough that they may merit Agency action.2 In general, the determination of whether a
disproportionate impact exists is ultimately a policy judgment which, while informed by
analysis, is the responsibility of the decision-maker. The terms "difference" or "differential"
indicate an analytically discernible distinction in impacts or risks across population groups. It is
the role of the analyst to assess and present differences in anticipated impacts across population
groups of concern for both the baseline and proposed regulatory options, using the best available
information (both quantitative and qualitative) to inform the decision-maker and the public.
1 See, e.g., "Environmental Justice." Epa.gov, U.S. Environmental Protection Agency, 4 Mar. 2021,
https://www.epa.gov/environmentaljustice.
2 See https://www.epa.gov/environmentaljustice/technical-guidance-assessing-environmental-justice-regulatory-
analysis.
7-1
-------
A regulatory action may involve potential environmental justice concerns if it could: (1)
create new disproportionate impacts on minority populations, low-income populations, and/or
Indigenous peoples; (2) exacerbate existing disproportionate impacts on minority populations,
low-income populations, and/or Indigenous peoples; or (3) present opportunities to address
existing disproportionate impacts on minority populations, low-income populations, and/or
Indigenous peoples through the action under development.
The Presidential Memorandum on Modernizing Regulatory Review (86 FR 7223;
January 20, 2021) calls for procedures to "take into account the distributional consequences of
regulations, including as part of a quantitative or qualitative analysis of the costs and benefits of
regulations, to ensure that regulatory initiatives appropriately benefit, and do not inappropriately
burden disadvantaged, vulnerable, or marginalized communities." Under Executive Order 13563,
federal agencies may consider equity, human dignity, fairness, and distributional considerations,
where appropriate and permitted by law. For purposes of analyzing regulatory impacts, EPA
relies upon its June 2016 "Technical Guidance for Assessing Environmental Justice in
Regulatory Analysis,"3 which provided recommendations that encourage analysts to conduct the
highest quality analysis feasible, recognizing that data limitations, time and resource constraints,
and analytical challenges will vary by media and circumstance.
A reasonable starting point for assessing the need for a more detailed environmental
justice analysis is to review the available evidence from the published literature and from
community input on what factors may make population groups of concern more vulnerable to
adverse effects (e.g., underlying risk factors that may contribute to higher exposures and/or
impacts). It is also important to evaluate the data and methods available for conducting an
environmental justice analysis. EJ analyses can be grouped into two types, both of which are
informative, but not always feasible for a given rulemaking:
1. Baseline: Describes the current (pre-control) distribution of exposures and risk,
identifying potential disparities.
2. Policy: Describes the distribution of exposures and risk after the control strategy has been
applied (post-control), identifying how potential disparities change in response to the
rulemaking.
3 See https://www.epa.gov/environmentaljustice/technical-guidance-assessing-environmental-justice-regulatory-
analysis.
7-2
-------
EPA's 2016 Technical Guidance does not prescribe or recommend a specific approach or
methodology for conducting an environmental justice analysis, though a key consideration is
consistency with the assumptions underlying other parts of the regulatory analysis when
evaluating the baseline and regulatory options.
7.2 Analyzing EJ Impacts in This Proposal
In addition to the benefits assessment (Chapter 5), EPA considers potential environmental
justice (EJ) concerns of this proposed rulemaking.4 Although EJ concerns for each rulemaking
are unique and should be considered on a case-by-case basis, EPA's EJ Technical Guidance5
states that "[t]he analysis of potential EJ concerns for regulatory actions should address three
questions:
1. Are there potential EJ concerns associated with environmental stressors affected
by the regulatory action for population groups of concern in the baseline?
2. Are there potential EJ concerns associated with environmental stressors affected
by the regulatory action for population groups of concern for the regulatory
option(s) under consideration?
3. For the regulatory option(s) under consideration, are potential EJ concerns created
or mitigated compared to the baseline?"
To address these questions, EPA developed an analytical approach that considers the
purpose and specifics of the proposed rulemaking, as well as the nature of known and potential
exposures and impacts. For example, while we recognize that the proposal is focused on
reducing NOx emissions to ensure states meet their obligations under the "Good Neighbor"
provision of the Clean Air Act to eliminate significant contributions to, or interference with
maintenance of, the 2015 ozone National Ambient Air Quality Standards (NAAQS), this
proposed rulemaking may also reduce other pollutant emissions such as NO2. Like other oxides
4 A potential EJ concern is defined as "the actual or potential lack of fair treatment or meaningful involvement of
minority populations, low-income populations, tribes, and indigenous peoples in the development, implementation
and enforcement of environmental laws, regulations and policies" (U.S. EPA, 2015a). For analytic purposes, this
concept refers more specifically to "disproportionate impacts on minority populations, low-income populations,
and/or indigenous peoples that may exist prior to or that may be created by the proposed regulatory action" (U.S.
EPA, 2015a).
5 U.S. Environmental Protection Agency (EPA), 2015. Guidance on Considering Environmental Justice During the
Development of Regulatory Actions.
7-3
-------
of nitrogen, NO2 can contribute to the formation of ozone downwind of sources; however, direct
emissions of NO2 can also lead to localized exposures that may be associated with respiratory
effects in nearby populations at sufficiently high concentrations. While NO2 exposures and
concentrations were not estimated as part of this proposal, the proximity analysis allows for the
possibility that such exposures may be relevant to the baseline and/or change due to this
proposed action. Due to the potential for reductions in NO2 concentration nearby affected
sources, EPA conducts two proximity analyses to evaluate the potential EJ implications of
changes in pollutants (Section 3): a demographic proximity analyses of populations residing near
affected facilities (Section 7.3.1), and tribal proximity analyses of affected facilities (Section
7.3.2). EPA also conducts an analysis of reductions in ozone concentrations nationwide resulting
from the NOx emission reductions projected to occur under the proposed rule, characterizing
distributional exposures both prior to and following implementation of the regulatory alternatives
in 2023 and 2026 (Section 7.4). Each analysis involves unique limitations and uncertainties,
which are described prior to presentation of analytic results in the subsections below.
7.3 Demographic Proximity Analyses
Demographic proximity analyses allow one to assess the proximity of vulnerable
populations to environmental hazards as a proxy for exposure and the potential for adverse health
impacts that may occur at a local scale due to economic activity at a given location such as noise,
odors, and traffic. We include the following proximity screening analyses to characterize the
potential for communities with EJ concerns to be impacted by emissions sources covered under
this EPA action.
Although baseline proximity analyses are presented here, several important caveats
should be noted. In most areas, emissions are not expected to increase from the proposed
rulemaking, so most communities nearby affected facilities should not experience increases in
exposure from directly emitted pollutants. However, facilities may vary widely in terms of the
risk they already pose to nearby populations and proximity to affected facilities does not capture
variation in baseline exposure across communities. Nor does it indicate that any exposures or
impacts will occur and should not be interpreted as a direct measure of exposure or impact.
These points limit the usefulness of proximity analyses when attempting to answer question 1 or
2 above from EPA's EJ Technical Guidance.
7-4
-------
• Electricity Generating Unit (EGU): Comparison of the percentage of various populations
(race/ethnicity, age, education, poverty status, income, and linguistic isolation) living
nearby covered EGU sources to average national levels.
• Non-EGU (non-electric seneratins units, or other stationary emissions sources):
Comparison of the percentage of various populations (race/ethnicity, age, education,
poverty status, income, and linguistic isolation) living nearby covered non-EGU sources
to average national levels.
• Tribal. Analysis of tribes and unique tribal lands within 50 miles of covered facilities.
7.3.1 EGU and Non-EGU Proximity Assessments
The current analysis identified all census blocks within a 5 km, 10 km and 50 km radius
of the latitude/longitude location of each facility, and then linked each block with census-based
demographic data.6 The total population within a specific radius around each facility is the sum
of the population for every census block within that specified radius, based on each block's
population provided by the decennial Census.7 Statistics on race, ethnicity, age, education level,
poverty status and linguistic isolation were obtained from the Census' American Community
Survey (ACS) 5-year averages for 2015-2019. These data are provided at the block group level.
For the purposes of this analysis, the demographic characteristics of a given block group - that
is, the percentage of people in different races/ethnicities, the percentage in different age groups
(<18, 18-64, and >64), the percentage without a high school diploma, the percentage that are
below the poverty level, and the percentage that are linguistically isolated - are presumed to also
describe each census block located within that block group.
In addition to facility-specific demographics, the demographic composition of the total
population within the specified radius (e.g., 50 km) for all facilities as a whole was also
computed (e.g., all EGUs or all non-EGU facilities). In calculating the total populations, to avoid
6 Five km and 50 km radii are the default distances currently used for proximity analyses. The 5 km distance is the
shortest distance that should be chosen to avoid excessive demographic uncertainty and provides information on
near-field populations. The 50 km distance offers a sub-regional perspective and represents the maximum 50 km
modeling domain for exposure modeling. The 10 km distance was added to this analysis as few to no people were
within 5 km of some affected facilities.
7 The location of the Census block centroid is used to determine if the entire population of the Census block is
assumed to be within the specified radius. It is unknown how sensitive these results may be to different methods of
population estimation, such as aerial apportionment.
7-5
-------
double-counting, each census block population was only counted once. That is, if a census block
was located within the selected radius (i.e., 50 km) for multiple facilities, the population of that
census block was only counted once in the total population. Finally, this analysis compares the
demographics at each specified radius (i.e., 5 km, 10 km, and 50 km) to the demographic
composition of the nationwide population.
For this action, a demographic analysis was conducted for 851 EGU facilities at the 5 km,
10 km, and 50 km radius distances (Table 7-1). Approximately 196 million people live within 50
km of the EGU facilities, representing roughly 60% of the 328 million total population of the
U.S. The percent demographic make-up of the population within 50 km of the EGU facilities is
very similar to the national average for each demographic investigated. Approximately 23.9
million and 65.5 million people live within 5 km and 10 km of the EGU facilities, respectively.
The demographic make-up of the population within 5 km and 10 km of EGU facilities are very
similar. Minorities constitute about 55% of the population within 5 km and 10 km of EGU
facilities, which is about 15% greater than the national average of 40% minorities. The higher
minority population is driven largely by a higher Hispanic/Latino population (about 10% above
national average) and a higher African American population (about 3% above national average).
The age distribution for the population within 5 km and 10 km of EGU facilities is similar to the
national average. The percent of people living below the poverty level is about 2-3% higher
within 5 km and 10 km of the EGU facilities than the national average. About 8% to 9% of the
population within 5 km and 10 km of the EGU facilities is living in linguistic isolation, this is
more than 1.5 times higher than the national average (about 5%).
7-6
-------
Table 7-1. Population Demographics for EGU Facilities
Demographic Group
Percent of Populati
Compared to t
on Within Each Distance
le National Average1
5km
10km
50km
National
Average
Race/
Ethnicity
White
44.1%
44.9%
58.2%
60.1%
Minority2
55.9%
55.1%
41.8%
39.9%
African American
14.7%
15.4%
12.9%
12.2%
Native American
0.3%
0.3%
0.4%
0.7%
Other and Multiracial
11.1%
11.2%
9.0%
8.2%
Hispanic or Latino3
29.7%
28.1%
19.4%
18.8%
Age
0-17 Years Old
22.4%
22.7%
22.5%
22.6%
18-64 Years Old
63.7%
63.1%
62.2%
61.7%
>=65 Years Old
13.9%
14.2%
15.3%
15.7%
Income
People Living Below
the Poverty Level
16.3%
15.5%
13.1%
13.4%
Education
>= 25 Years Old
Without a High
School Diploma
16.7%
15.8%
12.7%
12.1%
Language
People Living in
Linguistic Isolation
9.0%
8.4%
5.4%
5.4%
Total Population
23,863,069
65,522,012
196,411,623
328,016,242
1 Demographic percentage is based on the Census' 2015-2019 American Community Survey 5-year averages, at the block
group level, and include the 50 states, District of Columbia, and Puerto Rico. Total population is based on block level data from
the 2010 Decennial Census.
