-------
0.15%-
0.10%-
0.05%-

O)
bO
c

0.00%-

O

a?

-0.05%-
-0.10%-
-0.15%-

2026	2031	2036	2041

Year

Figure 5-6 Percent Change in Economy-wide Sectoral Output (All Sectors)

5.2.4.4 Output Price Impacts

Figure 5-7 presents the percent changes in real output prices for each sector in the SAGE
model in 2026, 2031, 2036, and 2041. CGE models report prices in relative terms.150 The largest
percent changes in real output prices occur in the natural gas, electricity, and coal sectors. The
estimated change in the electricity sector output price reflects the additional costs associated with
complying with the final rules as well as demand side reductions in electricity use from both
firms and households. Estimated price changes for natural gas and coal largely reflect the
changes in demand for those fuel types in the electricity sector.

150 Here, we denominate output prices in terms of the consumer price index (CPI) internal to the SAGE model,
which reflects the overall change in end-use prices for the bundle of goods demanded by households.
Characterizing prices relative to the CPI allows a comparison of changes in the magnitude of output prices to
overall trends in the economy (i.e., a percentage change that is positive reflects a price that increases more than
the average price changes across the economy).

5-25


-------
Healthcare services
Services

Truck transportation
T ransportation
Other manufacturing

Transportation equipment manufacturing
Electronics and technology manufacturing
Fabricated metal product manufacturing
Primary metal manufacturing
Cement manufacturing

Plastics and rubber products manufacturing

o

Chemical manufacturing

-------
explicitly accounting for labor market transitions to a new equilibrium may have minimal impact
on the aggregate welfare changes associated with new regulations, though the author notes that
this is a function of the transition dynamics assumed in the model. Slower transition dynamics
may widen the gap between social cost measures with and without accounting for short-term
transition dynamics in the labor market. Hafstead and Williams (2018) develop a two-sector
CGE model that incorporates several wage-setting mechanisms where the adjustment costs from
transitioning between unemployment and employment are realized at much smaller time steps
than are typical in a CGE framework. The authors estimate that the net employment impacts of
environmental policy may be small due to the offsets in the labor demand by unregulated sectors.

Figure 5-8 presents the percent change in net labor demand across the economy in 2026,
2031, 2036, and 2041. Shifts in aggregate labor demand are expected to occur as some sectors
require fewer hours worked, some require more hours worked, and wage rates adjust to ensure
there is adequate labor being voluntarily supplied by households to meet firms' demand for
labor. In model years 2026 and 2031, the model estimates a small aggregate increase in the labor
supply to accommodate additional labor demand across the economy needed to support
additional investments occurring in anticipation of the final regulatory requirements. In
subsequent model years expected reductions in output and investment result in small decreases in
labor supply. Figure 5-9 presents the estimated percent change in labor demand by electricity,
coal, and natural gas sectors in 2026, 2031, 2036, and 2041. In these sectors, changes in labor
demand are generally reflective of the estimated output changes.

Figure 5-10 presents the percent change in sectors other than electricity, natural gas, and
coal for 2026, 2031, 2036, and 2041. The increase in the labor supply in 2026 and 2031 is driven
by increases in demand for labor in sectors associated with capital formation (e.g., construction,
cement manufacturing) to support new investments.

5-27


-------
0.15%-

0.10%-

0.05%-

O)
bo

-0.05%-

-0.10%-

-0.15%

2026

2031

2036

2041

Time

Figure 5-8 Percent Change in Economy-wide Labor Demand (All Sectors)

Electric power'

0 Coal mining'

Natural gas'

"i	1	1		1	1	 -i	1	1		1	1	 •—i	1	1		1	1	 -i	1	1		1	1—

3% -2% -1% 0% 1% 2% -3% -2% -1% 0% 1% 2% -3% -2% -1% 0% 1% 2% -3% -2% -1% 0% 1% 2%

% Change

Figure 5-9 Percent Change in Labor Demand (Electricity, Coal, Natural Gas)

5-28


-------
T ransportation

Water, sewage, arid o
ther utilities

Healthcare services

Food and beverage ma
nufacturing

Petroleum refineries

Chemical manufacturi

ng

Metal ore and nonmet
alic mineral mining

Truck transportation

u Wood product manufac

o	turing
u

i>	Crude oil

Agriculture, forestr
y, fishing and hunting

Plastics and rubber
products manufacturing

Fabricated metal pro
duct manufacturing

Other manufacturing

Cement manufacturing

Construction

Primary metal manufa
cturing

Transportation equip
ment manufacturing

Electronics and tech
nology manufacturing

D

-0.2% -0.1% 0.0% 0.1% -0

3.1% -0.2% -0.1%

. 1% -0.2% -0.1% 0.0% 0.1%

% Change

Figure 5-10 Percent Change in Labor Demand (Rest of Economy)

5.2.4.6 Household Distributional Impacts

The social costs of regulation are ultimately borne by households through changes in
final goods prices or changes in labor, capital, and resource income. SAGE models
representative households by income quintiles in each of the four Census regions. This allows
the social costs to be separately estimated across the income distribution and for different regions

5-29


-------
of the country, as presented in Figure 5-11.151 In general, the annualized household costs increase
with income and are expected to be highest in the Western Census region and lowest in the
Southern Census Region.

t

!

Regions

Midwest
0 Northeast

South
A West

<30k	30-50k	50-70k	70-150k

Income Quintile

>150k

Figure 5-11 Distribution of General Equilibrium Social Costs

Estimates in Figure 5-11 reflect a combined effect of the final rules' requirements and
interactions with IRA subsidies that are expected to see increased use in response to the final
rules. A regulation may affect the value of government expenditures through relative prices of
goods and services purchased by the government. In addition, it may affect tax revenues through
impacts on the value of the base for ad valorem taxes (e.g., labor and capital taxes). In these
cases, a CGE model must implement a closure rule to ensure that the government has the funds
necessary to support its expenditures. A common assumption in CGE models is to balance the
government's budget through lump sum transfers between households and the government as a

151 Distributional cost estimates are annualized for the period 2024 to 2047 and divided by the total number of
households of a given income quintile and region using 2016 estimates from the Census' Current Population
Survey.

5-30


-------
non-distortionary approach to closing the model. This is the approach used in the SAGE model.
Given uncertainties in the accounting for the IRA subsidies in this analysis, we are unable to
determine the relative role of this effect in the distributional estimates at this time.

5.2.5 Limitations to Analysis

The SAGE model and methodology for aligning IPM outputs for use as inputs in SAGE
reflect the best available science for conducting economy-wide modeling of the final rules.
However, both the use of SAGE in a regulatory analysis and the framework for linking IPM with
the SAGE model are subject to uncertainty and limitations:

•	The costs of complying with existing regulations are largely reflected in the social
accounting matrix and in projections used to calibrate the SAGE model but are not
distinguished from non-regulatory related costs (i.e., there is no explicit characterization
of already existing regulations in the constructed baseline). Data underlying the SAGE
baseline ranges from 2016 to 2020, depending on the specific source. As a result, recent
changes in the economy, including new regulations, may not be captured in the source
data used to calibrate the model's baseline. For these reasons, SAGE may not explicitly
capture interactions that the final rules may have with compliance activities already
underway to meet existing regulatory requirements.

•	Since IPM provides inputs for this SAGE analysis, the SAGE estimates are subject to
many of the same uncertainties and limitations of the IPM methodology, which are
detailed in Section 3. In particular, this economy-wide analysis focuses on a single
illustrative compliance scenario for these final rules.

•	The methodology used to align IPM and SAGE accounts for partial equilibrium
feedbacks in IPM and represents an improvement over assuming the solution of one
model directly in the other. While a full model linkage, where the models iteratively pass
information back and forth until jointly converging to an equilibrium, may provide a
more complete representation of the economy-wide impacts of the final rules, it is
challenging to implement and not feasible at this time.

•	To align IPM outputs for use as SAGE inputs, we target the estimated change in
amortized payments to capital. However, because the representation of capital differs

5-31


-------
between IPM and SAGE, the projected stream of capital investments in response to the
final rules also likely differs between the two models. See Section 5.2.3.3 for a discussion
of this choice.

Production with extant and new capital in SAGE is not equivalent to differentiating
existing and new generation in the IPM modeling framework. This analysis assigns
incremental costs on existing generators in IPM to production with extant capital in
SAGE until model year 2051, and on production with new capital thereafter, as extant
capital in the electricity sector is mostly depreciated by that point. Extant capital in
SAGE is assumed to be relatively inflexible in its ability to accommodate changes in
production processes when compared to new capital. Therefore, it is possible that the
linked framework may over- or under-attribute incremental costs to less flexible
production processes in SAGE.

Given the level of sectoral aggregation in SAGE, subsidies on specific electricity-sector
technologies are reflected in the SAGE model through a sector-wide adjustment in output
taxes. This sector-wide adjustment is designed to approximate subsidies levied on
specific technologies but may add a degree of uncertainty to the social cost estimate
regarding the degree to which the subsidies interact with pre-existing distortions in the
economy. Furthermore, this treatment of subsidies is subject to additional uncertainties
related to the effective magnitude of these payments.

The purpose of this analysis is to quantify the social cost and the economy-wide impacts
of the final rules. To the extent possible, the analysis models the potential interactions
between the final rules and IRA subsidies, but it is beyond the scope of this RIA to
evaluate the social costs and benefits of the IRA subsidies in their entirety. Additional
effects of the IRA, as they relate to the final rules, beyond the specific subsidies modeled
in this RIA could result in a change in estimated social costs and other economy-wide
impacts.

SAGE assumes perfect competition within each sector, a standard assumption in CGE
modeling used to ensure tractability. However, market power is itself a distortion because
it moves private behavior away from the economically efficient outcome. Environmental
regulations can also potentially affect the number of producers and the market structure

5-32


-------
of the regulated sector by raising production costs, modifying economies of scale, or
affecting barriers to entry. A more concentrated market can result in higher prices and
lower output, increasing the social cost of a regulation relative to what is estimated under
an assumption of perfect competition.

• The economy-wide analysis is limited to an evaluation of social costs. SAGE does not
currently estimate changes in emissions nor account for environmental benefits. The SAB
(U.S. EPA 2017) noted that CGE models "have not achieved their potential for analysis
of the benefits of air regulations" because they do not account for potential interactions
between costs and benefits. While this means that estimates from SAGE - and CGE
models generally - are a partial representation of the total effects of regulation on the
economy, the SAB stated that this "does not invalidate the use of CGE models to estimate
costs."

5.3 Small Entity Analysis
5.3.1 Overview

For the final rules, EPA performed a small entity screening analysis for impacts on all
affected EGUs by comparing compliance costs to historic revenues at the ultimate parent
company level. This is known as the cost-to-revenue or cost-to-sales test, or the "sales test." The
sales test is an impact methodology EPA employs in analyzing entity impacts as opposed to a
"profits test," in which annualized compliance costs are calculated as a share of profits. The sales
test is frequently used because revenues or sales data are commonly available for entities
impacted by EPA regulations, and profits data normally made available are often not the true
profit earned by firms because of accounting and tax considerations. Also, the use of a sales test
for estimating small business impacts for a rulemaking is consistent with guidance offered by
EPA on compliance with the Regulatory Flexibility Act (RFA)152 and is consistent with guidance
published by the U.S. Small Business Administration's (SBA) Office of Advocacy that suggests

152 The RFA compliance guidance to EPA rule writers can be found at



5-33


-------
that cost as a percentage of total revenues is a metric for evaluating cost increases on small
entities in relation to increases on large entities.153

5.3.2 EGU Small Entity Analysis and Results

This section presents the methodology and results for estimating the impact of the New
Source Performance Standards for Greenhouse Gas Emissions from New, Modified, and
Reconstructed Fossil Fuel-Fired Electric Generating Units on small EGU entities in 2035 based
on the following endpoints:

•	annual economic impacts of the final rule on small entities, and

•	ratio of small entity impacts to revenues from electricity generation.

This rule would affect the buildout and operation of future NGCC and NGCT additions.
Costs are projected to peak in 2035, which is consistent with the imposition of the second phase
of the NSPS requirements on new NGCC builds, and as such, the analysis focuses on this year.
While IPM can provide important information about the future operation and addition of natural
gas capacity over the analysis period, the model does not project actions taken by individual
firms. Hence, as a proxy for the future gas capacity built by small entities EPA assumed that the
same small entities identified using the process outlined below would continue to build the same
share of future capacity additions projected by IPM over the forecast period. EPA reviewed
historical data and planned builds since 2017 to determine the universe of NGCC and NGCT
additions as outlined in EPA National Electric Energy Data System (NEEDS) v.7 database. The
NEEDS database includes operational capacity in the year of publication as well as capturing
planned/committed units that are likely to come online because ground has been broken,
financing obtained, or other demonstrable factors indicate a high probability that the unit will be
built before June 30, 2028.154

153	See U.S. SBA Office of Advocacy. (2017). A Guide For Government Agencies: How To Comply With The
Regulatory Flexibility Act. Available at: https://advocacy.sba.gOv/2017/08/31/a-guide-for-government-agencies-
how-to-comply-with-the-regulatory-flexibility-act

154	For details please see Chapter 4.3 IPM base case documentation, available at:
https://www.epa. gov/power-sector-modeling

5-34


-------
Based on these criteria, EPA identified a total of 58 GW of NGCC and 11 GW of NGCT
units greater than 25 MW built since 2017. Next, we determined power plant ownership
information, including the name of associated owning entities, ownership shares, and each
entity's type of ownership. Ownership information for these assets was obtained primarily using
data from Ventyx155, supplemented by research using S&P156 and publicly available data.

Majority owners of power plants with affected EGUs were categorized as one of the
seven ownership types.157 These ownership types are:

1.	Investor-Owned Utility (IOU): Investor-owned assets (e.g., a marketer, independent
power producer, financial entity) and electric companies owned by stockholders, etc.

2.	Cooperative (Co-Op): Non-profit, customer-owned electric companies that generate
and/or distribute electric power.

3.	Municipal: A municipal utility, responsible for power supply and distribution in a small
region, such as a city.

4.	Sub-division: Political subdivision utility is a county, municipality, school district,
hospital district, or any other political subdivision that is not classified as a municipality
under state law.

5.	Private: Similar to an investor-owned utility, however, ownership shares are not openly
traded on the stock markets.

6.	State: Utility owned by the state.

7.	Federal: Utility owned by the federal government.

Next, EPA used the D&B Hoover's online database, the Ventyx database, and the S&P
database to identify the ultimate owners of power plant owners identified in the NEEDS
database. This was necessary, as many majority owners of power plants (listed in Ventyx) are
themselves owned by other ultimate parent entities (listed in D&B Hoover's).158 In these cases,

155	The Ventyx Energy Velocity Suite database consists of detailed ownership and corporate affiliation information
at the EGU level. For more information, see: www.ventyx.com.

156	The S&P database consists of detailed ownership and corporate affiliation information at the EGU level. For
more information, see: www.capitaliq.spglobal.com

157	Throughout this analysis, EPA refers to the owner with the largest ownership share as the "majority owner" even
when the ownership share is less than 51 percent.

158	The D&B Hoover's online platform includes company records that can contain NAICS codes, number of
employees, revenues, and assets. For more information, see: https://www.dnb.com/products/marketing-
sales/dnb -hoovers, html.

5-35


-------
the ultimate parent entity was identified via D&B Hoover's, whether domestically or
internationally owned.

EPA followed SBA size standards to determine which non-government ultimate parent
entities should be considered small entities in this analysis. These SBA size standards are
specific to each industry, each having a threshold level of either employees, revenue, or assets
below which an entity is considered small.159 SBA guidelines list all industries, along with their
associated North American Industry Classification System (NAICS) code160 and SBA size
standard. Therefore, it was necessary to identify the specific NAICS code associated with each
ultimate parent entity in order to understand the appropriate size standard to apply. Data from
D&B Hoover's was used to identify the NAICS codes for most of the ultimate parent entities. In
many cases, an entity that is a majority owner of a power plant is itself owned by an ultimate
parent entity with a primary business other than electric power generation. Therefore, it was
necessary to consider SBA entity size guidelines for the range of NAICS codes listed in Table
5-5. This table represents the range of NAICS codes and areas of primary business of ultimate
parent entities that are majority owners of potentially affected EGUs in the historical record.

159	SBA's table of size standards can be located here: https://www.sba.gov/document/support-table-size-standards.

160	North American Industry Classification System can be accessed at the following link:
https://www.census.gov/naics/

5-36


-------
Table 5-5 SBA Size Standards by NAICS Code

NAICS Codes

NAICS U.S. Industry Title

Size
Standards
(number of
employees)

221111

Hydroelectric Power Generation

750

221112

Fossil Fuel Electric Power Generation

950

221113

Nuclear Electric Power Generation

1,150

221114

Solar Electric Power Generation

500

221115

Wind Electric Power Generation

1,150

221116

Geothermal Electric Power Generation

250

221117

Biomass Electric Power Generation

550

221118

Other Electric Power Generation

650

221121

Electric Bulk Power Transmission and Control

950

221122

Electric Power Distribution

1,100

221210

Natural Gas Distribution

1,150

Note: This table is an example of the NAICS codes that comprised this analysis. For a complete
list, please see the accompanying workbook. Based on size standards available at the following
link: https://www.sba.gov/document/support--table-size-standards). Source: SBA, 2023.

EPA compared the relevant entity size criterion for each ultimate parent entity to the SBA
size standard noted in Table 5-2. We used the following data sources and methodology to
estimate the relevant size criterion values for each ultimate parent entity:

1.	Employment, Revenue, and Assets: EPA used the D&B Hoover's database as the
primary source for information on ultimate parent entity employee numbers, revenue, and
assets.161 In parallel, EPA also considered estimated revenues from affected EGUs based
on analysis of IPM estimates for the baseline for 2035. EPA assumed that the ultimate
parent entity revenue was the larger of the two revenue estimates. In limited instances,
supplemental research was also conducted to estimate an ultimate parent entity's number
of employees, revenue, or assets.

2.	Population: Municipal entities are defined as small if they serve populations of less than
50,000.162 EPA primarily relied on data from the Ventyx database and the U.S. Census
Bureau to inform this determination.

161	Estimates of sales were used in lieu of revenue estimates when revenue data was unavailable.

162	The Regulatory Flexibility Act defines a small government jurisdiction as the government of a city, county,
town, township, village, school district, or special district with a population of less than 50,000

(5 U.S.C. section 601(5)). For the purposes of the RFA, States and Tribal governments are not
considered small governments. EPA's Final Guidance for EPA Rulewriters: Regulatory Flexibility Act is located
here: https://www.epa.gov/sites/default/files/2015-06/documents/guidance-regflexact.pdf.

5-37


-------
Ultimate parent entities for which the relevant measure is less than the SBA size standard
were identified as small entities and carried forward in this analysis. Using this analysis, EPA
identified 20 percent of the NGCC and 31 percent of the NGCT additions over the historical
period were attributed to small entities as summarized in Table 5-6 below.

