Regulatory Impact Analysis
for the
Clean Power Plan Final Rule
-------
[This page intentionally left blank]
11
-------
EPA-452/R-15-003
August 2015
Regulatory Impact Analysis for the Clean Power Plan Final Rule
U.S. Environmental Protection Agency
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
in
-------
CONTACT INFORMATION
This document has been prepared by staff from the Office of Air Quality Planning and
Standards, the Office of Atmospheric Programs, and the Office of Policy of the U.S.
Environmental Protection Agency. Questions related to this document should be addressed to
Alexander Macpherson, U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards, Research Triangle Park, North Carolina 27711 (email:
macpherson.alex@epa.gov).
ACKNOWLEDGEMENTS
Thank you to the many staff who worked on this document from EPA Offices including
the Office of Air Quality Planning and Standards, the Office of Atmospheric Programs, and the
Office of Policy. Contributions to this report were also made by ICE International and RTI
International.
IV
-------
Table of Contents
LIST OF TABLES x
LIST OF FIGURES xvii
ACRONYMS xx
EXECUTIVE SUMMARY ES-1
ES.l Background and Context ES-1
ES.2 Summary of Clean Power Plan Final Rule ES-1
ES.3 Illustrative Plan Approaches Examined in RIA ES-3
ES.4 Emissions Reductions ES-6
ES.5 Costs ES-8
ES.6 Monetized Climate Benefits and Health Co-benefits ES-10
ES.6.1 Estimating Global Climate Benefits ES-14
ES 6.2 Estimating Air Quality Health Co-Benefits ES-16
ES 6.3 Combined Benefits Estimates ES-19
ES.7 Net Benefits ES-21
ES.8 Economic Impacts ES-24
ES.9 Employment Impacts ES-24
ES.10 References ES-25
CHAPTER!: INTRODUCTION AND BACKGROUND FOR THE CLEAN POWER PLAN 1-1
1.1 Introduction 1-1
1.2 Legal, Scientific and Economic Basis for this Rulemaking 1-1
1.2.1 Statutory Requirement 1-1
1.2.2 Health and Welfare Impacts from Climate Change 1-2
1.2.3 Market Failure 1-3
1.3 Summary of Regulatory Analysis 1-4
1.4 Background for the Final Emission Guidelines 1-4
1.4.1 Base Case and Years of Analysis 1-4
1.4.2 Definition of Affected Sources 1-5
1.4.3 Regulated Pollutant 1-6
1.4.4 Emission Guidelines 1-6
1.4.5 State Plans 1-7
1.5 Organization of the Regulatory Impact Analysis 1-7
1.6 References 1-8
CHAPTER 2: ELECTRIC POWER SECTOR INDUSTRY PROFILE 2-1
2.1 Introduction 2-1
2.2 Power Sector Overview 2-1
2.2.1 Generation 2-1
2.2.2 Transmission 2-9
-------
2.2.3 Distribution 2-10
2.3 Sales, Expenses and Prices 2-11
2.3.1 Electricity Prices 2-11
2.3.2 Prices of Fossil Fuels Used for Generating Electricity 2-17
2.3.3 Changes in Electricity Intensity of the U.S. Economy Between 2002 to
2012 2-18
2.4 Deregulation and Restructuring 2-19
2.5 Emissions of Greenhouse Gases from Electric Utilities 2-24
2.6 Carbon Dioxide Control Technologies 2-27
2.6.1 Carbon Capture and Storage 2-29
2.6.2 Geologic and Geographic Considerations for Geologic Sequestration 2-33
2.6.3 Availability of Geologic Sequestration in Deep Saline Formations 2-37
2.6.4 Availability of COi Storage via Enhanced Oil Recovery (EOR) 2-37
2.7 State Policies on GHG and Clean Energy Regulation in the Power Sector 2-39
2.8 Revenues and Expenses 2-42
2.9 Natural Gas Market 2-43
2.10 References 2-47
CHAPTER 3: COST, EMISSIONS, ECONOMIC, AND ENERGY IMPACTS 3-1
3.1 Introduction 3-1
3.2 Overview 3-1
3.3 Power Sector Modelling Framework 3-1
3.4 Recent Updates to EPA's Base Case using IPM (v.5.15) 3-4
3.5 State Goals in this Final Rule 3-5
3.6 Illustrative Plan Approaches Analyzed 3-7
3.7 Demand-Side Energy Efficiency 3-12
3.7.1 Demand-Side Energy Efficiency Improvements (Electricity Demand
Reductions) 3-12
3.7.2 Demand-Side Energy Efficiency Costs 3-15
3.8 Monitoring, Reporting, and Recordkeeping Costs 3-16
3.9 Projected Power Sector Impacts 3-19
3.9.1 Projected Emissions 3-19
3.9.2 Projected Compliance Costs 3-21
3.9.3 Projected Compliance Actions for Emissions Reductions 3-23
3.9.4 Projected Generation Mix 3-25
3.9.5 Projected Incremental Retirements 3-30
3.9.6 Projected Capacity Additions 3-31
3.9.7 Projected Coal Production and Natural Gas Use for the Electric Power
Sector 3-33
3.9.8 Projected Fuel Price, Market, and Infrastructure Impacts 3-34
3.9.9 Projected Retail Electricity Prices 3-35
3.9.10 Projected Electricity Bill Impacts 3-40
3.11 Limitations of Analysis 3-43
3.12 Social Costs 3-45
3.13 References 3-48
VI
-------
APPENDIX 3A: ANALYSIS OF POTENTIAL UPSTREAM METHANE EMISSIONS CHANGES IN
NATURAL GAS SYSTEMS AND COAL MINING 3A-1
3A.1 General Approach 3A-2
3A.1.1 Analytical Scope 3A-2
3A.1.2 Coal Mining Source Description 3A-3
3A.1.3 Natural Gas Systems Source Description 3A-4
3A.1.4 Illustrative Plan Approaches Examined 3A-6
3A.1.5 Activity Drivers 3A-6
3A.2 Results 3A-7
3A.3 Uncertainties and Limitations 3A-8
3A.4 References 3A-9
CHAPTER 4: ESTIMATED CLIMATE BENEFITS AND HUMAN HEALTH CO-BENEFITS 4-1
4.1 Introduction 4-1
4.2 Estimated Climate Benefits from CO2 4-1
4.2.1 Climate Change Impacts 4-2
4.2.2 Social Cost of Carbon 4-3
4.3 Estimated Human Health Co-Benefits 4-11
4.3.1 Health Impact Assessment for PM2.5 and Ozone 4-13
4.3.2 Economic Valuation for Health Co-benefits 4-18
4.3.3 Benefit-per-ton Estimates for PM2.5 4-20
4.3.4 Benefit-per-ton Estimates for Ozone 4-21
4.3.5 Estimated Health Co-Benefits Results 4-22
4.3.6 Characterization of Uncertainty in the Estimated Health Co-benefits 4-36
4.4 Combined Climate Benefits and Health Co-Benefits Estimates 4-42
4.5 Unquantified Co-benefits 4-46
4.5.1 HAP Impacts 4-48
4.5.2 Additional NO2 Health Co-Benefits 4-52
4.5.3 Additional SO2 Health Co-Benefits 4-53
4.5.4 Additional NO2 and SO2 Welfare Co-Benefits 4-54
4.5.5 Ozone Welfare Co-Benefits 4-55
4.5.6 Carbon Monoxide Co-Benefits 4-55
4.5.7 Visibility Impairment Co-Benefits 4-56
4.6 References 4-56
APPENDIX 4A: GENERATING REGIONAL BENEFIT-PER-TON ESTIMATES 4A-1
4A.1 Overview of Benefit-per-Ton Estimates 4A-1
4A.2 Air Quality Modeling for the Proposed Clean Power Plan 4A-2
4A.3 Regional PM2.5 Benefit-per-Ton Estimates for EGUs Derived from Air
Quality Modeling of the Proposed Clean Power Plan 4A-5
4A.4 Regional Ozone Benefit-per-Ton Estimates 4A-15
4A.5 References 4A-18
CHAPTER 5: ECONOMIC IMPACTS - MARKETS OUTSIDE THE UTILITY POWER SECTOR 5-1
Vll
-------
5.1 Introduction 5-1
5.2 Methods 5-2
5.3 Summary of Secondary Market Impacts of Energy Price Changes 5-3
5.3.1 Share of Total Production Costs 5-5
5.3.2 Ability to Substitute between Inputs to the Production Process 5-5
5.3.3 Availability of Substitute Goods and Services 5-5
5.4 Effect of Changes in Input Demand from Electricity Sector 5-6
5.5 Conclusions 5-6
5.6 References 5-7
CHAPTER 6: EMPLOYMENT IMPACT ANALYSIS 6-1
6.1 Introduction 6-1
6.2 Economic Theory and Employment 6-2
6.3 Current State of Knowledge Based on the Peer-Reviewed Literature 6-6
6.3.1 Regulated Sector 6-7
6.3.2 Economy-Wide 6-9
6.3.3 Labor Supply Impacts 6-11
6.4 Recent Employment Trends 6-11
6.4.1 Electric Power Generation 6-12
6.4.2 Fossil Fuel Extraction 6-13
6.4.3 Clean Energy Employment Trends 6-14
6.5 Projected Sectoral Employment Changes due to the Final Emission Guidelines.... 6-18
6.5.1 Projected Changes in Employment in Electricity Generation and Fossil
Fuel Extraction 6-19
6.5.2 Projected Changes in Employment in Demand-Side Energy Efficiency
Activities 6-25
6.6 Conclusion 6-34
6.7 References 6-36
APPENDIX 6A: ESTIMATING SUPPLY SIDE EMPLOYMENT IMPACTS 6A-1
6A.1 General Approach 6A-1
6A.2 Employment Changes due to Heat Rate Improvements 6A-3
6A.2.1 Employment Changes Due to Building (or Avoiding) New Generation
Capacity 6A-5
6A.2.2 Employment Changes due to Coal and Oil/Gas Retirements 6A-8
6A.2.3 Employment Changes due to Changes in Fossil Fuel Extraction 6A-9
6A.3 References 6A-10
CHAPTER 7: STATUTORY AND EXECUTIVE ORDER ANALYSIS 7-1
7.1 Executive Order 12866: Regulatory Planning and Review, and Executive Order
13563: Improving Regulation and Regulatory Review 7-1
7.2 Paperwork Reduction Act (PRA) 7-6
7.3 Regulatory Flexibility Act (RFA) 7-7
7.4 Unfunded Mandates Reform Act (UMRA) 7-8
Vlll
-------
7.5 Executive Order 13132: Federalism 7-9
7.6 Executive Order 13175: Consultation and Coordination with Indian Tribal
Governments 7-14
7.7 Executive Order 13045: Protection of Children from Environmental Health Risks
and Safety Risks 7-16
7.8 Executive Order 13211: Actions Concerning Regulations That Significantly
Affect Energy Supply, Distribution, or Use 7-17
7.9 National Technology Transfer and Advancement Act (NTTAA) 7-17
7.10 Executive Order 12898: Federal Actions to Address Environmental Justice in
Minority Populations and Low-Income Populations 7-18
7.11 Congressional Review Act (CRA) 7-21
CHAPTER 8: COMPARISON OF BENEFITS AND COSTS 8-1
8.1 Comparison of Benefits and Costs 8-1
8.2 Uncertainty Analysis 8-5
8.2.1 Uncertainty in Costs and Illustrative Plan Approaches 8-5
8.2.2 Uncertainty Associated with Estimating the Social Cost of Carbon 8-6
8.2.3 Uncertainty Associated with PMi.5 and Ozone Health Co-Benefits
Assessment 8-7
8.3 References 8-9
IX
-------
LIST OF TABLES
Table ES-1. Emission Performance Rates (Adjusted Output-Weighted-Average Pounds
of CO2 Per Net MWh from All Affected Fossil Fuel-Fired EGUs) ES-2
Table ES-2. Climate and Air Pollutant Emission Reductions for the Rate-Based
Illustrative Plan Approach ES-6
Table ES-3. Climate and Air Pollutant Emission Reductions for the Mass-Based
Illustrative Plan Appproach ES-7
Table ES-4. Projected CO2 Emission Reductions, Relative to 2005 ES-8
Table ES-5. Compliance Costs for the Illustrative Rate-Based and Mass-Based Plan
Approaches ES-9
Table ES-6. Quantified and Unquantified Benefits ES-12
Table ES-7. Combined Estimates of Climate Benefits and Health Co-Benefits for Rate-
Based Approach (billions of 2011$) ES-18
Table ES-8. Combined Estimates of Climate Benefits and Health Co-benefits for Mass-
Based Approach (billions of 2011$) ES-21
Table ES-9. Monetized Benefits, Compliance Costs, and Net Benefits Under the Rate-
based Illustrative Plan Approach (billions of 2011$) ES-22
Table ES-10. Monetized Benefits, Compliance Costs, and Net Benefits under the Mass-
based Illustrative Plan Approach (billions of 2011$) ES-23
Table ES-11. Summary Table of Important Energy Market Impacts (Percent Change from
Base Case) ES-24
Table 2-1. Existing Electricity Generating Capacity by Energy Source, 2002 and 2012.... 2-3
Table 2-2. Net Generation in 2002 and 2013 (Trillion kWh = TWh) 2-5
Table 2-3. Coal and Natural Gas Generating Units, by Size, Age, Capacity, and
Thermal Efficiency (Heat Rate) 2-7
Table 2-4. Total U.S. Electric Power Industry Retail Sales in 2012 (billion kWh) 2-11
Table 2-5. Domestic Emissions of Greenhouse Gases, by Economic Sector (million
tons of CO2 equivalent) 2-25
Table 2-6. Greenhouse Gas Emissions from the Electricity Sector (Generation,
Transmission and Distribution), 2002 and 2012 (million tons of COi equivalent) 2-26
-------
Table 2-7. Fossil Fuel Emission Factors in EPA Base Case 5.14 IPM Power Sector
Modeling Application 2-27
Table 2-8. Total CO2 Storage Resource (DOE-NETL) 2-35
Table 2-9. Revenue and Expense Statistics for Major U.S. Investor-Owned Electric
Utilities for 2002, 2008 and 2012 (nominal $millions) 2-42
Table 3-1. Statewide COi Emission Performance Goals, Rate-based and Mass-based 3-6
Table 3-2. Demand-Side Energy Efficiency Plan Scenario: Net Cumulative Demand
Reductions [Contiguous U.S.] (GWh and as Percent of BAU Sales) 3-14
Table 3-3. Annualized Cost of Demand-Side Energy Efficiency Plan Scenario (at
discount rates of 3 percent and 7 percent, billions 2011$) 3-15
Table 3-4. Years 2020, 2025 and 2030: Summary of State and Industry Annual
Respondent Burden and Cost of Reporting and Recordkeeping Requirements
(2011$) 3-18
Table 3-5. Projected COi Emission Impacts, Relative to Base Case 3-19
Table 3-6. Projected COi Emission Impacts, Relative to 2005 3-20
Table 3-7. Projected Non-CO2 Emission Impacts, 2020-2030 3-20
Table 3-8. Annualized Compliance Costs Including Monitoring, Reporting and
Recordkeeping Costs Requirements (billions of 2011$) 3-22
Table 3-9. Total Power Sector Generating Costs (IPM) (billions 2011$) 3-23
Table 3-10. Projected Capacity Factor of Existing Coal Steam and Natural Gas
Combined Cycle Capacity 3-25
Table 3-11. Generation Mix (thousand GWh) 3-27
Table 3-12. Total Generation Capacity by 2020-2030 (GW) 3-31
Table 3-13. Projected Capacity Additions, Gas (GW) 3-32
Table 3-14. Projected Capacity Additions, Renewable (GW) 3-33
Table 3-15. Coal Production for the Electric Power Sector, 2025 3-33
Table 3-16. Power Sector Gas Use 3-34
Table 3-17. Projected Average Minemouth and Delivered Coal Prices (201 l$/MMBtu).. 3-35
XI
-------
Table 3-18. Projected Average Henry Hub (spot) and Delivered Natural Gas Prices
(2011$/MMBtu) 3-35
Table 3-19. 2020 Projected Contiguous U.S. and Regional Retail Electricity Prices
(cents/kWh) 3-37
Table 3-20. 2025 Projected Contiguous U.S. and Regional Retail Electricity Prices
(cents/kWh) 3-38
Table 3-21. 2030 Projected Contiguous U.S. and Regional Retail Electricity Prices
(cents/kWh) 3-39
Table 3-22. Projected Changes in Average Electricity Bills 3-40
Table 3A-1. Base Year Upstream Methane-Related Emissions in the
U.S. GHG Inventory 3A-6
Table 3A-2. Projected Coal Production Impacts 3A-7
Table 3A-3. Projected Natural Gas Production Impacts 3A-7
Table 3A-4. Potential Upstream Emissions Changes 3A-8
Table 4-1. Climate Effects 4-2
Table 4-2. Social Cost of CO2, 2015-2050 (in 2011$ per short ton) 4-8
Table 4-3. Estimated Global Climate Benefits of COi Reductions for the Final
Emission Guidelines in 2020 (billions of 2011$) 4-9
Table 4-4. Estimated Global Climate Benefits of CO2 Reductions for the Final
Emission Guidelines in 2025 (billions of 2011$) 4-9
Table 4-5. Estimated Global Climate Benefits of COi Reductions for the Final
Emission Guidelines in 2030 (billions of 2011$) 4-9
Table 4-6. Human Health Effects of Ambient PMi.5 and Ozone 4-14
Table 4-7. Summary of Regional PlVb.5 Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2020 (2011$) 4-23
Table 4-8. Summary of Regional PMi.5 Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2025 (2011$) 4-23
Table 4-9. Summary of Regional PM2.5 Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2030 (2011$) 4-24
Table 4-10. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Rate-based Illustrative Plan Approach in 2020 (thousands of short tons) 4-24
xn
-------
Table 4-11. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Rate-based Illustrative Plan Approach in 2025 (thousands of short tons) 4-24
Table 4-12. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Rate-based Illustrative Plan Approach in 2030 (thousands of short tons) 4-25
Table 4-13. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Mass-based Illustrative Plan Approach in 2020 (thousands of short tons) 4-25
Table 4-14. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Mass-based Illustrative Plan Approach in 2025 (thousands of short tons) 4-25
Table 4-15. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Mass-based Illustrative Plan Approach in 2030 (thousands of short tons) 4-25
Table 4-16. Summary of Estimated Monetized Health Co-Benefits for the Final
Emission Guidelines Rate-based Illustrative Plan Approach in 2020 (billions of
2011$) 4-26
Table 4-17. Summary of Estimated Monetized Health Co-Benefits for the Final
Emission Guidelines Rate-based Illustrative Plan Approach in 2025 (billions of
2011$) 4-26
Table 4-18. Summary of Estimated Monetized Health Co-Benefits for the Final
Emission Guidelines Rate-based Illustrative Plan Approach in 2030 (billions of
2011$) 4-27
Table 4-19. Summary of Estimated Monetized Health Co-Benefits for the Final
Emission Guidelines Mass-based Illustrative Plan Approach in 2020 (billions of
2011$) 4-27
Table 4-20. Summary of Estimated Monetized Health Co-Benefits for the Final
Emission Guidelines Mass-based Illustrative Plan Approach in 2025 (billions of
2011$) 4-28
Table 4-21. Summary of Estimated Monetized Health Co-Benefits for the Final
Emission Guidelines Mass-based Illustrative Plan Approach in 2030 (billions of
2011$) 4-28
Table 4-22. Summary of Avoided Health Incidences from PMi.s-Related and Ozone-
Related Co-benefits for the Final Emission Guidelines Rate-based Illustrative Plan
Approach in 2020 4-29
Table 4-23. Summary of Avoided Health Incidences from PMi.s-Related and Ozone-
Related Co-benefits for Final Emission Guidelines Rate-based Illustrative Plan
Approach in 2025 4-30
Xlll
-------
Table 4-24. Summary of Avoided Health Incidences from PMi.s-Related and Ozone-
Related Co-Benefits for Final Emission Guidelines Rate-based Illustrative Plan
Approach in 2030 4-31
Table 4-25. Summary of Avoided Health Incidences from PMi.5-Related and Ozone-
Related Co-benefits for the Final Emission Guidelines Mass-based Illustrative Plan
Approach in 2020 4-32
Table 4-26. Summary of Avoided Health Incidences from PM2.5-Related and Ozone-
Related Co-benefits for Final Emission Guidelines Mass-based Illustrative Plan
Approach in 2025 4-33
Table 4-27. Summary of Avoided Health Incidences from PMi.s-Related and Ozone-
Related Co-Benefits for Final Emission Guidelines Mass-based Illustrative Plan
Approach in 2030 4-34
Table 4-28. Population Exposure in the Clean Power Plan Proposal Option 1 State
Scenario Modeling (used to generate the benefit-per-ton estimates) Above and
Below Various Concentrations Benchmarks in the Underlying Epidemiology Studies 4-41
Table 4-29. Combined Climate Benefits and Health Co-Benefits for Final Emission
Guidelines in 2020 (billions of 2011$) 4-44
Table 4-30. Combined Climate Benefits and Health Co-Benefits for Final Emission
Guidelines in 2025 (billions of 2011$) 4-44
Table 4-31. Combined Climate Benefits and Health Co-Benefits for Final Emission
Guidelines in 2030 (billions of 2011$) 4-45
Table 4-32. Unquantified Health and Welfare Co-benefits Categories 4-47
Table 4A-1. State Total Annual EGU Emissions for NOx for the 2011 Base Year, 2025
Base Case, and 2025 Clean Power Plan Proposal (Option 1 State) (in thousands of
tons) 4A-3
Table 4A-2. State Total Annual EGU Emissions for SOi for the 2011 Base Year, 2025
Base Case, and 2025 Clean Power Plan Proposal (Option 1 State) (in thousands of
tons) 4A-4
Table 4A-3. Summary of Regional PM2.5 Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2020 (2011$) 4A-10
Table 4A-4. Summary of Regional PMi.5 Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2025 (2011$) 4A-10
Table 4A-5. Summary of Regional PM2.5 Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2030 (2011$) 4A-11
xiv
-------
Table 4A-6. Summary of Regional PlVb.5 Incidence-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2020 4A-12
Table 4A-7. Summary of Regional PMi.5 Incidence-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2025 4A-13
Table 4A-8. Summary of Regional PM2.5 Incidence-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2030 4A-14
Table 4A-9. Summary of Regional Ozone Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2020 (2011$) 4A-16
Table 4A-10. Summary of Regional Ozone Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2025 (2011$) 4A-16
Table 4A-11. Summary of Regional Ozone Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2030 (2011$) 4A-16
Table 4A-12. Summary of Regional Ozone Incidence-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2020 4A-17
Table 4A-13. Summary of Regional Ozone Incidence-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2025 4A-17
Table 4A-14. Summary of Regional Ozone Incidence-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2030 4A-17
Table 5-1. Estimated Percentage Changes in Average Energy Prices by Energy Type
for the Final Emission Guidelines, Rate-based and Mass-based Illustrative Plan
Approaches 5-4
Table 6-1. U. S. Green Goods and Services (GGS) Employment (annual average) 6-16
Table 6-2. Renewable Electricity Generation-Related Employment 6-17
Table 6-3. Energy and Resources Efficiency-Related Employment 6-18
Table 6-4. Engineering-Based51 Changes in Labor Utilization, Rate-based Scenario
(Number of Job-Years'5 of Employment in a Single Year) 6-24
Table 6-5. Engineering-Based51 Changes in Labor Utilization, Mass-Based Illustrative
Plan Approach (Number of Job-Years of Employment in a Single Year) 6-25
Table 6-6. Estimated Demand-Side Energy Efficiency Employment Impacts: Target 1
percent Growth in Energy Efficiency 6-31
Table 6A-1. Labor Productivity Growth Rate due to Heat Rate Improvement 6A-5
Table 6A-2. Capital Charge Rate and Duration Assumptions 6A-6
xv
-------
Table 6A-3. Expenditure Breakdown due to New Generating Capacity 6A-6
Table 6A-4. Labor Productivity due to New Generating Capacity 6A-7
Table 6A-5. Average FOM Costs for Existing Coal and Oil and Gas Steam Capacity
($/kW, 2011$) 6A-8
Table 6A-6. Labor Productivity due to Fossil Fuel Extraction 6A-9
Table 7-1. Monetized Benefits, Compliance Costs, and Net Benefits Under the Rate-
based Illustrative Plan Approach (billions of 2011 $)a 7-4
Table 7-2. Monetized Benefits, Compliance Costs, and Net Benefits under the Mass-
based Illustrative Plan Approach (billions of 2011 $)a 7-5
xvi
-------
LIST OF FIGURES
Figure 2-1. New Build and Retired Capacity (MW) by Fuel Type, 2002-2012 2-4
Figure 2-2. Cumulative Distribution in 2010 of Coal and Natural Gas Electricity
Capacity and Generation, by Age 2-8
Figure 2-3. Fossil Fuel-Fired Electricity Generating Facilities, by Size 2-9
Figure 2-4. Average Retail Electricity Price by State (cents/kWh), 2011 2-13
Figure 2-5. Nominal National Average Electricity Prices for Three Major End-Use
Categories 2-14
Figure 2-6. Relative Increases in Nominal National Average Electricity Prices for Major
End-Use Categories, With Inflation Indices 2-15
Figure 2-7. Real National Average Electricity Prices (2011$) for Three Major End-Use
Categories 2-16
Figure 2-8. Relative Change in Real National Average Electricity Prices (2011$) for
Three Major End-Use Categories 2-17
Figure 2-9. Relative Real Prices of Fossil Fuels for Electricity Generation; Change in
National Average Real Price per MBtu Delivered to EGU 2-17
Figure 2-10. Relative Growth of Electricity Generation, Population and Real GDP Since
2002 2-18
Figure 2-11. Relative Change of Real GDP, Population and Electricity Generation
Intensity Since 2002 2-19
Figure 2-12. Status of State Electricity Industry Restructuring Activities 2-21
Figures 2-13 and 2-14. Capacity and Generation Mix by Ownership Type, 2002 & 2012.... 2-23
Figures 2-15 and 2-16. Generation Capacity Built and Retired between 2002 and 2012 by
Ownership Type 2-24
Figure 2-17. Domestic Emissions of Greenhouse Gases from Major Sectors, 2002 and
2013 (million tons of CO2 equivalent) 2-25
Figure 2-18. Marketable products from Syngas Generation 2-29
Figure 2-19. Post-Combustion COi Capture for a Pulverized Coal Power Plant 2-30
Figure 2-20. Pre-Combustion CO2 Capture for an IGCC Power Plant 2-31
xvn
-------
Figure 2-21. Geologic Sequestration in the Continental United States 2-34
Figure 2-22. Relative Change Nominal and Real (2011$) Prices of Natural Gas Delivered
to the Power Sector ($/MMBtu) 2-44
Figure 2-23. Relative Change in Real (2011$) Prices of Fossil Fuels Delivered to the
Power Sector ($/MMBtu) 2-45
Figure 3-1. Illustrative Regions for Demand-Side Energy Efficiency/Renewable Energy
Procurement Used in this Analysis 3-9
Figure 3-2 Generation Mix (thousand GWh) 3-28
Figure 3-3. Nationwide Generation: Historical (1990-2014) and Base Case Projections
(2020,2025,2030) 3-29
Figure 3-4. Nationwide Generation: Historical (1990-2014) and Rate-Based Illustrative
Plan Approach Projections (2020, 2025, 2030) 3-29
Figure 3-5. Nationwide Generation: Historical (1990-2014) and Mass-Based Illustrative
Plan Approach Projections (2020, 2025, 2030) 3-30
Figure 3-6. Electricity Market Module Regions 3-40
Figure 4-1. Monetized Health Co-benefits of Rate-based and Mass-based Illustrative
Plan Approaches for the Final Emission Guidelines in 2025 3-35
Figure 4-2. Breakdown of Monetized Health Co-benefits by Precursor Pollutant at a 3%
Discount Rate for Rate-based and Mass-based Illustrative Plan Approaches for the
Final Emission Guidelines in 2025 3-36
Figure 4-3. Percentage of Adult Population (age 30+) by Annual Mean PMi.5 Exposure
in the Option 1 State Scenario Clean Power Plan Proposal Modeling (used to
generate the benefit-per-ton estimates) 3-41
Figure 4-4. Cumulative Distribution of Adult Population (age 30+) by Annual Mean
PMi.5 Exposure in the Option 1 State Scenario Clean Power Plan Proposal Modeling
(used to generate the benefit-per-ton estimates) 3-42
Figure 4-5. Breakdown of Combined Monetized Climate and Health Co-benefits of
Final Emission Guidelines in 2025 for Rate-based and Mass-based Illustrative Plan
Approaches and Pollutants (3% discount rate) 3-46
Figure 4A-1. Regional Breakdown 4A-7
Figure 6.1. Electric Power Industry Employment 6-12
Figure 6.2. Coal Production Employment 6-13
XVlll
-------
Figure 6.3. Oil and Gas Production Employment 6-14
Figure 6.4. Demand-Side Energy Efficiency Employment: Jobs per One Million Dollars
(2011$) 6-32
xix
-------
ACRONYMS
ACS American Cancer Society
AEO Annual Energy Outlook
AQ Air quality
ASM Annual Survey of Manufactures
ATSDR Agency for Toxic Substances and Disease Registry
BACT Best Available Control Technology
BenMAP Benefits Mapping and Analysis Program
BPT Benefit-per-Ton
BSER Best System of Emissions Reduction
Btu British Thermal Units
C Celsius
CAA Clean Air Act
CAIR Clean Air Interstate Rule
CCR Coal Combustion Residuals
CCS Carbon Capture and Sequestration or Carbon Capture and Storage
CCSP Climate Change Science Program
CFR Code of Federal Regulations
CH Methane
4
CO Carbon Monoxide
CO Carbon Dioxide
CRF Capital Recovery Factor
CSAPR Cross State Air Pollution Rule
CT Combustion Turbines
CUA Climate Uncertainty Adder
DICE Dynamic Integrated Climate and Economy Model
DOE U.S. Department of Energy
EAB Environmental Appeals Board
EC Elemental carbon
ECS Energy Cost Share
EG Emissions guidelines
EGR Enhanced Gas Recovery
EGU Electric Generating Unit
EIA U.S. Energy Information Administration
EMM Electricity Market Module
EO Executive Order
EOR Enhanced Oil Recovery
EPA U.S. Environmental Protection Agency
ER Enhanced Recovery
FERC Federal Energy Regulatory Commission
FGD Flue Gas Desulfurization
FOAK First of a Kind
FOM Fixed Operating and Maintenance
FR Federal Register
FRCC Florida Reliability Coordinating Council
FUND Framework for Uncertainty, Negotiation, and Distribution Model v
GDP Gross Domestic Product
GHG Greenhouse Gas
xx
-------
GS Geologic Sequestration
Gt Gigaton
H2S Hydrogen Sulfide
HAP Hazardous air pollutant
HC1 Hydrogen chloride
HFC Hydrofluorocarbons
HIA Health impact assessment
IARC International Agency for Research on Cancer
IAM Integrated Assessment Model
ICR Information Collection Request
IGCC Integrated Gasification Combined Cycle
IOU Investor Owned Utility
IPCC Intergovernmental Panel on Climate Change
IPM Integrated Planning Model
IRIS Integrated Risk Information System
IRP Integrated Resource Plan
ISA Integrated Science Assessment
kWh Kilowatt-hour
Ibs Pounds
LCOE Levelized Cost of Electricity
LML Lowest measured level
LNB Low NOx Burners
MATS Mercury and Air Toxics Standards
MEA Monoethanolamine
MECSA Manufacturing Energy Consumption Survey
MeHg Methylmercury
MGD Millions of Gallons per Day
mg/L Milligrams per Liter
MMBtu Million British Thermal Units
MW Megawatt
MWh Megawatt-hour
N O Nitrous Oxide
NAAQS National Ambient Air Quality Standards
NAICS North American Industry Classification System
NaOH Sodium Hydroxide
NATCARB National Carbon Sequestration Database and Geographic Information System
NEEDS National Electric Energy Data System
NEMS National Energy Modeling System
NERC North American Electric Reliability Corporation
NETL National Energy Technology Laboratory
NGCC Natural Gas Combined Cycle
NMMAPS National Morbidity, Mortality Air Pollution Study
NOAK Next of a Kind or Nth of a Kind
NO Nitrogen Oxide
A
NRC National Research Council
NSPS New Source Performance Standard
NSR New Source Review
NTTAA National Technology Transfer and Advancement Act
OC Organic carbon
xxi
-------
OFA Overfire Air
OMB Office of Management and Budget
PAGE Policy Analysis of the Greenhouse Gas Effect Model
PEC Perfluorocarbons
PM Fine Particulate Matter
2.5
ppm Parts per Million
PRA Paperwork Reduction Act
PSD Prevention of Significant Deterioration
RCSP Regional Carbon Sequestration Partnerships
RADS Relative Airways Dysfunction Syndrome
RES Renewable Electricity Standards
RFA Regulatory Flexibility Act
RGGI Regional Greenhouse Gas Initiative
RIA Regulatory Impact Analysis
RPS Renewable Portfolio Standards
SAB-CASAC Science Advisory Board Clean Air Scientific Advisory Committee
SAB-HES Science Advisory Board Health Effects Subcommittee of the Advisory Council on
Clean Air Compliance
SAB-EEAC Science Advisory Board Environmental Economics Advisory Committee
SBA Small Business Administration
SBREFA Small Business Regulatory Enforcement Fairness Act
SCC Social Cost of Carbon
SCPC Super Critical Pulverized Coal
SCR Selective Catalytic Reduction
SF Sulfur Hexafluoride
6
SIP State Implementation Plan
SO2 Sulfur Dioxide
Tcf Trillion Cubic Feet
IDS Total Dissolved Solids
TSD Technical Support Document
TSM Transportation Storage and Monitoring
UMRA Unfunded Mandates Reform Act
U.S.C. U.S. Code
USGCRP U.S. Global Change Research Program
USGS U.S. Geological Survey
USG SCC U.S. Government's Social Cost of Carbon
U.S. NRC U.S. Nuclear Regulatory Commission
VCS Voluntary Consensus Standards
VOC Volatile Organic Compounds
VOM Variable Operating and Maintenance
VSL Value of a statistical life
WTP Willingness to pay
xxn
-------
EXECUTIVE SUMMARY
This Regulatory Impact Analysis (RIA) discusses potential benefits, costs, and economic
impacts of the Final Carbon Pollution Emission Guidelines for Existing Stationary Sources:
Electric Utility Generating Units (herein referred to as "final emission guidelines" or the "Clean
Power Plan Final Rule").
ES.l Background and Context
The emission of greenhouse gases (GHGs) threatens Americans' health and welfare by
leading to long-lasting changes in our climate. Carbon dioxide (CCh) is the primary greenhouse
gas pollutant, accounting for roughly three-quarters of global greenhouse gas emissions in 2010
and 82 percent of U.S. greenhouse gas emissions in 2013. Fossil fuel-fired electric generating
units (EGUs) are by far the largest emitters of GHGs, primarily in the form of CO2, among
stationary sources in the U.S.
In this action, the Environmental Protection Agency (EPA) is establishing final emission
guidelines for states to follow in developing plans to reduce greenhouse gas emissions from
existing fossil fuel-fired EGUs. Specifically, the EPA is establishing: 1) COi emission
performance rates representing the best system of emission reduction (BSER) for two
subcategories of existing fossil fuel-fired EGUs - fossil fuel-fired electric utility steam
generating units and stationary combustion turbines, 2) state-specific CO2 goals reflecting the
COi emission performance rates, and 3) guidelines for the development, submittal and
implementation of state plans that establish emission standards or other measures to implement
the CO2 emission performance rates, which may be accomplished by meeting the state goals.
This final rule will continue progress already underway in the U.S. to reduce COi emissions
from the utility power sector.
ES.2 Summary of Clean Power Plan Final Rule
Under CAA section 11 l(d), states must establish standards of performance that reflect the
degree of emission limitation achievable through the application of the "best system of emission
reduction" (BSER) that, taking into account the cost of achieving such reduction and any non-air
quality health and environmental impacts and energy requirements, the Administrator determines
has been adequately demonstrated. The EPA has determined that the BSER is the combination of
ES-1
-------
emission rate improvements and limitations on overall emissions at affected EGUs that can be
accomplished through any combination of one or more measures from the following three sets of
measures or building blocks:
1. Improving heat rate at affected coal-fired steam EGUs.
2. Substituting increased generation from lower-emitting existing natural gas combined
cycle units for reduced generation from higher-emitting affected steam generating
units.
3. Substituting increased generation from new zero-emitting generating capacity for
reduced generation from affected fossil fuel-fired generating units.
Specifically, the EPA is establishing COi emission performance rates for two
subcategories of existing fossil fuel-fired EGUs, fossil fuel-fired electric steam generating units
and stationary combustion turbines. The rates are intended to represent COi emission rates
achievable by 2030 after a 2022-2029 interim period on an output-weighted-average basis
collectively by all affected EGUs. The interim and final emission performance rates are
presented in the following table:
Table ES-1. Emission Performance Rates (Adjusted Output-Weighted-Average Pounds of
CCh Per Net MWh from All Affected Fossil Fuel-Fired EGUs)
Subcategory Interim Rate Final Rate
Fossil Fuel-Fired Electric Steam Generating Units 1,534 1,305
Stationary Combustion Turbines 832 771
Also, states with one or more affected EGUs will be required to develop and implement
plans that set emission standards for affected EGU. These emission standards may incorporate
the subcategory-specific CO2 emission performance rates set by the EPA or, in the alternative,
may be set at levels that ensure that the state's affected EGUs, individually, in aggregate, or in
combination with other measures undertaken by the state achieve the equivalent of the interim
and final CO2 emission performance rates between 2022 and 2029 and by 2030, respectively.
EPA derived statewide rate-based CO2 emissions performance goals as a weighted
average of the uniform rate goals with weights based on baseline generation for the two types of
units (fossil steam and stationary combustion turbine) in the state. This blended rate reflects the
ES-2
-------
collective emission rate a state may expect to achieve when its baseline fleet of likely affected
EGUs continues to operate at baseline levels while meeting its subcategory-specific emission
performance rates reflecting the BSER.
The Clean Power Plan Final Rule also establishes an 8-year interim compliance period
that begins in 2022 with a glide path for meeting interim COi emission performance rates
separated into three steps: 2022-2024, 2025-2027, and 2028-2029. This results in interim and
final statewide goal values unique to each state's historical blend of fossil steam and NGCC
generation. Chapter 3 presents finalized state rate-based CO2 emissions performance goals.
The EPA is also establishing mass-based statewide COi emission performance goals for
each state, which are also presented in Chapter 3. For more detail on the methodology that
translates CO2 emission performance rates to mass-based COi performance goes, please refer to
the preamble of the Clean Power Plan Final Rule and the U.S. EPA's CCh Emission Performance
Rate and Goal Computation Technical Support Document for Final Rule, which is available in
the docket.
Given the flexibilities afforded states in complying with the emission guidelines, the
benefits, cost and economic impacts reported in this RIA are not definitive estimates. Rather, the
impact estimates are instead illustrative of approaches that states may take.
ES.3 Illustrative Plan Approaches Examined in RIA
In the final emission guidelines, the EPA has translated the source category-specific COi
emission performance rates into state-level rate-based and mass-based COi goals in order to
maximize the range of choices that states will have in developing their plans. Because of the
range of choices available to states and the lack of a priori knowledge about the specific choices
states will make in response to the final goals, this RIA presents two scenarios designed to
achieve these goals, which we term the "rate-based" illustrative plan approach and the "mass-
based" illustrative plan approach.
In this final rule, states may use trading or other multi-unit compliance approaches and
technologies or strategies that are not explicitly mentioned in any of the three building blocks as
1 U.S. EPA. 2015. Technical Support Document (TSD) the Final Carbon Pollution Emission Guidelines for Existing
Stationary Sources: Electric Utility Generating Units. COi Emission Performance Rate and Goal Computation.
ES-3
-------
part of their overall plans, as long as they achieve the required emission reductions from affected
fossil fuel-fired EGUs. In addition, the final rule provides additional options to allow individual
EGUs to use creditable out-of-state reductions to achieve required CO2 reductions, without the
need for up-front interstate agreements.
The modelled implementation plan approaches reflect states and affected EGUs pursuing
building block strategies such as heat rate improvements, shifting generation to less CCh -
intensive generation, and increased deployment of renewable energy, which are more completely
described in Chapter 3. However, the modelled strategies are not limited to the technologies and
measures included in the BSER. While the final rule no longer includes demand-side energy
efficiency potential as part of BSER, the rule does allow such potential to be used for
compliance. These scenarios include a representation of demand-side energy efficiency
compliance potential because energy efficiency is a highly cost-effective means for reducing
COi from the power sector, and it is reasonable to assume that a regulatory requirement to
reduce COi emissions will motivate parties to pursue all highly cost-effective means for making
emission reductions accordingly, regardless of what particular emission reduction measures were
assumed in determining the level of that regulatory requirement. In the rate-based approach,
energy efficiency activities are modeled as being used by EGUs as a low-cost method of
demonstrating compliance with their rate-based emissions standards. In the mass-based
approach, energy efficiency activities are assumed to be adopted by states to lower demand,
which in turn reduces the cost of achieving the mass limitations.
Alternative compliance approaches other than those modelled are also possible, which
may have different levels and distributions of emissions and electricity generation as well as
costs. While IPM finds a least cost way to achieve the state goals implemented through the rate-
based or mass-based emissions constraints imposed in the illustrative plan approaches, individual
states or multi-state regional groups may develop alternate approaches to achieve their state
goals.
It is very important to note that the differences between the analytical results for the rate-
based and mass-based illustrative plan approaches presented in this RIA may not be indicative of
likely differences between the approaches if implemented by states and affected EGUs in
response to the final guidelines. Rather, the two sets of analyses are intended to illustrate two
ES-4
-------
contrasting, stylized implementation approaches to accomplish the emission performance rates
finalized in the Clean Power Plan Final Rule. In other words, if one approach performs
differently than the other on a given metric during a given time period, this does not imply this
will apply in all instances.
To present a complete picture of costs and benefits of the final emission guidelines, this
RIA presents results for the analysis years 2020, 2025, and 2030. While 2020 is before the first
year of the interim compliance period (2022), the EPA expects states and affected EGUs to
perform voluntary activities that will facilitate compliance with interim and final goals. These
pre-compliance period activities might include investments in renewable energy or demand-side
energy efficiency projects, for example, that produce emissions reductions in the compliance
period. Activities might also include preparatory investments in transmission capacity or
monitoring, reporting, and recordkeeping systems. As a result, there are likely to be benefits and
costs in 2020, so these are reported in the illustrative analysis of this RIA. Meanwhile, cost and
benefits are estimated in this RIA for 2025, which is intended to represent a central period of the
interim compliance time-frame as states and tribes are on glide paths toward fully meeting the
final CO2 emission performance goals. Lastly, the RIA presents costs and benefits for 2030,
when the emission performance goals are fully achieved.
ES-5
-------
ES.4 Emissions Reductions
Table ES-2 shows the emission reductions associated with the modelled rate-based
illustrative plan approach.
Table ES-2. Climate and Air Pollutant Emission Reductions for the Rate-Based
Illustrative Plan Approach1
2020 Rate-Based Approach
Base Case
Final Guidelines
Emissions Change
2025 Rate-Based Approach
Base Case
Final Guidelines
Emissions Change
2030 Rate-Based Approach
Base Case
Final Guidelines
Emission Change
C02
(million
short tons)
2,155
2,085
-69
2,165
1,933
-232
2,227
1,812
-415
S02
(thousand
short tons)
1,311
1,297
-14
1,275
1,097
-178
1,314
996
-318
Annual NOx
(thousand
short tons)
1,333
1,282
-50
1,302
1,138
-165
1,293
1,011
-282
Source: Integrated Planning Model, 2015. Emissions change may not sum due to rounding.
1 COi emission reductions are used to estimate the climate benefits of the guidelines. SOi, and NOx reductions are
relevant for estimating air quality health co-benefits of the final guidelines. The final guidelines are also expected to
achieve reductions in directly emitted PMi.s, which we were not able to estimate for this RIA.
In 2020, the EPA estimates that COi emissions will be reduced by 69 million short tons
under the rate-based scenario compared to base case levels. In 2025, the EPA estimates that COi
emissions will be reduced by 232 million short tons under the rate-based approach compared to
base case levels. COi emission reductions increase to 415 million short tons annually in 2030
when compared to the base case emissions. Table ES-2 also shows emission reductions for
criteria air pollutants (in short tons).2
2 The final guidelines are also expected to achieve reductions in directly emitted PMi.s, which we were not able to
estimate for this RIA. However, the SOi and NOx reductions account for the large majority of the anticipated health
co-benefits. Based on analyses for the proposed rule which included benefits from reductions in directly emitted
PM2.s, those benefits accounted for less than 10 percent of total monetized health co-benefits.
ES-6
-------
Table ES-3 shows the emission reductions associated with the modeled mass-based
illustrative plan approach.
Table ES-3. Climate and Air Pollutant Emission Reductions for the Mass-Based
Illustrative Plan Appproach1
2020 Mass-Based Approach
Base Case
Final Guidelines
Emissions Change
2025 Mass-Based Approach
Base Case
Final Guidelines
Emissions Change
2030 Mass-Based Approach
Base Case
Final Guidelines
Emission Change
CO2
(million
short tons)
2,155
2,073
-82
2,165
1,901
-264
2,227
1,814
-413
SO2
(thousand
short tons)
1,311
1,257
-54
1,275
1,090
-185
1,314
1,034
-280
Annual NOx
(thousand
short tons)
1,333
1,272
-60
1,302
1,100
-203
1,293
1,015
-278
Source: Integrated Planning Model, 2015. Emissions change may not sum due to rounding.
1 COi emission reductions are used to estimate the climate benefits of the guidelines. SOi, and NOx reductions are
relevant for estimating air quality health co-benefits of the final guidelines. The final guidelines are also expected to
achieve reductions in directly emitted PM2.s, which we were not able to estimate for this RIA.
In 2020, the EPA estimates that COi emissions will be reduced by 82 million short tons under
the mass-based approach compared to base case levels. In 2025, the EPA estimates that COi
emissions will be reduced by 264 million short tons under the mass-based approach compared to
base case levels. COi emission reductions increase to 413 million short tons annually in 2030
when compared to the base case emissions. Table ES-3 also shows emission reductions for
criteria air pollutants (in short tons).
ES-7
-------
Table ES-4 presents COi emission reductions relative to 2005.
Table ES-4. Projected CCh Emission Reductions, Relative to 2005
Base Case
Rate-based
Mass-based
CCh Emissions
(million short tons)
2005
2,683
-
CCh Emissions:
Change from 2005
(million short tons)
2020 2025 2030
-528 -518 -456
-598 -750 -871
-610 -782 -869
CCh Emissions Reductions:
Percent Change from 2005
2020 2025 2030
-20% -19% -17%
-22% -28% -32%
-23% -29% -32%
Source: Integrated Planning Model, 2015.
In 2020, the EPA estimates that COi emissions will be reduced by 598 million short tons (22
percent) under the rate-based approach compared to 2005 levels. In 2025, the EPA estimates that
COi emissions will be reduced by 750 million short tons (28 percent) under the rate-based
approach compared to 2005 levels. Under the rate-based approach, CO2 emission reductions
increase to 871 million short tons (32 percent) in 2030 when compared to 2005 levels.
Under the mass-based approach in 2020, the EPA estimates that CO2 emissions will be
reduced by 610 million short tons (23 percent) under the rate-based approach compared to 2005
levels. In 2025, the EPA estimates that COi emissions will be reduced by 782 million short tons
(29 percent) under the mass-based approach compared to 2005 levels. Under the mass-based
approach, COi emission reductions increase to 869 million short tons (32 percent) in 2030 when
compared to 2005 levels.
ES.5 Costs
The compliance cost estimates for this final action are represented in this analysis as the
change in electric power generation costs between the base case and illustrative plan approach
policy cases, including the cost of demand-side energy efficiency measures and costs associated
with monitoring, reporting, and recordkeeping requirements (MR&R). In the rate-based
approach, energy efficiency activities are modeled as being used by EGUs as a low-cost method
of demonstrating compliance with their rate-based emissions standards. In the mass-based
approach, energy efficiency activities are assumed to be adopted by states to lower demand,
which in turn reduces the cost of achieving the mass limitations. The level of energy efficiency
measures is determined outside of IPM and is assumed to be the same in the two illustrative plan
approaches. The compliance assumptions, and therefore the projected "compliance costs" set
ES-8
-------
forth in this analysis, are illustrative in nature and do not represent the full suite of compliance
flexibilities states may ultimately pursue.
The annual incremental cost is the projected additional cost of complying with the final
rule in the year analyzed and includes the net change in the annualized cost of capital investment
in new generating sources and heat rate improvements at coal-fired steam generating units, the
change in the ongoing costs of operating pollution controls, shifts between or amongst various
fuels, demand-side energy efficiency measures, and other actions associated with compliance.
The total compliance cost estimates presented here include the costs associated with monitoring,
reporting, and recordkeeping.3 The costs for both illustrative plan approaches are reflected in
Table ES-5 below and discussed more extensively in Chapter 3 of this RIA. All dollar estimates
are in 2011 dollars.
The EPA estimates the annual incremental compliance cost for the rate-based approach
for final emission guidelines to be $2.5 billion in 2020, $1.0 billion in 2025 and $8.4 billion in
2030, including the costs associated with monitoring, reporting, and recordkeeping.4 The EPA
estimates the annual incremental compliance cost for the mass-based approach for final emission
guidelines to be $1.4 billion in 2020, $3.0 billion in 2025 and $5.1 billion in 2030, including the
costs associated with monitoring, reporting, and recordkeeping.
Table ES-5. Compliance Costs for the Illustrative Rate-Based and Mass-Based Plan
Approaches
Incremental Cost from Base Case (billions of 2011$)
Rate-based Approach Mass-based Approach
2020 $2.5 $1.4
2025 $1.0 $3.0
2030 $8.4 $5.1
Source: Integrated Planning Model, 2015, with post-processing to account for exogenous demand-side management
energy efficiency costs and monitoring, reporting, and recordkeeping costs. See Chapter 3 of this RIA for more
details.
3 These costs are estimated outside of the IPM modelling framework as IPM only models the contiguous U.S. and
does not incorporate monitoring, reporting, and recordkeeping requirements specific to the Clean Power Plan Final
Rules.
4The MR&R costs estimates are $67 million in 2020, $16 million in 2025 and $16 million in 2030 and are assumed
to be the same for both rate-based and mass-based illustrative plan approaches. Note the MR&R costs in 2020 are
related to facilities setting up net energy output monitoring and upgrading data acquisition systems.
ES-9
-------
The costs reported in Table ES-5 represent the estimated incremental electric utility
generating costs changes from the base case plus the estimates of demand-side energy efficiency
program costs (which are paid by electric utilities), demand-side energy efficiency participant
costs (which are paid by electric utility consumers), and MR&R costs. For example, in 2030,
under the rate-based approach, the incremental electric utility generating costs decline by about
$18.0 billion from the base case. MR&R requirements in 2030 are estimated at $16.0 million,
and demand-side energy efficiency costs in 2030 are estimated to be $26.3 billion, split equally
between program and participants using a 3 percent discount rate (see Chapter 3 of this RIA for
more details on these estimates). These cost estimates sum to the $8.4 billion shown in Table ES-
3 and represent the total costs of the rate-based illustrative plan approach in 2030. The same
approach applies in each year of analysis for the rate-based and the mass-based illustrative plan
approaches.
The compliance costs reported in Table ES-5 are not social costs. These costs represent
the estimated expenditures incurred by EGUs and states to comply with the BSER goals for the
Clean Power Plan Final Rule. These compliance cost estimates are compared to estimates of
social benefits to derive net benefits of the final emission guidelines, which are presented later in
this Executive Summary. For a more extensive discussion of social costs and benefits, see
Chapter 3 and Chapter 4, respectively, of this RIA.
ES.6 Monetized Climate Benefits and Health Co-benefits
Implementing the final emission guidelines is expected to reduce emissions of CO2 and
have ancillary emission reductions (i.e., co-benefits) of SCh, NO2, and directly emitted PlVh.5,
which would lead to lower ambient concentrations of PM2.5 and ozone. The climate benefits
estimates have been calculated using the estimated values of marginal climate impacts presented
in the Technical Support Document: Technical Update of the Social Cost of Carbon for
Regulatory Impact Analysis under Executive Order 12866 (May 2013, Revised July 2015),
henceforth denoted as the current SC-CCh TSD.5 Also, the range of combined benefits reflects
5 Technical Support Document: Technical Update of the Social Cost of Carbon for Regulatory Impact Analysis
Under Executive Order 12866, Interagency Working Group on Social Cost of Carbon, with participation by Council
of Economic Advisers, Council on Environmental Quality, Department of Agriculture, Department of Commerce,
Department of Energy, Department of Transportation, Domestic Policy Council, Environmental Protection Agency,
National Economic Council, Office of Management and Budget, Office of Science and Technology Policy, and
ES-10
-------
different concentration-response functions for the air quality health co-benefits, but it does not
capture the full range of uncertainty inherent in the health co-benefits estimates. Furthermore, we
were unable to quantify or monetize all of the climate benefits and health and environmental co-
benefits associated with the final emission guidelines, including reducing exposure to 862, NOx,
and hazardous air pollutants (e.g., mercury), as well as ecosystem effects and visibility
improvement. The omission of these endpoints from the monetized results should not imply that
the impacts are small or unimportant. Table ES-6 provides the list of the quantified and
unquantified health and environmental benefits in this analysis.
Department of Treasury (May 2013, Revised July 2015). Available at:
Acces sed 7/11/2015.
ES-11
-------
Table ES-6. Quantified and Unquantified Benefits
Benefits Category
Specific Effect
Effect Has Effect Has
Been Been More Information
Quantified Monetized
Improved
Environment
Reduced climate
effects
Global climate impacts from CO2 '
Climate impacts from ozone and black carbon (directly
emitted PM) ~~
Other climate impacts (e.g., other GHGs such as methane,
aerosols, other impacts)
S SC-C02 TSD
Ozone ISA, PM
~~ ISA2
IPCC2
Improved Human Health (co-benefits)
Reduced incidence of
premature mortality
from exposure to
PM2.5
Reduced incidence of
morbidity from
exposure to PM2.5
Reduced incidence of
mortality from
exposure to ozone
Adult premature mortality based on cohort study estimates ,
and expert elicitation estimates (age >25 or age >30)
Infant mortality (age < 1 ) /
Non-fatal heart attacks (age > 18) -S
Hospital admissions respiratory (all ages) S
Hospital admissions cardiovascular (age >20) /
Emergency room visits for asthma (all ages) S
Acute bronchitis (age 8-12) /
Lower respiratory symptoms (age 7-14) -S
Upper respiratory symptoms (asthmatics age 9-11) S
Asthma exacerbation (asthmatics age 6-18) /
Lost work days (age 18-65) -S
Minor restricted-activity days (age 1 8-65) /
Chronic Bronchitis (age >26)
Emergency room visits for cardiovascular effects (all ages)
Strokes and cerebrovascular disease (age 50-79)
Other cardiovascular effects (e.g., other ages)
Other respiratory effects (e.g., pulmonary function, non-
asthma ER visits, non-bronchitis chronic diseases, other
ages and populations)
Reproductive and developmental effects (e.g., low birth
weight, pre-term births, etc)
Cancer, mutagenicity, and genotoxicity effects
Premature mortality based on short-term study estimates (all ,
ages)
Premature mortality based on long-term study estimates
(age 30-99) ~~
S PM ISA
S PM ISA
S PM ISA
S PM ISA
S PM ISA
v' PM ISA
S PM ISA
S PM ISA
S PM ISA
S PM ISA
v' PM ISA
S PM ISA
PM ISA2
PM ISA2
PM ISA2
PM ISA3
PM ISA3
PM ISA3-4
PM ISA3-4
S Ozone ISA
Ozone ISA2
Reduced incidence of
morbidity from
exposure to ozone
Hospital admissions respiratory causes (age > 65) -S
Hospital admissions respiratory causes (age <2) /
Emergency department visits for asthma (all ages) /
Minor restricted-activity days (age 18-65) -S
School absence days (age 5-1 7) /
Decreased outdoor worker productivity (age 18-65)
Other respiratory effects (e.g., premature aging of lungs)
Cardiovascular and nervous system effects
Reproductive and developmental effects
S Ozone ISA
S Ozone ISA
v' Ozone ISA
S Ozone ISA
v' Ozone ISA
Ozone ISA2
Ozone ISA3
Ozone ISA3
Ozone ISA3-4
ES-12
-------
Table ES-6. Continued
Asthma hospital admissions (all ages)
Chronic lung disease hospital admissions (age > 65)
Respiratory emergency department visits (all ages)
Reduced incidence of Asthma exacerbation (asthmatics age 4-18)
morbidity from Acute respiratory symptoms (age 7-14)
exposure to NO2 Premature mortality
Other respiratory effects (e.g., airway hyperresponsiveness
and inflammation, lung function, other ages and
populations)
Respiratory hospital admissions (age > 65)
Asthma emergency department visits (all ages)
r,_J J:_:J r Asthma exacerbation (asthmatics age 4-1 2)
morbidif from Acute respiratory symptoms (age 7-14)
Ovrnr/t,, Qrv Premature mortality
Other respiratory effects (e.g., airway hyperresponsiveness
and inflammation, lung function, other ages and
populations)
Neurologic effects IQ loss
Reduced incidence of other neurologic effects (e_g developmental delays,
morbidity from memo^ behavk)r)
^"i'"1'"^ ^ Cardiovascular effects
Genotoxic, immunologic, and other toxic effects
N02 ISA2
N02 ISA2
NO2 ISA2
N02 ISA2
N02 ISA2
NO2 ISA2-3-4
NO2 ISA3-4
S02 ISA2
SO2 ISA2
S02 ISA2
S02 ISA2
SO2 ISA2-3-4
SO2 ISA2-3
IRIS; NRC, 20002
IRIS; NRC, 20003
IRIS; NRC, 20003-4
IRIS; NRC, 20003-4
Improved Environment (co-benefits)
Reduced visibility Visibility in Class 1 areas
impairment Visibility in residential areas
Reduced effects on Household soiling
materials Materials damage (e.g., corrosion, increased wear)
Reduced PM
deposition (metals and Effects on Individual organisms and ecosystems
organics)
Visible foliar injury on vegetation
Reduced vegetation growth and reproduction
Yield and quality of commercial forest products and crops
Reduced vegetation Damage to urban ornamental plants
and ecosystem effects Carbon sequestration in terrestrial ecosystems
from exposure to Recreational demand associated with forest aesthetics
ozone Other non-use effects
Ecosystem functions (e.g., water cycling, biogeochemical
cycles, net primary productivity, leaf-gas exchange,
community composition)
Recreational fishing
Tree mortality and decline
Reduced effects from Commercial fishing and forestry effects
acid deposition Recreational demand in terrestrial and aquatic ecosystems
Other non-use effects
Ecosystem functions (e.g., biogeochemical cycles)
PM ISA2
PM ISA2
PM ISA2-3
PM ISA3
PM ISA3
Ozone ISA2
Ozone ISA2
Ozone ISA2
Ozone ISA3
Ozone ISA2
Ozone ISA3
Ozone ISA3
Ozone ISA3
NOx SOx ISA2
NOx SOx ISA3
NOx SOx ISA3
NOx SOx ISA3
NOx SOx ISA3
NOx SOx ISA3
ES-13
-------
Table ES-6. Continued
Reduced effects from
nutrient enrichment
Reduced vegetation
effects from exposure
to SO2 and NOX
Reduced ecosystem
effects from exposure
to methylmercury
Species composition and biodiversity in terrestrial and
estuarine ecosystems
Coastal eutrophication
Recreational demand in terrestrial and estuarine ecosystems
Other non-use effects
Ecosystem functions (e.g., biogeochemical cycles, fire
regulation)
Injury to vegetation from SCh exposure
Injury to vegetation from NOX exposure
Effects on fish, birds, and mammals (e.g., reproductive
effects)
Commercial, subsistence and recreational fishing
NOx SOx ISA3
NOx SOx ISA3
NOx SOx ISA3
NOx SOx ISA3
NOx SOx ISA3
NOx SOx ISA3
NOx SOx ISA3
Mercury Study
~~ RTC3
Mercury Study
~~ RTC2
1 The global climate and related impacts of COi emissions changes, such as sea level rise, are estimated within each
integrated assessment model as part of the calculation of the SC-COi. The resulting monetized damages, which
are relevant for conducting the benefit-cost analysis, are used in this RIA to estimate the welfare effects of
quantified changes in COi emissions.
2 We assess these co-benefits qualitatively due to data and resource limitations for this analysis.
3 We assess these co-benefits qualitatively because we do not have sufficient confidence in available data or
methods.
4 We assess these co-benefits qualitatively because current evidence is only suggestive of causality or there are other
significant concerns over the strength of the association.
ES.6.1 Estimating Global Climate Benefits
We estimate the global social benefits of COi emission reductions expected from this
ralemaking using the SC-CCh estimates presented in the current SC-COi TSD. We refer to these
estimates, which were developed by the U.S. government, as "SC-COi estimates" for the
remainder of this document. The SC-CO2 is a metric that estimates the monetary value of
impacts associated with marginal changes in CCh emissions in a given year. It includes a wide
range of anticipated climate impacts, such as net changes in agricultural productivity and human
health, property damage from increased flood risk, and changes in energy system costs, such as
reduced costs for heating and increased costs for air conditioning. It is typically used to assess
the avoided damages as a result of regulatory actions (i.e., benefits of rulemakings that lead to an
incremental reduction in cumulative global CO2 emissions).
The SC-CO2 estimates used in this analysis have been developed over many years, using
the best science available, and with input from the public. The EPA and other federal agencies
ES-14
-------
have considered the extensive public comments on ways to improve SC-CCh estimation received
via the notice and comment period that was part of numerous rulemakings. In addition, OMB's
Office of Information and Regulatory Affairs recently issued a response to the public comments
it sought through a separate comment period on the approach used to develop the SC-COi
estimates.6
An interagency working group (IWG) that included the EPA and other executive branch
entities used three integrated assessment models (lAMs) to develop SC-CCh estimates and
recommended four global values for use in regulatory analyses. The SC-COi estimates represent
global measures because of the distinctive nature of the climate change problem. Emissions of
greenhouse gases contribute to damages around the world, even when they are released in the
United States, and the world's economies are now highly interconnected. Therefore, the SC-COi
estimates incorporate the worldwide damages caused by carbon dioxide emissions in order to
reflect the global nature of the problem, and we expect other governments to consider the global
consequences of their greenhouse gas emissions when setting their own domestic policies. See
RIA Chapter 4 for more discussion.
The IWG first released the estimates in February 2010 and updated them in 2013 using
new versions of each IAM. The SC-COi values was estimated using three integrated assessment
models (DICE, FUND, and PAGE)7, which the IWG harmonized across three key inputs: the
probability distribution for equilibrium climate sensitivity; five scenarios for economic,
population, and emissions growth; and three constant discount rates. The 2010 SC-COi
Technical Support Document (2010 SC-CCh TSD) provides a complete discussion of the
methodology and the current SC-CChTSD8 presents and discusses the updated estimates. The
four SC-COi estimates are as follows: $12, $40, $60, and $120 per short ton of COi emissions in
the year 2020 (2011$), and each estimate increases over time.9 These SC-COi estimates are
6 Seehttps://www.whitehouse.gov/sites/default/files/omb/inforeg/scc-response-to-comments-fmal-july-2015.pdf
7 The full models names are as follows: Dynamic Integrated Climate and Economy (DICE); Climate Framework for
Uncertainty, Negotiation, and Distribution (FUND); and Policy Analysis of the Greenhouse Gas Effect (PAGE).
8 The IWG published the updated TSD in 2013, then issued two minor corrections to it in July 2015.
9 The 2010 and 2013 TSDs present SC-CO2 in 2007$ per metric ton. The estimates were adjusted to (1) short tons
for using conversion factor 0.90718474 and (2) 2011$ using GDP Implicit Price Deflator,
http://www.gpo.gov/fdsys/pkg/ECONI-2013-02/pdf/ECONI-2013-02-Pg3.pdf.
ES-15
-------
associated with different discount rates. The first three estimates are the model average at 5
percent discount rate, 3 percent, and 2.5 percent, respectively, and the fourth estimate is the 95th
percentile at 3 percent.
The 2010 SC-CO2 TSD noted a number of limitations to the SC-COi analysis, including
the incomplete way in which the lAMs capture catastrophic and non-catastrophic impacts, their
incomplete treatment of adaptation and technological change, uncertainty in the extrapolation of
damages to high temperatures, and assumptions regarding risk aversion. Currently integrated
assessment models do not assign value to all of the important physical, ecological, and economic
impacts of climate change recognized in the climate change literature because of a lack of
precise information on the nature of damages and because the science incorporated into these
models understandably lags behind the most recent research. In particular, the IPCC Fourth
Assessment Report concluded that "It is very likely that [SC-COi estimates] underestimate the
damage costs because they cannot include many non-quantifiable impacts." Nonetheless, these
estimates and the discussion of their limitations represent the best available information about
the social benefits of COi emission reductions to inform the benefit-cost analysis.
In addition, after careful evaluation of the full range of comments submitted to OMB's
Office of Information and Regulatory Affairs, the IWG continues to recommend the use of these
SC-CO2 estimates in regulatory impact analysis. With the release of the response to comments,
the IWG announced plans to obtain expert independent advice from the National Academies of
Sciences, Engineering, and Medicine (Academies) to ensure that the SC-COi estimates continue
to reflect the best available scientific and economic information on climate change.10 The
Academies process will be informed by the public comments received and focus on the technical
merits and challenges of potential approaches to improving the SC-CCh estimates in future
updates.
ES 6.2 Estimating Air Quality Health Co-Benefits
The final emission guidelines would reduce emissions of precursor pollutants (e.g., SCh,
NOx, and directly emitted particles), which in turn would lower ambient concentrations of PMi.5
1 See.
ES-16
-------
and ozone. This co-benefits analysis quantifies the monetized benefits associated with the
reduced exposure to these two pollutants.11 Unlike the global SC-COi estimates, the air quality
health co-benefits are only estimated for the contiguous U.S. The estimates of monetized PM2.5
co-benefits include avoided premature deaths (derived from effect coefficients in two cohort
studies [Krewski et al. 2009 and Lepeule et al. 2012] for adults and one for infants [Woodruff et
al. 1997]), as well as avoided morbidity effects for ten non-fatal endpoints ranging in severity
from lower respiratory symptoms to heart attacks (U.S. EPA, 2012). The estimates of monetized
ozone co-benefits include avoided premature deaths (derived from the range of effect
coefficients represented by two short-term epidemiology studies [Bell et al. (2004) and Levy et
al. (2005)]), as well as avoided morbidity effects for five non-fatal endpoints ranging in severity
from school absence days to hospital admissions (U.S. EPA, 2008, 2011).
We use a "benefit-per-ton" approach to estimate the PMi.5 and ozone co-benefits in this
RIA. Benefit-per-ton approaches apply an average benefit per ton derived from modeling of
benefits of specific air quality scenarios to estimates of emissions reductions for scenarios where
no air quality modeling is available. The benefit-per-ton approach we use in this RIA relies on
estimates of human health responses to exposure to PM and ozone obtained from the peer-
reviewed scientific literature. These estimates are used in conjunction with population data,
baseline health information, air quality data and economic valuation information to conduct
health impact and economic benefits assessments.
Specifically, in this analysis, we multiplied the benefit-per-ton estimates by the
corresponding emission reductions that were generated from air quality modeling of the
proposed Clean Power Plan. Similar to the co-benefits analysis conducted for the RIA for this
rule at proposal, we generated regional benefit-per-ton estimates by aggregating the impacts in
BenMAP12 to the region (i.e., East, West, and California) rather than aggregating to the nation.
To calculate the co-benefits for the final emission guidelines, we then multiplied the regional
11 We did not estimate the co-benefits associated with reducing direct exposure to SOi and NOx. For this RIA, we
did not estimate changes in emissions of directly emitted particles. As a result, quantified PM2.s related benefits are
underestimated by a relatively small amount. In the proposal RIA, the benefits from reductions in directly emitted
PM2.s were less than 10 percent of total monetized health co-benefits across all scenarios and years.
12 BenMAP is a computer program developed by the EPA that calculates the number and economic value of air
pollution-related deaths and illnesses. The software incorporates a database that includes many of the concentration-
response relationships, population files, and health and economic data needed to quantify these impacts.
ES-17
-------
benefit-per-ton estimates for the EGU sector by the corresponding emission reductions. All
benefit-per-ton estimates reflect the geographic distribution of the modeled emissions, which
may not exactly match the emission reductions in this rulemaking, and thus they may not reflect
the local variability in population density, meteorology, exposure, baseline health incidence
rates, or other local factors for any specific location.
Our estimate of the monetized co-benefits is based on the EPA's interpretation of the best
available scientific literature (U.S. EPA, 2009) and methods and supported by the EPA's Science
Advisory Board and the NAS (NRC, 2002). Below are key assumptions underlying the estimates
for PMi.s-related premature mortality, which accounts for 98 percent of the monetized PlVb.5
health co-benefits:
1. We assume that all fine particles, regardless of their chemical composition, are
equally potent in causing premature mortality. This is an important assumption,
because PM2.5 varies considerably in composition across sources, but the scientific
evidence is not yet sufficient to allow differentiation of effect estimates by particle
type. The PM ISA concluded that "many constituents of PlVb.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, 2009b).
2. We assume that the health impact function for fine particles is log-linear without a
threshold in this analysis. Thus, the estimates include health co-benefits from
reducing fine particles in areas with varied concentrations of PM2.5, including both
areas that do not meet the National Ambient Air Quality Standard for fine particles
and those areas that are in attainment, down to the lowest modeled concentrations.
3. 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 PlVb.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, 2004c), which affects the valuation of mortality co-
benefits at different discount rates.
ES-18
-------
Every benefits analysis examining the potential effects of a change in environmental
protection requirements is limited, to some extent, by data gaps, model capabilities (such as
geographic coverage) and uncertainties in the underlying scientific and economic studies used to
configure the benefit and cost models. In addition, given the flexibilities afforded states in
complying with the emission guidelines, the co-benefits estimated presented in this RIA are not
definitive estimates, but are instead illustrative of approaches that states may take. Despite these
uncertainties, we believe this analysis provides a reasonable indication of the expected health co-
benefits of the air quality emission reductions for the final emission guidelines under a set of
reasonable assumptions. This analysis does not include the type of detailed uncertainty
assessment found in the 2012 PM2.5 National Ambient Air Quality Standard (NAAQS) RIA (U.S.
EPA, 2012) because we lack the necessary air quality input and monitoring data to conduct a
complete benefits assessment. In addition, using a benefit-per-ton approach adds another
important source of uncertainty to the benefits estimates.
ES 6.3 Combined Benefits Estimates
The EPA has evaluated the range of potential impacts by combining all four SC-COi
values with health co-benefits values at the 3 percent and 7 percent discount rates. Different
discount rates are applied to SC-CCh than to the health co-benefit estimates; because COi
emissions are long-lived and subsequent damages occur over many years. Moreover, several
discount rates are applied to SC-COi because the literature shows that the estimate of SC-COi is
sensitive to assumptions about discount rate and because no consensus exists on the appropriate
rate to use in an intergenerational context. The U.S. government centered its attention on the
average SC-COi at a 3 percent discount rate but emphasized the importance of considering all
four SC-CO2 estimates. Table ES-7 (rate-based illustrative plan approach) and Table ES-8
(mass-based illustrative plan approach) provide the combined climate benefits and health co-
benefits for the Clean Power Plan Final Rule estimated for 2020, 2025, and 2030 for each
discount rate combination. All dollar estimates are in 2011 dollars.
ES-19
-------
Table ES-7. Combined Estimates of Climate Benefits and Health Co-Benefits for Rate-
Based Approach (billions of 2011$)*
Climate Climate Benefits plus Health Co-benefits
SC-CO2 Discount Rate and Statistic** Benefits (Discount Rate Applied to Health Co-benefits)
3%
In 2020
5%
3%
2.5%
(95th percentile)
In 2025
5%
3%
2.5%
3% (95th percentile)
3%
In 2030
5%
3%
2.5%
(95th percentile)
Only
69
$0.80
$2.8
$4.1
$8.2
232
$3.1
$10
$15
$31
415
$6.4
$20
$29
$61
3%
million short
$1.5
$3.5
$4.9
$8.9
million short
$11
$18
$23
$38
million short
$21
$34
$43
$75
tons
to
to
to
to
tons
to
to
to
to
tons
to
to
to
to
CO2
$2.6
$4.6
$6.0
$10
CO2
$21
$28
$33
$49
CO2
$40
$54
$63
$95
$1.4
$3.5
$4.8
$8.9
$9.9
$17
$22
$38
$19
$33
$42
$74
7%
to
to
to
to
to
to
to
to
to
to
to
to
$2.5
$4.5
$5.9
$9.9
$19
$26
$31
$47
$37
$51
$60
$92
*A11 benefit estimates are rounded to two significant figures. Climate benefits are based on reductions in CO2
emissions. Co-benefits are based on regional benefit-per-ton estimates. Ozone co-benefits occur in analysis year, so
they are the same for all discount rates. The health co-benefits reflect the sum of the PM2.s and ozone co-benefits
and reflect the range based on adult mortality functions (e.g., from Krewski et al. (2009) with Bell et al. (2004) to
Lepeule et al. (2012) with Levy et al. (2005)). The monetized health co-benefits do not include reduced health
effects from reductions in directly emitted PM2.s, direct exposure to NOx, SO2, and HAP; ecosystem effects; or
visibility impairment. See Chapter 4 for more information about these estimates and for more information
regarding the uncertainty in these estimates.
**Unless otherwise specified, it is the model average.
ES-20
-------
Table ES-8. Combined Estimates of Climate Benefits and Health Co-benefits for Mass-
Based Approach (billions of 2011$)*
SC-CO2 Discount
3%
In
(95
In
Climate
Rate and Statistic** Benefits
2020
5%
3%
2.5%
* percentile)
2025
5%
3%
2
3% (95th
3%
In
(95
.5%
percentile)
2030
5%
3%
2.5%
* percentile)
Only
82
$0.94
$3.3
$4.9
$9.7
264
$3.6
$12
$17
$35
413
$6.4
$20
$29
$60
Climate Benefits plus Health Co-benefits
(Discount Rate Applied to Health Co-benefits)
3%
million short
$2.9
$5.3
$6.9
$12
million short
$11
$19
$24
$42
million short
$18
$32
$41
$72
tons
to
to
to
to
tons
to
to
to
to
tons
to
to
to
to
CO2
$5.7
$8.1
$9.7
$14
CO2
$21
$29
$35
$52
CO2
$34
$48
$57
$89
$2.8
$5.1
$6.7
$11
$10
$18
$24
$42
$17
$31
$40
$71
7%
to
to
to
to
to
to
to
to
to
to
to
to
$5.3
$7.7
$9.3
$14
$19
$27
$33
$51
$32
$46
$55
$86
*A11 benefit estimates are rounded to two significant figures. Climate benefits are based on reductions in CO2
emissions. Co-benefits are based on regional benefit-per-ton estimates. Ozone co-benefits occur in analysis year, so
they are the same for all discount rates. The health co-benefits reflect the sum of the PM2.s and ozone co-benefits
and reflect the range based on adult mortality functions (e.g., from Krewski et al. (2009) with Bell et al. (2004) to
Lepeule et al. (2012) with Levy et al. (2005)). The monetized health co-benefits do not include reduced health
effects from reductions in directly emitted PM2.s, direct exposure to NOx, SO2, and HAP; ecosystem effects; or
visibility impairment. See Chapter 4 for more information about these estimates and for more information
regarding the uncertainty in these estimates.
**Unless otherwise specified, it is the model average.
ES.7 Net Benefits
Table ES-9 and ES-10 provide the estimates of the climate benefits, health co-benefits,
compliance costs and net benefits of the final emission guidelines for rate-based and mass-based
approaches, respectively. There are additional important benefits that the EPA could not
monetize. Due to current data and modeling limitations, our estimates of the benefits from
reducing COi emissions do not include important impacts like ocean acidification or potential
tipping points in natural or managed ecosystems. Unquantified benefits also include climate
benefits from reducing emissions of non-CCh greenhouse gases and co-benefits from reducing
exposure to SOi, NOx, and hazardous air pollutants (e.g., mercury), as well as ecosystem effects
and visibility impairment. Upon considering these limitations and uncertainties, it remains clear
that the benefits of this final rule are substantial and far outweigh the costs.