2 Minority population is the total population minus the white population.
3 To avoid double counting, the "Hispanic or Latino" category is treated as a distinct demographic category for these analyses.
A person who identifies as Hispanic or Latino is counted as Hispanic/Latino for this analysis, regardless of what race this
person may have also identified as in the Census.
For this action, a demographic analysis was also conducted for 251 non-EGU facilities at
the 5 km, 10 km, and 50 km radius distances (Table 7-2). Approximately 92 million people live
within 50 km of the non-EGU facilities, representing roughly 37% of the 328 million total
population of the U.S. The percent demographic make-up of the population within 50 km of the
non-EGU facilities is very similar to the national average for each demographic investigated.
Approximately 2.8 million and 9.4 million people live within 5 km and 10 km of the non-EGU
facilities, respectively. The demographic make-up of the population within 5 km and 10 km of
non-EGU facilities are very similar. Minorities constitute about 39% of the population within 5
km and 40% of the population within 10 km of non-EGU facilities, which is the same or slightly
less than the national average of 40% minorities. The age distribution for the population within 5
7-7
-------
km and 10 km of non-EGU facilities is similar to the national average. The percent of people
living below the poverty level is about 3-4% higher within 5 km and 10 km of the non-EGU
facilities than the national average. The percent of the population within 5 km and 10 km of the
non-EGU facilities living in linguistic isolation is lower (about 4%) than the national average
(about 5%).
Table 7-2. Population Demographics for Non-EGU Facilities
Demographic Group
Percent of Population Within Each Distance
Compared to the National Average1
5km
10km
50km
National
Average
Race/
Ethnicity
White
61.3%
60.1%
61.2%
60.1%
Minority2
38.7%
39.9%
38.8%
39.9%
African American
13.0%
15.4%
13.3%
12.2%
Native American
0.6%
0.5%
0.4%
0.7%
Other and Multiracial
7.7%
6.9%
7.4%
8.2%
Hispanic or Latino3
17.4%
17.0%
17.7%
18.8%
Age
0-17 Years Old
23.7%
23.1%
22.3%
22.6%
18-64 Years Old
60.9%
61.3%
62.1%
61.7%
>=65 Years Old
15.4%
15.6%
15.5%
15.7%
Income
People Living Below the
Poverty Level
17.6%
16.0%
13.8%
13.4%
Education
>= 25 Years Old Without
a High School Diploma
14.2%
13.3%
12.9%
12.1%
Language
People Living in
Linguistic Isolation
4.3%
3.9%
4.7%
5.4%
Total Population
2,819,973
9,437,895
91,874,288
328,016,242
1 Demographic percentage is based on the Census' 2015-2019 American Community Survey 5-year averages, at the block
group level, and include the 50 states, District of Columbia, and Puerto Rico. Total population is based on block level data
from the 2010 Decennial Census.
2 Minority population is the total population minus the white population.
3 To avoid double counting, the "Hispanic or Latino" category is treated as a distinct demographic category for these analyses.
A person who identifies as Hispanic or Latino is counted as Hispanic/Latino for this analysis, regardless of what race this
person may have also identified as in the Census.
Overall, the baseline demographic proximity analyses suggest that larger percentages of
Hispanics, Blacks, people below the poverty level, people with less educational attainment, and
people linguistically isolated are living within 5 km and 10 km of an affected EGU, compared to
national averages. It also finds larger percentages of people below the poverty level and with less
educational attainment living within 5 km and 10 km of a non-EGU facility. Relating these
7-8
-------
results to question 1 from Section 7.1, we find that there may be potential EJ concerns associated
with environmental stressors affected by the regulatory action (e.g., NO2) for certain population
groups of concern in the baseline, although NO2 air quality modeling was not performed.
Additionally, concerns suggested by the proximity analyses results cannot be related to potential
impacts from this proposed rulemaking resulting from ozone concentration decreases due to
long-range transport.
For additional information on the proximity analyses, see the memorandum Analysis of
Demographic Factors For Populations Living Near EGU and Non-EGUFacilities, in the
proposed rulemaking docket.
7.3.2 Tribal Lands Proximity Assessment
We conducted a tribal analysis to identify the total number of EGU and non-EGU
facilities located on and within 50 miles of tribal lands (Table 7-3).8 For the purpose of this
assessment, tribal lands refer to all lands associated with Federally recognized tribal entities.9
Using Geographic Information System (GIS) to map tribal lands and facilities, EPA found that of
the 851 EGUs included in this action, 38 are located on tribal lands and 176 are located within a
50-mile distance. Of the 251 non-EGUs facilities included in this action, 9 are located on tribal
lands and 87 non-EGUs are located within a 50-mile distance.
Table 7-3. Tribal Proximity Assessment
Total Number of
Affected Sources
Number of Affected Sources
with Tribes Within 50
Miles*
Number of Affected
Sources Located on Tribal
Lands
EGUs
851
176
38
Non-EGUs
251
87
9
EGUs and
Non-EGUs
1,102
168
47
* The total number of tribes within 50 miles of facilities is not a direct sum. Tribes located within 50
miles of both an EGU and non-EGU facility are only counted once.
8 It has been established through tribal consultation that a 50-mile (not kilometer) radius from tribal lands is a
sufficient distance to accomplish a tribal proximity analysis to address any concerns that a tribe might have on a
specific action.
9 This includes Federally recognized Reservations, Off-Reservation Trust Lands, and Census Oklahoma Tribal
Statistical Areas (OTSA).
7-9
-------
7.4 Ozone Exposure Analysis
To evaluate the potential for EJ concerns among potentially vulnerable populations
resulting from exposure to ozone under the baseline and regulatory control alternatives in this
proposal, we also assess the impact of NOx emission reductions on downwind ozone
concentrations. EPA presents an analysis of ozone concentrations associated with upwind NOx
emissions, characterizing the distribution of exposures both prior to and following
implementation of the proposed rule, as well as of the more and less stringent regulatory
alternatives, in 2023 and 2026. Under the proposed rule and more stringent scenario, the year of
full compliance is 2026 for both EGUs and non-EGUs. Under the less stringent scenario the year
of full compliance is 2028 for EGUs and 2026 for non-EGUs. Population variables considered
include race/ethnicity, poverty status, educational attainment, age, and sex (Table 7-4).10'11
Table 7-4. Populations Included in the Ozone Exposure Analysis
Demographic
Characteristics
Description
Ethnicity
Hispanic, Non-Hispanic
Race
Asian, American Indian, Black, White
Educational Attainment
High school degree or more, No high school degree
Poverty Status
Above/below 200% of the poverty line, Above/below the poverty line
Age
Children (0-17), Adults (18-64), Older Adults (65-99)
Sex
Female, Male
As this analysis is based on the same ozone spatial fields as the benefits assessment (see
Chapter 3 for a discussion of the spatial fields), it is subject to similar types of uncertainty (see
Chapter 5, Section 5.1.3 for a discussion of the uncertainty). A particularly germane limitation is
that ozone, being a secondary pollutant, is the byproduct of complex atmospheric chemistry such
10 Due to the consent decree deadline, we did not have time to evaluate or bring in stratified baseline incidence rates
or concentration-response functions relating to potentially evaluate at-risk populations. As results of a risk analysis
lacking stratified concentration-response and/or baseline incidence rates would not provide additional information
regarding population group impacts beyond exposure differences and age-related difference in baseline incidence,
this EJ analysis was limited to exposure only.
11 Population projections stratified by race/ethnicity, age, and sex are based on economic forecasting models
developed by Woods and Poole (Woods and Poole, 2015). The Woods and Poole database contains county-level
projections of population by age, sex, and race out to 2050, relative to a baseline using the 2010 Census data.
Population projections for each county are determined simultaneously with every other county in the U.S to consider
patterns of economic growth and migration. County-level estimates of population percentages within the poverty
status and educational attainment groups were derived from 2015-2019 5-year average ACS estimates. Additional
information can be found in Appendix J of the BenMAP-CE User's Manual (https://www.epa.gov/benmap/benmap-
ce-manual-and-appendices).
7-10
-------
that direct linkages cannot be made between specific affected facilities and downwind ozone
concentration changes based on available air quality modeling (see Chapter 3, Section 3.4).
Ozone concentration and exposure metrics can take many forms, although only a small
number are commonly used. The analysis presented here is based on the average April-
September warm season maximum daily 8-hour average ozone concentrations (AS-M03). This
is consistent with the health impact functions used in the benefits assessment (Chapter 5). As
developing spatial fields is time and resource intensive, the same spatial fields used for the
benefits analysis were also used for the ozone exposure analysis performed here to assess EJ
impacts.
The construct of the AS-M03 ozone metric used for this analysis should be kept in mind
when attempting to relate the results presented here to the ozone NAAQS and when interpreting
the confidence in the association between exposures and health effects. Specifically, the seasonal
average ozone metric used in this analysis is not constructed in a way that directly relates to
NAAQS design values, which are based on daily maximum 8-hour concentrations.12 Thus, AS-
M03 values reflecting seasonal average concentrations well below the level of the NAAQS at a
particular location do not necessarily indicate that the location does not experience any daily (8-
hour) exceedances of the ozone NAAQS. Relatedly, EPA is confident that reducing the highest
ambient ozone concentrations will result in substantial improvements in public health, including
reducing the risk of ozone-associated mortality. However, the Agency is less certain about the
public health implications of changes in relatively low ambient ozone concentrations. Most
health studies rely on a metric such as the warm-season average ozone concentration; as a result,
EPA typically utilizes air quality inputs such as the AS-M03 spatial fields in the benefits
assessment, and we judge them also to be the best available air quality inputs for this EJ ozone
exposure assessment. To further support the use of the AS-M03 spatial fields in this ozone
analysis, we compared baseline AS-M03 spatial fields with average baseline maximum daily 1-
hour average (MDA1) ozone concentrations spatial fields, also over the April-September warm
season, and found that average population ozone concentration trends within populations were
similar when considering either the AS-M03 or the MDA1 spatial fields. Therefore, we
12 Level of 70 ppb with an annual fourth-highest daily maximum 8-hour concentration, averaged over 3 years.
7-11
-------
performed ozone analyses using only the AS-M03 metric over the April-September warm
season.
The metric and averaging season are also relevant inputs to consider when interpreting
the results as they can affect the sharpness of pollutant gradients, an important factor when
associating exposure for different demographic populations. Figure 3-2 and Figure 3-3 in
Chapter 3, Section 3.4 of this RIA show maps of the baseline 12 km gridded AS-M03
concentrations in 2023 and 2026, respectively. As the AS-M03 seasonal metric is based on the
average of concentrations over more than 180 days in the spring and summer, the resulting
spatial fields are relatively smooth and do not display sharp gradients, compared to what might
be expected when looking at the spatial patterns of the average maximum daily 8-hour average
ozone concentrations on individual high ozone episode days.