Table 5-6 Historical NGCC and N<

jCT Additions (2017-present)

Capacity Type

Total Additions
(GW)

Total Additions by Small
Entities (GW)

Share of Small Entities to
Total Build (%)

NGCC

57.9

11.5

20%

NGCT*

10.8

3.4

31%

Notes: (1) One small entity accounts for 1 GW of these builds (and owns 2.3 GW of currently operating capacity).
(2) As the scope of the FRFA is limited to the new source performance standards, this table presents small business
impacts for the capacity types of NGCC and NGCT. Small entities also own other capacity types not covered in the
scope of the FRFA.

In run year 2035, a new NGCC addition can comply with the final rule by implementing
efficiency improvements (if it operates at an annual capacity factor of below 50 percent),
installing CCS, or co-firing hydrogen. A new NGCT addition can comply with the final rule
through implementing efficiency improvements (if it operates at an annual capacity factor of
below 20 percent), installing CCS or co-firing hydrogen. The chosen compliance strategy will be
primarily a function of the unit's marginal control costs and its position relative to the marginal
control costs of other units.

To attempt to account for each potential control strategy, EPA estimates compliance costs
as follows:

Ccompliance A COperating+Retrofit A CFuel A R

where C represents a component of cost as labeled163, and A R represents the change in revenues,
calculated as the difference in value of electricity generation between the baseline case and the
rule in in 2035 for projected NGCC and NGCT additions (calculated separately), when the
second phase of the NSPS is assumed to be active under the final rule.

Realistically, compliance choices and market conditions can combine such that an entity
may actually experience a reduction in any of the individual components of cost. Under the rule,

163 Retrofit costs include the costs of installation of CCS.

5-38


-------
some units will generate less electricity (and thus revenues), and this impact will be lessened on
these entities by the projected increase in electricity prices under the rule. On the other hand,
those units increasing generation levels will see an increase in electricity revenues and as a
result, lower net compliance costs. If entities are able to increase revenue more than an increase
in fuel cost and other operating costs, ultimately, they will have negative net compliance costs
(or increased profit). Because this analysis evaluates the total costs along each of the compliance
strategies laid out above for each entity, it inevitably captures gains such as those described. As a
result, what we describe as cost is a measure of the net economic impact of the rule on small
entities.

For this analysis, EPA used IPM output to estimate costs based on the parameters above,
at the unit level. These impacts were then summed for each small entity, adjusting for ownership
share. Net impact estimates were based on the following: operating and retrofit costs, and the
change in fuel costs or electricity generation revenues under the final rule relative to the baseline.
These individual components of compliance costs were estimated as follows:

1.	Operating and retrofit costs (A Coperating+Retrofit)'. The change in operating and retrofit
costs under the final rule was estimated by taking the difference in projected FOM, VOM
and retrofit capital expenditures between the IPM estimates for the final rule and the
baseline for the NGCT and NGCC additions projected by the model.

2.	Fuel costs (A Cfubi): The change in fuel expenditures under the final rule was estimated
by taking the difference in projected fuel expenditures between the IPM estimates for the
final rule and the baseline for the NGCT and NGCC additions projected by the model.

3.	Revenue: To estimate the value of electricity generated, the projected level of electricity
generation is multiplied by the regional wholesale electricity price ($/MWh) projected by
IPM, and the accredited capacity multiplied by the projected regional capacity price
projected by IPM for the NGCT and NGCC additions projected by the model. The
difference between this value under the baseline and the final rule constitutes the
estimated change in revenue.

5-39


-------
Once the costs of the rule were calculated in the manner described above, the costs
attributed to small entities were calculated by multiplying the total costs to the share of the
historical build attributed to small entities. These costs were then shared to individual entities
using the ratio of their build to total small entity additions in the historical dataset.

Under the compliance modeling for the final rule, NGCT additions and dispatch are
higher as a result of reductions in existing coal-fired EGU capacity and generation. As a result,
economic NGCT additions experience negative compliance costs in 2035. Under the compliance
modeling for the final rule, economic NGCC additions dispatch at lower levels relative to the
baseline when the second phase of the NSPS is active. As such, they experience positive
compliance costs.

As indicated above, the use of a sales test for estimating small business impacts for a
rulemaking is consistent with guidance offered by EPA on compliance with the RFA and is
consistent with guidance published by the SBA's Office of Advocacy that suggests that cost as a
percentage of total revenues is a metric for evaluating cost increases on small entities in relation
to increases on large entities. The potential impacts, including compliance costs, of the final rule
on NGCCs owned by small entities are summarized in Table 5-7. All costs are presented in 2019
dollars. EPA estimated the annual net compliance cost to small entities to be approximately $46
million in 2035.

Table 5-7 Projected Impact of the Final Rule on Small Entities in 2035





Total Net





EGU
Ownership
Type

Number of

Compliance

Number of Small Entities

Number of Small Entities

Potentially
Affected

Cost
($2019

with Compliance Costs
>=1% of Generation

with Compliance Costs
>=3% of Generation

Entities

millions)

Revenues

Revenues

Private

8

27

3

0

Co-op

5

18

0

0

Municipal

1

1

0

0

Total

14

46

3

0

Source: IPM analysis

EPA assessed the economic and financial impacts of the rule using the ratio of
compliance costs to the value of revenues from electricity generation, focusing in particular on
entities for which this measure is greater than 1 percent. Of the 14 entities that own NGCC units
considered in this analysis, three are projected to experience compliance costs greater than or

5-40


-------
equal to 1 percent of generation revenues in 2035 and none are projected to experience
compliance costs greater than or equal to 3 percent of generation revenues in 2035.

5.4 Labor Impacts

This section discusses potential employment impacts of these regulations in the power
and related fuel sectors. As economic activity shifts in response to a regulation, typically there
will be a mix of declines and gains in employment in different parts of the economy over time
and across regions. To present a complete picture, an employment impact analysis will describe
the potential positive and negative changes in employment levels, in the power and related fuel
sectors. There are significant challenges when trying to evaluate the employment effects of an
environmental regulation due to a wide variety of other economic changes that can affect
employment, including the impact of the coronavirus pandemic on labor markets and the state of
the macroeconomy generally. Considering these challenges, we look to the economics literature
to provide a constructive framework and empirical evidence. In this section, we focus on impacts
on labor demand related to compliance behavior, providing detailed first-order estimates of
changes in construction and non-recurring construction labor utilization for pollution control
equipment and different capacity and fuel types. This analysis is a complement to the economy-
wide analysis provided in Section 5.2, which projects medium to long run shifts in the expected
use of labor across aggregate sectors as a result of the final rules.

Economic theory of labor demand indicates that employers affected by environmental
regulation may increase their demand for some types of labor, decrease demand for other types,
or for still other types, not change their demand at all (Berman and Bui, 2001; Deschenes, 2018;
Morgenstern et al., 2002). To study labor demand impacts empirically, a growing literature has
compared employment levels at facilities subject to an environmental regulation to employment
levels at similar facilities not subject to that environmental regulation; some studies find no
employment effects, and others find significant differences. For example, see Berman and Bui
(2001), Curtis (2018, 2020), Deschenes (2018), Ferris et al. (2014), Greenstone (2002), and
Morgenstern et al. (2002).

A variety of conditions can affect employment impacts of environmental regulation,
including baseline labor market conditions and employer and worker characteristics such as

5-41


-------
occupation and industry. Changes in employment may also occur in different sectors related to
the regulated industry, both upstream and downstream, or in sectors producing substitute or
complimentary products. Environmental regulation may also affect labor supply through changes
in worker health and productivity (Zivin and Neidell, 2018). We focus our labor impacts analysis
primarily on the directly regulated facilities, with an extension to other EGUs and related fuel
markets.

This section discusses and projects potential employment impacts for the utility power,
coal and natural gas production sectors that may result from the final rules. EPA has a long
history of analyzing the potential impacts of air pollution regulations on changes in the amount
of labor needed in the power generation sector and closely related sectors. The analysis
conducted for this RIA builds upon the approaches used in the past and takes advantage of newly
available data to improve the assumptions and methodology.164

The results presented in this section are based on a methodology that estimates
employment impacts based on differences in projections between two modeling scenarios: the
baseline scenario, and a scenario that represents the implementation of the final rules. The
estimated employment difference between these scenarios can be interpreted as the incremental
effect of the rules. As discussed in Section 3, there is uncertainty related to the future baseline
projections. Note that there is also uncertainty related to the employment factors applied in this
analysis, particularly factors informing job-years related to relatively new technologies, such as
energy storage, on which there is limited data to base assumptions.

Like previous analyses, this analysis represents an evaluation of "first-order employment
impacts" using a sectoral modeling approach. It includes some of the potential ripple effects of
these impacts on the broader economy. While these potential ripple effects include the secondary
job impacts on upstream fuel sectors including coal, natural gas, and uranium, the analysis does
not account for impacts on other fuel sectors, nor does it analyze potential impacts related to
transmission or distribution. This approach excludes the economy-wide employment effects of

164 For a detailed overview of this methodology, including all underlying assumptions, see the U.S. EPA
Methodology for Power Sector-Specific Employment Analysis, available in the docket.

5-42


-------
changes to energy markets (such as higher or lower forecasted electricity prices).165 This
approach also excludes labor impacts that are sometimes reflected in a benefits analysis for an
environmental policy, such as increased productivity from a healthier workforce and reduced
absenteeism due to fewer sick days of employees and dependent family members (e.g., children).

5.4.1 Overview of Methodology

The methodology includes the following two general approaches, based on the available
data. The first approach utilizes the rich employment data that is available for several types of
generation technologies in the 2020 U.S. Energy and Employment Report.166 Detailed
employment inventory data is available regarding recent employment related to coal, hydro,
natural gas, geothermal, wind, and solar generation technologies. The data enables the creation
of technology-specific factors that can be applied to model projections of capacity (reported in
megawatts, or MW) and generation (reported in megawatt-hours, or MWh) in order to estimate
impacts on employment. Since employment data is only available in aggregate by fuel type, it is
necessary to disaggregate by labor type in order to differentiate between types of jobs or tasks for
categories of workers. For example, some types of employment remain constant throughout the
year and are largely a function of the size of a generator, e.g., fixed operation and maintenance
activities, while others are variable and are related to the amount of electricity produced by the
generator, e.g., variable operation and maintenance activities. The approach can be summarized
in three basic steps:

• Quantify the total number of employees by fuel type in a given year;

165	Section 5.2.4.5 provides estimates of sectoral changes in employment accounting for consequent change in the
economy. Relative to the sectoral analysis in this section estimating sectoral employment impacts in the power
and fuels sectors, the economy-wide analysis sheds further light on the medium to longer run labor reallocation
across the economy in response to the rules, although without the same resolution at the level of capacity,
pollution control, and fuel type.

166	While more recent data is available in the 2023 version of this report, this section of the RIA utilizes 2019 data
because this year does not reflect any short-term trends related to the coronavirus pandemic. The 2023 report
states that: "In 2020, the energy sector was deeply impacted by the COVID-19 pandemic and subsequent
economic fallout. The energy sector lost nearly 840,000 jobs, contracting at a faster rate than jobs economy-
wide. Last year's United States Energy and Employment Report (USEER) showed that, by the end of 2020, the
energy sector was beginning to rebound, adding back 560,000 jobs. While the energy sector as a whole has not
recovered all of the jobs lost in 2020, nearly all technologies added energy jobs in 2021. Employment in
transmission, distribution, and storage; energy efficiency; and motor vehicles increased across all technologies.
However, energy jobs in the fuels category declined in 2021." The annual report is available at:
https://www.usenergyjobs.org/.

5-43


-------
•	Estimate total fixed operating & maintenance (FOM), variable operating & maintenance
(VOM), and capital expenditures by fuel type in that year; and

•	Disaggregate total employees into three expenditure-based groups and develop factors for
each group (FTE/MWh, FTE/MW-year, FTE/MW new capacity).

For employment related to electric power generation other than coal, hydro, natural gas,
geothermal, wind and solar, as well as employment required by pollution control technologies,
detailed employment data is not available. Thus, EPA implements a second approach that utilizes
information available in the U.S. Economic Census. These data are used to estimate labor
impacts using labor intensity ratios. These factors provide a relationship between employment
and economic output and are used to estimate employment impacts related to construction and
operation of pollution control retrofits, as well as some types of electric generation technologies.

For a detailed overview of this methodology, including all underlying assumptions and
the types of employment represented by this analysis, see the U.S. EPA Methodology for Power
Sector-Specific Employment Analysis, available in the docket.

5.4.2 Overview of Power Sector Employment

In this section we focus on employment related to electric power generation, as well as
coal and natural gas extraction because these are the segments of the power sector with available
data that are relevant to the projected impacts of the rule. Other segments not discussed here
include the extraction or production of other fuels (e.g., hydrogen), energy efficiency, and
transmission, distribution, and storage. The statistics presented here are based on the 2020
USEER, which reports data from 2019.167

167 While more recent data is available in the 2023 version of this report, this section of the RIA utilizes 2019 data
because this year does not reflect any short-term trends related to the coronavirus pandemic. The 2023 report
states that: "In 2020, the energy sector was deeply impacted by the COVID-19 pandemic and subsequent
economic fallout. The energy sector lost nearly 840,000 jobs, contracting at a faster rate than jobs economy-
wide. Last year's United States Energy and Employment Report (USEER) showed that, by the end of 2020, the
energy sector was beginning to rebound, adding back 560,000 jobs. While the energy sector as a whole has not
recovered all of the jobs lost in 2020, nearly all technologies added energy jobs in 2021. Employment in
transmission, distribution, and storage; energy efficiency; and motor vehicles increased across all technologies.
However, energy jobs in the fuels category declined in 2021." The annual report is available at:
https://www.usenergyjobs.org/.

5-44


-------
In 2019, the electric power generation sector employed nearly 900,000 people. Relative to
2018, this sector grew by over 2 percent, despite job losses related to nuclear and coal generation
which were offset by increases in employment related to other generating technologies, including
natural gas, solar, and wind. The largest component of total 2019 employment in this sector is
construction (33 percent). Other components of the electric power generation workforce include
utility workers (20 percent), professional and business service employees (20 percent),
manufacturing (13 percent), wholesale trade (8 percent), and other (5 percent). In 2019, jobs
related to solar and wind generation represent 31 percent and 14 percent of total jobs,
respectively, and jobs related to coal generation represent 10 percent of total employment.

In addition to generation-related employment we also look at employment related to coal
and natural gas in the electric power sector. In 2019, the coal industry employed about 75,000
workers. Mining and extraction jobs represent the vast majority of total coal-related employment
in 2019 (74 percent). The natural gas fuel sector employed about 276,000 employees in 2019.
About 60 percent of those jobs were related to mining and extraction.

5.4.3 Projected Sectoral Employment Changes due to the Final Rules

Electric generating units subject to these final rules will use various GHG mitigation
measures to comply. Under the modeling of the final rules, by 2030 19 GW of coal and gas
capacity is estimated to install CCS (while 11 GW of coal and gas capacity are projected to
install CCS under the baseline), 790 MW of coal-fired EGUs are projected to co-fire natural gas,
and 20 GW of coal-fired capacity are projected to undertake coal to gas conversion (7 GW
incremental to the baseline). By 2030, the final rules are projected to result in an additional 5
GW of coal retirements, by 2035 an incremental 21 GW of coal retirements, and by 2040 an
incremental 14 GW of coal retirements relative to the baseline. Under the final rules in 2035, the
modeling projects 2 GW fewer NGCC builds and an incremental 10 GW of NGCT additions
relative to the baseline. 0.9 GW of natural gas capacity is projected to co-fire with hydrogen by
2035. 15 GW of incremental wind and solar additions are also projected to occur relative to the
baseline by 2035.

Based on these power sector modeling projections, we estimate an increase of
approximately 7,900 construction-related job-years related to the installation of new pollution

5-45


-------
controls under the rule in 2030 and another 10,200 construction-related job-years for new
pollution controls in 2035. We estimate an increase of approximately 45,300 job-years in 2028
related to the construction of new capacity in that year, and another even larger increase of
approximately 181,300 construction-related job-years in 2035 as battery storage systems are
constructed. In 2030 and 2040, we estimate decreases of 21,000 construction-related job-years
and 107,500 construction-related job-years, respectively. The relatively large increase and
subsequent decrease results primarily from relatively small temporal changes in the projected
deployment of renewable energy and battery storage capacity in the modeling. The employment
factors related to battery storage are relatively high, and, as a relatively new technology on which
there is limited data to base assumptions, these factors are uncertain. Without including battery
storage in the total estimate, we would estimate increases in 2028, 2030, 2035, and 2045 of
46,100, 7,300, 18,300, and 42,600 job-years, respectively, related to the construction of new
capacity in those years, and a decrease of 11,300 job-years in 2040.

Construction-related job-year changes are one-time impacts, occurring during each year
of the multi-year periods during which construction of new capacity is completed. Construction-
related figures in Table 5-8 represent a point estimate of incremental changes in construction
jobs for each year (e.g., for a three-year construction projection, this table presents one-third of
the total jobs for that project).

Table 5-8 Changes in Labor Utilization: Construction-Related (number of job-years of
employment in a single year)	



2028

2030

2035

2040

2045

New Pollution Controls

<100

7,900

10,200

<100

<100

New Capacity

45,300

-21,000

181,300

-107,500

41,800

Note: These values describe changes under the final rules relative to a projected baseline. A large share of the
construction-related job years is attributable to construction of energy storage, a relatively new technology on which
there is limited data to base labor assumptions.

We also estimate changes in the number of job-years related to recurring non-
construction employment. Recurring employment changes are job-years associated with annual
recurring jobs including operating and maintenance activities and fuel extraction jobs. Newly
built generating capacity creates a recurring stream of positive job-years, while retiring
generating capacity, as well as avoided capacity builds, create a stream of negative job-years.
The rule is projected to result, generally, in a replacement of relatively labor-intensive coal

5-46


-------
capacity with less labor-intensive capacity, which results in an overall decrease of non-
construction jobs between 2028 and 2045. The total net estimated decrease in recurring
employment in any given analysis year is a small percentage of total 2019 power sector
employment reported in the 2020 USEER (approximately 900,000 generation-related jobs,
75,000 coal-related jobs, and 276,000 natural gas-related jobs). Table 5-9 provide detailed
estimates of recurring non-construction employment changes.