ES-21
-------
Table ES-9. Monetized Benefits, Compliance Costs, and Net Benefits Under the Rate-
based Illustrative Plan Approach (billions of 2011$) a
Rate-Based Approach
2020 2025 2030
Climate Benefits b
5% discount rate
3% discount rate
2.5% discount rate
95thpercentileat3%
$0.80
$2.8
$4.1
$8.2
$3.1
$10
$15
$31
$6.4
$20
$29
$61
Air Quality Co-benefits Discount Rate
3% 7% 3% 7% 3% 7%
Air Quality Health $0.70 to $1.8 $0.64 to $1.7 $7.4 to $18 $6.7 to $16 $14 to $34 $13 to $31
Co-benefits
Compliance Costsd $2.5 $1.0 $8.4
Net Benefits6 $1.0 to $2.1 $1.0 to $2.0 $17 to $27 $16 to $25 $26 to $45 $25 to $43
Non-monetized climate benefits
Reductions in exposure to ambient NO2 and SO2
Non-Monetized Reductions in mercury deposition
Benefits : l
Ecosystem benefits associated with reductions in emissions of NOx, SO2, PM, and mercury
Visibility impairment
a All are rounded to two significant figures, so figures may not sum.
b The climate benefit estimate in this summary table reflects global impacts from CO2 emission changes and does
not account for changes in non-CO2 GHG emissions. Also, different discount rates are applied to SC-CO2 than to the
other estimates because CO2 emissions are long-lived and subsequent damages occur over many years. The benefit
estimates in this table are based on the average SC-CO2 estimated for a 3 percent discount rate, however we
emphasize the importance and value of considering the full range of SC-CO2 values. As shown in the RIA, climate
benefits are also estimated using the other three SC-CO2 estimates (model average at 2.5 percent discount rate, 3
percent, and 5 percent; 95th percentile at 3 percent). The SC-CO2 estimates are year-specific and increase over time.
c The air quality health co-benefits reflect reduced exposure to PM2.s and ozone associated with emission reductions
of SO2 and NOx. The co-benefits do not include the benefits of reductions in directly emitted PM2.5. These
additional benefits would increase overall benefits by a few percent based on the analyses conducted for the
proposed rule. The range reflects the use of concentration-response functions from different epidemiology studies.
The reduction in premature fatalities each year accounts for over 98 percent of total monetized co-benefits from
PM2.s and ozone. These models assume that all fine particles, regardless of their chemical composition, are equally
potent in causing premature mortality because the scientific evidence is not yet sufficient to allow differentiation of
effect estimates by particle type. Estimates in the table are presented for three analytical years with air quality co-
benefits calculated using two discount rates. The estimates of co-benefits are annual estimates in each of the
analytical years, reflecting discounting of mortality benefits over the cessation lag between changes in PM2.s
concentrations and changes in risks of premature death (see RIA Chapter 4 for more details), and discounting of
morbidity benefits due to the multiple years of costs associated with some illnesses. The estimates are not the
present value of the benefits of the rule over the full compliance period.
d Total costs are approximated by the illustrative compliance costs estimated using the Integrated Planning Model for
the final emission guidelines and a discount rate of approximately 5 percent. This estimate also includes monitoring,
recordkeeping, and reporting costs and demand-side energy efficiency program and participant costs.
e The estimates of net benefits in this summary table are calculated using the global SC-CO2 at a 3 percent discount
rate (model average). The RIA includes combined climate and health estimates based on additional discount rates.
ES-22
-------
Table ES-10. Monetized Benefits, Compliance Costs, and Net Benefits under the Mass-
based Illustrative Plan Approach (billions of 2011$) a
Mass-Based Approach
2020 2025 2030
Climate Benefits b
5% discount rate
3% discount rate
2.5% discount rate
95thpercentileat3%
$0.94
$3.3
$4.9
$9.7
$3.6
$12
$17
$35
$6.4
$20
$29
$60
Air Quality Co-benefits Discount Rate
3% 7% 3% 7% 3% 7%
Air Quality Health $2.0 to $4.8 $1.8 to $4.4 $7.1 to $17 $6.5 to $16 $12 to $28 $11 to $26
Co-benefits
Compliance Costsd $1.4 $3.0 $5.1
Net Benefits6 $3.9 to $6.7 $3.7 to $6.3 $16 to $26 $15 to $24 $26 to $43 $25 to $40
Non-monetized climate benefits
Reductions in exposure to ambient NO2 and SO2
Non-Monetized Reductions in mercury deposition
1 Ecosystem benefits associated with reductions in emissions of NOx, SO2, PM, and
mercury
Visibility improvement
a All are rounded to two significant figures, so figures may not sum.
b The climate benefit estimate in this summary table reflects global impacts from CO2 emission changes and does
not account for changes in non-CO2 GHG emissions. Also, different discount rates are applied to SC-CO2 than to the
other estimates because CO2 emissions are long-lived and subsequent damages occur over many years. The benefit
estimates in this table are based on the average SC-CO2 estimated for a 3 percent discount rate, however we
emphasize the importance and value of considering the full range of SC-CO2 values. As shown in the RIA, climate
benefits are also estimated using the other three SC-CO2 estimates (model average at 2.5 percent discount rate, 3
percent, and 5 percent; 95th percentile at 3 percent). The SC-CO2 estimates are year-specific and increase over time.
c The air quality health co-benefits reflect reduced exposure to PM2.s and ozone associated with emission reductions
of, SO2 and NOx. The co-benefits do not include the benefits of reductions in directly emitted PM2.s. These
additional benefits would increase overall benefits by a few percent based on the analyses conducted for the
proposed rule. The range reflects the use of concentration-response functions from different epidemiology studies.
The reduction in premature fatalities each year accounts for over 98 percent of total monetized co-benefits from
PM2.s and ozone. These models assume that all fine particles, regardless of their chemical composition, are equally
potent in causing premature mortality because the scientific evidence is not yet sufficient to allow differentiation of
effect estimates by particle type. Estimates in the table are presented for three analytical years with air quality co-
benefits calculated using two discount rates. The estimates of co-benefits are annual estimates in each of the
analytical years, reflecting discounting of mortality benefits over the cessation lag between changes in PM2.5
concentrations and changes in risks of premature death (see RIA Chapter 4 for more details), and discounting of
morbidity benefits due to the multiple years of costs associated with some illnesses. The estimates are not the
present value of the benefits of the rule over the full compliance period.
d Total costs are approximated by the illustrative compliance costs estimated using the Integrated Planning Model for
the final emission guidelines and a discount rate of approximately 5 percent. This estimate also includes monitoring,
recordkeeping, and reporting costs and demand-side energy efficiency program and participant costs.
e The estimates of net benefits in this summary table are calculated using the global SC-CO2 at a 3 percent discount
rate (model average). The RIA includes combined climate and health estimates based on additional discount rates.
ES-23
-------
ES.8 Economic Impacts
The final emission guidelines have important energy market implications. Table ES-11
presents a variety of important energy market impacts for 2020, 2025, and 2030 for both the rate-
based and mass-based illustrative plan approaches.
Table ES-11. Summary Table of Important Energy Market Impacts (Percent Change from
Base Case)
Rate-Based
Retail electricity prices
Price of coal at minemouth
Coal production for power sector use
Price of natural gas delivered to power sector
Natural gas use for electricity generation
2020
3%
-1%
-5%
5%
3%
2025
1%
-5%
-14%
-8%
-1%
2030
1%
-4%
-25%
2%
-1%
Mass-Based
2020
3%
-1%
-7%
4%
5%
2025
2%
-5%
-17%
-3%
0%
2030
0%
-3%
-24%
-2%
-4%
Energy market impacts from the guidelines are discussed more extensively in Chapter 3 of this
RIA.
Additionally, changes in supply or demand for electricity, natural gas, and coal can
impact markets for goods and services produced by sectors that use these energy inputs in the
production process or that supply those sectors. Changes in cost of production may result in
changes in price and/or quantity produced by these sectors and these market changes may affect
the profitability of firms and the economic welfare of their consumers. The EPA recognizes that
these final emission guidelines provide flexibility, and states implementing the guidelines may
choose to mitigate impacts to some markets outside the EGU sector. Similarly, demand for new
generation or energy efficiency, for example, can result in changes in production and
profitability for firms that supply those goods and services.
ES.9 Employment Impacts
Executive Order 13563 directs federal agencies to consider the effect of regulations on
job creation and employment. According to the Executive Order, "our regulatory system must
protect public health, welfare, safety, and our environment while promoting economic growth,
innovation, competitiveness, and job creation. It must be based on the best available science"
(Executive Order 13563, 2011). Although standard benefit-cost analyses have not typically
included a separate analysis of regulation-induced employment impacts, we typically conduct
ES-24
-------
employment analyses. During the current economic recovery, employment impacts are of
particular concern and questions may arise about their existence and magnitude.
Given the wide range of approaches that may be used to meet the requirements of the
Clean Power Plan Final Rule, quantifying the associated employment impacts is difficult. The
EPA's illustrative employment analysis includes an estimate of projected employment impacts
associated with these guidelines for the utility power sector, coal and natural gas production, and
demand-side energy efficiency activities. These projections are derived, in part, from the detailed
model of the utility power sector used for this regulatory analysis, and U.S government data on
employment and labor productivity.
In the electricity, coal, and natural gas sectors, the EPA estimates that these guidelines
could result in a net decrease of approximately 25,000 job-years in 2025 for the final guidelines
under the rate-based illustrative plan approach and approximately 26,000 job-years in 2025
under the mass-based approach. For 2030 the estimates of the net decrease in job-years is 30,900
under the rate-based plan, and 33,700 under the mass-based plan. The Agency is also offering an
illustrative calculation of potential employment effects due to demand-side energy efficiency
programs. Employment impacts from demand-side energy efficiency programs in 2030 could
range from approximately 52,000 to 83,000 jobs under the final guidelines. More detail about
these analyses can be found in Chapter 6 of this RIA.
ES.10 References
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.Docket ID EPA-HQ-OAR-2009-0472-114577,
Technical Support Document: Social Cost of Carbon for Regulatory Impact Analysis Under
Executive Order 12866, Interagency Working Group on Social Cost of Carbon, with
participation by the Council of Economic Advisers, Council on Environmental Quality,
Department of Agriculture, Department of Commerce, Department of Energy, Department of
Transportation, Environmental Protection Agency, National Economic Council, Office of
Energy and Climate Change, Office of Management and Budget, Office of Science and
Technology Policy, and Department of Treasury (February 2010). Available at:
.
ES-25
-------
Docket ID EPA-HQ-OAR-2013-0602, Technical Support Document: Technical Update of the
Social Cost of Carbon for Regulatory Impact Analysis Under Executive Order 12866,
Interagency Working Group on Social Cost of Carbon, with Participation by Council of
Economic Advisers, Council on Environmental Quality, Department of Agriculture,
Department of Commerce, Department of Energy, Department of Transportation, Domestic
Policy Council, Environmental Protection Agency, National Economic Council, Office of
Management and Budget, Office of Science and Technology Policy, and Department of
Treasury (May 2013, Revised July 2015). Also available at: <
https://www.whitehouse.gov/sites/default/files/omb/inforeg/scc-tsd-final-july-2015.pdf>.
Accessed July 15, 2015.
Fann, N., K.R. Baker, and C.M. Fulcher. 2012. "Characterizing the PM2.5-Related Health
Benefits of Emission Reductions for 17 Industrial, Area and Mobile Emission Sectors Across
the U.S." Environment International. 49:41-151.
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.
Interagency Working Group on Social Cost of Carbon, with participation by Council of
Economic Advisers, Council on Environmental Quality, Department of Agriculture,
Department of Commerce, Department of Energy, Department of Transportation,
Environmental Protection Agency, National Economic Council, Office of Management and
Budget, Office of Science and Technology Policy, and Department of Treasury. Response to
to Comments: Social Cost of Carbon for Regulatory Impact Analysis Under Executive Order
12866. July 2015. Available at:
Accessed July 15, 2015.
Intergovernmental Panel on Climate Change (IPCC). 2007. Climate Change 2007: Synthesis
Report Contribution of Working Groups I, II and III to the Fourth Assessment Report of the
IPCC. Available at:
. Accessed June 6, 2015.
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.
ES-26
-------
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.
National Research Council (NRC). 2000. Toxicological Effects ofMethylmercury: Committee on
the Toxicological Effects ofMethylmercury." Board on Environmental Studies and
Toxicology. National Academies Press. Washington, DC.
National Research Council (NRC). 2002. Estimating the Public Health Benefits of Proposed Air
Pollution Regulations. National Academies Press. Washington, DC.
U.S. Environmental Protection Agency (U.S. EPA). 2008a. Integrated Science Assessment for
Sulfur OxidesHealth Criteria (Final Report). National Center for Environmental
Assessment - RTP Division, Research Triangle Park, NC. September. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2008b. Final Ozone NAAQS Regulatory
Impact Analysis. EPA-452/R-08-003. Office of Air Quality Planning and Standards Health
and Environmental Impacts Division, Air Benefit and Cost Group Research Triangle Park,
NC. March. Available at: < http://www.epa.gov/ttnecasl/regdata/RIAs/6-
ozoneriachapter6.pdf>. Accessed June 4, 2015.
U.S. EPA. 2008c. Integrated Science Assessment for Oxides of Nitrogen: Health Criteria (Final
Report). Research Triangle Park, NC: National Center for Environmental Assessment. July.
Available at < http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=194645>.
U.S. Environmental Protection Agency (U.S. EPA). 2008c. Integrated Science Assessment for
Oxides of Nitrogen - Health Criteria (Final Report). National Center for Environmental
Assessment, Research Triangle Park, NC. July. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2009b. Integrated Science Assessment for
Paniculate Matter (Final Report). EPA-600-R-08-139F. National Center for Environmental
Assessment - RTP Division, Research Triangle Park, NC. December. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2010d. Section 3: Re-analysis of the Benefits
of Attaining Alternative Ozone Standards to Incorporate Current Methods. Available at:
. Accessed June 4, 2015.
ES-27
-------
U.S. Environmental Protection Agency (U.S. EPA). 2012a. Regulatory Impact Analysis for the
Final Revisions to the National Ambient Air Quality Standards for Paniculate Matter. EPA-
452/R-12-003. Office of Air Quality Planning and Standards, Health and Environmental
Impacts Division, Research Triangle Park, NC. December. Available at: <
http://www.epa.gov/ttnecasl/regdata/RIAs/finalria.pdf>. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2013b. 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:
. Accessed June 4,
2015.
U.S. EPA. 2015. Technical Support Document (TSD) the Final Carbon Pollution Emission
Guidelines for Existing Stationary Sources: Electric Utility Generating Units. COi Emission
Performance Rate and Goal Computation.
Woodruff, T.J., J. Grillo, and K.C. Schoendorf. 1997. "The relationship between selected causes
of postneonatal infant mortality and particulate air pollution in the United States."
Environmental Health Perspectives. 105(6): 608-612.
ES-28
-------
CHAPTER 1: INTRODUCTION AND BACKGROUND FOR THE CLEAN POWER PLAN
1.1 Introduction
This document presents estimates of potential benefits, costs, and economic impacts of
illustrative approaches states may implement to comply with the Final Carbon Pollution
Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units (herein
referred to as "final emission guidelines" or the "Clean Power Plan Final Rule"). This chapter
contains background information on these rules and an outline of the chapters in the report.
1.2 Legal, Scientific and Economic Basis for this Rulemaking
1.2.1 Statutory Requirement
Clean Air Act section 111, which Congress enacted as part of the 1970 Clean Air Act
Amendments, establishes mechanisms for controlling emissions of air pollutants from stationary
sources. This provision requires the EPA to promulgate a list of categories of stationary sources
that the Administrator, in his or her judgment, finds "causes, or contributes significantly to, air
pollution which may reasonably be anticipated to endanger public health or welfare."13 The EPA
has listed more than 60 stationary source categories under this provision.14 Once the EPA lists a
source category, the EPA must, under CAA section lll(b)(l)(B), establish "standards of
performance" for emissions of air pollutants from new sources in the source categories.15 These
standards are known as new source performance standards (NSPS), and they are national
requirements that apply directly to the sources subject to them.
When the EPA establishes NSPS for new sources in a particular source category, the
EPA is also required, under CAA section lll(d)(l), to prescribe regulations for states to submit
plans regulating existing sources in that source category for any air pollutant that, in general, is
not regulated under the CAA section 109 requirements for the NAAQS or regulated under the
CAA section 112 requirements for hazardous air pollutants (HAP). CAA section lll(d)'s
mechanism for regulating existing sources differs from the one that CAA section 11 l(b) provides
13CAA§lll(b)(l)(A).
14 See 40 CFR 60 subparts Cb - OOOO.
15
CAA§lll(b)(l)(B),
1-1
-------
for new sources because CAA section lll(d) contemplates states submitting plans that establish
"standards of performance" for the affected sources and that contain other measures to
implement and enforce those standards.
"Standards of performance" are defined under CAA section lll(a)(l) as standards for
emissions that reflect the emission limitation achievable from the "best system of emission
reduction," considering costs and other factors, that "the Administrator determines has been
adequately demonstrated." CAA section lll(d)(l) grants states the authority, in applying a
standard of performance to a particular source, to take into account the source's remaining useful
life or other factors.
Under CAA section 11 l(d), a state must submit its plan to the EPA for approval, and the
EPA must approve the state plan if it is "satisfactory."16 If a state does not submit a plan, or if the
EPA does not approve a state's plan, then the EPA must establish a plan for that state.17 Once a
state receives the EPA's approval of its plan, the provisions in the plan become federally
enforceable against the entity responsible for noncompliance, in the same manner as the
provisions of an approved State Implementation Plan (SIP) under the Act.
1.2.2 Health and Welfare Impacts from Climate Change
According to the National Research Council, "Emissions of COi from the burning of
fossil fuels have ushered in a new epoch where human activities will largely determine the
evolution of Earth's climate. Because COi in the atmosphere is long lived, it can effectively lock
Earth and future generations into a range of impacts, some of which could become very severe.
Therefore, emission reduction choices made today matter in determining impacts experienced
not just over the next few decades, but in the coming centuries and millennia."18
In 2009, based on a large body of robust and compelling scientific evidence, the EPA
Administrator issued the Endangerment Finding under CAA section 202(a)(l).19 In the
16 CAA section lll(d)(2)(A).
17 CAA section lll(d)(2)(A).
18 National Research Council, Climate Stabilization Targets, p.3.
19 "Endangerment and Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the Clean Air
Act," 74 Fed. Reg. 66,496 (Dec. 15, 2009) ("Endangerment Finding").
1-2
-------
Endangerment Finding, the Administrator found that the current, elevated concentrations of
GHGs in the atmospherealready at levels unprecedented in human historymay reasonably
be anticipated to endanger public health and welfare of current and future generations in the
United States.
Since the administrative record concerning the Endangerment Finding closed following
the EPA's 2010 Reconsideration Denial, the climate has continued to change, with new records
being set for a number of climate indicators such as global average surface temperatures, Arctic
sea ice retreat, COi concentrations, and sea level rise. Additionally, a number of major scientific
assessments have been released that improve understanding of the climate system and strengthen
the case that GHGs endanger public health and welfare both for current and future generations.
These assessments are from the Intergovernmental Panel on Climate Change (IPCC), the U.S.
Global Change Research Program (USGCRP), and the National Research Council (NRC). These
and other assessments are discussed in more detail in the preamble and in Chapter 4 of this
Regulatory Impact Assessment (RIA).
1.2.3 Market Failure
Many regulations are promulgated to correct market failures, which otherwise lead to a
suboptimal allocation of resources within the free market. Air quality and pollution control
regulations address "negative externalities" whereby the market does not internalize the full
opportunity cost of production borne by society as public goods such as air quality are unpriced.
GHG emissions impose costs on society, such as negative health and welfare impacts,
that are not reflected in the market price of the goods produced through the polluting process.
For this regulatory action the good produced is electricity. If a fossil fuel-fired electricity
producer pollutes the atmosphere when it generates electricity, this cost will be borne not by the
polluting firm but by society as a whole, thus imposing a negative externality. The equilibrium
market price of electricity may fail to incorporate the full opportunity cost to society of
generating electricity. All else equal, given this externality, the composition of EGUs used to
generate electricity in a free market will not be socially optimal, and the quantity of electricity
generated may not be at the socially optimal level. Fossil fuel-fired EGUs may produce more
electricity than would occur if they had to account for the cost associated with this negative
externality. Consequently, absent a regulation on emissions, the composition of the fleet of
1-3
-------
EGUs used to generate electricity may not be socially optimal, and the marginal social cost of
the last unit of electricity produced may exceed its marginal social benefit. This regulation will
regulation will work towards addressing this market failure by causing affected EGUs to begin to
internalize the negative externality associated with CO2 emissions.
1.3 Summary of Regulatory Analysis
In accordance with Executive Order 12866, Executive Order 13563, OMB Circular A-4,
and the EPA's "Guidelines for Preparing Economic Analyses," the EPA prepared this RIA for
this "significant regulatory action." This action is an economically significant regulatory action
because it is expected to have an annual effect on the economy of $100 million or more or
adversely affect in a material way the economy, a sector of the economy, productivity,
competition, jobs, the environment, public health or safety, or state, local, or tribal governments
or communities.20
This RIA addresses the potential costs, emission reductions, and benefits of the final
emission guidelines that are the focus of this action. Additionally, this RIA includes information
about potential impacts on electricity markets, employment, and markets outside the electricity
sector.
In evaluating the impacts of the final guidelines, we analyzed a number of uncertainties.
For example, the analysis includes an evaluation of two illustrative plan approaches that states
and affected EGUs may take to accomplish state emission performance goals, a rate-based and a
mass-based approach. The RIA also examines key uncertainties in the estimated benefits of
reducing carbon dioxide and other air pollutants. For a further discussion of key evaluations of
uncertainty in the regulatory analyses for this rulemaking, see Chapter 8 of this RIA.
1.4 Background for the Final Emission Guidelines
1.4.1 Base Case and Years of Analysis
The rule analyzed in this RIA finalizes emission guidelines for states to limit COi
emissions from certain existing EGUs. The base case for this analysis, which uses the Integrated
20 The analysis in this RIA and the RIA that accompanied the proposal together constitute the economic assessment
required by CAA section 317. In the EPA's judgment, the assessment is as extensive as practicable taking into
account the EPA's time, resources, and other duties and authorities.
1-4
-------
Planning Model (IPM), includes state rules that have been finalized and/or approved by a state's
legislature or environmental agencies, as well as final federal rules. The IPM Base Case v.5.15
includes the Cross-State Air Pollution Rule (CSAPR), the Mercury and Air Toxics Rule
(MATS), the proposed Carbon Pollution Standards for New Power Plants, the Cooling Water
Intakes (316(b)) Rule, the Combustion Residuals from Electric Utilities (CCR), and other state
and Federal regulations to the extent that they contain measures, permits, or other air-related
limitations or requirements. Additional legally binding and enforceable commitments for GHG
reductions considered in the base case are discussed in the documentation for IPM.21
Costs and benefits are presented for illustrative plan approaches for the analysis years of
2020, 2025, and 2030. These years were selected because they represent initial build up, interim,
and full implementation years for the two illustrative approaches analyzed. Analyses of energy,
economic, and employment impacts are presented for illustrative plan approaches in 2020, 2025,
and 2030. All dollar estimates are presented in 2011 dollars.
1.4.2 Definition of Affected Sources
For the emission guidelines, an affected EGU is any fossil fuel-fired electric utility steam
generating unit or stationary combustion turbine that was in operation or had commenced
construction as of January 8, 2014,22 and that meets the following criteria, which differ
depending on the type of unit. To be an affected source, such a unit, if it is a steam generating
unit or integrated gasification combined cycle (IGCC), must serve a generator capable of selling
greater than 25 MW to a utility power distribution system and have a base load rating greater
than 260 GJ/h (250 MMBtu/h) heat input of fossil fuel (either alone or in combination with any
other fuel). If such a unit is a stationary combustion turbine, the unit must meet the definition of
a combined cycle or combined heat and power combustion turbine, serve a generator capable of
selling greater than 25 MW to a utility power distribution system, and have a base load rating of
greater than 260 GJ/h (250 MMBtu/h). Certain EGUs are exempt from inclusion in a state plan.
For specifics on these criteria see section IV of the preamble.
21 Detailed documentation for IPM v.5.15 is available at: http://www.epa.gov/powersectormodeling
22 Under Section 11 l(a) of the CAA, determination of affected sources is based on the date that the EPA proposes
action on such sources. January 8, 2014 is the date the proposed GHG standards of performance for new fossil fuel-
fired EGUs were published in the Federal Register (79 FR 1430).
1-5
-------
When considering and understanding applicability, the following definitions may be
helpful. Simple cycle combustion turbine means any stationary combustion turbine which does
not recover heat from the combustion turbine engine exhaust gases for purposes other than
enhancing the performance of the stationary combustion turbine itself. Combined cycle
combustion turbine means any stationary combustion turbine which recovers heat from the
combustion turbine engine exhaust gases to generate steam that is used to create additional
electric power output in a steam turbine. Combined heat and power (CHP) combustion turbine
means any stationary combustion turbine which recovers heat from the combustion turbine
engine exhaust gases to heat water or another medium, generate steam for useful purposes other
than exclusively for additional electric generation, or directly uses the heat in the exhaust gases
for a useful purpose.
1.4.3 Regulated Pollutant
The purpose of this CAA section lll(d) rule is to address CCh emissions from fossil
fuel-fired power plants in the U.S. because they are the largest domestic stationary source of
emissions of carbon dioxide (COi), the most prevalent of the greenhouse gases (GHG), which
are air pollutants that the EPA has determined endangers public health and welfare through their
contribution to climate change. This rule establishes for the first time federal emission guidelines
for existing power plants that will lead to significant reductions in COi emissions.
1.4.4 Emission Guidelines
In this action, the Environmental Protection Agency (EPA) is establishing final emission
guidelines for states to follow in developing plans to reduce greenhouse gas emissions from
existing fossil fuel-fired EGUs. Specifically, the EPA is establishing: 1) COi emission
performance rates representing the best system of emission reduction (BSER) for two
subcategories of existing fossil fuel-fired EGUs - fossil fuel-fired electric utility steam
generating units and stationary combustion turbines, 2) state-specific COi goals reflecting the
COi emission performance rates, and 3) guidelines for the development, submittal and
implementation of state plans that establish emission standards or other measures to implement
the CO2 emission performance rates, which may be accomplished by meeting the state goals
1-6
-------
1.4.5 State Plans
After the EPA establishes the emission guidelines that set forth the BSER, each state23
shall then develop, adopt and submit a state plan under CAA section 11 l(d) that establishes
standards of performance for the affected EGUs in its jurisdiction in order to implement the
BSER. The final guidelines include three approaches that states may adopt for purposes of
implementing the BSER, any one of which a state may use in its plan. These are: 1) establishing
standards of performance that apply the subcategory specific COi emission performance rates to
their affected EGUs, 2) adopting a combination of standards and/or other measures that achieve
state-specific rate-based goals that represent the weighted aggregate of the COi emission
performance rates applied to the affected EGUs in each state, and 3) adopting a program to meet
mass-based CO2 emission goals that represent the equivalent of the rate-based goal for each
state. These alternatives, as well as the other options we are finalizing, ensure that both states and
affected EGUs enjoy the maximum flexibility and latitude in meeting the requirements of the
emission guidelines and that the BSER is fully implemented by each state.
1.5 Organization of the Regulatory Impact Analysis
This report presents the EPA's analysis of the potential benefits, costs, and other
economic effects of the final emission guidelines to fulfill the requirements of an RIA. This RIA
includes the following chapters:
Chapter 2, Electric Power Sector Industry Profile
Chapter 3, Cost, Emissions, Economic, and Energy Impacts
Chapter 4, Estimated Climate Benefits and Health Co-benefits
Chapter 5, Economic Impacts - Markets Outside the Electricity Sector
Chapter 6, Employment Impact Analysis
Chapter 7, Statutory and Executive Order Analyses
Chapter 8, Comparison of Benefits and Costs
23 In this section, the term "state" encompasses the 48 contiguous states and the District of Columbia, and any Indian tribe that has been approved by the EPA pursuant to 40 CFR
49.9 as eligible to develop and implement a CAA section 11 l(d) plan.
1-7
-------
1.6 References
40 CFR Chapter I [EPA-HQ-OAR-2009-0171; FRL-9091-8] RIN 2060-ZA14,
"Endangerment and Cause or Contribute Findings for Greenhouse Gases under Section 202(a)
of the Clean Air Act," Federal Register / Vol. 74, No. 239 / Tuesday, December 15, 2009 /
Rules and Regulations.
75 FR 49556. August 13, 2010. "EPA's Denial of the Petitions to Reconsider the Endangerment
and Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the Clean
Air Act."
Melillo, J.M., T.C. Richmond, and G.W. Yohe, Eds., 2014: Climate Change Impacts in the
United States: The Third National Climate Assessment. U.S. Global Change Research
Program. Available at . Accessed June 4, 2015.
National Research Council. Climate Stabilization Targets: Emissions, Concentrations, and
Impacts over Decades to Millennia. Washington, DC: The National Academies Press, 2011.
U.S. Environmental Protection Agency. EPA's Power Sector Modeling Platform v.5.14. March
25, 2015. Available online at .
Accessed June 4, 2015.
1-8
-------
CHAPTER 2: ELECTRIC POWER SECTOR INDUSTRY PROFILE
2.1 Introduction
This chapter discusses important aspects of the power sector that relate to the Final Carbon
Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units,
including the types of power-sector sources affected by the regulation, and provides background
on the power sector and EGUs. In addition, this chapter provides some historical background on
trends in the past decade in the power sector, as well as about existing EPA regulation of the
power sector.
In the past decade there have been significant structural changes in the 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
US economy, growth and regional changes in the US 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, in particular the increased natural gas supply and
subsequent relatively low natural gas prices, have resulted in more gas being utilized as base load
energy in addition to supplying electricity during peak load. This chapter presents data on the
evolution of the power sector from 2002 through 2012. Projections of new capacity and the
impact of this rule on these new sources 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.
2.2.7 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 from an EGU in a typical hour, typically
2-1
-------
measured in megawatts (MW) or gigawatts (1 GW = 1000 MW). Electricity Generation refers to
the amount of electricity actually produced by EGUs, measured in kilowatt-hours (kWh) or
gigawatt-hours (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). In addition to producing electricity for sale to the
grid, generators 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
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 2002 and 2012.
2-2
-------
In 2012 the power sector consisted of over 19,000 generating units with a total capacity24
of 1,168 GW, an increase of 188 GW (or 19 percent) from the capacity in 2002 (980 GW). The
188 GW increase consisted primarily of natural gas fired EGUs (134 GW) and wind generators
(55 GW), with substantially smaller net increases and decreases in other types of generating
units.
Table 2-1. Existing Electricity Generating Capacity by Energy Source, 2002 and
2012
Energy Source
Coal
Natural Gas1
Nuclear
Hydro
Petroleum
Wind
Other
Renewable
Misc
Total
2002
Generator
Nameplate
Capacity
(MW)
338,199
352,128
104,933
96,344
66,219
4,531
14,208
3,023
979,585
% Total
Capacit
y
35%
36%
11%
10%
7%
0.5%
1.5%
0.3%
100%
2012
Generator
Nameplate
Capacity
(MW)
336,341
485,957
107,938
99,099
53,789
59,629
20,986
4,257
1,167,995
% Total
Capacit
y
29%
42%
9%
8%
5%
5.1%
1.8%
0.4%
100%
Change Between '02
%
Increas
e
-1%
38%
3%
3%
-19%
1216%
47.7%
40.8%
19%
Nameplate
Capacity
Change
(MW)
-1,858
133,829
3,005
2,755
-12,430
55,098
6,778
1,234
188,410
and '12
%of
Total
Capacity
Increase
-1%
71%
2%
1%
-7%
29%
3.6%
0.7%
100%
Note: This table presents generation capacity. Actual net generation is presented in Table 2-2.
Source: U.S. EIA. Downloaded from EIA Electricity Data Browser, Electric Power Plants Generating Capacity By
energy source, by producer, by state back to 2000 (annual data from EIA Form 860). Available online at:
Accessed 12/19/2014
1 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 (31 percent of 2012
natural gas-fired capacity), Gas Turbine (30 percent), Combined Cycle Steam (19 percent), Steam Turbine (17
percent), and miscellaneous (< 1 percent).
24 As with all data presented in this section, this includes generating capacity not only at EGUs primarily operated to
supply electricity to the grid, but also generating capacity at commercial and industrial facilities that produce both
electricity used onsite as well as dispatched to the grid. Unless otherwise indicated, capacity data presented in this
RIA is installed nameplate capacity (also known as nominal capacity), defined by EIA as "The maximum rated
output of a generator, prime mover, or other electric power production equipment under specific conditions
designated by the manufacturer." Nameplate capacity is consistently reported to regulatory authorities with a
common definition, where alternate measures of capacity (e.g., net summer capacity and net winter capacity) can
use a variety of definitions and specified conditions.
2-3
-------
The 19 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
capacity of individual existing units due to changes in operating equipment, changes in emission
controls, etc. During the period 2002 to 2012, a total of 315,752 MW of new generating capacity
was built and brought online, and 64,763 MW existing units were retired. The net effect of the
re-rating of existing units reduced the total capacity by 62,579 MW. The overall net change in
capacity was 188,410 MW, as shown in Table 2-1.
The newly built generating capacity was primarily natural gas (226,605 MW), which was
partially offset by gas retirements (29,859 MW). Wind capacity was the second largest type of
new builds (55,583 MW), augmented by 2,807 MW of solar.25 The overall mix of newly built
and retired capacity, along with the net effect, is shown on Figure 2-1.
ICoal Nat Gas WindS Solar BON& Other
Figure 2-1. New Build and Retired Capacity (MW) by Fuel Type, 2002-2012
Source: EIA Form 860
Not displayed: wind and solar retirements = 87 MW, net change in coal capacity = -56 MW
' Partially offset by 87 MW retired older wind or solar capacity.
2-4
-------
In 2012, electric generating sources produced a net 4,058 trillion kWh to meet electricity
demand, a 5 percent increase from 2002 (3,858 trillion kWh). As presented in Table 2-2, almost
70 percent of electricity in 2012 was produced through the combustion of fossil fuels, primarily
coal and natural gas, with coal accounting for the largest single share. Although the share of the
total generation from fossil fuels in 2012 (67 percent) was only modestly smaller than the total
fossil share in 2002 (71 percent), the mix of fossil fuel generation changed substantially during
that period. Coal generation declined by 18 percent and petroleum generation by 72 percent,
while natural gas generation increased by 60 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. Wind generation also grew from a very small portion of the overall
total in 2002 to 4.1 percent of the 2012 total.
Table 2-2. Net Generation in 2002 and 2013 (Trillion kWh = TWh)
Coal
Natural Gas
Nuclear
Hydro
Petroleum
Wind
Other Renewable
Misc
Total
2002
Net
Generation
(TWh)
1,933.1
702.5
780.1
255.6
94.6
10.4
68.8
13.5
3,858
Fuel
Source
Share
50%
18%
20%
7%
2.5%
0.3%
1.8%
0.4%
100%
2013
Net
Generation
(TWh)
1,514.0
1,237.8
769.3
271.3
23.2
140.8
77.5
12.4
4,046
Fuel
Source
Share
37%
31%
19%
7%
0.6%
3.5%
1.9%
0.3%
100%
Change Between '02 and '13
Net
Generation
Change
(TWh)
-419.1
535.3
-10.7
15.7
-71.4
130.5
8.8
-1.2
188
% Change in
Net
Generation
-21.7%
76.2%
-1.4%
6.1%
-75.5%
1260.0%
12.7%
-8.7%
5%
Source: U.S. EIA Monthly Energy Review, December 2014. Table 7.2a Electricity Net Generation: Total (All
Sectors). Available online at: . Accessed 12/19/2014
Coal-fired and nuclear generating units have historically supplied "base load" electricity,
the portion of electricity loads which are continually present, and typically operate throughout all
hours of the year. The coal units meet the part of demand that is relatively constant. 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
2-5
-------
compose 37 percent of the total number of coal-fired units, but only 6 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.
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 55 percent of the coal EGU fleet is over 500 MW per unit, 77 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.
2-6
-------
Table 2-3. Coal and Natural Gas Generating Units, by Size, Age, Capacity, and
Thermal Efficiency (Heat Rate)
Unit Size
Grouping
(MW)
No.
Units
% of All
Units
Avg.
Age
Avg. Net
Summer
Capacity
(MW)
Total Net
Summer Avg. Heat
Capacity % Total Rate
(MW) Capacity (Btu/kWh)
COAL
0-24
25-49
50-99
100 - 149
150 - 249
250 - 499
500 - 749
750 - 999
1000 - 1500
Total Coal
223
108
157
128
181
205
187
57
11
1257
18%
9%
12%
10%
14%
16%
15%
5%
1%
100%
40.7
44.2
49.0
50.6
48.7
38.4
35.4
31.4
35.7
42.6
11.4
36.7
74.1
122.7
190.4
356.2
604.6
823.9
1259.1
250.7
2,538
3,963
11,627
15,710
34,454
73,030
113,056
46,963
13,850
315,191
1%
1%
4%
5%
11%
23%
36%
15%
4%
100%
11,733
11,990
11,883
10,971
10,620
10,502
10,231
9,942
9,732
11,013
NATURAL GAS
0-24
25-49
50-99
100 - 149
150 - 249
250 - 499
500 - 749
750 - 1000
Total Gas
1992
410
962
802
167
982
37
14
5366
37%
8%
18%
15%
3%
18%
1%
0.3%
100%
37.6
21.8
15.6
23.4
28.7
24.6
40.0
35.9
27.7
7.0
125.0
174.2
39.9
342.4
71.1
588.8
820.9
79.2
13,863
51,247
167,536
31,982
57,179
69,788
21,785
11,492
424,872
3%
12%
39%
8%
13%
16%
5%
3%
100%
13,531
9,690
8,489
11,765
9,311
12,083
11,569
10,478
11,652
Source: National Electric Energy Data System (NEEDS) v.5.14
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 2013 or earlier, and excludes those units in NEEDS
with planned retirements in 2014 or 2015.
In terms of the age of the generating units, 50 percent of the total coal generating capacity
has been in service for more than 38 years, while 50 percent of the natural gas capacity has been
in service less than 15 years. Figure 2-2 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 2-2 also includes the distribution of generation.
2-7
-------
- Gas Gen
Figure 2-2. Cumulative Distribution in 2010 of Coal and Natural Gas Electricity
Capacity and Generation, by Age
Source: National Electric Energy Data System (NEEDS) v.5.13
Not displayed: coal units (376 MW total, 1 percent of total) and gas units (62 MW, < .01 percent of total)) over 70
years old for clarity. Figure is limited to coal-steam units in NEEDS v5.13 in operation in 2013 or earlier (excludes
-2,100 MW of coal-fired IGCC and fossil waste capacity), and excludes those units in NEEDS with planned
retirements in 2014 or 2015.
The locations of existing fossil units in EPA's National Electric Energy Data System
(NEEDS) v.5.13 are shown in Figure 2-3.
2-8
-------
Facility Capacity (MW)
010100
100 to 500
500 to 1.000
1,000 to 2,000
2,000 to 3,700
Figure 2-3. Fossil Fuel-Fired Electricity Generating Facilities, by Size
Source: National Electric Energy Data System (NEEDS) v.5.13
Note: This map displays fossil capacity at facilities in the NEEDS v.5.13 IPM frame. NEEDS v.5.13 reflects
generating capacity expected to be on-line at the end of 2015. 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,26 each operating synchronously. Within each of these transmission
26 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 Interconnection,
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
http://www.nerc.com/AboutNERC/keyplayers/Documents/NERC_Interconnections_Color_072512.jpg
2-9
-------
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
regional operator27; in others, individual utilities28 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, the transmission system had been developed by vertically integrated
utilities, establishing 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
27 E.g., PMJ Interconnection, LLC, Western Area Power Administration (which comprises 4 sub-regions).
28 E.g., Los Angeles Department of Power and Water, Florida Power and Light.
2-10
-------
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 commercial, industrial and
residential ultimate customers. Each of the three major ultimate categories consume roughly a
quarter to a third of the total electricity produced29 (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 2002 and 2012.
Table 2-4. Total U.S. Electric Power Industry Retail Sales in 2012 (billion kWh)
Sales
Residential
Commercial
Industrial
Transportation
Other
Total
Direct Use
Total End Use
2002
Sales/Direct
Use (Billion Share of Total
kWh)
1,265
1,104
990
NA
106
3,465
166
3,632
End Use
35%
30%
27%
3%
95%
5%
100%
2012
Sales/Direct
Use (Billion
kWh)
1,375
1,327
986
7
NA
3,695
138
3,832
Share of Total End
Use
35.9%
34.6%
25.7%
0.2%
96%
4%
100%
Source: Table 2.2, EIA Electric Power Annual, 2013
Notes: Retail sales are not equal to net generation (Table 2-2) because net generation includes net exported
electricity and loss of electricity that occurs through transmission and distribution.
Direct Use represents commercial and industrial facility use of onsite net electricity generation; and
electricity sales or transfers to adjacent or co-located facilities for which revenue information is not
available.
2.3.1 Electricity Prices
Electricity prices vary substantially across the United States, differing both between the
ultimate customer categories and also by state and region of the country. Electricity prices are
typically highest for residential and commercial customers because of the relatively high costs of
29 Transportation (primarily urban and regional electrical trains) is a fourth ultimate customer category which
accounts less than one percent of electricity consumption.
2-11
-------
distributing electricity to individual homes and commercial establishments. The high 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
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 considerable 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 2011 the national
average retail electricity price (all sectors) was 9.90 cents/KWh, with a range from 6.44 cents
(Idaho) to 31.59 (Hawaii). The Northeast, California and Alaska have average retail prices that
can be as much as double those of other states (see Figure 2-4), and Hawaii has the most
expensive retail price of electricity in the country.
2-12
-------
Average Price (cents per kilowatthour)
^] 6.44 - 7.80
HI 7.88-8.78
| | 8 80 - 9.39
| | 9.61 -12.81
I I 13.04-31.59
Note: Data are displayed as 5 groups of lOStates and the District of Columbia.
U.S. total average price per kilowatthour is 9.90 cents.
Source: U.S. Energy Information Administration, Annual Energy Review -
Electricity Section, Table 4, September 27,2012.
Figure 2-4. Average Retail Electricity Price by State (cents/kWh), 2011
Average national overall retail electricity prices increased between 2002 and 2012 by 36.7
percent in nominal (current year $) terms. The amount of increase differed for the three major
end use categories (residential, commercial and industrial). National average residential prices
increased the most (40.8 percent), and commercial prices increased the least (27.9 percent). The
nominal year prices for 2002 through 2012 are shown in Figure 2-5.
2-13
-------
14.0
2002 2004
^^ Residential
2006 2008 2010 2012
Commercial ^^^Industrial Total
Figure 2-5. Nominal National Average Electricity Prices for Three Major End-Use
Categories
Source: EIA AEO 2012, Table 2.4
Electricity prices for all three end-use categories increased more than overall inflation
through this period, measured by either the GDP implicit price deflator (23.5 percent) or the
consumer price index (CPI-U, which increased by 27.7 percent)30. Most of these electricity price
increases occurred between 2002 and 2008; since 2008 nominal electricity prices have been
relatively stable while overall inflation continued to increase. The increase in nominal electricity
prices for the major end use categories, as well as increases in the GDP price and CPI-U indices
for comparison, are shown in Figure 2-6.
30
Source: Federal Reserve Economic Data, FRB St. Louis. Available online at: http://research.stlouisfed.org/fred2/.
2-14
-------
45%
0%
2004
2006
Residential
Commercial
Industrial
CPI-U
. GDP Price
Figure 2-6. Relative Increases in Nominal National Average Electricity Prices for Major
End-Use Categories, With Inflation Indices
The real (inflation-adjusted) change in average national electricity prices can be calculated
using the GDP implicit price deflator. Figure 2-7 shows real31 (2011$) electricity prices for the
three major customer categories from 1960 to 2012, and Figure 2-8 shows the relative change in
real electricity prices relative to the prices in 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
2012 was relatively unchanged from 1960, residential and commercial real prices are 23 percent
and 28 percent lower respectively than in 1960.
31 All prices in this section are estimated as real 2011 prices adjusted using the GDP implicit price deflator unless
otherwise indicated.
2-15
-------
Real Electricity Prices, 1960-2014 (including taxes)
I960 1970
Residential
1980 1990 2000
Commercial ^^^ Industrial
2010
Figure 2-7. Real National Average Electricity Prices (2011$) for Three Major End-Use
Categories
Source: EIA Monthly Energy Review, April 2015, Table 9.8
Relative Change in Electricity Prices,
1960-2014 (including taxes)
-50%
1960
1970
Residential
1980
Commercial
1990
2000
Industrial
2010
Total
2-16
-------
Figure 2-8. Relative Change in Real National Average Electricity Prices (2011$) for
Three Major End-Use Categories
Source: EIA Monthly Energy Review, April 2015, Table 9.8
2.3.2 Prices of Fossil Fuels Used for Generating Electricity
Another important factor in the changes in electricity prices are the changes in fuel prices
for the three major fossil fuels used in electricity generation; coal, natural gas and oil. Relative to
real prices in 2002, the national average real price (in 2011$) of coal delivered to EGUs in 2012
had increased by 54 percent, while the real price of natural gas decreased by 22 percent. The real
price of oil increased by 203 percent, but with oil declining as an EGU fuel (in 2012 oil
generated only 1 percent of electricity) the doubling of oil prices had little overall impact in the
electricity market. The combined real delivered price of all fossil fuels in 2012 increased by 23
percent over 2002 prices. Figure 2-9 shows the relative changes in real price of all 3 fossil fuels
between 2002 and 2012.
-50%
2002
2004
Coal
2006
Oil
2008
Gas
2010
Average
2012
Figure 2-9. Relative Real Prices of Fossil Fuels for Electricity Generation; Change in
National Average Real Price per MBtu Delivered to EGU
Source: EIA AEO 2012, Table 9.9
2-17
-------
2.3.3 Changes in Electricity Intensity of the U.S. Economy Between 2002 to 2012
An important aspect of the changes in electricity generation (i.e., electricity demand)
between 2002 and 2012 is that while total net generation increased by 4.9 percent over that
period, the demand growth for generation has been low, and in fact was lower than both the
population growth (9.2 percent) and real GDP growth (19.8 percent). Figure 2-10 shows the
growth of electricity generation, population and real GDP during this period.
25%
20%
15%
10%
0%
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Real GDP Population Generation
Figure 2-10. Relative Growth of Electricity Generation, Population and Real GDP Since
2002
Sources: U.S. EIA Monthly Energy Review, December 2014. Table 7.2a Electricity Net Generation: Total (All
Sectors). U.S. Census.
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 dollar of output) during 2002 to 2012. On a per capita basis, real GDP per
capita grew by 10.9 percent, increasing from $44,900 (in 2011$) per person in 2002 to
$49,800/person in 2012. At the same time electricity generation per capita decreased by 3.9
percent, declining from 13.4 MWh/person in 2002 to 12.8 MWh/person in 2012. The combined
effect of these two changes improved the overall electricity efficiency of the U.S. market
economy. Electricity generation per dollar of real GDP decreased 12.5 percent, declining from
299 MWh per $1 million of GDP to 261 MWh/$l million GDP. These relative changes are
shown in Figure 2-11. Figures 2-10 and 2-11 clearly show the effects of the 2007 - 2009
2-18
-------
recession on both GDP and electricity generation, as well as the effects of the subsequent
economic recovery.
15%
-15%
2002 2003 2004 2005
Real GDP/Capita
2006 2007 2008 2009 2010 2011
Generation/Capita Generation/ Real GDP
2012
Figure 2-11. Relative Change of Real GDP, Population and Electricity Generation
Intensity Since 2002
Sources: U.S. EIA Monthly Energy Review, December 2014. Table 7.2a Electricity Net Generation: Total (All
Sectors). U.S. Census
2.4 Deregulation and Restructuring
The process of restructuring and deregulation of wholesale and retail electric 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
2-19
-------
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
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.
2-20
-------
Electricity Restructuring by State
IJ Not Active
Active
Suspended
Figure 2-12. Status of State Electricity Industry Restructuring Activities
Source: EIA 2010. "Status of Electricity Restructuring by State." Available online at:
.
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 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 (IPP). The resulting generating
sector differed by state or region, as the power sector adapted to the restructuring and
deregulation requirements in each state.
By 2002 the major impacts of adapting to changes brought about by deregulation and
restructuring during the 1990s were largely in place. The resulting ownership mix of generating
capacity (MW) in 2002 was 62 percent of the generating capacity owned by traditional utilities,
2-21
-------
35 percent owned by IPPs32, and 3 percent owned by commercial and industrial producers. The
mix of electricity generated (MWh) was more heavily weighted towards the utilities, with a
distribution in 2002 of 66 percent, 30 percent and 4 percent for utilities, IPPs and
commercial/industrial, respectively.
Since 2002 IPPs have expanded faster than traditional utilities, substantially increasing
their share by 2012 of both capacity (58 percent utility, 39 percent IPPs, and 3 percent
commercial/industrial) and generation (58 percent, 38 percent and 4 percent).
The mix of capacity and generation for each of the ownership types is shown in Figures 2-
13 (capacity) and 2-14 (generation). The capacity and generation data for commercial and
industrial owners are not shown on these figures due to the small magnitude of those ownership
types. Figures 2-13 and 2-14 present the mixes in 2002 and 2012. 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.
: IPP data presented in this section include both combined and non-combined heat and power plants.
2-22
-------
Capacity Mix, 2002 & 2012
700,000
600,000
500,000
-, 400,000
300,000
200,000
100,000
ii:,
Generation Mix, 2002 & 2012
3,000,000
2,500,000
2,000,000
(D
.2 1,500,000
(D
1,000,000
2002- 2012-
Utility Utility
2002-IPP 2012-IPP
I Nuclear BCoal Gas Hydro Wind BAH Other
500,000
ll,
I
2002- 2012-
Utility Utility
2002-IPP 2012-IPP
I Nuclear BCoal Gas Hydro Wind BAN Other
Figures 2-13 and 2-14. Capacity and Generation Mix by Ownership Type, 2002 &
2012
The mix of capacity by fuel types that have been built and retired between 2002 and 2012
also varies significantly by type of ownership. Figure 2-15 presents the new capacity built during
that period, showing that IPPs built the majority of both new wind and solar generating capacity,
as well as somewhat more natural gas capacity than the traditional utilities built. Figure 2-16
presents comparable data for the retired capacity, showing that utilities retired more coal and
"other" capacity (mostly oil-fired) than IPPs retired, while the IPPs retired more natural gas
capacity than the utilities retired. The retired gas capacity was primary (60 percent) steam and
combustion turbines.
2-23
-------
Capacity Built 2002-2012
by Ownership Type
I Coal
Utility IPP
IGas WindS Solar HOther
Capacity Retirements 2002-2012
by Ownership Type
I Coal
Utility IPP
IGas WindS Solar HOther
Figures 2-15 and 2-16.
Generation Capacity Built and Retired between 2002 and 2012
by Ownership Type
2.5 Emissions of Greenhouse Gases from Electric Utilities
The burning of fossil fuels, which generates about 69 percent of our electricity nationwide,
results in emissions of greenhouse gases. The power sector is a major contributor of COi in
particular, but also contributes to emissions of sulfur hexafluoride (SFe), CH4, and NiO. In 2012,
the electricity generation accounted for 38 percent of national COi emissions. Including both
generation and transmission (a source of SF6), the power sector accounted for 31 percent of total
nationwide greenhouse gas emissions, measured in COi equivalent. Table 2-5 and Figure 2-17
show the GHG emissions33 from the power sector relative to other major economic sectors.
Table 2-6 shows the contributions of COi and other GHGs from the power sector and other
major emitting economic sectors.
33 COi equivalent data in this section are calculated with the IPCC SAR (Second Assessment Report) GWP potential
factors.
2-24
-------
Table 2-5. Domestic Emissions of Greenhouse Gases, by Economic Sector (million tons of
CCh equivalent)
Sector/Source
Electric Power Industry
Transportation
Industry
Agriculture
Commercial
Residential
US Territories
Total GHG Emissions
Sinks and Reductions
Net GHG Emissions
2002
GHG
Emissions
2,550
2,158
1,564
618
402
412
58
7,762
% Total
GHG
Emissions
33%
28%
20%
8%
5%
5%
<1%
100%
-976
6,786
2013
GHG
Emissions
2,289
1,991
1,535
647
442
413
38
7,356
% Total
GHG
Emissions
31%
27%
21%
9%
6%
6%
<1%
100%
-972
6,384
Change
Change
in
Emissions
-260
-167
-29
29
40
1
-19
-406
4
-402
Between '02
%
Change in
Emissions
-10%
-8%
-2%
5%
10%
0%
-33%
-5%
0%
-6%
and '13
%of
Total
Change
in
Emissions
64%
41%
7%
-7%
-10%
0%
5%
100%
Source: EPA, 2014 "Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2012", Table 2-12. Includes
COi, CH4, NiO and SF6 emissions.
2002 2013
I Electric Power Industry Transportation Industry
(Agriculture Commercial Residential
Figure 2-17. Domestic Emissions of Greenhouse Gases from Major Sectors, 2002 and 2013
(million tons of CCh equivalent)
Source: EPA, 2015 "Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2013", Table 2-12.
Not Shown: CO2e emissions from US Territories
2-25
-------
The amount of CO2 emitted during the combustion of fossil fuels varies according to the
carbon content and heating value of the fuel used. The CO2 emission factors used in IPM v5.14
(same as used in v5.13) are shown in Table 2-7. Coal has higher carbon content than oil or
natural gas, and thus releases more COi during combustion. Coal emits around 1.7 times as much
carbon per unit of energy when burned as natural gas (EPA 2013).
Table 2-6. Greenhouse Gas Emissions from the Electricity Sector (Generation,
Transmission and Distribution), 2002 and 2012 (million tons of CCh equivalent)
Gas/Fuel Type or Source
CCh
CH4
N2O
SF6
Fossil Fuel
Combustion
Coal
Natural Gas
Petroleum
Geothermal
Incineration of
Waste
Other Process Uses
of Carbonates
Stationary
Combustion*
Incineration of
Waste
Stationary
Combustion*
Incineration of
Waste
Electrical
Transmission and
Distribution
Total GHG Emissions
2002
GHG
Emissions
%of
Total
GHG
Emissions
2,521
2,505
2,083
337
84.7
0.4
13.0
2.9
0.4
0.4
+
13.7
13.2
0.4
14.7
14.7
from
Power
Sector
98.9%
98.2%
81.7%
13.22%
3.32%
0.02%
0.51%
0.11%
0.02%
0.02%
0.54%
0.52%
0.02%
0.57%
0.57%
2,550
2013
GHG
Emissions
2,262
2,248
1,736
487
24.7
0.4
11.1
2.4
0.4
0.4
+
21.4
21.1
0.3
5.6
5.6
% of Total
GHG
Emissions
from Power
Sector
98.8%
98.2%
75.8%
21.28%
1.08%
0.02%
0.49%
0.11%
0.02%
0.02%
0.93%
0.92%
0.01%
0.25%
0.25%
2,289
Change Between '02
and '13
Change in
GHG
Emissions
-259
-257
-347
150
-60.0
0.0
-1.9
-0.4
0.0
0.0
7.7
7.8
-0.1
-9.0
-9.0
% Change
in
Emissions
-10%
-10%
-17%
45%
-71%
0%
-14%
-15%
0%
0%
56%
59%
-25%
-62%
-62%
-260
Source: EPA, 2015 "Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2015", Table 2-11
* Includes only stationary combustion emissions related to the generation of electricity.
** SF6 is not covered by this rule, which specifically regulates GHG emissions from combustion.
+ Does not exceed 0.05 Tg COi Eq. or 0.05 percent.
2-26
-------
Table 2-7. Fossil Fuel Emission Factors in EPA Base Case 5.14 IPM Power Sector
Modeling Application
Fuel Type
Coal
Bituminous
Subbituminous
Lignite
Natural Gas
Fuel Oil
Distillate
Residual
Biomass
Waste Fuels
Waste Coal
Petroleum Coke
Fossil Waste
Non-Fossil Waste
Tires
Municipal Solid Waste
Carbon Dioxide (Ib/MMBtu)
202.8 - 209.6
209.2-215.8
212.6-219.
117.1
161.4
161.4-173.9
195
204.7
225.1
321.1
0
189.5
91.9
Source: Documentation for IPM Base Case v.5.13, Table 11-5. The emission factors used in Base Case 5.14 are
identical to the emission factors in IPM Base Case 5.13.
Note: COi emissions presented here for biomass account for combustion only and do not reflect lifecycle
emissions from initial photosynthesis (carbon sink) or harvesting activities and transportation (carbon
source).
2.6 Carbon Dioxide Control Technologies
In the power sector, current approaches available for significantly reducing the CO2
emissions of new fossil fuel combustion sources to meet a 1,400 Ib COi/MWh emission rate
include the use of: (1) highly efficiency coal-fired designs (e.g., modern supercritical or ultra-
supercritical steam units) with up to 40 percent natural gas co-firing, (2), integrated coal
gasification combined cycle (IGCC) with < 10 percent CCS or co-firing with up to 10 percent
natural gas, (3) natural gas combined cycle (NGCC) combustion turbine/steam-turbine units,
and/or (4) conventional coal-fired generation with carbon capture and storage (CCS). While CCS
is not included in the BSER framework, it is an emerging technology with both new build and
retrofit commercial-scale EGUs coming into operation in 2014 and 2015 in the United States and
Canada. All of these units with CCS have received substantial subsidies to further develop and
demonstrate the feasibility of CCS at a commercial scale, and the costs of these new units with
2-27
-------
CCS are not indicative of anticipated future costs of new or retrofit CCS units. CCS is briefly
discussed in this section as existing (but still emerging) technology that may become
economically viable in the future.
Investment decisions for the optimal choice in a particular situation of the type of new
generating capacity capable of meeting the 1,400 Ib COi/MWh standard of performance depend
in part on the intended primary use of new generating capacity. Daily peak electricity demands,
involving operation for relatively few hours per year, are often most economically met by
simple-cycle combustion turbines (CT). Stationary CTs used for power generation can be
installed quickly, at relatively low capital cost. They can be remotely started and loaded quickly,
and can follow rapid demand changes. Full-load efficiencies of large current technology CTs are
typically 30-33 percent but can be has high as 40 percent or more (high heating value basis), as
compared to efficiencies of 50 percent or more for new combined-cycle units that recover and
use the exhaust heat otherwise wasted from a CT . A simple-cycle CT's lower efficiency causes
it to burn much more fuel to produce a MWh of electricity than a combined-cycle unit. Thus,
when burning natural gas its COi emission rate per MWh could be 40-60 percent higher than a
more efficient NGCC unit.
Base load electricity demand can be met with NGCC generation, coal and other fossil-fired
steam generation, and IGCC technology, as well as generation from sources that do not emit
COi, such as nuclear and hydro. IGCC employs the use of a gasifier to transform fossil fuels into
synthesis gas ("syngas") and heat. The syngas is used to fuel a combined cycle generator, and the
heat from the syngas conversion can produce steam for the steam turbine portion of the
combined cycle generator. Electricity can be generated through this IGCC process somewhat
more efficiently than through conventional boiler-steam generators. Additionally, with
gasification, some of the syngas can be converted into other marketable products such as
fertilizers and chemical feedstocks for Fisher-Tropsch processes to manufacture liquid
hydrocarbons (e.g., fuels and lubricants), and COi can be captured for use in FOR. Figure 2-18
shows the array of products (including electricity) and by-products that can be produced in a
syngas process (NETL).
2-28
-------
Fischer
Tropsch
Synthesis
Figure 2-18. Marketable products from Syngas Generation
Source: National Energy Technology Lab. Gasifipedia. Available online at:
http://www.netl.doe.gov/research/coal/energy-systems/gasification/gasifipedia/co-generation
2.6.1 Carbon Capture and Storage
CCS can be achieved through either pre-combustion or post-combustion capture of CCh
from a gas stream associated with the fuel combusted. Furthermore, CCS can be designed and
operated for full capture of the COi in the gas stream (i.e., above 90 percent) or for partial
capture (below 90 percent). Post-combustion capture processes remove COi from the exhaust
gas of a combustion system - such as a utility boiler. It is referred to as "post-combustion
capture" because the COi is the product of the combustion of the primary fuel and the capture
takes place after the combustion of that fuel. This process is described in more detail in the
preamble. (See preamble section V.D.) This process is illustrated for a pulverized coal power
plant in Figure 2-19. For post-combustion, a station's net generating output will be lower due to
the energy needs of the capture process.
2-29
-------
Flue Gas
Flue Gas
Volume %
CO2 12-14%
N2 -65%
H2O -18%
Air
Coal
CC>2 To Storage
Figure 2-19. Post-Combustion CCh Capture for a Pulverized Coal Power Plant
Source: Interagency Task Force on Carbon Capture and Storage 2010
Pre-combustion capture is mainly applicable to IGCC facilities, where the fuel is converted
into gaseous components ("syngas") under heat and pressure and some percentage of the carbon
contained in the syngas is captured before combustion.34 For pre-combustion technology, a
significant amount of energy is needed to gasify the fuel(s). This process is illustrated in Figure
2-20. Application of post-combustion CCS with IGCC can be designed to use no water-gas shift,
or single- or two-stage shift processes, to obtain varying percentages of COi removal - from a
"partial capture" percentage to 90 percent "full capture." Pre-combustion CCS typically has a
lesser impact on net energy output than does post-combustion CCS. For more detail on CCS
technology, see the "Report of the Interagency Task Force on Carbon Capture and Storage"
(2010).35
34 Note that pre-combustion CCS is not considered the best system of emission reduction for this standard. This
information is provided for background purposes.
35 For more information on the cost and performance of CCS, see http://www.netl.doe.gov/energy-
analyses/baseline_studies.html.
2-30
-------
Shifted Syngas
Hfe -50%
Oxygen
Coal
- Power Block
Power«
Steam
Turbine
-
Steam
Generator
-
Combustion
Turbines)
CO:, to Storage
i Power
Figure 2-20. Pre-Combustion CCh Capture for an IGCC Power Plant
Source: Interagency Task Force on Carbon Capture and Storage 2010
Carbon capture technology has been successfully applied since 1930 on several smaller
scale industrial facilities and more recently in a number of demonstration phase projects
worldwide for power sector applications. In October 2014 the first commercial-scale coal-fired
capture and storage project for electricity generation began operation at the Boundary Dam
Power Station in Saskatchewan, Canada. The Boundary Dam Station is owned by the Province
of Saskatchewan, and operated by SaskPower, a provincially owned corporation that is the
primary electric utility in the Province. The commercial-scale demonstration project retrofit Unit
3 (a 130 MW, coal fired built in 1970, and rebuilt in 2013) at a total cost of approximately $1.5
billion (Canadian, or about $1.2 billion US), including a partial subsidy of $240 million
(Canadian) by the Canadian federal government. The carbon capture system is a post-
combustion process designed to capture 90 percent of the COi emitted by Unit #3. Retrofitting
the carbon capture system reduced the capacity of the unit to 110 MW. The majority of the
captured COi is used for an EOR project in southern Saskatchewan. The portion of the COi is
being stored in a nearby research and monitoring geological storage facility, where the captured
COi will be injected 3.4 kilometers underground into a sandstone formation located below the
2-31
-------
major coal field supplying lignite to Unit # 3. The remaining captured CO2 will be injected into
deep saline formations.
In the United States, there are two commercial-scale CCS facilities nearing completion:
1. the Kemper County Carbon Dioxide Capture and Storage Project in Mississippi, and
2. The W.A. Parish Petra Nova CCA Project near Houston, Texas.
Construction began on the Kemper project in 2010, and the startup is currently scheduled
for May, 2016. The Kemper project is constructing a new 524 MW lignite unit as well as a 58
MW natural gas unit. Mississippi Power (a division of Southern Power) is building and will
operate the Kemper County project. The control system is designed to capture 65 percent of the
COi generated by the plant, and is projected to capture 3.5 million tons of CO2 per year. The
resulting COi emission rate is expected to be about 800 pounds per MWh produced. The current
total cost estimate is $5.6 billion, a substantial increase from the original $2.4 billion estimate.36
The construction has received a $270 million grant from the US Department of Energy, and $133
million in investment tax credits from the Internal Revenue Service. The captured CO2 will be
transported via a 60 mile pipeline and used for EOR projects in mature Mississippi oil fields.37
The only other commercial-scale electricity power sector CCS project currently under
construction in the United States is the W.A. Parish Petra Nova CCS Project near Houston,
Texas. The Parish Petra project is a 50/50 partnership between NRG Energy (an integrated
electricity company generating and supplying electricity to 1.6 million customers in Texas) and
the Nippon Oil and Gas Exploration Company. The Parish project will retrofit a post-combustion
CCS system on a portion of the flue gas from the existing 610 MW coal fired Unit # 8. The CCS
system will treat a 240 MW slipstream of the flue gas, and is designed to capture 90 percent of
the COi in the treated flue gas. The capacity rating of Unit # 8 will not be reduced due to the
36 The Mississippi Public Utilities Staff authorized an independent monitor to conduct a review of the project. The
findings of the review are provided in a summary report Available online at::
http://www.psc.state.ms.us/InsiteConnect/InSiteView.aspx?model=INSITE_CONNECT&queue=CTS_ARCHIVEQ
&docid=328417
37 Carbon Capture and Sequestration Technologies Program at MIT. Accessed 1/23/2015.
https://sequestration.mit.edu/tools/projects/kemper.html
2-32
-------
CCS project because an 85 MW custom-built natural gas fired combustion turbine co-generation
unit is being built on-site to provide both electricity and steam to the CCS unit. The total cost of
the CCS project is estimated to be $1 billion (including a $167 million grant from the US
Department of Energy), and the project is expected to extract 1.4-1.6 million tons of COi per
year. The construction contract was awarded in July, 2014, and operation is expected to begin in
early 2016. The CO2 will be piped 85 miles to a reservoir for EOR in the West Ranch Oil Field.38
2.6.2 Geologic and Geographic Considerations for Geologic Sequestration
Geologic sequestration (GS) (i.e., long-term containment of a COi stream in subsurface
geologic formations) is technically feasible and available throughout most of the United States.
GS is feasible in different types of geologic formations including deep saline formations
(formations with high salinity formation fluids) or in oil and gas formations, such as where
injected CO2 increases oil production efficiency through a process referred to as enhanced oil
recovery (EOR). COi may also be used for other types of enhanced recovery, such as for natural
gas production. Reservoirs, such as unmineable coal seams, also offer the potential for geologic
storage. The geographic availability of deep saline formations, EOR, and un-mineable coal
seams is shown in Figure 2-21. Estimates of COi storage resources by state compiled by the
DOE's National Carbon Sequestration Database and Geographic Information System
(NATCARB) and published in DOE's 2012a Carbon Utilization and Storage Atlas (discussed
below) are provided in Table 2-8.
38 US DOE (2010) "Recovery Act: W.A. Parish Post-Combustion CO2 Capture and Sequestration Project"
http://www.netl.doe.gov/research/proj?k=FE0003311 Accessed 1/23/2015
2-33
-------
Existing C02 pipeline (Department of Transportation)
Probable, planned, or under study CO2 pipeline
Counties with active CO2-EOR operations (EPA GHG Reporting Program)
Oil & Natural Gas Reservoirs (Department of Energy, NATCARB)
Deep Saline Formations (Department of Energy, NATCARB)
Unmineable Coal Seams (Department of Energy, NATCARB)
500
Miles
Figure 2-21. Geologic Sequestration in the Continental United States
Sources: EPA Greenhouse Gas Reporting Program; Department of Energy, NATCARB; Department of
Transportation, National Pipeline Management System.
2-34
-------
Table 2-8. Total CCh Storage Resource (DOE-NETL)39
State
Million Tons*
Low Estimate High Estimate
ALABAMA
ALASKA
ARIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
DISTRICT OF COLUMBIA
FLORIDA
GEORGIA
HAWAII
IDAHO
ILLINOIS
INDIANA
IOWA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
135,022
9,524
143
6,812
37,357
41,458
not assessed by DOE-NETL
44
not assessed by DOE-NETL
113,251
160,210
not assessed by DOE-NETL
44
11,045
35,296
11
11,993
3,219
186,842
not assessed by DOE-NETL
2,050
not assessed by DOE-NETL
765,422
21,771
1,290
70,184
463,665
393,734
not assessed by DOE-NETL
44
not assessed by DOE-NETL
611,793
175,322
not assessed by DOE-NETL
430
128,772
75,189
55
95,173
8,433
2,319,238
not assessed by DOE-NETL
2,127
not assessed by DOE-NETL
(Continued on next page)
39 The United States 2012 Carbon Utilization and Storage Atlas, Fourth Edition, U.S Department of Energy, Office
of Fossil Energy, National Energy Technology Laboratory (NETL).
2-35
-------
Table 2-8. Total CCh Storage Resource, continued
State
MICHIGAN
MINNESOTA
MISSISSIPPI
MISSOURI
MONTANA
NEBRASKA
NEVADA
NEW HAMPSHIRE
NEW JERSEY
NEW MEXICO
NEW YORK
NORTH CAROLINA
NORTH DAKOTA
Offshore Federal Only
OHIO
OKLAHOMA
OREGON
PENNSYLVANIA
RHODE ISLAND
SOUTH CAROLINA
SOUTH DAKOTA
TENNESSEE
TEXAS
UTAH
VERMONT
VIRGINIA
WASHINGTON
WEST VIRGINIA
WISCONSIN
WYOMING
U.S. Total
* States with a "zero" value
Million
Low Estimate
20,999
not assessed by DOE-NETL
159,846
11
93,233
26,202
not assessed by DOE-NETL
not assessed by DOE-NETL
-
47,135
5,115
1,477
73,954
539,956
14,837
62,777
7,507
24,361
not assessed by DOE-NETL
33,180
9,656
474
489,205
28,076
not assessed by DOE-NETL
485
40,367
18,353
0
80,127
2,531,653
Tons*
High Estimate
52,040
not assessed by DOE-NETL
1,306,270
187
1,006,100
124,826
not assessed by DOE-NETL
not assessed by DOE-NETL
-
395,828
5,115
20,271
162,569
7,098,976
14,837
269,570
103,286
24,361
not assessed by DOE-NETL
37,677
26,489
4,255
4,772,925
265,558
not assessed by DOE-NETL
3,208
547,550
18,353
0
754,917
22,147,811
represent estimates of minimal COi storage resource. States that have not yet been
assessed by the RCSPs have been identified.
2-36
-------
2.6.3 Availability of Geologic Sequestration in Deep Saline Formations
DOE and the United States Geological Survey (USGS) have independently conducted
preliminary analyses of the availability and potential COi sequestration capacity of deep saline
formations in the United States. DOE estimates are compiled by the DOE's National Carbon
Sequestration Database and Geographic Information System (NATCARB) using volumetric
models and published in a Carbon Utilization and Storage Atlas.40 DOE estimates that areas of
the United States with appropriate geology have a sequestration potential of at least 2,244 billion
tons of CO2 in deep saline formations. According to DOE and at least 39 states have geologic
characteristics that are amenable to deep saline GS in either onshore or offshore locations. In
2013, the USGS completed its evaluation of the technically accessible GS resources for COi in
U.S. onshore areas and state waters using probabilistic assessment.41 The USGS estimates a
mean of 3,307 billion tons of subsurface COi sequestration potential, including saline and oil and
gas reservoirs, across the basins studied in the United States. As shown in Figure 2-21, there are
39 states for which onshore and offshore deep saline formation storage capacity has been
identified.42
2.6.4 Availability of CO2 Storage via Enhanced Oil Recovery (EOR)
Although the regulatory impact analysis for this rule relies on GS in deep saline
formations, the EPA also recognizes the potential for securely sequestering COi via EOR. EOR
has been successfully used at numerous production fields throughout the United States to
increase oil recovery. The oil industry in the United States has over 40 years of experience with
EOR. An oil industry study in 2014 identified more than 125 EOR projects in 98 fields in the
United States.43 More than half of the projects evaluated in the study have been in operation for
40 The United States 2012 Carbon Utilization and Storage Atlas, Fourth Edition, U.S. Department of Energy, Office
of Fossil Energy, National Energy Technology Laboratory (NETL).
41 U.S. Geological Survey Geologic Carbon Dioxide Storage Resources Assessment Team, 2013, National
assessment of geologic carbon dioxide storage resourcesResults: U.S. Geological Survey Circular 1386, p. 41,
http://pubs.usgs.gov/circ/1386/.
42 Alaska is not shown in the figure; it has deep saline formation storage capacity, geology amenable to EOR
operations, and potential GS capacity in unmineable coal.
43 Koottungal, Leena, 2014, 2014 Worldwide EOR Survey, Oil & Gas Journal, Volume 112, Issue 4, April 7, 2014
(corrected tables appear in Volume 112, Issue 5, May 5, 2014).
2-37
-------
more than 10 years, and many have been in operation for more than 30 years. This experience
provides a strong foundation for demonstrating successful CO2 injection and monitoring
technologies, which are needed for safe and secure GS that can be used for deployment of CCS
across geographically diverse areas.
Currently, 12 states have active EOR operations and most have developed an extensive
COi infrastructure, including pipelines, to support the continued operation and growth of EOR.
An additional 18 states are within 100 kilometers (62 miles) of current EOR operations (see
Figure 2-21).44 The vast majority of EOR is conducted in oil reservoirs in the Permian Basin,
which extends through southwest Texas and southeast New Mexico. States where EOR is
utilized include Alabama, Colorado, Louisiana, Michigan, Mississippi, New Mexico, Oklahoma,
Texas, Utah, and Wyoming.
At the project level, the volume of COi already injected for EOR and the duration of
operations are of similar magnitude to the duration and volume of CO2 expected to be captured
from fossil fuel-fired EGUs. The volume of CO2 used in EOR operations can be large (e.g., 55
million tons of CO2 were stored in the SACROC unit in the Permian Basin over 35 years), and
operations at a single oil field may last for decades, injecting into multiple parts of the field.45
According to data reported to the EPA's Greenhouse Gas Reporting Program (GHGRP),
approximately 66 million tons of CO2 were supplied to EOR in the United States in 2013.46
Approximately 70 percent of this total CO2 supplied was produced from natural (geologic) CO2
sources, and approximately 30 percent was captured from anthropogenic sources.47
A DOE-sponsored study has analyzed the geographic availability of applying EOR in 11
major oil producing regions of the United States and found that there is an opportunity to
44 The distance of 100 kilometers reflects the assumptions in the DOE-NETL cost estimates.
45 Han, Weon S., McPherson, B J., Lichtner, P C., and Wang, F P. "Evaluation of COi trapping mechanisms at the
SACROC northern platform, Permian basin, Texas, site of 35 years of COi injection." American Journal of Science
310. (2010): 282-324.
46 Greenhouse Gas Reporting Program, data reported as of August 18, 2013.
47 Greenhouse Gas Reporting Program, data reported as of August 18, 2013.
2-38
-------
significantly increase the application of EOR to areas outside of current operations.48 DOE-
sponsored geologic and engineering analyses show that expanding EOR operations into areas
additional to the capacity already identified and applying new methods and techniques over the
next 20 years could utilize 20 billion tons of anthropogenic COi and increase total oil production
by 67 billion barrels. The availability of anthropogenic CO2 in areas outside of current sources
could drive new EOR projects by making more COi locally available.
2.7 State Policies on GHG and Clean Energy Regulation in the Power Sector
Several states have also established emission performance standards or other measures to
limit emissions of GHGs from new EGUs that are comparable to or more stringent than this
rulemaking.
In 2003, then-Governor George Pataki sent a letter to his counterparts in the Northeast and
Mid-Atlantic inviting them to participate in the development of a regional cap-and-trade program
addressing power plant CO2 emissions. This program, known as the Regional Greenhouse Gas
Initiative (RGGI), began in 2009 and sets a regional CO2 cap for participating states. The
currently participating states include: Connecticut, Delaware, Maine, Maryland, Massachusetts,
New Hampshire, New York, Rhode Island, and Vermont. The cap covers CO2 emissions from all
fossil-fired EGUs greater than 25 MW in participating states, and limits total emissions to 91
million short tons in 2014. The 2014 emissions cap is a 51 percent reduction below the initial cap
in 2009 to 2011 of 188 million tons. This emissions budget is reduced 2.5 percent annually from
2015 to 2020. RGGI COi allowances are sold in a quarterly auction. RGGI conducted their 27th
quarterly allowance auction in March, 2015 the market clearing price was $5.41 per ton of CO2
for current allowances, which was a record high price (the February ' 15 price of $5.21 was the
previous record). A total of allowances for 15.3 million tons were sold in the March ' 15 auction,
well below the record of 38.7 million tons sold in June ' 13 for $3.21.
48 "Improving Domestic Energy Security and Lowering COi Emissions with "Next Generation" COi-Enhanced Oil
Recovery", Advanced Resources International, Inc. (ARI), 2011. Available online at:
http://www.netl.doe.gov/research/energy-analysis/publications/details?pub=df02ffba-6b4b-4721-a7b4-
04a505al9185.
2-39
-------
In September 2006, California Governor Schwarzenegger signed into law Senate Bill
1368. The law limits long-term investments in baseload generation by the state's utilities to
power plants that meet an emissions performance standard jointly established by the California
Energy Commission and the California Public Utilities Commission. The Energy Commission
has designed regulations that establish a standard for new and existing baseload generation
owned by, or under long-term contract to publicly owned utilities, of 1,100 Ib CO2/MWh-net.
In 2006 Governor Schwarzenegger also signed into law Assembly Bill 32, the Global
Warming Solutions Act of 2006. This act includes a multi-sector GHG cap-and-trade program
which covers approximately 85 percent of the state GHG emissions. EGUs are included in phase
I of the program, which began in 2013. Phase II begins in 2020 and includes upstream sources.
The cap is based on a 2 percent reduction from total 2012 expected emissions, and declines 2
percent annually through 2014, then 3 percent each year until 2020. The AB32 cap and trade
program began functioning in 2011, and functioning market is now operating on the NYMEX
futures commodity market. The final 2014 market price for 2014 carbon allowances was
$13.65/ton of carbon. On April 17, 2015 the 2015 allowance futures price was $13.94/ton, and
the spot price was $13.73/ton.