The ozone exposure analyses begin with national- and state-level aggregated results
(Section 7.4.1) and then examine spatially resolved distributional results via figures (Section
7.4.2). Maps were not included as the magnitude of differences between populations observed is
relatively small and ozone gradients are relatively smooth (Section 7.1.1).
7.4.1 Aggregated Results
Results aggregated to the national and state levels provide an overview of the average
impacts within each population group. We provide baseline results in absolute terms (i.e., total
AS-M03 concentrations; Section 7.4.1.1) and regulatory alternative results in relative terms (i.e.,
the change in AS-M03 concentrations; Sections 7.4.1.2).
7.4.1.1 Baseline Assessment
Before evaluating the impacts of the proposal, it is important to understand baseline, or
pre-proposal, conditions. Below are the average baseline maximum daily 8-hour average ozone
concentrations in parts per billion (ppb) over the April-September warm season in the two
modeled future years, 2023 and 2026 (Figure 7-1). These concentrations represent the total
estimated ozone exposure burden averaged over the 6-month warm season each year and are
colored to more easily visualize differences in concentrations, with white coloring representing
the lowest concentrations and dark orange coloring representing the highest.
7-12
-------
Due to existing regulatory control programs reflected in the baseline, average ozone
concentrations are estimated to decrease across the overall population between 2023 and 2026 by
approximately 0.5 ppb. While many of the average ozone concentrations within the individual
population groups are estimated to be similar to or below average concentrations of the overall
reference group (i.e., total population of contiguous U.S.), certain populations are estimated to
experience higher average ozone concentrations in the baseline in both future years. The five
populations with the largest differences from the national average ozone concentration within the
subpopulation in both 2023 and 2026 as compared to the overall reference group were: American
Indians, Hispanics, Asians, the less educated, and children. These populations live in areas with
seasonal average baseline ozone concentrations of approximately 2.0, 1.9, 1.2, 0.3 and 0.2 ppb
higher than the national average concentrations, respectively.13 In contrast, national average
baseline ozone concentrations in the Black population are estimated to be about 1.2 ppb less than
the reference group in both 2023 and 2026. However, it is important to note that these are
aggregate results across broad areas and large numbers of people; aggregating results may
underestimate the impact in individual locations where there is both an ozone nonattainment
issue and a disproportionately large racial/ethnic population. Additionally, while AS-M03
exposures across all groups are relatively low, in the range of 40-43 ppb, these seasonal averages
do not necessarily indicate that individual locations do not experience exceedances of the
NAAQS. Thus, it is difficult to draw conclusions from this analysis about whether some
population subgroups experience higher daily maximum exposures than others in the baseline.
13 Differences in both 2023 and 2026 were calculated and averaged to generate these estimates, as differences
between the air quality in the two future years were similar.
7-13
-------
Population , , .
Groups Populations (Age Range)
Year
2023 2026
Reference All (0-99)
41.50 41.01
Ethnicity Non-Hispanic (0-99)
Hispanic (0-99)
41.04 40.50
43.36 42.94
Race White (0-99)
Asian (0-99)
Black (0-99)
American Indian (0-99)
41.59 41.10
42.65 42.19
40.33 39.79
43.53 43.07
Educational More educated (high school or more) (25-99)
Attainment Less educated (no high school) (25-99)
41.36 40.87
41.81 41.33
Poverty Above 200% of the poverty line (0-99)
Status Below 200% of the poverty line (0-99)
Above poverty line (0-99)
Below poverty line (0-99)
41.52 41.03
41.47 40.97
41.52 41.02
41.42 40.93
Age Children (0-17)
Adults (18-64)
Older Adults (64-99)
41.73 41.23
41.54 41.05
41.10 40.61
Sex Females (0-99)
Males (0-99)
41.49 41.00
41.51 41.02
Figure 7-1. Heat Map of the National Average AS-M03 Ozone Concentrations Across
Demographic Groups in the Baseline Assessment (ppb)
Connecting back to question 1 from EPA's EJ Technical Guidance, the national-level
baseline assessment of ozone concentrations suggests that there may be potential EJ concerns
associated with environmental stressors affected by the regulatory action for certain population
groups of concern in the baseline. Specifically, the data indicate that some population subgroups
evaluated may experience slightly elevated seasonal average ozone concentrations in the baseline
as compared to the reference group nationally.
7.4.1.2 Regulatory Alternatives Assessment
While the baseline provides information regarding overall ozone exposures, it does not
provide information regarding how the proposed rulemaking will impact various populations. To
better understand this, we evaluated how NOx emissions reductions affecting ozone
concentrations downwind affected average ozone concentrations experienced by each
subpopulation under the regulatory alternatives in 2023 and 2026, again with dark orange
7-14
-------
coloring representing the highest ozone concentration (Figure 7-2).14 Although NOx reductions
from this proposed rule will also reduce concentrations of fine particle (PM2.5) and NO2 and this
proposed rule is also projected to reduce carbon dioxide (CO2) emissions, this analysis is only a
partial representation of the distributions of potential impacts.
Figure 7-2 shows how ozone concentrations may change in 2023 (from EGU controls
only) and in 2026 (from EGUs controls, non-EGU controls, and EGU and non-EGU controls
combined) under the proposal, the less stringent alternative, and the more stringent alternative.
Under the proposed rule, the population-weighted seasonal average ozone reduction in the
overall reference group is approximately 0.02 ppb in 2023 and 0.36 ppb in 2026. In 2026,
roughly 0.17 ppb of ozone concentration reductions are attributable to affected EGUs and 0.20
ppb are attributable to non-EGU affected facilities. Hispanics, Asians, and American Indians are
estimated to experience reductions in AS-M03 that are slightly less than the reference group in
both 2023 and 2026.15 Pairing these results with the national baseline ozone concentrations
shown in Section 7.4.1.1 suggests that although this proposal lessens overall ozone
concentrations within each population as compared with the baseline levels, reductions are
smallest in populations with higher baseline ozone concentrations. However, the relative
differences in the policy impacts are small (e.g., on the order of -0.1 ppb less reduction in ozone
among these subpopulations as compared to the reference group) and substantially smaller than
the baseline differences across these subpopulations (~2 ppb). Conversely, Blacks and non-
Hispanics, who on average experience lower ozone concentrations than the reference group
under the baseline, are estimated to experience average ozone concentration reductions slightly
greater than the reference group in 2023 and 2026 (e.g., roughly 0.06 ppb greater reduction in
ozone concentrations than the reference group).16 Again, these differences are small relative to
14 The proposed rule identifies unit level emissions rates on EGUs in the 2025 run year, while the less stringent
alternative identifies these emissions rates in the 2028 run year. The unit level emissions rate limits drive much of
the EGU retirement activity, and retirements are delayed in the less stringent alternative relative to the proposed
rule. Since the power sector model is forward looking, it has an incentive to run units harder before they retire. This
incentive is lower in the less stringent alternative relative to the proposed rule due to delayed retirements. As such,
emissions are slightly lower in 2023 in some states in the less stringent alternative relative to the proposed rule,
leading to slightly greater emission reductions.
15 A smaller or greater ozone concentration reduction is defined as at least a 0.2 ppb less than the national average
ozone concentration within the subpopulation in 2026.
16 Due to the consent decree deadline, we did not have time to evaluate or bring in stratified baseline incidence rates
or concentration-response functions relating to potentially evaluate at-risk populations. As results of a risk analysis
lacking stratified concentration-response and/or baseline incidence rates would not provide additional information
7-15
-------
the overall reduction in ozone concentrations across all populations. We report analytics only to
the hundredths decimal place for ppb of ozone, as uncertainty with regard to modeling accuracy
is likely larger for very small differences.
Under the less stringent regulatory alternative in 2023 there are similar magnitudes of
ozone concentration reductions in the reference group as in the proposed rule, and a greater
reduction in average ozone concentration in the more stringent regulatory alternative, within all
population groups.17 In 2026 the less stringent and more stringent alternatives are estimated to
result in smaller and larger reductions in ozone concentrations, respectively, as compared to the
proposed rule. Notably, the less stringent alternative has smaller ozone concentration reductions
from EGUs than from non-EGUs, whereas the more stringent alternative has slightly larger
ozone concentration reductions from both EGUs and non-EGUs.
The relative population-weighted AS-M03 ozone concentration reduction contributions
from EGUs and non-EGUs can be directly compared in 2026. For all regulatory control
alternatives and across all populations, non-EGU NOx emission reductions are estimated to
result in greater ozone concentration reductions than the EGU NOx emissions reductions. The
difference is relatively small under the policy and more stringent alternatives but is greater under
the less stringent alternative.
regarding population group impacts beyond exposure differences and age-related difference in baseline incidence,
this EJ analysis was limited to exposure only.
17 The proposed rule identifies unit level emissions rates on EGUs in the 2025 run year, while the less stringent
alternative identifies these emissions rates in the 2028 run year. The unit level emissions rate limits drive much of
the EGU retirement activity, and retirements are delayed in the less stringent alternative relative to the proposed
rule. Since the power sector model is forward looking, it has an incentive to run units harder before they retire. This
incentive is lower in the less stringent alternative relative to the proposed rule due to delayed retirements. As such,
emissions are slightly lower in 2023 in some states in the less stringent alternative relative to the proposed rule,
leading to slightly greater emission reductions.
7-16
-------
Policy
2023
Less
More
Policy
2026
Less
More
Population
Groups
Populations (Age Range)
EGU
EGU
EGU
EGU
NonEGU
EGU+
NonEGU
EGU
NonEGU
EGU+
NonEGU
EGU
NonEGU
EGU+
NonEGU
Reference
All (0-99)
0.02
0.02
0.02
0.17
0.20
0.36
0.08
0.17
0.25
0.19
0.21
0.41
Ethnicity
Non-Hispanic (0-99)
0.02
0.02
0.02
0.18
0.21
0.39
0.08
0.18
0.27
0.21
0.22
0.44
Hispanic (0-99)
0.01
0.01
0.01
0.11
0.15
0.26
0.06
0.14
0.19
0.13
0.17
0.30
Race
White (0-99)
0.02
0.02
0.02
0.17
0.19
0.36
0.08
0.17
0.25
0.19
0.21
0.40
Asian (0-99)
American Indian (0-99)
0.01
0.01
0.01
0.01
0.01
0.01
0.11
0.12
0.17
0.16
0.28
0.28
0.05
0.06
0.15
0.14
0.20
0.20
0.14
0.14
0.18
0.17
0.32
0.31
Black (0-99)
0.02
0.02
0.02
0.20
0.22
0.42
0.09
0.19
0.29
0.23
0.24
0.47
Educational More educated (25-99)
Attainment Less educated (no high school) (25-99)
0.02
0.02
0.02
0.02
0.02
0.02
0.17
0.16
0.20
0.19
0.36
0.35
0.08
0.08
0.17
0.17
0.25
0.24
0.20
0.18
0.21
0.21
0.41
0.39
Poverty
Status
Above 200% of the poverty line (0-99)
Below 200% of the poverty line (0-99)
Above poverty line (0-99)
Below poverty line (0-99)
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.17
0.17
0.17
0.17
0.20
0.20
0.20
0.20
0.36
0.37
0.36
0.37
0.08
0.08
0.08
0.08
0.17
0.17
0.17
0.17
0.25
0.25
0.25
0.26
0.19
0.20
0.19
0.20
0.21
0.21
0.21
0.22
0.41
0.41
0.41
0.41
Age
Children (0-17)
Adults (18-64)
Older Adults (64-99)
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.17
0.17
0.17
0.20
0.20
0.19
0.37
0.36
0.36
0.08
0.08
0.08
0.18
0.17
0.17
0.26
0.25
0.25
0.20
0.19
0.19
0.22
0.21
0.21
0.41
0.41
0.40
Sex
Females (0-99)
Males (0-99)
0.02
0.02
0.02
0.02
0.02
0.02
0.17
0.17
0.20
0.20
0.37
0.36
0.08
0.08
0.17
0.17
0.25
0.25
0.20
0.19
0.21
0.21
0.41
0.41
Figure 7-2. Heat Map of the National Average AS-M03 Ozone Concentration Reductions
by Demographic Group, Regulatory Alternative, and Affected Facilities (ppb)
The goal of this proposed action is to require NOx emissions reductions that will
eliminate significant contribution to nonattainment or interference with maintenance of the 2015
ozone NAAQS in downwind areas.18 As upwind emissions reductions necessary to achieve this
goal will not affect ozone concentrations uniformly within each state, we provide AS-M03
ozone concentration reductions by state and demographic population for the combined EGU and
non-EGU proposed alternative in 2026 for the 48 states in the contiguous U.S. (Figure 7-3). In
this heat map dark orange indicates larger AS-M03 reductions, although the demographic
groups are now shown as columns and each state as a row. On average, the state-specific
reference populations are projected to experience reductions in AS-M03 concentrations by up to
1.02 ppb and populations potentially of concern are projected to experience reductions in AS-
M03 concentrations by up to 1.15 ppb.