Table 5-9 Changes in Labor Utilization: Recurring Non-Construction (number of job-
years of employment in a single year)	



2028

2030

2035

2040

2045

Pollution Controls

<100

-200

-300

<100

-100

Existing Capacity

-2,000

-3,900

-8,700

-5,100

-7,600

New Capacity

3,000

3,400

4,100

3,000

6,300

Fuels (Coal, Natural Gas, Uranium)

-1,200

-400

-800

1,100

-1,100

Coal

-900

-300

-1,800

1,100

-1,200

Natural Gas

-300

<100

1,100

<100

100

Uranium

<100

<100

<100

<100

<100

Note: These values describe changes under the final rules relative to a projected baseline. "<100" denotes an
increase or decrease of less than 100 job-years; Numbers may not sum due to rounding

5.4.4 Conclusions

Generally, there are significant challenges when trying to evaluate the employment effects
due to an environmental regulation from employment effects due to a wide variety of other
economic changes, including the impact of the coronavirus pandemic, on labor markets and the
state of the macroeconomy generally. The analysis of employment impacts in this section
evaluates first-order employment effects at a detailed level for construction and recurring non-
construction labor utilization for pollution control equipment and different capacity and fuel
types.168 For EGUs, these final rules may result in increases and decreases over time of
construction-related jobs related to the installation of new pollution controls and construction of
new capacity. The rule is also projected to result, generally, in a replacement of relatively labor-
intensive coal capacity with less labor-intensive capacity, which results in an overall decrease of
non-construction jobs. It is important to note that this analysis does not estimate the employment

168 In contrast, the economy-wide analysis in Section 5.2, assumes an economy with full employment, and is most
useful for understanding medium to long run shifts in the expected use of labor across aggregate sectors as a
result of the final rules.

5-47


-------
gains likely to result from the expected development and construction of new transmission and
distribution capacity throughout the U.S.

Speaking generally, a variety of federal programs are available to invest in communities
potentially affected by coal mine and coal power plant closures. An initial report by The
Interagency Working Group on Coal and Power Plant Communities and Economic
Revitalization (April 2021) identifies funding available to invest in such "energy communities"
through existing programs from agencies including Department of Energy, Department of
Treasury, Department of Labor and others.169 The Inflation Reduction Act also provides
numerous incentives, including through tax incentives, loans, and grants, to encourage
investment in communities affected by coal mine and coal power plant closures and, more
broadly, communities whose economies are more-reliant on fossil fuels.170

169	See "Initial Report to the President on Empowering Workers Through Revitalizing Energy Communities" April
2021 at https://energycommunities.gOv/wp-content/uploads/2021/l 1/Initial-Report-on-Energy-
Communities_Apr2021 .pdf

170	For more details see Congressional Research Service. "Inflation Reduction Act of 2022 (IRA): Provisions
Related to Climate Change" October 3, 2022 at https://crsreports.congress.gOv/product/pdf/R/R47262

5-48


-------
5.5 References

Berman, E., & Bui, L. T. M. (2001). Environmental regulation and labor demand: evidence from
the South Coast Air Basin. Journal of Public Economics, 79(2), 265-295.
doi: https:// doi. org/10.1016/50047-2727(99^)00101-2

Curtis, E. M. (2018). Who Loses under Cap-and-Trade Programs? The Labor Market Effects of
the NOx Budget Trading Program. The Review of Economics and Statistics, 100(1), 151-
166. doi: 10.1162/REST_a_00680

Curtis, E. M. (2020). Reevaluating the ozone nonattainment standards: Evidence from the 2004
expansion. Journal of Environmental Economics and Management, 99, 102261.
doi: 10.1016/j.jeem.2019.102261

Deschenes, O. (2018). Environmental regulations and labor markets. IZA World of Labor, 22.
doi: 10.1518 5/izawol. 22. v2

Ferris, A. E., Shadbegian, R., & Wolverton, A. (2014). The Effect of Environmental Regulation
on Power Sector Employment: Phase I of the Title IV S02 Trading Program. Journal of
the Association of Environmental and Resource Economists, 7(4), 521-553.
doi: 10.1086/679301

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. doi: 10.1086/342808

Hafstead, M. A. C., & Williams, R. C. (2018). Unemployment and environmental regulation in
general equilibrium. Journal of Public Economics, 160, 50-65.
doi:https://doi.org/10.1016/i.ipubeco.2018.01.013

Marten, A., Schreiber, A., and Wolverton, A. (2023). SAGE Model Documentation (2.1.0).
Washington DC. https://www.epa.gov/environmental-economics/cge-modeling-
regulatory-analvsis

McFarland, J. R., & Herzog, J. H. (2006). Incorporating carbon capture and storage technologies
in integrated assessment models. Energy Economics, 28(5-6), 632-652.

Morgenstern, R. D., Pizer, W. A., & Shih, J.-S. (2002). Jobs Versus the Environment: An
Industry-Level Perspective. Journal of Environmental Economics and Management,
43(3), 412-436. doi:https://doi.org/10.1006/ieem.2001.1191

OMB. (2004). Issuance of OMB's 'Final Information Quality Bulletin for Peer Review.
Washington, DC. https://cfpub.epa.gov/si/m05-03.pdf

Ortiz, D. S., Samaras, C., & Molina-Perez, E. (2013). The Industrial Base for Carbon Dioxide
Storage: Status and Prospects. Retrieved from

https://www.rand.org/content/dam/rand/pubs/technical reports/TR1300/TR1300/RAND
TR1300.pdf

5-49


-------
Rogerson, R. (2015). A Macroeconomic Perspective on Evaluating Environmental Regulations.
Review of Environmental Economics and Policy, 9(2), 219-238. doi:10.1093/reep/rev005

Schreiber, A., Evans, D., Marten, A., Wolverton, A., Davis, W. (2023). Evaluating Economy-
wide Effects of Power Sector Regulations Using the SAGE Model. Retrieved from
https://www.epa.gov/environmental-economics/evaluating-economy-wide-effects-power-
sector-regulations-using-sage-model

U.S. EPA. (2014). Guidelines for Preparing Economic Analyses. (EPA 240-R-10-001).

Washington DC: U.S. Environmental Protection Agency, Office of Policy, National
Center for Environmental Economics, https://www.epa.gov/environmental-
economics/guidelines-preparing-economic-analyses

U.S. EPA. (2015). Economy-Wide Modeling: Social Cost and Welfare White Paper.
https://www.epa.gov/svstem/files/documents/2023-
02/CGE%20social%20cost%20white%20paper%20final.pdf

U.S. EPA Science Advisory Board. (2017). SAB Advice on the Use of Economy-Wide Models in
Evaluating the Social Costs, Benefits, and Economic Impacts of Air Regulations. (EPA-
SAB-17-012). Washington DC

U.S. EPA Science Advisory Board. (2020). Technical Review ofEPA's Computable General
Equilibrium Model, SAGE. (EPA-SAB-20-010). Washington DC

Zivin, J. G., & Neidell, M. (2018). Air pollution's hidden impacts. Science, 359(6371), 39-40.
doi: doi: 10.1126/ science. aap7711

5-50


-------
6 ENVIRONMENTAL JUSTICE IMPACTS

6.1 Introduction

E.O. 12898 directs EPA to "achiev[e] environmental justice (EJ) by identifying and
addressing, as appropriate, disproportionately high and adverse human health or environmental
effects" (59 FR 7629, February 16, 1994), termed disproportionate impacts in this section.
Additionally, E.O. 13985 was signed to advance racial equity and support communities with EJ
concerns through Federal government actions (86 FR 7009, January 20, 2021). Most recently,
E.O. 14096 (88 FR 2521, April 26, 2023) strengthens the directives for achieving environmental
justice that are set out in E.O. 12898. EPA defines EJ as "the just treatment and meaningful
involvement of all people regardless of income, race, color, national origin, Tribal affiliation,
disability, or income with respect to the development, implementation, and enforcement of
environmental laws, regulations, and policies. EPA further defines the term just treatment to
mean that "no group of people should bear a disproportionate burden of environmental harms
and risks, including those resulting from the negative environmental consequences of industrial,
governmental, and commercial operations or programs and policies."171 Meaningful involvement
means that: (1) potentially affected populations have an appropriate opportunity to participate in
decisions about a proposed activity that will affect their environment and/or health; (2) the
public's contribution can influence the regulatory Agency's decision; (3) the concerns of all
participants involved will be considered in the decision-making process; and (4) the rule-writers
and decision-makers seek out and facilitate the involvement of those potentially affected.

The term "disproportionate impacts" refers to differences in impacts or risks that are
extensive enough that they may merit Agency action.172 In general, the determination of whether
a disproportionate impact exists is ultimately a policy judgment which, while informed by
analysis, is the responsibility of the decision-maker. The terms "difference" or "differential"
indicate an analytically discernible distinction in impacts or risks across population groups. It is
the role of the analyst to assess and present differences in anticipated impacts across population

171	See, e.g., "Environmental Justice." EPA.gov, U.S. Environmental Protection Agency, 4 Mar. 2021,
https://www.epa.gov/environmentaljustice

172	See https://www.epa.gov/environmentaljustice/technical-guidance-assessing-environmental-justice-regulatory-
analysis

6-1


-------
groups of concern for both the baseline and regulatory options, using the best available
information (both quantitative and qualitative) to inform the decision-maker and the public.

The Presidential Memorandum on Modernizing Regulatory Review (86 FR 7223;

January 20, 2021) calls for procedures to "take into account the distributional consequences of
regulations, including as part of a quantitative or qualitative analysis of the costs and benefits of
regulations, to ensure that regulatory initiatives appropriately benefit, and do not inappropriately
burden disadvantaged, vulnerable, or marginalized communities." Under E.O. 13563, federal
agencies may consider equity, human dignity, fairness, and distributional considerations, where
appropriate and permitted by law. For purposes of analyzing regulatory impacts, EPA relies upon
its June 2016 "Technical Guidance for Assessing Environmental Justice in Regulatory
Analysis,"173 which provides recommendations that encourage analysts to conduct the highest
quality analysis feasible, recognizing that data limitations, time, resource constraints, and
analytical challenges will vary by media and circumstance. The Technical Guidance states that a
regulatory action may involve potential EJ concerns if it could: (1) create new disproportionate
impacts; (2) exacerbate existing disproportionate impacts; or (3) present opportunities to address
existing disproportionate impacts on communities with EJ concerns through this action under
development.

A reasonable starting point for assessing the need for a more detailed EJ analysis is to
review the available evidence from the published literature and from community input on what
factors may make population groups of concern more vulnerable to adverse effects (e.g.,
underlying risk factors that may contribute to higher exposures and/or impacts). It is also
important to evaluate the data and methods available for conducting an EJ analysis. EJ analyses
can be grouped into two types, both of which are informative, but not always feasible for a given
rulemaking:

1. Baseline: Describes the current (pre-control) distribution of exposures and risk,
identifying potential disparities.

173 See https://www.epa.gov/environmentaljustice/technical-guidance-assessing-environmental-justice-regulatory-
analysis.

6-2


-------
2. Policy: Describes the distribution of exposures and risk after the regulatory option(s)
have been applied (post-control), identifying how potential disparities change in response to the
rulemaking.

EPA's 2016 Technical Guidance does not prescribe or recommend a specific approach or
methodology for conducting EJ analyses, though a key consideration is consistency with the
assumptions underlying other parts of the regulatory analysis when evaluating the baseline and
regulatory options.

6.2 Analyzing EJ Impacts in These Final Rules

In addition to the benefits assessment (Section 4), EPA considers potential EJ concerns of
these final rulemakings. A potential EJ concern is defined as "the actual or potential lack of fair
treatment or meaningful involvement of communities with EJ concerns in the development,
implementation and enforcement of environmental laws, regulations and policies."174 For
analytical purposes, this concept refers more specifically to "disproportionate impacts on
communities with EJ concerns that may exist prior to or that may be created by the regulatory
actions." Although EJ concerns for each rulemaking are unique and should be considered on a
case-by-case basis, EPA's EJ Technical Guidance states that "[t]he analysis of potential EJ
concerns for regulatory actions should address three questions:

1.	Are there potential EJ concerns associated with environmental stressors affected
by the regulatory actions for populations groups of concern in the baseline?

2.	Are there potential EJ concerns associated with environmental stressors affected
by the regulatory actions for population groups of concern for the regulatory
option(s) under consideration?

3.	For the regulatory option(s) under consideration, are potential EJ concerns created
[, exacerbated,] or mitigated compared to the baseline?"

To address these questions, EPA developed an analytical approach that considers the
purpose and specifics of the rulemakings, as well as the nature of known and potential exposures
across various demographic groups. As the final rules are focused on climate impacts resulting

174 See https://www.epa.gov/environmentaljustice/technical-guidance-assessing-environmental-justice-regulatory-
analysis

6-3


-------
from emissions reductions directly targeted in these rulemakings, we begin with a qualitative
discussion (Section 6.3). Insight into any potential near-source pollutant emission changes
associated with existing units is provided by demographic proximity analyses. Proximity
analyses for new units are not feasible as their locations are unknown (Section 6.4). PM2.5 and
ozone concentration reductions due to this action are also quantitatively evaluated relative to the
future baseline with respect to EJ impacts. This analysis characterizes aggregated and
distributional exposures both under a future baseline and following implementation of the final
regulatory options in 2028, 2030, 2035, 2040, and 2045. It is important to note that due to the
relatively low emissions projected under the baseline, and the small magnitude of projected
emissions and ozone and PM2.5 concentration changes, these rules are expected to have a small
impact on the distribution of exposures across each demographic group (Section 6.5). Potential
PM2.5 EJ health impacts (i.e., mortality impacts) and potential impacts of new sources are
discussed qualitatively, based on other recent national quantitative analyses (Sections 6.6, 6.7).

Unique limitations and uncertainties are specific to each type of analysis, which are
described prior to presentation of results in the subsections below.

6.3 GHG Impacts on Environmental Justice and other Populations of Concern

In the 2009 Endangerment Finding, the Administrator considered how climate change
threatens the health and welfare of the U.S. population. As part of that consideration, she also
considered risks to people of color and low-income individuals and communities, finding that
certain parts of the U.S. population may be especially vulnerable based on their characteristics or
circumstances. These groups include economically and socially disadvantaged communities;
individuals at vulnerable life stages, such as the elderly, the very young, and pregnant or nursing
women; those already in poor health or with comorbidities; persons with disabilities; those
experiencing homelessness, mental illness, or substance abuse; and Indigenous or other
populations dependent on one or limited resources for subsistence due to factors including but
not limited to geography, access, and mobility.

Scientific assessment reports produced over the past decade by the U.S. Global Change
Research Program (USGCRP), the IPCC, and the National Academies of Science, Engineering,
and Medicine add more evidence that the impacts of climate change raise potential EJ concerns

6-4


-------
(IPCC, 2018; Oppenheimer et al., 2014; Porter et al., 2014; Smith et al., 2014; USGCRP, 2016,
2018). These reports conclude that less-affluent, traditionally marginalized, predominately non-
White communities can be especially vulnerable to climate change impacts because they tend to
have limited resources for adaptation and are more dependent on climate-sensitive resources
such as local water and food supplies or have less access to social and information resources.
Some communities of color, specifically populations defined jointly by ethnic/racial
characteristics and geographic location (e.g., African-American, Black, and Hispanic/Latino
communities; Native Americans, particularly those living on Tribal lands and Alaska Natives),
may be uniquely vulnerable to climate change health impacts in the U.S., as discussed below. In
particular, the 2016 scientific assessment on The Impacts of Climate Change on Human Health
found with high confidence that vulnerabilities are place- and time-specific, life stages and ages
are linked to immediate and future health impacts, and social determinants of health are linked to
greater extent and severity of climate change-related health impacts (USGCRP, 2016).

Per the Fourth National Climate Assessment (NCA4), "Climate change affects human
health by altering exposures to heat waves, floods, droughts, and other extreme events; vector-,
food- and waterborne infectious diseases; changes in the quality and safety of air, food, and
water; and stresses to mental health and well-being" (Ebi et al., 2018). Many health conditions
such as cardiopulmonary or respiratory illness and other health impacts are associated with and
exacerbated by an increase in GHGs and climate change outcomes, which is problematic as these
diseases occur at higher rates within vulnerable communities. Importantly, negative public health
outcomes include those that are physical in nature, as well as mental, emotional, social, and
economic.

The scientific assessment literature, including the aforementioned reports, demonstrates
that there are myriad ways in which these populations may be affected at the individual and
community levels. Individuals face differential exposure to criteria pollutants, in part due to the
proximities of highways, trains, factories, and other major sources of pollutant-emitting sources
to less-affluent residential areas. Outdoor workers, such as construction or utility crews and
agricultural laborers, who frequently are comprised of already at-risk groups, are exposed to poor
air quality and extreme temperatures without relief. Furthermore, people in communities with EJ
concerns face greater housing, clean water, and food insecurity and bear disproportionate and
adverse economic impacts and health burdens associated with climate change effects. They have

6-5


-------
less or limited access to healthcare and affordable, adequate health or homeowner insurance
(USGCRP, 2016). Finally, resiliency and adaptation are more difficult for economically
vulnerable communities; these communities have less liquidity, individually and collectively, to
move or to make the types of infrastructure or policy changes to limit or reduce the hazards they
face. They frequently are less able to self-advocate for resources that would otherwise aid in
building resilience and hazard reduction and mitigation.

The assessment literature cited in EPA's 2009 and 2016 Endangerment and Cause or
Contribute Findings, as well as The Impacts of Climate Change on Human Health, also
concluded that certain populations and life stages, including children, are most vulnerable to
climate-related health effects (USGCRP, 2016). The assessment literature produced from 2016 to
the present strengthens these conclusions by providing more detailed findings regarding related
vulnerabilities and the projected impacts youth may experience. These assessments - including
the Fourth National Climate Assessment (USGCRP, 2018) and The Impacts of Climate Change
on Human Health in the United States (USGCRP, 2016) - 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
(USGCRP, 2016). In addition, children are among those especially susceptible to allergens, as
well as health effects associated with heatwaves, 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. More
generally, these reports note that extreme weather and flooding can cause or exacerbate poor
health outcomes by affecting mental health because of stress; contributing to or worsening
existing conditions, again due to stress or also as a consequence of exposures to water and air
pollutants; or by impacting hospital and emergency services operations (Ebi et al., 2018).

Further, in urban areas in particular, flooding can have significant economic consequences due to
effects on infrastructure, pollutant exposures, and drowning dangers. The ability to withstand and
recover from flooding is dependent in part on the social vulnerability of the affected population
and individuals experiencing an event (National Academy of Sciences, 2019). In addition,
children are among those especially susceptible to allergens, as well as health effects associated
with heat waves, storms, and floods. Additional health concerns may arise in low-income

6-6


-------
households, especially those with children, if climate change reduces food availability and
increases prices, leading to food insecurity within households.

The Impacts of Climate Change on Human Health also found that some communities of
color, low-income groups, people with limited English proficiency, and certain immigrant groups
(especially those who are undocumented) are subject to many factors that contribute to
vulnerability to the health impacts of climate change (USGCRP, 2016). While difficult to isolate
from related socioeconomic factors, race appears to be an important factor in vulnerability to
climate-related stress, with elevated risks for mortality from high temperatures reported for
Black or African American individuals compared to White individuals after controlling for
factors such as air conditioning use. Moreover, people of color are disproportionately more
exposed to air pollution based on where they live, and disproportionately vulnerable due to
higher baseline prevalence of underlying diseases such as asthma. As explained earlier, climate
change can exacerbate local air pollution conditions so this increase in air pollution is expected
to have disproportionate and adverse effects on these communities. Locations with greater health
threats include urban areas (due to, among other factors, the "heat island" effect where built
infrastructure and lack of green spaces increases local temperatures), areas where airborne
allergens and other air pollutants already occur at higher levels, and communities experienced
depleted water supplies or vulnerable energy and transportation infrastructure.