In May 2007, Washington Governor Gregoire signed Substitute Senate Bill 6001,
"Baseload Electric Generation Performance" which established statewide GHG emissions
reduction goals, and imposed an emission standard that applies to any baseload electric
generation that commenced operation after June 1, 2008 and is located in Washington, whether
or not that generation serves load located within the state. Baseload generation facilities must
initially comply with an emission limit of 1,100 Ib CCh/MWh-net. In 2013 the State of
Washington revised49 the emission limit to 970 Ib CCh/MWh-net based on a survey of available
NGCC generation units commercially available in the United States.
In 1997 Oregon required a new baseload gas fired power plants to meet a COi emission
standard that was 17 percent below the most efficient NGCC unit operating in the United States.
49 Washington Department of Commerce, 2013. "Greenhouse Gas Emission Performance Standard for Baseload
Electric Generation". Available online at: http://www.comnierce.wa.gov/Docunients/Concise-Expl-Stmt-WSR-13-
06-074.pdf.
2-40
-------
In 2000 Oregon established that the effective 17 percent below most efficient was 675 Ib
COi/MWh-net. In July 2009, Oregon Governor Kulongoski signed Senate Bill 101, which
mandated that facilities generating baseload electricity, whether gas- or coal-fired, must have
emissions equal to or less than 1,100 Ib COi/MWh-net regardless of fuel type, and prohibited
utilities from entering into long-term purchase agreements for baseload electricity with out-of-
state facilities that do not meet that standard. Natural gas- and petroleum distillate-fired facilities
that are primarily used to serve peak demand or to integrate energy from renewable resources are
specifically exempted from the performance standard.
In August 2011, New York Governor Cuomo signed the Power NY Act of 2011.
Implementing regulations established CO2 emission standards for new and modified electric
generators greater than 25 MW. The standards vary based on the type of facility: base load
facilities must meet a COi standard of 925 Ib/MWh-net or 120 Ib/MMBtu, and peaking facilities
must meet a CO2 standard of 1,450 Ibs/MWh-net or 160 Ibs/MMBtu.
Several other states have enacted COi regulations affecting EGUs that do not set emission
limits, but set other regulatory requirements limiting CO2 emissions from EGUs. For example,
Montana enacted a law in 2007 requiring the Public Service Commission to limit approvals of
new equity interests in or leases of a facility used to generate coal-based electricity to facilities
that capture and sequester at least half of their COi emissions. Minnesota enacted the Next
Generation Energy Act in 2007 requiring increases in power sector greenhouse gas emissions
from any new large coal energy facilities built in Minnesota or the import of electricity from
such a facility located out of state to be offset by equivalent emission reductions. New Mexico
enacted legislation in 2007 authorizing tax credits and cost recovery incentives for qualifying
coal-fired facilities. To qualify, plants must capture and store emissions so that they emit less
than 1,100 Ibs CO2/MWh, among other requirements.
Additionally, most states have implemented Renewable Portfolio Standards (RPS), or
Renewable Electricity Standards (RES). These programs are designed to increase the renewable
share of a state's total electricity generation. Currently 29 states, the District of Columbia, and
Guam have enforceable RPS or other mandatory renewable capacity policies, and eight states,
2-41
-------
Puerto Rico, and Guam have voluntary goals.50 These programs vary widely in structure,
enforcement, and scope.
2.8 Revenues and Expenses
Due to lower retail electricity sales, total utility operating revenues declined in 2012 to
$271 billion from a peak of almost $300 billion in 2008. Despite revenues not returning to 2008
levels in 2012, operating expenses were appreciably lower and as a result, net income also rose
in comparison to 2008 (see Table 2-9). Recent economic events have put downward pressure on
electricity demand, thus dampening electricity prices and consumption (utility revenues), but
have also reduced the price and cost of fossil fuels and other expenses. In 2012 electricity
generation was 1.28 percent below the generation in 2011, and has declined in 4 of the past 5
years.
Table 2-9 shows that investor-owned utilities (lOUs) earned income of about 13.0 percent
compared to total revenues in 2012. The 2012 return on revenue was the third highest year for
the period 2002 to 2012 (average: 11.9 percent range: 10.6 percent to 13.32 percent).
Table 2-9. Revenue and Expense Statistics for Major U.S. Investor-Owned Electric Utilities
for 2002, 2008 and 2012 (nominal $millions)
Utility Operating Revenues
Electric Utility
Other Utility
Utility Operating Expenses
Electric Utility
Operation
Production
Cost of Fuel
Purchased Power
Other
Transmission
Distribution
Customer Accounts
Customer Service
2002
219,609
200,360
19,250
189,062
171,604
116,660
90,715
24,149
58,810
7,776
3,560
3,117
4,168
1,820
2008
298,962
266,124
32,838
267,263
236,572
175,887
140,974
47,337
84,724
8,937
6,950
3,997
5,286
3,567
2012
270,912
249,166
21,745
235,694
220,722
152,379
111,714
38,998
54,570
18,146
7,183
4,181
5,086
5,640
'EIA2012a
2-42
-------
Sales
Admin, and
General
Maintenance
Depreciation
Taxes and Other
Other Utility
Net Utility Operating Income
264
13,018
10,861
16,199
26,716
17,457
30,548
225
14,718
14,192
19,049
26,202
30,692
31,699
221
18,353
15,489
23,677
29,177
14,972
35,218
Source: Table 8.3, EIA Electric Power Annual, 2012
Note: This data does not include information for public utilities, nor for Independent Power Producers (IPPs).
2.9 Natural Gas Market
The natural gas market in the United States has historically experienced significant price
volatility from year to year, between seasons within a year, can undergo major price swings
during short-lived weather events (such as cold snaps leading to short-run spikes in heating
demand), and has seen a dramatic shift since 2008 due to increased production from shale
formations. Over the last decade, the annual average nominal price of gas delivered to the power
sector peaked in 2008 at $9.02/MMBtu and has since fallen dramatically to a low of
$3.42/MMBtu in 2012. During that time, the daily price51 of natural gas reached as high as
$18.48/MMBtu and as low as $2.03. Adjusting for inflation using the GDP implicit price
deflator, in $2011 the annual average price of natural gas delivered to the power sector peaked at
$9.38/MMBtu in 2008 and has fallen dramatically to a low of $3.36 in 2012. The annual natural
gas prices in both nominal and real (2011$) terms are in Figure 2-22. A comparison of the trends
in the real price of natural gas with the real prices of delivered coal and oil are shown in Figure
2-23. Figure 2-23 shows that while the real price of coal and oil increased from 2002 to 2012
(+54 percent and +203 percent respectively), the real price of natural gas declined by 22 percent
in the same period. Most of the decline in real natural gas prices occurred between 2008 (the
peak price year) and 2012, during which real gas prices declined by 64 percent while coal and oil
51 Henry Hub daily prices. Henry Hub is a major gas distribution hub in Louisiana; Henry Hub prices are generally
seen as the primary metric for national gas prices for all end uses. The price of natural gas delivered to electricity
generation differs substantially in different regions of the country, and can be higher or lower than the Henry Hub
national benchmark price.
2-43
-------
prices both increased by 9 percent. The sharp decline in natural gas prices from 2008 to 2012
was primarily caused by the rapid increase in natural gas production from shale formations.
$0.00
2002
2004
2006
Nominal Price
2008
Real Price
2010
2012
Figure 2-22. Relative Change Nominal and Real (2011$) Prices of Natural Gas Delivered
to the Power Sector ($/MMBtu)
Source: http://www.eia.gov/totalenergy/data/monthly/ttprices. Downloaded 2/15/2015.
2-44
-------
-50%
2002
2012
Figure 2-23. Relative Change in Real (2011$) Prices of Fossil Fuels Delivered to the Power
Sector ($/MMBtu)
Source: http://www.eia.gov/totalenergy/data/monthly/ttprices. Downloaded 2/15/2015.
Current and projected natural gas prices are considerably lower than the prices observed
over the past decade, largely due to advances in hydraulic fracturing and horizontal drilling
techniques that have opened up new shale gas resources and substantially increased the supply of
economically recoverable natural gas. According to AEO 2012 (EIA 2012):
Shale gas refers to natural gas that is trapped within shale formations. Shales are fine-
grained sedimentary rocks that can be rich sources of petroleum and natural gas. Over the
past decade, the combination of horizontal drilling and hydraulic fracturing has allowed
access to large volumes of shale gas that were previously uneconomical to produce. The
production of natural gas from shale formations has rejuvenated the natural gas industry
in the United States.
The U.S. Energy Information Administration's Annual Energy Outlook 2014 estimates that
the United States possessed 2,266 trillion cubic feet (Tcf) of technically recoverable dry natural
gas resources as of January 1, 2012. Proven reserves make up 15 percent of the technically
2-45
-------
recoverable total estimate, with the remaining 85 percent from unproven reserves. Natural gas
from proven and unproven shale resources accounts for 611 Tcf of this resource estimate.
Many shale formations, especially the Marcellus52, are so large that only small portions of
the entire formations have been intensively production-tested. Furthermore, estimates from the
Marcellus and other emerging fields with few wells already drilled are likely to shift significantly
over time as new geological and production information becomes available. Consequently, there
is some uncertainty in the estimate of technically recoverable resources, and it is regularly
updated as more information is gained through drilling and production.
At the 2012 rate of U.S. consumption (about 25.6 Tcf per year), 2,266 Tcf of natural gas is
enough to supply nearly 90 years of use. The AEO 2014 estimate of the shale gas resource base
is modestly higher than the AEO 2012 estimate (2,214 Tcf) of shale gas production, driven by
lower drilling costs and continued drilling in shale plays with high concentrations of natural gas
liquids and crude oil, which have a higher value in energy equivalent terms than dry natural
gas.53
EIA's projections of natural gas conditions did not change substantially in AEO 2014 from
either the AEO 2012 or 2013, and EIA continues to forecast abundant reserves consistent with
the above findings. Recent historical data reported to EIA is also consistent with these trends,
with 2014 being the highest year on record54 for domestic natural gas production.55
52 The Marcellus formation, underlying most of Pennsylvania and West Virginia, along with portions of New York
and Ohio, in 2014 produced 36% of the U.S. total natural gas extracted from shale formations.
53 For more information, see: http://www.eia.gov/forecasts/archive/aeol l/IF_all.cfm#prospectshale;
http://www.eia.gov/energy_in_brief/about_shale_gas.cfm
54 The total dry gas production in 2012 from the lower 48 states, including both onshore and offshore production,
was 23.97 Tcf, a 1.5% increase from 2013 and a 7.9% total increase from 2011
55 http://www. eia. gov/oiaf/aeo/tablebrowser/#release=AEO2014&subj ect=8 - AEO2014&table=72-
AEO2014®ion=0-0&cases=ref2014-dl02413a
2-46
-------
2.10 References
Advanced Resources International. 2011. Improving Domestic Energy Security and Lowering
C02 Emissions with "Next Generation " CO2-Enhanced Oil Recovery (CO2-EOR). Available
online at: http://www.netl.doe.gov/research/energy-
analysis/publications/details?pub=df02ffba-6b4b-4721-a7b4-04a505al9185. Accessed June 9,
2015.
Clean Energy States Alliance (CESA). The State of State Renewable Portfolio Standards. June
2013. Available online at http://www.cesa.org/assets/2013-Files/RPS/State-of-State-RPSs-
Report-Final-June-2013.pdf
Han, Weon S., McPherson, B J., Lichtner, P C., and Wang, F P. 2010. Evaluation of CO2
trapping mechanisms at the SACROC northern platform, Permian basin, Texas, site of 35
years of COi injection. American Journal of Science 310: 282-324.
Independent Monitor's Prudency Evaluation Report for the Kemper County IGCC Project. April
15, 2014. Available online at:
http://www.psc.state.ms.us/InsiteConnect/InSite View.aspx?model=INSITE_CONNECT&que
ue=CTS_ARCHIVEQ&docid=328417.
Interagency Task Force on Carbon Capture and Storage. Report of the Inter agency Task Force
on Carbon Capture and Storage. August 2010. Available online at:
http://www.epa.gov/climatechange/downloads/CCS-Task-Force-Report-2010.pdf. Accessed
June 9, 2015.
Intergovernmental Panel on Climate Change. 2001. Climate Change 2001: The Scientific Basis.
Available online
at:http://www.grida.no/publications/other/ipcc_tar/?src=/climate/ipcc_tar/wgl/218.htm.
Accessed June 9, 2015.
International Energy Agency (IEA). 2013. Tracking Clean Energy Progress 2013. Input to the
Clean Energy Ministerial. Available online at: http://www.iea.org/etp/tracking/. Accessed
June 9, 2015.
Koottungal, Leena. 2014. 2014 Worldwide FOR Survey, Oil & Gas Journal, Volume 112, Issue
4, April 7, 2014 (corrected tables appear in Volume 112, Issue 5, May 5, 2014).
National Energy Technology Laboratory (NETL). 2008. Reducing COz Emissions by Improving
the Efficiency of Existing Coal-fired Power Plant Fleet. Available online at
http://www.netl.doe.gov/energy-analyses/pubs/CFPP%20Efficiency-FINAL.pdf. Accessed
June 9, 2015.
2-47
-------
National Energy Technology Laboratory (NETL). 2012. The United States 2012 Carbon
Utilization and Storage Atlas, Fourth Edition. Available online at:
http://www.netl.doe.gov/technologies/carbon_seq/refshelf/atlasIV/. Accessed June 9, 2015.
National Energy Technology Laboratory (NETL). 2013. Energy Analyses: Cost and
Performance Baselines for Fossil Energy Plants. Available online at:
http://www.netl.doe.gov/energy-analyses/baseline_studies.html. Accessed June 9, 2015.
Pacific Northwest National Laboratory (PNNL). 2009. An Assessment of the Commercial
Availability of Carbon Dioxide Capture and Storage Technologies as of June 2009. Available
online at: http://www.pnl.gov/science/pdf/PNNL-18520_Status_of_CCS_062009.pdf.
Accessed June 9, 2015.
U.S. Energy Information Administration (U.S. EIA). 2000. Carbon Dioxide Emissions from the
Generation of Electric Power in the United States. Available online at:
ftp://ftp.eia.doe.gov/environment/co2emissOO.pdf. Accessed June 9, 2015.
U.S. Energy Information Administration (U.S. EIA). 2003. Electric Power Annual 2003.
Available online at: http://www.eia.gov/electricity/annual/archive/03482003.pdf. Accessed
June 9, 2015.
U.S. Energy Information Administration (U.S. EIA). 2009. Electric Power Annual 2009.
Available online at: http://www.eia.gov/electricity/annual/archive/03482009.pdf. Accessed
June 9, 2015.
U.S. Energy Information Administration (U.S. EIA). 2013. Electric Power Annual 2013.
Available online at: http://www.eia.gov/electricity/annual/. Accessed June 9, 2015.
U.S. Energy Information Administration (U.S. EIA). 2010a. "Status of Electricity Restructuring
by State." Available online at:
http://www.eia.gov/cneaf/electricity/page/restructuring/restructure_elect.html. Accessed June
9,2015.
U.S. Energy Information Administration (U.S. EIA). 2010b. AEO 2010 Retrospective Review.
Available online at: http://www.eia.gov/forecasts/aeo/retrospective/. Accessed June 9, 2015.
U.S. Energy Information Administration (U.S. EIA). 2010c. Annual Energy Outlook 2010.
Available online at: http://www.eia.gov/oiaf/archive/aeolO/index.html. Accessed June 9,
2015.
U.S. Energy Information Administration (U.S. EIA). 2010d. Annual Energy Review 2010.
Available online at: http://www.eia.gov/totalenergy/data/annual/pdf/aer.pdf. Accessed June 9,
2015.
2-48
-------
U.S. Energy Information Administration (U.S. EIA). 2011. Annual Energy Outlook 2011.
Available online at: http://www.eia.gov/forecasts/archive/aeoll/. Accessed June 9, 2015.
U.S. Energy Information Administration (U.S. EIA). 2012. Annual Energy Outlook 2012 (Early
Release). Available online at: http://www.eia.gov/forecasts/aeo/. Accessed June 9, 2015.
U.S. Energy Information Administration (U.S. EIA). Today in Energy: Most states have
Renewable Portfolio Standards. 2012a. Available online at:
http://www.eia.gov/todayinenergy/detail.cfm?id=4850. Accessed June 9, 2015.
U.S. Energy Information Administration (U.S. EIA). 2013. Annual Energy Outlook 2013.
Available online at: http://www.eia.gov/forecasts/aeo/. Accessed June 9, 2015.
U.S. Energy Information Administration (U.S. EIA). 2015. Monthly Energy Review, April 2015.
Available online at: http://www.eia.gov/totalenergy/data/monthly/. Accessed June 9, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2013. Inventory of U.S. Greenhouse Gas
Emissions and Sinks: 1990-2011. Available online at:
http://www.epa.gov/climatechange/Downloads/ghgemissions/US-GHG-Inventory-2013-
Main-Text.pdf. Accessed June 9, 2015.
U.S. Geological Survey (USGS) Carbon Dioxide Storage Resources Assessment Team. 2013.
National assessment of geologic carbon dioxide storage resources -Results: U.S. Geological
Survey Circular 1386. Available online at: http://pubs.usgs.gov/circ/1386/. Accessed June 9,
2015
2-49
-------
CHAPTER 3: COST, EMISSIONS, ECONOMIC, AND ENERGY IMPACTS
3.1 Introduction
This chapter reports the compliance cost, emissions, economic, and energy impact
analysis performed for the Clean Power Plan Final Rule. EPA used the Integrated Planning
Model (IPM), developed by ICF International, to conduct most of the analysis discussed in this
Chapter. IPM is a dynamic linear programming model that can be used to examine air pollution
control policies for CO2, SCh, NOx, Hg, HC1, and other air pollutants throughout the contiguous
United States for the entire power system. The IPM electricity demand projections are based on
projections from the Energy Information Administration (EIA), adjusted for demand-side energy
efficiency measures that can be reasonably anticipated to occur under the Clean Power Plan.
3.2 Overview
This chapter of the RIA presents illustrative analyses of the final rule by making
assumptions about the possible approaches that States might pursue as they develop their state
plans. Over the last decade, EPA has conducted extensive analyses of regulatory actions
affecting the power sector. These efforts support the Agency's understanding of key variables
that influence the effects of a policy and provide the framework for how the Agency estimates
the costs and benefits associated with its actions.
3.3 Power Sector Modelling Framework
The Integrated Planning Model (IPM), developed by ICF Consulting, 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 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 Clean Power Plan
Final Rule. Additional demand side energy efficiency measures that may be adopted in response
to the regulation, and the resulting changes to future demand projections, are also accounted for
in the analyses. The level of demand side energy efficiency-driven reductions in electricity
demand, and their associated costs, are reported in section 3.7.
IPM is a multi-regional, dynamic, deterministic linear programming model of the
contiguous U.S. electric power sector. It provides forecasts of least cost capacity expansion,
3-1
-------
electricity dispatch, and emission control strategies while meeting energy demand and
environmental, transmission, dispatch, and reliability constraints. EPA has used IPM for over
two decades to better understand power sector behavior under future business-as-usual
conditions and to evaluate the economic and emission 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 information on the assumptions
discussed here as well as all other model assumptions and inputs.56
The model incorporates a detailed representation of the fossil-fuel supply system that is
used to forecast equilibrium fuel prices. The model includes an endogenous representation of the
North American natural gas supply system through a natural gas module that reflects a partial
supply/demand equilibrium of the North American gas market accounting for varying levels of
potential power sector and non-power sector gas demand and corresponding gas production and
price levels.57 This module consists of 118 supply, demand, and storage nodes and 15 liquefied
natural gas re-gasification facility locations that are tied together by a series of linkages (i.e.,
pipelines) that represent the North American natural gas transmission and distribution network.
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
56 Detailed information and documentation of EPA's Base Case using IPM (v5.15), including all the underlying
assumptions, data sources, and architecture parameters can be found on EPA's website at:
http://www.epa.gov/powersectormodeling
57 See Chapter 10 of EPA's Base Case using IPM (v5.154) documentation, available at:
http://www.epa.gov/powersectormodeling
3-2
-------
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.58
The costs presented in this RIA include both the IPM-projected annualized estimates of
private compliance costs as well as the estimated costs incurred by utilities and ratepayers to
achieve demand-side energy efficiency improvements. The IPM-projected annualized 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 final rule.
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. 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.59 It is important to note that there is no single CRF factor applied in
the model; rather, the CRF varies across technologies in the model in order to better simulate
power sector decisionmaking.
While the CRF is used to annualize costs within IPM, a discount rate is used to estimate
the net present value of the intertemporal flow of the annualized capital and operating costs. The
optimization model then identifies power sector investment decisions that minimize the net
present value of all costs over the full planning horizon while satisfying a wide range of demand,
capacity, reliability, emissions, and other constraints. As explained in Chapter 8 of the IPM
documentation, the discount rate is derived as a weighted average cost of capital that is a
function of capital structure, post-tax cost of debt, and post-tax cost of equity. While the detailed
formulation of this rate is presented in the IPM documentation, the rate estimated and used in the
current analysis is 4.77 percent. It is important to note that this discount rate is selected for the
purposes of best simulating power sector behavior, and not for the purposes of discounting social
costs or benefits.
58 See Chapter 9 of EPA's Base Case using IPM (v5.15) documentation, available at:
http://www.epa.gov/powersectormodeling
59 See Chapter 8 of EPA's Base Case using IPM (v5.15) documentation, available at:
http://www.epa.gov/powersectormodeling.
3-3
-------
EPA has used IPM extensively over the past two decades to analyze options for reducing
power sector emissions. Previously, the model has been used to forecast the costs, emission
changes, and power sector impacts for the Clean Air Interstate Rule, Cross-State Air Pollution
Rule (CSAPR), the Mercury and Air Toxics Standards (MATS), and the proposed Carbon
Pollution Standards for New Power Plants. Recently IPM has also been used to estimate the air
pollution reductions and power sector impacts of water and waste regulations affecting EGUs,
including Cooling Water Intakes (316(b)) Rule, Disposal of Coal Combustion Residuals from
Electric Utilities (CCR) and Steam Electric Effluent Limitation Guidelines (ELG).
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 the late
1990s, the Science Advisory Board reviewed IPM as part of the CAA Amendments Section 812
prospective studies that are periodically conducted. The model has also undergone considerable
interagency scrutiny when it was used to conduct over a dozen legislative analyses (performed at
Congressional request) over the past decade. The Agency has also used the model in a number of
comparative modeling exercises sponsored by Stanford University's Energy Modeling Forum
over the past 15 years. IPM has also been employed by states (e.g., for RGGI, the Western
Regional Air Partnership, Ozone Transport Assessment Group), other Federal and state agencies,
environmental groups, and industry.
3.4 Recent Updates to EPA's Base Case using IPM (v.5.15)
The "Base Case" for this analysis is a business-as-usual scenario that would be expected
under market and regulatory conditions in the absence of this rule. As such, the IPM base case
represents the baseline for this RIA. EPA frequently updates the IPM base case to reflect the
latest available electricity demand forecasts as well as expected costs and availability of new and
existing generating resources, fuels, emissions control technologies, and regulatory requirements.
3-4
-------
EPA's IPM modeling platform used to analyze this final rale (v.5.15) incorporates
updates to the version of the model used to analyze the impacts of the proposed rule (v.5.13).
These updates are primarily routine calibrations with the Energy Information Agency's (EIA)
Annual Energy Outlook (AEO), including updating the electric demand forecast consistent with
the AEO 2015 and an update to natural gas supply. Additional updates, based on the most up-to-
date information and/or public comments received by the EPA, include unit-level specifications
(e.g., pollution control configurations), planned power plant construction and closures, and
updated cost and performance for onshore wind and utility-scale solar technologies. This IPM
modeling platform incorporates federal and most state laws and regulations whose provisions
were either in effect or enacted and clearly delineated in March 2015. This update also includes
two non-air federal rules affecting EGUs: Cooling Water Intakes (316(b)) Rule and Combustion
Residuals from Electric Utilities (CCR). Additionally, all new capacity projected by the model is
compliant with Clean Air Act 11 l(b) standards, including the final standards of performance for
GHG emissions from new sources. For a detailed account of all updates made to the v.5.15
modeling platform, see the Incremental Documentation for EPA Base Case v.5.15 Using IPM.60
EPA also updated the National Electric Energy Data System (NEEDS). This database
contains the unit-level data that is used to construct the "model" plants that represent existing and
committed units in EPA modeling applications of IPM. NEEDS includes detailed information on
each individual EGU, including geographic, operating, air emissions, and other data on every
generating units in the contiguous U.S.61
3.5 State Goals in this Final Rule
In this final rale, the EPA is establishing COi emission performance rates for two
categories of existing fossil fuel-fired EGUs, fossil fuel-fired electric utility steam generating
units and stationary combustion turbines. The EPA has translated the source category-specific
COi emission performance rates into state-level rate-based and mass-based COi goals in order to
expand the range of choices that states have in developing their plans. Due to the range of
choices available to states, and the lack of a priori knowledge about the specific choices states
60 Available at: http://www.epa.gov/powersectormodeling/
61 The NEEDS database can be found on the EPA's website for the Base Case using IPM (v5.15),
.
3-5
-------
will make in response to the final goals, this RIA presents two scenarios designed to achieve
these goals, which we term the "rate-based" illustrative plan approach and the "mass-based"
illustrative plan approach. Table 3-1 presents the rate-based and mass-based state goals.
Table 3-1. Statewide CCh Emission Performance Goals, Rate-based and Mass-based
Rate-Based
(Adjusted Output- Weighted-
Average Pounds of CCh Per
Net MWh From All Affected
Fossil Fuel-Fired EGUs)
State
Alabama
Arkansas
Arizona
California
Colorado
Connecticut
Delaware
Florida
Lands of the Fort Mojave Tribe
Georgia
Iowa
Idaho
Illinois
Indiana
Kansas
Kentucky
Louisiana
Massachusetts
Maryland
Maine
Michigan
Minnesota
Missouri
Mississippi
Montana
Lands of the Navajo Nation
North Carolina
North Dakota
Nebraska
New Hampshire
New Jersey
New Mexico
Interim Goal Final Goal
1,157 1,018
1,304 1,130
1,173 1,031
907 828
1,362 1,174
852 786
1,023 916
1,026 919
832 771
1,198 1,049
1,505 1,283
832 771
1,456 1,245
1,451 1,242
1,519 1,293
1,509 1,286
1,293 1,121
902 824
1,510 1,287
842 779
1,355 1,169
1,414 1,213
1,490 1,272
1,061 945
1,534 1,305
1,534 1,305
1,311 1,136
1,534 1,305
1,522 1,296
947 858
885 812
1,325 1,146
Mass-Based
(Adjusted Output- Weighted-
Average Short Tons of CO2 From
All Affected Fossil Fuel-Fired
EGUs)
Interim Goal Final Goal
62,210,288 56,880,474
33,683,258 30,322,632
33,061,997 30,170,750
51,027,075 48,410,120
33,387,883 29,900,397
7,237,865 6,941,523
5,062,869 4,711,825
112,984,729 105,094,704
611,103 588,519
50,926,084 46,346,846
28,254,411 25,018,136
1,550,142 1,492,856
74,800,876 66,477,157
85,617,065 76,113,835
24,859,333 21,990,826
71,312,802 63,126,121
39,310,314 35,427,023
12,747,677 12,104,747
16,209,396 14,347,628
2,158,184 2,073,942
53,057,150 47,544,064
25,433,592 22,678,368
62,569,433 55,462,884
27,338,313 25,304,337
12,791,330 11,303,107
24,557,793 21,700,587
56,986,025 51,266,234
23,632,821 20,883,232
20,661,516 18,272,739
4,243,492 3,997,579
17,426,381 16,599,745
13,815,561 12,412,602
3-6
-------
Rate-Based
(Adjusted Output- Weighted-
Average Pounds of CCh Per
Net MWh From All Affected
Fossil Fuel-Fired EGUs)
State
Nevada
New York
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Lands of the Uintah and Ouray
Reservation
Utah
Virginia
Washington
Wisconsin
West Virginia
Wyoming
Interim Goal Final Goal
942 855
1,025 918
1,383 1,190
1,223 1,068
964 871
1,258 1,095
832 771
1,338 1,156
1,352 1,167
1,411 1,211
1,188 1,042
1,534 1,305
1,368 1,179
1,047 934
1,111 983
1,364 1,176
1,534 1,305
1,526 1,299
Mass-Based
(Adjusted Output- Weighted-
Average Short Tons of CO2 From
All Affected Fossil Fuel-Fired
EGUs)
Interim Goal Final Goal
14,344,092 13,523,584
33,595,329 31,257,429
82,526,513 73,769,806
44,610,332 40,488,199
8,643,164 8,118,654
99,330,827 89,822,308
3,657,385 3,522,225
28,969,623 25,998,968
3,948,950 3,539,481
31,784,860 28,348,396
208,090,841 189,588,842
2,561,445 2,263,431
26,566,380 23,778,193
29,580,072 27,433,111
11,679,707 10,739,172
31,258,356 27,986,988
58,083,089 51,325,342
35,780,052 31,634,412
3.6 Illustrative Plan Approaches Analyzed
To estimate the costs, benefits, and economic and energy market impacts of
implementing the CPP guidelines, the EPA modeled two illustrative plan approaches, each at the
state level, based on a rate-based approach and a mass-based approach. The rate-based plan
approach requires affected sources in each state to achieve a single average emissions rate in
each period as represented by the statewide goals. The mass-based plan approach requires
affected sources in each state to limit their aggregate emissions not to exceed the mass goal for
that state. The two plan types in these illustrative analyses represent two types of plans that are
available to the states.
In each of these scenarios, affected EGUs include:
Existing fossil steam boilers with nameplate capacity greater than 25 MW
Existing NGCC units with nameplate capacity greater than 25 MW
3-7
-------
In the rate-based scenario, generation (or avoided generation) from these additional sources
represented in the model is counted toward meeting state goals:
All renewable capacity (hydro, solar PV, wind, geothermal) that comes online
after 2012
Under-construction nuclear62
Demand-side energy efficiency in addition to levels implicit in base case
electricity demand.
In the rate-based illustrative plan approach analyzed in this RIA, the affected EGUs
within each state are required to achieve an average emissions rate that is less than or equal to the
state goals for each state. In order meet the goal for each state, the affected sources in this
scenario have the ability to do one or both of the following:
1) generate in amounts within that state such that the average emissions rate is achieved,
and/or
2) include in the average emissions rate calculation new renewable generation or
demand-side energy efficiency located outside of the state but within each of the
illustrative Interconnection-based regions shown in Figure 3-1 below.63
62 Includes three nuclear facilities at which construction has already commenced: Watts Barr (TN), Vogtle (GA), and
Summer (SC)
63 In this illustrative scenario, energy efficiency/renewable energy procurement is limited to within one of the three
illustrative regions. Since the interconnections do not always follow state borders, certain states that fall into more
than one region were grouped in regions where there was a majority of geographic territory (area) or generation.
Depending on the elements of their respective state's plan, sources in states that have adopted certain rate-based
plans may be able to procure energy efficiency/renewable energy from states outside of these illustrative regions.
See the preamble for discussion.
3-8
-------
Figure 3-1. Illustrative Regions for Demand-Side Energy Efficiency/Renewable Energy
Procurement Used in this Analysis
This rate-based implementation plan approach enables some sources to emit at emission
rates higher than their applicable state goal, as long as there is either corresponding generation
coming from affected sources in that state that emit at a lower rate and/or generation (or avoided
generation) from energy efficiency/renewable energy (which is procured from within the
illustrative regions, including within the source's state). In this illustrative analysis affected
EGUs may not procure emission reductions from (e.g., by averaging their emissions with)
affected EGUs located in other states (which may also have different emission performance
standards) in order to demonstrate compliance. Furthermore in this rate-based scenario, specific
generation (or avoided generation) from energy efficiency/renewable energy procurement may
only be used once for compliance toward a state goal; in other words, while emitting sources in
all states may avail themselves of qualifying energy efficiency/renewable energy across the
illustrative region, no particular energy efficiency/renewable energy MWh can be claimed by
more than one emitter as part of reaching a state goal.
Each illustrative plan approach assumes identical levels of demand-side energy efficiency
megawatt-hour (MWh) demand reductions and associated costs, which are specified
exogenously and consistent with the energy efficiency plan scenario performance levels
3-9
-------
described in section 3.7. Details of the implementation of the demand reduction are reported in
the following section.
The mass-based scenario presented in this chapter includes a 5 percent set-aside of
allowances that would be allocated to recognize deployment of new renewable capacity, which is
represented by lowering the capital cost of new renewable capacity in a compliance period by the
estimated value of the allowances in the set-aside in that period. The value of the set-aside is
estimated in each model run year (i.e., simulated year in IPM) as the total allowances in the set-
asides of each state in the contiguous U.S. multiplied by the projected average allowance price
over the contiguous U.S. for that year. This total value is then assumed to apply evenly to all new
renewable capacity.
Each of the two illustrative plan approaches assumes that sources within each state
comply with the applicable state goals without exchanging a compliance instrument (ERC or
allowance) with sources in any other state. However, in the rate-based scenario, sources are
allowed to procure renewable energy or demand-side energy efficiency beyond their own state in
order to adjust their effective emission rate, which is consistent with the conditions for rate-based
implementation in any state that are described in section VIII of the preamble.64 For example,
while the final rule enables states to achieve their mass goals with the flexibility of interstate
trading, this RIA presents analysis is an illustrative plan approach that assumes that each state
achieves its goal independently. Cooperation between the states that allows for trading across
states would provide EGUs with additional low cost abatement opportunities and would
therefore lower the overall cost of compliance across the affected states. While the illustrative
plan approaches assume particular plan types that may limit compliance options available to
affected EGUs, the equilibrium effects on generation, emissions, etc., in a particular state that are
forecast in these analyses depend on the behavior of generators in neighboring states in response
to the regulation.
The full array of estimates for the benefits, costs, and economic impacts of this action are
presented for both the illustrative rate-based and mass-based plan approaches. These illustrative
plan approaches are designed to reflect, to the extent possible, the scope and nature of the CPP
64 In this modeling scenario, sources were only able to procure such RE and EE within the same interconnection-
based region, while the rule does not impose a regional limitation to such claims in rate-based compliance.
3-10
-------
guidelines. However, there is considerable uncertainty with regard to the regulatory form and
precise measures that states will adopt to meet the requirements, since there are considerable
flexibilities afforded to the states in developing state plans. Nonetheless, the analysis of the
benefits, costs, and relevant impacts of the rule attempts to encapsulate some of those flexibilities
in order to inform states and stakeholders of the potential overall impacts of the CPP.
It is also important to note that the analysis does not specify any particular COi reduction
measure to occur, with the exception of the level of demand-side energy efficiency assumed to
be adopted in response to the CPP. In other words, aside from investments in energy efficiency,
the analysis allows the power system the flexibility to respond to average emissions rate or mass
constraints on affected sources in the illustrative scenarios to achieve the goals in the most cost-
effective manner determined by IPM, as specified below. Additionally, there are other zero-
emitting alternatives to replacing fossil generation beyond the renewable generation technologies
that are part of building block 3 and the energy efficiency measures that were analyzed in these
scenarios. For instance, while costs would be different, the impact of distributed zero-emitting
generation such as residential and commercial solar would displace fossil generation in the same
way that demand side energy efficiency would.
While IPM produces a cost-minimizing solution to achieve the state goals imposed in the
illustrative scenarios, there may be yet lower-cost approaches that the states may adopt to
achieve their state goals inasmuch as states and sources take advantage of emission reduction
opportunities in practice, and flexibilities afforded under the final rule, that are not represented in
this analysis and would yield different cost and emissions outcomes.
As previously noted, the power sector modeling and analysis presented in this chapter is
intended to be illustrative in nature, and reflects the EPA's best assessment of likely impacts of
the CPP under a range of approaches that states may adopt. The modeling is designed to reflect
the rule's requirements, including the timing, applicability to sources, and flexibilities across the
power system as accurately as possible to represent the nature and scope of the CPP. The
analysis is a reasonable expectation of the incremental effects of the rule, and is consistent with
past EPA analyses of power sector regulatory requirements. The EPA has separately analyzed
and considered the cost of implementing the emission reduction measures in BSER, which do
3-11
-------
not rely on energy efficiency measures. For this analysis, see section V.A.4.d. of the preamble to
this final rale.
For the CPP, the analysis and projections for the year 2025 reflect the impacts across the
power system of complying with the interim goals, and the analysis and projections for 2030
reflect the impacts of complying with the final goals. In addition to the 2025 and 2030
projections, modeling results and projections are also shown for 2020. There is no regulatory
requirement reflected in the 2020 run-year in IPM, consistent with the final rule. These years
reflect the basic run-year structure in IPM, as configured by EPA.
Although the analysis of the CPP does not include estimates of the costs and benefits of
the CPP across each year of the rule in a year-by-year manner, the EPA has reflected the
structure of the rale, including the interim and the final state goals of the CPP, in a manner that is
consistent with the regulatory requirements. This is also consistent with past practice, including
analysis of the Clean Air Interstate Rule, the Cross State Air Pollution Rule, the NOx SIP Call,
the Acid Rain Program, National Ambient Air Quality Standards, and state rules. These past
regulatory and legislative efforts included modeling and analysis in a similar manner, where
select analytic years reflected projections of policy impacts for rales that include multi-year
compliance periods.
3.7 Demand-Side Energy Efficiency
3.7.1 Demand-Side Energy Efficiency Improvements (Electricity Demand Reductions)65
While the final rale no longer includes demand-side energy efficiency potential as part of
BSER, the rale does allow such potential to be used for compliance. These scenarios include a
representation of demand-side energy efficiency compliance potential because energy efficiency
is a highly cost-effective means for reducing CO2 from the power sector, and it is reasonable to
assume that a regulatory requirement to reduce COi emissions will motivate parties to pursue all
highly cost-effective means for making emission reductions accordingly, regardless of what
particular emission reduction measures were assumed in determining the level of that regulatory
requirement. The EPA has included in our illustrative plan scenarios (both rate- and mass-based)
65 For a more detailed discussion of the demand-side energy efficiency demand reductions and their associated costs,
refer to U.S. EPA. 2015. Technical Support Document (TSD) the Final Carbon Pollution Emission Guidelines for
Existing Stationary Sources: Electric Utility Generating Units. Demand-Side Energy Efficiency.
3-12
-------
a level of demand reduction that could be achieved, and the associated costs incurred, through
implementation of demand-side energy efficiency measures. This "demand-side energy
efficiency plan scenario" represents a level of performance that has already been demonstrated or
is required by policies (e.g., energy efficiency resource standards) of leading energy efficiency
implementing states, and is consistent with a demonstrated or required annual pace of
performance improvement over time. The resulting levels of demand reduction are consistent
with recent studies of achievable demand reduction potential conducted throughout the U.S. For
these reasons, the demand-side energy efficiency plan scenario represents a reasonable
assumption about the level of demand-side energy efficiency investments that may be
encouraged in response to the final CPP.
For the illustrative demand-side energy efficiency plan scenario, electricity demand
reductions for each state for each year are developed by ramping up from a historical basis66 to a
target annual incremental demand reduction rate of 1.0 percent of electricity demand over a
period of years starting in 2020, and maintaining that rate throughout the modeling horizon.67
Nineteen leading states either have achieved, or have established requirements that will lead
them to achieve, this rate of incremental electricity demand reduction on an annual basis. Based
on historic performance and existing state requirements, for each state the pace of improvement
from the state's historical incremental demand reduction rate is set at 0.2 percent per year,
beginning in 2020, until the target rate of 1.0 percent is achieved. States already at or above the
1.0 percent target rate are assumed to achieve a 1.0 percent rate beginning in 2020 and sustain
that rate thereafter.68 The incremental demand reduction rate for each state, for each year, is used
to derive cumulative annual electricity demand reductions based upon information about the
average life of energy efficiency measures and the distribution of measure lives across energy
66 The historical basis of the percentage of reduced electricity consumption differs for each state and is drawn from
the data reported in Energy Information Administration (EIA) Form 861, 2013, available at
http://www.eia.gov/electricity/data/eia861/.
67 The incremental demand reduction percentage is applied to the previous year's electricity demand for the state.
68 This assumption may result in underestimating electricity demand reductions in these states in the illustrative plan
scenarios.
3-13
-------
efficiency programs.69 The cumulative annual electricity demand reduction derived using this
methodology is used to adjust base case electricity demand levels in the illustrative plan
approach modeling.
To reflect the implementation of the illustrative energy efficiency plan scenario in
modeling, the IPM base case electricity demand was adjusted exogenously to reflect the
estimated future-year demand reductions calculated as described above. State-level demand
reductions were scaled up to account for transmission losses and applied to base case generation
demand in each model year to derive adjusted demand for each state, reflecting the energy
efficiency plan scenario energy reductions. The demand adjustments were applied proportionally
across all segments (peak and non-peak) of the load duration curve.70 To reflect the adjusted
state-level demand within IPM model regions that cross state borders, energy reductions from a
bisected state were distributed between the applicable IPM model regions using a distribution
approach based on reported sales in 2013 as a proxy for the distribution of energy efficiency
investment opportunities.
Table 3-2 summarizes the results of the illustrative demand-side energy efficiency plan
scenario at the national level.
Table 3-2. Demand-Side Energy Efficiency Plan Scenario: Net Cumulative Demand
Reductions [Contiguous U.S.] (GWh and as Percent of BAU Sales)
Net Cumulative Demand Reduction (GWh)
Net Cumulative Demand Reduction as Percent of BAU Sales
2020
23,150
0.59%
2025
194,126
4.81%
2030
327,092
7.83%
Source: U.S. EPA. 2015. Technical Support Document (TSD) for the Final Carbon Pollution Emission Guidelines
for Existing Stationary Sources: Electric Utility Generating Units. Demand-Side Energy Efficiency.
69 The average life of demand-side energy efficiency measures used is 10.2 years. This average is represented using
a four-tier distribution of measure lives ranging from 6.5 to 21.2 years. This approach is based on 2015 analysis by
Lawrence Berkeley National Laboratory and is discussed in detail in section 8.2.6 of the Demand-Side Energy
Efficiency TSD.
70 Details and reasoning for this assumption are included in U.S. EPA. 2015. Technical Support Document (TSD)
for the Final Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating
Units. Demand-Side Energy Efficiency.
3-14
-------
3.7.2 Demand-Side Energy Efficiency Costs71
Total costs of achieving the demand-side energy efficiency plan scenario for each year
were calculated exogenous to the power sector modeling. The power system cost impacts
resulting from the illustrative plan approach analyses were captured within IPM and include the
effects of reduced demand levels driven by the energy efficiency scenario discussed above. The
integration of the exogenously calculated demand-side energy efficiency scenario costs with the
power system cost impacts of the illustrative plan approaches are discussed in section 3.9.2. In
addition to the demand reduction results, the demand-side energy efficiency costs were based
upon an estimate of the total first-year cost of saved energy (i.e., reduced demand), the average
life of the demand-side energy efficiency measures, the distribution of those measure lives, and
cost factors as greater levels of demand reductions are achieved. The total first-year cost of saved
energy accounts for both the costs of the demand-side energy efficiency programs, known as the
program costs, and the additional cost to electricity consumers participating in the program (e.g.,
purchasing a more energy efficient technology), known as the participant costs.
To calculate total annualized demand-side energy efficiency costs, first-year costs for
each year for each state were levelized (at 3 percent and 7 percent discount rates) over the
estimated distribution of measure lives and the results summed for each year for each state. For
example, the 2025 estimate of annualized energy efficiency cost includes levelized value of first-
year costs for energy efficiency investments made in 2020 through 2025. The annualized costs
rise in each analysis year as additional first-year costs are incurred. The annualized cost results
are summarized below in Table 3-3. The total levelized cost of saved energy was calculated
based upon the same inputs and using a 3 percent discount rate resulted in national average
values of 9.2 cents per kWh in 2020, 8.6 cents per kWh in 2025, and 8.1 cents per kWh in 2030.
Table 3-3. Annualized Cost of Demand-Side Energy Efficiency Plan Scenario (at discount
rates of 3 percent and 7 percent, billions 2011$)
Discount Rate
Source: U.S
Existing
EPA.
Stationary
at 3 percent
at 7 percent
2015. Technical Support
Sources:
Electric Utility
2020
2.1
2.6
Document (TSD)
Generating Units
2025
16.7
20.6
2030
26.3
32.5
the Final Carbon Pollution Emission
Demand-Side Energy
Efficiency.
Guidelines for
71 For a more detailed discussion of the demand-side energy efficiency cost analysis, refer to the Demand-Side
Energy Efficiency TSD.
3-15
-------
The funding for demand-side energy efficiency programs (to cover program costs) is
typically collected through a standard per kWh surcharge to the ratepayer; the regional retail
price impacts analyzed from this RIA's illustrative plan approaches assumes the recovery of
these program costs through the following procedure.72 For each state, the first-year energy
efficiency program costs are calculated for each year. These costs were distributed between the
applicable IPM regions using an approach based on reported sales in 2012 as a proxy for the
distribution of energy efficiency investment opportunities. These regionalized energy efficiency
program costs were then incorporated into the regional retail price calculation as discussed in
section 3.9.9.73 The U.S. EPA's_Demand-Side Energy Efficiency Technical Support Document
(U.S. EPA 2015) provides complete details on the calculations of annualized costs and first-year
costs as well as comprehensive results (by state, by year) for the illustrative demand-side energy
efficiency plan scenario.
3.8 Monitoring, Reporting, and Recordkeeping Costs
EPA projected monitoring, reporting and recordkeeping costs for both state entities and
affected EGUs for the compliance years 2020, 2025, and 2030. In calculating the costs for state
entities, EPA estimated personnel costs to oversee compliance, and review and report annually to
EPA on program progress relative to meeting the state's reduction goal. To calculate the national
costs, EPA estimated that 47 states and 1,028 facilities would be affected.
The EPA estimated that the majority of the cost to EGUs would be in calculating net
energy output, which is needed whether the state plan utilizes a rate-based or a mass-based
limit. Since the majority of EGUs do have some energy usage meters or other equipment
available to them, EPA believes a new system for calculating net energy output is not needed.
Under the final guidelines, states are required to use monitoring and reporting requirements for
their affected EGUs to ensure that the sources are meeting the appropriate CO2 emission
performance rates or emission goals.
72 The full retail price analysis method is discussed in section 3.7.9 of this chapter.
73 The effect on equilibrium supply and demand of electricity due to changing retail rates to fund energy efficiency
programs is not captured in the IPM modeling.
3-16
-------
The EPA has made it a priority to streamline reporting and monitoring requirements. In
this rale, the EPA is making implementation as efficient as possible for both the states and the
affected EGUs by allowing state plans to utilize the current monitoring and recordkeeping
requirements and pathways that have already been well established in other EPA rulemakings.
For example, under the Acid Rain Program's continuous emissions monitoring, 40 CFR Part 75,
the EPA has established requirements for the majority of the EGUs that would be affected by a
11 l(d) state plan to monitor CCh emissions and report that data using the Emissions Collection
and Monitoring Plan System (ECMPS). Additionally since the CO2 hourly data is already
reported to the EPA's ECMPS there is no additional burden associated with the reporting of that
data. Since the ECMPS pathway is already in place, the EPA will allow for states to utilize the
ECMPS system to facilitate the data reporting of the additional net energy output data required
under the emission guidelines. However, because the Acid Rain Program does not require net
energy output to be reported, there is some additional burden (Shown in Table 3-4) in updating
an affected EGUs monitoring system to be able to report the associated net energy output of an
affected EGU.
The EPA estimates that it would take three working months for a technician to retrofit
any existing energy meters to meet the requirements set in the state plan. Additionally EPA
believes that 50 hours will be needed for each EGU operator to read the rule and understand how
the facility will comply with the rale, based on an average reading rate of 100 words per minute
and a projected rale word count of 300,000 words.74 Also, after all modifications are made at a
facility to measure net energy output, each EGU's Data Acquisition System (DAS) would need
to be upgraded to supply the rate-based emissions value to either the state or EPA's Emissions
Collection and Monitoring Plan System (ECMPS). Note the costs to develop net energy output
monitoring and to upgrade each facility's DAS system are one-time costs incurred in 2020.
Recordkeeping and reporting costs substantially decrease for the period 2021-2030. The
projected costs for 2020, 2025, and 2030 are summarized below.
74 According to one source, the average person can proofread at about 200 words per minute on paper and 180 words
per minute on a monitor. (Source: Ziefle, M. 1988. "Effects of Display Resolution on Visual Performance." Human
Factors 40(4):554-68). Due to the highly technical nature of the rule requirements in subpart UUUU, a more
conservative estimate of 100 words per minute was used to determine the burden estimate for reading and
understanding rule requirements.
3-17
-------
In calculating the cost for states to comply, EPA estimates that each state will rely on the
equivalent of two full time staff to oversee program implementation, assess progress, develop
possible contingency measures, perform state plan revisions and host the subsequent public
meetings if revisions are indeed needed, download data from the ECMPS for their annual
reporting and develop their annual EPA report. The burden estimate was based on an analysis of
similar tasks performed under the Regional Haze Program, whereby states were required to
develop their list of eligible sources, draft implementation plans, revise initial drafts, identify
baseline controls, identify data gaps, identify initial strategies, conduct various reviews, and
manage their programs. A total estimate of 78,000 hours of labor performed by seven states over
a three-year period resulted in 3,714 hours per year, per entity. Due to the nature of this final rule
whereby we believe the air office and the energy office will both be involved in performing the
above-mentioned tasks, we rounded up to the equivalent of two full time staff, which totaled
4,160 hours per year.75 Table 3-4 shows estimates of the annual state and industry respondent
burden and costs of reporting and recordkeeping for 2020, 2025 and 2030.
Table 3-4. Years 2020, 2025 and 2030: Summary of State and Industry Annual Respondent
Burden and Cost of Reporting and Recordkeeping Requirements (2011$)
Nationwide
Totals
Total Annual
Labor Burden
(Hours)
T . , Total
Total . ,. .
. , Annuahzed
Annual .
Labor Costs .
Costs
Total
Annual
O&M Costs
Total
Annual- Total Annual
ized Respondent Costs
Costs
State
Year 2020
Year 2025
Year 2030
195,520
208,320
208,320
13,838,429
14,744,381
14,744,381
0
0
0
34,545
23,500
23,500
34,545
23,500
23,500
13,872,974
14,767,881
14,767,881
Industry
Year 2020
Year 2025
Year 2030
581,848
0
0
49,959,446
0
0
0
0
0
1,532,000
0
0
1,532,500
0
0
51,491,446
0
0
Total
Year 2020
Year 2025
Year 2030
777,368
208,320
208,320
63,797,875
14,744,381
14,744,381
0
0
0
1,566,545
23,500
23,500
1,566,545
23,500
23,500
65,364,420
14,767,881
14,767,881
75 Renewal of the ICR for the Regional Haze Rule, Section 6(a) Tables 1 through 4 based on 7 states' burden. EPA-
HQ-OAR-2003-0162-0001.
3-18
-------
3.9 Projected Power Sector Impacts
The following sections present projected impacts from the two illustrative scenarios
described above. The tables present impacts from 2020 (prior to the initial compliance year),
2025 (representative of the interim compliance period), and 2030 (representative of the final
compliance period). The narrative focuses on results during the initial and final compliance
periods.
3.9.1 Projected Emissions
Under the rate-based approach, EPA projects annual COi reductions of 3 percent below
the base case in 2020, 11 percent below the base case in 2025, and 19 percent below base case
projections in 2030 (reaching 28 percent to 32 percent below 2005 emissions76 in 2025 and 2030,
respectively). For the mass-based approach, EPA projects annual COi reductions of 4 percent
below the base case in 2020, 12 percent below the base case in 2025 and 19 percent below base
case projections in 2030 (reaching 29 percent to 32 percent below 2005 emissions77 in 2025 and
2030, respectively).78
Table 3-5. Projected CCh Emission Impacts, Relative to Base Case
Base Case
Rate-based
Mass-based
CCh Emissions
(million short tons)
2020 2025 2030
2,155 2,165 2,227
2,085 1,933 1,812
2,073 1,901 1,814
CCh Emissions: Change
from Base Case
(million short tons)
2020 2025 2030
-69 -232 -415
-81 -265 -413
CCh Emissions: Percent
Change from Base Case
2020 2025 2030
-3% -11% -19%
-4% -12% -19%
Source: Integrated Planning Model run by EPA, 2015
76 For purposes of these calculations, EPA has used historical COi emissions from eGRID for 2005, which reports
EGU emissions as 2,683 million short tons in the contiguous U.S.
77 For purposes of these calculations, EPA has used historical COi emissions from eGRID for 2005, which reports
EGU emissions as 2,683 million short tons in the contiguous U.S.
78 EPA also analyzed a mass-based scenario without any set-asides using IPM, which produced a 2030 emission
reduction estimate of 31 percent, relative to 2005 levels (approximately a 1 percent erosion of emission reductions
due to leakage to new sources of emissions, relative to both the mass-based scenario that includes the RE set-aside,
and the rate-based scenario. This equates to approximately 24 million short tons of COi.). The scenario can be found
in the docket for the final rule, and is called "Mass-based without set-aside."
3-19
-------
Table 3-6. Projected CCh Emission Impacts, Relative to 2005
Base Case
Rate-based
Mass-based
CCh Emissions
(million short tons)
2005
2,683
-
CCh Emissions: Change
from 2005
(million short tons)
2020 2025 2030
-528 -518 -456
-598 -750 -871
-610 -782 -869
CCh Emissions: Percent
Change from 2005
2020 2025 2030
-20% -19% -17%
-22% -28% -32%
-23% -29% -32%
Source: Integrated Planning Model run by EPA, 2015
Under the rate-based illustrative plan approach, EPA projects a 14 percent reduction of
862, 13 percent reduction of NOx, and all percent reduction of mercury in 2025, and a 24
percent reduction of SOi, 22 percent reduction of NOx, and a 17 percent reduction of mercury in
2030. Under the mass-based illustrative plan approach, EPA projects a 15 percent reduction of
SOi, 16 percent reduction of NOx, and a 12 percent reduction of mercury in 2025, and a 24
percent reduction of SO2, 22 percent reduction of NOx, and a 16 percent reduction of mercury in
2030. The projected non-COi reductions are summarized below in Table 3-7.
Table 3-7. Projected Non-CCh Emission Impacts, 2020-2030
Base Case
Rate-based
Mass-based
Rate-based
Mass-based
2020
SOi (thousand short tons)
NOx (thousand short tons)
Hg (short tons)
1,311
1,333
6.6
1,297
1,282
6.4
1,257
1,272
6.4
-1.0%
-3.8%
-2.8%
-4.1%
-4.5%
-3.3%
2025
SOi (thousand short tons)
NOx (thousand short tons)
Hg (short tons)
1,275
1,302
6.6
1,097
1,138
5.9
1,090
1,100
5.8
-14.0%
-12.6%
-10.8%
-14.5%
-15.6%
-12.2%
2030
SOi (thousand short tons)
NOx (thousand short tons)
Hg (short tons)
1,314
1,293
6.8
996
1,011
5.6
1,034
1,015
-24.2%
-21.8%
5.8 -17.2%
-21.3%
-21.5%
-15.6%
Source: Integrated Planning Model run by EPA, 2015. For this RIA, we did not estimate changes in emissions of
directly emitted particles (PM2.s).
While the EPA has not quantified the climate impacts of non-COi emissions changes or
CO2 emissions changes outside the electricity sector for the final emissions guidelines, the
Agency has analyzed the potential changes in upstream methane emissions from the natural gas
and coal production sectors that may result from the illustrative approaches examined in this
3-20
-------
RIA. The EPA assessed whether the net change in upstream methane emissions from natural gas
and coal production is likely to be positive or negative. The EPA also assessed the potential
magnitude of changes relative to CO2 emissions reductions anticipated at power plants. This
assessment included CO2 emissions from the flaring of methane, but did not evaluate potential
changes in other combustion-related CO2 emissions, such as emissions associated with drilling,
mining, processing, and transportation in the natural gas and coal production sectors. This
analysis found that the net upstream methane emissions from natural gas systems and coal mines
and CO2 emissions from flaring of methane will likely decrease under the final emissions
guidelines. Furthermore, the changes in upstream methane emissions are small relative to the
changes in direct CO2 emissions from power plants. The projections include voluntary and
regulatory activities to reduce emissions from coal mining and natural gas and oil systems,
including the 2012 Oil and Natural Gas NSPS. In addition, the EPA plans to issue a proposed
rule later this summer that would build on its 2012 Oil and Gas NSPS. When these standards are
finalized and implemented, they would further reduce projected emissions from natural gas and
oil systems. The technical details supporting this analysis can be found in the Appendix to this
chapter.
3.9.2 Projected Compliance Costs
The power industry's "compliance costs" are represented in this analysis as the change in
electric power generation costs between the base case and illustrative CPP scenarios, including
the cost of demand-side energy efficiency programs and measures and monitoring, reporting, and
recordkeeping (MR&R) costs. The system costs reflect the least cost power system outcome in
which the sector employs all the flexibilities assumed in the modeling, as discussed above, and
pursues the most cost-effective emission reduction opportunities in order to meet the rate- and
mass-based goals, as represented in the illustrative plan scenarios. In simple terms, these costs
are an estimate of the increased power industry expenditures required to meet demand
projections while complying with state goals, including the total demand-side energy efficiency
costs.79 The compliance costs for the final emissions guidelines for EGUs in the contiguous U.S.
79 The compliance costs also capture the effect of changes in equilibrium fuel prices on the expenditures of the
electricity sector to serve demand.
3-21
-------
states is forecast using IPM. The cost of demand-side energy efficiency programs assumed in the
IPM analysis are reported in section 3.7.2.
EPA projects that the annual compliance cost of the rate-based illustrative plan scenario
are $2.4 billion in 2020, $1.1 billion in 2025, and $8.5 billion in 2030 (Table 3-8). The annual
compliance cost of the mass-based illustrative plan approach are estimated to be $1.4 billion in
2020, $3.0 billion in 2025, and $5.1 billion in 2030. The different patterns of incremental cost in
each of these scenarios over 2020-2030 are consistent with the differences in the projected
pattern of gas use and price in these scenarios, consistent with the differences in the projected
pattern of gas use and price in these scenarios. The annual compliance cost is the projected
additional cost of complying with the rule in the year analyzed and reflects the net difference in
the sum of the annualized cost of capital investment in new generating sources and heat rate
improvements at coal steam facilities,80 the change in the ongoing costs of operating pollution
controls, the change in expenditures on various fuels (inclusive of changes in the price of these
fuels), demand-side energy efficiency measures, and other actions associated with compliance.
Relative to the base case, we expect a decrease in the total cost to generate sufficient supply for
demand, which, together with the costs of demand-side energy efficiency measures, we project
will result in net cost estimates of $8.4 billion in 2030 for the rate-based scenario and $5.1 billion
for the mass-based scenario.
Table 3-8. Annualized Compliance Costs Including Monitoring, Reporting and
Recordkeeping Costs Requirements (billions of 2011$)
2020 2025 2030
Rate-based $2.5 $1.0 $8.4
Mass-based $1.4 $3.0 $5.1
Source: Integrated Planning Model run by EPA, 2015, with post-processing to account for exogenous demand-side
energy efficiency costs and monitoring, reporting, and recordkeeping costs.
In order to contextualize EPA's projection of the additional costs in 2030 across the two
illustrative plan approaches evaluated in this RIA, it is useful to compare these incremental cost
estimates to total projected power sector expenditures. The power sector is expected in the base
case to expend over $201 billion in 2030 to generate, transmit, and distribute electricity to end-
use consumers. In 2014, according to EIA, the power sector generated $389 billion in revenue
80 See Chapter 2 of the GHG Mitigation Measures TSD and EPA's Base Case using IPM (v5.15) documentation,
available at: http://www.epa.gov/powersectormodeling
3-22
-------
from retail sales of electricity. For context, the projected costs of compliance with the final rale
amount to a 4 percent increase in the cost of meeting electricity demand, while securing public
health and welfare benefits that are several times greater (as described in Chapters 4 and 8).
The following example uses projected results for the year 2030 to illustrate how different
components of estimated expenditures are combined to form the full compliance costs presented
in Table 3-8. In Table 3-9, we present the IPM modeling results for the two illustrative plan
scenarios in 2030 (as well as 2020 and 2025). The results show that annualized expenditures
required to supply enough electricity to meet demand decline by $18 billion (rate) and $21
billion (mass) from the base case in 2030. This incremental decline is a net outcome of two
simultaneous effects that move in opposite directions. First, imposing the CO2 constraints
represented by each illustrative plan scenario on electric generators would, other things equal,
result in an incremental increase in expenditures to supply any given level of electricity.
However, once electricity demand is reduced to reflect demand-side energy efficiency
improvements, there is a substantial reduction in the expenditures needed to supply a
correspondingly lower amount of electricity demand.
Table 3-9. Total Power Sector Generating Costs (IPM) (billions 2011$)
Base Case
Rate-based
Mass-based
2020
$166.5
$166.8
$165.7
2025
$178.3
$162.6
$164.6
2030
$201.3
$183.3
$180.1
Source: Integrated Planning Model run by EPA, 2015
In order to reflect the full compliance cost attributable to the CPP scenarios, it is
necessary to include the annualized expenditures needed to secure the demand-side energy
efficiency improvements. As described in section 3.7.2, EPA has estimated these energy
efficiency-related expenditures to be $26.3 billion in 2030 (using a 3 percent discount rate). The
energy efficiency-related expenditures include costs incurred by parties administering energy
efficiency programs and costs incurred by participants in those programs. As a result, this
analysis finds the cost of the rate-based and mass-based illustrative plan approaches in 2030 to
be $8.4 billion and $5.1 billion, respectively.
3.9.3 Projected Compliance Actions for Emissions Reductions
Heat Rate Improvements (HRI): EPA analysis assumes that the existing coal steam electric
generating fleet has, on average, the ability to improve operating efficiency (i.e., reduce the
3-23
-------
average net heat rate, or the Btu of fuel energy needed to produce one kWh of net electricity
output). All else held constant, an HRI allows the EGU to generate the same amount of
electricity using less fuel. The decrease in required fossil fuel results in a lower output-based
CO2 emissions rate (Ibs/MWh), as well as a lower variable cost of electricity generation. In the
modeling conducted for these illustrative plan approaches, coal boilers have the choice to
improve heat rates by 4.3 percent in the eastern illustrative compliance region, 2.1 percent in the
western illustrative compliance region, and 2.3 percent in Texas, all at a capital cost of $100 per
kW.81 The option for heat rate improvement is only made available in the illustrative plan
approaches during the compliance period, in response to the final rule.
The majority of existing coal boilers are projected to adopt the aforementioned heat rate
improvements. Of the 183 GW of coal projected to operate in 2030, EPA projects that 99 GW of
existing coal steam capacity (greater than 25 MW) will improve operating efficiency (i.e., reduce
the average net heat rate) under the rate-based approach by 2030. Under the mass-based
approach, EPA projects that 88 GW of the 174 GW of coal projected to operate in 2030 will
improve operating efficiency by 2030.
Generation Shifting: Another approach for reducing the average emission rate from existing units
is to shift some generation from more CCh-intensive generation to less COi-intensive generation.
Compared to the base case, existing coal steam capacity is, on average, projected to operate at a
lower capacity factor for both illustrative plan approaches. Under the illustrative rate-based plan
approach, the average 2030 capacity factor is 69 percent, and under the mass-based approach, the
average capacity factor for existing coal steam is 75 percent. Existing natural gas combined cycle
units, which are less carbon-intensive than coal steam capacity on an output basis, operate at
noticeably higher capacity factor under both illustrative plan approaches, on average. The
utilization of existing natural gas combined cycle capacity is lower than the BSER level of 75
percent82 on an annual average basis in these illustrative plan approaches, reflecting the fact that,
81 The option for heat rate improvement is only made available in the illustrative plan scenarios, and is not available
in the base case. For an explanation of the regional differences in average ability to improve heat rates, see GHG
Mitigation Measures TSD.
82 See preamble section V.D.
3-24
-------
in practice, the most cost-effective CO2 reduction strategies to meet each state's goal may not
require that each building block be achieved in entirety. See Table 3-10.
Table 3-10. Projected Capacity Factor of Existing Coal Steam and Natural Gas
Combined Cycle Capacity
Base Case
Rate-based
Mass-based
Existing Coal Steam
2020 2025 2030
77% 76% 79%
78% 75% 69%
78% 75% 75%
Existing Natural Gas Combined Cycle
2020 2025
54% 56%
56% 60%
56% 58%
2030
51%
61%
54%
Source: Integrated Planning Model run by EPA, 2015
Demand-Side Energy Efficiency: Another approach for reducing emissions from affected EGUs
is to consider reductions in demand attributable to demand-side energy efficiency measures as
discussed in section 3.7. In the illustrative plan approaches presented in this RIA, each state is
credited for total demand-side energy efficiency implemented in, or procured by, that state,
consistent in aggregate with the state-by-state demand reductions that are represented by the
demand-side energy efficiency scenario discussed in section 3.7.1.
Deployment of Cleaner Generating Technologies: Another key opportunity to reduce emissions
from existing sources is to build more lower- or zero-emitting generating resources, in particular
renewable energy. These sources of electricity, including wind and solar, can displace higher
emitting existing sources, may be procured for compliance with the state goals in the rate-based
illustrative scenario, and are further incentivized as a generation option in the mass-based
illustrative scenario as they are not subject to the mass-based constraint and may receive the
renewable set-aside. Increased deployment results in CO2 reductions in both rate-based and
mass-based approaches. See sections below discussing projected impacts on generation mix and
capacity.
3.9.4 Projected Generation Mix
Table 3-11 and Figure 3-2 show the generation mix in the base case and under the two
illustrative plan approaches. In both scenarios, total generation declines relative to the base case
as a result of the reduction in total demand attributable to the demand-side energy efficiency
applied in the illustrative scenarios, by 5 percent in 2025 and 8 percent in 2030.
3-25
-------
Under the rate-based scenario, coal-fired generation is projected to decline 12 percent in
2025, and natural-gas-fired generation from existing combined cycle capacity is projected to
increase 5 percent relative to the base case. The coal-fired fleet in 2030 generates 23 percent less
than in the base case, while natural-gas-fired generation from existing combined cycles increases
18 percent relative to the base case. Gas-fired generation from new combined cycle capacity
decreases in 2025 and 2030, consistent with the decrease in new capacity (see section 3.9.6).
Relative to the base case, generation from non-hydro renewables decreases 1 percent in 2025 and
increases 9 percent in 2030.
Similarly, under the mass-based scenario, coal-fired generation is projected to decline 15
percent in 2025, and natural-gas-fired generation from existing combined cycle capacity is
projected to increase 2 percent relative to the base case. The coal-fired fleet in 2030 generates 22
percent less than in the base case, while natural-gas-fired generation from existing combined
cycles increases 5 percent relative to the base case. Gas-fired generation from new combined
cycle capacity decreases 8 percent and 36 percent relative to the base case in 2025 and 2030,
respectively. Relative to the base case, generation from non-hydro renewables decreases 3
percent in 2025 and increases 8 percent in 2030.
The results presented in these illustrative compliance scenarios suggest that existing
nuclear generation could be slightly more competitive under a mass-based implementation than
under a rate-based implementation, because the former tends to create more wholesale price
support for those generators. These scenarios do not include potential approaches that states can
take to incentivize zero-carbon baseload power.
3-26
-------
Table 3-11. Generation Mix (thousand GWh)
Base Case
Rate-based Mass-based
Rate-based Mass-based
2020
Coal
NG Combined Cycle (existing)
NG Combined Cycle (new)
Combustion Turbine
Oil/Gas Steam
Non-Hydro Renewables
Hydro
Nuclear
Other
Total
1,462
1,111
33
15
51
393
310
798
18
4,190
1,391
1,126
53
20
51
399
311
792
18
4,160
1,374
1,132
69
17
50
385
310
804
18
4,159
-5%
1%
61%
39%
0%
2%
0%
-1%
0%
-1%
-6%
2%
111%
14%
-1%
-2%
0%
1%
0%
-1%
2025
Coal
NG Combined Cycle (existing)
NG Combined Cycle (new)
Combustion Turbine
Oil/Gas Steam
Non-Hydro Renewables
Hydro
Nuclear
Other
Total
1,428
1,152
113
23
39
417
340
799
17
4,328
1,256
1,206
53
30
21
414
340
791
17
4,128
1,217
1,179
104
34
19
404
340
804
18
4,118
-12%
5%
-53%
31%
-46%
-1%
0%
-1%
0%
-5%
-15%
2%
-8%
46%
-52%
-3%
0%
1%
0%
-5%
2030
Coal
NG Combined Cycle (existing)
NG Combined Cycle (new)
Combustion Turbine
Oil/Gas Steam
Non-Hydro Renewables
Hydro
Nuclear
Other
Total
Note: "Other" mostly includes generation
EPA, 2015
1,466
1,042
324
22
22
450
340
783
17
4,467
fromMSW
1,131
1,230
100
27
11
488
341
111
17
4,122
and fuel cells.
1,144
1,090
207
32
11
485
340
785
17
4,110
-23%
18%
-69%
21%
-52%
9%
0%
-1%
0%
-8%
Source: Integrated Planning Model
-22%
5%
-36%
46%
-53%
8%
0%
0%
0%
-8%
run by
3-27
-------
1.2&S
2D2C
2C30
Figure 3-2 Generation Mix (thousand GWh)
Source: Integrated Planning Model run by EPA, 2015
Under both the rate-based and mass-based approaches, the projected rate of change in
coal-fired generation is consistent with recent historical declines in coal-fired generation.
Additionally, under both of these approaches, the trends for all other types will remain consistent
with what their trends would be in the absence of this rule. Specifically, natural-gas fired
generation and renewables would be expected to increase without this rule, and both are
expected to increase under this rule, with renewables increasing at a somewhat greater rate than
in the absence of this rule; and nuclear, oil-fired, and other types of generation are expected to be
little impacted by this rule generation mix is consistent with recent declines in coal-fired
generation and increases in gas-fired generation. See Figures 3-3, 3-4, and 3-5.
3-28
-------
2.5
2.0
,.
,
1.5
*
*
c
o
!
:*
05 ;..
**% J|«i
0.0
1990 1995 2000 2005 2010 2015 2020 2025 2030
Coal
Natural Gas
Nuclear
Hydro
Petroleum
Non-hydro
renewables
Other
Figure 3-3. Nationwide Generation: Historical (1990-2014) and Base Case Projections
(2020, 2025, 2030)
Sources: Historic data (i.e., 1990-2014): U.S. Energy Information Administration, June 2015 Monthly Energy
Review, Table 7.2a Electricity Net Generation: Total (All Sectors), Available at
. Projected data (i.e., 2020, 2025, 2030): Integrated Planning Model,
2015. Notes: Historic and projected data include generation from the power, industrial, and commercial sectors.
Historic data from U.S. EIA reflects all cogeneration, while projections from the Integrated Planning Model reflect
net cogeneration.
2.5
2.0
i.5
0
,»
*
.
!
*
05 '::
'«
0.0
1990 1995 2000 2005 2010 2015 2020 2025 2030
Coal
Natural Gas
Nuclear
Hydro
Petroleum
Non-hydro
renewables
Other
Figure 3-4. Nationwide Generation: Historical (1990-2014) and Rate-Based Illustrative
Plan Approach Projections (2020, 2025, 2030)
Sources: Historic data (i.e., 1990-2014): U.S. Energy Information Administration, June 2015 Monthly Energy
Review, Table 7.2a Electricity Net Generation: Total (All Sectors), Available at
. Projected data (i.e., 2020, 2025, 2030): Integrated Planning Model,
2015. Notes: Historic and projected data include generation from the power, industrial, and commercial sectors.
Historic data from U.S. EIA reflects all cogeneration, while projections from the Integrated Planning Model reflect
net cogeneration.
3-29
-------
2.5
2.0 .*
. *
*** ' *
5 1.5
(D . «
1
~ 1 n
5
f §'***
_*** _ *
OR *
;! *
nn «^s^-^^
Coal
Natural Gas
Nuclear
Hydro
Petroleum
Non-hydro
renewables
Other
1990 1995 2000 2005 2010 2015 2020 2025 2030
Figure 3-5. Nationwide Generation: Historical (1990-2014) and Mass-Based Illustrative
Plan Approach Projections (2020, 2025, 2030)
Sources: Historic data (i.e., 1990-2014): U.S. Energy Information Administration, June 2015 Monthly Energy
Review, Table 7.2a Electricity Net Generation: Total (All Sectors), Available at
. Projected data (i.e., 2020, 2025, 2030): Integrated Planning Model,
2015. Notes: Historic and projected data include generation from the power, industrial, and commercial sectors.
Historic data from U.S. EIA reflects all cogeneration, while projections from the Integrated Planning Model reflect
net cogeneration.
3.9.5 Projected Incremental Retirements
Relative to the base case, about 23 GW of additional coal-fired capacity is projected to be
uneconomic to maintain by 2025 under the rate-based illustrative scenario, increasing to 27 GW
in 2030 (about 11-13 percent respectively of all coal-fired capacity projected to be in service in
the base case). Under the mass-based scenario, about 29 GW of additional coal-fired capacity is
projected to be uneconomic to maintain by 2025, increasing to 38 GW by 2030 (about 14-19
percent respectively of all coal-fired capacity projected to be in service in the base case).
Capacity changes from the base case are shown in Table 3-12.83
83
EPA examined the implications of the illustrative plan scenarios for concerns about regional resource adequacy
and the potential for concerns about reliability. This examination can be found in U.S. EPA. 2015. Technical
Support Document (TSD) the Final Carbon Pollution Emission Guidelines for Existing Stationary Sources:
Electric Utility Generating Units. Resource Adequacy and Reliability Analysis.
3-30
-------
Table 3-12. Total Generation Capacity by 2020-2030 (GW)
Base Case
Rate-based
Mass-based
Rate-based
Mass-based
2020
Coal
NG Combined Cycle (existing)
NG Combined Cycle (new)
Combustion Turbine
Oil/Gas Steam
Non-Hydro Renewables
Hydro
Nuclear
Other
Total
208
233
4
141
88
130
106
100
5
1,016
195
231
7
137
81
132
106
100
5
994
193
232
9
137
80
128
106
101
5
992
-6%
-1%
62%
-3%
-8%
1%
0%
-1%
0%
-2%
-7%
0%
113%
-3%
-9%
-2%
0%
1%
0%
-2%
2025
Coal
NG Combined Cycle (existing)
NG Combined Cycle (new)
Combustion Turbine
Oil/Gas Steam
Non-Hydro Renewables
Hydro
Nuclear
Other
Total
208
233
15
143
82
139
112
100
5
1,037
187
231
7
138
71
137
112
99
5
988
181
232
14
137
69
134
112
101
5
985
-10%
-1%
-52%
-4%
-14%
-1%
0%
-1%
0%
-5%
-13%
0%
-9%
-4%
-16%
-3%
0%
1%
0%
-5%
2030
Coal
NG Combined Cycle (existing)
NG Combined Cycle (new)
Combustion Turbine
Oil/Gas Steam
Non-Hydro Renewables
Hydro
Nuclear
Other
Total
207
233
44
147
82
154
112
99
5
1,082
183
231
14
138
70
174
112
98
5
1,025
174
232
27
136
67
171
112
99
5
1,024
-11%
-1%
-68%
-6%
-15%
13%
0%
-1%
0%
-5%
-16%
0%
-38%
-7%
-18%
11%
0%
0%
0%
-5%
Source: Integrated Planning Model run by EPA, 2015
3.9.6 Projected Capacity Additions
Due largely to the electricity demand reduction attributable to the demand-side energy
efficiency improvements applied in the illustrative scenarios, the EPA projects less new natural
3-31
-------
gas combined cycle capacity built under the rate-based scenario than is built in the base case over
the period covered by the rule. While this new NGCC capacity cannot be directly counted
towards the average emissions rate used for compliance in the rate-based approach, it can
displace some generation from covered sources and thus indirectly lower the average emissions
rate from covered sources. Conversely, the EPA projects an overall increase in new renewable
capacity. New non-hydro renewables are able to contribute their generation to the average
emissions rate in each state or region.
Under the rate-based illustrative scenario, new natural gas combined cycle capacity is
projected to decrease by 8 GW in 2025 and 30 GW in 2030 (52 percent and 68 percent decrease
relative to the base case). New renewable capacity is projected to decrease by about 2 GW (3
percent decrease) below the base case in 2025, and increase by 20 GW (27 percent increase) by
2030.
Under the mass-based illustrative scenario, new natural gas combined cycle capacity is
projected to decrease by 1 GW in 2025 and decrease by 17 GW in 2030 (a 9 percent and 38
percent decrease relative to the base case). New renewable capacity is projected to decrease 4
GW (7 percent) relative to the base case in 2025, and increase 18 GW (24 percent increase) by
2030.
Table 3-13. Projected Capacity Additions, Gas (GW)
Base Case
Rate-based
Mass-based
Cumulative Capacity Additions: Gas
Combined Cycle
2020
4.4
7.1
9.3
2025
14.9
7.1
13.6
2030
44.0
13.9
27.2
Incremental Cumulative Capacity
Additions: Gas Combined Cycle
2020 2025 2030
2.7 -7.8 -30.1
4.9 -1.3 -16.8
Source: Integrated Planning Model run by EPA, 2015
3-32
-------
Table 3-14. Projected Capacity Additions, Renewable (GW)
Base Case
Rate-based
Mass-based
Cumulative Capacity Additions:
Renewables
2020
39.1
40.5
36.7
2025
59.1
57.4
54.9
2030
74.1
94.4
91.9
Incremental Cumulative Capacity
Additions: Renewables
2020 2025 2030
1.4 -1.8 20.2
-2.4 -4.2 17.8
Source: Integrated Planning Model run by EPA, 2015
3.9.7 Projected Coal Production and Natural Gas Use for the Electric Power Sector
Coal production is projected to decrease in 2025 and beyond in the illustrative scenarios
due to (1) improved heat rates (generating efficiency) at existing coal units, (2) electricity
demand reduction attributable to demand-side energy efficiency improvements, and (3) a shift in
generation from coal to less-carbon intensive generation. As shown in Table 3-15, the largest
decrease in coal production is projected to occur in the western region.