Air quality improvements across demographic groups within individual states are
variable. For example, although nationally Hispanics experienced a smaller improvement in air
quality than the overall average, this effect was observed in only 14 of the 48 states. In addition,
for Hispanics there were greater average improvements of AS-M03 concentrations in the two
states with the largest AS-M03 concentration reductions, Kentucky and Louisiana. Therefore,
18 See Section 1 of the proposal preamble for a discussion of the states included in the proposal and their proposed
requirements for EGUs and non-EGUs.
7-17
-------
small differences in air quality improvements observed at the national level are not experienced
consistently across geographic areas.
An important limitation of this state-level analysis is that the influence of the number of
people in the state is not reflected in the results, whereas the national-level results above weight
air quality changes by population. For example, even though there is only a small reduction in
AS-M03 concentration from this action in California, the state's large population will contribute
substantially to the national averages. Conversely, while the largest AS-M03 concentration
reduction occurs in Kentucky, as of 2021 it is the 26th most populated state with approximately
4.5 million people and will contribute less to the national population-weighted AS-M03
information than California.
7-18
-------
Year / Regulatory Alternative / Facilities / Population Groups / Populations
2026
Policy
EGU+ NonEGU
Ref
Race/Ethnicity
Poverty Status Education
Ages
Sex
>
3
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
0.39 0.39 0.39 0.40 0.39 0.38 0.41 0.40 0.40 0.39 0.39 0.39 0.40 0.39 0.39 0.40 0.39 0.39
0.04 0.05 0.04 0.04 0.04 0.04 0.06
0.04 0.04 0.04 0.05 0.04 0.04 0.04 0.04 0.05 0.04 0.04
0.89 0.90 0.80 0.87 0.81 1.02 0.79
0.89 0.89 0.89 0.89 0.87 0.89 0.89 0.89 0.89 0.89 0.89
0.11 0.10 0.11 0.11 0.10 0.11 0.11
0.11 0.11 0.11 0.11 0.11 0.10 0.10
0.10 0.11 0.11 0.11 0.11 0.10 0.11 0.11 0.10 0.11 0.11
0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11
0.23 0.23 0.24 0.23 0.23 0.23 0.23 0.23 0.23 0.23 0.23 0.23 0.23 0.23 0.23 0.23 0.23 0.23
0.38 0.38 0.39 0.38 0.40 0.39 0.38 0.39 0.39 0.38 0.38 0.38 0.38 0.39 0.39 0.38 0.39 0.38
0.09 0.10 0.07 0.09 0.10 0.09 0.11 0.10 0.09 0.09 0.09 0.09 0.09 0.10 0.09 0.09 0.09 0.10
0.27 0.27 0.28 0.28 0.29 0.27 0.28 0.28 0.28 0.27 0.27 0.27 0.28 0.27 0.27 0.27 0.27 0.27
0.02
0.02 0.02 0.02 0.02 0.02 0.03
0.02 0.02 0.02 0.02 0.02 0.02
0.03 0.02 0.02 0.02 0.02
0.71
0.72 0.66 0.71 0.66 0.73 0.69
0.71 0.71 0.72 0.73 0.71 0.71
0.71 0.71 0.72
0.71 0.71
0.84
0.84 0.78 0.84 0.82 0.81 0.81
0.84 0.84 0.83 0.83 0.83 0.84
0.83 0.84 0.84
0.84 0.84
0.38 0.38 0.37 0.38 0.39 0.40 0.35 0.38 0.38 0.38 0.39 0.38 0.38 0.38 0.38 0.38 0.38 0.38
0.45 0.45 0.43 0.45 0.47 0.47 0.47 0.45 0.45 0.45 0.45 0.44 0.45 0.45 0.45 0.45 0.45 0.45
1.02
0.97
1.02 1.07 1.01 1.09 1.15 1.03
0.97 1.00 0.97 1.00 0.98 0.87
1.05 1.04 0.97 0.96 0.94 1.03
0.98 0.98 0.96 0.96 0.96 0.98
1.03 1.02 1.02
0.97 0.98 0.97
1.02 1.02
0.98 0.97
0.10
0.43
0.17
0.62
0.27
0.10 0.10
0.43 0.43
0.17 0.16
0.62 0.65
0.27 0.27
0.10 0.10
0.44 0.44
0.17 0.16
0.62 0.63
0.27 0.28
0.10 0,
0.42 0.
0.16 0.
0.63 0.
0.27 0.
09 0.10
43 0.43
16 0.16
56 0.62
22 0.27
0.10
0.43
0.17
0.62
0.27
0.10 0
0.43 0.
0.17 0.
0.62 0
0.26 0.
10 0.10
43 0.43
17 0.17
62 0.62
26 0.26
0.10 0.10
0.43 0.43
0.17 0.17
0.62 0.63
0.27 0.27
0.10 0
0.43 0
0.16 0
0.62 0
0.27 0
10 0.10
43 0.43
17 0.17
61 0.62
26 0.27
0.10
0.43
0.17
0.62
0.27
0.73
0.81
0.73 0.76
0.81 0.73
0.74 0.79
0.79 0.88
0.72 0.
0.91 0.
68 0.73
72 0.82
0.73
0.81
0.73 0
0.78 0
73 0.72
79 0.80
0.73 0.73
0.81 0.80
0.73 0
0.81 0
73 0.73
81 0.81
0.73
0.81
0.03
0.26
0.07
0.16
0.32
0.11
0.28
0.32
0.09
0.76
0.62
0.00
0.49
0.18
0.24
0.17
0.56
0.46
0.21
0.20
0.45
0.00
0.03 0.03
0.26 0.26
0.07 0.07
0.16 0.16
0.32 0.31
0.10 0.11
0.29 0.27
0.32 0.32
0.10 0.09
0.03 0.02
0.26 0.27
0.07 0.08
0.16 0.16
0.32 0.32
0.11 0.10
0.29 0.27
0.32 0.32
0.10 0.10
0.03 0.
0.28 0.
0.08 0.
0.16 0.
0.32 0.
0.11 0.
0.27 0.
0.32 0.
0.10 0.
04 0.03
25 0.26
06 0.07
16 0.16
32 0.32
09 0.11
28 0.28
28 0.32
08 0.10
0.03
0.26
0.07
0.16
0.32
0.11
0.28
0.32
0.09
0.03 0
0.25 0
0.07 0.
0.16 0
0.32 0
0.11 0.
0.28 0
0.32 0
0.09 0
03 0.03
26 0.25
07 0.07
16 0.16
32 0.32
11 0.11
28 0.27
32 0.32
09 0.09
0.03 0.03
0.26 0.26
0.07 0.07
0.16 0.16
0.32 0.32
0.10 0.11
0.28 0.28
0.32 0.32
0.10 0.09
0.03 0.
0.26 0.
0.07 0.
0.16 0.
0.32 0.
0.11 0.
0.28 0.
0.32 0.
0.10 0.
03 0.03
25 0.26
07 0.07
16 0.16
32 0.32
10 0.11
29 0.28
32 0.32
09 0.10
0.03
0.26
0.07
0.16
0.32
0.11
0.28
0.32
0.09
0.76 0.74
0.63 0.61
0.00 0.00
0.50 0.45
0.18 0.18
0.24 0.24
0.17 0.18
0.55 0.58
0.51 0.41
0.21 0.23
0.20 0.20
0.45 0.45
0.00 0.00
0.76 0.78
0.62 0.63
0.00 0.00
0.50 0.47
0.18 0.18
0.24 0.25
0.17 0.18
0.54 0.58
0.45 0.50
0.21 0.23
0.20 0.20
0.46 0.45
0.00 0.00
0.76 0.
0.63 0.
000 0.
0.46 0.
0.18 0.
0.23 0.
0.19 0.
0.64 0.
0.53 0.
0.23 0.
0.20 0.
0.43 0.
0.00 0.
76 0.76
65 0.63
00 0.00
47 0.49
18 0.18
24 0.24
16 0.17
55 0.55
48 0.48
19 0.22
19 0.20
45 0.45
00 0.00
0.76
0.62
0.00
0.49
0.18
0.24
0.17
0.56
0.47
0.21
0.20
0.45
0.00
0.76 0.
0.62 0
0.00 0
0.49 0
0.18 0
0.24 0.
0.17 0
0.56 0
0.44 0
0.21 0
0.20 0
0.46 0
0.00 0
76 0.76
62 0.62
0.76 0.76
0.63 0.62
0.76 0.
0.62 0.
75 0.76
63 0.63
0.76
0.62
00 0.00
49 0.48
18 0.18
24 0.24
17 0.17
56 0.55
43 0.43
21 0.21
20 0.20
46 0.46
00 0.00
0.00 0.00
0.50 0.49
0.18 0.18
0.24 0.24
0.17 0.17
0.55 0.57
0.47 0.46
0.21 0.22
0.20 0.20
0.45 0.45
0.00 0.00
0.00 0
0.49 0
0.18 0
0.24 0
0.17 0
0.56 0
0.47 0
0.22 0
0.20 0
0.45 0
0.00 0
00 0.00
50 0.49
18 0.18
24 0.24
17 0.17
54 0.56
47 0.47
20 0.21
20 0.20
46 0.45
00 0.00
0.00
0.49
0.18
0.24
0.17
0.56
0.46
0.21
0.20
0.45
0.00
0.70
0.46
0.17
0.70 0.65
0.46 0.51
0.17 0.18
0.70 0.69
0.46 0.47
0.18 0.17
0.68 0.
0.52 0.
0.18 0.
70 0.70
43 0.46
13 0.18
0.70
0.46
0.17
0.70 0
0.47 0
0.17 0
71 0.69
47 0.47
18 0.18
0.70 0.70
0.46 0.47
0.17 0.18
0.70 0
0.47 0
0.18 0
70 0.70
46 0.47
17 0.17
0.70
0.46
0.17
Figure 7-3. Heat Map of State Average AS-M03 Ozone Concentration Reductions by
Demographic Group for EGUs and Non-EGUs Under the Proposed Rule (ppb)
7-19
-------
Connecting back to question 2 from EPA's EJ Technical Guidance again, the aggregated
analyses of ozone exposures under the various regulatory alternatives in 2023 and 2026 do not
suggest that there may be potential EJ concerns associated with environmental stressors affected
by the regulatory action for population groups evaluated.