The 2021 EPA report on climate change and social vulnerability examined four socially
vulnerable groups (individuals who are low income, minority, without high school diplomas,
and/or 65 years and older) and their exposure to several different climate impacts (air quality,
coastal flooding, extreme temperatures, and inland flooding) (U.S. EPA, 2021). This report
found that Black and African-American individuals were 40 percent more likely to currently live
in areas with the highest projected increases in mortality rates due to climate-driven changes in
extreme temperatures, and 34 percent more likely to live in areas with the highest projected
increases in childhood asthma diagnoses due to climate-driven changes in particulate air
pollution. The report found that Hispanic and Latino individuals are 43 percent more likely to
live in areas with the highest projected labor hour losses in weather-exposed industries due to
climate-driven warming, and 50 percent more likely to live in coastal areas with the highest
projected increases in traffic delays due to increases in high-tide flooding. The report found that
American Indian and Alaska Native individuals are 48 percent more likely to live in areas where

6-7


-------
the highest percentage of land is projected to be inundated due to sea level rise, and 37 percent
more likely to live in areas with high projected labor hour losses. Asian individuals were found
to be 23 percent more likely to live in coastal areas with projected increases in traffic delays
from high-tide flooding. Persons with low income or no high school diploma are about 25
percent more likely to live in areas with high projected losses of labor hours, and 15 percent
more likely to live in areas with the highest projected increases in asthma due to climate-driven
increases in particulate air pollution, and in areas with high projected inundation due to sea level
rise.

In a more recent 2023 report, Climate Change Impacts on Children's Health and Weil-
Being in the U.S., EPA considered the degree to which children's health and well-being may be
impacted by five climate-related environmental hazards—extreme heat, poor air quality, changes
in seasonality, flooding, and different types of infectious diseases (U.S. EPA, 2023). The report
found that children's academic achievement is projected to be reduced by 4-7 percent per child,
as a result of moderate and higher levels of warming, impacting future income levels. The report
also projects increases in the number of annual emergency department visits associated with
asthma, and that the number of new asthma diagnoses increases by 4-11 percent due to climate-
driven increases in air pollution relative to current levels. In addition, more than 1 million
children in coastal regions are projected to be temporarily displaced from their homes annually
due to climate-driven flooding, and infectious disease rates are similarly anticipated to rise, with
the number of new Lyme disease cases in children living in 22 states in the eastern and
midwestern U.S. increasing by approximately 3,000-23,000 per year compared to current levels.
Overall, the report confirmed findings of broader climate science assessments that children are
uniquely vulnerable to climate-related impacts and that in many situations, children in the U.S.
who identify as Black, Indigenous, and People of Color, are limited English-speaking, do not
have health insurance, or live in low-income communities may be disproportionately more
exposed to the most severe adverse impacts of climate change.

Indigenous communities face disproportionate and adverse risks from the impacts of
climate change, particularly those communities impacted by degradation of natural and cultural
resources within established reservation boundaries and threats to traditional subsistence
lifestyles. Indigenous communities whose health, economic well-being, and cultural traditions
depend upon the natural environment will likely be affected by the degradation of ecosystem

6-8


-------
goods and services associated with climate change. The IPCC indicates that losses of customs
and historical knowledge may cause communities to be less resilient or adaptable (Porter et al.,
2014). The NCA4 (USGCRP, 2018) noted that while Indigenous peoples are diverse and will be
impacted by the climate changes universal to all Americans, there are several ways in which
climate change uniquely threatens indigenous peoples' livelihoods and economies (Jantarasami
et al., 2018; USGCRP, 2018). In addition, as noted in the following paragraph, there can be
institutional barriers (including policy-based limitations and restrictions) to their management of
water, land, and other natural resources that could impede adaptive measures.

For example, Indigenous agriculture in the Southwest is already being adversely affected
by changing patterns of flooding, drought, dust storms, and rising temperatures leading to
increased soil erosion, irrigation water demand, and decreased crop quality and herd sizes. The
Confederated Tribes of the Umatilla Indian Reservation in the Northwest have identified climate
risks to salmon, elk, deer, roots, and huckleberry habitat. Housing and sanitary water supply
infrastructure are vulnerable to disruption from extreme precipitation events. Native Americans'
ability to respond to these conditions is impeded by limitations imposed by statutes including the
Dawes Act of 1887 and the Indian Reorganization Act of 1934, which ultimately restrict
Indigenous peoples' autonomy regarding land-management decisions through Federal trusteeship
of certain Tribal lands and mandated Federal oversight of these peoples' management decisions.
Additionally, NCA4 noted that Indigenous peoples generally are subjected to institutional racism
effects, such as poor infrastructure, diminished access to quality healthcare, and greater risk of
exposure to pollutants. Consequently, Native Americans often have disproportionately higher
rates of asthma, cardiovascular disease, Alzheimer's disease, diabetes, and obesity. These health
conditions and related effects (disorientation, heightened exposure to PM2.5, etc.) can all
contribute to increased vulnerability to climate-driven extreme heat and air pollution events,
which also may be exacerbated by stressful situations, such as extreme weather events, wildfires,
and other circumstances.

NCA4 and IPCC's Fifth Assessment Report also highlighted several impacts specific to
Alaskan Indigenous Peoples (Porter et al., 2014). Coastal erosion and permafrost thaw will lead
to more coastal erosion, rendering winter travel riskier and exacerbating damage to buildings,
roads, and other infrastructure—impacts on archaeological sites, structures, and objects that will
lead to a loss of cultural heritage for Alaska's indigenous people. In terms of food security, the

6-9


-------
NCA4 discussed reductions in suitable ice conditions for hunting, warmer temperatures
impairing the use of traditional ice cellars for food storage, and declining shellfish populations
due to warming and acidification. While the NCA4 also noted that climate change provided more
opportunity to hunt from boats later in the fall season or earlier in the spring, the assessment
found that the net impact was an overall decrease in food security.

6.4 Demographic Proximity Analyses of Existing Facilities

Demographic proximity analyses allow one to assess communities with EJ concerns
residing near affected facility as a proxy for exposure and the potential for adverse health
impacts that may occur at a local scale due to economic activity at a given location including
noise, odors, traffic, and emissions under these EPA actions.

Although baseline proximity analyses are presented here, several important caveats
should be noted. It should be noted that facilities may vary widely in terms of the impacts they
already pose to nearby populations. In addition, proximity to affected facilities does not capture
variation in baseline exposure across communities, nor does it indicate that any exposures or
impacts will occur and should not be interpreted as a direct measure of exposure or impact.

These points limit the usefulness of proximity analyses when attempting to answer questions
from EPA's EJ Technical Guidance.

Demographic proximity analyses were performed for all plants with at least one coal-
fired unit greater than 25 MW that do not have known retirement plans before 2032 (or has gas
conversion plans) that are affected by these rulemakings. Due to some plants having known
retirement plans, the following subsets of affected facilities were separately evaluated. For each
subset, comparisons of the percentage of various populations (race/ethnicity, age, education,
poverty status, income, and linguistic isolation) living near the facilities were made to average
national levels.

•	All Coal plants subject to the rules (114 facilities, 99 GW).

•	Coal plants subject to the rules that have known retirement plans between 2033 and 2040

(23 facilities, 29 GW).

•	Coal plants subject to the rules without known retirement plans before 2040 (94 facilities,

70 GW).

6-10


-------
The current analysis identified all census blocks with centroids within a 5 km, 10 km, and
50 km radius of the latitude/longitude location of each facility, and then linked each block with
census-based demographic data. The total population within a specific radius around each
facility is the sum of the population for every census block within that specified radius, based on
each block's population provided by the 2020 decennial Census.175 Statistics on race, ethnicity,
age, education level, poverty status and linguistic isolation were obtained from the Census'
American Community Survey (ACS) 5-year averages for 2016 to 2020. These data are provided
at the block group level. For the purposes of this analysis, the demographic characteristics of a
given block group - that is, the percentage of people in different races/ethnicities, the percentage
without a high school diploma, the percentage that are below the poverty level, the percentage
that are below two times the poverty level, and the percentage that are linguistically isolated -
are presumed to also describe each census block located within that block group.

In addition to facility-specific demographics, the demographic composition of the total
population within the specified radius (e.g., 5 km, 10 km, or 50 km) for all facilities was also
computed (e.g., all EGUs subject to the 111 rules). In calculating the total populations, to avoid
double-counting, each census block population was only counted once. That is, if a census block
was located within the selected radius (i.e., 5 km, 10 km, or 50 km) for multiple facilities, the
population of that census block was only counted once in the total population. Finally, this
analysis compares the demographics at each specified radius (i.e., 5 km, 10 km, or 50 km) to the
demographic composition of the nationwide population. The methodology and the results of the
demographic analyses for the final rules are presented in the technical report, Analysis of
Demographic Factors for Populations Living Near Coal-Fired Electric Generating Units
(EGUs) for the Section 111 NSPS and Emissions Guidelines - Final, available in the docket for
these actions. The docket also contains the detailed demographic spreadsheets with facility-
specific demographic data.

Table 6-1 through Table 6-3 show the results of the proximity analysis for the three sets
of affected facilities investigated at the 5 km radius, 10 km radius and the 50 km radius,
respectively. Approximately 564,000 people live within 5 km of the 114 coal plants, 2.6 million

175 The location of the Census block centroid is used to determine if the entire population of the Census block is
assumed to be within the specific radius. It is unknown how sensitive these results may be to different methods
of population estimation, such as aerial apportionment.

6-11


-------
people live within 10 km, and 40 million live within 50 km. It should be noted that at the 5 km
radius, two facilities have zero population living within 5 km and another 10 facilities have less
than 100 people living within 5 km. For facilities where the population is zero, there are no
demographics data, and for those where the population is very low, the uncertainty in
demographics data may be high. Therefore, in addition to the 5 km radius, we conducted
proximity analyses at 10 km and 50 km, which provide more robust population data. At the 10
km radius, there were no facilities with zero population data and only two facilities had
populations of less than 100 people living within 10 km. For the 50 km radius, at least 10,000
people were living within 50 km of each facility.

The analysis indicates that, on average for all 114 facilities subject to the final rules, the
percent of the population that is American Indian within 5 km and 10 km of the plants (1 percent
and 0.8 percent, respectively) is above the national average (0.6 percent). This is largely driven
by seven facilities that have a percent American Indian population living within 5 km and 10 km
ranging from 10 percent to just over 40 percent. The percent of the population within 5 km and
10 km that is living below poverty (14 percent for both) and below 2 times the poverty level (34
percent and 33 percent, respectively) is above the corresponding national averages (13 percent
and 29 percent). The percentage of the population living within 50 km of the facilities that is
Black (13 percent) is above the national average (12 percent). The age distributions of the
populations living within 5 km, 10 km, and 50 km are similar to the national average
distribution.

For the 23 facilities with known retirement plans from 2033 to 2040, the percentages of
the population living within 5 km and 10 km of these units that are living below the federal
poverty level (14 percent for both) and below 2 times the federal poverty level (33 percent and
31 percent, respectively) are above their corresponding national averages (13 percent and 29
percent). When we look at the population living within 50 km of these 23 facilities, we see that a
larger percentage of the population is Black (14 percent), which is above the national average (12
percent). The age distributions of the populations living within 5 km, 10 km, and 50 km are
similar to the national average distribution.

6-12


-------
Since the population living around the 94 facilities for which EPA is unaware of plans to
retire before 2040 accounts for about 85 percent of the population living around all 114 units
subject to the final rules, the demographics are nearly identical.

It is important to note that any incremental deployment of CCS under these rules could
occur within this group of 94 facilities. Several environmental justice organizations and
community representatives raised significant concerns about the potential health, environmental,
and safety impacts of CCS. As discussed in section VII.C of the preamble, the EPA recognizes
that use of this technology can, under some circumstances, result in the increase in emission of
certain co-pollutants at a coal-fired steam generating unit. While there are protections in place
that can mitigate these impacts, it is important to consider the population living nearby any
facility where there may be potential for these impacts to occur. Given the uncertainty regarding
where installations may occur and the extent to which local emissions might be affected, the
EPA is providing detailed information for each of the 94 facilities discussed above, at which
installation of CCS is possible. This information is being provided in the document titled:
Analysis of Demographic Factors for Populations Living Near Coal-Fired Electric Generating
Units (EGUs) for the Section 111 NSPS and Emissions Guidelines and Potential Emissions
Changes which is available in the docket. This document presents information on the populations
living within 5 km and 10 km of facility, as well as the potential emissions implications of
installing CCS absent the implementation of any protections discussed in section VII.C of the
preamble. While the EPA projects that only a subset of this capacity is likely to install this
technology, this information is being provided for all units out of abundance of caution, and to
assist all states and stakeholders in considering options for state plans.

6-13


-------
Table 6-1 Proximity Demographic Assessment Results Within 5 km of Coal-Fired Units
Greater than 25 MW Affected by these Final Rules a,b,c

	Population within 5 km	

Demographic
Group

Nationwide
Average for
Comparison

All Coal Plants
subject to the rules

Coal Plants with
known retirement
plans from 2033 to
2040 subject to the
rules

Coal Plants
without known
retirement plans
before 2040
subject to the rules

Total Population

329,824,950 564,492

87,230

494.112

Number of
Facilities

-

114°

23c

94c

Race and Ethnicity by Percent

White

60%

78%

90%

75%

Black

12%

8%

3%

9%

American Indian

0.6%

1%

0.2%

1.2%

Hispanic or Latino

19%

9%

4%

9%

Other and
Multiracial

9%

5%

4%

5%

Age By Percent

Age 0 to 17 years

22%

22%

22%

22%

Age 18 to 64 years

62%

60%

60%

60%

Age > 65 years

16%

18%

18%

18%

Income by Percent

Below Poverty
Level

13%

14%

14%

14%

Below 2x Poverty
Level

29%

34%

33%

34%

Education by Percent

>25 and w/o a HS
diploma

12%

12%

12%

12%

Linguistically Isolated by Percent

Linguistically
Isolated

5%

2%

1%

2%

3 The nationwide population count and all demographic percentages are based on the Census' 2016-2020 American
Community Survey five-year block group averages and include Puerto Rico. Demographic percentages based on
different averages may differ. The total population counts are based on the 2020 Decennial Census block
populations.

b To avoid double counting, the "Hispanic or Latino" category is treated as a distinct demographic category for these
analyses. A person is identified as one of five racial/ethnic categories above: White, Black, American Indian, Other
and Multiracial, or Hispanic/Latino. A person who identifies as Hispanic or Latino is counted as Hispanic/Latino for
this analysis, regardless of what race this person may have also identified as in the Census. Includes white and
nonwhite.

c For all coal plants subject to the rule, two facilities have zero population within 5 km and another 10 facilities have
less than 100 people living within 5 km. In the group of plants with known retirement plans from 2033 to 2040, one
facility had zero population within 5 km. In the group of plants without known retirement plans before 2040, one
facility had zero population within 5 km, and 10 facilities had less than 100 people living within 5 km. For facilities
where the population is zero, there is no demographics data and for those where the population is low, the
uncertainty in the demographics data may be high.

6-14


-------
Table 6-2 Proximity Demographic Assessment Results Within 10 km of Coal-Fired
Units Greater than 25 MW Affected by these Final Rules a,b

	Population within 10 km	

Demographic
Group

Nationwide
Average for
Comparison

All Coal Plants
subject to the rules

Coal Plants with
known retirement
plans from 2033 to
2040 subject to the
rules

Coal Plants
without known
retirement plans
before 2040
subject to the rules

Total Population

329,824,950 2,574,398

414,646

2.289,025

Number of



114

23

94

Facilities



Race and Ethnicity by Percent

White

60%

72%

87%

70%

Black

12%

10%

4%

11%

American Indian

0.6%

0.8%

0.4%

0.8%

Hispanic or Latino

19%

12%

4%

12%

Other and
Multiracial

9%

6%

5%

6%

Age By Percent

Age 0 to 17 years

22%

23%

21%

23%

Age 18 to 64 years

62%

60%

63%

61%

Age > 65 years

16%

17%

17%

17%

Income by Percent

Below Poverty
Level

13%

14%

14%

14%

Below 2x Poverty
Level

29%

33%

31%

33%

Education by Percent

>25 and w/o a HS
diploma

12%

11%

9%

11%

Linguistically Isolated by Percent

Linguistically
Isolated

5%

3%

1%

3%

3 The nationwide population count and all demographic percentages are based on the Census' 2016-2020 American
Community Survey five-year block group averages and include Puerto Rico. Demographic percentages based on
different averages may differ. The total population counts are based on the 2020 Decennial Census block
populations.

b To avoid double counting, the "Hispanic or Latino" category is treated as a distinct demographic category for these
analyses. A person is identified as one of five racial/ethnic categories above: White, Black, American Indian, Other
and Multiracial, or Hispanic/Latino. A person who identifies as Hispanic or Latino is counted as Hispanic/Latino for
this analysis, regardless of what race this person may have also identified as in the Census. Includes white and
nonwhite.

6-15


-------
Table 6-3 Proximity Demographic Assessment Results Within 50 km of Coal-Fired
Units Greater than 25 MW Affected by these Final Rules a,b	

Population within 50 km







Coal Plants with

Coal Plants

Demographic
Group

Nationwide
Average for
Comparison

All Coal Plants
subject to the rules

known retirement
plans from 2033 to
2040 subject to the
rules

without known
retirement plans
before 2040
subject to the rules

Total Population

329,824,950

40,143,893

12.196,836

34.070,688

Number of



114

23

94

Facilities



Race and Ethnicity by Percent

White

60%

69%

73%

69%

Black

12%

13%

14%

12%

American Indian

0.6%

0.5%

0.4%

0.5%

Hispanic or Latino

19%

11%

7%

12%

Other and
Multiracial

9%

6%

6%

6%

Age By Percent

Age 0 to 17 years

22%

22%

22%

22%

Age 18 to 64 years

62%

61%

61%

61%

Age > 65 years

16%

17%

17%

17%

Income by Percent

Below Poverty
Level

13%

12%

12%

12%

Below 2x Poverty
Level

29%

29%

28%

29%

Education by Percent

>25 and w/o a HS
diploma

12%

10%

10%

10%

Linguistically Isolated by Percent

Linguistically
Isolated

5%

3%

2%

3%

3 The nationwide population count and all demographic percentages are based on the Census' 2016-2020 American
Community Survey five-year block group averages and include Puerto Rico. Demographic percentages based on
different averages may differ. The total population counts are based on the 2020 Decennial Census block
populations.

b To avoid double counting, the "Hispanic or Latino" category is treated as a distinct demographic category for these
analyses. A person is identified as one of five racial/ethnic categories above: White, Black, American Indian, Other
and Multiracial, or Hispanic/Latino. A person who identifies as Hispanic or Latino is counted as Hispanic/Latino for
this analysis, regardless of what race this person may have also identified as in the Census. Includes white and
nonwhite.