Table 3-15. Coal Production for the Electric Power Sector, 2025
Appalachia
Interior
West
Waste Coal
Imports
Total
Coal Production
Base Case
92
250
379
6
1
729
(million short
Rate-based
71
242
306
6
1
626
tons) Percent Change
Mass-based Rate-based
69
236
293
6
1
606
-23%
-3%
-19%
0%
-37%
-14%
from Base Case
Mass-based
-25%
-6%
-23%
0%
-14%
-17%
Source: Integrated Planning Model run by EPA, 2015
Power sector natural gas use is projected to decrease by about 1 percent in 2025 and 2030
under the rate-based illustrative plan scenario. In the mass-based scenario, power sector natural
gas use is projected to decrease by 4.5 percent in 2030. These trends are consistent with the
change in generation mix described above in Section 3.9.4.
3-33
-------
Table 3-16. Power Sector Gas Use
Base Case
Rate-based
Mass-based
Power Sector Gas Use (TCF)
2020 2025 2030
8.62 9.38 9.72
8.91 9.28 9.59
9.02 9.39 9.28
Percent Change in Power Sector Gas Use
2020 2025 2030
3.4% -1.0% -1.3%
4.6% 0.2% -4.5%
Source: Integrated Planning Model run by EPA, 2015
3.9.8 Projected Fuel Price, Market, and Infrastructure Impacts
The impacts of the two illustrative plan scenarios on coal and natural gas prices before
shipment are shown below in Table 3-17 and Table 3-18 and are attributable to the changes in
overall power sector demand for each fuel due to the final guidelines. Coal demand decreases by
2030, resulting in a decrease in the price of coal delivered to the electric power sector. In 2030,
gas demand and price decrease below the base case projections, due to the cumulative impact of
demand-side energy efficiency improvements and the consequent reduced overall electricity
demand.
IPM modeling of natural gas prices uses both short- and long-term price signals to
balance supply and demand for the fuel across the modeled time horizon. As such, it should be
understood that the pattern of IPM natural gas price projections over time is not a forecast of
natural gas prices incurred by end-use consumers at any particular point in time. The natural gas
market in the United States has historically experienced some degree of price volatility from year
to year, between seasons within a year, and during short-lived weather events (such as cold snaps
leading to short-run spikes in heating demand). These short-term price signals are fundamental
for allowing the market to successfully align immediate supply and demand needs. However,
end-use consumers are typically shielded from experiencing these rapid fluctuations in natural
gas prices by retail rate regulation and by hedging through longer-term fuel supply contracts by
the power sector. IPM assumes these longer-term price arrangements take place "outside of the
model" and on top of the "real-time" shorter-term price variation necessary to align supply and
demand. Therefore, the model's natural gas price projections should not be mistaken for
traditionally experienced consumer price impacts related to natural gas, but a reflection of
expected average price changes over the period represented by the modeling horizon.
There are very small changes to natural gas pipeline infrastructure needs over time, in
response to the illustrative plan scenarios. These changes, compared to historical deployment of
3-34
-------
new infrastructure, are very modest. In both the rate-based and mass-based scenarios, pipeline
capacity construction through 2020 is projected to increase by less than two percent beyond base
case projections. By 2030, however, the total cumulative pipeline capacity construction built is
projected to decrease compared to the base case, consistent with the projected decrease in total
demand and natural gas use. The projected increase in pipeline capacity in the near term is
largely the result of building pipeline capacity a few years earlier than projected in the base case.
Table 3-17. Projected Average Minemouth and Delivered Coal Prices (2011$/MMBtu)
Base Case
Rate-based
Mass-based
Rate-based
Mass-based
Minemouth
2020
1.55
1.54
1.54
-0.8%
-0.7%
2025
1.67
1.58
1.59
-5.0%
-4.7%
2030
1.79
1.73
1.73
-3.8%
-3.2%
Delivered
2020
2.38
2.34
2.35
-1.7%
-1.6%
- Electric Power Sector
2025 2030
2.50 2.68
2.35 2.46
2.40 2.55
-6.2% -8.0%
-4.3% .4.6%
Source: Integrated Planning Model run by EPA, 2015
Table 3-18. Projected Average Henry Hub (spot) and Delivered Natural Gas Prices
(2011$/MMBtu)
Base Case
Rate-based
Mass-based
Rate-based
Mass-based
Henry Hub
2020
5.20
5.48
5.40
5.4%
3.9%
2025
5.12
4.73
4.97
-7.5%
-3.0%
2030
6.01
6.21
5.92
3.3%
-1.4%
Delivered
2020
5.25
5.53
5.45
5.3%
3.8%
- Electric Power
2025
5.17
4.77
5.00
-7.7%
-3.2%
Sector
2030
5.98
6.13
5.86
2.5%
-2.1%
Source: Integrated Planning Model run by EPA, 2015
3.9.9 Projected Retail Electricity Prices
EPA's analysis of the illustrative rate-based plan scenario shows an increase in the
national average (contiguous U.S.) retail electricity price of less than one percent in both 2025
and 2030, compared to the modeled base case price estimate in those years. Under the illustrative
mass-based plan scenario, EPA projects an increase in the national average (contiguous U.S.)
retail electricity price of 2 percent in 2025 and 0.01 percent in 2030.
Retail electricity prices embody generation, transmission, distribution, taxes, and
demand-side energy efficiency costs. IPM modeling projects changes in regional wholesale
power prices and capacity payments related to imposition of the represented CPP scenarios that
3-35
-------
are combined with EIA regional transmission and distribution costs to calculate changes to
regional retail prices using the Retail Price Model (RPM).84 As described in Section 3.7.2, the
funding for demand-side energy efficiency (to cover program costs) is typically collected
through a standard per kWh surcharge to the ratepayer and the regional retail price impacts
presented here assume that these costs are recovered by utilities in retail rates. This is an
approximation, since not every utility will pass through the entirety of demand-side energy
efficiency costs. For example, a distribution only utility may generate reductions from demand-
side energy efficiency, sell the associated reduction in generation to affected EGUs (which in
turn use them to demonstrate compliance), and then account for this revenue in rate
determination. Furthermore, this analysis assumes that ratepayers in the state producing zero-
emitting generation (or avoided generation) bear the costs of such production. However, in
practice, if such generation is claimed by an affected source in another state, part of the cost of
that generation may ultimately be borne by ratepayers in the claiming state rather than the state
in which that zero-emitting generation was located. There are many factors influencing the
estimated retail electricity price impacts, namely projected changes in generation mix, fuel
prices, and development of new generating capacity. These projected changes vary regionally
under each illustrative plan scenario in response to the goals under the two scenarios. The
projected changes also vary depending upon retail electricity market structure (e.g., cost-of-
service vs. competitive). In the mass-based approach, treatment of allowance allocations will
also have an impact on retail electricity prices. In competitive regions, this RIA assumes that
allowances are freely allocated to generators who then keep 100% of the freely allocated
allowance value without passing this value through to ratepayers in the form of lower retail
electricity prices. To the extent that implementing authorities choose to require this allowance
value to be passed through to ratepayers (such as by allocating allowances to load-serving
entities who could be subject to such a requirement), retail prices would be lower than those
shown here.
' See documentation available at: http://www.epa.gov/powersectormodeling/
3-36
-------
Table 3-19. 2020 Projected Contiguous U.S. and Regional Retail Electricity Prices
(cents/kWh)
2020 Projected
Retail Price
Base Case Rate-based
ERCT
FRCC
MROE
MROW
NEWE
NYCW
NYLI
NYUP
RFCE
RFCM
RFCW
SRDA
SRGW
SRSE
SRCE
SRVC
SPNO
SPSO
AZNM
CAMX
NWPP
RMPA
Contiguous U.S.
9.7
10.5
9.9
8.7
13.3
17.4
14.4
12.4
11.1
10.4
9.4
8.6
8.6
10.0
8.0
9.8
9.9
7.9
10.9
14.3
6.9
8.7
10.0
9.9
10.7
10.3
9.0
14.0
18.3
15.1
13.1
11.8
10.9
9.8
8.8
9.0
10.1
8.1
9.9
9.9
8.1
11.2
14.8
7.1
9.0
10.3
(cents/kWh)
Mass-based
9.9
10.7
10.3
9.0
14.0
18.3
15.1
13.1
11.8
10.9
9.8
8.7
9.0
10.1
8.1
9.9
9.9
8.1
11.2
14.7
7.1
8.9
10.3
Percent Change
Rate-based
2.5%
2.0%
4.2%
2.8%
5.1%
5.0%
4.6%
5.4%
6.1%
4.3%
5.1%
2.1%
4.1%
0.9%
1.1%
1.5%
-0.8%
3.2%
2.1%
3.3%
3.2%
3.1%
3.2%
from Base Case
Mass-based
2.1%
1.6%
3.8%
2.3%
5.5%
5.3%
5.1%
5.3%
6.1%
4.3%
4.8%
1.7%
4.8%
0.5%
0.8%
1.2%
-0.9%
2.4%
2.1%
3.0%
2.9%
2.9%
3.0%
Note: regions pictured on Figure 3-6.
3-37
-------
Table 3-20. 2025 Projected Contiguous U.S. and Regional Retail Electricity Prices
(cents/kWh)
2025
Projected Retail Price
Base Case Rate-based
ERCT
FRCC
MROE
MROW
NEWE
NYCW
NYLI
NYUP
RFCE
RFCM
RFCW
SRDA
SRGW
SRSE
SRCE
SRVC
SPNO
SPSO
AZNM
CAMX
NWPP
RMPA
Contiguous U.S.
10.7
10.2
9.7
8.7
12.6
17.0
14.0
11.8
10.3
10.4
9.8
8.6
9.1
9.6
7.8
9.3
9.8
8.1
10.7
13.2
6.8
8.6
9.9
11.1
10.2
10.0
9.0
12.4
16.9
13.7
11.7
10.2
10.4
9.7
8.6
9.0
9.7
8.0
9.5
10.0
8.3
10.9
13.3
6.9
8.7
9.9
(cents/kWh)
Mass-based
10.9
10.3
10.0
9.0
12.7
16.9
13.7
11.7
10.5
10.6
10.1
8.7
9.3
9.8
8.0
9.6
10.2
8.4
10.9
13.5
7.0
8.9
10.1
Percent Change
Rate-based
3.8%
-0.2%
2.4%
2.5%
-1.3%
-0.5%
-2.2%
-0.8%
-0.2%
0.5%
-1.4%
0.0%
-0.9%
1.4%
2.6%
1.7%
2.9%
2.7%
2.2%
0.8%
2.1%
2.0%
0.9%
from Base Case
Mass-based
1.5%
1.0%
2.6%
3.1%
0.5%
-0.5%
-1.7%
-1.3%
2.1%
1.9%
2.4%
1.4%
2.5%
2.1%
3.0%
2.4%
4.3%
4.4%
1.8%
2.4%
2.7%
4.3%
2.0%
Note: regions pictured on Figure 3-6.
3-38
-------
Table 3-21. 2030 Projected Contiguous U.S. and Regional Retail Electricity Prices
(cents/kWh)
2030 Projected
Retail Price
Base Case Rate-based
ERCT
FRCC
MROE
MROW
NEWE
NYCW
NYLI
NYUP
RFCE
RFCM
RFCW
SRDA
SRGW
SRSE
SRCE
SRVC
SPNO
SPSO
AZNM
CAMX
NWPP
RMPA
Contiguous U.S.
11.6
10.3
9.7
8.9
14.3
19.2
16.3
13.6
11.3
10.5
10.4
9.0
9.7
9.8
7.8
9.3
9.5
8.7
10.9
13.5
6.9
8.9
10.3
11.4
10.8
10.3
9.1
13.6
18.2
14.8
12.7
10.7
10.8
10.5
9.3
9.6
10.2
8.1
9.6
9.8
9.0
11.2
13.6
7.0
9.0
10.4
(cents/kWh)
Mass-based
11.3
10.5
10.3
9.1
13.4
18.0
14.6
12.5
10.6
10.7
10.5
9.2
9.7
10.0
8.0
9.5
10.1
8.9
11.1
13.7
7.1
9.3
10.3
Percent Change
Rate-based
-1.4%
4.6%
5.9%
2.7%
-5.4%
-5.2%
-9.0%
-7.0%
-5.6%
3.4%
1.2%
3.5%
-0.6%
3.9%
4.3%
3.2%
2.7%
3.9%
2.3%
1.1%
2.2%
0.7%
0.8%
from Base Case
Mass-based
-2.5%
2.3%
6.3%
2.8%
-6.9%
-6.4%
-10.1%
-8.4%
-6.5%
1.7%
0.7%
1.9%
0.4%
2.1%
3.3%
2.0%
5.8%
2.0%
2.0%
1.4%
2.6%
3.5%
0.01%
Note: regions pictured on Figure 3-6.
3-39
-------
Figure 3-6. Electricity Market Module Regions
Source: EIA (http://www.eia.gov/forecasts/aeo/pdf/nerc_map.pdf)
3.9.10 Projected Electricity Bill Impacts
The electricity price changes addressed in section 3.9.9 combine with the significant
reductions in electricity demand applied in the illustrative approaches to affect average electricity
bills. The estimated changes to average bills are summarized in Table 3-22, and are subject to the
same caveats described in section 3.9.9. Under the illustrative rate-based plan scenario, EPA
estimates an average monthly bill increase of 2.7 percent in 2020 and an average bill decrease of
3.8 percent in 2025 and 7 percent in 2030. Under the mass-based scenario, EPA estimates an
average bill increase of 2.4 percent in 2020 and an average bill decrease of 2.7 percent in 2025
and 7.7 percent in 2030. These reduced electricity bills reflect the combined effects of changes in
both average retail rates (driven by compliance approaches taken to achieve the state goals) and
lower electricity demand (driven by demand-side energy efficiency).
Table 3-22. Projected Changes in Average Electricity Bills
Rate-based
Mass-based
2020
2.7%
2.4%
2025
-3.8%
-2.7%
2030
-7.0%
-7.7%
3.10 Adoption of a Mix of State Plan Approaches
The impact of the EGs on the marginal cost of generating electricity may differ for
affected EGUs if a state adopts a rate-based or a mass-based plan. Analysts have observed, in the
3-40
-------
context of the proposed EG, that the different production incentives for rate and mass-based
plans may encourage greater generation by the affected EGUs in the rate-based state. This is
because the rate-based approach may yield lower marginal costs of electricity generation than the
mass-based approach for some otherwise similar EGUs. In a rate-based program, affected EGUs
may emit more if they generate more, whereas in a mass-based approach, if an affected EGU
generates more it must incur the full cost of increasing its emissions. Some analysts have
suggested that this implies that if a state with a rate-based plan shares an electricity market with
another state that adopted a mass-based plan, then total CO2 emissions may be higher than if
both states adopted the same form of implementation (e.g. Burtraw et al., 2015; Bushnell et al.,
2014). In each case, both states would still be able to demonstrate that their affected EGUs are in
compliance, such that the state is achieving its state goal (or the uniform rates).
While these analyses identify how emissions and costs may be influenced by the
variation in the types of plans that states adopt, they have not raised concerns about the ability of
the electricity system to provide reliable and affordable electricity when EGUs face different
regulatory incentives. The EPA believes that differences in state plans, along with differences in
incentives from those plans, will not detrimentally affect the operation of electricity markets
because EGUs in the same market are often subject to different regulatory incentives. For
example, the time-differentiated pattern of renewable portfolio standard (RPS) adoption, their
varying stringency and form, and the operation of their associated renewable energy credit
(REC) markets, across the U.S. demonstrates how interconnected electricity markets are able to
function successfully, even with differential regulatory incentives across states. RPS are adopted
at the state level and are required of load-serving entities (LSEs). In some states, LSEs and the
owners of most of the fossil generation are one and the same. In other states, LSEs own no
generation (either fossil or renewable), and in some states and markets, one LSE may own
generation, while another may not. Furthermore, RPS requirements for LSEs serving load in
multiple states will influence the behavior of all EGUs operating the electricity market. Even
with this non-uniform regulatory environment, electricity has been delivered affordably and
reliably while at the same time, the use of renewable energy has increased dramatically.
In the context of preexisting programs, evidence suggests that the effect of differential
regulatory structures on emissions is relatively modest. For example, Schennach (2000) finds
that in the early years of the Title IV cap and trade program, the increase in 862 emissions of
3-41
-------
Phase II units, which historically were subject to emission rate performance standards, offset the
decrease in 862 emissions by Phase I units in by about 5%. The EPA's prospective analysis of
the benefits and costs of the Cross-State Air Pollution Rule, which used IPM, forecast only a
small increase in SOi emissions from plants that were not subject to the rule (U.S.EPA 2011).
The Regional Greenhouse Gas Initiative (RGGI) produces an annual report monitoring the trends
in on CO2 emissions from electricity generation in the region and imports from outside of the
region. To date, RGGI's monitoring effort has not identified any significant change in COi
emissions or the COi emission rate from non-RGGI electric generation serving load in the RGGI
region (e.g., RGGI 2014). The effect on the relative costs of production across similar sources
affected by different regulatory approaches will, in part, depend on the relative stringency of the
different regulatory approaches, and the emission rate of the EGUs that represent the marginal
source of electricity supply in the long-run.
In practice, determining the direction and magnitude of the effect of variation in state
plan type on sector wide emissions, relative to the two illustrative plan scenarios evaluated in this
RIA, would be difficult. At the outset there is a lack of information as to what design features
states might adopt in their plans and in turn what patterns of spatial and plan variation would be
most appropriate to consider. Determining the change in sectoral costs and emissions for the
situation in which subsets of states adopt different types of plans would require many additional
assumptions regarding which states adopt which plan types and the specific features of those
plans. The effect on the relative costs of generation across states will be sensitive to these
analytical choices, and therefore so will the estimated results regarding the direction and
magnitude of state plan variation on aggregate sectoral costs and emissions.
The mere existence of variation among the design of state plans would not be sufficient
to conclude that there will be a notable change in emissions relative to a case with less variation.
The ultimate impact of the variation will depend upon the specific plan approaches, such as the
way mass-based states allocate allowances, the state's goals, as well as the states' existing
generating fleets, the transmission grid, spatial variation in future electricity demand, and the
degree of ERC and allowance trading available within the system, amongst other variables.
There are other features of the requirements of state plans in this final rulemaking that
would influence the scope of emissions changes that may result from states adopting a mix of
3-42
-------
mass and rate-based plans. For example, this final rulemaking also requires that states adopting
mass-based plans include a method for addressing leakage to new fossil-fired generation. These
approaches are described in the preamble for this final rule. If states adopt programs to address
leakage within their state, those programs may lead to reduced generation by EGUs in
neighboring rate-based states (relative to the scenario where those plans were not in place). For
example, as shown in Burtraw et al. (2015) and Demailly and Quirion (2006), as well as other
related studies, output-based allocation to sources covered by a mass requirement would lead to
reduced production by sources subject to rate-based (or no) regulation.
3.11 Limitations of Analysis
EPA's modeling is based on expert judgment of various input assumptions for variables
whose outcomes are in fact uncertain. As a general matter, the Agency reviews the best available
information from engineering studies of air pollution controls, the ability to improve operating
efficiency, and new capacity construction costs to support a reasonable modeling framework for
analyzing the cost, emission changes, and other impacts of regulatory actions.
The costs presented in this RIA include both the IPM-projected annualized estimates of
private compliance costs as well as the estimated costs incurred by utilities and program
participants to achieve demand-side energy efficiency improvements. The demand-side energy
efficiency costs are developed based on a review of energy efficiency data and studies, and
expert judgment. The EPA recognizes that significant variation exists in these analyses reflecting
data and methodological limitations. The method used for estimating the demand-side energy
efficiency costs is discussed in more detail in the Demand-Side Energy Efficiency Technical
Support Document (TSD). The evaluation, measurement and verification (EM&V) of demand-
side energy efficiency is addressed in the section VIII, State Plans, of the preamble for the final
rule.
The base case electricity demand in IPM v.5.15 is calibrated to reference case demand in
AEO 2015. AEO 2015 demand may reflect, to some extent, a continuation of the impacts of state
demand-side energy efficiency policies but does not explicitly represent the most significant
existing state policies in this area (e.g., energy efficiency resource standards). To some degree,
the implicit representation of state policies in the EPA's base case alters the impacts assessment,
3-43
-------
but the direction and magnitude of change is not known with certainty. This issue is discussed in
the Demand-Side Energy Efficiency TSD.
Cost estimates for the final emission guidelines are based on rigorous power sector
modeling using ICF's Integrated Planning Model.85 IPM assumes "perfect foresight" of market
conditions over the time horizon modeled; to the extent that utilities and/or energy regulators
misjudge future conditions affecting the economics of pollution control, costs may be
understated as well.
One important element of the final CPP is the flexibility afforded to states as they
develop requirements for their existing emitting sources. Each state has discretion on how to best
achieve the standards of performance and/or state goals. As such, states can apply requirements
to sources that achieve greater reductions than required during the interim period, and use those
earlier reductions in the final period (i.e., banking of reductions).
In the analysis and modeling for the RIA, such flexibilities were not explicitly modeled in
the compliance scenarios. Doing so would require additional assumptions about the specific
opportunities states may choose to adopt in their plans, including the form of the standard that
states apply, the manner in which it is applied, and the economic signal that such a mechanism
provides to sources over time, such that sources would have an incentive to make greater
reductions earlier. As previously stated, the analysis in the RIA is intended to be illustrative to
inform the broad impacts of the rule across the power sector, and not intended to forecast the
specific approaches that individual states might choose, and how sources might prefer to achieve
the emission reductions to reflect each state plan in response to particular policy signals or
requirements. Not representing banking of earlier reductions into the final period captures this
uncertainty that there is inadequate and incomplete information at this time regarding state plans
in the analytic approach.
The analysis does not fully reflect the potential under the final rule for recognition of pre-
compliance emission reduction measures. Under the final rule, states implementing a rate-based
plan can recognize eligible emission reduction measures, including RE and demand-side energy
' Full documentation for IPM can be found at .
3-44
-------
efficiency, implemented after 2012 for the emission reductions those measures provide during
the interim and final performance periods (see preamble Sec. VIII.K.l). In the analysis, this
treatment is appropriately applied in the compliance period to generation from renewable
capacity built after 2012. However, demand-side EE is limited to recognition of impacts
occurring in the compliance period that result from investments in demand-side EE that are
assumed to begin after 2019 (as represented in the illustrative demand-side EE plan scenario).
Additionally, under the final rule, states will have the opportunity to recognize certain RE and
demand-side EE measures implemented after the effective date of the rule for the emission
reductions they provide in 2020-2021 through the Clean Energy Incentive Program (see
preamble Sec. VIII.B.2). By committing to recognize these actions in 2020-2021, states will have
access to a capped pool of additional rate-based ERCs and mass-based allowances, based on
their plan type. The Clean Energy Incentive Program is not reflected in this analysis.
The illustrative mass-based implementation scenario presented in this chapter includes an
RE set-aside, which is only one component of a potential approach to address leakage to new
sources. Please see section VIII of the preamble for a description of how states must show that
they are addressing leakage under mass-based implementation.
3.12 Social Costs
As discussed in the EPA Guidelines for Preparing Economic Analyses, social costs are
the total economic burden of a regulatory action. 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. The social costs of
a regulatory action will not necessarily be equivalent to the expenditures associated with
compliance. Nonetheless, here we use compliance costs as a proxy for social costs. This section
provides a qualitative discussion of the relationship between social costs and compliance cost
estimates presented in this chapter.
The cost estimates for the illustrative plan scenarios presented in this chapter are the sum
of expenditures on demand-side energy efficiency and the change in expenditures required by the
electricity sector to comply with the final emission guidelines. These two components are
3-45
-------
estimated separately. The expenditures required to achieve the assumed demand reductions
through demand-side energy efficiency programs are estimated using historical data, analysis,
and expert judgment. The change in the expenditures required by the electricity sector to meet
demand and maintain compliance are estimated by IPM and reflect both the reduction in
electricity production costs due to the reduction in demand caused by the demand-side energy
efficiency measures and the increase in electricity production costs required to achieve the
additional emission reductions necessary to comply with the state goals.
As described in section 3.7.1, the illustrative plan approaches assume that, in achieving
their goals, demand-side energy efficiency measures are adopted which lead to demand
reductions in each year represented by the illustrative energy efficiency plan scenario. The
estimated expenditures required to achieve those demand reductions through demand-side energy
efficiency are presented in this chapter and detailed in the Demand-Side Energy Efficiency TSD.
The social cost of achieving these energy savings comes in the form of increased expenditures on
technologies and/or services that are required to lower electricity consumption beyond the
business as usual. Under the assumption of complete and well-functioning markets, the
expenditures required to reduce electricity consumption on the margin will represent society's
opportunity cost of the resources required to produce the energy savings.
Due to the flexibility held by states in implementing their compliance with the final
standards these energy efficiency expenditures may be borne by end-users through direct
participant expenditures or electricity rate increases, or by producers through reductions in their
profits. While the allocation of these expenditures between consumers and producers is
important for understanding the distributional impact of potential compliance strategies, it does
not necessarily affect the opportunity cost required for the production of the energy savings from
a social perspective. However, specific design elements of demand-side energy efficiency
measures included to address distributional outcomes may have an effect on the economic
efficiency of the programs and therefore the social cost.
Another reason the expenditures associated with demand-side energy efficiency may
differ from social costs is due to differences in the services provided by more energy efficient
technologies and services adopted under the program relative to the baseline. For example, if
under the program end-users adopted more energy efficient products which were associated with
3-46
-------
quality or service attributes deemed less desirable, then there would be an additional welfare loss
that should be accounted for in social costs but is not necessarily captured in the measure of
expenditures. However, there is an analogous possibility that in some cases the quality of
services, outside of the energy savings, provided by the more energy efficient products and
practices are deemed more desirable by some end-users. For example, weatherization of
buildings to reduced electricity demand associated with cooling will likely have a significant
impact on natural gas use associated with heating. In either case, these real welfare impacts are
not fully captured by end-use energy efficiency expenditure estimates.
The fact that such quality and service differences may exist in reality but may not be
reflected in the price difference between more and less energy efficient products is one potential
hypothesis for the energy paradox. The energy paradox is the observation that end-users do not
always purchase products that are more energy efficient when the additional cost is less than the
reduction in the net present value of expected electricity expenditures achieved by those
products.86 Such circumstances are present in the analysis presented in this chapter, whereby in
some regions the base case and illustrative approaches suggest that cost of reducing demand
through energy efficiency programs is less than the retail electricity price. In addition to
heterogeneity in product services and consumer preferences, there are other explanations for the
energy paradox, falling both within and outside the neoclassical rational expectations paradigm
that is used in benefit/cost analysis. The Demand-Side Energy Efficiency TSD discusses the
energy paradox and provides additional hypothesis for why consumers may not make energy
efficiency investments that ostensibly seem to be in their own interest. The TSD discussion also
provides details on how the presence of additional market failures can lead to levels of energy
efficiency investment that may be too low from society's perspective even if that is not the case
for the end-user. In such cases there is the potential for properly designed energy efficiency
programs to address the source of under-investment, such as principal-agent problems where
there is a disconnect between those making the purchase decision regarding energy efficient
investments and energy use and those that would receive the benefits associated with reduced
energy use through lower electricity bills.
86 An analogous situation is present when some EGUs have assumed to have the ability to make heat rate
improvements at a capital cost that is less than the anticipated fuel expenditure savings.
3-47
-------
The other component of compliance cost reported in this chapter is the change in resource
cost (i.e., expenditures) required by the electricity sector to fulfill the remaining demand while
making additional CO2 emissions reductions necessary to comply with the state goals. Included
in the estimate of these compliance costs, estimated using IPM, are the cost reductions associated
with the reduction in required electricity generation due to the demand reductions from demand-
side energy efficiency measures and improvements in heat rate. By shifting the demand curve for
electricity, demand-side energy efficiency reduces the production cost in the sector. The resource
cost estimates from IPM therefore account for the increased cost of providing electricity,
including changes in fuel prices associated with changes in their demand, while EGUs comply
with their regulatory obligations (net of the reduction in their production costs due to lower
demand resulting from demand-side energy efficiency measures).
3.13 References
Burtraw, Dallas, Karen Palmer, Sophie Pan, and Anthony Paul. 2015. A Proximate Mirror:
Greenhouse Gas Rules and Strategic Behavior under the U.S. Clean Air Act. Resources for
the Future Discussion Paper 15-02. March 2015.
Bushnell, James B., Stephen P. Holland, Jonathan E. Hughes, Christopher R. Knittel. 2015.
Strategic Policy Choice in State-Level Regulation: The EPA's Clean Power Plan. National
Bureau of Economic Research Working Paper No. w21259. June, 2015.
Demailly, Damien, and Philippe Quirion. 2006 "COi abatement, competitiveness and leakage in
the European cement industry under the EU ETS: grandfathering versus output-based
allocation." Climate Policy 6.1: 93-113.
Regional Greenhouse Gas Initiative (RGGI). 2014. CO2 Emissions from Electricity Generation
and Imports in the Regional Greenhouse Gas Initiative: 2012 Monitoring Report. August 11,
2014. Accessed July 22, 2015:
http://www.rggi.org/docs/Documents/Elec_monitoring_report_2012 _15_08_ll.pdf
U.S. EPA. 2010. EPA Guidelines for Preparing Economic Analyses Available at:
. Accessed 7/11/2015.
U.S. EPA. 2015. Technical Support Document (TSD) the Final Carbon Pollution Emission
Guidelines for Existing Stationary Sources: Electric Utility Generating Units. Demand-Side
Energy Efficiency.
3-48
-------
U.S. EPA. 2015. Technical Support Document (TSD) the Final Carbon Pollution Emission
Guidelines for Existing Stationary Sources: Electric Utility Generating Units. Resource
Adequacy and Reliability Analysis.
3-49
-------
APPENDIX 3A: ANALYSIS OF POTENTIAL UPSTREAM METHANE EMISSIONS
CHANGES IN NATURAL GAS SYSTEMS AND COAL MINING
The purpose of this appendix is to describe the methodology for estimating upstream
methane emissions related to natural gas systems and coal mining sectors that may result from
the illustrative plan approaches examined in the Regulatory Impact Analysis (RIA). The US
Environmental Protection Agency (EPA) assessed whether the net change in upstream methane
emissions from natural gas and coal production is likely to be positive or negative and also
assessed the potential magnitude of these upstream changes relative to COi emissions reductions
anticipated at power plants from the illustrative plan approaches examined in the RIA. In
addition to estimating changes in upstream methane emissions, this assessment included
estimating CCh from the flaring of methane, but did not examine other potential changes in other
upstream greenhouse gas emissions changes from natural gas systems and coal mining sectors.
The methodologies used to project upstream emissions were previously developed for the
purpose of the 2014 U.S. Climate Action Report, and were subject to peer review and public
review as part of the publication of that report. In section 3 A.I, the overall approach is described
in brief. In section 3A.2, results are presented. Section 3A.3 discusses uncertainties and
limitations of the analysis. Finally, section 3A.4 contains a bibliography of cited resources. In the
RIA for the Clean Power Plan proposal (in then section 3A.3), we presented the detailed
methodologies for how methane and flaring-related COi projections were estimated for coal
mining and natural gas systems. We rely on the same methods in this RIA, so we refer the
interested reader to the proposal RIA87 for the detailed methodological discussion. The
calculations have been updated to reflect input data from the most recent U.S. GHG Inventory,
published in April 2015.
87 Clean Power Plan proposal RIA can be found at < http://www2.epa.gov/sites/production/files/2014-
06/documents/20140602ria-clean-power-plan.pdf>.
3A-1
-------
3A.1 General Approach
3A. 1.1 Analytical Scope
Upstream methane and flaring-related CO2 emissions associated with coal mining and
natural gas systems were estimated for 2025 through 2030 using methodologies developed for
the 2014 U.S. Climate Action Report (U.S. Department of State 2014). The base year for the
projections is 2013, as reported in the 2015 U.S. GHG Inventory (EPA 2015a). The projection
methodologies use activity driver data outputs such as coal and natural gas production from the
base case and policy scenarios generated by the Integrated Planning Model (IPM), which was
used in the RIA to model illustrative plan approaches. The projection methodologies use similar
activity data and emissions factors as are used in the U.S. Greenhouse Gas (GHG) Inventory.
The projection methodologies estimate reductions associated with both voluntary and
regulatory programs affecting upstream methane-related emissions. In the case of the voluntary
programs, the rate of reductions is based on the historical average decrease from these programs
over recent years. In the case of regulatory reductions, the reductions are based on the reduction
rates estimated in the RIAs of relevant regulations. The projections include emissions reductions
projected to result from the 2012 Oil and Natural Gas New Source Performance Standards. The
methodologies to estimate upstream emissions were subject to expert peer review and public
review in the context of the 2014 U.S. Climate Action Report. For more information on the
review, or for the detailed methodologies used for non-COi source projections in that report,
including methane-related emissions from coal production and natural gas systems, see
"Methodologies for U.S. Greenhouse Gas Emissions Projections: Non-CCh and Non-Energy COi
Sources" (EPA, 2013). Uncertainties and limitations are discussed, including a side case which
incorporates additional geographic information for estimating methane from coal mining.
The term "upstream emissions" in this document refers to vented, fugitive and flared
emissions associated with fuel production, processing, transmission, storage, and distribution of
fuels prior to fuel combustion in electricity plants. For this analysis, the EPA focused on
upstream methane from the natural gas systems and coal mining sectors. In addition, the analysis
included CO2 resulting from flaring in natural gas production. This analysis does not assess other
upstream GHG emissions changes, such as CCh emissions from the combustion of fuel used in
3A-2
-------
natural gas and coal production activities or other non-combustion CO2 emissions from natural
gas systems, such as vented CCh and CCh emitted from acid-gas removal processes.
Also, the EPA assessed potential upstream methane emissions from natural gas systems
and coal mining sectors within the domestic U.S., but did not examine emissions from potential
changes in upstream emissions generated by changes in natural gas and coal production,
processing, and transportation activities outside of the US.88 Last, the EPA did not assess
potential changes in other upstream non-GHG emissions, such as nitrogen oxides, volatile
organic compounds, and particulate matter. Table 3A-1 presents estimates of the upstream
emissions discussed in this analyses for 2013, based on the 2015 U.S. GHG Inventory.
EPA defined the boundaries of this assessment in order to provide targeted insights into the
potential net change in methane emissions from natural gas systems and coal production
activities specifically. COi emissions from flared methane are included because regulatory and
voluntary programs influence the rate of methane flaring over time and the CO2 remaining after
flaring is a methane-related GHG. Because of the multiple strategies adopted in the illustrative
plan approaches, a more comprehensive assessment of upstream GHG emissions would require
examination of the broader power sector and related input markets and their potential changes in
response to the rule. This analysis would be complex and likely subject to data limitations and
substantial uncertainties. Rather, EPA chose to limit the scope of this upstream analysis to
evaluate the potential for changes in GHG emissions that may be of significant scale relative to
the impacts of the rule and for which EPA had previously-reviewed projection techniques, which
are presented in detail below.
3A. 1.2 Coal Mining Source Description
Within coal mining, this analysis covers fugitive methane emissions from coal mining
(including pre-mining drainage) and post-mining activities (i.e., coal handling), including both
underground and surface mining. Emissions from abandoned mines are not included. Energy-
88 While the analysis does not estimate methane emissions changes outside of the United States, activity factors
include imports and exports of natural gas to help estimate domestic methane emissions related to trade of natural
gas, such as emissions from LNG terminals in the US or from pipelines transporting imported natural gas within the
US (or transporting natural gas within the US while en route for export).
3A-3
-------
related CO2 emissions, such as emissions from mining equipment and vehicles transporting coal
are not included. Methane, which is contained within coal seams and the surrounding rock strata,
is released into the atmosphere when mining operations reduce the pressure above and/or
surrounding the coal bed. The quantity of methane emitted from these operations is a function of
two primary factors: coal rank and coal depth. Coal rank is a measure of the carbon content of
the coal, with higher coal ranks corresponding to higher carbon content and generally higher
methane content. Pressure increases with depth and prevents methane from migrating to the
surface; as a result, underground mining operations typically emit more methane than surface
mining. In addition to emissions from underground and surface mines, post-mining processing of
coal and abandoned mines also release methane. Post-mining emissions refer to methane retained
in the coal that is released during processing, storage, and transport of the coal.
3A. 1.3 Natural Gas Systems Source Description
Within natural gas systems, this analysis covers vented and fugitive methane emissions
from the production, processing, transmission and storage, and distribution segments of the
natural gas system. It also includes COi from flaring of natural gas. Not included are vented and
fugitive COi emissions from natural gas systems, such as vented COi emissions removed during
natural gas processing, or energy-related CO2 such as emissions from stationary or mobile
combustion. The U.S. natural gas system encompasses hundreds of thousands of wells, hundreds
of processing facilities, and over a million miles of transmission and distribution pipelines.
Methane and non-combustion89 CCh emissions from natural gas systems are generally process-
related, with normal operations, routine maintenance, and system upsets being the primary
contributors. There are four primary stages of the natural gas system which are briefly described
below.
Production: In this initial stage, wells are used to withdraw raw gas from underground
formations. Emissions arise from the wells themselves, gathering pipelines, and well-site gas
treatment facilities (e.g., dehydrators, separators). Major emissions source categories within the
production stage include pneumatic devices, gas wells with liquids unloading, and gas well
89 In this document, consistent with IPCC accounting terminology, the term "combustion emissions" refers to the
emissions associated with the combustion of fuel for useful heat and work, while "non-combustion emissions" refers
to emissions resulting from other activities, including flaring and COi removed from raw natural gas.
3A-4
-------
completions and re-completions (i.e., workovers) with hydraulic fracturing (EPA 2013). Flaring
emissions account for the majority of the non-combustion COi emissions within the production
stage.
Processing: In this stage, natural gas liquids and various other constituents from the raw gas are
removed, resulting in "pipeline-quality" gas, which is then injected into the transmission system.
Fugitive methane emissions from compressors, including compressor seals, are the primary
emissions source from this stage. In the U.S. GHG Inventory, the majority of non-combustion
COi emissions in the processing stage come from acid gas removal units, which are designed to
remove COi from natural gas.
Transmission and Storage: Natural gas transmission involves high-pressure, large-diameter
pipelines that transport gas long distances from field production and processing areas to
distribution systems or large-volume customers such as power plants or chemical plants.
Compressor station facilities, which contain large reciprocating and turbine compressors, are
used to move the gas throughout the U.S. transmission system. Fugitive methane emissions from
these compressor stations and from metering and regulating stations account for the majority of
the emissions from this stage. Pneumatic devices and non-combusted engine exhaust are also
sources of methane emissions from transmission facilities. Natural gas is also injected and stored
in underground formations, or liquefied and stored in above-ground tanks, during periods of
lower demand (e.g., summer), and withdrawn, processed, and distributed during periods of
higher demand (e.g., winter). Compressors and dehydrators are the primary contributors to
emissions from these storage facilities. Emissions from LNG import terminals are included
within the transportation and storage stage.
Distribution: Distribution pipelines take the high-pressure gas from the transmission system at
"city gate" stations, reduce the pressure, and then distribute the gas through primarily
underground mains and service lines to individual end users.
3A-5
-------
Table 3A-1. Base Year Upstream Methane-Related Emissions in the U.S. GHG Inventory
Emissions Source 2013 Emissions (million short tons CCh Eq.)
Methane from Coal Mining 71.2
Underground Mining and Post-Mining 58.2
Surface Mining and Post-Mining 13.0
Methane from Natural Gas Systems 173.5
Production 51.8
Processing 25.0
Transmission and Storage 60.0
Distribution 36.7
CCh from flaring of natural gas 17.1
Source: 2015 U.S. GHG Inventory (EPA, 2015). A Global Warming Potential of 25 was used to convert methane
emissions to COi Eq.
In Table 3A-1, CCh-equivalent methane emissions are presented using the Fourth
Assessment Report Global Warming Potential (GWP) of 25.
3A. 1.4 Illustrative Plan Approaches Examined
States will ultimately determine optimal approaches to comply with the goals established
in this regulatory action. The RIA depicts illustrative plan approaches for the final emissions
guidelines, reflecting a rate-based illustrative plan or mass-based illustrative plan approach.
3A.1.5 Activity Drivers
IPM-based activity driver projections from base case and illustrative plan approaches
underlie the estimates of upstream methane emissions. These activity drivers include domestic
coal and natural gas production, imports and exports, and natural gas consumption. Table 3A-2
and Table 3A-3 summarize the IPM-based coal and natural gas production activity driver results
from the baseline and illustrative scenario for the final guidelines.90
Under the final guidelines, both the rate-based and mass-based illustrative plan approaches
result in reduced coal production and little change in natural gas production. We estimate that the
illustrative plan approaches will result in reductions in coal production of 5 to 6 percent in 2020,
12 to 15 percent in 2025 and 21 to 22 percent in 2030, relative to base case coal production.
90
Uncertainties related to activity drivers are discussed in the uncertainties and limitations section.
3A-6
-------
Natural gas production in the illustrative plan approaches change by 1 percent or less in each of
the years of analysis relative to production in the base case.
Table 3A-2. Projected Coal Production Impacts
Coal Production
(million short tons)
Coal Production Change
from Base Case
(million short tons)
Coal Production Percent
Change from Base Case
2020
2025
2030 2020
2025
2030
2020
2025
2030
Base Case
Rate-based
Mass-based
832.4
791.5
779.9
828.7
725.7
705.6
860.1
674.4
678.9
-41.0
-52.0
-103.0
-123.0
-186.0
-181.0
-5%
-6%
-12%
-15%
-22%
-21%
Table 3A-3. Projected Natural Gas Production Impacts
Dry Gas Production
(trillion cubic feet)
Dry Gas Production
Change from Base Case
(trillion cubic feet)
Dry Gas Production
Percent Change from
Base Case
2020
2025
2030
2020
2025
2030
2020
2025
2030
Base Case
Rate-based
Mass-based
28.9
29.1
29.2
30.8
30.7
30.8
33.0
32.9
32.6
+0.2
+0.3
-0.1
+0.0
-0.1
-0.4
+1%
+1%
0%
0%
0%
-1%
3A.2 Results
The analytical results (Table 3A-4) for the final guidelines indicates decreases in methane
emissions from coal mining of 8 to 9 million short tons COi Eq. in 2025, and about 14 million
short tons CCh Eq. in 2030. Methane from natural gas systems decreases relative to the base case
by 0 to 1 million short tons CCh Eq. in 2025 and 1 to 2 million short tons COi Eq. in 2030. COi
from flaring in natural gas production does not show significant change relative to the base case.
Based on the actions modelled in the illustrative plan approaches, upstream methane
emissions and COi emissions are predicted to decline (see Table 3A-4). The final guidelines are
predicted to result in a net emissions reduction of 8 to 9 million short tons CO2 Eq. in 2025 and a
net emissions reduction of 15 to 16 million short tons CO2 Eq. in 2030. These net emissions
changes represent the sum of changes in methane from coal mining, methane from natural gas
systems, and COi from flaring in natural gas production. The projections include voluntary and
3A-7
-------
regulatory activities to reduce emissions from coal mining and natural gas and oil systems,
including the 2012 Oil and Natural Gas NSPS. In addition, the EPA plans to issue a proposed
rule later this summer that would build on its 2012 Oil and Gas NSPS. When these standards are
finalized and implemented, they would further reduce projected emissions from natural gas and
oil systems.
Table 3A-4. Potential Upstream Emissions Changes
Emissions
2020
Rate-based
Methane from Coal Mining
Methane from Natural Gas Systems
CO2 from NG flaring
Total Methane + CO2
Mass-based
Methane from Coal Mining
Methane from Natural Gas Systems
CO2 from NG flaring
Total Methane + CO2
-3.0
+ 1.1
+0.2
-1.7
-3.8
+ 1.5
+0.2
-2.2
(million short
2025
-7.5
-0.8
-0.1
-8.4
-9.0
-0.1
+0.0
-9.0
tonsCChEq.)
2030
-14.0
-0.6
-0.1
-14.8
-13.7
-2.2
-0.3
-16.1
Note: A Global Warming Potential of 25 was used to convert methane emissions to COiEq.
3A.3 Uncertainties and Limitations
Projections of upstream methane emissions and CO2 emitted from flaring of methane are
subject to a range of uncertainties and limitations. These uncertainties and limitations include
estimating the effect of the plan approach on activity drivers, uncertainty in base year emissions,
and uncertainties in changes in emissions factors over relatively long periods of time. For
example, EPA's application of IPM relies on EIA projections for coal imports and exports.
Consequently, coal imports and exports are not able to fully respond within the IPM framework
to significant fluctuations in power sector coal demand. To the extent international markets may
be expected to offset reduced domestic coal demand, changes in U.S. upstream emissions as a
result of the policy scenarios would be smaller than what is presented here.
Discussion of uncertainty in historical estimates of emissions from coal mining and natural
gas systems can be found in the 2015 U.S. GHG Inventory. Projected changes in activity drivers
and emissions factors are based on a combination of policy, macroeconomic, energy market, and
technology factors which are uncertain in both baseline and illustrative plan approaches.
3A-8
-------
Relatively higher or lower economic growth, or changes in the relative prices or availability of
various technologies could result in alternative estimates in the net change in upstream methane
emissions and related COi emissions.
3A.4 References
IPCC. 2006. 2006IPCC Guidelines for National Greenhouse Gas Inventories. Available at
. Accessed June 4, 2015.
U. S. Department of State. 2014. 2014 U.S. Climate Action Report to the UNFCCC. Available at
. Accessed June 4, 2015.
U. S. Environmental Protection Agency (U.S. EPA). 1993. Anthropogenic Methane Emissions in
the United States, Estimates for 1990: Report to Congress. EPA-430-R-93-003.
U. S. Environmental Protection Agency (U.S. EPA). 2011. Oil and Natural Gas Sector:
Standards of Performance for Crude Oil and Natural Gas Production, Transmission, and
Distribution: Background Technical Support Document for the Proposed Standards.
Available at . Accessed June 4,
2015.
U. S. Environmental Protection Agency (U.S. EPA). 2012a. Oil and Natural Gas Sector: New
Source Performance Standards and National Emission Standards for Hazardous Air
Pollutants Reviews. 40 CFR Parts 60 and 63 [EPA-HQ-OAR-2010-0505; FRL-9665-1], RIN
2060-AP76. Available at . Accessed June 4, 2015.
U. S. Environmental Protection Agency (U.S. EPA). 2012b. Oil and Natural Gas Sector:
Standards of Performance for Crude Oil and Natural Gas Production, Transmission, and
Distribution: Background Supplemental Technical Support Document for the Final New
Source Performance Standards. Available at
. Accessed June 4, 2015.
U. S. Environmental Protection Agency (U.S. EPA). 2015. Inventory of U.S. Greenhouse Gas
Emissions and Sinks: 1990-2013. EPA-430-R-15-004. Available at
. Accessed June 4,
2015.
U. S. Environmental Protection Agency (U.S. EPA). 2013. Methodologies for U.S. Greenhouse
Gas Emissions Projections: Non-CO2 and Non-Energy CO2 Sources. Available at
Accessed June 4, 2015.
3A-9
-------
CHAPTER 4: ESTIMATED CLIMATE BENEFITS AND HUMAN HEALTH CO-BENEFITS
4.1 Introduction
Implementing the Final Carbon Pollution Emission Guidelines for Existing Stationary
Sources: Electric Utility Generating Units (hereafter referred to as the "final emission
guidelines" or "Clean Power Plan Final Rule") is expected to reduce emissions of carbon dioxide
(CCh) and have ancillary human health benefits (i.e., co-benefits) associated with lower ambient
concentrations of criteria air pollutants. This chapter describes the methods used to estimate the
monetized climate benefits and the monetized air quality health co-benefits associated with
reducing exposure to ambient fine particulate matter (PIVb.s) and ozone by reducing emissions of
precursor pollutants (i.e., sulfur dioxide (SOi), nitrogen dioxide (NOi), and directly emitted
PMi.s). Data, resource, and methodological limitations prevent the EPA from monetizing the
benefits from several important co-benefit categories, including reducing direct exposure to 862,
NOi, and hazardous air pollutants (HAP), as well as ecosystem effects and visibility impairment.
We qualitatively discuss these unquantified benefits in this chapter.
This chapter provides estimates of the monetized climate benefits and air quality health co-
benefits associated with emission reductions for the illustrative rate-based and mass-based
illustrative plan approaches across several analysis years and discount rates. The estimated
benefits associated with these emission reductions are beyond those achieved by previous EPA
rulemakings, including the Mercury and Air Toxics Standards (MATS).
4.2 Estimated Climate Benefits from CCh
The primary goal of the final emission guidelines is to reduce emissions of CO2. In this
section, we provide a brief overview of the 2009 Endangerment Finding and climate science
assessments released since then. We also provide information regarding the economic valuation
of CO2 using the Social Cost of Carbon (SC-CCh), a metric that estimates the monetary value of
impacts associated with marginal changes in CO2 emissions in a given year. Table 4-1
summarizes the quantified and unquantified climate benefits in this analysis.
4-1
-------
Table 4-1. Climate Effects
Benefits Specific Effect
Category
Effect Has Been
Quantified
Effect Has Been More
Monetized Information
Improved Environment
Reduced Global climate impacts from COi
climate effects Climate impacts from ozone and black
carbon (directly emitted PM)
Other climate impacts (e.g., other GHGs
such as methane, aerosols, other impacts)
i
S SCC TSD
Ozone ISA, PM
ISA2
IPCC2
1 The global climate and related impacts of COi emissions changes, such as sea level rise, are estimated within each
integrated assessment model as part of the calculation of the SC-COi. The resulting monetized damages, which
are relevant for conducting the benefit-cost analysis, are used in this RIA to estimate the welfare effects of
quantified changes in COi emissions.
2 We assess these co-benefits qualitatively because we do not have sufficient confidence in available data or
methods.
4.2.1 Climate Change Impacts
Through the implementation of CAA regulations, the EPA addresses the negative
externalities caused by air pollution. In 2009, the EPA Administrator found that elevated
concentrations of greenhouse gases in the atmosphere may reasonably be anticipated both to
endanger public health and to endanger public welfare. It is these adverse impacts that make it
necessary for the EPA to regulate GHGs from EGU sources. The preamble summarizes the
public health and public welfare impacts that were detailed in the 2009 Endangerment Finding.
For health, these include the increased likelihood of heat waves, negative impacts on air quality,
more intense hurricanes, more frequent and intense storms and heavy precipitation, and impacts
on infectious and waterborne diseases. For welfare, these include reduced water supplies in some
regions, increased water pollution, increased occurrences of floods and droughts, rising sea
levels and damage to coastal infrastructure, increased peak electricity demand, changes in
ecosystems, and impacts on indigenous communities.
The preamble also summarizes new scientific assessments and recent climatic
observations. Major scientific assessments released since the 2009 Endangerment Finding have
improved scientific understanding of the climate, and provide even more evidence that GHG
emissions endanger public health and welfare for current and future generations. The National
Climate Assessment (NCA3), in particular, assessed the impacts of climate change on human
health in the United States, finding that, Americans will be impacted by "increased extreme
4-2
-------
weather events, wildfire, decreased air quality, threats to mental health, and illnesses transmitted
by food, water, and disease-carriers such as mosquitoes and ticks." These assessments also detail
the risks to vulnerable groups such as children, the elderly and low income households.
Furthermore, the assessments present an improved understanding of the impacts of climate
change on public welfare, higher projections of future sea level rise than had been previously
estimated, a better understanding of how the warmth in the next century may reach levels that
would be unprecedented relative to the preceding millions of years of history, and new
assessments of the impacts of climate change on permafrost and ocean acidification. The impacts
of GHG emissions will be realized worldwide, independent upon their location of origin, and
impacts outside of the United States will produce consequences relevant to the United States.
4.2.2 Social Cost of Carbon
We estimate the global social benefits of COi emission reductions expected from the final
emission guidelines using the SC-CCh estimates presented in the Technical Support Document:
Technical Update of the Social Cost of Carbon for Regulatory Impact Analysis Under Executive
Order 12866 (May 2013, Revised July 2015) ("current TSD").91 We refer to these estimates,
which were developed by the U.S. government, as "SC-CCh estimates." The SC-CCh is a metric
that estimates the monetary value of impacts associated with marginal changes in COi emissions
in a given year. It includes a wide range of anticipated climate impacts, such as net changes in
agricultural productivity and human health, property damage from increased flood risk, and
changes in energy system costs, such as reduced costs for heating and increased costs for air
conditioning. It is typically used to assess the avoided damages as a result of regulatory actions
(i.e., benefits of rulemakings that lead to an incremental reduction in cumulative global CO2
emissions).
91 Docket ID EPA-HQ-OAR-2013-0495, Technical Support Document: Technical Update of the Social Cost of
Carbon for Regulatory Impact Analysis Under Executive Order 12866, Interagency Working Group on Social Cost
of Carbon, with participation by Council of Economic Advisers, Council on Environmental Quality, Department of
Agriculture, Department of Commerce, Department of Energy, Department of Transportation, Environmental
Protection Agency, National Economic Council, Office of Energy and Climate Change, Office of Management and
Budget, Office of Science and Technology Policy, and Department of Treasury (May 2013, Revised July 2015).
Available at: Accessed
7/11/2015.
4-3
-------
The SC-CO2 estimates used in this analysis were developed over many years, using the
best science available, and with input from the public. Specifically, an interagency working
group (IWG) that included the EPA and other executive branch agencies and offices used three
integrated assessment models (lAMs) to develop the SC-CCh estimates and recommended four
global values for use in regulatory analyses. The SC-COi estimates were first released in
February 2010 and updated in 2013 using new versions of each IAM. As discussed further
below, the IWG published two minor corrections to the SC-COi estimates in July 2015.
The SC-CO2 estimates were developed using an ensemble of the three most widely cited
integrated assessment models in the economics literature with the ability to estimate the SC-COi.
A key objective of the IWG was to draw from the insights of the three models while respecting
the different approaches to linking GHG emissions and monetized damages taken by modelers in
the published literature. After conducting an extensive literature review, the interagency group
selected three sets of input parameters (climate sensitivity, socioeconomic and emissions
trajectories, and discount rates) to use consistently in each model. All other model features were
left unchanged, relying on the model developers' best estimates and judgments, as informed by
the literature. Specifically, a common probability distribution for the equilibrium climate
sensitivity parameter, which informs the strength of climate's response to atmospheric GHG
concentrations, was used across all three models. In addition, a common range of scenarios for
the socioeconomic parameters and emissions forecasts were used in all three models. Finally, the
marginal damage estimates from the three models were estimated using a consistent range of
discount rates, 2.5, 3.0, and 5.0 percent. See the 2010 TSD for a complete discussion of the
methods used to develop the estimates and the key uncertainties, and the current TSD for the
latest estimates.92
The SC-CO2 estimates represent global measures because of the distinctive nature of the
climate change, which is highly unusual in at least three respects. First, emissions of most GHGs
contribute to damages around the world independent of the country in which they are emitted.
The SC-CO2 must therefore incorporate the full (global) damages caused by GHG emissions to
address the global nature of the problem. Second, the U.S. operates in a global and highly
92 See https://www.whitehouse.gov/omb/oira/social-cost-of-carbon for both TSDs.
4-4
-------
interconnected economy, such that impacts on the other side of the world can affect our
economy. This means that the true costs of climate change to the U.S. are larger than the direct
impacts that simply occur within the U.S. Third, climate change represents a classic public goods
problem because each country's reductions benefit everyone else and no country can be excluded
from enjoying the benefits of other countries' reductions, even if it provides no reductions itself.
In this situation, the only way to achieve an economically efficient level of emissions reductions
is for countries to cooperate in providing mutually beneficial reductions beyond the level that
would be justified only by their own domestic benefits. In reference to the public good nature of
mitigation and its role in foreign relations, thirteen prominent academics noted that these "are
compelling reasons to focus on a global SCC" in a recent article on the SCC (Pizer et al., 2014).
In addition, as noted in OMB's Response to Comments on the SCC, there is no bright line
between domestic and global damages. Adverse impacts on other countries can have spillover
effects on the United States, particularly in the areas of national security, international trade,
public health and humanitarian concerns.93
The 2010 TSD noted a number of limitations to the SC-CCh analysis, including the
incomplete way in which the integrated assessment models capture catastrophic and non-
catastrophic impacts, their incomplete treatment of adaptation and technological change,
uncertainty in the extrapolation of damages to high temperatures, and assumptions regarding risk
aversion. Currently integrated assessment models do not assign value to all of the important
physical, ecological, and economic impacts of climate change recognized in the climate change
literature due to a lack of precise information on the nature of damages and because the science
incorporated into these models understandably lags behind the most recent research.94 The
limited amount of research linking climate impacts to economic damages makes the modeling
exercise even more difficult. These individual limitations do not all work in the same direction in
terms of their influence on the SC-CCh estimates, though taken together they suggest that the
93 See Endangerment and Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the Clean
Air Act, 74 Fed. Reg. 66,496, 66,535 (Dec. 15, 2009) and National Research Council 2013a.
94 Climate change impacts and SCC modeling is an area of active research. For example, see: (1) Howard, Peter,
"Omitted Damages: What's Missing from the Social Cost of Carbon." March 13, 2014,
http://costofcarbon.org/files/Omitted_Damages_Whats_Missing_From_the_Social_Cost_of_Carbon.pdf; and (2)
Electric Power Research Institute, "Understanding the Social Cost of carbon: A Technical Assessment," October
2014, www.epri.com.
4-5
-------
SC-CO2 estimates are likely conservative. In particular, the IPCC Fourth Assessment Report
(2007), which was the most current IPCC assessment available at the time of the IWG's 2009-
2010 review, concluded that "It is very likely that [SC-COi estimates] underestimate the damage
costs because they cannot include many non-quantifiable impacts." Since then, the peer-
reviewed literature has continued to support this conclusion. For example, the IPCC Fifth
Assessment report observed that SC-CCh estimates continue to omit various impacts that would
likely increase damages. The 95th percentile estimate was included in the recommended range
for regulatory impact analysis to address these concerns.
The EPA and other agencies have continued to consider feedback on the SC-CCh estimates
from stakeholders through a range of channels, including public comments on this rulemaking
and others that use the SC-CCh in supporting analyses and through regular interactions with
stakeholders and research analysts implementing the SC-COi methodology used by the
interagency working group. The SC-CCh comments received on this rulemaking covered a wide
range of topics including the technical details of the modeling conducted to develop the SC-COi
estimates, the aggregation and presentation of the SC-CCh estimates, and the process by which
the SC-COi estimates were derived. Many but not all commenters were supportive of the SC-
COi and its application to this rulemaking. Commenters also provided constructive
recommendations for potential opportunities to improve the SC-CCh estimates in future updates.
The EPA Response to Comments document provides a summary and response to the SC-CCh
comments submitted to this rulemaking.
Many of the comments EPA received were similar to those that OMB's Office of
Information and Regulatory Affairs received in response to a separate request for public
comment on the approach used to develop the estimates. After careful evaluation of the full
range of comments submitted to OMB's Office of Information and Regulatory Affairs, the IWG
continues to recommend the use of these SC-COi estimates in regulatory impact analysis. With
the release of the response to comments95, the IWG announced plans to obtain expert
independent advice from the National Academies of Sciences, Engineering, and Medicine
95
See https://www. whitehouse.gov/sites/default/files/omb/inforeg/scc-response-to-comments-final-july-2015.pdf
4-6
-------
(Academies) to ensure that the SC-COi estimates continue to reflect the best available scientific
and economic information on climate change.96 The Academies' process will be informed by the
public comments received and focus on the technical merits and challenges of potential
approaches to improving the SC-COi estimates in future updates.
Concurrent with OMB's publication of the response to comments on SC-CCh and
announcement of the Academies process, OMB posted a revised TSD that includes two minor
technical corrections to the current estimates. One technical correction addressed an inadvertent
omission of climate change damages in the last year of analysis (2300) in one model and the
second addressed a minor indexing error in another model. On average the revised SC-CO2
estimates are one dollar less than the mean SC-CO2 estimates reported in the November 2013
revision to the May 2013 TSD. The change in the estimates associated with the 95th percentile
estimates when using a 3% discount rate is slightly larger, as those estimates are heavily
influenced by the results from the model that was affected by the indexing error.
The four SC-CO2 estimates are as follows: $12, $40, $60, and $120 per short ton of CO2
emissions in the year 2020 (2011$).97 The first three values are based on the average SC-CO2
from the three lAMs, at discount rates of 5, 3, and 2.5 percent, respectively. SC-COi estimates
for several discount rates are included because the literature shows that the SC-COi is quite
sensitive to assumptions about the discount rate, and because no consensus exists on the
appropriate rate to use in an intergenerational context (where costs and benefits are incurred by
different generations). The fourth value is the 95th percentile of the SC-COi from all three
models at a 3 percent discount rate. It is included to represent higher-than-expected impacts from
temperature change further out in the tails of the SC-CO2 distribution (representing less likely,
but potentially catastrophic, outcomes).
96
See https://www. whitehouse.gov/blog/2015/07/02/estimating-benefits-carbon-dioxide-emissions-reductions.
97 The current version of the TSD is available at: https://www. whitehouse.gov/sites/default/files/omb/inforeg/scc-
tsd-fmal-july-2015.pdf. The 2010 and 2013 TSDs present SC-CO2 in 2007$ per metric ton. The unrounded
estimates from the current TSD were adjusted to (1) 2011$ using GDP Implicit Price Deflator (1.061374),
http://www.bea.gov/iTable/index_nipa.cfm and (2) short tons using the conversion factor of 0.90718474 metric tons
in a short ton. The estimates presented in the RIA were rounded to two significant digits.
4-7
-------
Table 4-2 presents the global SC-COi estimates in short tons for the years 2015 to 2050.98
In order to calculate the dollar value for emission reductions, the SC-COi estimate for each
emissions year would be applied to changes in CO2 emissions for that year, and then discounted
back to the analysis year using the same discount rate used to estimate the SC-COi." The SC-
COi increases over time because future emissions are expected to produce larger incremental
damages as physical and economic systems become more stressed in response to greater climate
change. Note that the interagency group estimated the growth rate of the SC-CCh directly using
the three integrated assessment models rather than assuming a constant annual growth rate. This
helps to ensure that the estimates are internally consistent with other modeling assumptions.
Tables 4-3 through 4-5 report the incremental climate benefits estimated in three analysis years
(2020, 2025, and 2030) for the rate-based and mass-based illustrative plan approaches.
Table 4-2. Social Cost of CCh, 2015-2050 (in 2011$ per short ton)*
Year
2015
2020
2025
2030
2035
2040
2045
2050
5% Average
$11
$12
$13
$15
$17
$20
$22
$25
Discount
3% Average
$35
$40
$44
$48
$53
$58
$62
$66
Rate and Statistic
2.5% Average
$54
$60
$65
$70
$75
$81
$86
$91
3%(95thpercentile)
$100
$120
$130
$150
$160
$180
$190
$200
* These SC-COi values are stated in $/short ton and rounded to two significant figures. The SC-COi values have
been converted from $/metric ton to $/short ton using the conversion factor 0.90718474 metric tons in a short ton
for consistency with this rulemaking. This calculation does not change the underlying methodology nor does it
change the meaning of the SC-CO2 estimates. For both metric and short tons denominated SC-COi estimates, the
estimates vary depending on the year of COi emissions and are defined in real terms, i.e., adjusted for inflation
using the GDP implicit price deflator.
98 For consistency with this rulemaking, the SC-COi values have been converted from $/metric ton to $/short ton
and applied to the COi reductions (short tons) to estimate climate benefits. Specifically, the $/metric ton estimates
were multiplied by the conversion factor 0.90718474 metric tons in a short ton to yield $/short ton. This calculation
does not change the underlying methodology, the meaning of the SC-COi estimates, or the final benefits estimates.
99 This analysis considered the climate impacts of only COi emission change. As discussed below, the climate
impacts of other pollutants were not calculated for the proposed guidelines. While COi is the dominant GHG
emitted by the sector, we recognize the representative facilities within these comparisons may also have different
emission rates for other climate forcers that will serve a minor role in determining the overall social cost of
generation.
4-8
-------
Table 4-3. Estimated Global Climate Benefits of CCh Reductions for the Final Emission
Guidelines in 2020 (billions of 2011$)*
Discount rate and statistic Rate-Based Mass-Based
Million short tons of CO2 reduced 69 82
5% (average) $0.80 $0.94
3% (average) $2.8 $3.3
2.5% (average) $4.1 $4.9
3%(95thpercentile) $8.2 $9.7
* The SC-COi values are dollar-year and emissions-year specific. SC-COi values represent only a partial accounting
of climate impacts.
Table 4-4. Estimated Global Climate Benefits of CCh Reductions for the Final Emission
Guidelines in 2025 (billions of 2011$)*
Discount rate and statistic Rate-Based Mass-Based
Million short tons of CO2 reduced 232 264
5% (average) $3.1 $3.6
3% (average) $10 $12
2.5% (average) $15 $17
3%(95thpercentile) $31 $35
* The SC-COi values are dollar-year and emissions-year specific. SC-COi values represent only a partial accounting
of climate impacts.
Table 4-5. Estimated Global Climate Benefits of CCh Reductions for the Final Emission
Guidelines in 2030 (billions of 2011$)*
Discount rate and statistic Rate-Based Mass-Based
Million short tons of COi reduced 415 413
5% (average) $6.4 $6.4
3% (average) $20 $20
2.5% (average) $29 $29
3%(95thpercentile) $61 $60
* The SC-COi values are dollar-year and emissions-year specific. SC-COi values represent only a partial accounting
of climate impacts.
It is important to note that the climate benefits presented above are associated with changes
in COi emissions only. Implementing these final emission guidelines, however, will have an
impact on the emissions of other pollutants that would affect the climate. Both predicting
reductions in emissions and estimating the climate impacts of these other pollutants, however, is
complex. The climate impacts of these other pollutants have not been calculated for the final
emission guidelines.
100
100 The SC-COi estimates used in this analysis are designed to assess the climate benefits associated with changes in
COi emissions only.
4-9
-------
The other emissions potentially reduced as a result of the final emission guidelines include
other greenhouse gases (such as methane), aerosols and aerosol precursors such as black carbon,
organic carbon, sulfur dioxide and nitrogen oxides, and ozone precursors such as nitrogen oxides
and volatile organic carbon compounds. Changes in emissions of these pollutants (both increases
and decreases) could directly result from changes in electricity generation, upstream fossil fuel
extraction and transport, and/or downstream secondary market impacts. Reductions in black
carbon or ozone precursors are projected to lead to further cooling, but reductions in the other
aerosol species and precursors are projected to lead to warming. Therefore, changes in non-CCh
pollutants could potentially augment or offset the climate benefits calculated here. These
pollutants can act in different ways and on different timescales than carbon dioxide. For
example, aerosols reflect (and in the case of black carbon, absorb) incoming radiation, whereas
greenhouse gases absorb outgoing infrared radiation. In addition, these aerosols are thought to
affect climate indirectly by altering properties of clouds. Black carbon can also deposit on snow
and ice, darkening these surfaces and accelerating melting. In terms of lifetime, while carbon
dioxide emissions can increase concentrations in the atmosphere for hundreds or thousands of
years, many of these other pollutants are short lived and remain in the atmosphere for short
periods of time ranging from days to weeks and can therefore exhibit large spatial and temporal
variability.
While the EPA has not quantified the climate impacts of these other pollutants for the final
emission guidelines, the Agency has analyzed the potential changes in upstream methane
emissions from the natural gas and coal production sectors that may result from the illustrative
plan approaches examined in this RIA in the appendix to Chapter 3. The EPA assessed whether
the net change in upstream methane emissions from natural gas and coal production is likely to
be positive or negative and also assessed the potential magnitude of changes relative to CO2
emissions reductions anticipated at power plants. This assessment included CO2 emissions from
the flaring of methane, but did not evaluate potential changes in other combustion-related CCh
emissions, such as emissions associated with drilling, mining, processing, and transportation in
the natural gas and coal production sectors. This analysis found that the net upstream CH4
emissions from natural gas systems and coal mines and CCh emissions from flaring of methane
will likely decrease under the final emission guidelines. Furthermore, the analysis suggests that
4-10
-------
the changes in upstream methane emissions are small relative to the changes in direct emissions
from power plants.
4.3 Estimated Human Health Co-Benefits
In addition to reducing emissions of COi, implementing these final emission guidelines is
expected to reduce emissions of SOi and NOx, which are precursors to formation of ambient
PMi.5, as well as directly emitted fine particles.101 Therefore, reducing these emissions would
also reduce human exposure to ambient PM2.5 and the incidence of PMi.s-related health effects.
In addition, in the presence of sunlight, NOx and VOCs can undergo a chemical reaction in the
atmosphere to form ozone. Depending on localized concentrations of volatile organic compounds
(VOCs), reducing NOx emissions would also reduce human exposure to ozone and the incidence
of ozone-related health effects. Although we do not have sufficient data to quantify these impacts
in this analysis, reducing emissions of SO2 and NOx would also reduce ambient exposure to SO2
and NOi and their associated health effects, respectively. In this section, we provide an overview
of the monetized PM2.5 and ozone-related co-benefits estimated for the final emission guidelines.
A full description of the underlying data, studies, and assumptions is provided in the PM
NAAQS RIA (U.S. EPA, 2012a) and Ozone NAAQS RIA (U.S. EPA, 2008b, 2010d). The
estimated co-benefits associated with these emission reductions are beyond those achieved by
previous EPA rulemakings, including MATS.
There are several important considerations in assessing the air quality-related health co-
benefits for a climate-focused rulemaking. First, these estimated health co-benefits do not
account for any climate-related air quality changes (e.g., increased ambient ozone associated
with higher temperatures) but rather changes in precursor emissions affected by this rulemaking.
Excluding climate-related air quality changes may underestimate ozone-related health co-
benefits. It is unclear how PM2.5-related health co-benefits would be impacted by excluding
101 In the RIA for the proposed rule, we estimated the health co-benefits associated with emission reductions of two
categories of directly emitted particles: elemental carbon plus organic carbon (EC+OC) and crustal. Crustal
emissions are composed of compounds associated with minerals and metals from the earth's surface, including
carbonates, silicates, iron, phosphates, copper, and zinc. Often, crustal material represents particles not classified as
one of the other species (e.g., organic carbon, elemental carbon, nitrate, sulfate, chloride, etc.). For this RIA, we did
not estimate changes in emissions of directly emitted particles. As a result, quantified PM2.5 related benefits are
underestimated by a relatively small amount. In the proposal RIA, the benefits from reductions in directly emitted
PM2.s were less than 10 percent of total monetized health co-benefits across all scenarios and years.
4-11
-------
climate-related air quality changes since the science is unclear as to how climate change may
affect PMi.5 exposure. Second, the estimated health co-benefits also do not consider temperature
modification of PM2.5 and ozone risks (Roberts 2004; Ren 2006a, 2006b, 2008a, 2008b). Third,
the estimated climate benefits reported in this RIA reflect global benefits, while the estimated
health co-benefits are calculated for the contiguous U.S. only. Excluding temperature
modification of air pollution risks and international air quality-related health benefits likely leads
to underestimation of quantified health co-benefits (Anenberg et al, 2009, Jhun et al, 2014).
Fourth, as noted earlier, we do not estimate the climate benefits associated with reductions in PM
and Os precursors.
Implementing the final emission guidelines may lead to reductions in ambient PMi.5
concentrations below the National Ambient Air Quality Standards (NAAQS) for PM and ozone
in some areas and assist other areas with attaining these NAAQS. Because the NAAQS RIAs
(U.S. EPA, 2012a, 2008b, 2010d) also calculated PM and ozone benefits, there are important
differences worth noting in the design and analytical objectives of each RIA. The NAAQS RIAs
illustrate the potential costs and benefits of attaining a revised air quality standard nationwide
based on an array of emission reduction strategies for different sources reflecting the application
of known and unknown controls, incremental to implementation of existing regulations and
controls needed to attain the current standards. In short, NAAQS RIAs hypothesize, but do not
predict, the reduction strategies that States may choose to enact when implementing a revised
NAAQS. The setting of a NAAQS does not directly result in costs or benefits, and as such, the
EPA's NAAQS RIAs are merely illustrative and the estimated costs and benefits are not intended
to be added to the costs and benefits of other regulations that result in specific costs of control
and emission reductions. Some of the emissions reductions estimated to result from
implementation of the final emission guidelines may achieve some of the air quality
improvements that resulted from the hypothesized attainment strategies presented in the
illustrative NAAQS RIAs. The emissions reductions from implementing the final emission
guidelines will decrease the remaining amount of emissions reductions needed in non-attainment
areas and reduce the costs and benefits attributable to meeting the NAAQS.
Similar to NAAQS RIAs, the emission reduction scenarios estimated for the final emission
guidelines are also illustrative. In contrast to NAAQS RIAs, all of the emission reductions for the
4-12
-------
illustrative plan approaches would occur in one well-characterized sector (i.e., the EGU sector).
In general, the EPA is more confident in the magnitude and location of the emission reductions
for rules which require specific emission reductions in a specific sector, for example, the recent
Mercury and Air Toxics Standards. As such, emission reductions achieved under these types of
promulgated rules will ultimately be reflected in the baseline of future NAAQS analyses, which
would reduce the incremental costs and benefits associated with attaining revised future
NAAQS. The EPA does not re-issue illustrative RIAs outside of the rulemaking process that
retroactively update the baseline to account for implementation rules promulgated after an RIA
was completed. For more information on the relationship between illustrative analyses, such as
for the NAAQS and this final emission guidelines, and implementation rules, please see section
1.3 of the PM NAAQS RIA (U.S. EPA, 2012a).
4.3.1 Health Impact Assessment for PM2.5 and Ozone
The Integrated Science Assessment for Paniculate Matter (PM ISA) (U.S. EPA, 2009b)
identified the human health effects associated with ambient PM2.5 exposure, which include
premature mortality and a variety of morbidity effects associated with acute and chronic
exposures. Similarly, the Integrated Science Assessment for Ozone and Related Photochemical
Oxidants (Ozone ISA) (U.S. EPA, 2013b) identified the human health effects associated with
ambient ozone exposure, which include premature mortality and a variety of morbidity effects
associated with acute and chronic exposures. Table 4-6 identifies the quantified and unquantified
co-benefit categories captured in the EPA's health co-benefits estimates for reduced exposure to
ambient PM2.5 and ozone. Although the table below does not list unquantified health effects such
as those associated with exposure to SOi, NO2, and mercury nor welfare effects such as
acidification and nutrient enrichment, these effects are described in detail in Chapters 5 and 6 of
the PM NAAQS RIA (U.S. EPA, 2012a) and summarized later in this chapter. It is important to
emphasize that the list of unquantified benefit categories is not exhaustive, nor is quantification
of each effect complete.
4-13
-------
Table 4-6. Human Health Effects of Ambient PM2.5 and Ozone
Category
Specific Effect
Effect Has Effect Has
More
Been Been T .
.,_ , ,, . , Information
Quantified Monetized
Improved Human Health
Reduced incidence of
premature mortality
from exposure to
PM2.5
Reduced incidence of
morbidity from
exposure to PM2.s
Reduced incidence of
mortality from
exposure to ozone
Reduced incidence of
morbidity from
exposure to ozone
Adult premature mortality based on cohort study
estimates and expert elicitation estimates (age >25
or age >30)
Infant mortality (age <1)
Non-fatal heart attacks (age > 18)
Hospital admissions respiratory (all ages)
Hospital admissions cardiovascular (age >20)
Emergency room visits for asthma (all ages)
Acute bronchitis (age 8-12)
Lower respiratory symptoms (age 7-14)
Upper respiratory symptoms (asthmatics age 9-11)
Asthma exacerbation (asthmatics age 6-18)
Lost work days (age 18-65)
Minor restricted-activity days (age 18-65)
Chronic Bronchitis (age >26)
Emergency room visits for cardiovascular effects
(all ages)
Strokes and cerebrovascular disease (age 50-79)
Other cardiovascular effects (e.g., other ages)
Other respiratory effects (e.g., pulmonary function,
non-asthma ER visits, non-bronchitis chronic
diseases, other ages and populations)
Reproductive and developmental effects (e.g., low
birth weight, pre-term births, etc.)