7.4.2 Distributional Results
While aggregated national- and state-level average ozone concentration results (Section
7.4.1) provide an overview of potential exposure differences across populations, detailed
information on the distribution of AS-M03 ozone exposures within populations, and specifically
the portions of each population experiencing ozone concentration changes due to the proposal,
can provide a more comprehensive understanding of analytical results. Figures in this section
present cumulative counts of each population exposed to ascending levels of AS-M03 ozone
concentrations across the contiguous U.S. Results allow evaluation of what percentage of each
subpopulation (e.g., Asians) in the contiguous U.S. experience average baseline ozone
concentrations at or below certain AS-M03 ozone concentrations (e.g., 40 ppb) compared to
what percentage of the overall reference group (i.e., the total population of contiguous U.S.)
experiences ozone concentrations in that same range. More specifically, to permit the direct
comparison of demographic populations with different absolute numbers (e.g., the large overall
reference population with the much smaller number of Asians), we plot the running sum of each
population as a percentage against the increasing baseline ozone concentration (Section 7.4.2.1)
and ozone concentration changes from NOx emission reductions under the regulatory
alternatives (Section 7.4.2.2). This distributional EJ analysis is also subject to additional
uncertainties related to more highly resolved input parameters and additional assumptions (U.S.
EPA 2021d, Section 6). For example, this analysis does not account for potential difference in
underlying susceptibility, vulnerability, or risk factors across populations to AS-M03 exposure.
We also did not evaluate whether concentrations experienced by different groups persist across
the distribution of air quality. Nor could we include information about differences in other
factors that could affect the likelihood of adverse impacts (e.g., exercise patterns) across groups.
Therefore, this analysis should not be used to assert that there are meaningful differences in
either the baseline or the proposal.
7-20
-------
7.4.2.1 Baseline Assessment
Under baseline conditions approximately 80% of the overall reference population (i.e.,
total population of the contiguous U.S.) resides in areas of AS-M03 ozone concentrations at or
less than about 45 ppb in 2023 and at or less than about 44 ppb in 2026 (Figure 7-4). Most of this
population experiences AS-M03 ozone concentrations between 30-44 ppb. In contrast, the 20%
of the overall reference population residing in areas of the highest baseline ozone concentrations
experiences concentrations of between about 45-70 ppb.
As was observed in the national average ozone concentration analysis (Section 7.4.1),
projected ozone concentration distributions for some populations visibly differed from the
reference population distribution in 2023 and 2026. Notably, there were proportionally more
Hispanics and American Indians residing in areas of ozone concentrations above approximately
40 ppb than in other demographic groups evaluated. Conversely, at 30-38 ppb AS-M03 the
Hispanic population is exposed to disproportionately lower ozone concentrations than the White
population, reducing the overall impact observed in the national average above. The distribution
of the Asian population's exposure to ozone concentrations also indicated proportionally higher
exposures as compared to the reference population, but to a lesser degree and across nearly the
full array of ozone concentrations. There was also a slight shift in the distribution of less
educated populations at ozone concentrations above about 43 ppb, indicating that a greater
proportion of less educated people reside in areas of slightly higher ozone concentrations than
more educated people. Exposure of populations differing by poverty status, age, and sex did not
differ from the reference population with regard to national average ozone concentrations (Figure
7-1). These populations also did not substantially differ across the distribution of baseline ozone
concentrations in 2023 and 2026 (Figure 7-4).
7-21
-------
Year
Population Groups 2023 2026
° 100%-
1 =
OJ .Q
Reference a.
3
1
3 o%.
I I Less educated
/ | Wore educated
XJ
T>
O
<
CD
00
21
C*
if)
Cumulative Percent ol
Population
cn o
o o o
2se vP
o>
/ ¦ Below 200% of me poverty line /
J Below poverty hne /
/ | Above 200% of the poverty line i
i 1 Above poverty bne i
>
co
0>
(Cumulative Percent of
Population
cn o
o o o
^ ^
j ¦ Children (0-17)
¦ Adults (18-64)
/ ¦ Older Adults (64-99)
y
° 100%-
1 =
Q-
42 t£
i
3 o%.
/ | Females
/ ¦ Mates
XLL
30 40 50 60 70 30 40 50 60 70
Ozone Concentration (ppb) Ozone Concentration (ppb)
Figure 7-4. Distributions of Baseline Ozone Concentrations Across Populations in 2023 and
2026
7-22
-------
Connecting back to question 1 from EPA's EJ Technical Guidance again, this
distributional analysis of baseline ozone concentrations further supports that there may be
potential EJ concerns associated with environmental stressors affected by the regulatory action
for certain population groups of concern evaluated in the baseline.
7.4.2.2 Regulatory Alternatives Assessment
Distributions of 12 km gridded ozone concentration reductions from NOx emission
reductions in 2023 and 2026 are shown in Figure 7-5 and Figure 7-6, respectively. As with the
national average results (Section 7.4.1.1), the horizontal axes scales are different than in the
baseline analyses, indicating the disproportionate impacts of the proposal are substantially
smaller than under baseline conditions.
NOx emission reductions from affected EGUs under the three regulatory alternatives
analyzed in this proposed rulemaking are evaluated in 2023 (Figure 7-5). Approximately 90% of
the overall reference population experienced an ozone concentration reduction of less than 0.04
ppb.
There are slight differences in the ozone concentration reductions across population
demographics and regulatory alternatives. Proportionally, Hispanics, Asians, and American
Indians experience smaller ozone concentration reductions under the regulatory alternatives than
the overall reference population in 2023, by a very small amount. Alternatively, the distribution
of ozone concentration reductions for Blacks is greater than the reference population only in the
smallest half of ozone concentration reductions.
7-23
-------
Population
Groups
Policy
2023
Facilities / Scenario / Year
EGU
Less Stnngent
2023
More Stringent
2023
100%"
5 ° Q
Reference p § ¦§ 50%-
5 u a
,? 0) o
O Q_ Q_
0%.
100%-
s ° i
Ethnicity & = c "5 cnn.
r, p 8 "5 50%-
Race = " a.
3 ® O
O Q_ Q_
0%.
/ZT'—
Asian
// ¦ Black
¦ Hispanic
Amencan Indian
lr ¦ Non-Hispanic
¦ White
w
100%
§ ° o
Educational = = T5
AU l P O = 50%-
Attainment | £ a.
= aj o
O Q. Q.
0%.
f 1 Less educated
J ¦ More educated
j
100%"
S "o o
Poverty | = 50%.
Status § H q.
O) o
O Q. Q.
0%.
/ ¦ Betow 200% of the poverty line
/ | Betow poverty line
y ¦ Above 200% of the poverty line
j ¦ Above poverty line
f
y
100%-
> ° 1
Age 1 | 1 50%-
= Sj O
O Q. Q.
0%.
/ ¦ Children (0-17)
J ¦ Adults (18-64)
/ IZ Older Adults (64-99)
100%"
.i 2 J
Sex I § 3 50%-
O Q_ Q.
0%.
/ ¦ Females
J ¦ Mates
/¦
y
0.00 0.02 0.04 0.06 0.08
Ozone Reduction (ppb) *
0.00 0 02 0.04 0 06 0.08
Ozone Reduction (ppb) *¦
0.00 0.02 0 04 0.06 008
Ozone Reduction (ppb) *
Figure 7-5. Distributions of Ozone Concentration Reductions from EGU NOx Emission
Reductions Across Regulatory Alternatives and Populations in 2023
NOx emission reductions from affected EGU and non-EGU facilities under the three
regulatory alternatives analyzed in this proposed rulemaking in 2026 are evaluated in Figure 7-5.
The magnitude of ozone concentration reductions from affected EGUs is estimated to be roughly
10-fold greater in 2026 compared to 2023. Approximately 90% of the overall reference
population experiences a fairly linear distribution of ozone concentration reductions, although
the steepness of the distribution varies by regulatory alternative and facility type.
7-24
-------
There are differences in the ozone concentration reductions across population
demographics and affected facility types. However, distributions are reasonably similar across
the three regulatory alternatives. Hispanics, Asians, and American Indians experience
proportionally smaller ozone concentration reductions from EGU and non-EGUNOx emission
reductions under the regulatory alternatives than the overall reference population in 2026.
Alternatively, the distribution of ozone concentration reductions for Blacks is greater than the
reference population. This shift is greatest in the 30% of the population experiencing the smallest
ozone concentration reductions from EGU NOx emission reductions.
There is a shift in the distribution of ozone concertation reductions between more and less
educated populations in 2026 that differs by affected facility. Less educated people experience
disproportionately smaller ozone concentration reductions from affected EGUs at lower ozone
concentration reductions (approximately less than 0.2 ppb), whereas less educated people
experience disproportionately smaller ozone concentration reductions from affected non-EGUs
at larger ozone concentration reductions (approximately greater than 0.2 ppb).
As shown in Figure 7-6, both above and below the poverty line and 200% of the poverty
line were evaluated, with comparisons between both being very similar. Across about the 60th-
90th percentiles of people below the poverty line or 200% of the poverty line in 2026 experience
disproportionately smaller ozone concentration reductions from affected EGUs and non-EGUs,
as shown by the small shifts to the right in the population distributions. Substantial differences in
ozone exposure reductions were not observed in the distributions of populations stratified by age
or sex.
7-25
-------
Population
Groups
Reference | g 3 50%
S P o
o Q- Q-
0%.
100%
JoS
! | ¦§ 50%
3 0) O
O Q- Q.
Ethnicity &
Race
Educational
Attainment
J o §
w c
0%.
100%
50%
0%
100%
Poverty
Status
I O o
111 50%
I 6 §•
0 Q. Q.
0%.
100%
-------
of that consideration, she also considered risks to minority and low-income individuals and
communities, finding that certain parts of the U.S. population may be especially vulnerable based
on their characteristics or circumstances. These groups include economically and socially
disadvantaged communities; individuals at vulnerable lifestages, such as the elderly, the very
young, and pregnant or nursing women; those already in poor health or with comorbidities; the
disabled; those experiencing homelessness, mental illness, or substance abuse; and/or Indigenous
or minority populations dependent on one or limited resources for subsistence due to factors
including but not limited to geography, access, and mobility.
Scientific assessment reports produced over the past decade by the U.S. Global Change
Research Program (USGCRP),19'20 the Intergovernmental Panel on Climate Change
19 USGCRP, 2018: Impacts, Risks, and Adaptation in the United States: Fourth National Climate Assessment,
Volume II [Reidmiller, D.R., C.W. Aveiy, D.R. Easterling, K.E. Kunkel, K.L.M. Lewis, T.K. Maycock, and B.C.
Stewart (eds.)]. U.S. Global Change Research Program, Washington, DC, USA, 1515 pp. doi: 10.7930/NCA4.2018.
20 USGCRP, 2016: The Impacts of Climate Change on Human Health in the United States: A Scientific Assessment.
Crimmins, A., J. Balbus, J.L. Gamble, C.B. Beard, J.E. Bell, D. Dodgen, R.J. Eisen, N. Fann, M.D. Hawkins, S.C.
Herring, L. Jantarasami, D.M. Mills, S. Saha, M.C. Sarofim, J. Trtanj, and L. Ziska, Eds. U.S. Global Change
Research Program, Washington, DC, 312 pp. http://dx.doi.org/10.7930/J0R49NQX
7-27
-------
(IPCC),21'22'23'24 and the National Academies of Science, Engineering, and Medicine25'26 add
more evidence that the impacts of climate change raise potential environmental justice concerns.