6-16


-------
6.5 EJ PM2.5 and Ozone Exposure Impacts

This EJ air pollutant exposure176 analysis aims to evaluate the potential for EJ concerns
related to PM2.5 and ozone exposures177 among communities with EJ concerns. To assess EJ
ozone and PM2.5 exposure impacts, we focus on the first and third of the three EJ questions from
EPA's 2016 EJ Technical Guidance,178 which ask if there are potential EJ concerns associated
with stressors affected by the regulatory actions for population groups of concern in the baseline
and if those potential EJ concerns in the baseline are exacerbated, unchanged, or mitigated under
the regulatory options.179 To address these questions with respect to the PM2.5 and ozone
exposures, EPA developed an analytical approach that considers the purpose and specifics of
these final rules, as well as the nature of known and potential exposures and impacts.

Specifically, as 1) these rules affect EGUs across the U.S., which typically have tall stacks that
result in emissions from these sources being dispersed over large distances, and 2) both ozone
and PM2.5 can undergo long-range transport, it is appropriate to conduct an EJ assessment of the
contiguous U.S. Given the availability of modeled PM2.5 and ozone air quality surfaces under the
baseline and regulatory options, we conduct an analysis of changes in PM2.5 and ozone
concentrations resulting from the emission changes projected by IPM180 to occur under these

176	The term exposure is used here to describe estimated PM2.5 and ozone concentrations and not individual dosage.

177	Air quality surfaces used to estimate exposures are based on 12 km grids. Additional information on air quality
modeling can be found in the air quality modeling information section.

178	U.S. Environmental Protection Agency (EPA), 2015. Guidance on Considering Environmental Justice During the
Development of Regulatory Actions, https://www.epa.gov/sites/default/files/2015-06/documents/considering-ej-
in-rulemaking-guide-final.pdf

179	EJ question 2, which asks if there are potential EJ concerns (i.e., disproportionate burdens across population
groups) associated with environmental stressors affected by the regulatory action for population groups of
concern for the regulatory options under consideration, was not focused on for several reasons. Importantly, the
total magnitude of differential exposure burdens with respect to ozone and PM2 5 among population groups at the
national scale has been fairly consistent pre- and post-policy implementation across recent rulemakings. As such,
differences in nationally aggregated exposure burden averages between population groups before and after the
rulemaking tend to be very similar. Therefore, as disparities in pre- and post-policy burden results appear
virtually indistinguishable, the difference attributable to the rulemaking can be more easily observed when
viewing the change in exposure impacts, and as we had limited available time and resources, we chose to provide
quantitative results on the pre-policy baseline and policy-specific impacts only, which related to EJ questions 1
and 3. We do however use the results from questions 1 and 3 to gain insight into the answer to EJ question 2 in
the summary (Section 6.8).

180	As discussed in greater detail in Section 3, IPM is a comprehensive electricity market optimization model that
can evaluate the impacts of regulatory actions affecting the power sector within the context of regional and
national electricity markets. IPM generates least-cost resource dispatch decisions based on user-specified
constraints such as environmental, demand, and other operational constraints. IPM uses a long-term dynamic
linear programming framework that simulates the dispatch of generating capacity to achieve a demand-supply
equilibrium on a seasonal basis and by region. The model computes optimal capacity that combines short-term

6-17


-------
rules as compared to the baseline scenario, characterizing average and distributional exposures
following implementation of the regulatory options in 2028, 2030, 2035, 2040, and 2045.
However, several important caveats of this analysis are as follows:

The GHG mitigation measures in this RIA are illustrative since States are
afforded flexibility to implement the final rules, and thus the impacts could be
different to the extent states make different choices than those assumed in the
illustrative analysis. Additionally, the way that EGUs comply with the GHG
mitigation measures may differ from the methods forecast in the modeling for this
RIA.

Although several future years were assessed for health benefits associated with
these final rulemakings, there was high year-to-year PM2.5 and ozone
concentration change variability across modeled future years.

The baseline scenarios for 2028, 2030, 2035, 2040, and 2045 represent EGU
emissions expected in 2028, 2030, 2035, 2040 and 2045 respectively, but
emissions from all other sources are projected to the year 2026. The 2028, 2030,
2035, 2040, and 2045 baselines therefore do not capture any anticipated changes
in ambient ozone and PM2.5 between 2026 and 2028, 2030, 2035, 2040, or 2045
that would occur due to emissions changes from sources other than EGUs.

Modeling of post-policy air quality concentration changes are based on state-level
emission data paired with facility-level baseline 2026 emissions that were
available in the summer 2021 version of IPM. While the baseline spatial patterns
represent 12 km grid resolution ozone and PM2.5 concentrations associated with
the facility level emissions described above, the post-policy air quality surfaces
will capture expected ozone and PM2.5 changes that result from state-to-state
emissions changes by fuel type but will not capture heterogenous changes in
emissions from multiple facilities of the same fuel-type within a single state (i.e.
all coal EGU sources within each state are assumed to increase or decrease in

dispatch decisions with long-term investment decisions. IPM runs under the assumption that electricity demand
must be met and maintains a consistent expectation of future load. IPM outputs include the air emissions
resulting from the simulated generation mix

6-18


-------
unison and all natural gas EGU sources within each state are assumed to increase
or decrease in unison for the purpose of creating air quality surfaces).

Air quality simulation input information is at a 12 km grid resolution, and
population information is either at the Census tract- or county-level, potentially
masking impacts at geographic scales more highly resolved than the input
information.

The two specific air pollutant metrics evaluated in this assessment, warm season
maximum daily eight-hour ozone average concentrations and average annual
PM2.5 concentrations, are focused on longer-term exposures that have been linked
to adverse health effects. This assessment does not evaluate disparities in other
potentially health-relevant metrics, such as shorter-term exposures to ozone and
PM25.

PM2.5 EJ impacts were limited to exposures, and do not extend to health effects,
given additional uncertainties associated with estimating health effects stratified
by demographic population and the ability to predict differential PM2.5-
attributable EJ health impacts.

6-19


-------
Population variables considered in this EJ exposure assessment include race, ethnicity,
educational attainment, employment status, health insurance status, life expectancy, linguistic
isolation, poverty status, redlined areas, Tribal land, age, and sex (Table 6-4).181'182'183'184

Note that these variables are different than those used in the proximity analysis because
criteria pollutants have nationwide impacts rather than the localized impacts that are investigated
in a proximity analysis. Variables such as life expectancy, redlining, and health insurance status
that are not included in the localized demographic proximity analysis are included in the
nationwide criteria pollutant exposure analysis as a way to account for various vulnerabilities
and susceptibilities that already exist in the nationwide population due to cumulative risks from
multiple exposures (such as other pollutant exposures, stress, lack of access to healthcare, etc).
There are also fewer demographic uncertainties at a national scale which allows us to use an
expanded set of variables for a nationwide analysis.

181	Population projections stratified by race/ethnicity, age, and sex are based on economic forecasting models
developed by Woods and Poole (Woods & Poole, 2015). The Woods and Poole database contains county-level
projections of population by age, sex, and race out to 2050, relative to a baseline using the 2010 Census data.
Population projections for each county are determined simultaneously with every other county in the U.S. to
consider patterns of economic growth and migration. County-level estimates of population percentages within
the poverty status and educational attainment groups were derived from 2015-2019 5-year average ACS
estimates. Additional information can be found in Appendix J of the BenMAP-CE User's Manual

(https ://www. epa. gov/benmap/benmap-ce-manual-and-appendices).

182	EPA acknowledges the recent comments about cumulative risk assessment and is currently in the process of
developing cumulative risk assessment methods for our quantitative environmental justice analyses. In the
interim, the 111 EGU rulemakings utilize the "life expectancy" and "redlining" variables as a proxy to identify
communities with higher or lower exposure to cumulative risks. The choice of comparing the top 75% vs the
bottom 25% of life expectancy was made by finding the natural cut-off point in the distribution of life
expectancies across the contiguous U.S. EPA continues to improve its methodology based on its framework for a
Cumulative Risk Assessment as well as guidance from multiple Executive Orders and intend to more accurately
assess cumulative risk in future rulemakings.

183	The Tribal Land variable was also added in response to recent Executive Orders that have emphasized the need
for more detailed analysis on the impacts on American Indians. The Tribal Lands variable focuses specifically on
populations who live on Tribal lands in addition to quantifying those whose race is Indian American but may or
may not live on Tribal lands.

184	An additional population variable that is not included in this analysis is persons with disability. Persons with
disability is a new environmental justice metric listed in E.O. 14096 (88 FR 25251, April 26, 2023), and EPA is
currently developing analytical techniques/tools to evaluate its impact on our environmental analyses.

6-20


-------
Table 6-4 Demographic Populations Included in the Ozone and PM2.5 EJ Exposure
Analysis	

Demographic

Groups

Ages

Spatial Scale of
Population Data

Race

Asian; American Indian; Black; White

0-99

Census tract

Ethnicity

Hispanic; Non-Hispanic

0-99

Census tract

Educational Attainment

High school degree or more; No high school degree

25-99

Census tract

Employment Status

Employed; Unemployed; Not in the labor force

0-99

County

Health Insurance

Insured; Uninsured

0-64

County

Life Expectancy

Top 75%; Bottom 25%

0-99

Census tract

Linguistic Isolation

Speaks English "well or better"; Speaks English <
"well"

0-99

Census tract

Poverty Status

Above the poverty line
Below the poverty line

0-99

Census tract

Redlined Areas

HOLC Grades A-C; HOLC Grade D; Not graded by
HOLC

0-99

Census tract

Tribal Land

Tribal land; Not Tribal land

0-99

Census tract

Age

Children
Adults
Older Adults

0-17

18-64

65-99

Census tract

Sex

Female; Male

0-99

Census tract

6.5.1 Populations Predicted to Experience PM2.5 and Ozone Air Quality Changes

The EPA analyzed several illustrative compliance scenarios185 representing potential
compliance outcomes and projects that, relative to a projected future baseline, these actions
achieve nationwide reductions in EGU emissions of multiple health-harming air pollutants
including NOx, SO2, and PM2.5, resulting in significant public health benefits. In all years, the
final rules are expected to result in modest but widespread reductions in ambient levels of PM2.5
and ozone concentrations over many areas of the US, although some areas may experience
increases in ozone concentrations relative to forecasted future baselines without the rule.

Relative to 2028 baseline conditions, our analysis indicates that ozone and PM2.5 will decline in
virtually areas of the country. However, some areas of the country may experience slower or
faster rates of decline in ozone and PM2.5 pollution over time as a result of the changes in
generation and utilization resulting from the rule.

To best assess these modeled increases and decreases in emissions and evaluate their
impact on communities with EJ concerns, the contiguous U.S. was first grouped into areas where

185 The GHG mitigation measures in this RIA are illustrative since States are afforded flexibility to implement the
final rules, and thus the impacts could be different to the extent states make different choices than those assumed
in the illustrative analysis. Additionally, the way that EGUs comply with the GHG mitigation measures may
differ from the methods forecast in the modeling for this RIA.

6-21


-------
air quality 1) does not change or improves, or 2) worsens as a result of the final rulemakings
relative to a projected year's future baseline. Note that national emissions reduction estimates
vary by year, with 2035 being the snapshot future year in which emission reductions are
projected to be largest (Table 3-5). In the contiguous U.S., it is estimated that at least 75 percent
of the U.S. population is predicted to experience air quality improvements (or a lack of change)
for PM2.5 under all scenarios analyzed except for the 2040 regulatory options, in which
approximately 26-58 percent of the U.S. population is predicted to experience a PM2.5 air quality
improvement (Figure 6-1). Similarly, it is estimated that at least 65 percent of the U.S.
population is predicted to experience ozone improvements (or a lack of change) due to the
rulemakings for ozone under all scenarios analyzed with the exception of the 2040 regulatory
options. In absolute terms, this equates to up to 77 million people experiencing worsening PM2.5
concentrations (or up to 292 million in the 2040 final rules regulatory option) and up to 126
million people experiencing worsening ozone concentrations (or up to 313 million in the 2040
alternative 2 regulatory option). The average magnitudes of worsening PM2.5 concentration
changes due to the rulemakings round to 0.00 |ig/m3 and are much smaller than the average
magnitudes of improving PM2.5 concentration changes (which round to 0.01 - 0.04 |ig/m3). The
average magnitudes of worsening ozone concentration changes (which round to 0.00 - 0.03 ppb)
are also smaller than that of improving ozone concentration changes (which round to 0.02 - 0.09
ppb).

6-22


-------
Year

Scenario

PM2.5

Pollutant

Ozone

2028

Final Rules

63 M • £ 296M

126M O

Q232M



Alternative 1

58M O £ 301M

115M O

0 244M



Alternative 2

5SM O £ 301M

121M O

Q 238M

2030

Final Rules

69M O0296M

111MO

02S4M



Alternative 1

77M O0288M

SIM O

^ 284M



Alternative 2

74M 2;:v

12M

^ 353M

2035

Final Rules

S3M • ^ 327M

84M ®

297M ^



Alternative i

52M O 0329M

78M O

303M ^



Alternative 2

52 M O ^ 328M

91M ©

289M ^

2040

Final Rules

292M Q9103 M

310 M

®85M



Alternative 1

237M Q^1S8M

310M £

®S4M



Alternative 2

165 M O® 230M

313 M £

• 82M

204S

Final Rules

30 M 0 379 M

2M

407M



Alternative 1

7M • £ 401M

2M

407M



Alternative 2

6M • ^^403M

2M

^^407M



-0.05 0.00 0.05
Average Reduction

0.10 -0.05 O.OO 0.05 0.10
Average Reduction

Figure 6-1 Number of People Residing in the Contiguous U.S. Areas Improving or Not
Changing (Blue) or Worsening (Orange) in 2028, 2030, 2035, 2040, and 2045 for PM2.5 and
Ozone and the National Average Magnitude of Pollutant Concentration Reductions (jig/m3
and ppb) for the 3 Regulatory Options

6.5.2 PM2 5 EJ Exposure A nalysis

We evaluated the potential for EJ concerns among communities with EJ concerns
resulting from exposure to PM2.5 under the baseline and regulatory options in these rules. This
was done by characterizing the average and distribution of PM2.5 exposures both prior to and
following implementation of the three regulatory options (the final rules option, as well as the
alternative regulatory options), in 2028, 2030, 2035, 2040, and 2045. As this analysis is based on
the same PM2.5 spatial fields as the benefits assessment (see Section 3 for a discussion of the
spatial fields), it is subject to similar types of uncertainty (see Sections 3.8 and 4.3.8 for
discussions of uncertainty). A particularly germane limitation for this analysis is that the
expected concentration changes are quite small, likely making uncertainties associated with the
various input data more relevant.

6-23


-------
6.5.2.1 National Aggregated Results

National average baseline PM2.5 concentrations in micrograms per cubic meter (|ig/m3) in
2028, 2030, 2035, 2040, and 2045 are shown in the Figure 6-2 heat map. Concentrations
represent the total estimated PM2.5 exposure burden averaged over the 12-month calendar year
and are colored to visualize differences more easily in average concentrations (lighter blue
coloring representing smaller average concentrations and darker blue coloring representing larger
average concentrations). When looking across all five future years, all demographic groups will
see improving PM2.5 concentrations over time in the baseline such that the average concentration
experienced by each demographic group will be less in 2030 than in 2028, less in 2035 than in
2030, and so on. When looking within each future year's baseline, average national disparities
observed in the baseline of these rules are similar to those described by recent rules (e.g., the
Reconsideration of the National Ambient Air Quality Standards for Particulate Matter186), that is,
populations with national average PM2.5 concentrations higher than the reference population
ordered from most to least difference are: those who are residents of HOLC Grade D (i.e.,
redlined) census tracts, linguistically isolated, residents of HOLC Grade A-C (i.e., not redlined)
census tracts, Hispanic populations, Asian populations, those without a high school diploma, and
Black populations (Figure 6-2).

In Figure 6-3, columns labeled "Final Rules" "Alternative 1," and "Alternative 2"
provide information regarding how all three regulatory options will impact PM2.5 concentrations
across various populations, respectively.187 While the national-level PM2.5 concentration
reductions were similar for all population groups evaluated in 2028, 2030, and 2045, the
reductions were higher in 2035 and lower in 2040. Differences in reductions were also more
notable in 2035. For example, (Figure 6-3), for all scenarios, the linguistically isolated, Asian
population, and Hispanic population, which have higher average baseline exposures, are
estimated to experience slightly smaller PM2.5 concentration reductions than the overall reference
population.

186	https://www.federalregister.gov/documents/2023/01/27/2023-00269/reconsideration-of-the-national-ambient-air-
quality-standards-for-particulate-matter

187	We report average exposure results to the decimal place where difference between demographic populations
become visible, as we cannot provide a quantitative estimate of the air quality modeling precision uncertainty.
Using this approach allows for a qualitative consideration of uncertainties and the significance of the relative
magnitude of differences

6-24


-------
The national-level assessment of PM2.5 before and after implementation of these final
rulemakings suggests that while EJ exposure disparities are present in the pre-final rules
scenario, EJ exposure concerns are not likely created or exacerbated by the rules for the
population groups evaluated, due to the small difference in magnitudes of PVI2.5 concentration
reductions across demographic groups. It is also important to note that, at the national-level the
PM2.5 concentrations before and after implementation for all five future years evaluated, the
concentrations for each demographic group are below the recently revised national ambient air
quality standard of 9 ug/m3. ;sx









Year





Population

Qualifier

2028

2030

2035

2040

2045

Reference

Reference (0-99)

7.2

7.1

7.1

7.1

7.0

Race

White (0-99)

7.1

7.0

7.0

7.0

6.9



American Indian (0-99)

6.7

6.7

6.6

6.6

6.6



Asian (0-99)

7.7

7.7

7.6

7.6

7.5



Black (0-99)

7.4

7.4

7.3

7.2

7.2

Ethnicity

Non-Hispanic (0-99)

6.9

6.9

6.8

6.8

6.8



Hispanic (0-99)

7.9

7.9

7.8

7.8

7.7i

Educational Attainment More educated (>24: HS or more)

7.1

7.0

7.0

7.0

6.9



Less educated (>24; no HS)

7.5

7.5

7.4

7.4

7.4

Employment Status

Employed (0-99)

7.1

7.1

7.1

7.0

7.0



Unemployed (0-99)

7.3

7.3

7.2

7.2

7.2



Not in the labor force (0-99)

7.2

7.1

7.1

7.1

7.0

Insurance Status

Insured (0-64)

7.2

7.2

7.1

7.1

7.1



Uninsured (0-64)

7.3

7.2

7.2

7.2

7.1

Life Expectancy

Top 75% life expectancy (0-99)

7.1

7.1

7.1

7.1

7.0



Bottom 25% life expectancy (0-99)

7.2

7.1

7.1

7.1

7.0



No life expectancy data (0-99)

7.1

7.1

7.0

7.0

7.0

Linguistic Isolation

English well or better (0-99)

7.1

7.1

7.0

7.0

7.0



English < well (0-99)

8.1

8.1

8.0

8.0

8.0

Poverty Status

>200%ofthe poverty line(0-99)

7.1

7.0

7.0

7.0

7.0



<200%ofthe poverty line(0-99)

7.3

7.3

7.2

7.2

7.2



>Povertyline (0-99)

7.1

7.1

7.0

7.0

7.0




-------
Year / Scenario

2028	I	2030	2035	2040	2045





ju





10

JU

0)

>

1

ju

Q>

>



ju



0

JU



(U

>





3
CL

(B

In

Z3
DC

In

ID

3

a.