Cancer, mutagenicity, and genotoxicity effects
Premature mortality based on short-term study
estimates (all ages)
Premature mortality based on long-term study
estimates (age 30-99)
Hospital admissions respiratory causes (age > 65)
Hospital admissions respiratory causes (age <2)
Emergency department visits for asthma (all ages)
Minor restricted-activity days (age 18-65)
School absence days (age 5-17)
Decreased outdoor worker productivity (age 18-65)
Other respiratory effects (e.g., premature aging of
lungs)
Cardiovascular and nervous system effects
Reproductive and developmental effects
S S PM ISA
-------
We follow a "damage-function" approach in calculating benefits, which estimates changes
in individual health endpoints (specific effects that can be associated with changes in air quality)
and assigns values to those changes assuming independence of the values for those individual
endpoints. Because the EPA rarely has the time or resources to perform new research to measure
directly, either health outcomes or their values for regulatory analyses, our estimates are based
on the best available methods of benefits transfer, which is the science and art of adapting
primary research from similar contexts to estimate benefits for the environmental quality change
under analysis. In addition to transferring information from other contexts to the context of this
regulation, we also use a "benefit-per-ton" approach to estimate the PMi.5 and ozone co-benefits
in this RIA. Benefit-per-ton approaches apply an average benefit per ton derived from modeling
of benefits of specific air quality scenarios to estimates of emissions reductions for scenarios
where no air quality modeling is available. Thus, to develop estimates of benefits for this RIA,
we are transferring both the underlying health and economic information from previous studies
and information on air quality responses to emissions reductions from previous air quality
modeling. This section describes the underlying basis for the health and economic valuation
estimates that inform the benefit-per-ton estimates, and the subsequent section provides an
overview of the benefit-per-ton estimates,102 which are described in detail in the appendix to this
chapter.
The benefit-per-ton approach we use in this RIA relies on estimates of human health
responses to exposure to PM and ozone obtained from the peer-reviewed scientific literature.
These estimates are used in conjunction with population data, baseline health information, air
quality data and economic valuation information to conduct health impact and economic benefits
assessments. These assessments form the key inputs to calculating benefit-per-ton estimates. The
next sections provide an overview of the health impact assessment (HIA) methodology and
additional details on several key elements.
The HIA quantifies the changes in the incidence of adverse health impacts resulting from
changes in human exposure to PM2.5 and ozone. We use the environmental Benefits Mapping and
102 We have updated the benefit-per-ton estimates since the proposal RIA. In this RIA, we apply benefit-per-ton
estimates that were derived from air quality modeling of the proposed Clean Power Plan (Option 1 State).
4-15
-------
Analysis Program - Community Edition (BenMAP-CE) (version 1.1) to systematize health
impact analyses by applying a database of key input parameters, including population
projections, health impact functions, and valuation functions (Abt Associates, 2012). For this
assessment, the HIA is limited to those health effects that are directly linked to ambient PM2.5
and ozone concentrations. There may be other indirect health impacts associated with reducing
emissions, such as occupational health exposures. Epidemiological studies generally provide
estimates of the relative risks of a particular health effect for a given increment of air pollution
(often per 10 |^g/m3 for PlVb.5 or ppb for ozone). These relative risks can be used to develop risk
coefficients that relate a unit reduction in PMi.5 to changes in the incidence of a health effect. We
refer the reader to the PM NAAQS RIA (U.S. EPA, 2012a) and Ozone NAAQS RIA (U.S. EPA,
2008b, 2010d) for more information regarding the epidemiology studies and risk coefficients
applied in this analysis, and we briefly elaborate on adult premature mortality below. The size of
the mortality effect estimates from epidemiological studies, the serious nature of the effect itself,
and the high monetary value ascribed to reducing risks of premature death make mortality risk
reduction the most significant health endpoint quantified in this analysis.
4.3.1.1 Mortality Concentration-Response Functions for PM2.5
Considering a substantial body of published scientific literature and reflecting thousands
of epidemiology, toxicology, and clinical studies, the PM ISA documents the association
between elevated PM2.5 concentrations and adverse health effects, including increased premature
mortality (U.S. EPA, 2009b). The PM ISA, which was twice reviewed by the Clean Air
Scientific Advisory Committee of the EPA's Science Advisory Board (SAB-CASAC) (U.S.
EPA-SAB, 2009b, 2009c), concluded that there is a causal relationship between mortality and
both long-term and short-term exposure to PMi.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 function. In addition to
adult mortality discussed in more detail below, we use effect coefficients from Woodruff et al.
(1997) to estimate PM-related infant mortality.
For adult PM-related mortality, we use the effect coefficients from the most recent
epidemiology studies examining two large population cohorts: the American Cancer Society
4-16
-------
cohort (Krewski et al, 2009) and the Harvard Six Cities cohort (Lepeule et al, 2012). The PM
ISA (U.S. EPA, 2009b) concluded that the ACS and Six Cities cohorts produce the strongest
evidence of the association between long-term PM2.5 exposure and premature mortality with
support from a number of additional cohort studies. The SAB's Health Effects Subcommittee
(SAB-HES) also supported using these two cohorts for analyses of the benefits of PM reductions
(U.S. EPA-SAB, 2010a). As both the ACS and Six Cities cohort studies have inherent strengths
and weaknesses, we present PM2.5 co-benefits estimates based on benefits-per-ton derived using
relative risk estimates from both these cohorts.
As a characterization of uncertainty regarding the adult PMi.5-mortality relationship, the
EPA graphically presents the PMi.5 co-benefits based on benefits-per-ton estimated using C-R
functions derived from EPA's expert elicitation study (Roman et al, 2008; ffic, 2006). The
primary goal of the 2006 study was to elicit from a sample of health experts probabilistic
distributions describing uncertainty in estimates of the reduction in mortality among the adult
U.S. population resulting from reductions in ambient annual average PM2.5 concentrations. In
that study, twelve experts provided independent opinions regarding the PMi.s-mortality
concentration-response function. Because the experts relied upon the ACS and Six Cities cohort
studies to inform their concentration-response functions, the benefits estimates based on the
expert responses generally fall between benefits estimates based on these studies (see Figure 4-
1). We do not combine the expert results in order to preserve the breadth and diversity of opinion
on the expert panel. This presentation of the expert-derived results is generally consistent with
SAB advice (U.S. EPA-SAB, 2008), which recommended that the EPA emphasize that
"scientific differences existed only with respect to the magnitude of the effect of PlVb.5 on
mortality, not whether such an effect existed" and that the expert elicitation "supports the
conclusion that the benefits of PlVh.5 control are very likely to be substantial". Although it is
possible that newer scientific literature could revise the experts' quantitative responses if elicited
again, we believe that these general conclusions are unlikely to change.
4.3.1.2 Mortality Concentration-Response Functions for Ozone
In 2008, the National Academies of Science (NRC, 2008) issued a series of
recommendations to the EPA regarding the quantification and valuation of ozone-related short-
term mortality. Chief among these was that".. .short-term exposure to ambient ozone is likely to
4-17
-------
contribute to premature deaths" and the committee recommended that "ozone-related mortality
be included in future estimates of the health benefits of reducing ozone exposures..." The NAS
also recommended that ".. .the greatest emphasis be placed on the multicity and NMMAPS
[National Morbidity, Mortality, and Air Pollution Study] studies without exclusion of the meta-
analyses" (NRC, 2008). In view of the findings of the National Academies panel, we estimate the
co-benefits of avoiding short-term ozone mortality using the Bell et al. (2004) NMMAPS
analysis, the Schwartz (2005) multi-city study, the Huang et al. (2005) multi-city study as well as
effect estimates from the three meta-analyses (Bell et al. (2005), Levy et al. (2005), and Ito et al.
(2005)). These studies are consistent with the studies used in the Ozone NAAQS RIA (U.S.
EPA, 2008b, 2010d).103 For simplicity, we report the ozone mortality estimates in this RIA as a
range reflecting application of dollar-per-ton estimates based on Bell et al. (2004) and Levy et al.
(2005) to represent the lowest and the highest co-benefits estimates based on these six ozone
mortality studies. In addition, we graphically present in Figure 4-1 the estimated co-benefits
based on dollar-per-ton estimates derived from all six studies mentioned above as a
characterization of uncertainty regarding the ozone -mortality relationship.
4.3.2 Economic Valuation for Health Co-benefits
After quantifying the change in adverse health impacts, we estimate the economic value of
these avoided impacts. Reductions in ambient concentrations of air pollution generally lower the
risk of future adverse health effects by a small amount for a large population. Therefore, the
appropriate economic measure is willingness to pay (WTP) for changes in risk of a health effect.
For some health effects, such as hospital admissions, WTP estimates are generally not available,
so we use the cost of treating or mitigating the effect. These cost-of-illness (COI) estimates
generally (although not necessarily in every case) understate the true value of reductions in risk
of a health effect. They tend to reflect the direct expenditures related to treatment but not the
value of avoided pain and suffering from the health effect. The unit values applied in this
103 Since the EPA received NAS advice, the Agency published the Ozone ISA (U.S. EPA, 2013b) and the second
draft Ozone Health Risk and Exposure Assessment (U.S. EPA, 2014a). Therefore, the ozone mortality studies
applied in this analysis, while current at the time of the previous Ozone NAAQS RIAs, do not reflect the most
updated literature available. The selection of ozone mortality studies used to estimate benefits in RIAs will be
revisited in the forthcoming RIA accompanying the on-going review of the Ozone NAAQS.
4-18
-------
analysis are provided in Table 5-9 of the PM NAAQS RIA for each health endpoint (U.S. EPA,
2012a).
Avoided premature deaths account for 98 percent of monetized PM-related co-benefits and
over 90 percent of monetized ozone-related co-benefits. The economics literature concerning the
appropriate method for valuing reductions in premature mortality risk is still developing. The
adoption of a 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), the 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 reductions in mortality risk (U.S. EPA-SAB,
2000). The VSL approach is a summary measure for the value of small changes in mortality risk
experienced by a large number of people.
The 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, the EPA has decided to apply the VSL
that was vetted and endorsed by the SAB in the Guidelines for Preparing Economic Analyses
(U.S. EPA, 2014)104 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 $6.3 million (2000$).105 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 VSLs applied in
104 In the updated Guidelines for Preparing Economic Analyses (U.S. EPA, 2010e), the EPA retained the VSL
endorsed by the SAB with the understanding that further updates to the mortality risk valuation guidance would be
forthcoming.
105 In 1990$, this base VSL is $4.8 million.
4-19
-------
this analysis in 2011$ after adjusting for income growth are $9.9 million for 2020 and $10.1
million for 2025 and 2030.106
The Agency is committed to using scientifically sound, appropriately reviewed evidence in
valuing mortality risk reductions and has made significant progress in responding to the SAB-
EEAC's specific recommendations. In the process, the Agency has identified a number of
important issues to be considered in updating its mortality risk valuation estimates. These are
detailed in a white paper, "Valuing Mortality Risk Reductions in Environmental Policy" (U.S.
EPA, 2010c), which recently underwent review by the SAB-EEAC. A meeting with the SAB on
this paper was held on March 14, 2011 and formal recommendations were transmitted on
July 29, 2011 (U.S. EPA-SAB, 2011). The EPA is taking SAB's recommendations under
advisement.
In valuing PMi.s-related premature mortality, we discount the value of premature mortality
occurring in future years using rates of 3 percent and 7 percent (OMB, 2003). We assume that
there is a "cessation" lag between changes in PM exposures and the total realization of changes
in health effects. Although the structure of the lag is uncertain, the EPA follows the advice of the
SAB-HES to assume a segmented lag structure characterized by 30 percent of mortality
reductions in the first year, 50 percent over years 2 to 5, and 20 percent over the years 6 to 20
after the reduction in PMi.5 (U.S. EPA-SAB, 2004c). Changes in the cessation lag assumptions
do not change the total number of estimated deaths but rather the timing of those deaths. Because
short-term ozone-related premature mortality occurs within the analysis year, the estimated
ozone-related co-benefits are identical for all discount rates.
4.3.3 Benefit-per-ton Estimates for PM2.5
We used a "benefit-per-ton" approach to estimate the PMi.5 co-benefits in this RIA. The
EPA has applied this approach in several previous RIAs (e.g., U.S. EPA, 201 Ib, 201 Ic, 2012b,
2014a). These benefit-per-ton estimates provide the total monetized human health co-benefits
(the sum of premature mortality and premature morbidity), of reducing one ton of PMi.5 (or
PM2.5 precursor such as NOx or SOi) from a specified source. Specifically, in this analysis, we
106 Income growth projections are only currently available in BenMAP through 2024, so both the 2025 and 2030
estimates use income growth only through 2024 and are therefore likely underestimates.
4-20
-------
multiplied the benefit-per-ton estimates by the corresponding emission reductions that were
generated from air quality modeling of the proposed Clean Power Plan.
The method used to calculate the regional benefit-per-ton estimates is similar to the
average EGU sector estimates used for the proposal (U.S. EPA, 2013a), but relies on air quality
modeling of the proposed Clean Power Plan. Similar to the proposal, we generated regional
benefit-per-ton estimates by aggregating the impacts in BenMAP to the region (i.e., East, West,
and California) rather than aggregating to the nation. The appendix to this chapter provides
additional detail regarding these calculations.
As noted below in the characterization of uncertainty, all benefit-per-ton estimates have
inherent limitations. Specifically, all benefit-per-ton estimates reflect the geographic distribution
of the modeled proposal, which may not match the emission reductions anticipated by the final
emission guidelines, and they may not reflect local variability in population density,
meteorology, exposure, baseline health incidence rates, or other local factors for any specific
location. The regional benefit-per-ton estimates, although less subject to these types of
uncertainties than national estimates, still should be interpreted with caution. Even though we
assume that all fine particles have equivalent health effects, the benefit-per-ton estimates vary
between precursors depending on the location and magnitude of their impact on PMi.5 levels,
which drive population exposure.
4.3.4 Benefit-per-ton Estimates for Ozone
Similar to PlVh.5, we used a "benefit-per-ton" approach in this RIA to estimate the ozone
co-benefits, which represent the total monetized human health co-benefits (the sum of premature
mortality and premature morbidity) of reducing one ton of NOx (an ozone precursor). Also
consistent with the PM2.5 estimates, we generated regional benefit-per-ton estimates for ozone
based on air quality modeling for the proposed Clean Power Plan. In contrast to the PlVb.5
estimates, the ozone estimates are not based on changes to annual emissions. Instead, the
regional estimates (i.e., East, West, and California) correspond to NOx emissions from U.S.
EGUs during the ozone-season (May to September). Because we estimate ozone health impacts
from May to September only, this approach underestimates ozone co-benefits in areas with a
longer ozone season such as southern California and Texas. These estimates assume that EGU-
4-21
-------
attributable ozone formation at the regional-level is due to NOx alone. Because EGUs emit little
VOC relative to NOx emissions, it is unlikely that VOCs emitted by EGUs would contribute
substantially to regional ozone formation. As noted above, all benefit-per-ton estimates have
inherent limitations and should be interpreted with caution. We provide more detailed
information regarding the generation of these estimates in the appendix to this chapter.
4.3.5 Estimated Health Co-Benefits Results
Tables 4-7 through 4-9 provide the regional benefit-per-ton estimates for three analysis
years: 2020, 2025, and 2030. Tables 4-10 through 4-12 and 4-13 through 4-15 provide the
emission reductions estimated to occur in each analysis year for the rate-based and mass-based
illustrative plan approaches, respectively, by region (i.e., East, West, and California).107 Tables
4-16 through 4-18 and 4-19 through 4-21 summarize the national monetized PM and ozone-
related health co-benefits estimated to occur in each analysis year for the illustrative rate-based
and mass-based plan approaches, respectively, by precursor pollutant using discount rates of 3
percent and 7 percent. Tables 4-22 through 4-24 and 4-25 through 4-27 provide national
summaries of the reductions in estimated health incidences associated with the illustrative rate-
based and mass-based plan approaches, respectively, in each analysis year.108 Figure 4-1
provides a visual representation of the range of estimated PlVb.5 and ozone-related co-benefits
using benefit-per-ton estimates based on concentration-response functions from different studies
and expert opinion for the illustrative rate-based and mass-based plan approaches evaluated in
2025 as an illustrative analysis year. Figure 4-2 provides a breakdown of the monetized health
co-benefits for the rate-based and mass-based plan approaches evaluated in 2025 as an
illustrative analysis year by precursor pollutant.
107 See Chapter 3 of this RIA for more information regarding the expected emission reductions used to calculate the
health co-benefits in this chapter. Chapter 3 also provides more information regarding the illustrative plan approach.
108 Incidence estimates were generated using the same "per ton" approach as used to generate the dollar benefit per
ton values. See Appendix 4-A for details.
4-22
-------
Table 4-7. Summary of Regional PM2.5 Benefit-per-Ton Estimates Based on Air Quality
Modeling from Proposed Clean Power Plan in 2020 (2011$)*
Pollutant Discount Rat
SO2
Directly emitted PM2.s
(EC+OC)
Directly emitted PM2.s
(crustal)
NOX (as PM2.5)
NOx (as Ozone)
3%
7%
3%
7%
3%
7%
3%
7%
N/A
East
$33,000 to $75,000
$30,000 to $68,000
$140,000 to $320,000
$130,000 to $290,000
$23,000 to $52,000
$21,000 to $47,000
$3, 100 to $7,000
$2,800 to $6,300
$6,500 to $28,000
Regional
West
$6,200 to $14,000
$5,600 to $13,000
$27,000 to $60,000
$24,000 to $54,000
$11,000 to $25,000
$9,900 to $22,000
$0,670 to $1,500
$0,6 10 to $1,400
$2,000 to $8,900
California
$95,000 to $210,000
$85,000 to $190,000
$370,000 to $830,000
$330,000 to $740,000
$73,000 to $160,000
$66,000 to $150,000
$22,000 to $49,000
$19,000 to $44,000
$14,000 to $59,000
* The range of estimates reflects the range of epidemiology studies for avoided premature mortality for PM2.s and
ozone. All estimates are rounded to two significant figures. The monetized co-benefits do not include reduced
health effects from direct exposure to NO2, SO2, ecosystem effects, or visibility impairment. All fine particles are
assumed to have equivalent health effects, but the benefit-per-ton estimates vary depending on the location and
magnitude of their impact on PM2.s concentrations, which drive population exposure. The monetized co-benefits
incorporate the conversion from precursor emissions to ambient fine particles and ozone. Benefit-per-ton
estimates for ozone are based on ozone season NOx emissions. Ozone co-benefits occur in analysis year, so they
are the same for all discount rates. Confidence intervals are unavailable for this analysis because of the benefit-
per-ton methodology. In general, the 95th percentile confidence interval for monetized PM2.5 benefits ranges from
approximately -90 percent to +180 percent of the central estimates based on Krewski et al (2009) and Lepeule et
al. (2012).
Table 4-8. Summary of Regional PM2.5 Benefit-per-Ton Estimates Based on Air Quality
Modeling from Proposed Clean Power Plan in 2025 (2011$)*
Pollutant Discount Rat
SO2
Directly emitted PM2.5
(EC+OC)
Directly emitted PM2.5
(crustal)
NOx (as PM2.5)
NOx (as Ozone)
3%
7%
3%
7%
3%
7%
3%
7%
N/A
East
$37,000 to $83,000
$33,000 to $75,000
$160,000 to $360,000
$140,000 to $320,000
$25,000 to $58,000
$23,000 to $52,000
$3,300 to $7,500
$3,000 to $6,800
$7, 100 to $30,000
Regional
West
$7, 100 to $16,000
$6,400 to $14,000
$30,000 to $68,000
$27,000 to $61,000
$12,000 to $28,000
$11,000 to $25,000
$0,750 to $1,700
$0,670 to $1,500
$2,300 to $10,000
California
$110,000 to $240,000
$97,000 to $220,000
$410,000 to $930,000
$370,000 to $830,000
$82,000 to $180,000
$74,000 to $170,000
$24,000 to $54,000
$22,000 to $49,000
$15,000 to $66,000
* The range of estimates reflects the range of epidemiology studies for avoided premature mortality for PM2.s and
ozone. All estimates are rounded to two significant figures. The monetized co-benefits do not include reduced
health effects from direct exposure to NO2, SO2, ecosystem effects, or visibility impairment. All fine particles are
assumed to have equivalent health effects, but the benefit-per-ton estimates vary depending on the location and
magnitude of their impact on PM2.s concentrations, which drive population exposure. The monetized co-benefits
incorporate the conversion from precursor emissions to ambient fine particles and ozone. Benefit-per-ton
estimates for ozone are based on ozone season NOx emissions. Ozone co-benefits occur in analysis year, so they
are the same for all discount rates. Confidence intervals are unavailable for this analysis because of the benefit-
per-ton methodology. In general, the 95th percentile confidence interval for monetized PM2.5 benefits ranges from
approximately -90 percent to +180 percent of the central estimates based on Krewski et al. (2009) and Lepeule et
al. (2012).
4-23
-------
Table 4-9. Summary of Regional PM2.5 Benefit-per-Ton Estimates Based on Air Quality
Modeling from Proposed Clean Power Plan in 2030 (2011$)*
Pollutant
SO2
Directly emitted PM2.s
(EC+OC)
Directly emitted PM2.s
(crustal)
NOX (as PM2.5)
NOx (as Ozone)
Discount
Rate
3%
7%
3%
7%
3%
7%
3%
7%
N/A
East
$40,000 to $89,000
$36,000 to $8 1,000
$170,000 to $380,000
$150,000 to $340,000
$28,000 to $62,000
$25,000 to $56,000
$3,500 to $8,000
$3,200 to $7,200
$7,600 to $33,000
Regional
West
$7,800 to $18,000
$7, 100 to $16,000
$33,000 to $75,000
$30,000 to $68,000
$14,000 to $31,000
$13,000 to $28,000
$0,820 to $1,900
$0,740 to $1,700
$2,600 to $11, 000
California
$120,000 to $270,000
$110,000 to $240,000
$450,000 to $1,000,000
$410,000 to $920,000
$90,000 to $200,000
$8 1,000 to $180,000
$26,000 to $60,000
$24,000 to $54,000
$17,000 to $73,000
* The range of estimates reflects the range of epidemiology studies for avoided premature mortality for PM2.s and
ozone. All estimates are rounded to two significant figures. The monetized co-benefits do not include reduced
health effects from direct exposure to NO2, SO2, ecosystem effects, or visibility impairment. All fine particles are
assumed to have equivalent health effects, but the benefit-per-ton estimates vary depending on the location and
magnitude of their impact on PM2.s concentrations, which drive population exposure. The monetized co-benefits
incorporate the conversion from precursor emissions to ambient fine particles and ozone. Benefit-per-ton
estimates for ozone are based on ozone season NOx emissions. Ozone co-benefits occur in analysis year, so they
are the same for all discount rates. Confidence intervals are unavailable for this analysis because of the benefit-
per-ton methodology. In general, the 95th percentile confidence interval for monetized PM2.s benefits ranges from
approximately -90 percent to +180 percent of the central estimates based on Krewski et al (2009) and Lepeule et
al. (2012).
Table 4-10. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Rate-based Illustrative Plan Approach in 2020 (thousands of short tons)*
Region SO2 All-year NOx Ozone-Season NOx
East
West
California
13
1
0
50
1
0
19
0
0
National Total
14
50
19
*A11 emissions shown in the table are rounded, so regional emission reductions may appear to not sum to national
total. The final emissions guidelines are also expected to result in reductions in directly emitted PM2.s, which we
were not able to estimate for this RIA.
Table 4-11. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Rate-based Illustrative Plan Approach in 2025 (thousands of short tons)*
Region
SO2
All-year NOx
Ozone-Season NOx
East
West
California
171
7
1
155
8
2
67
3
0
National Total
178
165
70
*A11 emissions shown in the table are rounded, so regional emission reductions may appear to not sum to national
total. The final emissions guidelines are also expected to result in reductions in directly emitted PM2.s, which we
were not able to estimate for this RIA.
4-24
-------
Table 4-12. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Rate-based Illustrative Plan Approach in 2030 (thousands of short tons)*
Region SOi All-year NOx Ozone-Season NOx
East
West
California
306
11
1
263
15
4
109
9
0
National Total 318 282 118
*A11 emissions shown in the table are rounded, so regional emission reductions may appear to not sum to national
total. The final emissions guidelines are also expected to result in reductions in directly emitted PMi.s, which we
were not able to estimate for this RIA.
Table 4-13. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Mass-based Illustrative Plan Approach in 2020 (thousands of short tons)*
Region SOi All-year NOx Ozone-Season NOx
East
West
California
49
4
0
57
4
0
22
1
0
National Total 54 60 23
*A11 emissions shown in the table are rounded, so regional emission reductions may appear to not sum to national
total. The final emissions guidelines are also expected to result in reductions in directly emitted PMi.s, which we
were not able to estimate for this RIA.
Table 4-14. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Mass-based Illustrative Plan Approach in 2025 (thousands of short tons)*
Region SOi All-year NOx Ozone-Season NOx
East
West
California
156
29
0
169
34
0
74
14
0
National Total 185 203 88
*A11 emissions shown in the table are rounded, so regional emission reductions may appear to not sum to national
total. The final emissions guidelines are also expected to result in reductions in directly emitted PM2.s, which we
were not able to estimate for this RIA.
Table 4-15. Emission Reductions of Criteria Pollutants for the Final Emission Guidelines
Mass-based Illustrative Plan Approach in 2030 (thousands of short tons)*
Region SOi All-year NOx Ozone-Season NOx
East
West
California
243
36
1
229
48
1
99
21
1
National Total 280 279 121
*A11 emissions shown in the table are rounded, so regional emission reductions may appear to not sum to national
total. The final emissions guidelines are also expected to result in reductions in directly emitted PM2.s, which we
were not able to estimate for this RIA.
4-25
-------
Table 4-16. Summary of Estimated Monetized Health Co-Benefits for the Final Emission
Guidelines Rate-based Illustrative Plan Approach in 2020 (billions of 2011$) *
Pollutant 3% Discount Rate 7% Discount Rate
SO2 $0.44 to $0.99 $0.39 to $0.89
NOx(asPM2.5) $0.14 to $0.33 $0.13 to $0.30
NOx (as Ozone) $0.12 to $0.52 $0.12 to $0.52
Total $0.70 to $1.8 $0.64 to $1.7
* All estimates are rounded to two significant figures so numbers may not sum down columns. The estimated
monetized co-benefits do not include climate benefits or reduced health effects from direct exposure to NO2, SO2,
ecosystem effects, or visibility impairment. All fine particles are assumed to have equivalent health effects, but the
benefit-per-ton estimates vary depending on the location and magnitude of their impact on PM2.s levels, which drive
population exposure. The monetized co-benefits incorporate the conversion from precursor emissions to ambient
fine particles and ozone. Co-benefits for PM2.5precursors are based on regional benefit-per-ton estimates. Co-
benefits for ozone are based on ozone season NOx emissions. Ozone co-benefits occur in analysis year, so they are
the same for all discount rates. Confidence intervals are unavailable for this analysis because of the benefit-per-ton
methodology. In general, the 95th percentile confidence interval for monetized PM2.s benefits ranges from
approximately -90 percent to +180 percent of the central estimates based on Krewski et al. (2009) and Lepeule et al.
(2012). For this RIA, we did not estimate changes in emissions of directly emitted particles. As a result, quantified
PM2.s related benefits are underestimated by a relatively small amount. In the proposal RIA, the benefits from
reductions in directly emitted PM2.s were less than 10 percent of total monetized health co-benefits across all
scenarios and years.
Table 4-17. Summary of Estimated Monetized Health Co-Benefits for the Final Emission
Guidelines Rate-based Illustrative Plan Approach in 2025 (billions of 2011$) *
Pollutant 3% Discount Rate 7% Discount Rate
SO2 $6.4 to $14 $5.7 to $13
NOx(asPM2.5) $0.56 to $1.3 $0.50 to $1.1
NOx (as Ozone) $0.49 to $2.1 $0.49 to $2.1
Total $7.4 to $18 $6.7 to $16
* All estimates are rounded to two significant figures so numbers may not sum down columns. The estimated
monetized co-benefits do not include climate benefits or reduced health effects from direct exposure to NO2, SO2,
ecosystem effects, or visibility impairment. All fine particles are assumed to have equivalent health effects, but the
benefit-per-ton estimates vary depending on the location and magnitude of their impact on PM2.s levels, which drive
population exposure. The monetized co-benefits incorporate the conversion from precursor emissions to ambient
fine particles and ozone. Co-benefits for PM2.sprecursors are based on regional benefit-per-ton estimates. Co-
benefits for ozone are based on ozone season NOx emissions. Ozone co-benefits occur in analysis year, so they are
the same for all discount rates. Confidence intervals are unavailable for this analysis because of the benefit-per-ton
methodology. In general, the 95th percentile confidence interval for monetized PM2.s benefits ranges from
approximately -90 percent to +180 percent of the central estimates based on Krewski et al. (2009) and Lepeule et al.
(2012). For this RIA, we did not estimate changes in emissions of directly emitted particles. As a result, quantified
PM2.s related benefits are underestimated by a relatively small amount. In the proposal RIA, the benefits from
reductions in directly emitted PM2.s were less than 10 percent of total monetized health co-benefits across all
scenarios and years.
4-26
-------
Table 4-18. Summary of Estimated Monetized Health Co-Benefits for the Final Emission
Guidelines Rate-based Illustrative Plan Approach in 2030 (billions of 2011$) *
Pollutant 3% Discount Rate 7% Discount Rate
SO2 $12 to $28 $11 to $25
NOx(asPM2.5) $1.0 to $2.3 $0.93 to $2.1
NOx (as Ozone) $0.86 to $3.7 $0.86 to $3.7
Total $14 to $34 $13 to $31
* All estimates are rounded to two significant figures so numbers may not sum down columns. The estimated
monetized co-benefits do not include climate benefits or reduced health effects from direct exposure to NO2, SO2,
ecosystem effects, or visibility impairment. All fine particles are assumed to have equivalent health effects, but the
benefit-per-ton estimates vary depending on the location and magnitude of their impact on PM2.s levels, which drive
population exposure. The monetized co-benefits incorporate the conversion from precursor emissions to ambient
fine particles and ozone. Co-benefits for PM2.5precursors are based on regional benefit-per-ton estimates. Co-
benefits for ozone are based on ozone season NOx emissions. Ozone co-benefits occur in analysis year, so they are
the same for all discount rates. Confidence intervals are unavailable for this analysis because of the benefit-per-ton
methodology. In general, the 95th percentile confidence interval for monetized PM2.s benefits ranges from
approximately -90 percent to +180 percent of the central estimates based on Krewski et al. (2009) and Lepeule et al.
(2012). For this RIA, we did not estimate changes in emissions of directly emitted particles. As a result, quantified
PM2.s related benefits are underestimated by a relatively small amount. In the proposal RIA, the benefits from
reductions in directly emitted PM2.s were less than 10 percent of total monetized health co-benefits across all
scenarios and years.
Table 4-19. Summary of Estimated Monetized Health Co-Benefits for the Final Emission
Guidelines Mass-based Illustrative Plan Approach in 2020 (billions of 2011$) *
Pollutant 3% Discount Rate 7% Discount Rate
SO2 $1.7 to $3.8 $1.5 to $3.4
NOx(asPM2.5) $0.17 to $0.39 $0.16 to $0.36
NOx (as Ozone) $0.14 to $0.61 $0.14 to $0.61
Total $2.0 to $4.8 $1.8 to $4.4
* All estimates are rounded to two significant figures so numbers may not sum down columns. The estimated
monetized co-benefits do not include climate benefits or reduced health effects from direct exposure to NO2, SO2,
ecosystem effects, or visibility impairment. All fine particles are assumed to have equivalent health effects, but the
benefit-per-ton estimates vary depending on the location and magnitude of their impact on PM2.s levels, which drive
population exposure. The monetized co-benefits incorporate the conversion from precursor emissions to ambient
fine particles and ozone. Co-benefits for PM2.sprecursors are based on regional benefit-per-ton estimates. Co-
benefits for ozone are based on ozone season NOx emissions. Ozone co-benefits occur in analysis year, so they are
the same for all discount rates. Confidence intervals are unavailable for this analysis because of the benefit-per-ton
methodology. In general, the 95th percentile confidence interval for monetized PM2.s benefits ranges from
approximately -90 percent to +180 percent of the central estimates based on Krewski et al. (2009) and Lepeule et al.
(2012). For this RIA, we did not estimate changes in emissions of directly emitted particles. As a result, quantified
PM2.s related benefits are underestimated by a relatively small amount. In the proposal RIA, the benefits from
reductions in directly emitted PM2.s were less than 10 percent of total monetized health co-benefits across all
scenarios and years.
4-27
-------
Table 4-20. Summary of Estimated Monetized Health Co-Benefits for the Final Emission
Guidelines Mass-based Illustrative Plan Approach in 2025 (billions of 2011$) *
Pollutant 3% Discount Rate 7% Discount Rate
SO2 $6.0 to $13 $5.4 to $12
NOx(asPM2.5) $0.58 to $1.3 $0.52 to $1.2
NOx (as Ozone) $0.56 to $2.4 $0.56 to $2.4
Total $7.1 to $17 $6.5 to $16
* All estimates are rounded to two significant figures so numbers may not sum down columns. The estimated
monetized co-benefits do not include climate benefits or reduced health effects from direct exposure to NO2, SO2,
ecosystem effects, or visibility impairment. All fine particles are assumed to have equivalent health effects, but the
benefit-per-ton estimates vary depending on the location and magnitude of their impact on PM2.s levels, which drive
population exposure. The monetized co-benefits incorporate the conversion from precursor emissions to ambient
fine particles and ozone. Co-benefits for PM2.sprecursors are based on regional benefit-per-ton estimates. Co-
benefits for ozone are based on ozone season NOx emissions. Ozone co-benefits occur in analysis year, so they are
the same for all discount rates. Confidence intervals are unavailable for this analysis because of the benefit-per-ton
methodology. In general, the 95th percentile confidence interval for monetized PM2.5 benefits ranges from
approximately -90 percent to +180 percent of the central estimates based on Krewski et al (2009) and Lepeule et al.
(2012). For this RIA, we did not estimate changes in emissions of directly emitted particles. As a result, quantified
PM2.5 related benefits are underestimated by a relatively small amount. In the proposal RIA, the benefits from
reductions in directly emitted PM2.5 were less than 10 percent of total monetized health co-benefits across all
scenarios and years.
Table 4-21. Summary of Estimated Monetized Health Co-Benefits for the Final Emission
Guidelines Mass-based Illustrative Plan Approach in 2030 (billions of 2011$) *
Pollutant 3% Discount Rate 7% Discount Rate
SO2 $10 to $23 $9.0 to $20
NOx(asPM2.5) $0.87 to $2.0 $0.79 to $1.8
NOx (as Ozone) $0.82 to $3.5 $0.82 to $3.5
Total $12 to $28 $11 to $26
* All estimates are rounded to two significant figures so numbers may not sum down columns. The estimated
monetized co-benefits do not include climate benefits or reduced health effects from direct exposure to NO2, SO2,
ecosystem effects, or visibility impairment. All fine particles are assumed to have equivalent health effects, but the
benefit-per-ton estimates vary depending on the location and magnitude of their impact on PM2.s levels, which drive
population exposure. The monetized co-benefits incorporate the conversion from precursor emissions to ambient
fine particles and ozone. Co-benefits for PM2.sprecursors are based on regional benefit-per-ton estimates. Co-
benefits for ozone are based on ozone season NOx emissions. Ozone co-benefits occur in analysis year, so they are
the same for all discount rates. Confidence intervals are unavailable for this analysis because of the benefit-per-ton
methodology. In general, the 95th percentile confidence interval for monetized PM2.5 benefits ranges from
approximately -90 percent to +180 percent of the central estimates based on Krewski et al. (2009) and Lepeule et al.
(2012). For this RIA, we did not estimate changes in emissions of directly emitted particles. As a result, quantified
PM2.5 related benefits are underestimated by a relatively small amount. In the proposal RIA, the benefits from
reductions in directly emitted PM2.5 were less than 10 percent of total monetized health co-benefits across all
scenarios and years.
4-28
-------
Table 4-22. Summary of Avoided Health Incidences from PIVh.s-Related and Ozone-
Related Co-benefits for the Final Emission Guidelines Rate-based Illustrative
Plan Approach in 2020*
PM2.s-related Health Effects
Avoided Premature Mortality
Krewski et al (2009) (adult)
Lepeule et al. (2012) (adult)
Woodruff et al. (1997) (infant)
64
140
0
Avoided Morbidity
Emergency department visits for asthma (all ages)
Acute bronchitis (age 8-12)
Lower respiratory symptoms (age 7-14)
Upper respiratory symptoms (asthmatics age 9-1 1)
Minor restricted-activity days (age 18-65)
Lost work days (age 18-65)
Asthma exacerbation (age 6-18)
Hospital admissions respiratory (all ages)
Hospital admissions cardiovascular (age > 18)
Non-Fatal Heart Attacks (age >18)
Peters et al. (2001)
Pooled estimate of 4 studies
34
94
1,200
1,700
47,000
7,900
4,200
19
23
73
8
Ozone-related Health Effects
Avoided Premature Mortality
Bell et al. (2004) (all ages)
Levy et al. (2005) (all ages)
11
51
Avoided Morbidity
Hospital admissions respiratory causes (ages > 65)
Hospital admissions respiratory causes (ages < 2)
Emergency room visits for asthma (all ages)
Minor restricted-activity days (ages 18-65)
School absence days
66
33
37
66,000
23,000
* All estimates are rounded to whole numbers with two significant figures. Co-benefits for PMi.5precursors are
based on regional incidence-per-ton estimates for all precursors. Co-benefits for ozone are based on ozone season
NOx emissions. Confidence intervals are unavailable for this analysis because of the incidence-per-ton
methodology. In general, the 95th percentile confidence interval for the health impact function alone ranges from
approximately +30 percent for mortality incidence based on Krewski et al (2009) and +46 percent based on Lepeule
et al. (2012).
4-29
-------
Table 4-23. Summary of Avoided Health Incidences from PIVh.s-Related and Ozone-
Related Co-benefits for Final Emission Guidelines Rate-based Illustrative Plan
Approach in 2025*
PM2.s-related Health Effects
Avoided Premature Mortality
Krewski et al (2009) (adult)
Lepeule et al. (2012) (adult)
Woodruff et al. (1997) (infant)
740
1,700
2
Avoided Morbidity
Emergency department visits for asthma (all ages)
Acute bronchitis (age 8-12)
Lower respiratory symptoms (age 7-14)
Upper respiratory symptoms (asthmatics age 9-1 1)
Minor restricted-activity days (age 18-65)
Lost work days (age 18-65)
Asthma exacerbation (age 6-18)
Hospital admissions respiratory (all ages)
Hospital admissions cardiovascular (age > 18)
Non-Fatal Heart Attacks (age >18)
Peters et al. (2001)
Pooled estimate of 4 studies
380
1,100
14,000
20,000
530,000
89,000
48,000
220
270
860
93
Ozone-related Health Effects
Avoided Premature Mortality
Bell et al. (2004) (all ages)
Levy et al. (2005) (all ages)
44
200
Avoided Morbidity
Hospital admissions respiratory causes (ages > 65)
Hospital admissions respiratory causes (ages < 2)
Emergency room visits for asthma (all ages)
Minor restricted-activity days (ages 18-65)
School absence days
280
130
140
250,000
87,000
* All estimates are rounded to whole numbers with two significant figures. Co-benefits for PMi.5precursors are
based on regional incidence-per-ton estimates for all precursors. Co-benefits for ozone are based on ozone season
NOx emissions. Confidence intervals are unavailable for this analysis because of the incidence-per-ton
methodology. In general, the 95th percentile confidence interval for the health impact function alone ranges from
approximately +30 percent for mortality incidence based on Krewski et al (2009) and +46 percent based on Lepeule
et al. (2012).
4-30
-------
Table 4-24. Summary of Avoided Health Incidences from PIVh.s-Related and Ozone-
Related Co-Benefits for Final Emission Guidelines Rate-based Illustrative Plan
Approach in 2030*
PM2.s-related Health Effects
Avoided Premature Mortality
Krewski et al (2009) (adult)
Lepeule et al. (2012) (adult)
Woodruff et al. (1997) (infant)
1,400
3,200
3
Avoided Morbidity
Emergency department visits for asthma (all ages)
Acute bronchitis (age 8-12)
Lower respiratory symptoms (age 7-14)
Upper respiratory symptoms (asthmatics age 9-1 1)
Minor restricted-activity days (age 18-65)
Lost work days (age 18-65)
Asthma exacerbation (age 6-18)
Hospital admissions respiratory (all ages)
Hospital admissions cardiovascular (age > 18)
Non-Fatal Heart Attacks (age >18)
Peters et al. (2001)
Pooled estimate of 4 studies
540
2,000
26,000
37,000
970,000
160,000
90,000
440
530
1,700
180
Ozone-related Health Effects
Avoided Premature Mortality
Bell et al. (2004) (all ages)
Levy et al. (2005) (all ages)
73
330
Avoided Morbidity
Hospital admissions respiratory causes (ages > 65)
Hospital admissions respiratory causes (ages < 2)
Emergency room visits for asthma (all ages)
Minor restricted-activity days (ages 18-65)
School absence days
500
200
220
400,000
140,000
* All estimates are rounded to whole numbers with two significant figures. Co-benefits for PMi.5precursors are
based on regional incidence-per-ton estimates for all precursors. Co-benefits for ozone are based on ozone season
NOx emissions. Confidence intervals are unavailable for this analysis because of the incidence-per-ton
methodology. In general, the 95th percentile confidence interval for the health impact function alone ranges from
approximately +30 percent for mortality incidence based on Krewski et al (2009) and +46 percent based on Lepeule
et al. (2012).
4-31
-------
Table 4-25. Summary of Avoided Health Incidences from PIVh.s-Related and Ozone-
Related Co-benefits for the Final Emission Guidelines Mass-based Illustrative
Plan Approach in 2020*
PM2.s-related Health Effects
Avoided Premature Mortality
Krewski et al (2009) (adult)
Lepeule et al. (2012) (adult)
Woodruff et al. (1997) (infant)
200
460
0
Avoided Morbidity
Emergency department visits for asthma (all ages)
Acute bronchitis (age 8-12)
Lower respiratory symptoms (age 7-14)
Upper respiratory symptoms (asthmatics age 9-1 1)
Minor restricted-activity days (age 18-65)
Lost work days (age 18-65)
Asthma exacerbation (age 6-18)
Hospital admissions respiratory (all ages)
Hospital admissions cardiovascular (age > 18)
Non-Fatal Heart Attacks (age >18)
Peters et al. (2001)
Pooled estimate of 4 studies
110
300
3,800
5,500
150,000
25,000
13,000
59
73
230
25
Ozone-related Health Effects
Avoided Premature Mortality
Bell et al. (2004) (all ages)
Levy et al. (2005) (all ages)
13
61
Avoided Morbidity
Hospital admissions respiratory causes (ages > 65)
Hospital admissions respiratory causes (ages < 2)
Emergency room visits for asthma (all ages)
Minor restricted-activity days (ages 18-65)
School absence days
78
40
43
78,000
27,000
* All estimates are rounded to whole numbers with two significant figures. Co-benefits for PMi.5precursors are
based on regional incidence-per-ton estimates for all precursors. Co-benefits for ozone are based on ozone season
NOx emissions. Confidence intervals are unavailable for this analysis because of the incidence-per-ton
methodology. In general, the 95th percentile confidence interval for the health impact function alone ranges from
approximately +30 percent for mortality incidence based on Krewski et al (2009) and +46 percent based on Lepeule
et al. (2012).
4-32
-------
Table 4-26. Summary of Avoided Health Incidences from PIVh.s-Related and Ozone-
Related Co-benefits for Final Emission Guidelines Mass-based Illustrative Plan
Approach in 2025*
PM2.s-related Health Effects
Avoided Premature Mortality
Krewski et al (2009) (adult)
Lepeule et al. (2012) (adult)
Woodruff et al. (1997) (infant)
700
1,600
2
Avoided Morbidity
Emergency department visits for asthma (all ages)
Acute bronchitis (age 8-12)
Lower respiratory symptoms (age 7-14)
Upper respiratory symptoms (asthmatics age 9-1 1)
Minor restricted-activity days (age 18-65)
Lost work days (age 18-65)
Asthma exacerbation (age 6-18)
Hospital admissions respiratory (all ages)
Hospital admissions cardiovascular (age > 18)
Non-Fatal Heart Attacks (age >18)
Peters et al. (2001)
Pooled estimate of 4 studies
350
1,000
13,000
19,000
500,000
84,000
46,000
210
260
810
88
Ozone-related Health Effects
Avoided Premature Mortality
Bell et al. (2004) (all ages)
Levy et al. (2005) (all ages)
51
230
Avoided Morbidity
Hospital admissions respiratory causes (ages > 65)
Hospital admissions respiratory causes (ages < 2)
Emergency room visits for asthma (all ages)
Minor restricted-activity days (ages 18-65)
School absence days
320
150
160
290,000
100,000
* All estimates are rounded to whole numbers with two significant figures. Co-benefits for PMi.5precursors are
based on regional incidence-per-ton estimates for all precursors. Co-benefits for ozone are based on ozone season
NOx emissions. Confidence intervals are unavailable for this analysis because of the incidence-per-ton
methodology. In general, the 95th percentile confidence interval for the health impact function alone ranges from
approximately +30 percent for mortality incidence based on Krewski et al (2009) and +46 percent based on Lepeule
et al. (2012).
4-33
-------
Table 4-27. Summary of Avoided Health Incidences from PIVh.s-Related and Ozone-
Related Co-Benefits for Final Emission Guidelines Mass-based Illustrative Plan
Approach in 2030*
PM2.s-related Health Effects
Avoided Premature Mortality
Krewski et al (2009) (adult)
Lepeule et al. (2012) (adult)
Woodruff et al. (1997) (infant)
1,200
2,600
2
Avoided Morbidity
Emergency department visits for asthma (all ages)
Acute bronchitis (age 8-12)
Lower respiratory symptoms (age 7-14)
Upper respiratory symptoms (asthmatics age 9-1 1)
Minor restricted-activity days (age 18-65)
Lost work days (age 18-65)
Asthma exacerbation (age 6-18)
Hospital admissions respiratory (all ages)
Hospital admissions cardiovascular (age > 18)
Non-Fatal Heart Attacks (age >18)
Peters et al. (2001)
Pooled estimate of 4 studies
440
1,600
21,000
30,000
790,000
130,000
74,000
360
430
1,400
150
Ozone-related Health Effects
Avoided Premature Mortality
Bell et al. (2004) (all ages)
Levy et al. (2005) (all ages)
70
320
Avoided Morbidity
Hospital admissions respiratory causes (ages > 65)
Hospital admissions respiratory causes (ages < 2)
Emergency room visits for asthma (all ages)
Minor restricted-activity days (ages 18-65)
School absence days
470
200
210
380,000
130,000
* All estimates are rounded to whole numbers with two significant figures. Co-benefits for PMi.5precursors are
based on regional incidence-per-ton estimates for all precursors. Co-benefits for ozone are based on ozone season
NOx emissions. Confidence intervals are unavailable for this analysis because of the incidence-per-ton
methodology. In general, the 95th percentile confidence interval for the health impact function alone ranges from
approximately +30 percent for mortality incidence based on Krewski et al (2009) and +46 percent based on Lepeule
et al. (2012).
4-34
-------
PM2.5
Ozone
Rate-based
3%
7%
$25,000
$20,000
jj $15,000 x
§
£
I $10,000 x
; - Krewski
$5,000
$0
mi
Benefits estimates derived from 2 epidemiology and 12 expert functions
Rate-based
Levy etal. (2005)
Benefits estimates derived from 6 epidemiology functions
Mass-based
Mass-based
$25,000
$20,000
Benefits estimates derived from 2 epidemiology and 12 expert functions
Benefits estimates derived from 6 epidemiology functions
Figure 4-1. Monetized Health Co-benefits of Rate-based and Mass-based Illustrative
Plan Approaches for the Final Emission Guidelines in 2025 *
*The PM2.s graphs show the estimated PM2.s co-benefits at discount rates of 3% and 7% using effect coefficients
derived from the Krewski et al (2009) study and the Lepeule et al (2012) study, as well as 12 effect coefficients
derived from EPA's expert elicitation on PM mortality (Roman et al., 2008). The results shown are not the direct
results from the studies or expert elicitation; rather, the estimates are based in part on the concentration-response
functions provided in those studies. The ozone graphs show the estimated ozone co-benefits derived from six ozone
mortality studies (i.e., Bell et al (2004), Schwartz (2005), Huang et al (2005), Bell et al (2005), Levy et al (2005),
and Ito et al (2005). Ozone co-benefits occur in the analysis year, so they are the same for all discount rates. These
estimates do not include benefits from reductions in CO2. The monetized co-benefits do not include climate benefits
from changes in NO2 and SO2or reduced health effects from direct exposure to NO2, SO2, ecosystem effects, or
visibility impairment. For this RIA, we did not estimate changes in emissions of directly emitted particles. As a
result, quantified PM2.s related benefits are underestimated by a relatively small amount. In the proposal RIA, the
benefits from reductions in directly emitted PM2.s were less than 10 percent of total monetized health co-benefits
across all scenarios and years.
4-35
-------
Low Health Co-benefits
High Health Co-benefits
Rate-based
Mass-based
Rate-based
Mass-based
Figure 4-2. Breakdown of Monetized Health Co-benefits by Precursor Pollutant at a 3%
Discount Rate for Rate-based and Mass-based Illustrative Plan Approaches for
the Final Emission Guidelines in 2025*
* "Low Health Co-benefits" refers to the combined health co-benefits estimated using the Bell et al. (2004)
mortality study for ozone with the Krewski et al. (2009) mortality study for PMi.s. "High Health Co-benefits" refers
to the combined health co-benefits estimated using the Levy et al. (2005) mortality study for ozone with the Lepeule
et al. (2012) mortality study for PMi.s. For this RIA, we did not estimate changes in emissions of directly emitted
particles. As a result, quantified PM2.s related benefits are underestimated by a relatively small amount. In the
proposal RIA, the benefits from reductions in directly emitted PM2.s were less than 10 percent of total monetized
health co-benefits across all scenarios and years.
4.3.6 Characterization of Uncertainty in the Estimated Health Co-benefits
In any complex analysis using estimated parameters and inputs from numerous models,
there are likely to be many sources of uncertainty. This analysis is no exception. This analysis
4-36
-------
includes many data sources as inputs, including emission inventories, air quality data from
models (with their associated parameters and inputs), population data, population estimates,
health effect estimates from epidemiology studies, economic data for monetizing co-benefits,
and assumptions regarding the future state of the world (i.e., regulations, technology, and human
behavior). Each of these inputs may be uncertain and would affect the estimate of co-benefits.
When the uncertainties from each stage of the analysis are compounded, even small uncertainties
can have large effects on the total quantified benefits. In addition, the use of the benefit-per-ton
approach adds additional uncertainties beyond those for analyses based directly on air quality
modeling. Therefore, the estimates of co-benefits in each analysis year should be viewed as
representative of the general magnitude of co-benefits of the illustrative plan approach, rather
than the actual co-benefits anticipated from implementing the final emission guidelines.
This RIA does not include the type of detailed uncertainty assessment found in the PM
NAAQS RIA (U.S. EPA, 2012a) or the Ozone NAAQS RIA (U.S. EPA, 2008b) because we lack
the necessary air quality modeling input and/or monitoring data to run the benefits model.
However, the results of the quantitative and qualitative uncertainty analyses presented in the PM
NAAQS RIA and Ozone NAAQS RIA can provide some information regarding the uncertainty
inherent in the estimated co-benefits results presented in this analysis. For example, sensitivity
analyses conducted for the PM NAAQS RIA indicate that alternate cessation lag assumptions
could change the estimated PMi.s-related mortality co-benefits discounted at 3 percent by
between 10 percent and -27 percent and that alternative income growth adjustments could
change the PMi.s-related mortality co-benefits by between 33 percent and -14 percent. Although
we generally do not calculate confidence intervals for benefit-per-ton estimates and they can
provide an incomplete picture about the overall uncertainty in the benefits estimates, the PM
NAAQS RIA provides an indication of the random sampling error in the health impact and
economic valuation functions using Monte Carlo methods. In general, the 95th percentile
confidence interval for monetized PMi5 benefits ranges from approximately -90 percent to +180
percent of the central estimates based on Krewski et al. (2009) and Lepeule et al. (2012). The
95th percentile confidence interval for the health impact function alone ranges from approximately
±30 percent for mortality incidence based on Krewski et al. (2009) and ±46 percent based on Lepeule
et al. (2012).
4-37
-------
Unlike RIAs for which the EPA conducts scenario-specific air quality modeling, we do not
have information on the specific location of the air quality changes associated with the final
emission guidelines. As such, it is not feasible to estimate the proportion of co-benefits occurring
in different locations, such as designated nonattainment areas. Instead, we applied benefit-per-
ton estimates, which reflect specific geographic patterns of emissions reductions and specific air
quality and benefits modeling assumptions. For example, these estimates may not reflect local
variability in population density, meteorology, exposure, baseline health incidence rates, or other
local factors that might lead to an over-estimate or under-estimate of the actual co-benefits of
controlling PM and ozone precursors. Use of these benefit-per-ton values to estimate co-benefits
may lead to higher or lower benefit estimates than if co-benefits were calculated based on direct
air quality modeling. Great care should be taken in applying these estimates to emission
reductions occurring in any specific location, as these are all based on a broad emission reduction
scenario and therefore represent average benefits-per-ton over the entire region. The benefit-per-
ton for emission reductions in specific locations may be very different than the estimates
presented here. To the extent that the geographic distribution of the emissions reductions
achieved by implementing the final emission guidelines is different than the emissions in the air
quality modeling of the proposal, the co-benefits may be underestimated or overestimated.
Our estimate of the total monetized co-benefits is based on the 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
PM2.5-related premature mortality, which accounts for 98 percent of the monetized PMi.5 health
co-benefits.
1. We assume that all fine particles, regardless of their chemical composition, are equally
potent in causing premature mortality. This is an important assumption, because PMi.5
varies considerably in composition across sources, but the scientific evidence is not yet
sufficient to allow differentiation of effect estimates by particle type. The PM ISA
concluded that "many constituents of PlVb.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, 2009b).
2. We assume that the health impact function for fine particles is log-linear without a
threshold. Thus, the estimates include health co-benefits from reducing fine particles in
areas with varied concentrations of PM2.5, including both areas that do not meet the fine
4-38
-------
particle standard and those areas that are in attainment, down to the lowest modeled
concentrations.
3. 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, 2004c), which affects the valuation of mortality co-benefits at different discount
rates.
In general, we are more confident in the magnitude of the risks we estimate from simulated
PMi.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 PlVb.5 concentrations that fall below the bulk of the observed
data in these studies. Concentration benchmark analyses (e.g., lowest measured level [LML], one
standard deviation below the mean of the air quality data in the study, etc.) allow readers to
determine the portion of population exposed to annual mean PlVb.5 levels at or above different
concentrations, which provides some insight into the level of uncertainty in the estimated PMi.5
mortality benefits. In this analysis, we apply two concentration benchmark approaches (LML and
one standard deviation below the mean) that have been incorporated into recent RIAs and the
EPA's Policy Assessment for Paniculate Matter (U.S. EPA, 201 Id). There are uncertainties
inherent in identifying any particular point at which our confidence in reported associations
becomes appreciably less, and the scientific evidence provides no clear dividing line. However,
the EPA does not view these concentration benchmarks as a concentration threshold below
which we would not quantify health co-benefits of air quality improvements.109 Rather, the co-
benefits estimates reported in this RIA are the best estimates because they reflect the full range
of air quality concentrations associated with the emission reduction strategies. The PM ISA
concluded that the scientific evidence collectively is sufficient to conclude that the relationship
between long-term PMi.5 exposures and mortality is causal and that overall the studies support
109 por a summary of the scientific review statements regarding the lack of a threshold in the PMi.s-mortality
relationship, see the TSD entitled Summary of Expert Opinions on the Existence of a Threshold in the
Concentration-Response Function for PM2.s-related Mortality (U.S. EPA, 2010b).
4-39
-------
the use of a no-threshold log-linear model to estimate PM-related long-term mortality (U.S. EPA,
2009b).
For this analysis, policy-specific air quality data is not available, and the plan scenarios are
illustrative of what states may choose to do. However, we believe that it is still important to
characterize the distribution of exposure to baseline concentrations. As a surrogate measure of
mortality impacts, we provide the percentage of the population exposed at each PM2.5
concentration in the baseline of the air quality modeling used to calculate the benefit-per-ton
estimates for this final RIA using 12 km grid cells across the contiguous U.S. It is important to
note that baseline exposure is only one parameter in the health impact function, along with
baseline incidence rates population and change in air quality. In other words, the percentage of
the population exposed to air pollution below the LML is not the same as the percentage of the
population experiencing health impacts as a result of a specific emission reduction policy. The
most important aspect, which we are unable to quantify without rule-specific air quality
modeling, is the shift in exposure anticipated by implementing the final emission guidelines.
Therefore, caution is warranted when interpreting the LML assessment in this RIA because these
results are not consistent with results from RIAs that had air quality modeling.
Table 4-28 provides the percentage of the population exposed above and below two
concentration benchmarks (i.e., LML and one standard deviation below the mean) in the Clean
Power Plan proposal modeling. Figure 4-3 shows a bar chart of the percentage of the population
exposed to various air quality levels in the proposal modeling, and Figure 4-4 shows a
cumulative distribution function of the same data. Both figures identify the LML for each of the
major cohort studies.
4-40
-------
Table 4-28. Population Exposure in the Clean Power Plan Proposal Option 1 State
Scenario Modeling (used to generate the benefit-per-ton estimates) Above and
Below Various Concentrations Benchmarks in the Underlying Epidemiology
Studies*
Epidemiology Study
Below 1 Standard
Deviation.
Below AQ Mean
At or Above 1
Standard Deviation
Below AQ Mean
Below LML
At or Above LML
Krewski et al. (2009)
Lepeule ef a/. (2012)
3%
N/A
97%
N/A
12%
54%
88%
46%
*One standard deviation below the mean is equivalent to the middle of the range between the 10th and 25th
percentile. For Krewski, the LML is 5.8 Lig/m3 and one standard deviation below the mean is 11.0 Lig/m3. For
Lepeule et al, the LML is 8 Lig/m3 and we do not have the data for one standard deviation below the mean. It is
important to emphasize that although we have lower levels of confidence in levels below the LML for each study,
the scientific evidence does not support the existence of a level below which health effects from exposure to PM2.s
do not occur.
LM L of Krewski et
al. (2009) study
c
qj
0 20/6
£
C
o
re
3
s. 15/0
s
QJ
OJ
S.
_ 1
LML of Lepeule etal.
(2012) study
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Base ine Annual Mean PM25 Level (ug/m3)
Among the populations exposed to PM2.5 in the baseline:
88% are exposed to PM2.s levels at or above the LML of the Krewski et al. (2009) study
46% are exposed to PM2.s levels at or above the LML of the Lepeule et al. (2012) study
Figure 4-3. Percentage of Adult Population (age 30+) by Annual Mean PMi.s Exposure in
the Option 1 State Scenario Clean Power Plan Proposal Modeling (used to
generate the benefit-per-ton estimates)
4-41
-------
7 8 9 10 11 12 13 14 15 16 17 18
Baseline Annual Mean PM2 5 Level (ug/m3)
Among the populations exposed to PM2.5 in the baseline:
88% are exposed to PM2.5 levels at or above the LML of the Krewski et al (2009) study
46% are exposed to PM2.s levels at or above the LML of the Lepeule et al. (2012) study
Figure 4-4. Cumulative Distribution of Adult Population (age 30+) by Annual Mean
PMi.s Exposure in the Option 1 State Scenario Clean Power Plan Proposal
Modeling (used to generate the benefit-per-ton estimates)
4.4 Combined Climate Benefits and Health Co-Benefits Estimates
In this analysis, we were able to monetize the estimated benefits associated with the
decreased emissions of COi and co-benefits of reduced exposure to PMi.5 and ozone, but we
were unable to monetize the co-benefits associated with reducing exposure to mercury, carbon
monoxide, SOi, and NOi, as well as ecosystem effects and visibility impairment. In addition,
there are expected to be unquantified health and welfare impacts associated with changes in
hydrogen chloride. Specifically, we estimated combinations of climate benefits at discount rates
4-42
-------
of 5 percent, 3 percent, 2.5 percent, and 3 percent (95th percentile) (as recommended by the
interagency working group), and health co-benefits at discount rates of 3 percent and 7 percent
(as recommended by the EPA's Guidelines for Preparing Economic Analyses [U.S. EPA, 2014]
and OMB's Circular A-4 [OMB, 2003]).
Different discount rates are applied to SC-CCh than to the health co-benefit estimates
because CO2 emissions are long-lived and subsequent damages occur over many years.
Moreover, several rates are applied to SC-COi because the literature shows that it is sensitive to
assumptions about discount rate and because no consensus exists on the appropriate rate to use in
an intergenerational context. The SC-COi interagency group centered its attention on the 3
percent discount rate but emphasized the importance of considering all four SC-COi estimates.110
The EPA has evaluated the range of potential impacts by combining all SC-COi values with
health co-benefits values at the 3 percent and 7 percent discount rates. Combining the 3 percent
SC-CO2 values with the 3 percent health benefit values assumes that there is no difference in
discount rates between intragenerational and intergenerational impacts.
Tables 4-29 through 4-31 provide the combined climate and health benefits for the
illustrative plan approaches evaluated for each analysis year: 2020, 2025, and 2030. Figure 4-5
shows the breakdown of the monetized benefits by pollutant for the illustrative plan approaches
evaluated in 2025 as an illustrative analysis year using a 3 percent discount rate for both climate
and health benefits.
110 See the 2010 SCC TSD. Docket ID EPA-HQ-OAR-2009-0472-114577 or
http://www.whitehouse.gov/sites/default/files/omb/inforeg/for-agencies/Social-Cost-of-Carbon-for-RIA.pdffor
details.
4-43
-------
Table 4-29. Combined Climate Benefits and Health Co-Benefits for Final Emission
Guidelines in 2020 (billions of 2011$)*
SCC Discount Rate
Rate-based
5%
3%
2.5%
3% (95th percentile)
Mass-based
5%
3%
2.5%
3% (95th percentile)
Climate and Health Benefits (Discount Rate Applied to
Climate Health Co-Benefits)
Benefits Only
69
$0.80
$2.8
$4.1
$8.2
82
$0.94
$3.3
$4.9
$9.6
million short tons CO2
$1.5 to $2.6
$3.5 to $4.6
$4.9 to $6.0
$8.9 to $10
million short tons CO2
$2.9 to $5.7
$5. 3 to $8.1
$6.9 to 9.7
$12 to $14
$1.4 to $2.5
$3.5 to $4.5
$4.8 to $5.9
$8.9 to $9.9
$2.8 to $5.3
$5.1 to $7.7
$6.7 to $9.3
$11 to $14
*A11 estimates are rounded to two significant figures. Climate benefits are based on reductions in CO2 emissions.
Co-benefits are based on regional benefit-per-ton estimates. Co-benefits for ozone are based on ozone season NOx
emissions. Ozone co-benefits occur in analysis year, so they are the same for all discount rates. The health co-
benefits reflect the sum of the PM2.s and ozone co-benefits and reflect the range based on adult mortality functions
(e.g., from Krewski et al (2009) with Bell et al (2004) to Lepeule et al. (2012) with Levy et al. (2005)). The
monetized health co-benefits do not include reduced health effects from directly emitted PM2.s, direct exposure to
NO2, SO2, and HAP; ecosystem effects; or visibility impairment.
Table 4-30. Combined Climate Benefits and Health Co-Benefits for Final Emission
Guidelines in 2025 (billions of 2011$)*
SCC Discount Rate
Rate-based
5%
3%
2.5%
3% (95th percentile)
Mass-based
5%
3%
2.5%
3% (95* percentile)
Climate and Health Benefits (Discount Rate Applied to
Climate Health Co-Benefits)
Benefits Only J% ?%
232
$3.1
$10
$15
$31
264
$3.6
$12
$17
$35
million short tons CO2
$11 to $21
$18 to $28
$23 to $33
$38 to $49
million short tons CO2
$11 to $21
$19 to $29
$24 to $34
$42 to $52
$9.9 to $19
$17 to $26
$22 to $31
$38 to $47
$10 to $19
$18 to $27
$24 to $33
$42 to $51
*A11 estimates are rounded to two significant figures. Climate benefits are based on reductions in CO2 emissions.
Co-benefits are based on regional benefit-per-ton estimates. Co-benefits for ozone are based on ozone season NOx
emissions. Ozone co-benefits occur in analysis year, so they are the same for all discount rates. The health co-
benefits reflect the sum of the PM2.s and ozone co-benefits and reflect the range based on adult mortality functions
(e.g., from Krewski et al (2009) with Bell et al (2004) to Lepeule et al. (2012) with Levy et al. (2005)). The
monetized health co-benefits do not include reduced health effects from directly emitted PM2.s, direct exposure to
NO2, SO2, and HAP; ecosystem effects; or visibility impairment.
4-44
-------
Table 4-31. Combined Climate Benefits and Health Co-Benefits for Final Emission
Guidelines in 2030 (billions of 2011$)*
SCC Discount Rate
Rate-based
5%
3%
2.5%
3% (95th percentile)
Mass-based
5%
3%
2.5%
3% (95th percentile)
Climate and Health Benefits (Discount Rate Applied to
Climate Health Co-Benefits)
Benefits Only J% ?%
415
$6.4
$20
$29
$61
413
$6.4
$20
$29
$60
million short tons CCh
$21 to $40
$34 to $54
$43 to $63
$75 to $95
million short tons CCh
$18 to $34
$32 to $48
$41 to $57
$72 to $89
$19 to $37
$33 to $51
$42 to $60
$74 to $92
$17 to $32
$31 to $46
$40 to $55
$71 to $86
*A11 estimates are rounded to two significant figures. Climate benefits are based on reductions in COi emissions.
Co-benefits are based on regional benefit-per-ton estimates. Co-benefits for ozone are based on ozone season NOx
emissions. Ozone co-benefits occur in analysis year, so they are the same for all discount rates. The health co-
benefits reflect the sum of the PM2.s and ozone co-benefits and reflect the range based on adult mortality functions
(e.g., from Krewski et al (2009) with Bell et al (2004) to Lepeule et al. (2012) with Levy et al. (2005)). The
monetized health co-benefits do not include reduced health effects from directly emitted PM2.s, direct exposure to
NOi, SOi, and HAP; ecosystem effects; or visibility impairment.
4-45
-------
Low Health Co-benefits
High Health Co-benefits
Rate-based
Rate-based
NOx (as
PM2.5)
5%
NOx (as
Ozone)
7%
Mass-based
Mass-based
NOx (as
Ozone)
3%
NOx (as
PM2.5)
Figure 4-5. Breakdown of Combined Monetized Climate and Health Co-benefits of Final
Emission Guidelines in 2025 for Rate-based and Mass-based Illustrative Plan
Approaches and Pollutants (3% discount rate)*
* "Low Health Co-benefits" refers to the combined health co-benefits estimated using the Bell et al. (2004)
mortality study for ozone with the Krewski et al. (2009) mortality study for PM2.5. "High Health Co-benefits" refers
to the combined health co-benefits estimated using the Levy et al. (2005) mortality study for ozone with the Lepeule
et al. (2012) mortality study for PM2.s. For this RIA, we did not estimate changes in emissions of directly emitted
particles. As a result, quantified PM2.s related benefits are underestimated by a relatively small amount. In the
proposal RIA, the benefits from reductions in directly emitted PM2.s were less than 8 percent of total monetized
benefits across all scenarios and years.
4.5 Unquantified Co-benefits
The monetized co-benefits estimated in this RIA reflect a subset of co-benefits attributable
to the health effect reductions associated with ambient fine particles and ozone. Data, time, and
4-46
-------
resource limitations prevented the EPA from quantifying the impacts to, or monetizing the co-
benefits from several important benefit categories, including co-benefits associated with
exposure to several HAP (including mercury), SOi and NO2, as well as ecosystem effects, and
visibility impairment due to the absence of air quality modeling data for these pollutants in this
analysis. This does not imply that there are no co-benefits associated with changes in emissions
of HAP or reductions in exposures to 862 and NOi. In this section, we provide a qualitative
description of these benefits, which are listed in Table 4-32.
Table 4-32. Unquantified Health and Welfare Co-benefits Categories
Category
Specific Effect
Effect Has Effect Has
Been Been More Information
Quantified Monetized
Improved Human Health
Reduced incidence of
morbidity from exposure
toNO2
Reduced incidence of
morbidity from exposure
to SO2
Reduced incidence of
morbidity from exposure
to CO
Reduced incidence of
morbidity from exposure
to methylmercury
Asthma hospital admissions (all ages)
Chronic lung disease hospital admissions (age >
65)
Respiratory emergency department visits (all
ages)
Asthma exacerbation (asthmatics age 4-18)
Acute respiratory symptoms (age 7-14)
Premature mortality
Other respiratory effects (e.g., airway
hyperresponsiveness and inflammation, lung
function, other ages and populations)
Respiratory hospital admissions (age > 65)
Asthma emergency department visits (all ages)
Asthma exacerbation (asthmatics age 4-12)
Acute respiratory symptoms (age 7-14)
Premature mortality
Other respiratory effects (e.g., airway
hyperresponsiveness and inflammation, lung
function, other ages and populations)
Cardiovascular effects
Respiratory effects
Central nervous system effects
Premature mortality
Neurologic effects IQ loss
Other neurologic effects (e.g., developmental
delays, memory, behavior)
Cardiovascular effects
Genotoxic, immunologic, and other toxic effects
_ _ N02 ISA1
_ _ N02 ISA1
_ _ NO2 ISA1
_ _ NO2 ISA1
_ _ NO2 ISA1
_ _ NO2 ISA1-2-3
_ _ NO2 ISA2-3
_ _ S02 ISA1
_ _ SO2 ISA1
_ _ SO2 ISA1
_ _ SO2 ISA1
_ _ S02 ISA1-2-3
_ _ S02 ISA1-2
_ _ CO ISA !-2
_ _ CO ISA i'2-3
_ _ CO ISA i'2-3
_ _ CO ISA i'2-3
IRIS; NRC,
20001
IRIS; NRC,
20002
IRIS; NRC,
~~ ~~ 20002-3
IRIS; NRC,
~~ ~~ 20002-3
Improved Environment
Reduced visibility
impairment
Visibility in Class 1 areas
Visibility in residential areas
_ _ PM ISA1
_ _ PM ISA1
4-47
-------
Category
materials
Reduced effects from PM
deposition (metals and
organics)
ecosystem effects from
exposure to ozone
Reduced effects from
acid deposition
Reduced effects from
nutrient enrichment
Reduced vegetation
effects from ambient
exposure to SCh and NOX
Reduced ecosystem
methylmercury
Effect Has Effect Has
Specific Effect Been Been
Quantified Monetized
Household soiling
Materials damage (e.g., corrosion, increased
wear)
Effects on Individual organisms and ecosystems
Visible foliar injury on vegetation
Reduced vegetation growth and reproduction
Yield and quality of commercial forest products
and crops
Damage to urban ornamental plants
Carbon sequestration in terrestrial ecosystems
Recreational demand associated with forest
aesthetics
Other non-use effects
Ecosystem functions (e.g., water cycling,
biogeochemical cycles, net primary productivity,
leaf-gas exchange, community composition)
Recreational fishing
Tree mortality and decline
Commercial fishing and forestry effects
Recreational demand in terrestrial and aquatic
ecosystems
Other non-use effects
Ecosystem functions (e.g., biogeochemical
cycles)
Species composition and biodiversity in terrestrial
and estuarine ecosystems
Coastal eutrophication
Recreational demand in terrestrial and estuarine
ecosystems
Other non-use effects
Ecosystem functions (e.g., biogeochemical
cycles, fire regulation)
Injury to vegetation from SO2 exposure
Injury to vegetation from NOX exposure
Effects on fish, birds, and mammals (e.g.,
reproductive effects)
Commercial, subsistence and recreational fishing
More Information
PM ISA1-2
PM ISA2
PM ISA2
Ozone ISA1
Ozone ISA1
Ozone ISA1
Ozone ISA2
Ozone ISA1
Ozone ISA2
Ozone ISA2
Ozone ISA2
NOx SOx ISA1
NOx SOx ISA2
NOx SOx ISA2
NOx SOx ISA2
NOx SOx ISA2
NOx SOx ISA2
NOx SOx ISA2
NOx SOx ISA2
NOx SOx ISA2
NOx SOx ISA2
NOx SOx ISA2
NOx SOx ISA2
NOx SOx ISA2
Mercury Study
RTC2
Mercury Study
RTC1
1 We assess these co-benefits qualitatively due to data and resource limitations for this RIA.
2We assess these co-benefits qualitatively because we do not have sufficient confidence in available data or methods.
3 We assess these co-benefits qualitatively because current evidence is only suggestive of causality or there are other significant
concerns over the strength of the association.
4.5.1 HAP Impacts
Due to methodology and resource limitations, we were unable to estimate the impacts
associated with changes in emissions of the hazardous air pollutants in this analysis. The EPA's
SAB-HES concluded that "the challenges for assessing progress in health improvement as a
4-48
-------
result of reductions in emissions of hazardous air pollutants (HAPs) are daunting...due to a lack
of exposure-response functions, uncertainties in emissions inventories and background levels, the
difficulty of extrapolating risk estimates to low doses and the challenges of tracking health
progress for diseases, such as cancer, that have long latency periods" (U.S. EPA-SAB, 2008). In
2009, the EPA convened a workshop to address the inherent complexities, limitations, and
uncertainties in current methods to quantify the benefits of reducing HAP. Recommendations
from this workshop included identifying research priorities, focusing on susceptible and
vulnerable populations, and improving dose-response relationships (Gwinn etal, 2011).
4.5.1.1 Mercury
Mercury in the environment is transformed into a more toxic form, methylmercury
(MeHg). Because Hg is a persistent pollutant, MeHg accumulates in the food chain, especially
the tissue of fish. When people consume these fish, they consume MeHg. In 2000, the NAS
Study was issued which provides a thorough review of the effects of MeHg on human health
(NRC, 2000).m Many of the peer-reviewed articles cited in this section are publications
originally cited in the Mercury Study.112 In addition, the EPA has conducted literature searches
to obtain other related and more recent publications to complement the material summarized by
the NRC in 2000.
In its review of the literature, the NAS found neurodevelopmental effects to be the most
sensitive and best documented endpoints and appropriate for establishing a reference dose (RfD)
(NRC, 2000); in particular NAS supported the use of results from neurobehavioral or
neuropsychological tests. The NAS report noted that studies on animals reported sensory effects
as well as effects on brain development and memory functions and supported the conclusions
based on epidemiology studies. The NAS noted that their recommended endpoints for a RfD are
associated with the ability of children to learn and to succeed in school. They concluded the
following: "The population at highest risk is the children of women who consumed large
amounts of fish and seafood during pregnancy. The committee concludes that the risk to that
111 National Research Council (NRC). 2000. Toxicological Effects of Methylmercury. Washington, DC: National
Academies Press.
112 U.S. Environmental Protection Agency (U.S. EPA). 1997. Mercury Study Report to Congress, EPA-HQ-OAR-
2009-0234-3054. December. Available on the Internet at .
4-49
-------
population is likely to be sufficient to result in an increase in the number of children who have to
struggle to keep up in school."
The NAS summarized data on cardiovascular effects available up to 2000. Based on these
and other studies, the NRC concluded that "Although the data base is not as extensive for
cardiovascular effects as it is for other end points (i.e., neurologic effects), the cardiovascular
system appears to be a target for MeHg toxicity in humans and animals." The NRC also stated
that "additional studies are needed to better characterize the effect of methylmercury exposure on
blood pressure and cardiovascular function at various stages of life."
Additional cardiovascular studies have been published since 2000. The EPA did not
develop a quantitative dose-response assessment for cardiovascular effects associated with
MeHg exposures, as there is no consensus among scientists on the dose-response functions for
these effects. In addition, there is inconsistency among available studies as to the association
between MeHg exposure and various cardiovascular system effects. The pharmacokinetics of
some of the exposure measures (such as toenail Hg levels) are not well understood. The studies
have not yet received the review and scrutiny of the more well-established neurotoxicity data
base.
The Mercury Study noted that MeHg is not a potent mutagen but is capable of causing
chromosomal damage in a number of experimental systems. The NAS concluded that evidence
that human exposure to MeHg caused genetic damage is inconclusive; they note that some earlier
studies showing chromosomal damage in lymphocytes may not have controlled sufficiently for
potential confounders. One study of adults living in the Tapajos River region in Brazil (Amorim
et al., 2000) reported a direct relationship between MeHg concentration in hair and DNA damage
in lymphocytes, as well as effects on chromosomes.113 Long-term MeHg exposures in this
population were believed to occur through consumption of fish, suggesting that genotoxic effects
113 Amorim, M.I.M., D. Mergler, M.O. Bahia, H. Dubeau, D. Miranda, J. Lebel, R.R. Burbano, and M. Lucotte.
2000. Cytogenetic damage related to low levels of methyl mercury contamination in the Brazilian Amazon. An.
Acad. Bras. Cienc. 72(4): 497-507.
4-50
-------
(largely chromosomal aberrations) may result from dietary and chronic MeHg exposures similar
to and above those seen in the Faroes and Seychelles populations.
Although exposure to some forms of Hg can result in a decrease in immune activity or an
autoimmune response (ATSDR, 1999), evidence for immunotoxic effects of MeHg is limited
(NRC, 2000).114
Based on limited human and animal data, MeHg is classified as a "possible" human
carcinogen by the International Agency for Research on Cancer (IARC, 1994)115 and in IRIS
(U.S. EPA, 2002).116 The existing evidence supporting the possibility of carcinogenic effects in
humans from low-dose chronic exposures is tenuous. Multiple human epidemiological studies
have found no significant association between Hg exposure and overall cancer incidence,
although a few studies have shown an association between Hg exposure and specific types of
cancer incidence (e.g., acute leukemia and liver cancer) (NRC, 2000).