These reports conclude that poorer or predominantly non-White communities can be especially
vulnerable to climate change impacts because they tend to have limited adaptive capacities and
are more dependent on climate-sensitive resources such as local water and food supplies or have
less access to social and information resources. Some communities of color, specifically
populations defined jointly by ethnic/racial characteristics and geographic location, may be
uniquely vulnerable to climate change health impacts in the United States. In particular, the 2016
scientific assessment on the Impacts of Climate Change on Human Health27 found with high
confidence that vulnerabilities are place- and time-specific, lifestages and ages are linked to
immediate and future health impacts, and social determinants of health are linked to greater
extent and severity of climate change-related health impacts.
21 Oppenheimer, M., M. Campos, R.Warren, J. Birkmann, G. Luber, B. O'Neill, and K. Takahashi, 2014: Emergent
risks and key vulnerabilities. In: Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and
Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel
on Climate Change [Field, C.B., V.R. Barros, D.J. Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatteijee,
K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, and
L.L.White (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 1039-
1099.
22 Porter, J.R., L. Xie, A.J. Challinor, K. Cochrane, S.M. Howden, M.M. Iqbal, D.B. Lobell, and M.I. Travasso,
2014: Food security and food production systems. In: Climate Change 2014: Impacts, Adaptation, and Vulnerability.
Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change [Field, C.B., V.R. Barros, D.J. Dokken, K.J. Mach, M.D. Mastrandrea,
T.E. Bilir, M. Chatteijee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken,
P.R. Mastrandrea, and L.L.White (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York,
NY, USA, pp. 485-533.
23 Smith, K.R., A.Woodward, D. Campbell-Lendrum, D.D. Chadee, Y. Honda, Q. Liu, J.M. Olwoch, B. Revich, and
R. Sauerborn, 2014: Human health: impacts, adaptation, and co-benefits. In: Climate Change 2014: Impacts,
Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Field, C.B., V.R. Barros, D.J. Dokken, K.J.
Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatteijee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S.
Kissel,A.N. Levy, S. MacCracken, P.R. Mastrandrea, and L.L.White (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, pp. 709-754.
24 IPCC, 2018: Global Warming of 1.5°C.An IPCC Special Report on the impacts of global warming of 1.5°C above
pre-industrial levels and related global greenhouse gas emission pathways, in the context of strengthening the global
response to the threat of climate change, sustainable development, and efforts to eradicate poverty [Masson-
Delmotte, V., P. Zhai, H.-O. Portner, D. Roberts, J. Skea, P.R. Shukla, A. Pirani, W. Moufouma-Okia, C. Pean, R.
Pidcock, S. Connors, J.B.R. Matthews, Y. Chen, X. Zhou, M.I. Gomis, E. Lonnoy, T. Maycock, M. Tignor, and T.
Waterfield (eds.)]. In Press.
25 National Research Council. 2011. America's Climate Choices. Washington, DC: The National Academies Press.
https://doi.org/10.17226/12781.
26 National Academies of Sciences, Engineering, and Medicine. 2017. Communities in Action: Pathways to Health
Equity. Washington, DC: The National Academies Press, https://doi.org/10.17226/24624.
27 USGCRP, 2016: The Impacts of Climate Change on Human Health in the United States: A Scientific Assessment
7-28
-------
In a 2021 report, EPA considered the degree to which four socially vulnerable
populations—defined based on income, educational attainment, race and ethnicity, and age—
may be more exposed to the highest impacts of climate change.28 The report found that Blacks
and African Americans are approximately 40% more likely to live in areas of the U.S. projected
to experience the highest increases in mortality rates due to changes in extreme temperatures.
Additionally, Hispanic and Latino individuals in weather-exposed industries were found to be
43% more likely to currently live in areas with the highest projected labor hour losses due to
extreme temperatures. American Indian and Alaska Native individuals are projected to be 48%
more likely to currently live in areas where the highest percentage of land may be inundated by
sea level rise. Overall, the report confirmed findings of broader climate science assessments that
Americans identifying as people of color, those with low income, and those without a high
school diploma face disproportionate risks of experiencing the most damaging impacts of climate
change.
These findings suggest that CO2 reductions from NOx emission reductions under this
proposed rule may benefit disproportionately impacted populations. However, as we have not
conducted the wide-ranging analyses that would be needed to assess the specific impacts of this
proposed rule on the multiple climate-EJ interactions described above, we cannot analyze the
potential impacts of the proposed rule quantitatively.
7.6 Qualitative Assessment of PM2.5
Health studies have shown a significant association between exposure to particle
pollution and health risks, including premature death (U.S. EPA 2019 and Chapter 5). Particulate
matter with a mean aerodynamic diameter less than or equal to 2.5 |im (PM2.5) reductions are
expected from this proposed action but were not modeled for baseline or regulatory alternatives
under this proposed rulemaking. Therefore, similar analyses of disproportionate PM2.5 impacts,
as was done for ozone concentrations in Section 7.4, could not be performed. However, a brief
qualitative discussion of the potential for disproportionate PM2.5 impacts is provided.
In general, both recent publications and analyses by the EPA suggest that the burden of
PM2.5 exposures and impacts may disproportionately affect certain groups, such as Black and
28 EPA 2021. Climate Change and Social Vulnerability in the United States: A Focus on Six Impacts. U.S.
Environmental Protection Agency, EPA 430-R-21-003.
7-29
-------
Hispanic populations (e.g., Bell 2012, Bravo 2016, Kelly 2021, U.S. EPA 2020, U.S. EPA
2021a, U.S. EPA 2021c). PM2.5 reductions from NOx emission reductions under this proposed
rule may have benefits for disproportionately impacted populations. However, as we have not
conducted air quality modeling of PM2.5, we cannot analyze these potential impacts of the
proposed rule quantitatively.
7.7 Summary
As with all EJ analyses, data limitations make it quite possible that disparities may exist
that our analysis did not identify. This is especially relevant for potential EJ characteristics,
environmental impacts, and more granular spatial resolutions that were not evaluated.
Additionally, EJ concerns for each rulemaking are unique and should be considered on a case-
by-case basis, so results similar to those presented here should not be assumed for other
rulemakings.
For the proposal, we quantitatively evaluate 1) the proximity of affected facilities to
potentially disadvantaged populations (Section 7.3.1), 2) the potential for disproportionate total
ozone concentrations in the baseline across different demographic groups (Sections 7.4.1.1 and
7.4.2.1), and 3) how regulatory alternatives differentially impact the ozone concentration
changes experienced by different demographic populations (Sections 7.4.1.2 and 7.4.2.2). Each
of these analyses depends on mutually exclusive assumptions, was performed to answer separate
questions, and is associated with unique limitations and uncertainties.
Baseline demographic proximity analyses are relevant for identifying which populations
may be exposed to near-source pollutants, such as NO2 emitted from affected sources in this
proposed rule, however such analyses do not account for the potential impacts from this
proposed rulemaking from long-range ozone concentration decreases. Baseline demographic
proximity analyses can also provide information as to whether there may be potential EJ
concerns associated with environmental stressors affected by the regulatory action for certain
population groups of concern in the baseline. The baseline demographic proximity analysis finds
larger percentages of Hispanic individuals, Black individuals, people below the poverty level,
people with less educational attainment, and people linguistically isolated living within 5 km and
10 km of an affected EGU, compared to national averages. It also finds larger percentages of
people below the poverty level and with less educational attainment living within 5 km and 10
7-30
-------
km of an affected non-EGU. Separately, the tribal proximity analysis finds multiple tribes and
unique tribal lands located within 50 miles of an affected facility. These results cannot be used to
demonstrate disproportionate impacts of affected facilities in the baseline but could suggest that
population groups of concern in the baseline may be disproportionately impacted by any
potential local environmental stressors affected by the regulatory action.
While the demographic proximity analyses may appear to parallel the baseline analysis of
nationwide AS-M03 ozone concentrations in certain ways, the two should not be directly
compared. This is because the demographic proximity analysis does not include information on
baseline or policy-specific ozone concentration information. The AS-M03 ozone concentration
assessment is in effect an analysis of total ozone burden in the contiguous U.S. in 2023 and 2026,
including various assumptions such as the implementation of promulgated regulations. It serves
as a starting point for both the estimated ozone changes due to this proposal as well as a snapshot
of AS-M03 ozone concentrations in the near future.
The baseline analysis of AS-M03 ozone concentrations responds to question 1 from
EPA's EJ Technical Guidance document more directly than the proximity analyses, as it
evaluates a form of the environmental stressor primarily affected by the regulatory action.
Baseline AS-M03 analyses show that certain populations, such as American Indians, Hispanics,
and Asians, may experience disproportionately higher AS-M03 concentrations compared to the
national average. The less educated and children may also experience higher concentrations
compared to the national average, but to a lesser extent. Conversely, Black populations may
experience lower AS-M03 concentrations than the national average. Therefore, there likely are
potential EJ concerns associated with environmental stressors affected by the regulatory action
for population groups of concern in the baseline.
The third type of EJ analysis presented here evaluates how regulatory alternatives of this
proposal are expected to differentially impact demographic populations, informing questions 2
and 3 from EPA's EJ Technical Guidance with regard to AS-M03 exposure changes. Overall,
AS-M03 concentrations under the proposal, more stringent, and less stringent alternatives are
predicted to impact demographic groups very similarly in both future years and across both
EGUs and non-EGUs. While national-level results found slightly smaller AS-M03 ozone
concentration improvements for Hispanic, Asian, and Native American populations and greater
7-31
-------
but consistent AS-M03 ozone concentration improvements for Black populations, state-level
results showed this difference was highly variable across areas. Additionally, the magnitude of
these differences in air quality improvement is at or near the limit of uncertainty with regard to
our ability to distinguish meaningful health impacts as well as air quality modeling accuracy.
Therefore, regarding AS-M03 concentrations, there may be potential baseline EJ
concerns that will be affected by the regulatory action for certain population groups of concern
(question 1). However, we do not find evidence of meaningful EJ concerns associated with AS-
M03 concentrations after imposition of the proposed regulatory action or alternatives under
consideration (question 2). We also do not find evidence that any potential EJ concerns related to
AS-M03 would be meaningfully exacerbated in the regulatory alternatives under consideration,
compared to the baseline (question 3). Importantly, the action described in this proposal is
expected to lower ozone in many areas, including residual ozone nonattainment areas, and thus
mitigate some pre-existing health risks of ozone across all populations evaluated.
7.8 References
Bell, M.L. and Ebisu, K., (Bell 2012). Environmental inequality in exposures to airborne
particulate matter components in the United States. Environmental health perspectives,
120(12), pp.1699-1704.
Bravo, M.A., Anthopolos, R., Bell, M.L. and Miranda, M.L., (Bravo 2016). Racial isolation and
exposure to airborne particulate matter and ozone in understudied US populations:
Environmental justice applications of downscaled numerical model output. Environment
international, 92, pp.247-255.
Kelly, J.T., Jang, C., Timin, B., Di, Q., Schwartz, J., Liu, Y., van Donkelaar, A., Martin, R.V.,
Berrocal, V. and Bell, M.L (Kelly 2021). Examining PM2.5 concentrations and exposure
using multiple models. Environmental Research, 196, 110432.