ID

ID

3

ct

ID

Id

3

cz

ID

'±»
ID





~S

E

E

m

E

E

IB

E

E



E

E

]§

E

E





tZ

-M

£

tE





iL



*->

ill



±!

ul











<

<



<

<



<1

<



<

<



<

<

Population

Qualifier































Reference

Reference (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01

Race

White (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0 03

0.03

0.00

0.00

0.00

0.01

0.01

0.01



American Indian (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.02

0-02

0.02

0.00

0.00

0.00

0.01

0.01

0.01



Asian (0-99)

0.01

0.01

0.01

0.01

0.00

0.00

0.02

002

0.02

0.00

0.00

0.00

0.01

0.01

0.01



Black (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0 03

0.00

0.00

0.00

0.01

0.01

0.01

Ethnicity

Non-Hispanic (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

003

0.00

0.00

0.00

0.01

0.01

0.01



Hispanic (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.02

0.02

0.02

0.00

0.00

0.00

0.01

0.01

0.01

Educational

More educated (>24: HS or more)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0-03

0.03

0.00

0.00

0.00

0.01

0.01

0.01

Attainment

Less educated (>24; no HS)

0.01

0.01

0.01

0.01

0.01

0.01

0.02

0.03

0.02

0.00

0.00

0.00

0.01

0.01

0.01

Employment

Employed (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01

Status

Unemployed (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01



Not in the labor force (0-99)

0-01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01

Insurance

Insured (0-64)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01

Status

Uninsured (0-64)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01

Life Expectancy Top 75% life expectancy (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01



Bottom 25% life expectancy (0-99)

0.02

0.01

0.01

0.01

0.01

0.01

0.03

0 03

003

0.00

0.00

0.00

0.01

0.01

0.01



No life expectancy data (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0 03

0.03

0.00

0.00

0.00

0.01

0.01

0.01

Linguistic

English well or better (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01

Isolation

English < well (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.02

0.02

0.02

0.00

0.00

0.00

0.01

0.01

0.01

Poverty Status

>200%oxme poverty line (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01



<200% of the poverty line (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01



>Poverty line (0-99)

0.01

0.01

0.01

0.01

0.01

0.01

0.03

0.03

0.03

0.00

0.00

0.00

0.01

0.01

0.01




-------
map, only colors are used to show relative PM2.5 concentrations and only the overall reference
group (i.e., everyone ages 0-99) is included. The magnitude of state-level PM2.5 concentration
changes are very similar across all three scenarios. However, due to EGU-specific estimated
emission changes, the magnitude of state-level PM2.5 concentration changes varies considerably
across states. Depending on the regulatory scenario and year of analysis, average population-
weighted state-level PM2.5 concentrations are predicted to be reduced by up to 0.09 |ig/m3 (as
seen in Nebraska in 2035) which is 1.3 percent of the baseline PM2.5 concentration in 2035.
Increases in PM2.5 concentrations for state-level average populations were rare and largest in
2040 under all regulatory options in Mississippi, and only to a very small magnitude (0.02
|ig/m3) which is 0.3 percent of the baseline PM2.5 concentration in 2040. When considering
differences between demographic populations and the reference population affected by a
particular policy within a given year, average PM2.5 concentration changes at the state-level only
differ from the reference population by up to 0.02 |ig/m3 which is 0.3 percent of the baseline
PM2.5 concentration in 2040.190 While the percent changes relative to the baseline are notable,
the magnitude of the changes is too small to have a discernible impact on public health
outcomes. Therefore, whereas PM2.5 exposure impacts vary by state, the small magnitude of
differential impacts expected from the final rules is not likely to exacerbate or mitigate EJ
concerns within individual states.

190 Please note that population counts vary greatly by state, and that averaging results of the 48 states shown here
will not reflect national population-weighted exposure estimates.

6-27


-------
Year Group
2028 a«'V*flc«

Air*r>eafl Incilfl

Ajiin pm, , (na/fn*)	

Sl«c»

Hispanic 4jW	ft3i

.«}

Unamptoyad
yronsarad

Bottom S'S^-s Iff* « "?*Cv»rcy
Srjlisft < wall
«Pwa#ty fcra
H0LC6r»d»0
'rrBal lare

2030

AmtrtCMfi Incur

Asian

Stoic

Hiaparie

.*» MyOtM

ijramployad

UPinsuf«d

Scttom SSS i>|Ss5



Figure 6-4 Heat Map of the State Average PM2.5 Concentration Reductions (Blue) and
Increases (Red) Due to the Final Rules Scenario Across Demographic Groups in 2028,
2030, 2035, 2040, and 2045 (jug/m3) (Alternative Scenarios are shown in Appendix C)

6.5.2.3 Distributional Results

We also present the cumulative proportion of each population exposed to ascending
levels of PM2.5 concentration changes across the contiguous U.S. averaged at the county level.
Results allow evaluation of what percentage of each subpopulation (e.g., Hispanic population) in
the contiguous U.S. experience what change in PM2.5 concentrations compared to what
percentage of the overall reference group (i.e., the total population of the contiguous U.S.)
experiences similar concentration changes from EGU emission changes under the three
regulatory options in 2028, 2030, 2035, 2040, and 2045. Concentration reductions due to the

6-28


-------
policy at the county-level are shown in Figure 6-5 (reductions due to the alternative regulatory
options are shown in Appendix C).

This distributional EJ analysis is also subject to additional uncertainties related to more
highly- resolved input parameters and additional assumptions. For example, this analysis does
not explicitly account for potential difference in underlying susceptibility, vulnerability, or risk
factors across populations to PM2.5 exposure although we have incorporated variables such as
life expectancy and redlining into our EJ analyses to serve as proxies for cumulative exposures
within communities that lead to possible underlying susceptibility and vulnerability. Nor could
we include information about differences in other factors that could affect the likelihood of
adverse impacts (e.g., exercise patterns) across groups. As the baseline scenario is similar to that
described by other RIAs (e.g., the Regulatory Impact Analysis for the Reconsideration of the
National Ambient Air Quality Standards for Particulate Matter)191, we focus on the PM2.5
changes due to these rulemakings. The vast majority of each demographic population are
predicted to experience PM2.5 concentration changes less than 0.06 |ig/m3 at the state-level under
any of the regulatory options for all five future years analyzed. While the greatest impacts, and
the greatest differential impacts across population, occur in 2035, the distributions of PM2.5
concentration changes across population demographics are all fairly similar, and the small
difference in impacts shown in the 2028, 2030, 2035, 2040, and 2045 distributional analyses of
PM2.5 reductions under the various regulatory options suggests that the regulatory options are not
likely to exacerbate or mitigate EJ PM2.5 exposure concerns for population groups evaluated.

191 https://www.federalregister.gov/documents/2023/01/27/2023-00269/reconsideration-of-the-national-ambient-air-
quality-standards-for-particulate-matter

6-29


-------
I Employed (0-99)

¦	Unemployed (0-99)

Not in the labor force (0-99)

¦	>?overty lirte (0-99)

¦	24' HS or more)

¦	Less educated (>24; no HS)

¦	White (0-99)

H American Indian (0-99)

¦	Asian (0-99)

¦	Black (0-99)

H Top 75% life expectancy (0-99)

¦	Bottom 25% life expectancy (0-99)

¦	Mo life expectancy (0-99)
I E-glssh well or better (0-99)
B English < well (0-99)

¦	HOLC Grades A-C (0-99)

¦	HOLC Grade 0(0-99)

¦	Not Graded by HOLC (0-99)

¦	Not T ribal land (0*99)

¦	Tribal land (0-99)

Figure 6-5 Distribution of PM2.5 Concentration (jig/m3) Reductions Across Populations,
Future Years for the Final Rules Scenario (Alternative scenarios are shown in Appendix C)

6.5.3 Ozone EJ Exposure A nafysis

To evaluate the potential for EJ concerns among communities with EJ concerns resulting
from exposure to ozone under the baseline and regulatory options in these rules, we characterize
the distribution of ozone exposures both prior to and following implementation of the rules, as
well as under the alternative regulatory options, in 2028, 2030, 2035, 2040, and 2045.

As this analysis is based on the same ozone spatial fields as the benefits assessment (see
Section 3 for a discussion of the spatial fields), it is subject to similar types of uncertainty. In
addition to the small magnitude of differential ozone concentration changes associated with these
final rulemakings when comparing across demographic populations, a particularly germane
limitation is that ozone, being a secondary pollutant, is the byproduct of complex atmospheric

Population

Ethnicity

Educational
Attainment

2028	2030

r r

Employment o-
Status

Insurance
Status

Life Expectancy ^

Linguistic	a.

Isolation

Poverty Status -5

#

Redlining

r

F
F
F
F

r
r
r
r

•0.W 0.00 0.0-4 0.08 -0.04 0.00 0.04 0.08 -0.04 0.00 0 04 0.03 -0.04 0.00 0.04 0.08 -0.04 0.00 0.04 0.03

6-30


-------
chemistry such that direct linkages cannot be made between specific affected facilities and
downwind ozone concentration changes based on available air quality modeling.

Ozone concentration and exposure metrics can take many forms, although only a small
number are commonly used. The analysis presented here is based on the average April-
September warm season maximum daily eight-hour average ozone concentrations (AS-M03),
consistent with the health impact functions used in the benefits assessment (Section 4). As
developing spatial fields is time and resource intensive, the same spatial fields used for the
benefits analysis were also used for the ozone exposure analysis performed here to assess
potential EJ impacts.

The construct of the AS-M03 ozone metric used for this analysis should be kept in mind
when attempting to relate the results presented here to the ozone NAAQS and when interpreting
the confidence in the association between exposures and health effects. Specifically, the seasonal
average ozone metric used in this analysis is not constructed in a way that directly relates to
NAAQS design values, which are based on daily maximum eight-hour concentrations.192 Thus,
AS-M03 values reflecting seasonal average concentrations well below the level of the NAAQS
at a particular location do not necessarily indicate that the location does not experience any daily
(eight-hour) exceedances of the ozone NAAQS. Relatedly, EPA is confident that reducing the
highest ambient ozone concentrations will result in substantial improvements in public health,
including reducing the risk of ozone-associated mortality. However, the Agency is less certain
about the public health implications of changes in relatively low ambient ozone concentrations.
Most health studies rely on a metric such as the warm-season average ozone concentration; as a
result, EPA typically utilizes air quality inputs such as the AS-M03 spatial fields in the benefits
assessment, and we judge them also to be the best available air quality inputs for this EJ ozone
exposure assessment.

6.5.3.1 National Aggregated Results

National average baseline ozone concentrations in ppb in 2028, 2030, 2035, 2040, and
2045 are shown in a heat map (Figure 6-6). Concentrations represent the total estimated daily
eight-hour maximum ozone exposure burden averaged over the 6-month April-September ozone

192 Level of 70 ppb with an annual fourth-highest daily maximum eight-hour concentration, averaged over three
years.

6-31


-------
season and are colored to visualize differences more easily in average concentrations, with
lighter green coloring representing smaller average concentrations and darker green coloring
representing larger average concentrations. When looking across all five future years, all
demographic groups will see improving ozone concentrations over time in the baseline such that
the average concentration experienced by each demographic group will be less in 2030 than in
2028, less in 2035 than in 2030, and so on. When looking within each future year's baseline,
populations with national average ozone concentrations higher than the reference population
ordered from most to least difference were: American Indian populations, Hispanic populations,
those linguistically isolated, Asian populations, those living on Tribal land, residents of HOLC
Grade A-C (i.e., not redlined) census tracts, those without a high school diploma, the
unemployed, those with the top 75 percent life expectancy or no life expectancy data available,
and children. Average national disparities observed in the baseline of these rules are fairly
consistent across the five future years and similar to those described by recent rules (e.g., the
Regulatory Impact Analysis for Federal Implementation Plan Addressing Regional Ozone
Transport for the 2015 Ozone National Ambient Air Quality Standard).193

In Figure 6-7, columns labeled "Final Rules" "Alternative 1," and "Alternative 2"
provide information regarding how all three regulatory options will impact ozone concentrations
across various populations.194 All national-level ozone concentration changes of these final
rulemakings across population groups, years, and regulatory options are predicted to be relatively
small in absolute magnitude (i.e., <0.09 ppb) relative to the magnitude of disparities in the
baseline across populations. When comparing the small changes across demographic groups,
there are some disparate impacts in 2035 for Asian populations, Hispanic populations, those
living on Tribal lands, and those linguistically isolated (Figure 6-7). However, in the other years
and regulatory options analyzed, populations are estimated to experience similar ozone
concentration reductions to that of the reference populations with the exception of the Tribal land
demographic group which is estimated to experience the largest ozone reductions of any

193	https://www.federalregister.gov/documents/2023/01/27/2023-00269/reconsideration-of-the-national-ambient-air-
quality-standards-for-particulate-matter

194	We report average exposure results to the decimal place where difference between demographic populations
become visible, as we cannot provide a quantitative estimate of the air quality modeling precision uncertainty.
Using this approach allows for a qualitative consideration of uncertainties and the significance of the relatively
small difference

6-32


-------
demographic group/year in 2028 and 2030 (up to 0.09 ppb), followed by the largest increases in
ozone of any demographic group/year in 2040 (up to 0.03 ppb).

The national-level assessment of ozone burden concentrations in the baseline and ozone
exposure changes due to the regulatory options suggests that while most policy options and
future years analyzed will not likely mitigate or exacerbate ozone EJ exposure disparities for the
population groups evaluated, ozone EJ exposure disparities may be slightly exacerbated for some
population groups analyzed in 2035 and those living on Tribal lands in 2040 as well as slightly
mitigated for those living on Tribal lands in 2028 and 2030 under all regulatory options.
However, the extent to which disparities may be exacerbated is likely modest across population
groups, at most between 0.1 percent and 0.2 percent of baseline ozone levels. Note that while we
were able to compare the annual average PM2.5 concentrations to the newly revised NAAQS, the
estimated ozone impacts in terms of annual average change are difficult to compare to the ozone
NAAQS which is reported as the fourth-highest daily maximum 8-hour concentration.

6-33


-------
Year

Population Qualifier

2028 2030 2035 2040 2045

Reference Reference (0-99)

40.3 40.2 40.0 39.9 39.8

Race White (0-99)

American Indian (0-99)

40.3 40.3 40.1 40.0 40.0

42.6 42.6 42.4 42.3 42.3

Asian (0-99)
Black (0-99)

41.6 41.5 41.3 41.1 40.9

38.9 38.8 38.6 38.4 38.3

Ethnicity Non-Hispanic (0-99)
Hispanic (0-99)

39.6 39.6 39.3 39.2 39.1

42.5 42.4 42.2 42.0 41.9

Educational Attainment More educated (>24: HS or more)
Less educated (>24; no HS)

40.1 40.0 39.8 39.7 39.6
40.8 40.7 40.6 40.5 40.4

Employment Status Employed (0-99)

Unemployed (0-99)

Not in the labor force (0-99)

40.3 40.2 40.0 39.9 39.8
40.7 40.7 40.5 40.4 40.3
40.2 40.2 40.0 39.9 39.8

Insurance Status Insured (0-64)

Uninsured (0-64)

40.4 40.4 40.2 40.0 40.0
40.0 39.9 39.7 39.6 39.5

Life Expectancy Top 75% life expectancy (0-99)

40.5 40.5 40.3 40.2 40.1

Bottom 25% life expectancy (0-99)

Life expectancy data unavailable (0-99)

39.1 39.1 38.8 38.7 38.6

40.6 40.5 40.3 40.2 40.1

Linguistic Isolation English well or better (0-99)
English < well (0-99)

40.2 40.1 39.9 39.8 39.7

41.9 41.8 41.6 41.5 41.4

Poverty Status >200%ofthe poverty line (0-99)
<200%ofthe poverty line (0-99)
>Povertyline (0-99)


-------
Year / Scenario
2035

Population	Qualifier

Reference	Reference (0-99)

Race	White (0-99)

American Indian (0-99)

Asian (0-99)

	Black (0-99)	

Ethnicity	Non-Hispanic (0-99)

Hispanic (0-99)

Educational Atta.. More educated (>24: HS or more)

Less educated (>24; no HS)

Employment Employed (0-99)

Status	Unemployed (0-99)

Not in the labor force (0-99)

Insurance Status insured (0-64)

Uninsured (0-64)

Life Expectancy Top 75% life expectancy (0-99)

Bottom 25% life expectancy (0-99)

Life expectancy data unavailable (0-99)
Linguistic Isolati.. English well or better (0-99)

English < well (0-99)

Poverty Status >200% of the poverty I ine (0-99)
<200% of the poverty line (0-99)
>Povertyline (0-99)


-------
California, a 0.3 percent change relative to the baseline ozone level in 2035. This is notable
given that California's central valley and LA basin have higher baseline ozone concentrations
than other parts of the country in all modeled years, and these regions are home to many
disadvantaged communities.195 Importantly, Figure 6-8 shows that demographic groups within
most states are predicted to experience very similar exposure impacts as the state reference
populations, with a few potential exceptions (e.g., Iowa, Nebraska, South Dakota, and West
Virginia in 2035, 2040, and 2045). When comparing exposure impacts across demographic
groups within states, most states display similar impacts across demographic groups in 2028,
2035, 2040, and 2045. However, some populations with higher exposures have larger differences
in reductions between groups. For example, within several states, the largest difference in
reductions between a population and the reference population is 0.11 ppb. Therefore, the state-
level assessment of ozone exposure changes due to the regulatory options suggests that while
most policy options and future years analyzed will not likely mitigate or exacerbate ozone EJ
exposure disparities for the population groups evaluated in 2028, 2035, 2040, and 2045, ozone
EJ exposure disparities at the state level may be either mitigated or exacerbated for some
population groups analyzed in 2035, 2040, and 2045 under the various regulatory options.
However, the extent to which disparities may be exacerbated or mitigated is likely modest, due
to the small magnitude of the ozone concentration changes relative to the magnitude of baseline
ozone exposure disparities (between 0.3 percent to 1.4 percent of baseline ozone levels).