There is also some evidence of reproductive and renal toxicity in humans from MeHg
exposure. However, overall, human data regarding reproductive, renal, and hematological
toxicity from MeHg are very limited and are based on either studies of the two high-dose
poisoning episodes in Iraq and Japan or animal data, rather than epidemiological studies of
chronic exposures at the levels of interest in this analysis.
4.5.1.2 Hydrogen Chloride
Hydrogen chloride (HC1) is a corrosive gas that can cause irritation of the mucous
membranes of the nose, throat, and respiratory tract. Brief exposure to 35 ppm causes throat
114 Agency for Toxic Substances and Disease Registry (ATSDR). 1999. Toxicological Profile for Mercury. U.S.
Department of Health and Human Services, Public Health Service, Atlanta, GA.
115 International Agency for Research on Cancer (IARC). 1994. IARC Monographs on the Evaluation of
Carcinogenic Risks to Humans and their Supplements: Beryllium, Cadmium, Mercury, and Exposures in the Glass
Manufacturing Industry. Vol. 58. Jalili, H.A., and A.H. Abbasi. 1961. Poisoning by ethyl mercury toluene
sulphonanilide. Br. J. Indust. Med. 18(Oct.):303-308 (as cited in NRC, 2000).
116 U.S. Environmental Protection Agency (EPA). 2002. Integrated Risk Information System (IRIS) on
Methylmercury. National Center for Environmental Assessment. Office of Research and Development. Available at
http://www.epa.gov/iris/subst/0073.htm.
4-51
-------
irritation, and levels of 50 to 100 ppm are barely tolerable for 1 hour.117 Concentrations in typical
human exposure environments are much lower than these levels and rarely exceed the reference
concentration. 118The greatest impact is on the upper respiratory tract; exposure to high
concentrations can rapidly lead to swelling and spasm of the throat and suffocation. Most
seriously exposed persons have immediate onset of rapid breathing, blue coloring of the skin,
and narrowing of the bronchioles. Exposure to HC1 can lead to Reactive Airways Dysfunction
Syndrome (RADS), a chemically, or irritant-induced type of asthma. Children may be more
vulnerable to corrosive agents than adults because of the relatively smaller diameter of their
airways. Children may also be more vulnerable to gas exposure because of increased minute
ventilation per kg and failure to evacuate an area promptly when exposed. Hydrogen chloride has
not been classified for carcinogenic effects.119
4.5.2 Additional NO2 Health Co-Benefits
In addition to being a precursor to PMi.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 co-benefits associated with reduced NOi exposure in this analysis. Therefore, this analysis
only quantified and monetized the PM2.5 and ozone co-benefits associated with the reductions in
NOi emissions.
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, 2008c) 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
117 Agency for Toxic Substances and Disease Registry (ATSDR). Medical Management Guidelines for Hydrogen
Chloride. Atlanta, GA: U.S. Department of Health and Human Services. Available at
http://www.atsdr.cdc.gov/mmg/mmg. asp?id=758&tid=147#bookmark02.
118 Table of Prioritized Chronic Dose-Response Values: http://www2.epa.gov/sites/production/files/2014-
05/documents/table 1 .pdf
119 U.S. Environmental Protection Agency (U.S. EPA). 1995. "Integrated Risk Information System File of Hydrogen
Chloride." Washington, DC: Research and Development, National Center for Environmental Assessment. This
material is available at http://www.epa.gov/iris/subst/0396.htm.
4-52
-------
NOx ISA also concluded that the relationship between short-term NOi 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 NOi 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.
4.5.3 Additional SO2 Health Co-Benefits
In addition to being a precursor to PMi.5, 862 emissions are also linked to a variety of
adverse health effects associated with direct exposure. We were unable to estimate the health co-
benefits associated with reduced SOi in this analysis because we do not have air quality
modeling data available. Therefore, this analysis only quantifies and monetizes the PM2.5 co-
benefits associated with the reductions in 862 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 862 (U.S. EPA, 2008a). The immediate effect of 862 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 relationship has been demonstrated in laboratory studies following exposures to 862 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 862 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 862 ISA. The 862 ISA also
concluded that the relationship between short-term 862 exposure and premature mortality was
"suggestive of a causal relationship" because it is difficult to attribute the mortality risk effects to
862 alone. Although the 862 ISA stated that studies are generally consistent in reporting a
relationship between 862 exposure and mortality, there was a lack of robustness of the observed
associations to adjustment for other pollutants. We did not quantify these co-benefits due to data
constraints.
4-53
-------
4.5.4 Additional NO2 and 862 Welfare Co-Benefits
As described in the Integrated Science Assessment for Oxides of Nitrogen and Sulfur
Ecological Criteria (NOx/SOx ISA) (U.S. EPA, 2008d), SO2 and NOx emissions also contribute
to a variety of adverse welfare effects, including those associated with acidic deposition,
visibility impairment, and nutrient enrichment. Deposition of nitrogen 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, 2008d)
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,
2008d)
4-54
-------
Reductions in emissions of NOi and SOi 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).
4.5.5 Ozone Welfare Co-Benefits
Exposure to ozone has been associated with a wide array of vegetation and ecosystem
effects in the published literature (U.S. EPA, 2013b). 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.
4.5.6 Carbon Monoxide Co-Benefits
CO in ambient air is formed primarily by the incomplete combustion of carbon-containing
fuels and photochemical reactions in the atmosphere. The amount of CO emitted from these
reactions, relative to carbon dioxide (COi), is sensitive to conditions in the combustion zone,
such as fuel oxygen content, burn temperature, or mixing time. Upon inhalation, CO diffuses
through the respiratory system to the blood, which can cause hypoxia (reduced oxygen
availability). Carbon monoxide can elicit a broad range of effects in multiple tissues and organ
systems that depend on concentration and duration of exposure. The Integrated Science
Assessment for Carbon Monoxide (U.S. EPA, 2010a) concluded that short-term exposure to CO
is "likely to have a causal relationship" with cardiovascular morbidity, particularly in individuals
with coronary heart disease. Epidemiologic studies associate short-term CO exposure with
increased risk of emergency department visits and hospital admissions. Coronary heart disease
includes those who have angina pectoris (cardiac chest pain), as well as those who have
4-55
-------
experienced a heart attack. Other subpopulations potentially at risk include individuals with
diseases such as chronic obstructive pulmonary disease (COPD), anemia, or diabetes, and
individuals in very early or late life stages, such as older adults or the developing young. The
evidence is suggestive of a causal relationship between short-term exposure to CO and
respiratory morbidity and mortality. The evidence is also suggestive of a causal relationship for
birth outcomes and developmental effects following long-term exposure to CO, and for central
nervous system effects linked to short- and long-term exposure to CO.
4.5.7 Visibility Impairment Co-Benefits
Reducing secondary formation of PM2.5 would improve levels visibility in the U.S. because
suspended particles and gases degrade visibility by scattering and absorbing light (U.S. EPA,
2009b). Fine particles with significant light-extinction efficiencies include sulfates, nitrates,
organic carbon, elemental carbon, and soil (Sisler, 1996). Visibility has direct significance to
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, 2009b). Previous analyses (U.S.
EPA, 201 la) show that visibility co-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.
4.6 References
40 CFR Chapter I [EPA-HQ-OAR-2009-0171; FRL-9091-8] RIN 2060-ZA14,
"Endangerment and Cause or Contribute Findings for Greenhouse Gases under Section 202(a)
of the Clean Air Act," Federal Register, Vol. 74, No. 239, Tuesday, December 15, 2009,
Rules and Regulations.
4-56
-------
Abt Associates, Inc. 2012. "BenMAP User's Manual Appendices," prepared for U.S. Research
Triangle Park, NC: U. S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Available at:
. Accessed June
6,2015.
Anenberg SC, West IJ, Fiore AM, Jaffe DA, Prather MJ, Bergmann D, Cuvelier K, Dentener FJ,
Duncan BN, Gauss M, Hess P, Jonson JE, Lupu A, Mackenzie IA, Manner E, Park RJ,
Sanderson MG, Schultz M, Shindell DT, Szopa S, Vivanco MG, Wild O, Zeng G. 2009.
Intercontinental impacts of ozone pollution on human mortality. Environmental Science and
Technology. 43(17): 6482-7.
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.
Docket ID EPA-HQ-OAR-2009-0472-114577, Technical Support Document: Social Cost of
Carbon for Regulatory Impact Analysis Under Executive Order 12866, Interagency Working
Group on Social Cost of Carbon, with participation by the Council of Economic Advisers,
Council on Environmental Quality, Department of Agriculture, Department of Commerce,
Department of Energy, Department of Transportation, Environmental Protection Agency,
National Economic Council, Office of Energy and Climate Change, Office of Management
and Budget, Office of Science and Technology Policy, and Department of Treasury (February
2010). Available at: .
4-57
-------
Docket ID EPA-HQ-OAR-2013-0602, Technical Support Document: Technical Update of the
Social Cost of Carbon for Regulatory Impact Analysis Under Executive Order 12866,
Interagency Working Group on Social Cost of Carbon, with Participation by Council of
Economic Advisers, Council on Environmental Quality, Department of Agriculture,
Department of Commerce, Department of Energy, Department of Transportation, Domestic
Policy Council, Environmental Protection Agency, National Economic Council, Office of
Management and Budget, Office of Science and Technology Policy, and Department of
Treasury (May 2013, Revised July 2015). Also available at: <
https://www.whitehouse.gov/sites/default/files/omb/inforeg/scc-tsd-final-july-2015.pdf>.
Accessed July 15, 2015.
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.
Huang Y., F. Dominici, and M. Bell. 2005. "Bayesian Hierarchical Distributed Lag Models for
Summer Ozone Exposure and Cardio-Respiratory Mortality." Environmetrics. 16:547-562.
Industrial Economics, Incorporated (lEc). 2006. Expanded Expert Judgment Assessment of the
Concentration-Response Relationship Between PM2.5 Exposure and Mortality. Prepared for:
Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency,
Research Triangle Park, NC. September. Available at:
. Accessed June 6, 2015.
4-58
-------
Interagency Working Group on Social Cost of Carbon, with participation by Council of
Economic Advisers, Council on Environmental Quality, Department of Agriculture,
Department of Commerce, Department of Energy, Department of Transportation,
Environmental Protection Agency, National Economic Council, Office of Management and
Budget, Office of Science and Technology Policy, and Department of Treasury. Response to
to Comments: Social Cost of Carbon for Regulatory Impact Analysis Under Executive Order
12866. July 2015. Available at
Accessed July 15, 2015.
Intergovernmental Panel on Climate Change (JPCC). 2007. Climate Change 2007: Synthesis
Report Contribution of Working Groups I, II and III to the Fourth Assessment Report of the
IPCC. Available at:
. Accessed June 6, 2015.
Intergovernmental Panel on Climate Change (IPCC). 2012: Managing the Risks of Extreme
Events and Disasters to Advance Climate Change Adaptation. A Special Report of Working
Groups I and II of the Intergovernmental Panel on Climate Change. Cambridge University
Press, Cambridge, UK, and New York, NY, USA.
Intergovernmental Panel on Climate Change (IPCC). 2013. Climate Change 2013: The Physical
Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA.
Intergovernmental Panel on Climate Change (IPCC). 2014a. Climate Change 2014: Impacts,
Adaptation, and Vulnerability. Contribution of Working Group II to the Fifth Assessment
Report of the Intergovernmental Panel on Climate Change. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA.
Intergovernmental Panel on Climate Change (IPCC). 2014b. Climate Change 2014: Mitigation
of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA.
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.
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.
4-59
-------
Karl, T., J. Melillo, and T. Peterson, Eds. 2009. Global Climate Change Impacts in the United
States. Cambridge, United Kingdom: Cambridge University Press.
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 Paniculate 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.
Melillo, Jerry M., Terese (T.C.) Richmond, and Gary W. Yohe, Eds. 2014. Climate Change
Impacts in the United States: The Third National Climate Assessment. U.S. Global Change
Research Program. Available at
National Research Council (NRC). 2000. Toxicological Effects ofMethylmercury: Committee on
the Toxicological Effects ofMethylmercury." Board on Environmental Studies and
Toxicology. National Academies Press. Washington, DC.
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.
National Research Council (NRC). 2010. Ocean Acidification: A National Strategy to Meet the
Challenges of a Changing Ocean. National Academies Press. Washington, DC.
National Research Council (NRC). 201 la. Climate Stabilization Targets: Emissions,
Concentrations, and Impacts over Decades to Millennia. National Academies Press,
Washington, DC.
National Research Council (NRC) 20 lib. National Security Implications of Climate Change for
U.S. Naval Forces. National Academies Press. Washington, DC.
National Research Council (NRC, 201 Ic). Understanding Earth's Deep Past: Lessons for Our
Climate Future. National Academies Press. Washington, DC
National Research Council (NRC). 2012a. Sea-Level Rise for the Coasts of California, Oregon,
and Washington: Past, Present, and Future. National Academies Press. Washington, DC.
4-60
-------
National Research Council (NRC). 2013a. Climate and Social Stress: Implications for Security
Analysis. National Academies Press. Washington, DC.
National Research Council (NRC). 2013b. Abrupt Impacts of Climate Change: Anticipating
Surprises. National Academies Press. Washington, DC.
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. William, L. Morawska, K. Mengensen, and S. Tong. 2008a. "Ozone Modifies
Associations between Temperature and Cardiovascular Mortality: Analysis of the NMMAPS
Data." Occupational and Environmental Medicine. 65:255-260.
Ren, C., G.M. Williams, K. Mengersen, L. Morawska, and S. Tong. 2008b. "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. and S. Tong. 2006b. "Temperature Modifies the Health Effects of Particulate Matter in
Brisbane, Australia." International Journal ofBiometeorology. 51:87-96.
Ren. C., G.M. Williams, and S. Tong. 2006a. "Does Particulate Matter Modify the Association
between Temperature and Cardiorespiratory Diseases? Environmental Health Perspectives,
114:1690-1696.
Roberts, S. 2004. "Interactions between Particulate Air Pollution and Temperature in Air
Pollution Mortality Time Series Studies." Environmental Research. 96:328-337.
Roman, H., K. D. Walker, T.L. Walsh, L. Conner, H. M. Richmond, BJ. Hubbell, and P.L.
Kinney. 2008. "Expert Judgment Assessment of the Mortality Impact of Changes in Ambient
Fine Particulate Matter in the U.S." Environmental Science and Technology. 42(7):2268-
2274.
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.
4-61
-------
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.
U.S. Environmental Protection Agency (U.S. EPA). 1995. Regulatory Impact Analysis for the
Petroleum Refinery NESHAP. Revised Draft for Promulgation. Office of Air Quality Planning
and Standards, Research Triangle Park, N.C. Available on the Internet at <
http://yosemitel.epa.gov/ee/epa/ria.nsf/vwTD/9F39F2C26150BB21852564620063317F>.
Accessed June 6, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2008a. Integrated Science Assessment for
Sulfur OxidesHealth Criteria (Final Report). National Center for Environmental
Assessment - RTP Division, Research Triangle Park, NC. September. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2008b. Final Ozone NAAQS Regulatory
Impact Analysis. EPA-452/R-08-003. Office of Air Quality Planning and Standards Health
and Environmental Impacts Division, Air Benefit and Cost Group Research Triangle Park,
NC. March. Available at: < http://www.epa.gov/ttnecasl/regdata/RIAs/6-
ozoneriachapter6.pdf>. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2008c. Integrated Science Assessment for
Oxides of Nitrogen - Health Criteria (Final Report). National Center for Environmental
Assessment, Research Triangle Park, NC. July. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2008d. 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:
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2009. Integrated Science Assessment for
Paniculate Matter (Final Report). EPA-600-R-08-139F. National Center for Environmental
Assessment - RTP Division, Research Triangle Park, NC. December. Available at:
. Accessed June 4, 2015.
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:
. Accessed June 4, 2015.
4-62
-------
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 PMi.s-related Mortality. Research Triangle Park, NC. June. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2010c. Valuing Mortality Risk Reductions
for Environmental Policy: A White Paper: SAB Review Draft. National Center for
Environmental Economics December. Available at:
.
Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2010d. Section 3: Re-analysis of the Benefits
of Attaining Alternative Ozone Standards to Incorporate Current Methods. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 201 la. The Benefits and Costs of the Clean
Air Act from 1990 to 2020. Office of Air and Radiation, Washington, DC. March. Available
at: . Accessed June 4,
2015.
U.S. Environmental Protection Agency (U.S. EPA). 201 Ib. Regulatory Impact Analysis for the
Final Mercury and Air Toxics Standards. EPA-452/R-11-011. December. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 201 Ic. Regulatory Impact Analysis:
National Emission Standards for Hazardous Air Pollutants for Industrial, Commercial, and
Institutional Boilers and Process Heaters. February. Available at:
. Accessed June
4,2015.
U.S. Environmental Protection Agency (U.S. EPA). 201 Id. Policy Assessment for the Review of
the Paniculate 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: .
Accessed June 4, 2015.
4-63
-------
U.S. Environmental Protection Agency (U.S. EPA). 2012a. Regulatory Impact Analysis for the
Final Revisions to the National Ambient Air Quality Standards for Paniculate Matter. EPA-
452/R-12-003. Office of Air Quality Planning and Standards, Health and Environmental
Impacts Division, Research Triangle Park, NC. December. Available at: <
http://www.epa.gov/ttnecasl/regdata/RIAs/finalria.pdf>. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2012b. Regulatory Impact Analysis:
Petroleum Refineries New Source Performance Standards Ja. Office of Air Quality Planning
and Standards, Health and Environmental Impacts Division. June. Available at:
. Accessed June
4,2015.
U.S. Environmental Protection Agency (U.S. EPA). 2013a. Technical Support Document:
Estimating the Benefit per Ton of Reducing PM2.5 Precursors from 17 Sectors. Office of Air
Quality Planning and Standards, Research Triangle Park, NC. February. Available at: <
http://www2.epa.gov/sites/production/files/2014-
10/documents/sourceapportionmentbpttsd.pdf >. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2013b. 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:
. Accessed June 4,
2015.
U.S. Environmental Protection Agency (U.S. EPA). 2014. Guidelines for Preparing Economic
Analyses. EPA 240-R-10-001. National Center for Environmental Economics, Office of the
Administrator. Washington, DC. Available at:
. Accessed June 6, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2014a. Regulatory Impact Analysis for the
Proposed Carbon Pollution Guidelines for Existing Power Plants and Emission Standards for
Modified and Reconstructed Power Plants. EPA-542/R-14-002. Office of Air Quality
Planning and Standards, Research Triangle Park, NC. June. Available at
. Accessed June
4,2015.
4-64
-------
U.S. Environmental Protection Agency (U.S. EPA). 2014b. Health Risk and Exposure
Assessment for Ozone: Final Report. Office of Air Quality Planning and Standards, Research
Triangle Park, NC. EPA-452/R-14-004a. Available at:
. Accessed June
4,2015.
U.S. Environmental Protection Agency (U.S. EPA). 2014c. Welfare Risk and Exposure
Assessment for Ozone: Final. EPA-452/R-14-005 a. Office of Air Quality Planning and
Standards, Research Triangle Park, NC. August. Available at:
. Accessed
June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2014d. Regulatory Impact Analysis of the
Proposed Revisions to the National Ambient Air Quality Standards for Ground-Level Ozone.
EPA-452/P-14-006. Office of Air Quality Planning and Standards, Research Triangle Park,
NC. November. Available at .
Accessed June 4, 2015.
U.S. Environmental Protection AgencyScience Advisory Board (U.S. EPA-SAB). 2002.
Workshop on the Benefits of Reductions in Exposure to Hazardous Air Pollutants:
Developing Best Estimates of Dose-Response Functions An SAB Workshop Report of an
EPA/SAB Workshop (Final Report). EPA-SAB-EC-WKSHP-02-001. January. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection AgencyScience Advisory Board (U.S. EPA-SAB). 2008.
Characterizing Uncertainty in Paniculate Matter Benefits Using Expert Elicitation. EPA-
COUNCIL-08-002. July. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection AgencyScience 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:
. Accessed June 4, 2015.
4-65
-------
U.S. Environmental Protection AgencyScience Advisory Board (U.S. EPA-SAB). 2004c.
Advisory Council on Clean Air Compliance Analysis Response to Agency Request on
Cessation Lag. EPA-COUNCIL-LTR-05-001. December. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection AgencyScience Advisory Board (U.S. EPA-SAB). 2008.
Benefits of Reducing Benzene Emissions in Houston, 1990-2020. EPA-COUNCIL-08-001.
July. Available at:
. Accessed June 4, 2015.
U.S. Environmental Protection AgencyScience Advisory Board (U.S. EPA-SAB). 2009b.
Review of EPA's Integrated Science Assessment for Paniculate Matter (First External Review
Draft, December 2008). EPA-COUNCIL-09-008. May. Available at:
.
Accessed June 4, 2015.
U.S. Environmental Protection AgencyScience Advisory Board (U.S. EPA-SAB). 2009c.
Review of Integrated Science Assessment for Paniculate Matter (Second External Review
Draft, July 2009). EPA-CASAC-10-001. November. Available at:
. Accessed
June 4, 2015.
U.S. Environmental Protection AgencyScience Advisory Board (U.S. EPA-SAB). 2010a.
Review of EPA's DRAFT Health Benefits of the Second Section 812 Prospective Study of the
Clean Air Act. EPA-COUNCIL-10-001. June. Available at: <
http://yosemite.epa.gov/sab/sabproduct.nsf/9288428b8eeea4c885257242006935a3/72D4EFA
39E48CDB28525774500738776/$File/EPA-COUNCIL-10-001-unsigned.pdf>. Accessed
June 4, 2015.
U.S. Environmental Protection AgencyScience Advisory Board (U.S. EPA-SAB). 2011.
Review of Valuing Mortality Risk Reductions for Environmental Policy: A White Paper
(December 10, 2010). EPA-SAB-11-011 July. Available at:
. Accessed June 4, 2015.
4-66
-------
Woodruff, T.J., J. Grille, and K.C. Schoendorf. 1997. "The Relationship between Selected of
postneonatal infant mortality and participate air pollution in the United States."
Environmental Health Perspectives. 105(6): 608-612.
4-67
-------
APPENDIX 4A: GENERATING REGIONAL BENEFIT-PER-TON ESTIMATES
The purpose of this appendix is to provide additional detail regarding the generation of
the benefit-per-ton estimates applied in Chapter 4 of this Regulatory Impact Analysis (RIA).
Specifically, this appendix describes the methods for generating benefit-per-ton estimates by
region for the contiguous U.S. for PMi.5 and ozone precursors emitted by the electrical
generating unit (EGU) sector in the Final Carbon Pollution Emission Guidelines for Existing
Stationary Sources: Electric Utility Generating Units (hereafter referred to as the "final emission
guidelines" or "Clean Power Plan Final Rule").
4A.1 Overview of Benefit-per-Ton Estimates
As described in the Technical Support Document: Estimating the Benefit per Ton of
Reducing PMz.s Precursors from 17 Sectors (U.S. EPA, 2013), the general procedure for
calculating average benefit-per-ton coefficients generally follows three steps. As an example, in
order to calculate regional average benefit-per-ton estimates for the key precursor pollutants
emitted from EGU sources, we:
1. Use air quality modeling to predict changes in ambient concentrations of primary PM2.5,
nitrate, sulfate, and ozone at a 12km2 grid resolution across the contiguous U.S. that are
attributable to the proposed Clean Power Plan.
2. For each grid cell, estimate the health impacts, and the economic value of these impacts,
associated with the attributable ambient concentrations using the environmental Benefits
Mapping and Analysis Program - Community Edition (BenMAP-CE vl.l).120'121
Aggregate those impacts and economic values to the three regions of East, West, and
California.
3. Divide the regional health impacts attributable to each precursor, and the regional
monetary value of these impacts, by the amount of associated regional precursor
emissions. That is, directly emitted PMi.5 benefits are divided by directly emitted PlVh.5
emissions, sulfate benefits are divided by SOi emissions, nitrate benefits are divided by
NOx emissions, and ozone benefits are divided by ozone-season NOx emissions.
120 When estimating these impacts we apply effect coefficients that relate changes in total PM mass to the risk of adverse health outcomes; we do not apply effect coefficients
that are differentiated by PM species.
121 Previous RIAs have used earlier versions of the BenMAP software. BenMAP-CE vl.l provides results consistent
with earlier versions of BenMAP and is available for download at http://www.epa.gov/air/benmap/.
4A-1
-------
4A.2 Air Quality Modeling for the Proposed Clean Power Plan
The EPA ran the Comprehensive Model with Extensions (CAMx) photochemical model
(ENVIRON, 2014) to predict ozone and PlVb.5 concentrations for the following emissions
scenarios: a 2011 base year, a 2025 base case, and the 2025 proposed Clean Power Plan (Option
1 State) scenario. Each of the CAMx model simulations was performed for a nationwide
modeling domain122 using a full year of meteorological conditions for 2011. The modeling for
2011 was used as the anchor point for projecting ozone and annual PMi.5 concentration values
for the 2025 base case and for the 2025 Clean Power Plan proposal scenario using methodologies
consistent with the EPA's air quality modeling guidance (U.S. EPA, 2007). The air quality
modeling results for the 2025 base case served as the baseline for gauging the future year
impacts on ozone and annual PM2.5 of the Clean Power Plan proposal scenario. The 2025 base
case reflects emissions reductions between 2011 and 2025 that are expected to result from
regional and national rules including the Clean Air Interstate Rule (CAIR), the Mercury and Air
Toxics Standards (MATS), mobile source rules up through Tier-3, and various state emissions
control programs and consent decrees. The methods for estimating the EGU emissions for the
proposal are described in Chapter 3 of the RIA for the Clean Power Plan proposal (U.S. EPA,
2014). State total annual EGU emissions for NOx and SO2 for each of the scenarios modeled are
provided in Tables 4A-1 and 4A-2, respectively. The data indicate that, overall nationwide, EGU
SO2 and NOx emissions with proposed Option 1 (state) would be about 28% lower than the 2025
base case.
122 The modeling domain (i.e., region modeled) includes all of the lower 48 states plus adjacent portions of Canada
and Mexico) at a spatial resolution of 12 km.
4A-2
-------
Table 4A-1. State Total Annual ECU Emissions
Base Case, and 2025 Clean Power Plan
of tons)
for NOx for the 2011 Base Year, 2025
Proposal (Option 1 State) (in thousands
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
Tribal Data
Utah
Vermont
2011 Base Year
63
35
38
6
51
1
4
61
54
-
73
121
40
44
92
47
2
19
5
75
32
26
66
20
37
7
4
6
23
22
46
51
104
82
5
149
0
25
11
27
146
65
51
0
2025 Base Case
38
17
43
33
29
1
1
52
33
1
38
97
24
28
59
18
4
11
2
73
27
15
61
16
38
5
1
7
7
11
35
51
63
52
3
106
0
13
13
16
144
33
49
0
2025 Clean Power Plan
Proposal
(Option 1 State)
19
4
9
28
21
1
1
15
18
0
32
90
24
27
74
14
2
11
1
51
13
3
58
15
35
3
0
2
6
7
23
48
60
26
3
71
1
8
8
13
64
33
33
0
4A-3
-------
State
Virginia
Washington
West Virginia
Wisconsin
Wyoming
National Total
2011 Base Year
38
7
58
32
53
2,024
2025 Base Case
21
3
49
19
50
1,508
2025 Clean Power Plan
Proposal
(Option 1 State)
12
2
46
11
38
1,084
Table 4A-2. State Total Annual ECU Emissions for SCh for the 2011 Base Year, 2025
Base Case, and 2025 Clean Power Plan Proposal (Option 1 State) (in thousands
of tons)
2025 Clean Power Plan
State 2011 Base Year 2025 Base Case Proposal
(Option 1 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
186
28
74
1
45
1
11
95
187
-
227
382
100
39
246
93
1
32
23
228
40
43
205
19
73
5
24
5
6
41
78
93
79
18
30
4
15
-
1
70
37
0
45
126
18
15
109
14
1
5
1
122
21
10
80
18
25
1
0
7
4
4
36
15
45
4
5
4
10
-
1
7
12
0
48
121
18
15
119
11
1
9
0
95
12
3
76
17
24
1
0
1
4
2
33
14
4A-4
-------
State
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Tribal Data
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
National Total
2011 Base Year
594
96
13
338
0
68
11
120
426
18
22
0
75
1
103
92
55
4,665
2025 Base Case
105
21
1
67
-
19
11
38
149
19
14
0
8
1
78
17
23
1,504
2025 Clean Power Plan
Proposal
(Option 1 State)
102
6
1
47
-
12
7
31
48
19
10
0
4
1
47
11
17
1,077
As indicated above, the air quality modeling was used to project gridded ozone and
annual PlVb.5 concentrations at the 12km2 resolution for the 2025 base case and the Clean Power
Plan proposal scenario modeled for this analysis. The air quality modeling results were combined
with monitored ozone and PM2.5 data to create projected spatial fields of annual PMi.5 and
seasonal mean (May through September) 8-hour daily maximum ozone for the 2025 base case
and for the proposal scenario. These spatial fields were then used as inputs to estimate the health
co-benefits of the proposed Clean Power Plan as described below.
4A.3 Regional PM2.5 Benefit-per-Ton Estimates for EGUs Derived from Air Quality
Modeling of the Proposed Clean Power Plan
After estimating the 12km2 resolution PMi.5 benefits for each of the analysis years
applied in this RIA (i.e., 2020, 2025, and 2030), we aggregated the benefits results regionally
(i.e., East, West, and California), as shown in Figure 4A-1.123 While a small percentage of
benefits from emissions reductions in a particular region may occur in one of the other regions,
we selected each region to minimize this percentage. Thus, the benefits per ton in each region
123 This aggregation is identified as the shapefile "Report Regions" in BenMAP's grid definitions.
4A-5
-------
will represent well the match between where the emissions reductions and air quality benefits are
occurring. Due to the low emissions of SOi, NOx, and directly emitted particles from EGUs in
California and the high population density, we separated out California in order not to bias the
benefit-per-ton estimates for the rest of the Western U.S. In order to calculate the benefit-per-ton
estimates, we divided the regional benefits estimates by the corresponding emissions, as shown
in Table 4A-1. Lastly, we adjusted the benefit-per-ton estimates for a currency year of 2011$.124
This method provides estimates of the regional average benefit-per-ton for a subset of the
major PMi.5 precursors emitted from EGU sources. For precursor emissions of NOx, there is
generally a non-linear relationship between emissions and formation of PlVb.5. This means that
each ton of NOx reduced would have a different impact on ambient PM2.5 depending on the
initial level of emissions and potentially on the levels of emissions of other pollutants. In
contrast, SOi is generally linear in forming PMi.5. For precursors like NOx which form PMi.5
non-linearly, a marginal benefit-per-ton approach would better approximate the specific benefits
associated with an emissions reduction scenario for a given set of base case emissions, because it
would allow the benefit-per-ton to vary depending on the level of emissions reductions and the
baseline emissions levels. However, we do not have sufficient air quality modeling data to
calculate marginal benefit-per-ton estimates for the EGU sector. Therefore, using an average
benefit-per-ton estimate for NOx adds uncertainty to the co-benefits estimated in this RIA.
Because most of the estimated co-benefits for the proposed guidelines are attributable to
reductions in SO2 emissions, the added uncertainty is likely to be small.
124 Currently, BenMAP does not have an inflation adjustment to 2011$. We ran BenMAP for a currency year of 2010$ and calculated the benefit-per-ton estimates in 2010$. We
then adjusted the resulting benefit-per-ton estimates to 2011$ using the Consumer Price Index.
4A-6
-------
Figure 4A-1. Regional Breakdown
In this RIA, we estimate emission reductions from EGUs using IPM.125 IPM outputs
provide endogenously projected unit level emissions of SOi, NOx, COi, Hg, hydrogen chloride
(HC1) from EGUs, but carbon monoxide, volatile organic compounds, ammonia and total
directly emitted PMi.5 and PMio emissions are post-calculated.126 In addition, directly emitted
particle emissions calculated from IPM outputs do not include speciation, i.e. they are only the
total emissions. In order to conduct air quality modeling, directly emitted PMi.5 from EGUs is
speciated into components during the emissions modeling process based on emission profiles for
EGUs by source classification code. Even though these speciation profiles are not unit-specific,
an emission profile based on the source classification code is highly sophisticated and reflects the
fuel and the unit configuration. Model-predicted concentrations of nitrate and sulfate include
125 See Chapter 3 of this RIA for additional information regarding j-jjg Integrated Planning Model (
/-IPM).
126 Detailed documentation of this post-processing is available at
http://www.epa.gov/powersectormodeling/docs/v5 13/FlatFile_Methodology.pdf
4A-7
-------
both the directly emitted nitrate and sulfate from speciated PlVb.5 and secondarily formed nitrate
and sulfate from emissions of NOx and SOi, respectively.
In order to estimate the benefits associated with reduced emissions of directly emitted
particles without performing air quality modeling, we must determine the fraction of total PlVb.5
emissions comprised of elemental carbon and organic carbon (EC+OC) and crustal emissions.127
Based on the work by Fann, Baker, and Fulcher (2012), the national average EC+OC fraction of
emitted PMi.5 is 10% with a range of 5% to 63% in different states due to the different proportion
of fuels. The national average is similar to the averages for the east and west regions at 10% and
7%, respectively. Only five states had EC+OC fractions greater than 30%. For crustal emissions,
the national average fraction of emitted PM2.5 from EGUs is 78% with a range of 26% to 83%.
The national average is similar to the averages for the east and west regions at 78% and 81%,
respectively. Only four states had crustal fractions less than 50%. In calculating the PMi.5 co-
benefits in this RIA, we estimate the emission reductions of EC+OC and crustal emissions by
applying the national average fractions (i.e., 78% crustal and 10% EC+OC) to the emission
reductions of all directly emitted particles from EGUs. Because the benefit-per-ton estimates for
reducing emissions of EC+OC are larger than the benefit-per-ton estimate for crustal emissions,
this assumption underestimates the monetized PMi.5 co-benefits in certain states with higher
EC+OC fractions, such as California and North Dakota.
Although it is possible to calculate 95th percentile confidence intervals using the approach
described in this appendix (e.g., U.S. EPA, 201 Ib), we generally do not calculate confidence
intervals for benefit-per-ton estimates because of the additional unquantified uncertainties that
result from the benefit transfer methods, including those related to the transfer of air quality
modeling information. Instead, we refer the reader to Chapter 5 of PM NAAQS RIA (U.S. EPA,
2012a) for an indication of the combined random sampling error in the health impact and
economic valuation functions using Monte Carlo methods. In general, the 95th percentile
confidence interval for the total monetized PMi.5 benefits ranges from approximately -90% to
+180% of the central estimates based on concentration-response functions from Krewski et al.
(2009) and Lepeule et al (2012). The 95th percentile confidence interval for the health impact
127 Crustal emissions are composed of compounds associated with minerals and metals from the earth's surface,
including carbonates, silicates, iron, phosphates, copper, and zinc. Often, crustal material represents particles not
classified as one of the other species (e.g., organic carbon, elemental carbon, nitrate, sulfate, chloride, etc.).
4A-8
-------
function alone ranges from approximately ±30% for mortality incidence based on Krewski et al.
(2009) and ±46% based on Lepeule et al. (2012). These confidence intervals do not reflect other
sources of uncertainty inherent within the estimates, such as baseline incidence rates, populations
exposed, and transferability of the effect estimate to diverse locations. As a result, the reported
confidence intervals and range of estimates give an incomplete picture about the overall
uncertainty in the benefits estimates.
Tables 4A-3 through 4A-5 provide the regional benefit-per-ton estimates for the EGU
sector at discount rates of 3% and 7% in 2020, 2025, and 2030 respectively. The benefit-per-ton
values for 2020 and 2030 are based on applying the air quality modeling from 2025 to population
and health information from 2020 and 2030. Estimated benefit-per-ton for these years have
additional uncertainty relative to 2025 because of potential differences in atmospheric responses
to reductions in PlVh.5 precursors in those years, however, these uncertainties are likely to be
relatively small. Tables 4A-6 through 4A-8 provide the incidence per ton estimates (which
follows the same general methodology as for the benefit-per-ton calculations) for the EGU sector
in 2020, 2025, and 2030 respectively, for the set of health endpoints used to calculate the benefit-
per-ton estimates.
4A-9
-------
Table 4A-3. Summary of Regional PM2.5 Benefit-per-Ton Estimates Based on Air Quality Modeling from Proposed Clean Power Plan
in 2020 (2011$)*
Pollutant
SO2
Directly emitted
PM2.s(EC+OC)
Directly emitted
PM2.5(Crustal)
NOx (as PM2.5)
Discount
Rate
3%
7%
3%
7%
3%
7%
3%
7%
National
$32,000 to $71,000
$28,000 to $64,000
$140,000 to $3 10,000
$120,000 to $270,000
$22,000 to $49,000
$20,000 to $44,000
$3,000 to $6,800
$2,700 to $5,600
East
$33,000 to $75,000
$30,000 to $68,000
$140,000 to $320,000
$130,000 to $290,000
$23,000 to $52,000
$21,000 to $47,000
$3, 100 to $7,000
$2,800 to $6,300
Region
West
$6,200 to $14,000
$5,600 to $13,000
$27,000 to $60,000
$24,000 to $54,000
$11, 000 to $25,000
$9,900 to $22,000
$0,670 to $1,500
$0,6 10 to $1,400
California
$95,000 to $210,000
$85,000 to $190,000
$370,000 to $830,000
$330,000 to $740,000
$73,000 to $160,000
$66,000 to $150,000
$22,000 to $49,000
$19,000 to $44,000
* The range of estimates reflects the range of epidemiology studies for avoided premature mortality for PM2.s. All estimates are rounded to two significant figures. All fine
particles are assumed to have equivalent health effects, but the benefit-per-ton estimates vary depending on the location and magnitude of their impact on PM2.5 levels,
which drive population exposure. The monetized benefits incorporate the conversion from precursor emissions to ambient fine particles. The estimates do not include
reduced health effects from direct exposure to ozone, NO2, SO2, ecosystem effects, or visibility impairment.
Table 4A-4. Summary of Regional PM2.5 Benefit-per-Ton Estimates Based on Air Quality Modeling from Proposed Clean Power Plan
in 2025 (2011$)*
Pollutant
SO2
Directly emitted
PM2.5(EC+OC)
Directly emitted
PM2.5(Crustal)
NOx (as PM2.5)
Discount
Rate
3%
7%
3%
7%
3%
7%
3%
7%
National
$35,000 to $78,000
$3 1,000 to $70,000
$150,000 to $340,000
$130,000 to $290,000
$24,000 to $55,000
$22,000 to $49,000
$3,200 to $7,300
$2,900 to $6,000
East
$37,000 to $83,000
$33,000 to $75,000
$160,000 to $360,000
$140,000 to $320,000
$25,000 to $58,000
$23,000 to $52,000
$3,300 to $7,500
$3,000 to $6,800
Region
West
$7, 100 to $16,000
$6,400 to $14,000
$30,000 to $68,000
$27,000 to $61,000
$12,000 to $28,000
$11, 000 to $25,000
$0,750 to $1,700
$0,670 to $1,500
California
$110,000 to $240,000
$97,000 to $220,000
$410,000 to $930,000
$370,000 to $830,000
$82,000 to $180,000
$74,000 to $170,000
$24,000 to $54,000
$22,000 to $49,000
* The range of estimates reflects the range of epidemiology studies for avoided premature mortality for PM2.s. All estimates are rounded to two significant figures. All fine
particles are assumed to have equivalent health effects, but the benefit-per-ton estimates vary depending on the location and magnitude of their impact on PM2.s levels,
which drive population exposure. The monetized benefits incorporate the conversion from precursor emissions to ambient fine particles. The estimates do not include
reduced health effects from direct exposure to ozone, NO2, SO2, ecosystem effects, or visibility impairment.
4A-10
-------
Table 4A-5. Summary of Regional PM2.5 Benefit-per-Ton Estimates Based on Air Quality Modeling from Proposed Clean Power Plan
in 2030 (2011$)*
Pollutant
SO2
Directly emitted
PM2.s(EC+OC)
Directly emitted
PM2.5(Crustal)
NOx (as PM2.5)
Discount
Rate
3%
7%
3%
7%
3%
7%
3%
7%
National
$37,000 to $85,000
$34,000 to $76,000
$160,000 to $360,000
$150,000 to $320,000
$26,000 to $59,000
$24,000 to $53,000
$3,400 to $7,800
$3, 100 to $6,400
East
$40,000 to $89,000
$36,000 to $8 1,000
$170,000 to $380,000
$150,000 to $340,000
$28,000 to $62,000
$25,000 to $56,000
$3,500 to $8,000
$3,200 to $7,200
Region
West
$7,800 to $18,000
$7, 100 to $16,000
$33,000 to $75,000
$30,000 to $68,000
$14,000 to $3 1,000
$13,000 to $28,000
$0,820 to $1,900
$0,740 to $1,700
California
$120,000 to $270,000
$110,000 to $240,000
$450,000 to $1,000,000
$410,000 to $920,000
$90,000 to $200,000
$81,000 to $180,000
$26,000 to $60,000
$24,000 to $54,000
* The range of estimates reflects the range of epidemiology studies for avoided premature mortality for PM2.s. All estimates are rounded to two significant figures. All fine
particles are assumed to have equivalent health effects, but the benefit-per-ton estimates vary depending on the location and magnitude of their impact on PM2.5 levels,
which drive population exposure. The monetized benefits incorporate the conversion from precursor emissions to ambient fine particles. The estimates do not include
reduced health effects from direct exposure to ozone, NO2, SO2, ecosystem effects, or visibility impairment.
4A-11
-------
Table 4A-6. Summary of Regional PM2.5 Incidence-per-Ton Estimates Based on Air Quality Modeling from Proposed Clean Power
Plan in 2020*
Health Endpoint
Premature Mortality
Krewski et al. (2009) - adult
Lepeule et al. (2012) - adult
Woodruff et al. (1997) - infants
Morbidity
Emergency department visits for asthma
Acute bronchitis
Lower respiratory symptoms
Upper respiratory symptoms
Minor restricted-activity days
Lost work days
Asthma exacerbation
Hospital Admissions, Respiratory
Hospital Admissions, Cardiovascular
Non-fatal Heart Attacks (age>18)
Peters etal (2001)
Pooled estimate of 4 studies
East
SO2 NOx EC+OC Crustal
0.003700 0.000340 0.016000 0.002500
0.008300 0.000770 0.036000 0.005700
0.000009 0.000001 0.000037 0.000006
0.001900 0.000190 0.007800 0.001300
0.005400 0.000510 0.023000 0.003700
0.069000 0.006500 0.300000 0.047000
0.098000 0.009300 0.420000 0.068000
2.700000 0.250000 11.000000 1.900000
0.450000 0.043000 1.900000 0.310000
0.240000 0.023000 1.000000 0.170000
0.001100 0.000100 0.004500 0.000720
0.001300 0.000120 0.005600 0.000910
0.004100 0.000390 0.018000 0.002800
0.000450 0.000042 0.001900 0.000310
West
SO2 NOx EC+OC Crustal
0.000680 0.000073 0.002900 0.001200
0.001500 0.000170 0.006600 0.002700
0.000002 0.000000 0.000007 0.000003
0.000290 0.000031 0.001200 0.000470
0.001300 0.000200 0.005200 0.002100
0.016000 0.002500 0.067000 0.026000
0.023000 0.003600 0.095000 0.038000
0.580000 0.078000 2.400000 0.920000
0.098000 0.013000 0.410000 0.160000
0.056000 0.008800 0.230000 0.091000
0.000150 0.000015 0.000640 0.000260
0.000200 0.000019 0.000820 0.000330
0.000650 0.000064 0.002800 0.001200
0.000070 0.000007 0.000300 0.000130
California
SO2 NOx EC+OC Crustal
0.010000 0.002400 0.040000 0.008000
0.023000 0.005400 0.091000 0.018000
0.000023 0.000007 0.000097 0.000019
0.005300 0.001400 0.022000 0.004200
0.019000 0.005000 0.077000 0.015000
0.240000 0.064000 0.970000 0.190000
0.340000 0.092000 1.400000 0.270000
9.400000 2.200000 35.000000 6.800000
1.600000 0.380000 6.000000 1.100000
0.840000 0.220000 3.400000 0.650000
0.002500 0.000580 0.009400 0.001900
0.003000 0.000680 0.011000 0.002200
0.011000 0.002400 0.041000 0.007900
0.001100 0.000260 0.004400 0.000850
* All estimates are rounded to two significant figures. All fine particles are
the location and magnitude of their impact on PMi.s levels, which drive
precursor emissions to ambient fine particles.
assumed to have equivalent health effects, but the incidence-per-ton estimates vary depending on
population exposure. The incidence benefit-per-ton estimates incorporate the conversion from
4A-12
-------
Table 4A-7. Summary of Regional PM2.5 Incidence-per-Ton Estimates Based on Air Quality Modeling from Proposed Clean Power
Plan in 2025*
Health Endpoint
Premature Mortality
Krewski et al. (2009) - adult
Lepeule et al. (2012) - adult
Woodruff et al. (1997) - infants
Morbidity
Emergency department visits for
asthma
Acute bronchitis
Lower respiratory symptoms
Upper respiratory symptoms
Minor restricted-activity days
Lost work days
Asthma exacerbation
Hospital Admissions, Respiratory
Hospital Admissions, Cardiovascular
Non-fatal Heart Attacks (age>18)
Peters etal (2001)
Pooled estimate of 4 studies
East
SO2 NOx EC+OC Crustal
0.003900 0.000350 0.017000 0.002700
0.008900 0.000800 0.038000 0.006200
0.000008 0.000001 0.000035 0.000006
0.002000 0.000200 0.006300 0.001300
0.005700 0.000520 0.024000 0.003900
0.072000 0.006700 0.310000 0.050000
0.100000 0.009600 0.440000 0.071000
2.800000 0.250000 12.000000 1.900000
0.470000 0.043000 2.000000 0.320000
0.250000 0.023000 1.100000 0.170000
0.001200 0.000110 0.005100 0.000810
0.001400 0.000130 0.006200 0.001000
0.004600 0.000430 0.020000 0.003100
0.000490 0.000046 0.002100 0.000340
West
SO2 NOx EC+OC Crustal
0.000750 0.000079 0.003200 0.001300
0.001700 0.000180 0.007300 0.003000
0.000002 0.000000 0.000007 0.000003
0.000320 0.000033 0.001000 0.000510
0.001300 0.000210 0.005600 0.002200
0.017000 0.002700 0.071000 0.028000
0.024000 0.003800 0.100000 0.040000
0.610000 0.083000 2.500000 0.970000
0.100000 0.014000 0.430000 0.160000
0.059000 0.009300 0.250000 0.097000
0.000180 0.000017 0.000740 0.000300
0.000220 0.000022 0.000930 0.000380
0.000740 0.000071 0.003200 0.001300
0.000080 0.000008 0.000340 0.000140
California
SO2 NOx EC+OC Crustal
0.011000 0.002600 0.044000 0.008700
0.026000 0.005800 0.099000 0.020000
0.000022 0.000007 0.000093 0.000018
0.005500 0.001500 0.018000 0.004400
0.020000 0.005300 0.080000 0.015000
0.250000 0.067000 1.000000 0.200000
0.360000 0.096000 1.500000 0.280000
9.600000 2.300000 36.000000 6.900000
1.600000 0.390000 6.100000 1.200000
0.880000 0.230000 3.500000 0.680000
0.002800 0.000650 0.011000 0.002100
0.003300 0.000750 0.012000 0.002400
0.012000 0.002700 0.046000 0.008900
0.001300 0.000290 0.004900 0.000950
* All estimates are rounded to two significant figures. All fine
the location and magnitude of their impact on PM2.s levels,
precursor emissions to ambient fine particles.
particles are assumed to have equivalent health effects, but the incidence-per-ton estimates vary depending on
which drive population exposure. The incidence benefit-per-ton estimates incorporate the conversion from
4A-13
-------
Table 4A-8. Summary of Regional PM2.5 Incidence-per-Ton Estimates Based on Air Quality Modeling from Proposed Clean Power
Plan in 2030*
Health Endpoint
Premature Mortality
Krewski et al. (2009) - adult
Lepeule et al. (2012) - adult
Woodruff et al. (1997) - infants
Morbidity
Emergency department visits for
asthma
Acute bronchitis
Lower respiratory symptoms
Upper respiratory symptoms
Minor restricted-activity days
Lost work days
Asthma exacerbation
Hospital Admissions, Respiratory
Hospital Admissions, Cardiovascular
Non-fatal Heart Attacks (age>18)
Peters etal (2001)
Pooled estimate of 4 studies
East
SO2 NOx EC+OC Crustal
0.004200 0.000380 0.018000 0.002900
0.009600 0.000850 0.041000 0.006700
0.000008 0.000001 0.000033 0.000005
0.001600 0.000160 0.006600 0.001100
0.005900 0.000540 0.025000 0.004100
0.075000 0.006800 0.320000 0.052000
0.110000 0.009800 0.460000 0.074000
2.900000 0.260000 12.000000 2.000000
0.480000 0.043000 2.000000 0.330000
0.260000 0.024000 1.100000 0.180000
0.001300 0.000120 0.005600 0.000900
0.001600 0.000150 0.006800 0.001100
0.005000 0.000460 0.021000 0.003500
0.000540 0.000049 0.002300 0.000370
West
SO2 NOx EC+OC Crustal
0.000840 0.000087 0.003600 0.001500
0.001900 0.000200 0.008100 0.003400
0.000002 0.000000 0.000007 0.000003
0.000260 0.000027 0.001100 0.000420
0.001400 0.000220 0.005900 0.002300
0.018000 0.002800 0.075000 0.030000
0.026000 0.004000 0.110000 0.042000
0.650000 0.088000 2.700000 1.000000
0.110000 0.015000 0.450000 0.170000
0.063000 0.009800 0.260000 0.100000
0.000200 0.000019 0.000830 0.000340
0.000250 0.000024 0.001000 0.000420
0.000830 0.000079 0.003600 0.001500
0.000090 0.000009 0.000380 0.000160
California
SO2 NOx EC+OC Crustal
0.013000 0.002800 0.048000 0.009600
0.029000 0.006400 0.110000 0.022000
0.000021 0.000006 0.000088 0.000017
0.004500 0.001200 0.019000 0.003500
0.021000 0.005400 0.083000 0.016000
0.260000 0.069000 1.100000 0.200000
0.370000 0.099000 1.500000 0.290000
9.800000 2.300000 37.000000 7.100000
1.700000 0.400000 6.300000 1.200000
0.920000 0.240000 3.700000 0.710000
0.003200 0.000740 0.012000 0.002400
0.003800 0.000850 0.014000 0.002700
0.014000 0.003100 0.052000 0.010000
0.001500 0.000330 0.005600 0.001100
* All estimates are rounded to two significant figures. All fine
the location and magnitude of their impact on PM2.s levels,
precursor emissions to ambient fine particles.
particles are assumed to have equivalent health effects, but the incidence-per-ton estimates vary depending on
which drive population exposure. The incidence benefit-per-ton estimates incorporate the conversion from
4A-14
-------
4A.4 Regional Ozone Benefit-per-Ton Estimates
The process for generating the regional ozone benefit-per-ton estimates is consistent with
the process for PMi.5. Ozone is not directly emitted, and is a non-linear function of NOx and
VOC emissions. For the purpose of estimating benefit-per-ton for this RIA, we assume that all of
the ozone impacts from EGUs are attributable to NOx emissions. VOC emissions, which are also
a precursor to ambient ozone formation, are insignificant from the EGU sector relative to both
NOx emissions from EGUs and the total VOC emissions inventory. Therefore, we believe that
our assumption that EGU-attributable ozone formation at the regional-level is due to NOx alone
is reasonable.
Similar to PM2.5, this method provides estimates of the regional average benefit-per-ton.
Due to the non-linear chemistry between NOx emissions and ambient ozone, using an average
benefit-per-ton estimate for NOx adds uncertainty to the ozone co-benefits estimated for the
proposed guidelines. Because most of the estimated co-benefits for the proposed guidelines are
attributable to changes in ambient PMi.5, the added uncertainty is likely to be small.
In the ozone co-benefits estimated in this RIA, we apply the benefit-per-ton estimates
calculated using NOx emissions derived from modeling the Clean Power Plan proposal during
the ozone-season only (May to September). As shown in Table 4A-1, ozone-season NOx
emissions from EGUs are slightly less than half of all-year NOx emissions. Because we estimate
ozone health impacts from May to September only, this approach underestimates ozone co-
benefits in areas with longer ozone seasons such as southern California and Texas. When the
underestimated benefit-per-ton estimate is multiplied by ozone-season only NOx emission
reductions, this results in an underestimate of the monetized ozone co-benefits. For illustrative
purposes, Tables 4A-9 through 4A-11 provide the ozone benefit-per-ton estimates using both all-
year NOx emissions and ozone-season only NOx for 2020, 2025, and 2030, respectively. Tables
4A-12 through 4A-14 provide the ozone season incidence-per-ton estimates for 2020, 2025, and
2030, respectively. Similar to PMi.5, the ozone benefit-per-ton values for 2020 and 2030 are
based on applying the air quality modeling from 2025 to population and health information from
2020 and 2030. Estimated benefit-per-ton for these years have additional uncertainty relative to
2025 because of potential differences in atmospheric responses to reductions in ozone precursors
4A-15
-------
in those years. Uncertainties may be somewhat larger in the case of ozone due to high degree of
dependence of ozone responses to baseline meteorology and emissions levels.
Table 4A-9. Summary of Regional Ozone Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2020 (2011$)*
Ozone precursor , Regional
Pollutant East West California
Ozone season NOx $6,000 to $26,000 $6,500 to $28,000 $2,000 to $8,900 $14,000 to $59,000
* The range of estimates reflects the range of epidemiology studies for avoided premature mortality for ozone. All
estimates are rounded to two significant figures. The monetized benefits incorporate the conversion from NOx
precursor emissions to ambient ozone.
Table 4A-10. Summary of Regional Ozone Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2025 (2011$)*
Ozone precursor , Regional
Pollutant ai°n East West California
Ozone season NOx $6,600 to $27,000 $7,100 to $30,000 $2,300 to $10,000 $15,000 to $66,000
* The range of estimates reflects the range of epidemiology studies for avoided premature mortality for ozone. All
estimates are rounded to two significant figures. The monetized benefits incorporate the conversion from NOx
precursor emissions to ambient ozone.
Table 4A-11. Summary of Regional Ozone Benefit-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2030 (2011$)*
Ozone precursor , Regional
Pollutant East West California
Ozone season NOx $7,100 to $29,000 $7,600 to $33,000 $2,600 to $11,000 $17,000 to $73,000
* The range of estimates reflects the range of epidemiology studies for avoided premature mortality for ozone. All
estimates are rounded to two significant figures. The monetized benefits incorporate the conversion from NOx
precursor emissions to ambient ozone.
4A-16
-------
Table 4A-12. Summary of Regional Ozone Incidence-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2020*
Health Endpoint
Premature Mortality - adult
Bell et al. (2004)
Levy et al (2005)
Morbidity
Hospital Admissions, Respiratory (ages > 65)
Hospital Admissions, Respiratory (ages < 2)
Emergency Room Visits, Respiratory
Acute Respiratory Symptoms
School Loss Days
East
0.000600
0.002800
0.003500
0.001800
0.002000
3.500000
1.200000
West
0.000190
0.000880
0.000900
0.000780
0.000500
1.300000
0.490000
California
0.001300
0.005800
0.006600
0.003300
0.003900
8.800000
3.000000
* All estimates are rounded to two significant figures. The incidence benefit-per-ton estimates incorporate the
conversion from NOx precursor emissions to ambient ozone. These estimates reflect ozone-season NOx
emissions.
Table 4A-13. Summary of Regional Ozone Incidence-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2025*
Health Endpoint
Premature Mortality - adult
Bell et al. (2004)
Levy et al. (2005)
Morbidity
Hospital Admissions, Respiratory (ages > 65)
Hospital Admissions, Respiratory (ages < 2)
Emergency Room Visits, Respiratory
Acute Respiratory Symptoms
School Loss Days
East
0.000640
0.002900
0.004100
0.001800
0.002000
3.600000
1.300000
West
0.000210
0.000970
0.001100
0.000820
0.000540
1.400000
0.520000
California
0.001400
0.006400
0.007800
0.003400
0.004100
8.900000
3.200000
* All estimates are rounded to two significant figures. The incidence benefit-per-ton estimates incorporate the
conversion from NOx precursor emissions to ambient ozone. These estimates reflect ozone-season NOx
emissions.
Table 4A-14. Summary of Regional Ozone Incidence-per-Ton Estimates Based on Air
Quality Modeling from Proposed Clean Power Plan in 2030*
Health Endpoint
Premature Mortality - adult
Bell et al (2004)
Levy et al (2005)
Morbidity
Hospital Admissions, Respiratory (ages > 65)
Hospital Admissions, Respiratory (ages < 2)
Emergency Room Visits, Respiratory
Acute Respiratory Symptoms
School Loss Days
East
0.000640
0.002900
0.004400
0.001800
0.002000
3.500000
1.200000
West
0.000230
0.001100
0.001300
0.000860
0.000580
1.500000
0.550000
California
0.001800
0.008200
0.011000
0.004100
0.005000
11.000000
3.800000
* All estimates are rounded to two significant figures. The incidence benefit-per-ton estimates incorporate the
conversion from NOx precursor emissions to ambient ozone. These estimates reflect ozone-season NOx
emissions.
4A-17
-------
4A.5 References
Abt Associates, Inc. 2010. "User's Guide: Modeled Attainment Test Software." Available at:
. Accessed June 6, 2015.
Abt Associates, Inc. 2012. "BenMAP User's Manual Appendices," prepared for U.S. Research
Triangle Park, NC: U. S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Available at:
. Accessed June
6,2015.
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.
ENVIRON International Corporation. 2014. User's Guide: Comprehensive Air Quality Model
with Extensions, Version 6.1, Novato, CA. April. Available at .
Accessed June 6, 2015.
Fann, N., C.M. Fulcher, and B.J. Hubbell. 2009. "The Influence of Location, Source, and
Emission Type in Estimates of the Human Health Benefits of Reducing a Ton of Air
Pollution." Air Quality and Atmospheric Health. 2:169-176.
Fann, N., K.R. Baker, and C.M. Fulcher. 2012. "Characterizing the PM2.5-Related Health
Benefits of Emission Reductions for 17 Industrial, Area and Mobile Emission Sectors Across
the U.S." Environment International. 49:41-151.
Fann, N., K.R. Baker, and C.M. Fulcher. 2013. "The Recent and Future Health Burden of Air
Pollution Apportioned Across 23 U.S. Sectors." Environmental Scientific Technology. 47(8):
3580-3589.
Huang Y., F. Dominici, and M. Bell. 2005. "Bayesian Hierarchical Distributed Lag Models for
Summer Ozone Exposure and Cardio-Respiratory Mortality." Environmetrics. 16:547-562.
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.
4A-18
-------
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 Paniculate 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.
Roman, H.A., K. D. Walker, T. L. Walsh, L. Conner, H. M. Richmond, B. J. Hubbell, and P. L.
Kinney. 2008. "Expert Judgment Assessment of the Mortality Impact of Changes in Ambient
Fine Particulate Matter in the U.S." Environmental Scientific Technology. 42(7):2268-2274.
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.
U.S. Environmental Protection Agency (U.S. EPA). 2007. Guidance on the Use of Models and
Other Analyses for Demonstrating Attainment of Air Quality Goals for Ozone, PM2.5, and
Regional Haze. Office of Air Quality Planning and Standards, Research Triangle Park, NC.
Available at .
Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 201 Ib Regulatory Impact Analysis for the
Final Mercury and Air Toxics Standards. Research Triangle Park, NC: Office of Air Quality
Planning and Standards, Health and Environmental Impacts Division. (EPA document
number EPA-452/R-11-011, December). Available at
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2012a. 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. December. Available at: <
http://www.epa.gov/ttnecasl/regdata/RIAs/finalria.pdf>. Accessed June 4, 2015.
4A-19
-------
U.S. Environmental Protection Agency (U.S. EPA). 2013. Technical Support Document:
Estimating the Benefit per Ton of Reducing PM2.5 Precursors from 17 Sectors. Office of Air
Quality Planning and Standards, Research Triangle Park, NC. February. Available at: <
http://www2.epa.gov/sites/production/files/2014-
10/documents/sourceapportionmentbpttsd.pdf >. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2014. Regulatory Impact Analysis for the
Proposed Carbon Pollution Guidelines for Existing Power Plants and Emission Standards for
Modified and Reconstructed Power Plants. EPA-542/R-14-002. Office of Air Quality
Planning and Standards, Research Triangle Park, NC. June. Available at
. Accessed June
4,2015.
4A-20
-------
CHAPTER 5: ECONOMIC IMPACTS - MARKETS OUTSIDE THE UTILITY POWER
SECTOR
5.1 Introduction
The energy sector impacts presented in Chapter 3 of this RIA include potential changes in
the prices for electricity, natural gas, and coal potentially resulting from the Clean Power Plan
Final Rule. This chapter addresses the impact of these potential changes on other markets and
discusses some of the determinants of the magnitude of these impacts. We refer to these changes
as secondary market impacts.
Under the final emission guidelines, states are not required to use any of the measures that
the EPA determines constitute BSER, or use those measures to the same degree of stringency
that the EPA determines is achievable at reasonable cost. Rather, CAA section lll(d) allows
each state to determine the appropriate combination of, and the extent of its reliance on,
measures for its state plan, by way of meeting its state-specific goal. Given the flexibilities
afforded states in complying with the emission guidelines, the benefits, cost and economic
impacts reported in this RIA are illustrative of actions that states may take. The implementation
approaches adopted by the states, and the strategies adopted by affected EGUs, will ultimately
drive the magnitude and timing of secondary impacts from changes in the price of electricity, and
the demand for inputs by the electricity sector, on other markets that use and produce these
inputs.
The flexibility afforded to states in their state plans allows them to encourage compliance
methods by affected EGUs, which include design elements that may mitigate or promote
particular impacts based on their priorities. For example, states in the Regional Greenhouse Gas
Initiative use the revenues from allowance auctions to support direct bill assistance for retail
consumers, fund investments in clean energy and electricity demand reduction for business
consumers, and support employment in the development of clean and renewable energy
technologies. In its recent regulations to limit greenhouse gas (GHG) emissions, California's Air
Resources Board designated a portion of allowances to be allocated to electric distribution
companies in order to mitigate potential electricity rate increases and their associated impacts.
Other states may encourage compliance methods by affected EGUs with particularly robust
5-1
-------
deployment of renewables, energy efficiency, or natural gas to promote manufacturing demand
or employment in those sectors. For example, energy efficiency investments may be targeted
towards reducing both electricity consumption and natural gas or heating oil consumption, such
as weatherization projects. The state plan approach and the composition of these programs will
influence the effects of compliance with the final rulemaking.
To estimate the costs, benefits, and impacts of implementing the CPP guidelines, the EPA
modeled two illustrative plan approaches: a rate-based approach and a mass-based approach.
Chapter 3 provides a description of the illustrative plan approaches analyzed. This chapter
provides a quantitative assessment of the energy price impacts for these illustrative approaches
and a qualitative assessment of the factors that will in part determine the timing and magnitude
of effects in other markets.
5.2 Methods
One potential quantitative approach to evaluating the secondary market impacts is to use a
computable general equilibrium (CGE) model. CGE models are able to provide aggregated
representations of the whole economy in equilibrium in the baseline and potentially with
regulation in place. As such, CGE model may be able to capture interactions between economic
sectors and provide information on changes outside of the directly regulated sector. In support of
previous rulemakings, such as the 2008 Final Ozone NAAQS (U.S. EPA 2008) and the 2010
Transport Rule proposal (U.S. EPA 2010), the EPA used the Economic Model for Policy
Analysis (EMPAX) CGE model to estimate the secondary market effects based on the cost
impacts projected by the Integrated Planning Model (IPM) for the directly regulated sector.
When considering the secondary market impacts of a regulation both the effects of the
costs, the benefits of improved air quality, and their interaction may be relevant. Therefore, in
the Second Prospective Analysis under Section 812 of the Clean Air Act Amendments the EPA
incorporated a set of health benefits arising from air quality improvement into the EMPAX CGE
model when studying the economy-wide impacts of the Clean Air Act (U.S. EPA 2011). While
the external Council on Clean Air Compliance Analysis (Council) review of this study stated that
inclusion of benefits in an economy-wide model "represented] a significant step forward in
benefit-cost analysis" (Hammitt 2010), the EPA recognizes that serious technical challenges
5-2
-------
remain when attempting to evaluate the benefits and costs of potential regulatory actions using
economy-wide models.
In light of these challenges, the EPA has established a Science Advisory Board (SAB)
panel on economy-wide modeling to consider the technical merits and challenges of using this
analytical tool to evaluate costs, benefits, and economic impacts in regulatory development. In
addition, EPA is asking the panel to identify potential paths forward for improvements that could
address the challenges posed when using economy-wide models to evaluate the effects of
regulations. The final panel membership was announced in March 2015 and the first of multiple
face-to-face meetings of the SAB panel has been scheduled for October 2015. The EPA will use
the recommendations and advice of this panel as an input into its process for improving benefit-
cost and economic impact analyses used to inform decision-making at the agency.
The advice from the Science Advisory Board (SAB) panel formed specifically to address
the subject of economy-wide modeling was not available in time for this final action. Given the
ongoing SAB panel on economy-wide modeling, the uncertain nature of the ultimate energy
price impacts due to the state flexibility in choosing a plan and the compliance flexibility for
affected EGUs, and the ongoing challenges of accurately representing costs, benefits, energy
efficiency improvements in economy-wide modeling, this chapter considers the energy impacts
associated with the illustrative plan approaches analyzed and a qualitative assessment of the
factors that will, in part, determine the timing and magnitude of effects in other markets.
5.3 Summary of Secondary Market Impacts of Energy Price Changes
Electricity, natural gas, and coal are important inputs to the production of other goods and
services. Therefore, changes in the price of these commodities will shift the production costs for
sectors that use electricity, natural gas, and coal in the production of other goods and services.
Changes in the types and levels of inputs used by producers in response to electricity and fuel
price changes may mitigate the production cost changes in these sectors. Such changes in
production costs may lead to changes in the quantities and/or prices of the goods or services
produced and changes in imports and exports.
The EPA used IPM to estimate electricity, natural gas, and coal price changes based on the
illustrative plan approaches modeled for this rule. IPM is a multi-regional, dynamic,
5-3
-------
deterministic linear programming model of the U.S. electric power sector described in more
detail in Chapter 3. The Retail Price Model (RPM) uses forecast changes in wholesale prices and
the cost of demand-side energy efficiency programs to forecast changes in average retail prices.
The prices are average prices over consumer classes and regions weighted by the amount used.
Table 5-1 shows these estimated price changes. For other results generated by IPM and the RPM,
please refer to Chapter 3.
There are many factors influencing the projected natural gas prices. IPM (and its integrated
gas resource and supply module) models natural gas reserves appropriate natural gas supplies
based on a multitude of factors. Since the model simulates perfect foresight, it anticipates future
demand for natural gas and responds accordingly. In addition, IPM (and the natural gas module)
are viewing a very long time horizon (through 2050), such that the impacts in certain years may
be responsive to other modeling assumptions or drivers. The modeling framework is
simultaneously solving for all of these key market and policy parameters (both electric and
natural gas), resulting in the impacts shown.
Table 5-1. Estimated Percentage Changes in Average Energy Prices by Energy Type for
the Final Emission Guidelines, Rate-based and Mass-based Illustrative Plan
Approaches
Rate-based
Electricity Price Change
Delivered Natural Gas Price Change
Delivered Coal Price Change
2020
3.2%
5.3%
-1.7%
2025
0.9%
-7.7%
-6.2%
2030
0.8%
2.5%
-8.0%
Mass-based
Electricity Price Change
Delivered Natural Gas Price Change
Delivered Coal Price Change
2020
3.0%
3.8%
-1.6%
2025
2.0%
-3.2%
-4.3%
2030
0.0%
-2.1%
-4.6%
For years when the price of electricity, natural gas, or coal increased, one would expect
decreases in production and increases in market prices in sectors for which these commodities
are inputs, ceteris paribus. Conversely, for years when prices of these inputs decreased, one
would expect increases in production and decreases in market prices within these sectors.
Smaller changes in input price changes would lead to smaller impacts within secondary markets.
However, a number of factors in addition to the magnitude and sign of the energy price changes,
5-4
-------
influence the magnitude of the impact on production and market prices for sectors using
electricity, natural gas, or coal as inputs to production. These factors are discussed below.
5.3.1 Share of Total Production Costs
The impact of energy price changes in a particular sector depends, in part, on the share of
total production costs attributable to those commodities. For sectors in which the directly
affected inputs are only a small portion of production costs, the impact will be smaller than for
sectors in which these inputs make up a greater portion of total production costs. Therefore, more
energy-intensive sectors would potentially experience greater cost increases when electricity,
natural gas, or coal prices increase, but would also experience greater reduced costs when these
input prices decrease.128
5.3.2 Ability to Substitute between Inputs to the Production Process
The ease with which producers are able to substitute other inputs for electricity, natural
gas, or coal, or even amongst those commodities, influences the impact of price changes for
these inputs. Those sectors with a greater ability to substitute across energy inputs or to other
inputs will be able to, at least partially, offset the increased cost of these inputs resulting in
smaller market impacts. Similarly, when prices for electricity, natural gas, or coal decrease, some
sectors may choose to use more of these inputs in place of other more costly substitutes.
5.3.3 Availability of Substitute Goods and Services
The ability of producers in sectors experiencing changes in their input prices to pass along
the increased costs to their customers in the form of higher prices for their products depends, in
part, on the availability of substitutes for the sectors' products. Substitutes may be either other
domestic products or foreign imports. If close substitutes exist, the demand for the product will
in general be more elastic and the producers will be less able to pass on the added cost through a
price increase.
128 The net direct effect of this rulemaking on the production costs of a sector that is attributable to a change in the
electricity price also depends on the expenditures the sector makes to reduce its demand for electricity under any
energy efficiency program that was adopted to achieve a state goal. That said, those expenditures may lead to other
reduced expenditures for the sector, such as reduced natural gas use from weatherization projects.
5-5
-------
Such substitution can also take place between foreign and domestic goods within the same
sector. Changes in the price of electricity, natural gas, and coal can influence the quantities of
goods imported or exported from sectors using these inputs. When the cost of domestic
production increases due to more expensive inputs, imports may increase as consumers substitute
towards relatively less costly foreign-produced goods. If imports increase because of a regulation
and those imports come from countries with higher emissions per unit of production, this can
result in foreign emission increases that offset some portion of domestic decreases, an effect
commonly referred to as "leakage." Alternatively, if those imports are less emissions-intensive
than the sectors that have contracted, emissions may fall even further. The potential for changes
in global pollutants such as carbon dioxide (CCh) and other GHG emissions is noteworthy.
Unlike most criteria pollutants and hazardous air pollutants, the impacts of COi emissions are not
affected by the location from which those emissions originate. A more complete evaluation of
the effect of this regulation on GHG emissions from other countries would account for whether
those countries have, or are expected to implement, policies affecting their GHG emissions. This
may include the potential that the present regulation could change the likelihood that other
countries will adopt policies affecting their GHG emissions.
5.4 Effect of Changes in Input Demand from Electricity Sector
Section 5.2 focuses on the effects of changes in energy prices, and possible responses to
those price changes, on sectors outside of the electricity sector. A change in demand for inputs in
the electricity sector, as well as changes in demand for energy efficiency services and products,
will also influence economic activity in other sectors of the economy. For example, there will be
changes in the demand for new generation sources such as natural gas combined cycle units and
renewables, and therefore sectors producing these technologies may expand. Therefore, while a
sector that produces say, wind turbine blades, may face higher natural gas and electricity prices,
production in that sector may ultimately increase due to higher demand from the electricity
sector for wind turbines.
5.5 Conclusions
Changes in the price of electricity, natural gas, and coal can affect markets for goods and
services produced by sectors that use these energy inputs in the production process. The direction
5-6
-------
and magnitude of these impacts are influenced by a number of factors. For example, the more
able producers in these sectors are to substitute away from the use of these energy inputs, the
smaller the effect of energy prices changes will be on their production cost. Changes in cost of
production may lead to changes in price, quantity produced, and profitability of firms within
secondary markets. Furthermore, the demand inputs in the electricity sector, as well as changes
in the demand for energy efficiency services and products, will also affect secondary markets. If
regulation results in changes in domestic markets that lead to an increase in imports, increases in
production in countries with more energy-intensive production may lead to changes in CO2
emissions elsewhere. The presence and adoption of policies affecting GHG emissions in other
countries, which may be influenced by the adoption of this final rule, may affect the change in
emissions elsewhere.
Modeling choices in IPM influence the forecast changes in electricity, natural gas, and coal
prices in this RIA. Actual market conditions, as will the plan approaches that states adopt, will
ultimately influence the price changes of these energy inputs and consequent effects on
secondary markets.
5.6 References
Hammitt, J.K. 2010. Review of the Final Integrated Report for the Second Section 812
Prospective Study of the Benefits and Costs of the Clean Air Act. Available at
Accessed
June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2008. Final Ozone NAAQS Regulatory
Impact Analysis. Office of Air Quality Planning and Standards, Research Triangle Park, NC.
Available at . Accessed June
4,2015.
U.S. Environmental Protection Agency (U.S. EPA). 2010. Regulatory Impact Analysis for the
Proposed Federal Transport Rule. Office of Air Quality Planning and Standards, Research
Triangle Park, NC. Available at
. Accessed June 4, 2015.
5-7
-------
U.S. Environmental Protection Agency (U.S. EPA). 2011. The Benefits and Costs of the Clean
Air Act from 1990 to 2020, Final Report - Rev A. Office of Air and Radiation, Washington,
DC. Available at .
Accessed June 4, 2015.
5-8
-------
CHAPTER 6: EMPLOYMENT IMPACT ANALYSIS
6.1 Introduction
Executive Order 13563 directs federal agencies to consider regulatory impacts on job
creation and employment. According to the Executive Order, "our regulatory system must
protect public health, welfare, safety, and our environment while promoting economic growth,
innovation, competitiveness, and job creation. It must be based on the best available science"
(Executive Order 13563, 2011). Although standard benefit-cost analyses have not typically
included a separate analysis of regulation-induced employment impacts,129 we typically conduct
employment analyses for economically significant rules. While the economy continues moving
toward full-employment, employment impacts are of particular concern and questions may arise
about their existence and magnitude. This chapter discusses and projects potential employment
impacts for the utility power, coal and natural gas production, and demand-side energy efficiency
sectors which may result from the Final Carbon Pollution Emission Guidelines for Existing
Stationary Sources: Electric Utility Generating Units (herein referred to as "final emission
guidelines" or the "Clean Power Plan Final Rule").130
Section 6.2 describes the theoretical framework used to analyze regulation-induced
employment impacts, discussing how economic theory alone cannot predict whether such
impacts are positive or negative. Section 6.3 presents an overview of the peer-reviewed literature
relevant to evaluating the effect of environmental regulation on employment. Section 6.4
provides background regarding recent employment trends in the electricity generation, coal and
natural gas extraction, renewable energy, and demand-side energy efficiency-related sectors.
Section 6.5 presents the EPA's quantitative projections of potential employment impacts in these
sectors. These projections are based in part on a detailed model of the electricity production
sector used for this regulatory analysis. Additionally, this section discusses projected
employment impacts due to demand-side energy efficiency activities. Section 6.6 offers several
conclusions.
129 Labor expenses do, however, contribute toward total costs in the EPA's standard benefit-cost analyses.
130 The employment analysis in this RIA is part of EPA's ongoing effort to "conduct continuing evaluations of
potential loss or shifts of employment which may result from the administration or enforcement of [the Act]"
pursuant to CAA section 321(a).
6-1
-------
6.2 Economic Theory and Employment
Regulatory employment impacts are difficult to disentangle from other economic changes
affecting employment decisions over time and across regions and industries. Labor market
responses to regulation are complex. They depend on labor demand and supply elasticities and
possible labor market imperfections (e.g., wage stickiness, long-term unemployment, etc.). The
unit of measurement (e.g., number of jobs, types of job hours worked, and earnings) may affect
observability of that response. Net employment impacts are composed of a mix of potential
declines and gains in different areas of the economy (e.g., the directly regulated sector, the
environmental protection sector, upstream and downstream sectors, etc.) over time. In light of
these difficulties, economic theory provides a constructive framework for analysis.
Microeconomic theory describes how firms adjust their use of inputs in response to
changes in economic conditions.131 Labor is one of many inputs to production, along with capital,
energy, and materials. In competitive markets, firms choose inputs and outputs to maximize
profit as a function of market prices and technological constraints.132'133 Berman and Bui (2001)
adapt this model to analyze how environmental regulations affect labor demand.134 They model
environmental regulation as effectively requiring certain factors of production, such as pollution
abatement capital, at levels that firms would not otherwise choose. Berman and Bui (2001)
model two components that drive changes in firm-level labor demand: output effects and
substitution effects.135 Regulation affects the profit-maximizing quantity of output by changing
the marginal cost of production. If a regulation causes marginal cost to increase, it will place
upward pressure on output prices, leading to a decrease in demand, and resulting in a decrease in
production. The output effect describes how, holding labor intensity constant, a decrease in
production causes a decrease in labor demand. As noted by Berman and Bui, although many
131 See Layard and Walters (1978), a standard microeconomic theory textbook, for a discussion, in Chapter 9.
132 See Hamermesh (1993), Ch. 2, for a derivation of the firm's labor demand function from cost-minimization.
133 In this framework, labor demand is a function of quantity of output and prices (of both outputs and inputs).
134 Morgenstern, Pizer, and Shih (2002) develop a similar model.
135 The authors also discuss a third component, the impact of regulation on factor prices, but conclude that this effect
is unlikely to be important for large competitive factor markets, such as labor and capital. Morgenstern, Pizer and
Shih (2002) use a very similar model, but they break the employment effect into three parts: 1) a demand effect; 2) a
cost effect; and 3) a factor-shift effect.