U.S. Environmental Protection Agency (U.S. EPA 2019). Integrated Science Assessment (ISA)
for Particulate Matter (Final Report). U.S. EPA. Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Research and Development, Center for
Public Health and Environmental Assessment Available at:
https://www.epa.gov/naaqs/particulate-matter-pm-standards-integrated-science-
assessments-current-review
U.S. Environmental Protection Agency (U.S. EPA 2020). Integrated Science Assessment (ISA)
for Ozone and Related Photochemical Oxidants (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-20/012, 2020.
7-32
-------
U.S. Environmental Protection Agency (U.S. EPA 2021a). Regulatory Impact Analysis for the
Final Revised Cross-State Air Pollution Rule (CSAPR) Update for the 2008 Ozone
National Ambient Air Quality Standards (NAAQS). U.S. Environmental Protection
Agency, Research Triangle Park, NC, EPA-452/R-21-002, March 2021.
U.S. Environmental Protection Agency (U.S. EPA 2021b). Supplement to the 2019 Integrated
Science Assessment for Particulate Matter (External Review Draft). Center for Public
Health and Environmental Assessment. Office of Research and Development, U.S.
Environmental Protection Agency, Research Triangle Park, NC. EPA/600/R-21/198,
September 2021. Available at:
http s: //cfpub. epa. gov/ncea/i sa/ recordi spl ay. cfm? dei d=3 52823
U.S. Environmental Protection Agency (U.S. EPA 2021c). Draft Policy Assessment for the
Review of the Ozone National Ambient Air Quality Standards. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Health and
Environmental Impacts Division. Research Triangle Park, NC. U.S. EPA. EPA-452/P-21-
001. October 2021. Available at: https://www.epa.gov/system/files/documents/2021-
12/draft-policy-assessment-for-the-reconsideration-of-the-pm-naaqs_october-
2021_ed3.pdf
U.S. Environmental Protection Agency (U.S. EPA 2021d). Technical Support Document (TSD)
for the Final Revised Cross-State Air Pollution Rule Update for the 2008 Ozone Season
NAAQS. Estimating PM2.5- and Ozone-Attributable Health Benefits. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Health and
Environmental Impacts Division. Research Triangle Park, NC. U.S. EPA. EPA-HQ-
OAR-2020-0272. March 2021. Available at:
https://www.epa.gov/sites/default/files/2021-03/documents/estimating_pm2.5-
_and_ozone-attributable_health_benefits_tsd.pdf
Woods & Poole (2015). Complete Demographic Database.
7-33
-------
CHAPTER 8: COMPARISON OF BENEFITS AND COSTS
Overview
EPA performed an analysis to estimate the costs and benefits of compliance with the
proposed Federal Implementation Plan (FIP) Addressing Regional Ozone Transport for the 2015
Ozone National Ambient Air Quality Standards (FIP for the 2015 ozone NAAQS) and more and
less stringent alternatives. EPA is proposing to promulgate new or revised FIPs for 25 states that
include new NOx ozone season emission budgets for electric generating unit (EGU) sources,
with implementation of these emission budgets beginning in the 2023 ozone season. EPA is also
proposing to adjust these states' emission budgets for each ozone season thereafter to maintain
the initial stringency of the emissions budget, accounting for retirements and other changes to the
fleet over time. EPA is also proposing to extend the CSAPR NOx Ozone Season Group 3
Trading Program beginning in the 2023 ozone season through the 2025 ozone season. EPA is
proposing to establish new emissions budgets for the CSAPR NOx Ozone Season Group 3
Trading Program beginning in the 2026 ozone season, as discussed in Section VII.B.l. of the
preamble.
EPA is also proposing to promulgate new FIPs for 23 states that include new NOx
emissions limitations for non-electric generating unit (non-EGU) sources, with initial compliance
dates for these emissions limitations beginning in 2026.
For the RIA, in order to implement the OMB Circular A-4 requirement for fulfilling
Executive Order (E.O.) 12866 to assess one less stringent and one more stringent alternative to
the proposed rule, for the EGUs, all three alternatives use emission budgets that were developed
using uniform control stringency represented by $1,800 per ton of NOx (2016$) in 2023 and
$11,000 per ton of NOx (2016$) in 2026. The less stringent alternative imposes backstop
emission rate limits in the 2028 run year (reflective of imposition in the 2027 calendar year),
while the proposed rule and more stringent alternative impose backstop emission rate limits in
the 2025 run year (reflective of imposition in the 2026 calendar year) that force uncontrolled
units to either install NOx controls or retire. For the proposed rule and more stringent alternative,
backstop emission rate limits are imposed on all coal units within the 23-state region that are
greater than 100 MW and lack SCR controls. Emission rate limits are also imposed on all oil/gas
8-1
-------
steam units within the linked states that are greater than 100 MW and lack SCR controls that
operated at a greater than 20 percent historical capacity factor. In addition to the backstop rate
limits present in the proposed rule and the less stringent alternative, the more stringent
alternative also imposes backstop emission rate limits on all oil/gas steam units in the affected
states that are greater than 100 MW, lack SCR controls and have operated at below a 20 percent
capacity factor historically.
The proposal also includes NOx emissions limitations with an initial compliance date of
2026 applicable to certain non-EGU stationary sources in 23 states. The proposed rule
establishes NOx emissions limitations during the ozone season for the following unit types:
reciprocating internal combustion engines in Pipeline Transportation of Natural Gas; kilns in
Cement and Cement Product Manufacturing; boilers and furnaces in Iron and Steel Mills and
Ferroalloy Manufacturing; furnaces in Glass and Glass Product Manufacturing; and impactful
boilers in Basic Chemical Manufacturing, Petroleum and Coal Products Manufacturing, and
Pulp, Paper, and Paperboard Mills. In order to implement the OMB Circular A-4 requirement for
fulfilling Executive Order (E.O.) 12866 to assess one less stringent and one more stringent
alternative to the proposed rule, we analyzed a less stringent non-EGU alternative that assumes
there are emissions limits for all emission units from the proposed rule alternative except for
boilers in Basic Chemical Manufacturing, Petroleum and Coal Products Manufacturing, and
Pulp, Paper, and Paperboard Mills. We analyzed a more stringent non-EGU alternative that
assumes emissions limits for all emission units from the proposed rule alternative and all boilers,
not just impactful boilers, in Basic Chemical Manufacturing, Petroleum and Coal Products
Manufacturing, and Pulp, Paper, and Paperboard Mills. A summary of the proposed emissions
limits can be found in Section I.B. of the preamble.
8.1 Results
This RIA evaluates how EGUs and non-EGUs covered by the proposed rule are expected
to reduce their emissions in response to the requirements and flexibilities provided by the remedy
implemented by the proposed FIP for the 2015 ozone NAAQS and the benefits, costs and
impacts of their expected compliance behavior. This chapter summarizes these results. Table 8-1
shows the ozone season NOx emissions reductions expected from the proposed rule as well as
the more and less stringent alternatives analyzed from 2023 through 2030, and for 2035 and
8-2
-------
2042. In addition, Table 8-1 shows the annual NOx, SO2, PM2.5, and CO2 emissions reductions
expected from the proposed rule as well as the more and less stringent alternatives analyzed from
2023 through 2030, and for 2035 and 2042. Table 8-2 below provides a summary of the 2019
ozone season emissions for non-EGUs for the 23 states subject to the proposed FIP in 2026,
along with the estimated ozone season reductions for the proposal and the less and more
stringent alternatives.
Table 8-1. EGU Ozone Season NOx Emissions Changes and Annual Emissions Changes
for NOx, SO2, PM2.5, and CO2 for the Regulatory Control Alternatives from 2023 - 2042
Proposed Rule
Less Stringent
Alternative
More Stringent
Alternative
2023
NOx (ozone season)
6,000
6,000
7,000
NOx (annual)
10,000
10,000
10,000
SO2 (annual)*
--
1,000
2,000
CO2 (annual, thousand metric)
--
--
--
PM2.5 (annual)
2024
NOx (ozone season)
26,000
14,000
29,000
NOx (annual)
42,000
22,000
45,000
S02 (annual)
42,000
20,000
43,000
C02 (annual, thousand metric)
18,000
10,000
19,000
PM2.5 (annual)
4,000
1,000
4,000
2025
NOx (ozone season)
46,000
22,000
51,000
NOx (annual)
73,000
33,000
80,000
SO2 (annual)
83,000
39,000
84,000
CO2 (annual, thousand metric)
37,000
19,000
38,000
PM2.5 (annual)
9,000
2,000
9,000
2026
NOx (ozone season)
47,000
32,000
53,000
NOx (annual)
81,000
55,000
87,000
SO2 (annual)
106,000
76,000
108,000
CO2 (annual, thousand metric)
40,000
26,000
42,000
PM2.5 (annual)
9,000
5,000
9,000
2027
NOx (ozone season)
49,000
42,000
54,000
NOx (annual)
88,000
76,000
95,000
SO2 (annual)
129,000
113,000
131,000
CO2 (annual, thousand metric)
43,000
34,000
46,000
PM2.5 (annual)
10,000
7,000
10,000
2030
NOx (ozone season)
52,000
52,000
57,000
NOx (annual)
96,000
98,000
100,000
S02 (annual)
104,000
100,000
103,000
C02 (annual, thousand metric)
50,000
45,000
50,000
8-3
-------
Proposed Rule
Less Stringent
Alternative
More Stringent
Alternative
PM2.5 (annual)
9,000
9,000
9,000
2035
NOx (ozone season)
49,000
50,000
52,000
NOx (annual)
90,000
93,000
93,000
SO2 (annual)
96,000
93,000
98,000
CO2 (annual, thousand metric)
38,000
36,000
38,000
PM2.5 (annual)
11,000
12,000
10,000
2042
NOx (ozone season)
47,000
47,000
48,000
NOx (annual)
70,000
75,000
71,000
SO2 (annual)
54,000
50,000
54,000
CO2 (annual, thousand metric)
25,000
23,000
24,000
PM2.5 (annual)
8,000
9,000
8,000
* SO2 emissions reductions under the proposed rule are 350 tons and rounded to zero. SO2 emissions
reductions under the less stringent alternative are 507 tons and rounded to 1000 tons. SO2 emissions
reductions are 1,699 tons under the more stringent alternative and rounded to 2,000 tons. Given the
rounding, the difference between the reductions under the proposed rule and the less stringent alternative
is approximately 160 tons.
8-4
-------
Table 8-2. Non-EGU Ozone Season (OS) NOx Emissions and Emissions Reductions for the
Proposed Rule and the Less and More Stringent Alternatives
2019 OS NOx
Emissions
Proposed Rule -
Less Stringent
More Stringent
State
OS NOx
Alternative - OS
Alternative - OS
Reductions
NOx Reductions
NOx Reductions
AR
8,265
1,654
922
1,654
CA
14,579
1,666
1,598
1,777
IL
16,870
2,452
2,452
2,553
IN
19,604
3,175
2,787
3,175
KY
11,934
2,291
2,291
2,291
LA
35,831
6,769
4,121
6,955
MD
2,365
45
45
45
MI
18,996
2,731
2,731
3,093
MN
17,591
673
673
789
MO
9,109
3,103
3,103
3,103
MS
12,284
1,761
1,577
1,761
NJ
2,025
0
0
29
NV
2,418
0
0
0
NY
6,003
500
389
613
OH
19,729
2,790
2,611
2,814
OK
22,146
3,575
3,575
3,871
PA
15,861
3,284
3,132
3,340
TX
47,135
4,440
4,440
6,596
UT
6,276
757
757
757
VA
7,041
1,563
1,465
1,660
WI
6,571
2,150
677
2,234
WV
9,825
982
982
982
WY
10,335
826
826
826
Totals
322,793
47,186
41,153
50,918
As shown in Chapter 4, the estimated annual compliance costs to implement the proposed
rule, as described in this RIA, are approximately $-210 million in 2023 and $1,100 million in
2026 (2016$). Compliance costs are negative because in 2023 the EGU compliance costs are
negative. While seemingly counterintuitive, estimating negative compliance costs in a single
year is possible given IPM's objective function is to minimize the discounted net present value
(NPV) of a stream of annual total cost of generation over a multi-decadal time period. This
results in delayed retrofit and retirement at EGU facilities, which in turn leads to negative total
cost point estimates in 2023.