195 See Tables B8-12 in the Appendix for more information about the modeled air quality in California and the
California Air Resources Board's CalEnviroScreen website about more information regarding demographic
groups living in affected areas with higher ozone concentrations

(https://experience.arcgis.com/experience/lc21c53da8de48flb946f3402fbae55c/page/SB-535-Disadvantaged-
Communities/)

6-36


-------
Year
2028

Group

P«fflren«

Air*n«r Inf—

Asian 0ion« (ppt>)

Black [ t JiH

Kjpjnic -012

less

UitwnployM
Unirs^rec

Bsttsm 25^ We «i&*ct*ran
Asian
Bteek

Kigpinii

less ec.catec

Unemployed

Uninsured

Bsttam 23=% l.fe e*pectarcy

Engl «sh < well

«?9Verty line

KDtCGra<3eD

Tribal land

Reference

«r^n;jr InCur

Asian

Black

Hispanic

less educated

Unemployed

Uninsured

Batten SS4^ life expectancy
Engl !$h < well
^Poverty ftne
HOlC Grade 0
Tribal land

3S23S!

• i=5S-Si-£_i£ssS2SSS2Z22

vS>>i*a< = uozxi
:22200058l!iwK*;

35|§is





Figure 6-8 Heat Map of the State Average Ozone Concentrations Reductions (Green)
and Increases (Red) Due to the Final Rules Scenario Across Demographic Groups in 2028,
2030, 2035, 2040, and 2045 (ppb) (Alternative Scenarios are shown in Appendix C)

6.5.3.3 Distributional Results

We also present cumulative proportion of each population exposed to ascending levels of
ozone concentration changes across the contiguous U.S. Results allow evaluation of what
percentage of each subpopulation (e.g., Hispanic population) in the contiguous U.S. experience
what change in ozone concentrations compared to what percentage of the overall reference group
(i.e., the total population of contiguous U.S.) experiences similar concentration changes from
EGU emission changes under the three regulatory options in 2028, 2030, 2035, 2040, and 2045.

This distributional EJ analysis is also subject to additional uncertainties related to more
highly resolved input parameters and additional assumptions. For example, this analysis does not

6-37


-------
explicitly account for potential differences in underlying susceptibility, vulnerability, or risk
factors across populations expected to experience post-policy ozone exposure changes although
we have incorporated variables such as life expectancy and redlining into our demographic to
serve as proxies for cumulative exposures within communities that lead to possible underlying
susceptibility and vulnerability. Nor could we include information about differences in other
factors that could affect the likelihood of adverse impacts (e.g., exercise patterns) across groups.

As the baseline scenario is similar to that described by other RIAs (the Regulatory Impact
Analysis for Federal Implementation Plan Addressing Regional Ozone Transport for the 2015
Ozone National Ambient Air Quality Standard)196, we focus on the ozone changes due to these
rulemakings. Distributions of 12 km gridded ozone concentration changes from EGU control
strategies of affected facilities under the final rules scenario analyzed in these final rulemakings
are shown in Figure 6-9 (alternative regulatory options are shown in Appendix C). When
comparing distributional exposure impacts across demographic groups, similar impacts are
predicted to occur across demographic groups in 2028, 2030, 2040, and 2045. However, certain
groups, specifically Asian populations, Hispanic populations, those linguistically isolated, and
those living on Tribal land may experience smaller ozone exposure reductions across the
population distributions in 2035, as compared to the overall reference distribution. Additionally,
those living on Tribal lands may experience small ozone exposure increases across the
population distributions in 2040 as well as larger ozone exposure reductions across the
population distributions in 2028 and 2030. Therefore, the distributional assessment of ozone
exposure changes due to the regulatory options suggests that while most regulatory options and
future years analyzed will not likely mitigate or exacerbate ozone EJ exposure disparities for the
population groups evaluated in 2028, 2035, 2040, and 2045, distributional ozone EJ exposure
disparities may be slightly exacerbated for some population groups analyzed in 2035 and those
living on Tribal lands in 2040 as well as slightly mitigated for those living on Tribal lands in
2028 and 2030 under all regulatory options. However, the extent to which disparities may be
exacerbated is likely modest, due to the small magnitude of the ozone concentration changes
(will all changes between 0.3 to 1.4 percent of baseline ozone levels).

196 https://www.federalregister.gov/documents/2023/01/27/2023-00269/reconsideration-of-the-national-ambient-air-
quality-standards-for-particulate-matter

6-38


-------
Population

ftac e

Ethnicity

Educational
Attainment

Employment
Status

insurance
Status

Life Expectancy

Linguistic
Isolation

Poverty Status

Redlining

¦fc ^

n

r
r
r
r
r
r
r
r
r
r

| Employed (0-99)

| Unemployed (0-99)

Not In the labor force (0-99)

| >Poverty line (0*99)

| 24: HS or more)

I Less educated (>24; no HS)

| White (0-99)

American Indian (0-99)

| Asian (0-99)

I Black (0*99)

| Top 75% iife expectancy (0-99.1
I Bottom 25% life expectancy (0-99)
I Life expectancy data unavailable (0-99)
| English well or better (0-99)

| English < well (0-99)

|H0LC Grades A-C (0-99)

| HOLC Grade D (0-99)

I Not Graded by HOLC (0-99)

| Not Tribal land (0-99)

| Tribal land (0-99)

¦0.3 -0.1 0.1 0 3 0.5 -0.3 -0.1 0.1 0.3 0.5 -0.3 -0.1 0.1 0.3 0.5 -0.3 -0.1 01 0.3 0.5 -0.3 -0.1 0.1 0.3 0.5

Figure 6-9 Distributions of Ozone Concentration Changes (ppb) Across Populations,
Future Years for the Final Rules Scenario (Alternative Scenarios are shown in Appendix
C)

6.6 Qualitative Discussion of E J PM2.5 Health Impacts

While the potential for EJ concerns related to PM2.5 health outcomes (i.e., premature
mortality) among populations potentially at increased risk of or to PM2.5 exposures have been
evaluated previously (U.S. EPA, 2022a), EJ health impacts of PM2.5 exposures were not
quantitatively evaluated here, due to resource limitations and the lack of substantial differential
EJ impacts of the final rulemaking (Section 3.8).

While quantitative impacts are not analyzed, we can qualitatively speak to the expected
PM2.5-attributable mortality EJ impacts of these final rules, based on prior quantitative results
and the PM2.5 EJ exposure results provided here. For context, the PM ISA and PM ISA

6-39


-------
Supplement provided evidence that there are consistent racial and ethnic disparities in PM2.5
exposure across the U.S., particularly for Black populations, as compared to non-Hispanic White
populations. Additionally, some studies provided evidence of increased PIVh.s-related mortality
and other health effects from long-term exposure to PM2.5 among Black populations. Taken
together, the 2019 PM ISA concluded that the evidence was adequate to conclude that race and
ethnicity modify PM2.5-related risk, and that non-White individuals, particularly Black
individuals, are at increased risk for PM2.5-related health effects, in part due to disparities in
exposure ISA (U.S. EPA, 2019, 2022b).

Qualitatively, as the PM2.5 exposure changes are fairly consistent across demographic
populations, differential impacts are expected to reflect the epidemiologic hazard ratios. This
suggests that PM2.5 improvements would be most beneficial for Black populations, followed by
Hispanic and Asian populations. Conversely, worsening air quality would be disproportionately
harmful to the same groups in the same hierarchy.

6.7	Qualitative Discussion of New Source EJ Impacts

EJ impacts of new sources subject to 111(b) are highly uncertain as the location of new
sources is unknown. Therefore, we do not make predictions regarding potential EJ impacts from
new sources. However, the regulatory options do account for emissions changes at existing
facilities that are expected to result from the 111(b) policy.

6.8	Summary

As with all EJ analyses, data limitations make it quite possible that disparities may exist
that our analysis did not identify. This is especially relevant for potential EJ characteristics,
environmental impacts, and more granular spatial resolutions that were not evaluated. Therefore,
this analysis is only a partial representation of the distributions of potential impacts.

Additionally, EJ concerns for each rulemaking are unique and should be considered on a case-
by-case basis.

For these final rules, we quantitatively evaluate the proximity of affected facilities
populations of potential EJ concern (Section 4) and the potential for disproportionate pre- and
post-policy PM2.5 and ozone exposures and exposure changes across different demographic

6-40


-------
groups (Section 5). Each of these analyses was performed to answer separate questions and is
associated with unique limitations and uncertainties.

Baseline demographic proximity analyses provide information as to whether there may
be potential EJ concerns associated with environmental stressors. In this case, the proximity
analysis of the full population of potentially affected units greater than 25 MW (114 facilities)
indicated that the demographic percentages of the population within 5 km and 10 km of the
facilities are relatively similar to the national averages with the exception of the American Indian
population (1 percent and 0.8 percent, respectively) that is higher than the national average (0.6
percent). This higher percentage is driven mostly by 7 facilities that have an American Indian
percentage within a 5 km and 10 km radius that ranges from 10 percent to just above 40 percent
which is substantially above the national average (0.6 percent). The population living below the
federal poverty line (14 percent for both distances) as well as the population living below 2x the
federal poverty level (34 percent and 33 percent, respectively) that are both higher than the
national averages (13 percent and 29 percent, respectively). The proximity analysis of the 23
plants with known retirement plans from 2033 to 2040, (a subset of the total 114 plants) found
that the percentages of the population within 5 km and 10 km that is below the poverty line (14
percent both distances) and below 2x the federal poverty line (33 percent and 31 percent,
respectively) are both higher than the national average percentages (13 percent and 29 percent,
respectively). The proximity analysis for the 94 plants without known retirement plans before
2040, (a subset of the total 114 units) shows demographics similar to the total 114 facilities'
proximity analysis.

While the demographic proximity analyses may appear to parallel the baseline analysis of
nationwide ozone and PM2.5 exposures in certain ways, the two should not be directly compared.
The baseline ozone and PM2.5 exposure assessments are in effect an analysis of total burden in
the contiguous U.S., and include various assumptions, such as the implementation of
promulgated regulations. It serves as a starting point for both the estimated ozone and PM2.5
changes due to these final rules as well as a snapshot of air pollution concentrations in several
near future years.

The baseline ozone and PM2.5 exposure analyses respond to question 1 from EPA's EJ
Technical Guidance document more directly than the proximity analyses, as they evaluate a form

6-41


-------
of the environmental stressor primarily affected by the regulatory action (Section 5). Certain
populations, such as those who are residents of HOLC Grade D (i.e., redlined) census tracts,
those linguistically isolated, residents of HOLC Grade A-C (i.e., not redlined) census tracts,
Hispanic populations, Asian populations, and those without a high school diploma may
experience disproportionately higher PM2.5 and ozone concentrations than the reference group.
Black populations may experience disproportionately higher PM2.5 concentrations than the
reference group, and populations that are American Indian, living on Tribal land, residents of
HOLC Grade D (i.e., redlined) census tracts, the unemployed, those with the top 75 percent life
expectancy or no life expectancy data available, and children may also experience
disproportionately higher ozone concentrations than the reference group. Therefore, there likely
are potential EJ concerns associated with environmental stressors affected by the regulatory
action for population groups of concern in the baseline.

Finally, we evaluate how the post-policy options of these final rulemakings are expected
to differentially impact demographic populations, informing questions 2 and 3 from EPA's EJ
Technical Guidance regarding ozone and PM2.5 exposure changes. PM2.5 and ozone exposure
analyses show that the final rules will result in modest but widespread reductions in PM2.5 and
ozone concentrations, although some limited areas may experience small increases in ozone
concentrations relative to forecasted conditions without the rule. We infer that baseline
disparities in ozone and PM2.5 concentration burdens are likely to remain after implementation of
any of the regulatory options under consideration due to the small magnitude of the
concentration changes associated with these rulemakings across demographic populations,
relative to baseline burden disparities (with the largest changes being only 1.4 percent of baseline
concentrations) (EJ question 2). Also, due to the very small differences in the distributional
analyses of post-policy exposure impacts across demographic populations, we do not find
evidence that disparities in populations with potential EJ concerns will be exacerbated or
mitigated by the regulatory alternatives under consideration regarding PM2.5 exposures in all
future years evaluated and ozone exposures in 2028, 2030, 2040, and 2045. However, in 2035,
Asian populations, Hispanic populations, those linguistically isolated, and those living on Tribal
land may experience a slight exacerbation of ozone exposure disparities (up to 0.05 ppb different
than the reference group) at the national level under all regulatory options. Additionally, those
living on Tribal lands in 2040 may experience a slight exacerbation of ozone exposure disparities

6-42


-------
in 2040 (up to 0.03 ppb different than the reference group) as well as a slight mitigation of ozone
exposure disparities in 2028 and 2030 (up to 0.07 ppb different than the reference group) (EJ
question 3). At the state level, ozone exposure disparities may be either mitigated or exacerbated
for certain demographic groups analyzed in 2035, also to a small degree (up to 0.12 ppb different
than the reference group).

This EJ air quality analysis concludes that there are disparities across various populations
in the pre-policy baseline scenario (EJ question 1) and infer that these disparities are likely to
persist after promulgation of these final rulemakings (EJ question 2). This EJ assessment also
suggests that this action is unlikely to mitigate or exacerbate PM2.5 exposures disparities across
populations of EJ concern analyzed. Regarding ozone exposures, while most snapshot years for
the regulatory options analyzed will not likely mitigate or exacerbate ozone exposure disparities
for the population groups evaluated, ozone exposure disparities may be slightly exacerbated for
some population groups analyzed in 2035, slightly exacerbated for those living on Tribal lands in
2040, and slightly mitigated for those living on Tribal lands in 2028 and 2030 under all
regulatory options. However, the extent to which disparities may be exacerbated or mitigated is
likely modest, due to the small magnitude of the ozone concentration changes relative to baseline
disparities across populations (EJ question 3). Importantly, the action described in these final
rules is expected to lower PM2.5 and ozone in many areas, and thus mitigate some pre-existing
health risks of air pollution across all populations evaluated.

6-43


-------
6.9 References

Ebi, K. L., Hasegawa, T., Hayes, K., Monaghan, A., Paz, S., & Berry, P. (2018). Health risks of
warming of 1.5 °C, 2 °C, and higher, above pre-industrial temperatures. Environmental
Research Letters, 13(6), 063007. doi:10.1088/1748-9326/aac4bd

IPCC. (2018). Global Warming of 1.5°C. An IPCC Special Report on the impacts of global
warming of 1.5°C above pre-industrial levels and related global greenhouse gas
emission pathways, in the context of strengthening the global response to the threat of
climate change, sustainable development, and efforts to eradicate poverty (V. Masson-
Delmotte, P. Zhai, H.-O. Portner, D. Roberts, J. Skea, P. R. Shukla, A. Pirani, W.
Moufouma-Okia, C. Pean, R. Pidcock, S. Connors, J. B. R. Matthews, Y. Chen, X. Zhou,
M. I. Gomis, E. Lonnoy, T. Maycock, a. M. Tignor, & T. Waterfield Eds.).

Jantarasami, L., Novak, R., Delgado, R., Marino, E., McNeeley, S., Narducci, C., . . . Powys

Whyte, K. (2018). Tribes and Indigenous Peoples. In Impacts, Risks, and Adaptation in
the United States: Fourth National Climate Assessment (Vol. II). Washington, DC: U.S.
Global Change Research Program.

National Academy of Sciences. (2019). Climate Change and Ecosystems. Washington DC: The
National Academies Press.

Oppenheimer, M., Campos, M., Warren, R., Birkmann, J., Luber, G., O'Neill, B., & Takahashi,
K. (2014). Emergent risks and key vulnerabilities. In C.B. Field, V.R. Barros, D.J.
Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatteijee, K.L. Ebi, Y.O. Estrada,
R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, &
L.L.White (Eds.), 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 (pp. 1039-1099). Cambridge,
United Kingdom and New York, NY: Cambridge University Press.

Porter, J. R., Xie, L., Challinor, A. J., Cochrane, K., Howden, M., Iqbal, M. M., & Lobell, D. B.
(2014). Food security and food production systems. In C.B. Field, V.R. Barros, D.J.
Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatteijee, K.L. Ebi, Y.O. Estrada,
R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, &
L.L.White (Eds.), 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 (pp. 485-533). Cambridge,
United Kingdom and New York, NY: Cambridge University Press.

Smith, K. R., Woodward, A., Campbell-Lendrum, D., Chadee, D. D., Honda, Y., Liu, Q., . . .
Sauerborn, R. (2014). Human Health: Impacts, Adaptation, and Co-Benefits. In C.B.
Field, V.R. Barros, D.J. Dokken, K.J. 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, & L.L.White (Eds.), Climate Change 2014: Impacts, Adaptation, and
Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to

6-44


-------
the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (pp. 709-
754). Cambridge, United Kingdom and New York, NY: Cambridge University Press.

U.S. EPA. (2019). Integrated Science Assessment (ISA) for Particulate Matter (Final Report).
(EPA/600/R-19/188). Research Triangle Park, NC: U.S. Environmental Protection
Agency, Office of Research and Development, Center for Public Health and
Environmental Assessment, https://www.epa.gov/naaqs/particulate-matter-pm-standards-
integrated-science-assessments-current-review

U.S. EPA. (2021). Climate Change and Social Vulnerability in the United States: A Focus on Six
Impacts. (EPA 43O-R-21-003). Washington DC.

https://www.epa.gov/system/files/documents/2021-09/climate-vulnerabilitv september-
2021 508.pdf

U.S. EPA. (2022a). Regulatory Impact Analysis for the Proposed Reconsideration of the

National Ambient Air Quality Standards for Particulate Matter. (EPA-452/P-22-001).
Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Health and Environmental Impact Division.
https://www.epa.gov/svstem/files/documents/2023-01/naaqs-pm ria proposed 2022-
12.pdf

U.S. EPA. (2022b). Supplement to the 2019 Integrated Science Assessment for Particulate
Matter (FinalReport). (EPA/600/R-22/028). Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Research and Development, Center for
Public Health and Environmental Assessment.
https://cfpub.epa. gov/ncea/isa/recordisplav.cfm?deid=354490

U.S. EPA. (2023). Climate Change and Children's Health and Weil-Being in the United States.
(EPA 430-R-23-001). https://www.epa.gov/svstem/files/documents/2023-
04/CLiME Final%20Report.pdf

USGCRP. (2016). The Impacts of Climate Change on Human Health in the United States: A
Scientific Assessment. Washington DC: U.S. Global Change Research Program.
http://dx.doi.org/10.7930/J0R49NQX

USGCRP. (2018). Impacts, Risks, and Adaptation in the United States: Fourth National Climate
Assessment, Volume II. Washington DC: U.S. Global Change Research Program.
http://dx.doi.org/10.7930/NCA4.2018

Woods & Poole. (2015). Complete Demographic Database. Retrieved from
https://www.woodsandpoole.com/

6-45


-------
7 COMPARISON OF BENEFITS AND COSTS

7.1	Introduction

This section presents the estimates of the climate benefits, health benefits, compliance
costs, and net benefits associated with the illustrative scenarios analyzed in this RIA. There are
potential benefits and costs that may result from the final rules that have not been quantified or
monetized. Due to data and modeling limitations, there are still many categories of climate
impacts and associated damages that are not reflected yet in the monetized climate benefits from
reducing C02 emissions. For example, the modeling omits most of the consequences of changes
in precipitation, damages from extreme weather events, the potential for nongradual damages
from passing critical thresholds (e.g., tipping elements) in natural or socioeconomic systems, and
non-climate mediated effects of GHG emissions (e.g., ocean acidification). Unquantified benefits
also include climate benefits from reducing emissions of non-CCh greenhouse gases and benefits
from reducing exposure to SO2, NOx, and hazardous air pollutants (e.g., mercury), as well as
ecosystem effects and visibility impairment. Additionally, there may be health, ecological, and
productivity damages associated with water effluent and intake from coal generation that will be
avoided by these final rules.