6-2
-------
assume that regulations must increases marginal cost, it need not be the case. A regulation could
induce a firm to upgrade to less polluting and more efficient equipment that lowers the marginal
cost of production. In such a case, output could increase after firms comply with the regulation.
For example, in the context of the current rule, improving the heat rate of utility boiler increases
fuel efficiency, lowering marginal production costs, and thereby potentially increasing the utility
boiler's generation. An unregulated profit-maximizing firm may not have chosen to install such
an efficiency-improving technology if the return on investment were too low, but once the
investment is required it lowers marginal production costs.
The substitution effect describes how, holding output constant, regulation affects the
labor-intensity of production. Although increased environmental regulation may increase use of
pollution control equipment and energy to operate that equipment, the impact on labor demand is
ambiguous. For example, equipment inspection requirements, specialized waste handling,
completing required paperwork, or pollution technologies that alter the production process may
affect the number of workers necessary to produce a unit of output. Berman and Bui (2001)
model the substitution effect as the effect of regulation on pollution control equipment and
expenditures required by the regulation and the corresponding change in the labor-intensity of
production.
In summary, as output and substitution effects may be positive or negative, economic
theory alone cannot predict the direction of the net effect of regulation on labor demand at the
level of the regulated firm. Operating within the bounds of standard economic theory, however,
empirical estimation of net employment effects on regulated firms is possible when methods and
data of sufficient detail and quality are available. The extant literature, however, illustrates
difficulties with empirical estimation. For example, there is a paucity of publicly-available data
on plant-level employment, thus most studies must rely on confidential plant-level employment
data from the U.S. Census Bureau, typically combined with pollution abatement expenditure
data, that are too dated to be reliably informative, or other measures of the stringency of
regulation. In addition, the most commonly used empirical methods, for example, Greenstone
(2002), likely overstate employment impacts because they rely on relative comparisons between
more regulated and less regulated counties, which can lead to "double counting" of impacts
6-3
-------
when production and employment shift from more regulated towards less regulated areas. Thus
these empirical methods cannot be used to estimate net employment effects.136.
The conceptual framework described thus far focused on regulatory effects on plant-level
decisions within a regulated industry. Employment impacts at an individual plant do not
necessarily represent impacts for the sector as a whole. The theoretical approach must be
modified when applied at the industry level.
At the industry-level, labor demand is more responsive if: (1) the price elasticity of
demand for the product is high, (2) other factors of production can be easily substituted for labor,
(3) the supply of other factors is highly elastic, or (4) labor costs are a large share of total
production costs.137 For example, if all firms in an industry are faced with the same regulatory
compliance costs and product demand is inelastic, then industry output may not change much,
and output of individual firms may change slightly.138 In this case, the output effect may be small,
while the substitution effect depends on input substitutability. Suppose, for example, that new
equipment for heat rate improvements requires labor to install and operate. In this case, the
substitution effect may be positive, and with a small output effect, the total effect may be
positive. As with potential effects for an individual firm, theory cannot determine the sign or
magnitude of industry-level regulatory effects on labor demand. Determining these signs and
magnitudes requires additional sector-specific empirical study. For environmental rules, much of
the data needed for these empirical studies is not publicly available, would require significant
time and resources in order to access confidential U.S. Census data for research, and also would
not be necessary for other components of a typical regulatory impact analysis (RIA).
In addition to changes to labor demand in the regulated industry, net employment impacts
encompass changes in other related sectors. For example, the final guidelines may increase
demand for heat rate improving equipment and services. This increased demand may increase
revenue and employment in the firms supporting this technology. At the same time, the regulated
136 See Greenstone (2002) p. 1212.
137 See Ehrenberg & Smith, p. 108.
138 This discussion draws from Berman and Bui (2001), pp. 293.
6-4
-------
industry is purchasing the equipment, and these costs may impact labor demand at regulated
firms. Therefore, it is important to consider the net effect of compliance actions on employment
across multiple sectors or industries.
If the U.S. economy is at full employment, even a large-scale environmental regulation is
unlikely to have a noticeable impact on aggregate net employment.139 Instead, labor in affected
sectors would primarily be reallocated from one productive use to another (e.g., from producing
electricity or steel to producing high efficiency equipment), and net national employment effects
from environmental regulation would be small and transitory (e.g., as workers move from one
job to another).140 Some workers may retrain or relocate in anticipation of new requirements or
require time to search for new jobs, while shortages in some sectors or regions could bid up
wages to attract workers. These adjustment costs can lead to local labor disruptions. Although
the net change in the national workforce is expected to be small, localized reductions in
employment may adversely impact individuals and communities just as localized increases may
have positive impacts.
If, on the other hand, the economy is operating at less than full employment, economic
theory does not clearly indicate the direction or magnitude of the net impact of environmental
regulation on employment; it could cause either a short-run net increase or short-run net decrease
(Schmalansee and Stavins, 2011). For example, the Congressional Budget Office considered
EPA's Mercury Air Toxics Standards and regulations for industrial boilers and process heaters as
potentially leading to short-run net increases in economic growth and employment, driven by
capital investments for compliance with the regulations (Congressional Budget Office, 2011). An
important research question is how to accommodate unemployment as a structural feature in
economic models. This feature may be important in assessing large-scale regulatory impacts on
employment (Smith, 2012).
139 Full employment is a conceptual target for the economy where everyone who wants to work and is available to
do so at prevailing wages is actively employed. The unemployment rate at full employment is not zero.
HO Arrow ej ai 1995; see discussion on bottom of p. 8. In practice, distributional impacts on individual workers can
be important, as discussed in later paragraphs of this section.
6-5
-------
Environmental regulation may also affect labor supply and productivity. In particular,
pollution and other environmental risks may impact labor productivity or employees' ability to
work.141 While the theoretical framework for analyzing labor supply effects is analogous to that
for labor demand, it is more difficult to study empirically. There is a small emerging literature
described in the next section that uses detailed labor and environmental data to assess these
impacts.
To summarize, economic theory provides a framework for analyzing the impacts of
environmental regulation on employment. The net employment effect incorporates expected
employment changes (both positive and negative) in the regulated sector and other related
sectors. Labor demand impacts for regulated firms, and also for the regulated industry, can be
decomposed into output and substitution effects which may be either negative or positive.
Estimation of net employment effects for regulated sectors is possible when data of sufficient
detail and quality are available. Finally, economic theory suggests that labor supply effects are
also possible. In the next section, we discuss the empirical literature.
6.3 Current State of Knowledge Based on the Peer-Reviewed Literature
The labor economics literature contains an extensive body of peer-reviewed empirical
work analyzing various aspects of labor demand, relying on the theoretical framework discussed
in the preceding section.142 This work focuses primarily on effects of employment policies such
as labor taxes and minimum wages.143 In contrast, the peer-reviewed empirical literature
specifically estimating employment effects of environmental regulations is growing, but is more
limited. In this section, we present an overview of the latter. As discussed in the preceding
section on theory, determining the direction of employment effects in regulated industries is
challenging because of the complexity of the output and substitution effects. Complying with a
new or more stringent regulation may require additional inputs, including labor, and may alter
the relative proportions of labor and capital used by regulated firms (and firms in other relevant
industries) in their production processes.
141 E.g. Graff Zivin and Neidell (2012).
142 Again, see Hamermesh (1993) for a detailed treatment.
143
See Ehrenberg & Smith (2000), Chapter 4: "Employment Effects: Empirical Estimates" for a concise overview.
6-6
-------
Empirical studies, such as Berman and Bui (2001), suggest that net employment impacts
due to regulation were not statistically different from zero in the regulated sector. Other research,
such as Greenstone (2002), suggests that more highly regulated counties may generate fewer jobs
than less regulated ones, but the methodology used likely overstates employment impacts
because it relies on relative comparisons between more regulated and less regulated counties,
which can lead to "double counting" of impacts when production and employment shift from
more regulated towards less regulated areas.144. Moreover, environmental regulations may affect
sectors that support pollution reduction earlier than the regulated industry. Rules are usually
announced well in advance of their effective dates and then typically provide a period of time for
firms to invest in technologies and process changes to meet the new requirements. When a
regulation is promulgated, the initial response of firms is often to order pollution control
equipment and services to enable compliance when the regulation becomes effective. Estimates
of short-term increases in demand for specialized labor within the environmental protection
sector have been prepared for several EPA regulations in the past, including the Mercury and Air
Toxics Standards (MATS).145 Overall, the peer-reviewed literature does not contain evidence that
environmental regulation has a large impact on net employment (either negative or positive) in
the long run across the whole economy.
6.3.1 Regulated Sector
Several empirical studies, including Berman and Bui (2001) and Ferris, Shadbegian, and
Wolverton (2014), suggest that regulation-induced net employment impacts may be zero or
slightly positive, but small in the regulated sector. Gray et al (2014) find that pulp mills that had
to comply with both the air and water regulations in EPA's 1998 "Cluster Rule" experienced
relatively small, and not always statistically significant, decreases in employment. Other research
on regulated sectors suggests that employment growth may be lower in more regulated areas
(Greenstone 2002, Walker 2011, 2013). However since these latter studies compare more
regulated to less regulated counties this methodological approach likely overstates employment
impacts to the extent that regulation causes plants to locate in one area of the country rather than
144 See Greenstone (2002) p. 1212.
145 U.S. EPA (201 Ib).
6-7
-------
another, which would lead to "double counting" of the employment impacts. List et al. (2003)
find some evidence that this type of geographic relocation may be occurring.
A small literature examines impacts of environmental regulations on manufacturing
employment. Greenstone (2002) and Walker (2011, 2013) study the impact of air quality
regulations on manufacturing employment, estimating the effects in non-attainment areas relative
to attainment areas. Kahn and Mansur (2013) study environmental regulatory impacts on
geographic distribution of manufacturing employment, controlling for electricity prices and labor
regulation (right to work laws). Their methodology identifies employment impacts by focusing
on neighboring counties with different air quality regulations. They find limited evidence that
environmental regulations may cause employment to be lower within "county-border-pairs."
This result suggests that regulation may cause an effective relocation of labor across a county
border, but since one county's loss is another's gain, such shifts cannot be transformed into an
estimate of a national net effect on employment. Moreover this result is sensitive to model
specification choices.
The few studies in peer-reviewed journals evaluating employment impacts of policies
that reduce CCh emissions in the electric power generation sector are in the European context. In
a sample of 419 German firms, 13 percent of which were in the electricity sector, Anger and
Oberndorfer (2008) find that the initial allocation of emission permits did not significantly affect
employment growth in the first year of the European Union (EU) Emissions Trading Scheme
(ETS). Examining European firms from 1996-2007, Commins et al. (2011) find that a 1 percent
increase in energy taxes is associated with a 0.01 percent decrease in employees in the electricity
and gas sector. Chan et al (2013) estimate the impact of the EU ETS on a panel of almost 6,000
firms in 10 European countries from 2005-2009. They find that firms in the power sector that
participated in the ETS had 2-3 percent fewer employees relative to those that did not participate,
but this effect is not statistically significant.
This literature suggests that the employment impacts of controlling CO2 emissions in the
European power sector were small. The degree to which these studies' results apply to the U.S.
context is unclear. European policies analyzed in these studies effectively put a price on
emissions of both existing and new sources either through taxes or tradable permits with an
emissions cap. An emission rate-based regulatory approach may not generate similar
6-8
-------
employment effects. Moreover, European firms face relative fuel prices and market regulatory
structures different from their U.S. counterparts, further complicating attempts to transfer
quantitative results from the EU experience to evaluate this rule.
6.3.2 Economy- Wide
As noted above it is very difficult to estimate the net national employment impacts of
environmental regulation. Given the difficulty with estimating national impacts of regulations,
EPA has not generally estimated economy-wide employment impacts of its regulations in its
benefit-cost analyses. However, in its continuing effort to advance the evaluation of costs,
benefits, and economic impacts associated with environmental regulation, EPA has formed a
panel of experts as part of EPA's Science Advisory Board (SAB) to advise EPA on the technical
merits and challenges of using economy-wide economic models to evaluate the impacts of its
regulations, including the impact on net national employment.146 Once EPA receives guidance
from this panel it will carefully consider this input and then decide if and how to proceed on
economy-wide modeling of employment impacts of its regulations.
EPA received several comments regarding the potential net national employment impact
of the proposed emission guidelines. Many of these comments referred to analyses that pre-dated
the Clean Power Plan proposal, or focused on only one component of the proposal.147 However,
one comment was based on an "economy-wide assessment of the employment impacts
associated with the U.S. Environmental Protection Agency's (EPA's) proposed Clean Power
Plan" using the Long-term Inter-industry Forecasting Tool (LIFT) model.148 The LIFT model,
which is from the Interindustry Forecasting Project (Inforum) at the University of Maryland, has
been used in the peer-reviewed academic literature149 and has also been used to examine the
we por father information see:
http://yosemite.epa.gOv/sab/sabproduct.nsf/0/07E67CF77B54734285257BB0004F87ED7OpenDocument
147 See, for example, comments EPA-HQ-OAR-2013-0602-6743 andEPA-HQ-OAR-2013-0602-23140, within
Docket ID: EPA-HQ-OAR-2013-0602.
148 See comment EPA-HQ-OAR-2013-0602-22960, within Docket ID: EPA-HQ-OAR-2013-0602.
149 See, for example, Almon (1991) and McCarthy (1991).
6-9
-------
economic impacts of other national policies [Meade (2009); Werling (2011)].150 The commenter
noted that "While EPA's analysis provides a reasonable first approximation of the proposed
rule's employment effects, its focus on direct employment impacts does not capture various
indirect employment impacts that may be of interest to policymakers and the public." [...]
"These include the employment impact associated with changes in electricity and other energy
prices (both positive and negative, depending on the year), the productivity impacts associated
with heat rate improvements at power plants, households and businesses re-directing
expenditures to other uses because of increased demand-side energy efficiency, expenditures
crowded out by energy efficiency expenditures, and changes in investments for air pollution
control devices."
As mentioned previously, EPA is currently engaged in an SAB process on economy-wide
modeling. EPA will not make any determinations on whether modeling the economy-wide
impacts of its regulations - including employment impacts - is feasible and, if so, how and when
to do this until it receives guidance from the SAB panel. While the purpose of the SAB process
is not to peer review any particular economy-wide model, it is worth noting that the use of
models such as LIFT may be addressed by one of the charge questions to the SAB: "Are there
other economy-wide modeling approaches that EPA could consider in conjunction with CGE
models to evaluate the short run implications of an air regulation (e.g., macro-economic,
disequilibrium, input/output models)? What are the advantages or disadvantages of these
approaches?"151
150 The commenter provides the following description of the LIFT model: "LIFT is a 97-sector dynamic
representation of the U.S. national economy. The model combines an interindustry input / output (I-O) formulation
with extensive use of regression analysis to employ a 'bottom-up' approach to macroeconomic modeling. That is,
the model works like the actual economy, building macroeconomic totals from details of industry activity, rather
than distributing predetermined macroeconomic quantities among industries. The commenter also describes LIFT
also captures interactions between industries across the economy, enabling the model to gauge how changes in
prices, investment, or productivity in one industry cascade across the economy. In the context of the Clean Power
Plan, this is an important feature for understanding how the rule's direct impacts for the electric power sector affect
other industries."
151 See p. 10,
http://yosemite.epa.gov/sab/sabproduct.nsf/0/07E67CF77B54734285257BB0004F87ED/$File/Charge+Questions+2
-26-15.pdf
6-10
-------
6.3.3 Labor Supply Impacts
The empirical literature on environmental regulatory employment impacts focuses
primarily on labor demand. However, there is a nascent literature focusing on regulation-induced
effects on labor supply.152 Although this literature is limited by empirical challenges, researchers
have found that air quality improvements lead to reductions in lost work days (e.g., Ostro, 1987).
Limited evidence suggests worker productivity may also improve when pollution is reduced.
Graff Zivin and Neidell (2012) used detailed worker-level productivity data from 2009 and 2010,
paired with local ozone air quality monitoring data for one large California farm growing
multiple crops, with a piece-rate payment structure. Their quasi-experimental structure identifies
an effect of daily variation in monitored ozone levels on productivity. They find "ozone levels
well below federal air quality standards have a significant impact on productivity: a 10 parts per
billion (ppb) decreases in ozone concentrations increases worker productivity by 5.5 percent."
(Graff Zivin and Neidell, 2012, p. 3654).153
This section (section 6.3) has outlined the challenges associated with estimating
regulatory effects on both labor demand and supply for specific sectors. These challenges make
it difficult to estimate net national employment estimates that would appropriately capture the
way in which costs, compliance spending, and environmental benefits propagate through the
macro-economy.
6.4 Recent Employment Trends
The U.S. electricity system includes employees that support electric power generation,
transmission and distribution; the extraction of fossil fuels; renewable energy generation; and
supply-side and demand-side energy efficiency. This section describes recent employment trends
in the electricity system.
152
For a recent review see Graff-Zivin and Neidell (2013).
153 The EPA is not quantifying productivity impacts of reduced pollution in this rulemaking using this study. In light
of this recent research, however, the EPA is considering how best to incorporate possible productivity effects in the
future.
6-11
-------
6.4.1 Electric Power Generation
In 2014, the electric power generation, transmission and distribution sector (NAICS
2211) employed about 390,000 workers in the U.S.154 Installation, maintenance, and repair
occupations accounted for the largest share of workers (25 percent).155 These categories include
inspection, testing, repairing and maintaining of electrical equipment and/or installation and
repair of cables used in electrical power and distribution systems. Other major occupation
categories include office and administrative support (18 percent), production occupations (16
percent), architecture and engineering (10 percent), business and financial operations (7 percent)
and management (7 percent). As shown in Figure 6.1, employment in the Electric Power
Industry averaged about 420,000 workers 2000 to 2005, declining to an average of about
400,000 workers for the rest of the decade, and then declining to about 390,000 workers in 2014.
Power Generation and Supply Employment
(NAICS = 2211, Annual Average, 1000s of Employees)
500
450
400
350
300
250
200
150
100
50
0
Figure 6.1. Electric Power Industry Employment
154 U.S. Bureau of Labor Statistics. "Current Employment Survey Seasonally Adjusted Employment for Electric
Power Generation, Transmission, and Distribution (national employment)." Series ID: CES4422110001. Available
at . Accessed June 9, 2015.
155 U.S. Bureau of Labor Statistics, Occupational Employment Statistics, May 2014 National Industry-Specific
Occupational Employment and Wage Estimates, Electric Power Generation, Transmission, and Distribution (NAICS
2211). Available at: .
6-12
-------
6.4.2 Fossil Fuel Extraction
6.4.2.1 Coal Mining
The coal mining sector (NAICS 2121) is primarily engaged in coal mining and coal mine
site development, excluding metal ore mining and nonmetallic mineral mining and quarrying. In
2014, BLS reported about 74,000 coal mining employees (Figure 6.2). During the 2000 to 2014,
period, coal mining employment peaked in 2011 at about 87,000 employees.
Coal Mining Employment
(NAICS = 2121, Annual Average, 1000s of Employees)
100
90
80
70
60
50
40
30
20
10
0
'V?
Figure 6.2. Coal Production Employment
6.4.2.2 Oil and Gas Extraction
In 2014, there were close to 200,000 employees in the oil and gas extraction sector
(NAICS 2II).156 This sector includes production of crude petroleum, oil from oil shale and oil
sands, production of natural gas, sulfur recovery from natural gas, and recovery of hydrocarbon
liquids. Activities include the development of gas and oil fields, exploration activities for crude
petroleum and natural gas, drilling, completing, and equipping wells, and other production
156 BLS, Current Employment Statistics. Seasonally adjusted employment for oil and gas extraction (national
employment), NAICS 211. Series ID: CES1021100001. Available at . Accessed June 9,
2015.
6-13
-------
activities.157 In contrast with coal, and looking at Figure 6.3, there has been a sharp increase in
employment in this sector over the past decade.
Tern
9nn
1 en
4
1 nn
Rn
.
6-14
-------
efficient technology, workers that supply inputs and technicians who install service or operate
equipment. As such, there are a variety of definitions of clean or green jobs used, some more
expansive than others.
6.4.3.1 Defining Clean Energy Jobs
Two U.S. Government sources, the 2010 Department of Commerce (DOC) report,
Measuring the Green Economy and the 2010 and 2011 BLS Green Goods and Services surveys
have subdivided industrial classifications into "green" categories. In both cases the approach was
to determine which product classifications, rather than industries, were green. They multiplied
green production by product revenue and defined an industrial sector as green if it met a
threshold of green revenue as a proportion of total revenue.
DOC broadly defined green jobs in 2010 as those "created and supported in businesses
that produce green products and services."158 They further classified green jobs into a broad and a
narrow category. The narrow category includes only products deemed to be green without
disagreement, while the broad category is more inclusive definition of green goods and services
to over 22,000 product codes in the 2007 Economic Census to estimate their contribution to the
U.S. economy. The report found that the number of green jobs in 2007 ranged from 1.8 million
to 2.4 million jobs, accounting for between 1.5 and 2 percent of total private sector
employment.159
BLS used an expansive definition of clean or green jobs in 2010 and 2011. It goes
beyond direct clean energy-related investments and includes "those in businesses that produce
goods and provide services that benefit the environment or conserve natural resources. These
goods and services, which are sold to customers, include research and development, installation,
and maintenance services for renewable energy and energy efficiency and education and training
related to green technologies and practices" but also include recycling and natural resource
158 U.S. Department of Commerce Economics and Statistics Administration. 2010. "Measuring the Green
Economy," April. Available at:
.
159 U.S. Department of Commerce Economics and Statistics Administration. 2010. "Measuring the Green Economy,
April. Available at: .
6-15
-------
conservation, such as forestry management.160 Based on surveys across the 325 industries it
identified as potential producers of green goods and services, BLS counts approximately 2.3
million jobs in the green economy in 2010, rising 7.4 percent to 2.5 million in 2011,161 compared
to increases of about one percent across all occupations in the entire economy over the same
period.162 The table below, Table 6-1, presents BLS green job estimates nationally and for the
utility sector.
Table 6-1. U. S. Green Goods and Services (GGS) Employment (annual average)
Total GGS Utility GGS Total GGS Growth Utility GGS Growth
Employment Employment 2010-11 2010-11
2010 2,342,562 69,031 NA NA
2011 2,515,200 71,129 7.4% 3.0%
Source: Bureau of Labor Statistics
6.4.3.2 Renewable Electricity Generation Employment Trends
The DOC report does not separate renewable energy data and the BLS data include only
privately owned electricity generating facilities. As such, neither source isolates renewable
electricity generation employment. For historical trends in this sector, we therefore, rely on a
Brookings Institution study, Muro et al. (2011). This study built a national database of "clean
economy" jobs from the bottom up, verifying each company individually.163 They include a list
of categories similar but not identical to that of BLS, including agricultural and natural resources
conservation, education and compliance, energy and resource efficiency, greenhouse gas
reduction, environmental management and recycling, and renewable energy. This study found
about 138,000 jobs in the renewable energy sector in 2010, with an overall average annual
growth rate of 3.1 percent from 2003-2010. Table 6-2 details the national results by energy
source.
160 BLS has identified 325 detailed industries (6-digit NAICS) as potential producers of green goods and services.
Available at: . (Accessed on 1-14-14, last modified date: March 19,
2013.
161 U.S. Department of Labor, U.S. Bureau of Labor Statistics, (n.d.). 2011. "Green Goods and Services 2010-2011.'
(Retrieved on January 14, 2014). Available at :< http://www.bls.gov/ggs/ggsoverview.htm>.
162U.S. Department of Labor, U.S. Bureau of Labor Statistics. 2010. National Occupational Employment and Wage
Estimates, United States. Available at: , May. National
Occupational Employment and Wage Estimates, United States http://www.bls.gov/oes/2010/may/oes_nat.htm.
163 p. 15.
6-16
-------
Table 6-2. Renewable Electricity Generation-Related Employment
Sector
B iofuels/B iomas s
Geothermal
Hydropower
Renewable Energy Services
Solar Photovoltaic
Solar Thermal
Waste-to-Energy
Wave/Ocean Power
Wind
Total
Jobs, 2010
20,680
2,720
55,467
1,981
24,152
5,379
3,320
371
24,294
138,364
2003-2010 Average Annual Growth Rate (%)
8.9
6.7
-3.6
6.3
10.7
18.4
3.7
20.9
14.9
3.1
Source: http://www.brookings.edU/~/media/Series/resources/0713_clean_economy.pdf, Appendix A.
More recent industry data, from 2014, indicate higher employment numbers and growth in the
solar sector.164
6.4.3.3 Employment Trends in Demand-Side Energy Efficiency Activities
U.S. government data used for calculating the historical trends in the demand-side energy
efficiency sector come from the BLS green goods and services surveys. BLS reports an energy
efficiency category, finding 1.49 million private sector energy efficiency jobs in 2010 and 1.64
million in 2011.
In addition to the "clean energy" jobs, Muro et al (2011) found about 428,000 jobs in the
Energy and Resource Efficiency sector in 2010, with an overall average annual growth rate of
2.6 percent from 2003-2010.165 Table 6-3 details the results by energy sector.
164 The Solar Foundation, National Solar Jobs Census 2014. < http://www.thesolarfoundation.org/national-solar-
jobs-census-2014/>
165 p. 15.
6-17
-------
Table 6-3. Energy and Resources Efficiency-Related Employment
Sector
Appliances
Energy-saving Building Materials
Energy-saving Consumer Products
Green Architecture and Construction Services
HVAC and Building Control Systems
Lighting
Professional Energy Services
Smart Grid
Total
Jobs, 2010
36,608
161,896
19,210
56,190
73,600
14,298
49,863
15,987
427,652
2003-2010 Average Annual
Growth Rate (%)
-3.1
2.5
-2.9
6.4
3.3
-1.8
6.9
8.6
2.6
Source: http://www.brookings.edu/~/media/Series/resources/0713_clean_economy.pdf, Appendix A
In addition, other research institutes and industry groups have clean economy or clean
energy employment databases. While definitions and timeframes vary, all show positive
employment trends of 1.9 percent or more growth in clean energy-related jobs annually.
6.5 Projected Sectoral Employment Changes due to the Final Emission Guidelines
Affected EGUs may respond to these final COi emission performance rates by placing
new orders for efficiency-related or renewable energy equipment and services to reduce GHG
emissions. Implementing the CPP Final Rule will involve changes in the amount of labor needed
in different parts of the utility power sector. Installing and operating new equipment or
improving heat rate efficiency could increase labor demand in the electricity generating sector
itself, as well as associated equipment and services sectors. Specifically, the direct employment
effects of supply-side initiatives include increases in labor demand during the implementation
phase for manufacturing, installing, and operating higher efficiency and renewable energy
electricity generating assets, as well as making heat rate improvements at existing fossil units.
Additional supply-side direct employment impacts are the reductions in labor demand for labor
that would have been used by less efficient or higher emitting generating assets. Once
implemented, increases in operating efficiency and shifting generation to existing or new NGCC
units and renewable energy generation will impact the utility power sector's demand for fossil
fuels and plans for EGU retirement and new construction.
In addition, EPA expects state plans may also include demand-side energy efficiency
policies and programs that typically change energy consumption patterns of business and
residential consumers by reducing the quantity of energy required for a given level of production
or service. Demand-side initiatives generally aim to increase the use of cost-effective energy
6-18
-------
efficiency technologies (e.g., including more efficient appliances and air conditioning systems,
more efficient lighting devices, more efficient design of homes and businesses), and advance
efficiency improvements in motor systems and other industrial processes. Demand-side
initiatives can also directly reduce energy consumption, such as through programs encouraging
changing the thermostat during the hours a building is unoccupied or motion-detecting room
light switches. Such demand-side energy efficiency initiatives directly affect employment by
encouraging firms and consumers to shift to more efficient products and processes than would
otherwise be the case. Employment in the sectors that provide these more efficient devices and
services would be expected to increase, while employment in the sectors that produce less
efficient devices would be expected to contract.
This generation-side employment analysis uses the cost projections from the engineering-
based Integrated Planning Model (IPM) to project labor demand impacts of the final emission
guidelines on affected EGUs in the electricity power sector and the fuel production sector (coal
and natural gas). These projections include effects attributable to heat rate improvements,
construction of new EGUs, generation shifts, changes in fuel use, and reductions in electricity
generation due to demand-side energy efficiency activities. To project labor requirements for
demand-side energy efficiency activities, the analysis uses a different approach that combines
data on historic changes in employment and expenditures in the energy efficiency sector with
projected changes in expenditures in the sector arising from state implementation of the emission
guidelines. We project labor impacts for the rate-based and mass-based illustrative plan
approach.
6.5.1 Projected Changes in Employment in Electricity Generation and Fossil Fuel Extraction
The analytical approach used in this analysis is a bottom-up engineering method
combining EPA's cost analysis of the emission guidelines with data on labor productivity,
engineering estimates of the amount and types of labor needed to manufacture, construct, and
operate different types of generating units, and prevailing wage rates for skilled and general
labor categories. This approach is different from the economy-wide types of economic analyses
discussed in section 6.2. Lacking robust peer-reviewed methods to estimate economy-wide
impacts, the engineering-based analysis focuses on the supply-side direct impact on labor
demand in industries closely involved with electricity generation. The engineering approach
6-19
-------
projects labor changes measured as the change in each analysis year in job-years166 employed in
the utility power sector and directly related sectors (e.g., equipment manufacturing, fuel supply,
EGU construction and generating efficiency services). For example, this approach projects the
amounts and types of labor required to implement improvements in generating efficiency. The
generation efficiency improvements reduce the amount of fossil fuel needed. The efficiency-
driven change in fuel use is included in the estimates of the CPP's impact on the overall changes
in labor required to extract fossil fuels. Some of the quantified employment impacts in this
analysis are one-time impacts, such as changes associated with building new NGCC or
renewable generating units. Other labor impacts will continue, such as changes associated with
operating and maintaining generating units that will be retired, and labor providing the fuel
supplied to newly built, retired and improved EGUs.
This analysis relies on projections and the cost analysis from IPM, which uses industry-
specific data and assumptions to estimate costs and energy impacts of the final guidelines (see
Chapter 3). The EPA uses IPM to predict coal generating capacity that is likely to undertake
improvements in heat rate efficiency (HRI).167 IPM also predicts the guidelines' impacts on fuel
use, retirement of existing units, and construction of new ones.
The methods EPA uses to estimate the labor impacts are based on the analytical methods
used in many previous EPA regulatory analyses. The most relevant prior analysis was the
Regulatory Impact Analysis for the Mercury and Air Toxics Standards (MATS). While the
methods used in this analysis to estimate the recurring labor impacts (e.g., labor associated with
operating and maintaining generating units, as well as labor needed to mine coal and natural gas)
are the same as we used in MATS (with updated data where available), the methods used to
estimate the labor associated with installing new capacity and implementing heat rate
improvements were developed for the purpose of the Clean Power Plan RIA.
166 Job-years are not individual jobs, but rather the amount of work performed by the equivalent of one full-time
individual for one year. For example, 20 job-years in 2020 may represent 20 full-time jobs or 40 half-time jobs in
that year.
167 HRI could include a range of activities in the power plant to lower the heat rate required to generate a net
electrical output. Assuming all other things being equal, a lower heat rate is more efficient because more electricity
is generated from each ton of coal.
6-20
-------
The bottom-up engineering-based labor analysis in the MATS RIA primarily was
concerned with the labor needs of retrofitting pollution control equipment. A central feature of
the supply-side labor analysis for this RIA, however, involves the quantity and timing of the
labor needs of building new renewable and NGCC units and retiring coal units. The estimated
response of the utility power sector involves changes in the amount and timing of retirements of
existing coal and oil/gas units, as well as changes in the amount and timing of building new
NGCC units and renewable generating capacity. In addition to the changes in retirements and
construction of new units, there are also estimated changes in the utilization of existing
generating units, and changes in the gas and coal supply sectors.
For example, as presented in Chapter 3, the IPM analysis of the rate-based illustrative
plan approach scenario finds that in 2025 (part way through the 2022-2029 interim plan
performance period) less total generating capacity is needed than in the base case. The estimated
reduction in capacity by 2025 with the rate-based scenario is 49.4 GW less than the estimated
base case capacity (a 4.8 percent net capacity reduction). This 49.4 GW net reduction includes
more retirements of coal units (an additional 22.9 GW of coal-fired capacity retired) and oil/gas
steam units (an additional 9.3 GW of oil/gas retirement) compared to the base case, as well as a
reduction in the amount of new natural gas units needed to be built by 2025 (a decrease of 10.9
GW in new capacity from the amount forecast in the base case) and non-hydro renewables (1.7
GW less renewable capacity built). Fossil fuel utilization will also be impacted. The rate-based
scenario finds that in 2025 less coal will be used (102.9 million tons less, a 14.1 percent decrease
from the base case) and less gas will also be used (0.1 TCP less, a 1.0 percent decrease).
An important aspect of the labor analysis is that building new units, and all the associated
construction-related labor, occurs before the new units become operational. While the financial
costs of building the new units are amortized and recouped over the book life of the new
equipment, the labor involved with manufacturing equipment and constructing the new units
occurs, and is actually paid for, in a concentrated amount of time before the new capacity begins
to generate electricity. IPM assumes168 that new NGCC units take 3 years to build, and both
natural gas combustion turbines and wind-powered renewables take 2 years.
1 Table 4.13, IPM 5.13 Documentation.
6-21
-------
Avoiding some of the need for new capacity due to both demand and supply efficiency
improvements results in both a significant cost savings to consumers and the power sector, as
well as reduced emissions of both CO2 and non-COi pollutants from fossil fuel-fired generation.
The avoided new capacity, however, also has significant labor impacts. A portion of the labor
that would have been used to build the new capacity in the base case will not be employed in the
power generation sector with the implementation of the GHG guidelines, though it likely will be
employed in construction elsewhere. Similarly, less labor involved with operating and providing
fuel for new units will be needed with the emissions guidelines than in the base case.
A critical component of the overall labor impacts of implementing the GHG guidelines is
the impact of the labor associated with the demand-side energy efficiency activities. The
demand-side labor impacts are presented in section 6.5.2. The demand-side energy efficiency
activities are increases in labor needs and estimated in units of jobs, while the supply-side
employment impacts are estimated as job-years. The IPM labor expenditure projections are
distributed across different labor categories (e.g., general construction labor, boilermakers and
engineering) using data from engineering analyses of labor's overall share of total expenditures,
and apportionment of total labor cost to various labor categories. Hourly labor expenditures
(including wages, fringe benefits, and employer-paid costs including taxes, insurance and
administrative costs) for each category are used to estimate the labor quantity (measured in full-
time job-years) consistent with the compliance scenario projections. Projected labor impacts
arising from changes in fuel demand are primarily derived from labor productivity data for coal
mining (tons mined per employee hour) and natural gas extraction (MMBtu produced/job-year).
Tables 6.4 and 6.5 present projected changes relative to the baseline of four labor categories:
1. manufacturing, engineering and construction for building, designing and implementing
heat rate improvements;
2. manufacturing and construction for new generating capacity;
3. operating and maintenance for existing generating capacity; and
4. extraction of coal and natural gas fuel.
All of the employment estimates presented in Tables 6-4 and 6-5 are estimates occurring
in a single year. For the construction-related (one-time) labor impacts, including the installation
of HRI, Tables 6-4 and 6-5 present the average annual impact occurring in each year of three
6-22
-------
different intervals. The three intervals are from 2018 through 2020 (a three year interval), during
which there are modest labor impacts from the early changes in the power utility sectors
operations, from 2021 through 2025 (five years), and 2026 through 2030 (5 years). The
construction-related labor analysis are based on the IPM estimates of the net change in capital
investment that occurs during each multi-year interval to fund building new units completed
during that interval. The new build labor analysis uses the net change in capital investment to
estimate the amount and type of labor needed during the interval to build the new capacity. The
analysis assumes that the new build labor within each interval is evenly distributed throughout
the interval. Tables 6-4 and 6-5 reflect this assumption by presenting the average labor utilization
per year during each of the three intervals.
The HRI-related labor impacts are estimated based on the assumed capital cost of
$100/kw (see section 3.9.3). The labor estimates for operating and maintaining generating units
annually are based on IPMs estimates of Fixed Operating and Maintenance (FOM) and Variable
Operating and Maintenance (VOM) costs. IPM estimates FOM and VOM for each year
individually, so the net changes in O&M-related labor estimates in Tables 6-4 and 6-5 are single
year estimates for 2020, 2025 and 2030. These single year O&M labor estimates are not merely
the average annual averages labor needs throughout each multi-year interval. There are O&M
labor changes occurring in the all years throughout the entire period 2020-2030, but the labor
impacts in each labor category change each year. The fuel-related labor estimates are also single-
year estimates, and not multi-year averages. The labor analysis of the impacts on the fuel
extraction industries uses IPM's estimates of the net changes in the amount of coal and natural
gas in 2020, 2025 and 2030, which are inherently estimates of the fuel usage in a single year. As
with the O&M labor impacts, the fuels-related labor impacts occur in every year throughout
2020-2030, and the labor impact changes every year.
It should be noted that the supply-side labor impact estimates in Tables 6-4 and 6-5
reflect the supply-side changes that will potentially occur with each illustrative plan scenario.
These labor impacts include not only the direct supply-side impacts of the illustrative
implementation scenarios of the CPP, but also the changes in total generation activity that result
from the demand-side energy efficiency activities expected to be an important component of
state compliance strategies. The additional labor impact estimates from demand-side energy
efficiency activities are presented below in section 6.5.2.
6-23
-------
More details on methodology, assumptions, and data sources used to estimate the supply-
side labor impacts discussed in this section can be found in Appendix 6A.
Table 6-4. Engineering-Based3 Changes in Labor Utilization, Rate-based Scenario
(Number of Job-Yearsb of Employment in a Single Year)
Construction-related (One-time) Changes*
Heat Rate Improvement: Total
Boilermakers and General Construction
Engineering and Management
Equipment-related
Material-related
New Capacity Construction: Total
Renewables
Natural Gas
Operation and Maintenance: Total
Changes in Renewables
Changes in Gas***
Changes in Coal***
Retired Oil and Gas
Fuel Extraction: Total
Coal
Natural Gas
Supply-Side Employment Impacts - Quantified
2018-2020
0
0
0
0
0
500
700
-200
2020
-9,100
600
300
-8,000
-2,000
100
-1,300
1,400
-8,500
2021-2025
15,400
11,000
2,800
1,200
400
-15,600
-5,000
-10,600
Recurring Changes**
2025
-17,000
-100
-1,100
-13,300
-2,500
-7,800
-7,300
-500
-25,000
2026-2030
2,200
1,600
400
200
0
400
23,300
-22,900
2030
-19,600
1,100
-3,700
-14,700
-2,300
-13,900
-13,300
-600
-30,900
a Job-year estimates are derived from IPM investment and O&M cost estimates, as well as IPM fuel use estimates
(tons coals or MMBtu gas).
b All job-year estimates on this are full-time equivalent (FTE) jobs. Job estimates in the demand-side energy
efficiency section (below) include both full-time and part-time jobs.
*Construction-related job-year changes are one-time impacts, occurring during each year of the multi-year period
during which construction and HRI installation activities occur. Construction-related figures in table are the average
annual job-years in each year between the years in the range. Negative construction job-year estimates occur when
additional generating capacity must be built in the base case, but is avoided in the final rule.
**Recurring 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.
***O&M job-year changes include changes from new, retired and avoided capacity, and also changes arising from
changes in utilization (i.e., capacity factor changes) of existing EGU capacity that continue to operate but generate a
different amount of MWh/year.
6-24
-------
Table 6-5. Engineering-Based3 Changes in Labor Utilization, Mass-Based Illustrative Plan
Approach (Number of Job-Years of Employment in a Single Year)
Construction-related (One-time) Changes*
Heat Rate Improvement: Total
Boilermakers and General Construction
Engineering and Management
Equipment-related
Material-related
New Capacity Construction: Total
Renewables
Natural Gas
Operation and Maintenance: Total
Changes in Renewables
Changes in Gas***
Changes in Coal***
Retired Oil and Gas
Fuel Extraction: Total
Coal
Natural Gas
Supply-Side Employment Impacts - Quantified
2018-2020
0
0
0
0
0
-1,700
-1,400
-300
2020
-11,700
-900
500
-9,100
-2,200
200
-1,800
2,000
-13,100
2021-2025
14,900
10,700
2,700
1,200
300
-11,100
-3,500
-7,600
Recurring Changes**
2025
-21,200
-1,000
-400
-17,000
-2,800
-8,600
-8,700
100
-26,000
2026-2030
800
600
100
100
0
4,700
21,300
-16,600
2030
-25,000
700
-2,300
-20,800
-2,600
-14,300
-12,200
-2,100
-33,700
a Job-year estimates are derived from IPM investment and O&M cost estimates, as well as IPM fuel use estimates
(tons coals or MMBtu gas).
b All job-year estimates on this are full-time equivalent (FTE) jobs. Job estimates in the demand-side energy
efficiency section (below) include both full-time and part-time jobs.
*Construction-related job-year changes are one-time impacts, occurring during each year of the multi-year period
during which construction and HRI installation activities occur. Construction-related figures in table are the average
annual job-years in each year between the years in the range. Negative construction job-year estimates occur when
additional generating capacity must be built in the base case, but is avoided in the final rule.
**Recurring 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.
***O&M job-year changes include changes from new, retired and avoided capacity, and also changes arising from
changes in utilization (i.e., capacity factor changes) of existing EGU capacity that continue to operate but generate a
different amount of MWh/year.
6.5.2 Projected Changes in Employment in Demand-Side Energy Efficiency Activities
As described in Chapter 3, EPA anticipates that this rule may stimulate investment in
clean energy technologies and services, resulting in considerable increases in energy efficiency.
6-25
-------
Many of these investments may support demand-side energy efficiency activities such as:
reducing energy required for a given activity by encouraging more efficient technologies (e.g.,
ENERGY STAR appliances), implementing energy improvements for existing systems (e.g.,
weatherization of older homes), or encouraging changes in behavior (e.g., reducing air
conditioning during periods of high electricity demand).We expect these increases in energy
efficiency, specifically, to support a significant number of jobs in related industries. For more
information on EPA's illustrative investment levels in demand-side energy efficiency activities,
assumed to be adopted in response to the CPP, please see Section 3.7 "Demand-Side Energy
Efficiency" in Chapter 3 of this RIA.
In this section, we project employment impacts in demand-side energy efficiency
activities arising from these guidelines using illustrative calculations. The approach uses
information from power sector modeling and projected impacts on energy efficiency investments
analyzed (see Chapter 3), and U.S. government data on employment and expenditures in energy
efficiency.169 This approach is limited by the fact that we do not know which options states will
choose for demand-side energy efficiency activities and by uncertainties associated with
methods. These illustrative employment projections are gross; thus they do not include impacts
in other sectors of any shift in resources from other sectors to implement the demand-side energy
efficiency activities. Nor does this analysis attempt to quantify the positive employment impacts
in other sectors arising from changes in consumer expenditures on electricity due to reduced
electricity bills. In other words, these projections are not attempts at estimating net national job
creation. Also, this approach attempts to calculate the number of employees (full-time and part-
time) rather than full-time supply-side job-years estimated in section 6.5.1.
Employment impacts of demand-side energy efficiency programs have not been
extensively studied in the peer-reviewed, published economics literature. Instead, most research
has focused on consumer response to and amount of energy savings achieved by these programs
(e.g., Allcott (201 la, 201 Ib), Arimura et al. (2012)). Results suggest that demand-side energy
169 Investments in demand-side energy efficiency reduce energy required for a given activity by encouraging more
efficient technologies (e.g., ENERGY STAR appliances), implementing energy improvements for existing
systems (e.g., weatherization of older homes), or encouraging changes in behavior (e.g., reducing air
conditioning during periods of high electricity demand).
6-26
-------
efficiency programs reduce energy use and generate small increases in consumer welfare. These
policy impacts are due to low investment in energy efficiency as described in "energy paradox"
literature (Gillingham, Newell, and Palmer (2009), Gillingham and Palmer (2014)).170
Two recent articles discuss employment effects of demand-side energy efficiency
programs. Aldy (2013) describes clean energy investments funded by the American Recovery
and Reinvestment Act of 2009, which "included more than $90 billion for strategic clean energy
investments intended to promote job creation and the deployment of low-carbon technologies"
(p. 137), with nearly $20 billion for energy efficiency investments. The Council of Economic
Advisors (CEA) (2011) estimated higher economic activity and employment than would have
otherwise occurred without the American Recovery and Reinvestment Act. Using CEA's
methods to quantify job creation for the Recovery Act, Aldy uses the share of stimulus funds for
clean energy investments to estimate job-years supported by the Recovery Act. The largest
sources of job creation in clean energy are those that received the largest shares of stimulus
funds: renewable energy, energy efficiency, and transit. Aldy's estimates, while informative, are
not directly applicable for employment analysis in this rulemaking as there are important
differences in expected employment impacts from a historically large fiscal stimulus specifically
targeting job creation during a period of exceptionally high unemployment versus environmental
regulations taking effect several years from now.
Yi (2013) analyzes clean energy policies and employment for U.S. metropolitan areas in
2006, prior to the Recovery Act, to evaluate impacts on clean energy job growth. Implementing
an additional state clean energy policy tool (renewable energy policies, GHG emissions policies,
and energy efficiency polices such as energy efficiency resource standards, appliance or
equipment energy efficiency standards, tax incentives, and public building energy efficiency
standards) is associated with 1 percent more clean energy employment within that MSA. These
estimates are not transferable to this rulemaking since states are likely to change intensity as well
as number of clean energy programs.
170 por more information on this efficiency paradox see Chapter 3 and the Technical Support Document (TSD) for
the Final Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units.
Demand-Side Energy Efficiency
6-27
-------
Lacking a peer-reviewed methodology, we use the following approach to illustrate
possible effects on labor demand in the energy efficiency sector due to demand-side management
strategies. We use U.S. government data and divide energy efficiency employment by
expenditures on energy efficiency activities to calculate an estimate of jobs per million dollars.
We then multiply this fraction by projected expenditure in energy efficiency activities
undertaken in response to these final guidelines.
Data used for calculating employment in energy efficiency sectors comes from the
"energy efficiency" industry category of the BLS Green Goods and Services survey.171 Using
BLS Green Goods and Services (GGS) information on 132 energy efficiency industries, as
identified by BLS,172 we adjusted the list to remove ten industries expected to not be directly
affected by the rule, e.g. transportation.173 Next we used this detailed list of 122 industries to
extract 2011 BLS data on green employment.174 Employment data at the most-detailed industry
level (6-digit NAICS) is available only for a portion of these 122 industries. Therefore we use
both the most-detailed industry level (6-digit NAICS) and also a more aggregate level (4-digit
NAICS) to estimate a range of energy efficiency employment with the 2011 BLS Green Goods
and Services data.175
171For more details on this survey, see section 6.5.3.1.
172 See detailed listing available here: http://www.bls.gov/ggs/naics_2012.xlsx. Category 2 is "Energy Efficiency".
More information is available here: http://www.bls.gov/ggs/ggsfaq.htmt3.
173 The ten industries we removed from the list were: NAICS 483114 Coastal and Great Lakes passenger transport,
483212 Inland water passenger transportation, 485 111 Mixed mode transit systems, 485112 Commuter rail systems,
485113 Bus and other motor vehicle transit systems, 485119 Other urban transit systems, 485210 Interurban and
rural bus transportation, 485410 School and employee bus transportation, 485999 All other ground passenger
transportation, and 926120 Transportation program administration. It is possible that certain energy efficiency
services and products produced by the remaining sectors may also be applied in activities that may not be creditable
for compliance with rate-based plans in, or may otherwise be directly incentivized by, this final rule. However, there
is no reason to categorically exclude the remaining sectors from this analysis.
174 BLS Green Goods and Services data is available here:
http://download.bls.gOv/pub/time.series/gg/gg.data.l.AllData. A BLS technical note indicates that the scope of the
GGS survey changed between 2010 and 2011 (http://www.bls.gov/news.release/ggqcew.tn.htm). Some industries
and establishments that were not previously included in the 2010 survey were included in 2011. Rather than using
the change from 2010 to 2011, here we use only the 2011 data.
175 At the 6-digit NAICS level, 27 of the 122 energy efficiency industries listed have employment data available. At
the 4-digit NAICS level, 46 of the 122 energy efficiency industries listed have employment data available.
6-28
-------
BLS does not collect data on energy efficiency expenditures directly, however. Instead,
BLS collects data on the share of revenues associated with green goods and services, at the
establishment level.176 We multiply data on total revenues by NAICS by the share of green
revenues reported by BLS to obtain a measure of green revenues by industry. The only U.S.
Government data source containing this revenue information for all NAICS sectors is the U.S.
Economic Census. This Census is conducted at 5-year intervals (the latest available year is
2012), however, making it unsuitable for directly pairing with 2011 data from BLS. Instead, we
use U.S. Census Bureau data on total value of shipments by industry, for 2011, from the Annual
Survey of Manufacturers.177 The disadvantage of this data source is that the manufacturing sector
makes up only 50 percent of the 132 NAICS codes belonging to the energy efficiency sector as
defined by the BLS Green Goods and Services surveys, with the remainder in the construction or
service sectors. Thus, this analysis implicitly projects that the same number of jobs per dollar are
supported in construction and service sectors as in manufacturing. Also, the Annual Survey of
Manufacturers contains data for some, but not all, detailed industry codes, e.g. 4-digit and 6-digit
NAICS. We pair our BLS GGS data by industry, either by 4-digit or 6-digit NAICS, with data
from the Annual Survey of Manufacturers. At the more detailed, 6-digit level, 17 industries have
data available for both employment and total value of shipments.178 At the less detailed, 4-digit
level, 15 industries have data available for both employment and total value of shipments.179
Using this approach we obtain estimates of 2.07 demand-side energy efficiency jobs per million
2011 dollars of expenditure, using the less-detailed industry level (4-digit NAICS), and 3.29
demand-side energy efficiency jobs per million 2011 dollars of expenditure, using the more-
detailed industry level (6-digit NAICS).
176
More information is available here: http://www.bls.gov/ggs/ggsfaq.htmt5.
177 Census data on total value of shipments, by industry, for 2011 is available here:
http://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ASM_2011_31GS101&prodType
=table.
178 The 17 industries are: NAICS 321219, 321991, 321992, 327993, 327999, 332913, 332996, 333414, 333415,
334513, 334514, 334515, 335110, 335222, 335311, 335312, and 335999.
179 The 15 industries are: NAICS 314100, 321100, 324100, 326100, 327100, 327300, 327400, 332100, 333300,
334200, 334300, 334400, 335200, 336300, and 337900.
6-29
-------
Having calculated estimates of jobs per million dollars of energy efficiency expenditure,
we use EPA's illustrative energy efficiency investment levels of the first-year costs expected for
states to attain a target of 1 percent growth in demand-side efficiency improvements (see Chapter
3.7 of this RIA for more information). If some states were to target rates of energy efficiency
savings greater than one percent, they may see increased energy efficiency employment impacts,
relative to the one percent growth assumed in this analysis. The first year cost of saved energy
(i.e., reduced electricity demand) accounts for both the costs to the utilities that are funding the
demand-side energy efficiency programs (known as the program costs), and the additional cost to the
end-user purchasing a more energy efficient technology (known as the participant costs).180 Total
costs were divided evenly, 50 percent each, between program costs and participant costs. First-year
costs are not annualized; they are the projected expenditures on demand-side energy efficiency
activities in that year. As shown in the Demand-Side Energy Efficiency Technical Support
Document181, first-year costs for achieving a 1 percent growth target182 in energy efficiency activities
are projected to be $18.1 billion (2011 $) in 2020. Multiplying this dollar expenditure by the jobs
per dollar estimates results in projected employment impacts for demand-side energy efficiency
activities ranging from 37,570 to 59,700 jobs in 2020 depending on the jobs per million dollars
estimate used: low or high. Employment impacts for demand-side energy efficiency activities
range from 52,590 to 83,590 jobs in 2025, and from 52,440 to 83,360 jobs in 2030. These
estimates are shown in Table 6-6 below.
180
See Section 3.6.2, "Demand-Side Energy Efficiency Total Costs", in this RIA, for more information.
181 U.S. EPA. 2015. Technical Support Document (TSD) the Final Carbon Pollution Emission Guidelines for
Existing Stationary Sources: Electric Utility Generating Units. Demand-Side Energy Efficiency.
182 The illustrative demand-side energy efficiency plan scenario reflects each state ramping up to the 1% incremental
savings target from their 2013 level, beginning in 2020. Thus, most states are below 1% in 2020. All states have
achieved the 1% target no later than 2025 and the plan scenario has each state remain at that level through 2030.
6-30
-------
Table 6-6. Estimated Demand-Side Energy Efficiency Employment Impacts: Target 1
percent Growth in Energy Efficiency
Employment impact (jobs)*
Source
EPA low estimate, using BLS and Census data,
and power sector modeling projections
EPA high estimate, using BLS and Census data,
and power sector modeling projections
Factor
2.07
3.29
2020
37,570
59,700
2025
52,590
83,590
2030
52,440
83,360
*Since these figures represent number of employees (full- or part-time) they should not be added to the full-time
equivalent job-years reported in Table 6-4 and Table 6-5. Energy efficiency costs are from 1 percent growth target
projections for the continental U.S.. First-year energy efficiency costs are the same for rate-based and mass-based
scenarios. See Chapter 3 of this RIA and Demand-Side Energy Efficiency TSD for more information.
Although this approach has the advantage of using a range of estimates, derived from
U.S. government data, on energy efficiency employment per million dollars in industry
shipments, this approach is limited by its focus on manufacturing sectors and direction of bias
(overestimation or underestimation) cannot be determined at this time. As stated earlier, if, rather
than a one percent target, some states were to target rates of energy efficiency savings greater
than one percent, they may see increased energy efficiency employment impacts, relative to the
one percent growth assumed in this analysis. Finally, because states can choose to reduce
emissions by means of adopting more and broader demand-side energy efficiency programs,
there is uncertainty around the mix of energy efficiency programs and their associated demand-
side energy efficiency employment impacts.
Our estimates of 2.07 to 3.29 demand-side energy efficiency jobs per million dollars of
2011 expenditure fit with other estimates in the literature, focused on government data sources,
and are on the smaller end of the range.183 Figure 6.4 shows the range of estimates, including
EPA low (2.07) and EPA high (3.29). The Department of Commerce report estimates overall
employment per million dollar values, ranging from 4.65 to 4.85 (Department of Commerce
2010a).184 The report also contains some case studies, and for those focused on energy efficiency,
the Department of Commerce estimates 6.21 jobs per million dollars in green buildings activities
183
See Section 6.5.3 for more information on sources in the literature.
184 Calculated values using data on employment and shipments reported in Department of Commerce (2010a), Table
2, p. 12.
6-31
-------
and 7.53 jobs per million dollars for energy efficiency appliances.185 Lawrence Berkeley National
Lab (Goldman et al. 2010) reports a wide range of estimates: from 2.5 jobs per million dollars
for energy service companies (ESCOs), to 8.9 jobs per million dollars for low income
weatherization activities. The Pacific Northwest National Labs report (Anderson et al. 2014), in
surveying the literature, estimates 11 jobs per million dollars of initial energy efficiency
investments.
Demand-Side EE Employment
EPA low
LBNL ESCOs
EPA high
DOC overall low
DOC overall high
LBNL ratepayer funded EE
DOC green buildings
LBNL state and local gov't EER
DOC EE appliances
LBNL low income weatherization
PNNL initial investments
0 5 10
Jobs per Million $
Sources: EPA calculation, LBNL (2010), PNNL (2014), DOC (2010)
Figure 6.4. Demand-Side Energy Efficiency Employment: Jobs per One Million Dollars
(2011$)
There is more uncertainty involved in this approach than the standard bottom-up
engineering analysis used to estimate electricity generation and fuel production employment
impacts of this rulemaking. For those, the EPA was able to identify a limited set of activities
(e.g., constructing a new NGCC power plant), and study associated labor requirements. Demand-
185 Calculated values using data on employment and shipments reported in Department of Commerce (2010a),
Appendix 2, Table 2B, p. 4. http://www.esa.doc.gov/sites/default/files/appendix2_0.pdf
6-32
-------
side energy efficiency improvements, in contrast, encompass a wide array of activities (subsidies
for efficient appliances, "smart meters," etc.). In addition, there is considerable uncertainty
regarding which activities a state will choose. Thus, the validity of the jobs per dollar approach
used here relies on the assumption that states will use a mix of activities similar to the 2011
composition of energy efficiency sectors identified by BLS. Finally, this approach recognizes
that shifts in economic activity towards investments in demand-side energy efficiency are
accompanied by potential shifts in employment.
In addition, the EPA does not have access to bottom-up information regarding labor
requirements for these activities. Use of a constant job per dollar fraction is at best a crude
approximation of these labor requirements. The EPA has identified several other limitations of
this approach, outlined below.
Job Reclassification. Job numbers in this chapter represent gross changes in the affected sector.
As such they may over-estimate impacts to the extent that jobs created displace workers
employed elsewhere in the economy. For demand-side efficiency activities this potential over-
statement may be higher than in other sectors. If states encourage consumers to purchase
ENERGY STAR appliances, for example, currently employed workers in factories and retail
outlets may simply be given a different task. This approach, however, would count these workers
as jobs created.
Imports. The job per dollar fraction used in the employment projection is calculated based on
jobs per dollar of revenue for domestic firms only. To the extent that spending on demand-side
energy efficiency activities goes toward the purchase of imported goods this projection will
overstate the U.S. employment impact of those expenditures.
Fixed Coefficient. Implicit in this approach is the assumption that employment impacts can be
projected decades into the future on the basis of a single calculation from 2011 data. The labor
intensity of demand-side energy efficiency will likely change with technological innovation in
the sector. In addition, even absent technological change, labor intensity of expenditures will
likely change over time as states alter their portfolio of efficiency activities (e.g., by moving to
higher cost activities after exhausting low-cost efficiency activities).
Non-additional Activities. Here we assume that all activities financed by demand-side energy
efficiency expenditures are additional to what would have been undertaken in the absence of
6-33
-------
these programs. For example, if utilities finance some actions customers would have undertaken
in the absence of these programs (e.g., if a customer receives a rebate for an energy efficient
appliance that would have been purchased without the rebate), these numbers would
overestimate employment impacts of the final emissions guidelines.
6.6 Conclusion
This chapter presents qualitative and quantitative discussions of potential employment
impacts of the final guidelines for electricity generation, fuel production, and demand-side
energy efficiency sectors. The qualitative discussion identifies challenges associated with
estimating net employment effects and discusses anticipated impacts related to the rule. It
includes an in-depth discussion of economic theory underlying analysis of employment impacts.
The employment impacts for regulated firms can be decomposed into output and substitution
effects, both of which may be positive or negative. Consequently, economic theory alone cannot
predict the direction or magnitude of a regulation's employment impact. It is possible to combine
theory with empirical studies specific to the regulated firms and other relevant sectors if data and
methods of sufficient detail and quality are available. Finally, economic theory suggests that
environmental regulations may have positive impacts on labor supply and productivity as well.
We examine the peer-reviewed economics literature analyzing various aspects of labor
demand, relying on the above theoretical framework. Determining the direction of employment
effects in regulated industries is challenging because of the complexity of the output and
substitution effects. Complying with a new or more stringent regulation may require additional
inputs, including labor, and may alter the relative proportions of labor and capital used by
regulated firms (and firms in other relevant industries) in their production processes. The
available literature illustrates some of the difficulties for empirical estimation: for example, there
is a paucity of publicly data on plant-level employment, thus most studies must rely on
confidential plant-level employment data from the U.S. Census Bureau, typically combined with
pollution abatement expenditure data, that are too dated to be reliably informative, or other
measures of the stringency of regulation. In addition, the most commonly used empirical
methods, for example, Greenstone (2002), likely overstate employment impacts because they
rely on relative comparisons between more regulated and less regulated counties, which can lead
to "double counting" of impacts when production and employment shift from more regulated
6-34
-------
towards less regulated areas. Thus these empirical methods cannot be used to estimate net
employment effects.. Empirical analysis at the industry level requires estimates of product
demand elasticity; production factor substitutability; supply elasticity of production factors; and
the share of total costs contributed by wages, by industry, and perhaps even by facility.
Econometric studies of environmental rules converge on the finding that employment effects,
whether positive or negative, have been small in regulated sectors.
The illustrative quantitative analysis in this chapter projects a subset of potential
employment impacts in the electricity generation, fuel production, and demand-side energy
efficiency sectors. States have the responsibility and flexibility to implement plans that satisfy
final emissions guidelines, while affected EGUs may choose their compliance strategies from
requirements imposed by these plans. As such, given the wide range of approaches that may be
used, quantifying the associated employment impacts is difficult. EPA's employment analysis
includes projected employment impacts associated with these guidelines assuming two
illustrative plan approach scenarios for the electric power industry, coal and natural gas
production, and demand-side energy efficiency activities. These projections are derived, in part,
from a detailed model of the electricity production sector used for this regulatory analysis, and
U.S. government data on employment and labor productivity. In the electricity, coal, and natural
gas sectors, the EPA estimates that these guidelines could have an employment impact of
roughly -8,500 job-years in 2020, -25,000 job-years in 2025, and -30,900 job-years in 2030 for
the rate-based scenario. For the mass-based scenario, the EPA estimates that these guidelines
could have an employment impact of roughly -13,100 job-years in 2020, -26,000 job-years in
2025, and -33,700 job-years in 2030 (see Tables 6-4 and 6-5).
Employment impacts from demand-side energy efficiency activities are based on historic
data on jobs supported per million dollars of expenditure on energy efficiency. Demand-side
energy efficiency employment impacts would approximately range from 37,570 to 59,700 jobs in
2020, 52,590 to 83,590 jobs in 2025, and from 52,440 to 83,360 jobs in 2030 for both the rate-
based and mass-based approaches and a 1 percent growth target for energy efficiency
expenditures (see Table 6-6).
The IPM-generated job-year numbers for the electricity, coal and natural gas sectors
should not be added to the demand-side efficiency job impacts, since the former are reported in
6-35
-------
full-time equivalent job-years, whereas the latter do not distinguish between full- and part-time
employment. Finally, note again that this analysis is based on two an illustrative plan
approaches, and CAA section lll(d) allows each state to determine its state plan, by way of
meeting its state-specific goal. Given the flexibilities afforded states in implementing plans that
satisfy the emission guidelines, and in the compliance options affected EGUs may take, the
impacts reported in this chapter are illustrative of actions states may take.
6.7 References
Aldy, I.E. 2013. "A Preliminary Assessment of the American Recovery and Reinvestment Act's
Clean Energy Package." Review of Environmental Economics and Policy. 7(1): 136-155.
Allcott, H. 201 la. "Social Norms and Energy Conservation." Journal of Public Economics.
95(9-10):1082-1095.
Allcott, H. 201 Ib. "Rethinking Real-Time Electricity Pricing," Resource and Energy Economics.
33:820-842.
Almon, C. 1991. "The Inforum Approach to Interindustry Modeling," Economic Systems
Research 3(1): 1-8.
Anderson, D., O. Livingston, D. Belzer, D., and M. Scott. 2014. "Assessing National
Employment Impacts of Investment in Residential and Commercial Sector Energy Efficiency:
Review and Example Analysis" [Technical Report]. Pacific Northwest National Laboratory .
Available at < http://www.pnnl. gov/main/publications/external/technical_reports/PNNL-
23402.pdf >. Accessed June 4, 2015.
Anger, N. and U. Oberndorfer. 2008. "Firm Performance and Employment in the EU Emissions
Trading Scheme: An Empirical Assessment for Germany." Energy Policy. (36): 12-22.
Arimura, T. H.; S. Li; R. G. Newell, and K. Palmer. 2012 "Cost-Effectiveness of Electricity
Energy Efficiency Programs." Energy Journal. 33(2):63-99.
Arrow, K. J.; M. L. Cropper; G. C. Eads; R. W. Hahn; L. B. Lave; R. G. Noll; Paul R. Portney;
M. Russell; R. Schmalensee; V. K. Smith and R. N. Stavins. 1996. "Benefit-Cost Analysis in
Environmental, Health, and Safety Regulation: A Statement of Principles." American
Enterprise Institute, the Annapolis Center, and Resources for the Future; AEI Press. Available
at . Accessed June 5, 2015.
6-36
-------
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.
Bezdek, R. H.; R. M. Wendling, and P. Diperna. 2008. "Environmental Protection, the Economy,
and Jobs: National and Regional Analyses." Journal of Environmental Management. 86(1):
63-79.
Brown, R.S. and L. R. Christensen. 1981. "Estimating the Elasticities of Substitution in a Model
of Partial Static Equilibrium: An Application to U.S. Agriculture 1947 to 1974," in: Berndt,
E.R.; B. C. Field. (Eds.), Modeling and Measuring Natural Resource Substitution.
Chan, H. S., S. Li, and F. Zhang. (2013) "Firm Competitiveness and the European Union
Emissions Trading Scheme." Energy Policy. (63): 1056-1064.
Commins, N., S. Lyons, M. Schiffbauer, and R. SJ. Tol. 2011. "Climate Policy and Corporate
Behaviour." The Energy Journal. (32):51-68.
Congressional Budget Office (2011), Statement of Douglas W. Elmendorf, Director, before the
Senate Budget Committee, "Policies for Increasing Economic Growth and Employment in
2012 and 2013" (November 15)
Council of Economic Advisers. 2011. "The Economic Impact of the American Recovery and
Reinvestment Act of 2009." Sixth quarterly report.
Ehrenberg, R. G. and R. S. Smith. 2000. Modern Labor Economics: Theory and Public Policy.
Addison Wesley Longman, Inc., Chapter 4.
Executive Order 13563 (January 21, 2011). "Improving Regulation and Regulatory Review.
Section 1. General Principles of Regulation." Federal Register 76(14): 3821-3823.
Ferris, A. E., R. J. Shadbegian, A. Wolverton. 2014. "The Effect of Environmental Regulation on
Power Sector Employment: Phase I of the Title IV SO2 Trading Program." Journal of the
Association of Environmental and Resource Economists. 1(4): 521-553.
Gillingham, K.; R. Newell and Karen Palmer. 2009. "Energy Efficiency Economics and Policy."
Annual Review of Resource Economics. 1(1): 597-619.
Gillingham, K. and K. Palmer. 2014. "Bridging the Energy Efficiency Gap: Policy Insights from
Economic Theory and Empirical Evidence." Review of Environmental Economics and Policy.
8(l):18-38.
6-37
-------
Global Insight. 2008. U.S. Metro Economies: Current and Potential Green Jobs in the U.S.
Economy. U.S. Conference of Mayors. Available at:
.
Goldman, C.; M. C. Fuller; E. Stuart; J. S. Peters; M. McRae and N. Albers, et al 2010. Energy
Efficiency Services Sector: Workforce Size and Expectations for Growth. Lawrence Berkeley
National Laboratory, LBNL-3987E.
Graff Zivin J. and M. Neidell. 2012. "The Impact of Pollution on Worker Productivity."
American Economic Review. 102(7):3652-73.
Graff Zivin, J. and M. Neidell. 2013. "Environment, Health, and Human Capital." Journal of
Economic Literature. 51:(3):689-730.
Gray, W., R. J. Shadbegian, C. Wang and M. Meral. 2014 "Do EPA Regulations Affect Labor
Demand? Evidence from the Pulp and Paper Industry", Journal of Environmental Economics
and Management. 68: 188-202.
Greenstone, M. 2002. "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(6): 1175-1219.
Hamermesh, D. S. 1993. Labor Demand. Princeton, NJ: Princeton University Press. Chapter 2.
Kahn, M.E. and E. T. Mansur. 2013. "Do Local Energy Prices and Regulation Affect the
Geographic Concentration of Employment?" Journal of Public Economics. 101:105-114.
Layard, P.R.G. and A. A. Walters. 1978. Microeconomic Theory. McGraw-Hill, Inc. Chapter 9.
List, J. A.; D. L. Millimet; P. G. Fredriksson and W. W. McHone. 2003. "Effects of
Environmental Regulations on Manufacturing Plant Births: Evidence from a Propensity Score
Matching Estimator." The Review of Economics and Statistics. 85(4): 944-952.
Marshall, A. 1920. Principles of Economics. Library of Economics and Liberty. Available at:
.
McCarthy, M.B. 1991. "Lift: Inforum's Model of the U.S. Economy," Economic Systems
Research 3(1): 15-36.
Meade, D.S. 2009. "An Analysis of the Economic Impacts of the 2007 Energy Independence and
Security Act", in Energy Policy and International Competitiveness, M. Grassini and R.
Bardazzi, eds. Firenze University Press, pgs. 3-28.
Muro, M.; J. Rothwell; D. Saha & Partnership, w. B. 2011. Sizing the Clean Economy: A
National and Regional Green Jobs Assessment. Brookings Institution, Metropolitan Policy
6-38
-------
Program. Available at
. Accessed
June 4, 2015.
Ostro, B.D. 1987. "Air Pollution and Morbidity Revisited: A Specification Test." Journal of
Environmental Economics Management. 14:87-98.
Schmalansee, R. and R. Stavins (2011). "A Guide to Economic and Policy Analysis for the
Transport Rule." White Paper. Boston, MA. Exelon Corp.
Smith, K. 2012. "Reflections: In Search of Crosswalks between Macroeconomics and
Environmental Economics." Review of Environmental Economics and Policy. 6(2):298-317.
Southwest Energy Efficiency Project. 2002. "The New Mother Lode: The Potential for More
Efficient Electricity Use in the Southwest." Available at:
. Accessed June 4, 2015.
The Pew Charitable Trust. 2009. "The Clean Energy Economy: Repowering Jobs, Businesses
and Investments Across America." Available at: < http://community-
wealth.org/content/clean-energy-economy-repowering-jobs-businesses-and-investments-
across-americax Accessed June 4, 2015.
The Solar Foundation in partnership with BW Research and GW Institute. 2014. "The National
Solar Jobs Census." Available at: .
Sommers, D. 2013. "BLS Geen Jobs Overview," Monthly Labor Review. U.S. Bureau of Labor
Statistics, January, p. 3-16. Available at .
Accessed June 4, 2015.
U.S. Department of Commerce, Economics and Statistics Administration. 2010a. "Measuring the
Green Economy." Available at: . Accessed June 4, 2015.
U.S. Department of Labor, Bureau of Labor Statistics. 2012a. Databases, Tables & Calculators
by Subject. Available at . Acccessed on April
16,2012.
6-39
-------
U.S. Department of Labor, Bureau of Labor Statistics (2012b, April 6). "The Employment
SituationMarch 2012." News Release. Available at
. Accessed April 16, 2012.
U.S. Department of Labor, U.S. Bureau of Labor Statistics, (n.d.). "Employment Projections
Program." and available at: .
U.S. Department of Labor, U.S. Bureau of Labor Statistics, (n.d.). "Green Goods and Services,
2010-2011." Available at: . Retrieved January 14,
2014,
U.S. Energy Information Agency. 2013. "Annual Energy Outlook 2014: Early Release."
Available at: http://www.eia.gov/forecast/aeo/er/early_elecgen.cfm.
U.S. Environmental Protection Agency (U.S. EPA). 201 la. The Benefits and Costs of the Clean
Air Act from 1990 to 2020, Final Report - Rev A. Office of Air and Radiation, Washington,
DC. Available at .
Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 201 Ib. Regulatory Impact Analysis for the
Mercury and Air Toxics Standards (MATS). Office of Air Quality Planning and Standards,
Research Triangle Park, NC. Available at
. Acessed June 4, 2015.
U.S. EPA. 2015. Technical Support Document (TSD) the Final Carbon Pollution Emission
Guidelines for Existing Stationary Sources: Electric Utility Generating Units. GHG
Mitigation Measures.
U.S. International Trade Commission. "Environmental and Related Services." Publication No.
4389, Investigation No. 332-533, March, 2013. Available at
. Accessed June 4, 2015.
Walker, W. R. 2011."Environmental Regulation and Labor Reallocation." American Economic
Review. 101(2): 442-47.
Walker, W. R. 2013." The Transitional Costs of Sectoral Reallocation: Evidence From the Clean
Air Act and the Workforce." The Quarterly Journal of Economics 128 (4): 1787-1835.
Wei, M., S. Patadia, and D. M. Kammen. 2010. "Putting Renewables And Energy Efficiency to
Work: How Many Jobs Can the Clean Energy Industry Generate in the US?" Energy Policy,
38(2):919-931.
6-40
-------
Werling, J.F. 2011. "Preliminary Economic Analysis for OSHA's Proposed Crystalline Silica
Rule: Industry and Macroeconomic Impacts", Final Report for the Occupational Safety and
Health Administration, Available at
. Accessed June 9, 2015.
Yi, H. 2013. "Clean Energy Policies and Green Jobs: An Evaluation of Green Jobs in U.S.
Metropolitan Areas." Energy Policy. 56:644-65.
6-41
-------
APPENDIX 6A: ESTIMATING SUPPLY SIDE EMPLOYMENT IMPACTS
This appendix presents the methods used to estimate the supply-side employment impacts
of the Final Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric
Utility Generating Units (herein referred to as "final emission guidelines" or the "Clean Power
Plan Final Rule"). The focus of the employment analysis is limited to the direct changes in the
amount of labor needed in the power, fuels and generating equipment sectors directly influenced
by the illustrative plan approaches analyzed for the final emission guidelines. It does not include
the ripple effects of these impacts on the broader economy (i.e., the "multiplier" effect), nor does
it include the wider economy-wide effects of the changes to the energy markets, such as changes
in electricity prices.
The methods used to estimate the supply-side employments are based on methods
previously developed for the Mercury and Air Toxics Standards (MATS) Regulatory Impact
Analysis (RIA). The methods used in this analysis to estimate the recurring labor impacts (e.g.,
labor associated with operating and maintaining generating units, as well as labor needed to mine
coal and natural gas) are the same as was used in MATS (with updated data where available).
The labor analysis in the MATS RIA was primarily concerned with the labor needs of
retrofitting pollution control equipment. The analysis for the Clean Power Plan Final Rule,
however, involves the quantity and timing of the labor needs of building new renewable and
natural gas, as well as making heat rate improvements (HRI) at existing coal fired EGUs. These
construction-related compliance activities in the Clean Power Plan Final Rule required
developing additional appropriate analytical methods that were not needed for the MATS
analysis. The newly developed analytical methods for the construction-related activities are
similar in structure and overall approach to the methods used in MATS, but required additional
data and engineering information not needed in the MATS RIA.
6A.1 General Approach
The analytical approach used in this analysis is a bottom-up engineering method
combining the EPA's cost analysis of the final emission guidelines with data on labor
6A-1
-------
productivity, engineering estimates of the amount and types of labor needed to manufacture,
construct, and operate different types of generating units, and prevailing wage rates for skilled
and general labor categories. The approach involved using utility power sector projections and
various energy market implications under the final emission guidelines from modeling conducted
with the EPA Base Case version 5.15, using the Integrated Planning Model (IPM)186, along with
data from secondary sources, to estimate the first order employment impacts for 2020, 2025, and
2030.
Throughout the supply-side labor analysis the engineering approach projects labor changes
measured as the change in each analysis year in job-years employed in the power generation and
directly related sectors (e.g., equipment manufacturing, fuel supply and generating efficiency
services). Job-years are not individual jobs, nor are they necessarily permanent nor full time jobs.
Job-years are the amount of work performed by one full time equivalent (FTE) employee in one
year. For example, 20 job-years in 2020 may represent 20 full-time jobs or 40 half-time jobs in
that year, or any combination of full- and part-time workers such that total 20 FTEs.
The estimates of the employment impacts (both positive and negative) are divided into
five categories:
additional employment to make HRI187 at existing coal fired EGUs;
additional construction-related employment to manufacture and install additional new
generating capacity (renewables, and natural gas combined cycle or combustion turbine
units) when needed as part of early compliance actions;
lost construction-related employment opportunities due to reductions in the total amount
of new generating capacity needed to be built in the later years because of reduced
overall demand for electricity because of demand-side energy efficiency activities;
186 Results for this analysis were developed using various outputs fromEPA's Base Case v.5.15 using ICF's
Integrated Planning Model (IPM). See http://www.epa.gov/powersectormodeling/ for more information.
187 Heat rate improvements could include a range of activities in the power plant to lower the heat rate required to
generate a net electrical output. Assuming all other things being equal, a lower heat rate is more efficient because
less fuel is needed per unit of electric output.
6A-2
-------
lost operating and maintenance employment opportunities due to increased retirements of
coal and small oil/gas units;
changes (both positive and negative) in coal mining and natural gas extraction
employment due to the aggregate net changes in fuel demands arising from all the
activities occurring due to compliance with the final emission guidelines.
Some of the changes are one-time labor effects which are associated with the building (or
avoiding building) new generating capacity and installing HRI. This type of employment effects
involves project-specific labor that is used for 2 to 4 years to complete a specific construction
and installation type of project. There are other labor effects, however, which continue year after
year. For example, bringing new generating capacity online creates an ongoing need for labor to
operate and maintain the new generating capacity throughout the expected service life of the
unit. New generating capacity also creates a need for additional employment to provide the fuel
annually to run the new capacity. There are also continuing effects from the lost operations and
maintenance (O&M) and fuel sector labor opportunities from decisions to retire existing
capacity, as well as similar lost labor opportunities from decisions to reduce a portion of the
amount of additional capacity needed in the base case.