This RIA uses compliance costs as a proxy for social costs as mentioned in Chapter 4. As
shown in Chapter 5, the estimated monetized benefits from reduced PM2.5 and ozone
concentrations from implementation of the proposed rule are approximately $100 and $500
million in 2023 (2016$, based on a real discount rate of 3 percent). For 2026, the estimated
8-5
-------
monetized benefits from implementation of the proposed rule are approximately $9,300 and
$18,000 million (2016$, based on a real discount rate of 3 percent).
EPA calculates the monetized net benefits of the proposal by subtracting the estimated
monetized compliance costs from the estimated monetized benefits in 2023, 2026, and 2030. The
benefits include those to public health associated with reductions in PM2.5 and ozone
concentrations. The annual monetized net benefits of the proposed rule in 2023 (in 2016$) are
approximately $310 and $710 million using a 3 percent real discount rate. The annual monetized
net benefits of the proposed rule in 2026 are approximately $8,200 and $17,000 million using a 3
percent real discount rate. The annual monetized net benefits of the rule in 2030 are
approximately $7,700 and $18,000 million using a 3 percent real discount rate. Table 8-3
presents a summary of the monetized benefits, costs, and net benefits of the proposed rule and
the more and less stringent alternatives for 2023. Table 8-4 presents a summary of these impacts
for the proposed rule and the more and less stringent alternatives for 2026.
Table 8-5 presents a summary of these impacts for the proposed rule and the more and less
stringent alternatives for 2030. These results present an incomplete overview of the effects of the
proposal, because important categories of benefits — including benefits from reducing climate
pollution, other types of air pollutants, and water pollution - were not monetized and are
therefore not reflected in the cost-benefit tables. We anticipate that taking non-monetized effects
into account would show the proposal to be more net beneficial than this table reflects.
Table 8-3. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less and
More Stringent Alternatives for 2023 for the U.S. (millions of 2016$) a,b
Proposed Rule
Less Stringent
Alternative
More Stringent
Alternative
Benefits0
$100 and $500
$120 and $520
$250 and $720
Costs'1
-$210
-$170
-$180
Net Benefits
$310 and $710
$290 and $690
$430 and $900
a We focus results to provide a snapshot of costs and benefits in 2023, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Rows may not appear to add correctly due to rounding.
0 Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates and are presented at a real discount
rate of 3 percent. Several categories of benefits remain unmonetized and are thus not reflected in the table. Non-
monetized benefits include important climate benefits from reductions in C02 emissions. The U.S. District Court for
the Western District of Louisiana has issued an injunction concerning the monetization of the benefits of greenhouse
gas emission reductions by EPA and other defendants. See Louisiana v. Biden, No. 21-cv-01074-JDC-KK (W.D.
La. Feb. 11, 2022). Therefore, such values are not presented in the benefit-cost analysis of this proposal conducted
8-6
-------
pursuant to E.O. 12866. Please see Chapter 5, Section 5.2 for more discussion. In addition, there are important
unqualified water quality benefits and benefits associated with reductions in other air pollutants.
d The costs presented in this table are 2023 annual estimates for each alternative analyzed. An NPV of costs was
calculated using a 3.76% real discount rate consistent with the rate used in IPM's objective function for cost-
minimization.
Table 8-4. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less and
More Stringent Alternatives for 2026 for the U.S. (millions of 2016$) a,b
Proposed Rule
Less Stringent
Alternative
More Stringent
Alternative
Benefits0
$9,300 and $18,000
$4,300 and $10,000
$9,100 and $19,000
Costs'1
$1,100
-$49
$1,600
Net Benefits
$8,200 and $17,000
$4,300 and $10,000
$7,500 and $17,000
a We focus results to provide a snapshot of costs and benefits in 2026, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Rows may not appear to add correctly due to rounding.
0 Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates and are presented at a real discount
rate of 3 percent. Several categories of benefits remain unmonetized and are thus not reflected in the table. Non-
monetized benefits include important climate benefits from reductions in CO2 emissions. The U.S. District Court for
the Western District of Louisiana has issued an injunction concerning the monetization of the benefits of greenhouse
gas emission reductions by EPA and other defendants. See Louisiana v. Biden, No. 21-cv-01074-JDC-KK (W.D.
La. Feb. 11, 2022). Therefore, such values are not presented in the benefit-cost analysis of this proposal conducted
pursuant to E.O. 12866. Please see Chapter 5, Section 5.2 for more discussion. In addition, there are important
unqualified water quality benefits and benefits associated with reductions in other air pollutants.
d The costs presented in this table are 2026 annual estimates for each alternative analyzed. An NPV of costs was
calculated using a 3.76% real discount rate consistent with the rate used in IPM's objective function for cost-
minimization.
Table 8-5. Monetized Benefits, Costs, and Net Benefits of the Proposed Rule and Less and
More Stringent Alternatives for 2030 for the U.S. (millions of 2016$) a,b
Proposed Rule
Less Stringent
Alternative
More Stringent
Alternative
Benefits0
$9,400 and $20,000
$4,300 and $11,000
$9,200 and $21,000
Costs'1
$1,600
$1,600
$2,200
Net Benefits
$7,700 and $18,000
$2,800 and $9,700
$7,000 and $19,000
a We focus results to provide a snapshot of costs and benefits in 2030, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Rows may not appear to add correctly due to rounding.
0 Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates and are presented at a real discount
rate of 3 percent. Several categories of benefits remain unmonetized and are thus not reflected in the table. Non-
monetized benefits include important climate benefits from reductions in C02 emissions. The U.S. District Court for
the Western District of Louisiana has issued an injunction concerning the monetization of the benefits of greenhouse
gas emission reductions by EPA and other defendants. See Louisiana v. Biden, No. 21-cv-01074-JDC-KK (W.D.
La. Feb. 11, 2022). Therefore, such values are not presented in the benefit-cost analysis of this proposal conducted
pursuant to E.O. 12866. Please see Chapter 5, Section 5.2 for more discussion. In addition, there are important
unqualified water quality benefits and benefits associated with reductions in other air pollutants.
d The costs presented in this table are 2030 annual estimates for each alternative analyzed. An NPV of costs was
calculated using a 3.76% real discount rate consistent with the rate used in IPM's objective function for cost-
minimization.
8-7
-------
As part of fulfilling analytical guidance with respect to E.O. 12866, EPA presents
estimates of the present value (PV) of the monetized benefits and costs over the twenty-year
period 2023 to 2042. To calculate the present value of the social net-benefits of the proposed
rule, annual benefits and costs are discounted to 2022 at 3 percent and 7 discount rates as
directed by OMB's Circular A-4. EPA also presents the equivalent annualized value (EAV),
which represents a flow of constant annual values that, had they occurred in each year from 2023
to 2042, would yield a sum equivalent to the PV. The EAV represents the value of a typical cost
or benefit for each year of the analysis, in contrast to the year-specific estimates mentioned
earlier in the RIA.
For the twenty-year period of 2023 to 2042, the PV of the net benefits, in 2016$ and
discounted to 2022, is $220,000 million when using a 3 percent discount rate and $130,000
million when using a 7 percent discount rate. The EAV is $15,000 million per year when using a
3 percent discount rate and $12,000 million when using a 7 percent discount rate. The
comparison of benefits and costs in PV and EAV terms for the proposed rule can be found in
Table 8-6. Estimates in the table are presented as rounded values.
-------
Table 8-6. Summary of Present Values and Equivalent Annualized Values for the 2023-
2042 Timeframe for Estimated Monetized Compliance Costs, Benefits, and Net Benefits for
the Proposed Rule (millions of 2016$, discounted to 2022)a'b
Benefits
Cost0
Net Benefits
3%
7%
3%
7%
3%
7%
2023
$500
$450
($210)
$710
$660
2024
$520
$460
$710
-$190
-$240
2025
$530
$470
$710
-$180
-$230
2026
$18,000
$16,000
$1,100
$17,000
$15,000
2027
$19,000
$17,000
$2,000
$17,000
$15,000
2028
$18,000
$16,000
$2,000
$16,000
$14,000
2029
$19,000
$17,000
$2,000
$17,000
$15,000
2030
$20,000
$18,000
$1,600
$18,000
$16,000
2031
$20,000
$18,000
$1,600
$19,000
$16,000
2032
$21,000
$18,000
$2,100
$18,000
$16,000
2033
$20,000
$18,000
$2,100
$18,000
$16,000
2034
$21,000
$18,000
$2,100
$19,000
$16,000
2035
$21,000
$19,000
$2,100
$19,000
$16,000
2036
$21,000
$19,000
$2,100
$19,000
$17,000
2037
$22,000
$19,000
$2,100
$19,000
$17,000
2038
$21,000
$19,000
$1,300
$20,000
$18,000
2039
$22,000
$19,000
$1,300
$20,000
$18,000
2040
$22,000
$19,000
$1,300
$21,000
$18,000
2041
$22,000
$19,000
$1,300
$21,000
$18,000
2042
$22,000
$20,000
$1,300
$21,000
$18,000
PV
$250,000
$150,000
$22,000
$14,000
$220,000
$130,000
2023-2042
EAV
$17,000
$14,000
$1,500
$1,300
$15,000
$12,000
2023-2042
aRows may not appear to add correctly due to rounding.
b The annualized present value of costs and benefits are calculated over a 20-year period from 2023 to 2042. The
benefits values use the larger of the two benefits estimates presented in Table ES-9 and Table ES-10, as well as for
all other years. Monetized benefits include those related to public health associated with reductions in PM2 5 and
ozone concentrations. The health benefits are associated with several point estimates and are presented at a real
discount rate of 3 percent. Several categories of benefits remain unmonetized and are thus not reflected in the table.
Non-monetized benefits include important climate benefits from reductions in CO2 emissions. The U.S. District
Court for the Western District of Louisiana has issued an injunction concerning the monetization of the benefits of
greenhouse gas emission reductions by EPA and other defendants. See Louisiana v. Biden, No. 21-cv-01074-JDC-
KK (W.D. La. Feb. 11, 2022). Therefore, such values are not presented in the benefit-cost analysis of this proposal
conducted pursuant to E.O. 12866. Please see Chapter 5, Section 5.2 for more discussion. In addition, there are
important unqualified water quality benefits and benefits associated with reductions in other air pollutants.
0 The costs presented in this table are consistent with the costs presented in Chapter 4. To estimate these annualized
costs, EPA uses a conventional and widely accepted approach that applies a capital recovery factor (CRF) multiplier
to capital investments and adds that to the annual incremental operating expenses. Costs were calculated using a
3.76% real discount rate consistent with the rate used in IPM's objective function for cost-minimization.
8-9
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-452/D-22-001
Environmental Protection Health and Environmental Impacts Division February 2022
Agency Research Triangle Park, NC
------- |