6A.2 Employment Changes due to Heat Rate Improvements
The employment changes due to HRI were estimated based on the incremental MW
capacity estimated to implement such improvements between 2021 and 2030 as indicated by the
IPM analysis presented in Chapter 3. The labor analysis assumes there will be no HRI-related
costs jobs associated with operating or maintaining an EGU after HRI improvements are made.
As described in Chapter 3 of this RIA, EPA modeled the heat rate improvements exogenously to
IPM using the assumption that all "relevant" units can improve their heat rate at a capital cost of
$100/kW. The labor analysis assumes that the cost of implementing the HRI investments at a
particular EGU will occur over a four year period. Hence, the labor analysis calculates the per-
year cost of implementing the HRI is calculated to be $25/kW over a four year period, and these
HRI cost occur in the 4 years prior to the HRI improvements at an individual EGU being
operational.
6A-3
-------
The HRI costs were then allocated to four categories based on the estimates provided by
Andover Technology Partners (ATP), which were adapted from proxy projects involving
installation of combustion control retrofits, such as those installed under the Best Available
Retrofit Technology (BART) submissions from coal-fired power plants located in Wyoming and
Arizona. For more details, refer to the Staudt (2014) report.188 The data from the BART
submissions are used as proxies that are representative of the activities (and their associated
costs) EGUs will improve use to implement the HRI.
Information on cost for these proxies were then extrapolated to approximate the labor
requirements for four broad categories of labor - boilermakers and general construction,
engineering and management support labor, labor required to produce the equipment in upstream
sectors, and labor required to supply the materials (assumed to be primarily steel) in upstream
sectors. More details about these estimates are provided in the Staudt (2014) report.
Based on the cost allocated to each labor categories, output per worker estimates for
respective labor categories, and the assumed growth in labor productivity during the period 2021
through 2030 (with the bulk of HRI occurring between 2021 and 2025), the employment gains
for heat rate improvement were estimated for 2025 using the assumptions summarized in Table
6A-1 below. Output per workers in future years were adjusted to account for growth in labor
productivity, based on historical evidence of productivity growth rates for the relevant sectors.
189
188 Staudt, James, Andover Technology Partners, Inc. Estimating Labor Effects of Heat Rate Improvements. Report
prepared for the Clean Power Plan Proposed Rule, March 6, 2014.
189 Total value of shipments or receipts in 20012 and total employees were taken from 2012 Economic Census,
Statistics by Industry for Mining and Manufacturing sectors. The average annual growth rate of labor productivity
was taken from the Bureau of Labor Statistics. Average growth rate calculated for years 1988-2007, applied to 2012
productivity to determine 2025 estimates of productivity. For the construction sector, BLS productivity growth rate
data was unavailable. Because of this, and lack of reliable data on construction sector productivity growth, the
output per worker for the construction sector was not forecasted to 2030, and the most recent available value from
2012 was used.
6A-4
-------
Table 6A-1. Labor Productivity Growth Rate due to Heat Rate Improvement
Share of the Output/Worker Labor Productivity
Total Capital Cost (2025) Growth Rate
Boilermaker and Gen. Const. 40% $78,500 0%
Management/Engineering 20% $156,000 0.8%
Equipment 30% $542,000 2.7%
Materials 10% $600,000 1.0%
For these output per worker figures, a power sector construction industry (NAICS 237130) was
used for general construction and boilermakers, Engineering Services (NAICS 54133) was used
for the engineering and management component, Machinery Manufacturing (NAICS 333) was
used for the equipment sector, and steel manufacturing (NAICS 3312) was used for materials.
Use of machinery manufacturing for equipment and steel for materials was based on an analysis
of the types of materials and equipment needed for these projects, and what EPA determined to
be the most appropriate industry sectors for those. For more details, refer to the Staudt (2014)
report.
6A.2.1 Employment Changes Due to Building (or Avoiding) New Generation Capacity
Employment changes due to new generation units were based on the incremental changes
in capacity (MW), capital costs ($MM), and fixed operations and maintenance (FOM) costs
($MM) between the policy scenarios and the base case in a given year.
New capacities were aggregated by generation type into the following categories:
Combined Cycle,
Combustion Turbine, and
Renewables (which includes biomass, geothermal, landfill gas, onshore wind, and
solar).
For each category, the analysis estimated the impacts due to both the construction and
operating labor requirements for corresponding capacity changes. The construction labor was
estimated using information on the capital costs, while the operating labor was estimated using
the FOM costs.
Because IPM outputs provide annualized capital costs ($MM), EPA first converted the
annualized capital costs to changes in the total capital investment using the corresponding capital
6A-5
-------
charge rates.190 These total capital investments were then converted to annual capital investments
using assumptions about the estimated duration of the construction phase, in order to estimate the
annual impacts on construction phase labor. Duration estimates were based on assumptions for
construction lengths used in EPA's IPM modeling.191 Specific assumptions used for different
generating technologies are shown in Table 6A-2 below.
Table 6A-2. Capital Charge Rate and Duration Assumptions
New Investment Technology
Capital Charge Rate
Construction Duration
(Years)
Advanced Combined Cycle
Advanced Combustion Turbine
Renewables
Dedicated Biomass
Wind (Onshore)
Landfill Gas
Solar
Geothermal
10.3%
10.6%
9.5%
10.9%
10.9%
10.9%
10.9%
Annual capital costs for each generation type were then broken down into four categories:
equipment, material (which is assumed to be primarily steel), installation labor, and support labor
in engineering and management. The percentage breakdowns shown in Table 6A-3 were
estimated using information provided by Staudt (2014), based primarily on published budgets for
new unit assembled in a study for the National Energy Technology Laboratory (NETL). For
more details, refer to the Staudt (2014) report. Annual capital costs for each generation type
provided by the IPM output were allocated according to this breakdown.
Table 6A-3. Expenditure Breakdown due to New Generating Capacity
Equipment
Material
Labor
Eng. and Const. Mgt
Renewables
Combined Cycle
Combustion Turbine
54%
65%
65%
6%
10%
10%
31%
18%
18%
9%
7%
7%
The short-term construction labor of the new generation units were based on output ($ per
worker) figures for the respective sectors. The total direct workers per $1 million of output for
the baseline year 20012 were forecasted to the years under analysis using the relevant labor
190 Capital charge rates obtained from EPA's resource, EPA #450R13002: Documentation for EPA Base Case v.5.13
using the Integrated Programming Model (IPM).
191 Ibid.
6A-6
-------
productivity growth rate. Table 6A-4 shows the figures for each of the five productivities:
general power plant construction; engineering and management; material use; equipment use;
and plant operators. The resulting values were multiplied by the capital costs to get the job
impact.
Table 6A-4. Labor Productivity due to New Generating Capacity
General Power Plant Construction
Engineering and Management
Material Use (Steel)
Equipment Use (Machinery)
Plant Operators
Labor Productivity
Growth Rate
0.0%
0.8%
1.0%
2.7%
1.7%
Workers per
Million $ (20012)
5.0
5.24.7
1.9
2.1
9.9
General installation labor, assumed to be mostly related to the general power plant
construction phase, was matched with the power industry specific construction sector.
Engineering/management was matched to the engineering services sector to determine their
respective output per worker. For materials, EPA assumed steel to be the proxy and used the
steel manufacturing sector for this productivity. Equipment was assumed to primarily come from
machinery manufacturing sector (such as turbines, engines and fans).
The net labor impact for construction labor for a given year was adjusted to account for
changes in capacity that has already taken place in the prior IPM run year. Because IPM reports
cumulative changes for new generating capacity for any given run year, this adjustment ensured
that the short-term construction phase job impacts in any given run year does not reflect the
cumulative effects of prior construction changes for the given policy scenario. The estimated
amount of the change in construction-related labor in a single IPM run year (e.g., 2025)
represents the average labor impact that occurs in all years between that IPM run year and the
previous run year (i.e., the labor estimates derived from the 2025 IPM run year are the average
annual labor impacts in 2021 through 2025). The construction labor results for 2020 represent
the average labor impacts in 2017 through 2020.
The plant operating employment estimates used a simpler methodology as the one
described above. The operating employment estimates use the IPM estimated change in FOM
costs for the IPM run year. Because the FOM costs are inherently estimates for a single year, the
operating employment estimates are for a single year only. While there are obviously operating
6A-7
-------
employment effects occurring in every year throughout the entire IPM estimation period (2017-
2030), the labor analysis only estimates the single year labor impacts in the IPM run years: 2020,
2025 and 2030. The total direct workers for $1 million and labor productivity growth rate
provided for plant operators in Table 6A-4 were used to estimate the employment impact.
6A.2.2 Employment Changes due to Coal and Oil/Gas Retirements
Employment changes due to plant retirements were calculated using the IPM projected
changes in retirement capacities for coal and oil/gas units for the relevant year and the estimated
changes in total FOM costs due to those retiring units. Thus, the basic assumption in this analysis
is that increased retirements (over the base case) will lead to reduced FOM expenditures at those
plants which were assumed to lead to direct job losses for plant workers.
In order to estimate the total FOM changes due to retirements, EPA first estimated the
average FOM costs ($/kW) for existing coal-fired and oil/gas-fired units in the base case, as
shown in Table 6A-5 below. It was assumed that the average FOM cost of existing units in the
base case can be used as a proxy for the lost economic output due to fossil retirements. Thus,
changes in the FOM costs for these retiring units were derived by taking the product of the
incremental change in capacity and the average FOM costs. These values were converted to lost
employment using data from the Economic Census and BLS on the output/worker estimates for
the utility sector.192
Table 6A-5. Average FOM Costs for Existing Coal and Oil and Gas Steam Capacity
($/kW, 2011$)
2020 2025 2030
Coal $70 $73 $74
Oil and Gas $34 $33 $33
Note that the retirement related employment losses are assumed to include losses directly
affecting the utility sector, and do not include losses in upstream sectors that supply other inputs
to the EGU sector (except fuel related job losses, which are estimated separately and discussed in
the next section).
; The same specific sources as cited before, however, used workers and total payroll.
6A-8
-------
6A.2.3 Employment Changes due to Changes in Fossil Fuel Extraction
Two types of employment impacts due to projected fuel use changes were estimated in this
section. First, employment losses due to either reductions or shifts in coal demand were
estimated using an approach similar to EPA's coal employment analyses under Title IV of the
Clean Air Act Amendments. Using this approach, changes in coal demand (in short tons) for
various coal supplying regions were taken from EPA's base case and illustrative plan scenario
model runs for the final EGU GHG NSPS. These changes were converted to job-years using U.S.
Energy information Administration (EIA) data on regional coal mining productivity (in short
tons per employee hour), using 2012 labor productivity estimates.193'194
Specifically, the incremental changes to coal demand were calculated based on the coal
supply regions in IPM -- Appalachia, Interior, and West and Waste Coal (which was estimated
using U.S. total productivity). Worker productivity values used for estimating coal related job
impacts are shown in Table 6A-6 below.
Table 6A-6. Labor Productivity due to Fossil Fuel Extraction
Labor Productivity
Coal (Short tons/ employee hour)
Appalachia 2.32
Interior 4.73
West 17.09
Waste 5.19
Natural Gas (MMBtu/ employee hour) 122.0
Pipeline Construction (Workers per $Million) 4.2
For natural gas demand, labor productivity per unit of natural gas was unavailable, unlike
coal labor productivities used above. Most secondary data sources (such as Census and EIA)
provide estimates for the combined oil and gas extraction sector. This section thus used an
adjusted labor productivity estimate for the combined oil and gas sector that accounts for the
relative contributions of oil and natural gas in the total sector output (in terms of the value of
193
EIA Annual Energy Review. 2012.
194 Unlike the labor productivity estimates for various equipment resources which were forecasted to 2020 using
BLS average growth rates, the labor analysis uses the most recent historical productivity estimates for fuel sectors.
In general, labor productivity for the fuel sectors (both coal and natural gas) showed a significantly higher degree of
variability in recent years than the manufacturing sectors, which would have introduced a high degree of uncertainty
in forecasting productivity growth rates for future years.
6A-9
-------
energy output in MMBtu). This estimate of labor productivity was then used with the
incremental natural gas demand for the respective IPM runs to estimate the job-years for the
specific year (converting the TCP of gas used projected by IPM into MMBtu using the
appropriate conversion factors). In addition, the pipeline construction costs were estimated using
endogenously determined gas market model parameters in IPM used by EPA for the MATS rule
(using assumptions for EPA's Base Case v4.10). This analysis assumed that the need for
additional pipeline would be proportionate to those projected for the MATS rule and were hence
extrapolated from those estimates (U.S. EPA, 2011). The job-years associated with the pipeline
construction were included in the natural gas employment estimates. Worker productivity values
used for estimating natural gas related job impacts are shown in Table 6A-6.
6A.3 References
Staudt, James. 2014. "Estimating Labor Effects of Heat Rate Improvements. Report prepared for
the Clean Power Plan Proposed Rule, March 6, 2014."
U.S. Energy Information Agency. 2012. "Annual Energy Review 2012: Coal Mining
Productivity Data." Availableat http://www.eia.gov/totalenergy/data/monthly/previous.cfm..
Accessed June 4, 2015
U.S. Environmental Protection Agency (U.S. EPA). 2011. Regulatory Impact Analysis for the
Mercury and Air Toxics Standards (MATS). Office of Air Quality Planning and Standards,
Research Triangle Park, NC. Available at
. Acessed June 4, 2015.
6 A-10
-------
CHAPTER 7: STATUTORY AND EXECUTIVE ORDER ANALYSIS
7.1 Executive Order 12866: Regulatory Planning and Review, and Executive Order
13563: Improving Regulation and Regulatory Review
This final action is an economically significant regulatory action that was submitted to
the OMB for review. Any changes made in response to OMB recommendations have been
documented in the docket. The EPA prepared an analysis of the potential costs and benefits
associated with this action.
Consistent with Executive Order 12866 and Executive Order 13563, the EPA estimated
the costs and benefits for illustrative plan approaches of implementing the guidelines. The final
rule establishes: 1) state-specific carbon dioxide (COi) goals reflecting COi emission
performance rates for two source categories of existing fossil fuel-fired EGUs, fossil fuel-fired
electric utility steam generating units and stationary combustion turbines, and 2) guidelines for
the development, submittal and implementation of state plans that establish emission standards
or other measures to implement the COi emission performance rates. Actions taken to comply
with the guidelines will also reduce the emissions of directly-emitted PM2.5, SOi and NOx. The
benefits associated with these PM2.5, SOi and NOx reductions are referred to as co-benefits, as
these reductions are not the primary objective of this rule.
The EPA has used the social cost of carbon estimates presented in the Technical Support
Document: Technical Update of the Social Cost of Carbon for Regulatory Impact Analysis
Under Executive Order 12866 (May 2013, Revised July 2015) ("current TSD") to analyze CO2
climate impacts of this rulemaking. We refer to these estimates, which were developed by the
U.S. government, as "SC-COi estimates." The SC-COi is an estimate of the monetary value of
impacts associated with a marginal change in COi emissions in a given year. The four SC-COi
estimates are associated with different discount rates (model average at 2.5 percent discount rate,
3 percent, and 5 percent; 95th percentile at 3 percent), and each increases over time. In this
summary, the EPA provides the estimate of climate benefits associated with the SC-CO2 value
deemed to be central in the current TSD: the model average at 3 percent discount rate.
In the final emission guidelines, the EPA has translated the source category-specific COi
emission performance rates into equivalent state-level rate-based and mass-based COi goals in
7-1
-------
order to maximize the range of choices that states will have in developing their plans. Because of
the range of choices available to states and the lack of a priori knowledge about the specific
choices states will make in response to the final goals, the Regulatory Impact Analysis (RIA) for
this rule analyzed two implementation scenarios designed to achieve these goals, which we term
the "rate-based" illustrative plan approach and the "mass-based" illustrative plan approach.
It is very important to note that the differences between the analytical results for the rate-
based and mass-based illustrative plan approaches presented in the RIA may not be indicative of
likely differences between the approaches if implemented by states and affected EGUs in
response to the final guidelines. Rather, the two sets of analyses are intended to illustrate two
contrasting, stylized approaches to accomplish the emission performance rates finalized in the
Clean Power Plan Final Rule. In other words, if one approach performs better than the other on a
given metric during a given time period, this does not imply this will apply in all instances in all
time periods in all places.
The EPA estimates that, in 2020, the final guidelines will yield monetized climate
benefits (in 2011$) of approximately $2.8 billion for the rate-based approach and $3.3 billion for
the mass-based approach (3 percent model average). For the rate-based approach, the air
pollution health co-benefits in 2020 are estimated to be $0.7 billion to $1.8 billion (2011$) for a
3 percent discount rate and $0.64 billion to $1.7 billion (2011$) for a 7 percent discount rate. For
the mass-based approach, the air pollution health co-benefits in 2020 are estimated to be $2.0
billion to $4.8 billion (2011$) for a 3 percent discount rate and $1.8 billion to $4.4 billion
(2011$) for a 7 percent discount rate. The annual, illustrative compliance costs estimated by IPM
and inclusive of demand-side energy efficiency program and participant costs and MRR costs in
2020, are approximately $2.5 billion for the rate-based approach and $1.4 billion for the mass-
based approach (2011$). The quantified net benefits (the difference between monetized benefits
and compliance costs) in 2020 are estimated to range from $1.0 billion to $2.1 billion (2011$) for
the rate-based approach and from $3.9 billion to 6.7 billion (2011$) for the mass-based approach,
using a 3 percent discount rate (model average).
The EPA estimates that, in 2025, the final guidelines will yield monetized climate
benefits (in 2011$) of approximately $10 billion for the rate-based approach and $12 billion for
the mass-based approach (3 percent model average). For the rate-based approach, the air
7-2
-------
pollution health co-benefits in 2025 are estimated to be $7.4 billion to $18 billion (2011$) for a 3
percent discount rate and $6.7 billion to $16 billion (2011$) for a 7 percent discount rate. For the
mass-based approach, the air pollution health co-benefits in 2025 are estimated to be $7.1 billion
to $17 billion (2011$) for a 3 percent discount rate and $6.5 billion to $16 billion (2011$) for a 7
percent discount rate. The annual, illustrative compliance costs estimated by IPM and inclusive
of demand side energy efficiency program and participant costs and MRR costs in 2025, are
approximately $1.0 billion for the rate-based approach and $3.0 billion for the mass-based
approach (2011$). The quantified net benefits (the difference between monetized benefits and
compliance costs) in 2025 are estimated to range from $17 billion to $27 billion (2011$) for the
rate-based approach and $16 billion to $26 billion (2011$) for the mass-based approach, using a
3 percent discount rate (model average).
The EPA estimates that, in 2030, the final guidelines will yield monetized climate
benefits (in 2011$) of approximately $20 billion for the rate-based approach and $20 billion for
the mass-based approach (3 percent model average). For the rate-based approach, the air
pollution health co-benefits in 2030 are estimated to be $14 billion to $34 billion (2011$) for a 3
percent discount rate and $13 billion to $31 billion (2011$) for a 7 percent discount rate. For the
mass-based approach, the air pollution health co-benefits in 2030 are estimated to be $12 billion
to $28 billion (2011$) for a 3 percent discount rate and $11 billion to $26 billion (2011$) for a 7
percent discount rate. The annual, illustrative compliance costs estimated by IPM and inclusive
of demand side energy efficiency program and participant costs and MRR costs in 2030, are
approximately $8.4 billion for the rate-based approach and $5.1 billion for the mass-based
approach (2011$). The quantified net benefits (the difference between monetized benefits and
compliance costs) in 2030 are estimated to range from $26 billion to $45 billion (2011$) for the
rate-based approach and from $26 billion to $43 billion (2011$) for the mass-based approach,
using a 3 percent discount rate (model average).
Table 7-1 and 7-2 provide the estimates of the climate benefits, health co-benefits,
compliance costs and net benefits of the final emission guidelines for rate-based and mass-based
illustrative plan approaches, respectively.
7-3
-------
Table 7-1. Monetized Benefits, Compliance Costs, and Net Benefits Under the Rate-based
Illustrative Plan Approach (billions of 2011$)a
Rate-Based Approach
2020
2025
2030
Climate Benefits b
5% discount rate
3% discount rate
2.5% discount rate
95thpercentileat3%
$0.80
$2.8
$4.1
$8.2
$3.1
$10
$15
$31
$6.4
$20
$29
$61
Air Quality Co-benefits Discount Rate
3%
7%
3%
7%
3%
7%
Air Quality Health
Co-benefits c
Compliance Costs d
Net Benefits e
$0.7 to $1.8 $0.6 to $1.7
$2.5
$1.0 to $2.1 $1.0 to $2.0
$7.4 to $18 $6.7 to $16
$1.0
$17 to $27 $16 to $25
$14 to $34 $13 to $31
$8.4
$26 to $45 $25 to $43
Non-monetized climate benefits
Reductions in exposure to ambient NO2 and SO2
Non-Monetized Reductions in mercury deposition
1 Ecosystem benefits associated with reductions in emissions of NOx, SO2, PM, and
mercury
Visibility impairment
a All are rounded to two significant figures, so figures may not sum.
b The climate benefit estimate in this summary table reflects global impacts from CO2 emission changes and does
not account for changes in non-CO2 GHG emissions. Also, different discount rates are applied to SC-CO2 than to the
other estimates because CO2 emissions are long-lived and subsequent damages occur over many years. The benefit
estimates in this table are based on the average SC-CO2 estimated for a 3 percent discount rate, however we
emphasize the importance and value of considering the full range of SC-CO2 values. As shown in the RIA, climate
benefits are also estimated using the other three SC-CO2 estimates (model average at 2.5 percent discount rate, 3
percent, and 5 percent; 95th percentile at 3 percent). The SC-CO2 estimates are year-specific and increase over time.
c The air quality health co-benefits reflect reduced exposure to PM2.s and ozone associated with emission reductions
of directly emitted PM2.5, SO2 and NOx. The range reflects the use of concentration-response functions from
different epidemiology studies. The reduction in premature fatalities each year accounts for over 98 percent of total
monetized co-benefits from PM2.s and ozone. These models assume that all fine particles, regardless of their
chemical composition, are equally potent in causing premature mortality because the scientific evidence is not yet
sufficient to allow differentiation of effect estimates by particle type. Estimates in the table are presented for three
analytical years with air quality co-benefits calculated using two discount rates. The estimates of co-benefits are
annual estimates in each of the analytical years, reflecting discounting of mortality benefits over the cessation lag
between changes in PM2.s concentrations and changes in risks of premature death (see RIA Chapter 4 for more
details), and discounting of morbidity benefits due to the multiple years of costs associated with some illnesses. The
estimates are not the present value of the benefits of the rule over the full compliance period.
d Total costs are approximated by the illustrative compliance costs estimated using the Integrated Planning Model for
the final emission guidelines and a discount rate of approximately 5 percent. This estimate also includes monitoring,
recordkeeping, and reporting costs and demand-side energy efficiency program and participant costs.
e The estimates of net benefits in this summary table are calculated using the global SC-CO2 at a 3 percent discount
rate (model average). The RIA includes combined climate and health estimates based on additional discount rates.
7-4
-------
Table 7-2. Monetized Benefits, Compliance Costs, and Net Benefits under the Mass-based
Illustrative Plan Approach (billions of 2011$)a
Mass-Based Approach
2020
2025
2030
Climate Benefits b
5% discount rate
3% discount rate
2.5% discount rate
95thpercentileat3%
$0.9
$3.3
$4.9
$9.6
$3.6
$12
$17
$35
$6.4
$20
$29
$60
Air Quality Co-benefits Discount Rate
3%
7%
3%
7%
3%
7%
Air Quality Health
Co-benefits c
Compliance Costs d
Net Benefits e
$2.0 to $4.8 $1.8 to $4.4
$1.4
$3.9 to $6.7 $3.7 to $6.3
$7 to $17 $7 to $16
$3.0
$16 to $26 $15 to $24
$12 to $28 $11 to $26
$5.1
$26 to $43 $25 to $40
Non-monetized climate benefits
Reductions in exposure to ambient NO2 and SO2
Non-Monetized Reductions in mercury deposition
1 Ecosystem benefits associated with reductions in emissions of NOx, SO2, PM, and
mercury
Visibility improvement
a All are rounded to two significant figures, so figures may not sum.
b The climate benefit estimate in this summary table reflects global impacts from CO2 emission changes and does
not account for changes in non-CO2 GHG emissions. Also, different discount rates are applied to SC-CO2 than to the
other estimates because CO2 emissions are long-lived and subsequent damages occur over many years. The benefit
estimates in this table are based on the average SC-CO2 estimated for a 3 percent discount rate, however we
emphasize the importance and value of considering the full range of SC-CO2 values. As shown in the RIA, climate
benefits are also estimated using the other three SC-CO2 estimates (model average at 2.5 percent discount rate, 3
percent, and 5 percent; 95th percentile at 3 percent). The SC-CO2 estimates are year-specific and increase over time.
c The air quality health co-benefits reflect reduced exposure to PM2.s and ozone associated with emission reductions
of directly emitted PM2.5, SO2 and NOx. The range reflects the use of concentration-response functions from
different epidemiology studies. The reduction in premature fatalities each year accounts for over 98 percent of total
monetized co-benefits from PM2.s and ozone. These models assume that all fine particles, regardless of their
chemical composition, are equally potent in causing premature mortality because the scientific evidence is not yet
sufficient to allow differentiation of effect estimates by particle type. Estimates in the table are presented for three
analytical years with air quality co-benefits calculated using two discount rates. The estimates of co-benefits are
annual estimates in each of the analytical years, reflecting discounting of mortality benefits over the cessation lag
between changes in PM2.s concentrations and changes in risks of premature death (see RIA Chapter 4 for more
details), and discounting of morbidity benefits due to the multiple years of costs associated with some illnesses. The
estimates are not the present value of the benefits of the rule over the full compliance period.
d Total costs are approximated by the illustrative compliance costs estimated using the Integrated Planning Model for
the final emission guidelines and a discount rate of approximately 5 percent. This estimate also includes monitoring,
recordkeeping, and reporting costs and demand-side energy efficiency program and participant costs.
e The estimates of net benefits in this summary table are calculated using the global SC-CO2 at a 3 percent discount
rate (model average). The RIA includes combined climate and health estimates based on additional discount rates.
7-5
-------
There are additional important benefits that the EPA could not monetize. Due to current
data and modeling limitations, our estimates of the benefits from reducing CO2 emissions do not
include important impacts like ocean acidification or potential tipping points in natural or
managed ecosystems. Unquantified benefits also include climate benefits from reducing
emissions of non-COi greenhouse gases (e.g., nitrous oxide and methane) and co-benefits from
reducing direct exposure to SOi, NOx and hazardous air pollutants (e.g., mercury and hydrogen
chloride), as well as from reducing ecosystem effects and visibility impairment. Based upon the
foregoing discussion, it remains clear that the benefits of this final action are substantial, and far
exceed the costs. Additional details on benefits, costs, and net benefits estimates are provided in
this RIA.
7.2 Paperwork Reduction Act (PRA)
The information collection requirements in this rule have been submitted for approval to
OMB under the PRA. The Information Collection Request (ICR) document prepared by the EPA
has been assigned the EPA ICR number 2503.02. You can find a copy of the ICR in the docket
for this rule, and it is briefly summarized here. The information collection requirements are not
enforceable until OMB approves them.
This rule does not directly impose specific requirements on EGUs located in states or
areas of Indian country. The rule also does not impose specific requirements on tribal
governments that have affected EGUs located in their area of Indian country. For areas of Indian
country, the rule establishes CO2 emission performance goals that could be addressed through
either tribal or federal plans. A tribe would have the opportunity under the Tribal Authority Rule
(TAR), but not the obligation, to apply to the EPA for Treatment as State (TAS) for purposes of
a CAA section 11 l(d) plan and, if approved by the EPA, to establish a CAA section 11 l(d) plan
for its area of Indian country. To date, no tribe has requested or obtained TAS eligibility for
purposes of a CAA section 11 l(d) plan. For areas of Indian country with affected EGUs where a
tribe has not applied for TAS and submitted any needed plan, if the EPA determines that a CAA
section 11 l(d) plan is necessary or appropriate, the EPA would have the responsibility to
establish the plans. Because tribes are not required to implement section lll(d) plans and
because no tribe has yet sought TAS eligibility for this purpose, this action is not anticipated to
7-6
-------
impose any information collection burden on tribal governments over the 3-year period covered
by this ICR.
This rale does impose specific requirements on state governments with affected EGUs.
The information collection requirements are based on the recordkeeping and reporting burden
associated with developing, implementing, and enforcing a plan to limit COi emissions from
existing sources in the utility power sector. These recordkeeping and reporting requirements are
specifically authorized by CAA section 114 (42 U.S.C. 7414). All information submitted to the
EPA pursuant to the recordkeeping and reporting requirements for which a claim of
confidentiality is made is safeguarded according to agency policies set forth in 40 CFR part 2,
subpart B.
The annual burden for this collection of information for the states (averaged over the first
3 years following promulgation) is estimated to be a range of 505,000 to 821,000 hours at a total
annual labor cost of $35.8 to $58.1 million. The lower bound estimate reflects the assumption
that some states already have energy efficiency and renewable energy programs in place. The
higher bound estimate reflects the overly-conservative assumption that no states have energy
efficiency and renewable energy programs in place.
The total annual burden for the federal government associated with the state collection of
information (averaged over the first 3 years following promulgation) is estimated to be 54,000
hours at a total annual labor cost of $3.00 million. Burden is defined at 5 CFR 1320.3(b).
An agency may not conduct or sponsor, and a person is not required to respond to, a
collection of information unless it displays a currently valid OMB control number. The OMB
control numbers for the EPA's regulations in 40 CFR are listed in 40 CFR part 9. When OMB
approves this ICR, the agency will announce that approval in the Federal Register and publish a
technical amendment to 40 CFR part 9 to display the OMB control number for the approved
information collection activities contained in this final rule.
7.3 Regulatory Flexibility Act (RFA)
The EPA certifies that this action will not have a significant economic impact on a
substantial number of small entities under the RFA. This action will not impose any
requirements on small entities. Specifically, emission guidelines established under CAA section
7-7
-------
11 l(d) do not impose any requirements on regulated entities and, thus, will not have a significant
economic impact upon a substantial number of small entities. After emission guidelines are
promulgated, states establish emission standards on existing sources, and it is those requirements
that could potentially impact small entities.
Our analysis here is consistent with the analysis of the analogous situation arising when
the EPA establishes NAAQS, which do not impose any requirements on regulated entities. As
here, any impact of a NAAQS on small entities would only arise when states take subsequent
action to maintain and/or achieve the NAAQS through their state implementation plans. See
American Trucking Assoc. v. EPA, 175 F.3d 1029, 1043-45 (D.C. Cir. 1999) (NAAQS do not
have significant impacts upon small entities because NAAQS themselves impose no regulations
upon small entities).
Nevertheless, the EPA is aware that there is substantial interest in the rule among small
entities and, as detailed in section III.A of the preamble to the proposed carbon pollution
emission guidelines for existing EGUs (79 FR 34845-34847; June 18, 2014) and in section II.D
of the preamble to the proposed carbon pollution emission guidelines for existing EGUs in
Indian Country and U.S. Territories (79 FR 65489; November 4, 2014), has conducted an
unprecedented amount of stakeholder outreach. As part of that outreach, agency officials
participated in many meetings with individual utilities and electric utility associations, as well as
industry leaders and trade association representatives from various industries. While formulating
the provisions of the rule, the EPA considered the input provided over the course of the
stakeholder outreach as well as the input provided in the many public comments.
7.4 Unfunded Mandates Reform Act (UMRA)
This action does not contain an unfunded mandate of $100 million or more as described
in UMRA, 2 U.S.C. 1531-1538, and does not significantly or uniquely affect small governments.
The emission guidelines do not impose any direct compliance requirements on EGUs located in
states or areas of Indian country. As explained in section XII.B above, the rule also does not
impose specific requirements on tribal governments that have affected EGUs located in their area
of Indian country. The rule does impose specific requirements on state governments that have
affected EGUs. Specifically, states are required to develop plans to implement the guidelines
under CAA section 11 l(d) for affected EGUs. The burden for states to develop CAA section
7-8
-------
11 l(d) plans in the 3-year period following promulgation of the rale was estimated and is listed
in section XII.B above, but this burden is estimated to be below $100 million in any one year.
Thus, this rale is not subject to the requirements of section 202 or section 205 of the UMRA.
This rule is also not subject to the requirements of section 203 of UMRA because it contains no
regulatory requirements that might significantly or uniquely affect small governments.
Specifically, the state governments to which rale requirements apply are not considered small
governments.
In light of the interest among governmental entities, the EPA conducted outreach with
national organizations representing state and local elected officials and tribal governmental
entities while formulating the provisions of this rule. Sections III.A and XI.F of the preamble to
the proposed carbon pollution emission guidelines for existing EGUs (79 FR 34845-34847; June
18, 2014) and sections II.D and VI.F of the preamble to the proposed carbon pollution emission
guidelines for existing EGUs in areas of Indian Country and U.S. Territories (79 FR 65489;
November 4, 2014) describes the extensive stakeholder outreach the EPA has conducted on
setting emission guidelines for existing EGUs. The EPA considered the input provided over the
course of the stakeholder outreach as well as the input provided in the many public comments
when developing the provisions of these emission guidelines.
7.5 Executive Order 13132: Federalism
The EPA has concluded that this action may have federalism implications, pursuant to
agency policy for implementing the Order, because it imposes substantial direct compliance
costs on state or local governments, and the federal government will not provide the funds
necessary to pay those costs. As discussed in the Supporting Statement found in the docket for
this ralemaking, the development of state plans will entail many hours of staff time to develop
and coordinate programs for compliance with the rule, as well as time to work with state
legislatures as appropriate, to develop a plan submittal. Consistent with this determination, the
EPA provides the following federalism summary impact statement.
The EPA consulted with state and local officials early in the process of developing the
proposed action to permit them to have meaningful and timely input into its development. As
described in the Federalism discussion in the preamble to the proposed standards of performance
for GHG emissions from new EGUs (79 FR 1501; January 8, 2014), the EPA consulted with
7-9
-------
state and local officials in the process of developing the proposed standards for newly
constructed EGUs. This outreach addressed planned actions for new, reconstructed, modified and
existing sources. The EPA invited the following 10 national organizations representing state and
local elected officials to a meeting on April 12, 2011, in Washington DC: (1) National Governors
Association; (2) National Conference of State Legislatures, (3) Council of State Governments,
(4) National League of Cities, (5) U.S. Conference of Mayors, (6) National Association of
Counties, (7) International City/County Management Association, (8) National Association of
Towns and Townships, (9) County Executives of America, and (10) Environmental Council of
States. The National Association of Clean Air Agencies also participated. On February 26, 2014,
the EPA re-engaged with those governmental entities to provide a pre-proposal update on the
emission guidelines for existing EGUs and emission standards for modified and reconstructed
EGUs. In addition, as described in section III.A of the preamble to the proposed carbon pollution
emission guidelines for existing EGUs (79 FR 34845-34847; June 18, 2014), extensive
stakeholder outreach conducted by the EPA allowed state leaders, including governors, state
attorneys general, environmental commissioners, energy officers, public utility commissioners,
and air directors, opportunities to engage with EPA officials and provide input regarding
reducing carbon pollution from power plants.
In the spirit of Executive Order 13132, and consistent with the EPA's policy to promote
communications between the EPA and state and local governments, the EPA specifically
solicited comment on the proposed action from state and local officials. The EPA received
comments from over 400 entities representing state and local governments.
Several themes emerged from state and local government comments. Commenters raised
concerns with the building blocks that comprise the best system of emission reduction (BSER),
including the stringency of the building blocks, and the timing of achieving interim COi levels.
They also identified the potential for electric system reliability issues and stranded assets due to
the proposed timeframe for plan submittals and COi emission reductions. In addition, states
commented on state plan development and implementation topics, including state plan
approaches, early actions, trading programs, interstate crediting for RE, and EPA guidance and
outreach.
7-10
-------
Commenters identified overarching concerns regarding the stringency of the COi goals
and the timeframe for achieving reductions that encompassed the building blocks, the BSER, and
associated timing for achievement of interim COi levels. State commenters, in particular,
identified changes to the stringency of the building blocks, concerns with the timeframe over
which reductions must be achieved, and concerns with the approaches and measures used for the
BSER. For the final rule, in response to stakeholder comments, the EPA has made refinements to
the building blocks, the period of time over which measures are deployed, and the stringency of
emission limitations that those measures can achieve in a practical and reasonable cost way. The
final BSER reflects those refinements.
To many commenters, the proposal's 2020 compliance date, together with the stringency
of the interim CCh goal, bore significant reliability implications. In this final rule, the agency is
addressing those concerns via adjustments to the compliance timeframe (an 8-year interim period
that begins in 2022) and to the approach for meeting interim COi emission performance rates (a
glide path separated into three steps, 2022-2024, 2025-2027, and 2028-2029), as well as a more
gradual phase in of the emission reduction expectations. These adjustments provide more time
for planning, consultation and decision making in the formulation of state plans and in EGUs'
choices of compliance strategies. The final rule also retains flexibilities presented in the proposal
and offers additional opportunities, including opportunities for trading within and between states,
and other multi-state compliance approaches that will further support electric system reliability.
The EPA is also requiring states to consult with relevant ISOs/RTOs and/or planning/reliability
authorities during plan development, and to document recommendations in their plans - and is
providing the time for states to do so. Even with this foundation of flexibility in place, these final
guidelines further provide states with the option of proposing amendments to approved plans in
the event that unanticipated and significant reliability challenges arise.
Commenters provided compelling information indicating that it will take longer than the
agency initially anticipated to adjust investments and achieve interim COi reductions.
Recognizing this, as well as the urgent need for actions to reduce GHG emissions, the EPA is
requiring states to frame an initial plan by August 31, 2016, and is allowing states two additional
years to submit a final plan, if justified (to be submitted by August 31, 2018).
7-11
-------
States commented on state plan development and implementation topics that included
state plan approaches, early actions being taken into account, trading programs being allowed,
interstate crediting for RE being allowed, and guidance and outreach being provided by the EPA.
For the state plan approaches, commenters expressed concerns with the proposed "portfolio
approach" for state plans, including concerns with enforceability of requirements, and identified
a "state commitment approach" with backstop measures as an option for state plans. In this final
rule, in response to stakeholder comments on the portfolio approach and alternative approaches,
the EPA is finalizing a "state measures" approach that includes a requirement for the inclusion of
backstop measures.
State commenters supported providing incentives for states and utilities to deploy COi-
reducing investments, such as RE and demand-side EE measures, as early as possible. The EPA
recognizes the value of such early actions, and in this final rule is establishing a state-federal
Clean Energy Incentive Program to reward investment in certain RE and demand-side EE
projects that commence construction after the effective date of this rule and that generate MWh
or reduce end-use energy demand during 2020 and 2021.
Many state commenters supported the use of mass-based and rate-based emission trading
programs in state plans, including interstate emission trading programs. The EPA also received a
number of comments from states and stakeholders about the value of EPA support in developing
and/or administering tracking systems to support state administration of rate-based and mass-
based emission trading programs. In this final rule, states may use trading or averaging
approaches and technologies or strategies that are not explicitly mentioned in any of the three
building blocks as part of their overall plans, as long as they achieve the required emission
reductions from affected fossil-fuel-fired EGUs. In addition, in response to concerns from states
and power companies that the need for up-front interstate cooperation in developing multi-state
plans could inhibit the development of interstate programs that could lower cost, the final rule
provides additional options to allow individual EGUs to use creditable out-of-state reductions to
achieve required COi reductions, without the need for up-front interstate agreements. The EPA is
committed to working with states to provide support for tracking of emissions and allowances or
credits, to help implement multi-state trading or averaging approaches.
7-12
-------
In their comments, many states identified the need for the EPA to provide guidance,
including guidance on RE and EE emission measurement and verification (EM&V), and to
maintain regular contact/forums with states throughout the implementation process. To provide
state and local governments and other stakeholders with an understanding of the rule
requirements, and to provide efficiencies where possible and reduce the cost and administrative
burden, the EPA will continue outreach throughout the plan development and submittal process.
Outreach will include opportunities for states to participate in briefings, teleconferences, and
meetings about the final rule. The EPA's 10 regional offices will continue to be the entry point
for states and tribes to ask technical and policy questions. The agency will host (or partner with
appropriate groups to co-host) a number of webinars about various components of the final rule
during the first two months after the final rule is issued. The EPA will use information from this
outreach process to inform the training and other tools that will be of most use to the states and
tribes that are implementing the final rule. The EPA expects to issue guidance on specific topics,
including evaluation, measurement and verification (EM&V) for RE and demand-side EE, state-
community engagement, and resources and financial assistance for RE and demand-side EE. As
guidance documents, tools, templates and other resources become available, the EPA, in
consultation with the U.S. Department of Energy and other federal agencies, will continue to
make these resources available via a dedicated website.
A list of the state and local government commenters has been provided to OMB and has
been placed in the docket for this rulemaking. In addition, the detailed response to comments
from these entities is contained in the EPA's response to comments document on this final
rulemaking, which has also been placed in the docket for this rulemaking.
As required by section 8(a) of Executive Order 13132, the EPA included a certification
from its Federalism Official stating that the EPA had met the Executive Order's requirements in
a meaningful and timely manner when it sent the draft of this final action to OMB for review
pursuant to Executive Order 12866. A copy of the certification is included in the public version
of the official record for this final action.
7-13
-------
7.6 Executive Order 13175: Consultation and Coordination with Indian Tribal
Governments
This action has tribal implications. However, it will neither impose substantial direct
compliance costs on federally recognized tribal governments, nor preempt tribal law. Tribes are
not required to develop or adopt CAA programs, but they may apply to the EPA for treatment in
a manner similar to states (TAS) and, if approved, do so. As a result, tribes are not required to
develop plans to implement the guidelines under CAA section 11 l(d) for affected EGUs in their
areas of Indian country. To the extent that a tribal government seeks and attains TAS status for
that purpose, these emission guidelines would require that planning requirements be met and
emission management implementation plans be executed by the tribes. The EPA notes that this
rule does not directly impose specific requirements on affected EGUs, including those located in
areas of Indian country, but provides guidance to any tribe approved by the EPA to address CO2
emissions from EGUs subject to section 11 l(d) of the CAA. The EPA also notes that none of the
affected EGUs are owned or operated by tribal governments.
As described in sections III.A and XI.F of the preamble to the proposed carbon
pollution emission guidelines for existing EGUs (79 FR 34845-34847; June 18, 2014) and
sections II.D and VI.F of the preamble to the proposed carbon pollution emission guidelines for
existing EGUs in Indian Country and U.S. Territories (79 FR 65489; November 4, 2014), the
rule was developed after extensive and vigorous outreach to tribal governments. These tribes
expressed varied points of view. Some tribes raised concerns about the impacts of the regulations
on EGUs located in their areas of Indian country and the subsequent impact on jobs and revenue
for their tribes. Other tribes expressed concern about the impact the regulations would have on
the cost of water covered under treaty to their communities as a result of increased costs to the
EGU that provide energy to transport the water to the tribes. Other tribes raised concerns about
the impacts of climate change on their communities, resources, ways of life and hunting and
treaty rights. The tribes were also interested in the scope of the guidelines being considered by
the agency (e.g., over what time period, relationship to state and multi-state plans) and how
tribes will participate in these planning activities.
The EPA consulted with tribal officials under the EPA Policy on Consultation and
Coordination with Indian Tribes early in the process of developing this action to permit them to
have meaningful and timely input into its development. A summary of that consultation follows.
7-14
-------
Prior to issuing the supplemental proposal on November 4, 2014, the EPA consulted with
tribes as follows. The EPA held a consultation with the Ute Tribe, the Crow Nation, and the
Mandan, Hidatsa, Arikara (MHA) Nation on July 18, 2014. On August 22, 2014, the EPA held a
consultation with the Fort Mojave Tribe. On September 15, 2014, the EPA held a consultation
with the Navajo Nation. The Navajo Nation sent a letter to the EPA on September 18, 2014,
summarizing the information presented at the consultation and the Navajo Nation's position on
the supplemental proposal. One issue raised by tribal officials was the potential impacts of the
June 18, 2014 proposal and the supplemental proposal on tribes with budgets that are dependent
on revenue from coal mines and power plants, as well as employment at the mines and power
plants. The tribes noted the high unemployment rates and lack of access to basic services on their
lands. Tribal officials also asked whether the rules will have any impact on a tribe's ability to
seek TAS. Tribal officials also expressed interest in agency actions with regard to facilitating
power plant compliance with regulatory requirements. The Navajo Nation made the following
recommendations in their letter of September 18, 2014: the Navajo Nation supports a mass-based
COi emission standard based on the highest historical COi emissions since 1996; the Navajo
Nation requests that the EPA grant the Navajo Nation carbon credits and that the Navajo Nation
retains ownership and control of such credits; building block 2 is not appropriate for the Navajo
Nation because there are no NGCC plants located on the Navajo Nation; building block 3 is not
appropriate for the Navajo Nation because the Navajo people already receive virtually all of their
electricity from carbon-free sources (mostly hydroelectric power) and their use of electricity is
negligible compared to the generation at the power plants; building block 4 is not appropriate for
the Navajo Nation because of the inadequate access to electricity, and the goal should allow for
an increase in energy consumption on the Navajo Nation; the supplemental proposal should
consider the useful life of the power plants located on the Navajo Nation; and the supplemental
proposal should clarify that RE projects located within the Navajo Nation that provide electricity
outside the Navajo Nation should be counted toward meeting the relevant state's RE goals under
the Clean Power Plan.
After issuing the supplemental proposal, the EPA held additional consultation with tribes.
On November 18, 2014, the EPA held consultations with the following tribes: Fort McDowell
Yavapai Nation, Fort Mojave Tribe, Hopi Tribe, Navajo Nation, and Ak-Chin Indian
Community. A consultation with the Ute Indian Tribe of the Uintah and Ouray Reservation was
7-15
-------
held on December 16, 2014 and with the Gila River Indian Community on January 15, 2015. The
Navajo Nation reiterated the concerns raised during the previous consultation. Several tribes also
again indicated that they wanted to ensure they would be included in the development of any
tribal or federal plans for areas of Indian country. The Fort Mojave Tribe and the Navajo Nation
expressed concern with using data from 2012 as the basis for the goal for their areas of Indian
country; in their view, that year was not representative for the affected EGU. On April 28, 2015,
the EPA held an additional consultation with the Navajo Nation. The issues raised by the Navajo
Nation during the consultation included whether the EPA has the authority to set less stringent
standards on a case-by-case basis, and a suggested "parity glide path" that would account and
adjust for the very low electricity usage by the Navajo Nation and promote Navajo Nation
economic growth and demand. Furthermore, on July 7, 2015 the EPA conducted an additional
consultation with the Navajo Nation. One of the goals of the consultation was for the new
government of the Navajo Nation to deepen their understanding of the rulemaking. The questions
raised by the nation had to do with goal setting and carbon credits, the timing of the rulemaking,
and the proposed federal plan. Additionally, on July 14, 2015 the EPA conducted an additional
consultation with the Fort Mojave Tribe. The Fort Mojave tribes expressed concerns that 2012 is
not a representative year, that natural gas-fired combined cycle power plants should be treated
differently from coal-fired power plants, and that the proposed goal for Fort Mojave was not
appropriate. Additionally, they also expressed interest in being engaged in the federal plan
process. Responses to these comments and others received are available in the Response to
Comment Document that is in the docket for this rulemaking. As required by section 7(a), the
EPA's Tribal Consultation Official has certified that the requirements of the executive order
have been met in a meaningful and timely manner. A copy of the certification is included in the
docket for this action.
7.7 Executive Order 13045: Protection of Children from Environmental Health Risks
and Safety Risks
This action is subject to Executive Order 13045 (62 FR 19885, April 23, 1997) because it
is an economically significant regulatory action as defined by Executive Order 12866, and the
EPA believes that the environmental health or safety risk addressed by this action has a
disproportionate effect on children. Accordingly, the agency has evaluated the environmental
health and welfare effects of climate change on children.
7-16
-------
CO2 is a potent greenhouse gas that contributes to climate change and is emitted in
significant quantities by fossil fuel-fired power plants. The EPA believes that the COi emission
reductions resulting from implementation of these final guidelines, as well as substantial ozone
and PM2.5 emission reductions as a co-benefit, will further improve children's health.
The assessment literature cited in the EPA's 2009 Endangerment Finding concluded that
certain populations and lifestages, including children, the elderly, and the poor, are most
vulnerable to climate-related health effects. The assessment literature since 2009 strengthens
these conclusions by providing more detailed findings regarding these groups' vulnerabilities
and the projected impacts they may experience.
These assessments describe how children's unique physiological and developmental
factors contribute to making them particularly vulnerable to climate change. Impacts to children
are expected from heat waves, air pollution, infectious and waterborne illnesses, and mental
health effects resulting from extreme weather events. In addition, children are among those
especially susceptible to most allergic diseases, as well as health effects associated with heat
waves, storms, and floods. Additional health concerns may arise in low income households,
especially those with children, if climate change reduces food availability and increases prices,
leading to food insecurity within households.
7.8 Executive Order 13211: Actions Concerning Regulations That Significantly Affect
Energy Supply, Distribution, or Use
This action, which is a significant regulatory action under EO 12866, is likely to have a
significant effect on the supply, distribution, or use of energy. The EPA has prepared a Statement
of Energy Effects for this action as follows. We estimate a 1 to 2 percent change in retail
electricity prices on average across the contiguous U.S. in 2025, and a 22 to 23 percent reduction
in coal-fired electricity generation as a result of this rule. The EPA projects that utility power
sector delivered natural gas prices will increase by up to 2.5 percent in 2030. For more
information on the estimated energy effects, please refer to the economic impact analysis for this
proposal. The analysis is available in the RIA, which is in the public docket.
7.9 National Technology Transfer and Advancement Act (NTT A A)
This rulemaking does not involve technical standards.
7-17
-------
7.10 Executive Order 12898: Federal Actions to Address Environmental Justice in
Minority Populations and Low-Income Populations
Executive Order 12898 (59 FR 7629; February 16, 1994) establishes federal executive
policy on environmental justice. Its main provision directs federal agencies, to the greatest extent
practicable and permitted by law, to make environmental justice part of their mission by
identifying and addressing, as appropriate, disproportionately high and adverse human health or
environmental effects of their programs, policies, and activities on minority populations and low-
income populations in the U.S. The EPA defines environmental justice 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. The EPA has this goal for all communities and persons across this
Nation. It will be achieved when everyone enjoys the same degree of protection from
environmental and health hazards and equal access to the decision-making process to have a
healthy environment in which to live, learn, and work.
Leading up to this rulemaking the EPA summarized the public health and welfare effects
of GHG emissions in its 2009 Endangerment Finding. See, section VIII.A of this preamble
where the EPA summarizes the public health and welfare impacts from GHG emissions that
were detailed in the 2009 Endangerment Finding under CAA section 202(a)(l).195 As part of the
Endangerment Finding, the Administrator considered climate change risks to minority
populations and low-income populations, finding that certain parts of the population may be
especially vulnerable based on their characteristics or circumstances. Populations that were
found to be particularly vulnerable to climate change risks include the poor, the elderly, the very
young, those already in poor health, the disabled, those living alone, and/or indigenous
populations dependent on one or a few resources. See sections XII.F and XII.G, above, where the
EPA discusses Consultation and Coordination with Tribal Governments and Protection of
Children. The Administrator placed weight on the fact that certain groups, including children, the
elderly, and the poor, are most vulnerable to climate-related health effects.
195 "Endangerment and Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the Clean Air
Act," 74 Fed. Reg. 66,496 (Dec. 15, 2009) ("Endangerment Finding").
7-18
-------
The record for the 2009 Endangerment Finding summarizes the strong scientific evidence
in the major assessment reports by the U.S. Global Change Research Program (USGCRP), the
Intergovernmental Panel on Climate Change (IPCC), and the National Research Council (NRC)
of the National Academies that the potential impacts of climate change raise environmental
justice issues. These reports concluded that poor communities can be especially vulnerable to
climate change impacts because they tend to have more limited adaptive capacities and are more
dependent on climate-sensitive resources such as local water and food supplies. In addition,
Native American tribal communities possess unique vulnerabilities to climate change,
particularly those impacted by degradation of natural and cultural resources within established
reservation boundaries and threats to traditional subsistence lifestyles. Tribal communities whose
health, economic well-being, and cultural traditions that depend upon the natural environment
will likely be affected by the degradation of ecosystem goods and services associated with
climate change. The 2009 Endangerment Finding record also specifically noted that Southwest
native cultures are especially vulnerable to water quality and availability impacts. Native
Alaskan communities are already experiencing disruptive impacts, including coastal erosion and
shifts in the range or abundance of wild species crucial to their livelihoods and well-being.
The most recent assessments continue to strengthen scientific understanding of climate
change risks to minority populations and low-income populations in the United States.196 The
new assessment literature provides more detailed findings regarding these populations'
vulnerabilities and projected impacts they may experience. In addition, the most recent
assessment reports provide new information on how some communities of color (more
specifically, populations defined jointly by ethnic/racial characteristics and geographic location)
196 Melillo, Jerry M., Terese (T.C.) Richmond, and Gary W. Yohe, Eds., 2014: Climate Change Impacts in the
United States: The Third National Climate Assessment. U.S. Global Change Research Program, 841 pp.
IPCC, 2014: 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, DJ. Dokken, KJ. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatterjee, 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, 1132pp.
IPCC, 2014: Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part B: Regional Aspects. Contribution
of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Barros,
V.R., C.B. Field, DJ. Dokken, M.D. Mastrandrea, KJ. Mach, T.E. Bilir, M. Chatterjee, 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, 688 pp.
7-19
-------
may be uniquely vulnerable to climate change health impacts in the United States. These reports
find that certain climate change related impactsincluding heat waves, degraded air quality, and
extreme weather eventshave disproportionate effects on low-income populations and some
communities of color, raising environmental justice concerns. Existing health disparities and
other inequities in these communities increase their vulnerability to the health effects of climate
change. In addition, assessment reports also find that climate change poses particular threats to
health, well-being, and ways of life of indigenous peoples in the United States.
As the scientific literature presented above and as the 2009 Endangerment Finding
illustrates, low income populations and some communities of color are especially vulnerable to
the health and other adverse impacts of climate change. The EPA believes that communities will
benefit from this final rulemaking because this action directly addresses the impacts of climate
change by limiting GHG emissions through the establishment of COi emission guidelines for
existing affected fossil fuel-fired EGUs.
In addition to reducing CO2 emissions, the guidelines finalized in this rulemaking would
reduce other emissions from affected EGUs that reduce generation due to higher adoption of
energy efficiency and renewable energy. These emission reductions will include SCh and NOx,
which form ambient PlVb.5 and ozone in the atmosphere, and hazardous air pollutants (HAP),
such as mercury and hydrochloric acid. In the final rule revising the annual PlVh.5 NAAQS,197 the
EPA identified low-income populations as being a vulnerable population for experiencing
adverse health effects related to PM exposures. Low-income populations have been generally
found to have a higher prevalence of pre-existing diseases, limited access to medical treatment,
and increased nutritional deficiencies, which can increase this population's susceptibility to PM-
related effects.198 In areas where this rulemaking reduces exposure to PMi.5, ozone, and
methylmercury, low-income populations will also benefit from such emissions reductions. The
RIA for this rulemaking, included in the docket for this rulemaking, provides additional
197 "National Ambient Air Quality Standards for Particulate Matter, Final Rule," 78 FR 3086 (Jan. 15, 2013).
198 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 - RTF Division. December.
Available on the Internet at .
7-20
-------
information regarding the health and ecosystem effects associated with these emission
reductions.
Additionally, as outlined in the community and environmental justice considerations
section IX of this preamble, the EPA has taken a number of actions to help ensure that this action
will not have potential disproportionately high and adverse human health or environmental
effects on overburdened communities. The EPA consulted its May 2015, Guidance on
Considering Environmental Justice During the Development of Regulatory Actions, when
determining what actions to take.199 As described in the community and environmental justice
considerations section of this preamble the EPA also conducted a proximity analysis, which is
available in the docket of this rulemaking and is discussed in section IX. Additionally, as
outlined in sections I and IX of this preamble, the EPA has engaged with communities
throughout this rulemaking and has devised a robust outreach strategy for continual engagement
throughout the implementation phase of this rulemaking.
7.11 Congressional Review Act (CRA)
This final action is subject to the CRA, and the EPA will submit a rule report to each
House of the Congress and to the Comptroller General of the United States. This action is a
"major rule" as defined by 5 U.S.C. 804(2).
199 Guidance on Considering Environmental Justice During the Development of Regulatory Actions.
http://epa.gov/environmentaljustice/resources/policy/considering-ej-in-rulemaking-guide-final.pdf. May 2015.
7-21
-------
CHAPTER 8: COMPARISON OF BENEFITS AND COSTS
8.1 Comparison of Benefits and Costs
The benefits, costs, and net benefits of the illustrative plan scenarios are presented in this
chapter of the Regulatory Impact Analysis (RIA) for the Final Carbon Pollution Emission
Guidelines for Existing Stationary Sources: Electric Utility Generating Units. As discussed in
Chapter 1, the EPA is establishing carbon dioxide (COi) emission performance rates for two
source categories of existing fossil fuel-fired EGUs, fossil fuel-fired electric utility steam
generating units and stationary combustion turbines. Given the flexibilities afforded states in
complying with the emission guidelines, the benefits, cost and economic impacts reported in this
RIA are not definitive estimates, but are instead illustrative of plan approaches states may take.
The EPA has used the social cost of carbon estimates presented in the Technical Support
Document: Technical Update of the Social Cost of Carbon for Regulatory Impact Analysis
Under Executive Order 12866 (May 2013, Revised July 2015) ("current SC-CO2 TSD") to
analyze CO2 climate impacts of this rulemaking.200 We refer to these estimates, which were
developed by the U.S. government, as "SC-CCh estimates." The SC-CCh is an estimate of the
monetary value of impacts associated with a marginal change in CO2 emissions in a given year.
The four SC-COi estimates are associated with different discount rates (model average at 2.5
percent discount rate, 3 percent, and 5 percent; 95th percentile at 3 percent), and each increases
over time. In this comparison of benefits and costs, the EPA provides the estimate of climate
benefits associated with the SC-COi value deemed to be central in the current SC-COi TSD (the
model average at 3 percent discount rate). In addition to reducing CCh emissions, implementing
these final emission guidelines is expected to reduce emissions of SOi and NOx, which are
precursors to formation of ambient PM2.5, as well as directly emitted fine particles.201 Therefore,
200 Docket ID EPA-HQ-OAR-2013-0495, Technical Support Document: Technical Update of the Social Cost of
Carbon for Regulatory Impact Analysis Under Executive Order 12866, Interagency Working Group on Social Cost
of Carbon, with participation by Council of Economic Advisers, Council on Environmental Quality, Department of
Agriculture, Department of Commerce, Department of Energy, Department of Transportation, Environmental
Protection Agency, National Economic Council, Office of Energy and Climate Change, Office of Management and
Budget, Office of Science and Technology Policy, and Department of Treasury (May 2013, Revised July 2015).
Available at: .
201 We did not estimate the co-benefits associated with reducing direct exposure to SOi and NOx. For this RIA, we
did not estimate changes in emissions of directly emitted particles. As a result, quantified PM2.s related benefits are
8-1
-------
reducing these emissions would also reduce human exposure to ambient PlVb.5 and ozone
precursors, and the associated PlVb.5 and ozone related health effects. Tables 8-1 and 8-2 provide
a summary of the climate benefits, air quality co-benefits, and costs for the illustrative rate-based
and mass-based plan scenarios.
The EPA could not monetize important categories of impacts. Due to current data and
modeling limitations, our estimates of the benefits from reducing COi emissions do not include
important impacts like ocean acidification or potential tipping points in natural or managed
ecosystems. Unquantified impacts also include those associated with changes in emissions of
other pollutants that affect the climate, such as methane. In addition, the analysis does not
quantify co-benefits from reducing exposure to 862, NOX, and hazardous air pollutants (e.g.,
mercury), as well as ecosystem effects and visibility impairment.
Based upon the foregoing discussion, it remains clear that this final rule's combined
climate benefits and human health co-benefits associated with the reduction in other air
pollutants substantially outweigh the costs for both illustrative plan scenarios.
underestimated by a relatively small amount. In the proposal RIA, the benefits from reductions in directly emitted
PM2.s were less than 10 percent of total monetized health co-benefits across all scenarios and years.
8-2
-------
Table 8-1. Monetized Benefits, Costs, and Net Benefits Under the Rate-based
Illustrative Plan Approach (billions of 2011$)a
Rate-Based Scenario
2020 2025 2030
Climate Benefits b
5% discount rate
3% discount rate
2.5% discount rate
95th percentile at
3% discount rate
$0.80
$2.8
$4.1
$8.2
$3.1
$10
$15
$31
Air Quality Co-benefits Discount Rate
$6.4
$20
$29
$61
3% 7% 3% 7% 3% 7%
Air Quality Health $0.70to$1.8 $0.64 to $1.7 $7.4 to $18 $6.7 to $16 $14 to $34 $13 to $31
Co-benefits
Compliance Costsd $2.5 $1.0 $8.4
Net Benefits6 $1.0 to $2.1 $1.0 to $2.0 $17 to $27 $16 to $25 $26 to $45 $25 to $43
Non-monetized climate benefits
Reductions in exposure to ambient NO2 and SO2
Non-Monetized Reductions in mercury deposition
Benefits : l
Ecosystem benefits associated with reductions in emissions of NOx, SO2, PM, and mercury
Visibility improvement
a All are rounded to two significant figures, so figures may not sum.
b The climate benefit estimate in this summary table reflects global impacts from CO2 emission changes and does not
account for changes in non-CO2 GHG emissions. Also, different discount rates are applied to SC-CO2 than to the
other estimates because CO2 emissions are long-lived and subsequent damages occur over many years. The benefit
estimates in this table are based on the average SC-CO2 estimated for a 3 percent discount rate. However we
emphasize the importance and value of considering the full range of SC-CO2 values. As shown in the RIA, climate
benefits are also estimated using the other three SC-CO2 estimates (model average at 2.5 percent discount rate, 3
percent, and 5 percent; 95th percentile at 3 percent). The SC-CO2 estimates are year-specific and increase over time.
cThe air quality health co-benefits reflect reduced exposure to PM25and ozone associated with emission reductions
of SO2 and NOX. The co-benefits do not include the benefits of reductions in directly emitted PM2.s. These
additional benefits would increase overall benefits by a few percent based on the analyses conducted for the
proposed rule. The range reflects the use of concentration-response functions from different epidemiology studies.
The reduction in premature fatalities each year accounts for over 98 percent of total monetized co-benefits from
PM25 and ozone. These models assume that all fine particles, regardless of their chemical composition, are equally
potent in causing premature mortality because the scientific evidence is not yet sufficient to allow differentiation of
effect estimates by particle type.
d Total costs are approximated by the illustrative plan scenario costs estimated using the Integrated Planning Model
for the final emission guidelines and a discount rate of approximately 5 percent. This estimate includes monitoring,
recordkeeping, and reporting costs and demand side energy efficiency program and participant costs.
e The estimates of net benefits in this summary table are calculated using the global SC-CO2at a 3 percent discount
rate (model average). The RIA includes combined climate and health estimates based on additional discount rates.
f Estimates in the table are presented for three analytical years with air quality co-benefits calculated using two
discount rates. The estimates of co-benefits are annual estimates in each of the analytical years, reflecting
discounting of mortality benefits over the cessation lag between changes in PM2.5 concentrations and changes in
risks of premature death (see RIA Chapter 4 for more details), and discounting of morbidity benefits due to the
multiple years of costs associated with some illnesses. The estimates are not the present value of the benefits of the
rule over the full compliance period.
8-3
-------
Table 8-2. Monetized Benefits, Costs, and Net Benefits Under the Mass-based
Illustrative Plan Approach Scenario (billions of 2011$)a
Mass-Based Scenario
2020 2025 2030
Climate Benefits b
5% discount rate
3% discount rate
2.5% discount rate
95th percentile at
$0.94
$3.3
$4.9
$9.7
$3.6
$12
$17
$35
$6.4
$20
$29
$60
Air Quality Co-benefits Discount Rate
3% 7% 3% 7% 3% 7%
Air Quality
Health Co- $2.0 to $4.8 $1.8 to $4.4 $7.1 to $17 $6.5 to $16 $12 to $28 $11 to $26
benefits c
Costsd $1.4 $3.0 $5.1
Net Benefits6 $3.9 to $6.7 $3.7 to $6.3 $16 to $26 $15 to $24 $26 to $43 $25 to $40
Non-monetized climate benefits
Reductions in exposure to ambient NOi and SOi
Non-Monetized Reductions in mercury deposition
Benefits
Ecosystem benefits associated with reductions in emissions of NOx, SOi, PM, and mercury
Visibility improvement
a All are rounded to two significant figures, so figures may not sum.
bThe climate benefit estimate in this summary table reflects global impacts from CO2 emission changes and does not
account for changes in non-CO2 GHG emissions. Also, different discount rates are applied to SC-CO2 than to the
other estimates because CO2 emissions are long-lived and subsequent damages occur over many years. The benefit
estimates in this table are based on the average SC-CO2 estimated for a 3 percent discount rate. However we
emphasize the importance and value of considering the full range of SC-CO2 values. As shown in the RIA, climate
benefits are also estimated using the other three SC-CO2 estimates (model average at 2.5 percent discount rate, 3
percent, and 5 percent; 95th percentile at 3 percent). The SC-CO2 estimates are year-specific and increase over time.
cThe air quality health co-benefits reflect reduced exposure to PM25and ozone associated with emission reductions
of SO2 and NOX. The co-benefits do not include the benefits of reductions in directly emitted PMi.s. These
additional benefits would increase overall benefits by a few percent based on the analyses conducted for the
proposed rule. The range reflects the use of concentration-response functions from different epidemiology studies.
The reduction in premature fatalities each year accounts for over 98 percent of total monetized co-benefits from
PM25 and ozone. These models assume that all fine particles, regardless of their chemical composition, are equally
potent in causing premature mortality because the scientific evidence is not yet sufficient to allow differentiation of
effect estimates by particle type.
d Total costs are approximated by the illustrative plan scenario costs estimated using the Integrated Planning Model
for the final emission guidelines and a discount rate of approximately 5 percent. This estimate includes monitoring,
recordkeeping, and reporting costs and demand side energy efficiency program and participant costs.
e The estimates of net benefits in this summary table are calculated using the global SC-CO2at a 3 percent discount
rate (model average). The RIA includes combined climate and health estimates based on additional discount rates.
f Estimates in the table are presented for three analytical years with air quality co-benefits calculated using two
discount rates. The estimates of co-benefits are annual estimates in each of the analytical years, reflecting
discounting of mortality benefits over the cessation lag between changes in PM2.s concentrations and changes in
risks of premature death (see RIA Chapter 4 for more details), and discounting of morbidity benefits due to the
multiple years of costs associated with some illnesses. The estimates are not the present value of the benefits of the
rule over the full compliance period.
8-4
-------
8.2 Uncertainty Analysis
The Office of Management and Budget's circular Regulatory Analysis (Circular A-4)
provides guidance on the preparation of regulatory analyses required under E.O. 12866, and
requires an uncertainty analysis for rules with annual benefits or costs of $1 billion or more.202
This final rulemaking surpasses that threshold for both benefits and costs. Throughout the RIA,
we considered a number of sources of uncertainty, both quantitatively and qualitatively, on
benefits and costs. We summarize three key elements of our analysis of uncertainty here:
Evaluating uncertainty in the illustrative plan approaches that states will implement,
which influences both costs and benefits.
Assess uncertainty in the methods used to calculate the health co-benefits associated with
the reduction in PMi.5 and ozone and the use of a benefits-per-ton approach in estimating
these co-benefits.
Characterizing uncertainty in monetizing climate-related benefits.
Some of these elements are evaluated using probabilistic techniques, whereas for others
the underlying likelihoods of certain outcomes are unknown and we use scenario analysis to
evaluate their potential effect on the benefits and costs of this rulemaking.
8.2.1 Uncertainty in Costs and Illustrative Plan Approaches
The calculation of the state goals is based on an evaluation of methods for reducing the
carbon emissions intensity of electricity generation that may be achieved at reasonable cost. Our
best estimates of the costs of these methods of intensity reduction are reported within the cost
analysis of this rule and are included in the cost modeling in the RIA.
A source of uncertainty under this regulation is the ultimate approach states will adopt in
response to the guidelines, which will affect both the costs and benefits of this rule. For this
reason we modeled two potential illustrative plan scenarios: the rate-based illustrative plan
scenario and the mass-based illustrative plan scenario.
202 Office of Management and Budget (OMB), 2003, Circular A-4,
http://www.whitehouse.gov/omb/circulars_a004_a-4 and OMB, 2011. Regulatory Impact Analysis: A Primer.
http://www.whitehouse.gov/sites/default/files/omb//n/bre^/regpol/circular-a-4_regulatory-impact-analysis-a-
primer.pdf
8-5
-------
8.2.2 Uncertainty Associated with Estimating the Social Cost of Carbon
The 2010 SC-CO2 TSD noted a number of limitations to the SC-COi analysis, including
the incomplete way in which the integrated assessment models (IAM) capture catastrophic and
non-catastrophic impacts, their incomplete treatment of adaptation and technological change,
uncertainty in the extrapolation of damages to high temperatures, and assumptions regarding risk
aversion.203 Currently integrated assessment models do not assign value to all of the important
physical, ecological, and economic impacts of climate change recognized in the climate change
literature due to a lack of precise information on the nature of damages and because the science
incorporated into these models understandably lags behind the most recent research. These
individual limitations do not all work in the same direction in terms of their influence on the SC-
COi estimates, though taken together they suggest that the SC-CCh estimates are likely
conservative. In particular, the IPCC Fourth Assessment Report (2007) concluded that "It is very
likely that [SC-CCh estimates] underestimate the damage costs because they cannot include
many non-quantifiable impacts" and the IPCC Fifth Assessment report observed that SC-CCh
estimates continue to omit various impacts that would likely increase damages. The 95th
percentile estimate was included in the recommended range for regulatory impact analysis, in
part, to address these concerns.
The modeling underlying the development of the SC-CCh estimates addressed
uncertainty in several ways. An ensemble of three lAMs were used to generate the SC-CCh
estimates to capture differences in model structures that, in part, reflect uncertainty in the
scientific literature about these relationships. Parametric uncertainty was explicitly addressed in
each IAM, though to differing degrees, through Monte Carlo simulations in which explicit
probability distributions for key parameters were specified, including the equilibrium climate
sensitivity, which represents the long-run responsiveness of the climate to increasing GHG
concentrations. Furthermore, the analysis considered five different socioeconomic and emissions
203 Technical Support Document: Social Cost of Carbon for Regulatory Impact Analysis Under Executive Order
12866, Interagency Working Group on Social Cost of Carbon, with participation by the Council of Economic
Advisers, Council on Environmental Quality, Department of Agriculture, Department of Commerce, Department of
Energy, Department of Transportation, Environmental Protection Agency, National Economic Council, Office of
Energy and Climate Change, Office of Management and Budget, Office of Science and Technology Policy, and
Department of Treasury (February 2010). Available at:
.
8-6
-------
forecasts to capture the sensitivity of the SC-CCh estimates to key exogenous projections used in
the modeling. Finally, the results were calculated for three discount rates, which were selected, in
part, to reflect uncertainty about how interest rates may change over time and the possibility that
climate damages are positively correlated with uncertain future economic activity. This analysis
produced 45 different distributions of the SC-CCh estimates for each emissions year. To produce
a range of plausible estimates that are manageable in regulatory analysis but still reflects the
uncertainty in the results four point estimates were recommended. The use of this range of point
estimates in this rulemaking helps to reflect the uncertainty in the SC-CCh estimates. Chapter 4
of this RIA provides a comprehensive discussion about the methodology and application of the
SC-CO2; see both the 2010 TSD and current SC-CO2TSD for a full description.
In addition, OMB's Office of Information and Regulatory Affairs received comments
regarding uncertainty and the SC-COi estimates in response to a separate request for public
comment on the approach used to develop the estimates. Commenters discussed the analyses and
presentation of uncertainty in the TSD as well as the implications of uncertainty for use of the
SC-CO2 estimates in regulatory impact analysis. In their response, the interagency working
group (IWG) acknowledged uncertainty in the SC-CCh estimates but disagreed with commenters
that suggested the uncertainty undermines use of the SC-CCh estimates in regulatory impact
analysis. The IWG went on to note that the uncertainty in the SC-CO2 estimates is fully
acknowledged and comprehensively discussed in the TSDs and supporting academic literature,
and that while all regulatory impact analysis involves uncertainty, these analyses can provide
useful information to decision makers and the public. See the IWG Response to Comments for
the complete response.204
8.2.3 Uncertainty Associated with PM2.5 and Ozone Health Co-Benefits Assessment
Our estimate of the total monetized co-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
PM2.5-related premature mortality, which accounts for 98 percent of the monetized PM2.5 health
co-benefits:
' Seehttps://www.whitehouse.gov/sites/default/files/omb/inforeg/scc-response-to-comments-fmal-july-2015.pdf
8-7
-------
We assume that all fine particles, regardless of their chemical composition, are
equally potent in causing premature mortality. This is an important assumption,
because PMi.5 varies considerably in composition across sources, but the scientific
evidence is not yet sufficient to allow differentiation of effect estimates by particle
type. The PM ISA concluded that "many constituents of PMi.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).
We assume that the health impact function for fine particles is log-linear without a
threshold in this analysis. Thus, the estimates include health co-benefits from
reducing fine particles in areas with varied concentrations of PM2.5, including both
areas that do not meet the fine particle standard and those areas that are in
attainment, down to the lowest modeled concentrations.
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 PMi.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 co-
benefits at different discount rates. EPA quantitatively assessed uncertainty in the air
quality health co-benefits, including probabilistic approaches.
In addition, EPA provides the 95th percentile confidence interval for avoided PM-related
premature deaths and the associated economic valuation using two key epidemiology studies.
EPA provides the PM-related results using alternate concentration-response relationship
provided by an expert elicitation and alternate ozone-related results using concentration-response
relationships provided by alternate epidemiology studies. In addition, we include an assessment
of the distribution of population exposure in the modeling underlying the benefit-per-ton
estimates. For further discussion and characterization of those uncertainties influencing the
benefit assessment, see Chapter 4 of this RIA.
As noted and described in Chapter 4 of this RIA, we use a benefit-per-ton approach to
quantify health co-benefits. All benefit-per-ton estimates have inherent limitations, including that
8-8
-------
the estimates reflect the geographic distribution of the modeled sector emissions, which may not
match the emission reductions anticipated by the final emission guidelines, and they may not
reflect local variability in population density, meteorology, exposure, baseline health incidence
rates, or other local factors for any specific location. In addition, these estimates reflect the
regional average benefit-per-ton for each ambient PMi.5 precursor emitted from EGUs, which
assumes a linear atmospheric response to emission reductions. The regional benefit-per-ton
estimates, although less subject to these types of uncertainties than national estimates, still
should be interpreted with caution. Even though we assume that all fine particles have equivalent
health effects, the benefit-per-ton estimates vary between precursors depending on the location
and magnitude of their impact on PlVb.5 levels, which drive population exposure.
8.3 References
Docket ID EPA-HQ-OAR-2013-0602, Technical Support Document: Technical Update of the
Social Cost of Carbon for Regulatory Impact Analysis Under Executive Order 12866,
Interagency Working Group on Social Cost of Carbon, with Participation by Council of
Economic Advisers, Council on Environmental Quality, Department of Agriculture,
Department of Commerce, Department of Energy, Department of Transportation, Domestic
Policy Council, Environmental Protection Agency, National Economic Council, Office of
Management and Budget, Office of Science and Technology Policy, and Department of
Treasury (May 2013, Revised July 2015). Also available at:
.
Accessed July 15,2015.
Interagency Working Group on Social Cost of Carbon, with participation by Council of
Economic Advisers, Council on Environmental Quality, Department of Agriculture,
Department of Commerce, Department of Energy, Department of Transportation,
Environmental Protection Agency, National Economic Council, Office of Management and
Budget, Office of Science and Technology Policy, and Department of Treasury. Response to
to Comments: Social Cost of Carbon for Regulatory Impact Analysis Under Executive Order
12866. July 2015. Available at
Accessed July 15, 2015.
8-9
-------
Intergovernmental Panel on Climate Change (IPCC). 2007. Climate Change 2007: Synthesis
Report Contribution of Working Groups I, II and III to the Fourth Assessment Report of the
IPCC. Available at
. Accessed June 6, 2015.
National Research Council (NRC). 2002. Estimating the Public Health Benefits of Proposed Air
Pollution Regulations. National Academies Press. Washington, DC.
U.S. Environmental Protection AgencyScience 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
. Accessed June 4, 2015.
U.S. Environmental Protection Agency (U.S. EPA). 2009. Integrated Science Assessment for
Paniculate Matter (Final Report). EPA-600-R-08-139F. National Center for Environmental
Assessment - RTP Division, Research Triangle Park, NC. Available at
. Accessed June 4, 2015.
8-10
-------
United States Office of Air Quality Planning and Standards Publication No.
Environmental Health and Environmental Impacts Division EPA-452/R-15-003
Protection Agency Research Triangle Park, NC August 2015
------- |