The compliance costs reported in this Section are not social costs; instead, we use
compliance costs as a proxy for social costs. Economy-wide social costs are separately estimated
and discussed in Section 5.2, but those estimates are not applied in this section. Therefore, in this
section, we do not account for changes in costs and benefits due to changes in economic welfare
in the broader economy arising from shifts in production and consumption that may be induced
by the final requirements. Furthermore, costs and benefits due to interactions with pre-existing
market distortions outside the electricity sector are omitted, as are social costs that may be
associated with the net change in power sector subsidies under the final rules. Additional
limitations of the analysis and sources of uncertainty are described throughout the RIA and
summarized in the executive summary.

7.2	Methods

EPA calculated the PV of costs, benefits, and net benefits for the years 2024 through
2047, using the discount rates of two percent, three percent, and seven percent from the

7-1


-------
perspective of 2024. All dollars are in 2019 dollars. In order to implement the OMB Circular A-4
requirement for fulfilling E.O. 12866, we assess two scenarios representing alternative sets of
requirements.

This calculation of a PV requires an annual stream of values for each year of the 2024 to
2047 timeframe. All cost and benefit analysis begins in 2028, except MR&R costs which are
estimated to begin in 2024. EPA used IPM to estimate cost and emission changes for the
projection years 2028, 2030, 2035, 2040 and 2045. The final rules have requirements that come
into effect in different years, and the snapshot years approximate the different rule requirements
over the timeframe of analysis in this RIA. For details on how the three illustrative scenarios
reflect the requirements of the rules, see Section 3.2.

In the IPM modeling for this RIA, the 2028 projection year is representative of 2028 and
2029, the 2030 projection year is representative of 2030 and 2031, the 2035 projection year is
representative of 2032 to 2037, the 2040 projection year is representative of 2038 to 2041, and
the 2045 projection year is representative of 2042 through 2047. Estimates of costs and emission
changes in other years are determined from the mapping of projection years to the calendar years
that they represent. Consequently, the cost and emission estimates from IPM in each projection
year are applied to the years which it represents.

Climate benefits estimates are based on these projection year emission estimates and also
account for year-specific SC-CO2 values. Health benefits are based on projection year emission
estimates and also account for year-specific variables that influence the size and distribution of
the benefits. These variables include population growth, income growth, and the baseline rate of
death.

7.3 Results

Table 7-1 through Table 7-4 present the estimates of the projected compliance costs,
climate benefits, health benefits, and net benefits across the three illustrative scenarios for the
snapshot years 2028, 2030, 2035, 2040 and 2045, respectively. The comparison of benefits and
costs in PV and EAV terms for the final rules can be found in Table 7-6 for the illustrative final
rules scenario; Table 7-7 presents the results for the alternative 1 illustrative scenario; and Table
7-8 presents results for the alternative 2 illustrative scenario. Estimates in the tables are presented

7-2


-------
as rounded values. In this net benefits analysis, climate benefits are discounted using a two
percent discount rate only.197 Therefore, in Table 7-6 through Table 7-8, the net benefits
estimates under all discount rates reflect this two percent discounting of climate benefits.

As discussed in Section 4 of this RIA, the monetized benefits estimates provide an
incomplete overview of the beneficial impacts of the final rule. In particular, the monetized
climate benefits are incomplete and an underestimate as explained in Section 4.2. In addition,
important health, welfare, and water quality benefits anticipated under these final rules are not
quantified or monetized. EPA anticipates that taking non-monetized effects into account would
show the final rules to have greater benefits than the tables in this section reflect.

Simultaneously, the estimates of compliance costs used in the net benefits analysis may provide
an incomplete characterization of the true costs of the rule. The balance of unquantified benefits
and costs is ambiguous but is unlikely to change the result that the benefits of the final rules
exceed the costs by billions of dollars annually.

We also note that the RIA follows EPA's historical practice of using a technology-rich
partial equilibrium model of the electricity and related fuel sectors to estimate the incremental
costs of producing electricity under the requirements of proposed and final major EPA power
sector rules. In Section 5.2 of this RIA, EPA has also included an economy-wide analysis that
considers additional facets of the economic response to the final rules, including the full resource
requirements of the expected compliance pathways, some of which are paid for through
subsidies. The social cost estimates in the economy-wide analysis discussed in Section 5.2 are
still far below the projected benefits of the final rules.

197 Monetized climate benefits are discounted using a 2 percent discount rate, consistent with EPA's updated

estimates of the SC-CO2. OMB has long recognized that climate effects should be discounted only at appropriate
consumption-based discount rates. Because the SC-CO2 estimates reflect net climate change damages in terms of
reduced consumption (or monetary consumption equivalents), the use of the social rate of return on capital (7
percent under OMB Circular A-4 (2003)) to discount damages estimated in terms of reduced consumption would
inappropriately underestimate the impacts of climate change for the purposes of estimating the SC-CO2. See
Section 4.2 for more discussion.

7-3


-------
Table 7-1 Net Benefits of the Three Illustrative Scenarios in 2028 (billion 2019
dollars)a,b	

Final Rules

Alternative 1

Alternative 2

Climate Benefitsc



8.4





7.9





7.1



PM2.5 and O3-
related Health
Benefits d

2.6

and

5.8

2.3

and

5.2

2.1

and

4.8

Total Benefits

11

and

14

10

and

13

9.2

and

12

Compliance Costse



-1.3



-1.1

-1.1

Net Benefits

12

and

15

11

and

14

10

and

13

Non-Monetized Benefits'

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 We focus results to provide a snapshot of costs and benefits in 2028, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
c Monetized climate benefits are based on reductions in CO2 emissions and are calculated using three different
estimates of the social cost of CO2 (SC-CO2) (under 1.5 percent, 2.0 percent, and 2.5 percent near-term Ramsey
discount rates). For the presentational purposes of this table, we show the climate benefits associated with the SC-
CO2 at the 2 percent near-term Ramsey discount rate. Please see Table 4-5 for the full range of monetized climate
benefit estimates. See Section 4.2 for a discussion of the uncertainties associated with the climate benefit estimates.
d Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The two columns for each scenario of the health benefits represent different studies to estimate
premature deaths among adults. The health benefits are associated with several point estimates and are presented at a
real discount rate of 2 percent.

e The discount rate in IPM is 3.76 percent, as described in Section 3.

f Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

7-4


-------
Table 7-2 Net Benefits of the Three Illustrative Scenarios in 2030 (billion 2019
dollars)a,b	

Final Rules

Alternative 1

Alternative 2

Climate Benefitsc



11



11





6.2

PM2.5 and O3-
related Health
Benefits d

1.8

and

4.0

1.5 and

3.6

1.2

and 2.8

Total Benefits

13

and

15

12 and

14

7.4

and 9.0

Compliance Costse



-0.22



-0.046

-0.72

Net Benefits

13

and

16

13 and

15

8.1

and 9.7

Non-Monetized Benefitsf

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 We focus results to provide a snapshot of costs and benefits in 2030, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
c Monetized climate benefits are based on reductions in CO2 emissions and are calculated using three different
estimates of the social cost of CO2 (SC-CO2) (under 1.5 percent, 2.0 percent, and 2.5 percent near-term Ramsey
discount rates). For the presentational purposes of this table, we show the climate benefits associated with the SC-
CO2 at the 2 percent near-term Ramsey discount rate. Please see Table 4-5 for the full range of monetized climate
benefit estimates. See Section 4.2 for a discussion of the uncertainties associated with the climate benefit estimates.
d Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates and are presented at a real discount
rate of 2 percent.

e The discount rate in IPM is 3.76 percent, as described in Section 3.

f Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

7-5


-------
Table 7-3 Net Benefits of the Three Illustrative Scenarios in 2035 (billion 2019
dollars)a,b	

Final Rules

Alternative 1

Alternative 2

Climate Benefitsc



30





30





30



PM2.5 and O3-
related Health
Benefits d

6.9

and

15

7.1

and

15

7.1

and

15

Total Benefits

37

and

45

37

and

46

37

and

45

Compliance Costse



1.3



1.2

1.2

Net Benefits

36

and

44

36

and

44

36

and

44

Non-Monetized Benefitsf

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 We focus results to provide a snapshot of costs and benefits in 2035, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
c Monetized climate benefits are based on reductions in CO2 emissions and are calculated using three different
estimates of the social cost of CO2 (SC-CO2) (under 1.5 percent, 2.0 percent, and 2.5 percent near-term Ramsey
discount rates). For the presentational purposes of this table, we show the climate benefits associated with the SC-
CO2 at the 2 percent near-term Ramsey discount rate. Please see Table 4-5 for the full range of monetized climate
benefit estimates. See Section 4.2 for a discussion of the uncertainties associated with the climate benefit estimates.
d Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates and are presented at a real discount
rate of 2 percent.

e The discount rate in IPM is 3.76 percent, as described in Section 3.

f Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

7-6


-------
Table 7-4 Net Benefits of the Three Illustrative Scenarios in 2040 (billion 2019
dollars)a,b	

Final Rules

Alternative 1

Alternative 2

Climate Benefitsc



14





14





14



PM2.5 and O3-
related Health
Benefits d

-0.14

and

-0.35

0.012

and

0.00043

0.091

and 0.087

Total Benefits

14

and

14

14

and

14

14

and

14

Compliance Costse



0.59



0.64

0.60

Net Benefits

13

and

13

13

and

13

13

and

13

Non-Monetized Benefitsf

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 We focus results to provide a snapshot of costs and benefits in 2040, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
c Monetized climate benefits are based on reductions in CO2 emissions and are calculated using three different
estimates of the social cost of CO2 (SC-CO2) (under 1.5 percent, 2.0 percent, and 2.5 percent near-term Ramsey
discount rates). For the presentational purposes of this table, we show the climate benefits associated with the SC-
CO2 at the 2 percent near-term Ramsey discount rate. Please see Table 4-5 for the full range of monetized climate
benefit estimates. See Section 4.2 for a discussion of the uncertainties associated with the climate benefit estimates.
d Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates and are presented at a real discount
rate of 2 percent.

e The discount rate in IPM is 3.76 percent, as described in Section 3.

f Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

7-7


-------
Table 7-5 Net Benefits of the Three Illustrative Scenarios in 2045 (billion 2019
dollars)a,b	

Final Rules

Alternative 1

Alternative 2

Climate Benefitsc



12





11





11



PM2.5 and O3-
related Health
Benefits d

3.6

and

8.2

3.7

and

8.2

3.7

and

8.2

Total Benefits

16

and

20

15

and

20

15

and

20

Compliance Costse



3.3



3.3

3.6

Net Benefits

12

and

17

12

and

16

11

and

16

Non-Monetized Benefits'

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 We focus results to provide a snapshot of costs and benefits in 2045, using the best available information to
approximate social costs and social benefits recognizing uncertainties and limitations in those estimates.
b Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
c Monetized climate benefits are based on reductions in CO2 emissions and are calculated using three different
estimates of the social cost of CO2 (SC-CO2) (under 1.5 percent, 2.0 percent, and 2.5 percent near-term Ramsey
discount rates). For the presentational purposes of this table, we show the climate benefits associated with the SC-
CO2 at the 2 percent near-term Ramsey discount rate. Please see Table 4-5 for the full range of monetized climate
benefit estimates. See Section 4.2 for a discussion of the uncertainties associated with the climate benefit estimates.
d Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates and are presented at a real discount
rate of 2 percent.

e The discount rate in IPM is 3.76 percent, as described in Section 3.

f Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

7-8


-------
Table 7-6 Net Benefits of the Final Rules Illustrative Scenario for 2024 to 2047 (billion
2019 dollars) a	



Climate
Benefitsb

PM2.5 and 03-related
Health Benefits0

Compliance

Pneted



Net
Benefits0





2%

2%

3%

7%

01SI IS

2%

3%

7%

2024

-

-

-

-

0.011

-0.011

-0.011

-0.011

2025

-

-

-

-

0.011

-0.011

-0.011

-0.011

2026

-

-

-

-

0.011

-0.011

-0.011

-0.011

2027

-

-

-

-

0.000070

-0.000070

-0.000070

-0.000070

2028

8.4

5.8

5.6

5.0

-1.3

15

15

15

2029

8.5

5.9

5.8

5.1

-1.3

16

16

15

2030

11

4.0

3.9

3.5

-0.22

16

15

15

2031

12

4.1

4.0

3.5

-0.22

16

16

15

2032

29

14

13

12

1.3

41

41

39

2033

29

14

14

12

1.3

42

42

40

2034

30

14

14

12

1.3

43

43

41

2035

30

15

14

13

1.3

44

43

42

2036

31

15

15

13

1.3

44

44

42

2037

31

15

15

13

1.3

45

45

43

2038

14

-0.34

-0.33

-0.29

0.59

13

13

13

2039

14

-0.34

-0.33

-0.30

0.59

13

13

13

2040

14

-0.35

-0.34

-0.30

0.59

13

13

13

2041

14

-0.36

-0.35

-0.31

0.59

13

13

13

2042

11

7.9

7.7

6.8

3.3

16

16

15

2043

12

8.0

7.7

6.9

3.3

16

16

15

2044

12

8.1

7.8

6.9

3.3

16

16

15

2045

12

8.2

7.9

7.0

3.3

17

16

16

2046

12

8.3

8.0

7.1

3.3

17

17

16

2047

12

8.3

8.1

7.2

3.3

17

17

16



Climate
Benefitsb

PM2.5 and 03-related
Health Benefits0

Compliance
Costs'1



Net
Benefits0













Discount Rate









2%

2%

3%

7%

2% 3% 7%

2%

3%

7%

PV

270

120

100

59

19 15 7.5

370

360

320

EAV

14

6.3

6.1

5.2

0.98 0.91 0.65

20

19

19

Non-Monetized Benefits'

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Annual values from 2024 to 2047 are not discounted. PV and EAV values discounted to 2024. Values have been
rounded to two significant figures. Rows may not appear to add correctly due to rounding.
b Monetized climate benefits are based on reductions in CO2 emissions and are calculated using three different
estimates of the social cost of CO2 (SC-CO2) (under 1.5 percent, 2.0 percent, and 2.5 percent near-term Ramsey
discount rates). For the presentational purposes of this table, we show the climate benefits associated with the SC-
CO2 at the 2 percent near-term Ramsey discount rate. Please see Table 4-3 for the full range of monetized climate
benefit estimates. See Section 4.2 for a discussion of the uncertainties associated with the climate benefit estimates.

7-9


-------
c The health benefits estimates use the larger of the two benefits estimates presented in Table 4-15 through Table 4-
19. Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates.
d The discount rate in IPM is 3.76 percent, as described in Section 3.

e In this net benefits analysis, health benefits and costs are discounted at the rates shown in the table (i.e., two
percent, three percent, and seven percent). Climate benefits are discounted using a two percent discount rate only in
this net benefits analysis.

f Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

7-10


-------
Table 7-7 Net Benefits of the Alternative 1 Illustrative Scenario for 2024 to 2047
(billion 2019 dollars) a	



Climate
Benefitsb

PM2.5 and 03-related
Health Benefits0

Compliance
Costs'1



Net
Benefits0





2%

2%

3%

7%

2%

3%

7%

2024

-

-

-

-

0.011

-0.011

-0.011

-0.011

2025

-

-

-

-

0.011

-0.011

-0.011

-0.011

2026

-

-

-

-

0.011

-0.011

-0.011

-0.011

2027

-

-

-

-

0.000070

-0.000070

-0.000070

-0.000070

2028

7.9

5.2

5.0

4.5

-1.1

14

14

13

2029

8.0

5.3

5.2

4.6

-1.1

14

14

14

2030

11

3.6

3.4

3.0

-0.046

15

14

14

2031

11

3.6

3.5

3.1

-0.046

15

15

14

2032

29

14

14

12

1.2

42

42

40

2033

29

15

14

13

1.2

43

42

41

2034

30

15

15

13

1.2

44

43

42

2035

30

15

15

13

1.2

44

44

42

2036

31

16

15

13

1.2

45

45

43

2037

31

16

15

14

1.2

46

45

44

2038

14

0.0026

0.0025

0.0013

0.64

13

13

13

2039

14

0.0016

0.0016

0.00044

0.64

13

13

13

2040

14

0.00043

0.00045

-0.00065

0.64

13

13

13

2041

14

-0.0011

-0.00097

-0.0019

0.64

14

14

14

2042

11

8.0

7.7

6.9

3.3

16

15

15

2043

11

8.1

7.8

6.9

3.3

16

16

15

2044

11

8.1

7.9

7.0

3.3

16

16

15

2045

11

8.2

8.0

7.1

3.3

16

16

15

2046

12

8.3

8.1

7.2

3.3

17

16

15

2047

12

8.4

8.1

7.2

3.3

17

17

16



Climate
Benefitsb

PM2.5 and 03-related
Health Benefits0

Compliance
Costs'1



Net
Benefits0



Discount Rate



2%

2%

3%

7%

2%

3%

7%

2%

3%

7%

PV

270

120

110

60

19

16

7.8

370

360

320

EAV

14

6.5

6.2

5.2

0.99

0.93

0.68

20

19

19

Non-Monetized Benefits'

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Annual values from 2024 to 2047 are not discounted. PV and EAV values discounted to 2024. Values have been
rounded to two significant figures. Rows may not appear to add correctly due to rounding.
b Monetized climate benefits are based on reductions in CO2 emissions and are calculated using three different
estimates of the social cost of CO2 (SC-CO2) (under 1.5 percent, 2.0 percent, and 2.5 percent near-term Ramsey
discount rates). For the presentational purposes of this table, we show the climate benefits associated with the SC-
CO2 at the 2 percent near-term Ramsey discount rate. Please see Table 4-5 for the full range of monetized climate
benefit estimates. See Section 4.2 for a discussion of the uncertainties associated with the climate benefit estimates.

7-11


-------
c The health benefits estimates use the larger of the two benefits estimates presented in Table 4-15 through Table 4-
19. Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates.
d The discount rate in IPM is 3.76 percent, as described in Section 3.

e In this net benefits analysis, health benefits and costs are discounted at the rates shown in the table (i.e., two
percent, three percent, and seven percent). Climate benefits are discounted using a two percent discount rate only in
this net benefits analysis.

f Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

7-12


-------
Table 7-8 Net Benefits of the Alternative 2 Illustrative Scenario for 2024 to 2047
(billion 2019 dollars) a	



Climate
Benefitsb

PM2.5 and 03-related
Health Benefits0

Compliance

Pneted



Net
Benefits0





2%

2%

3%

7%

O IS US

2%

3%

7%

2024

-

-

-

-

0.011

-0.011

-0.011

-0.011

2025

-

-

-

-

0.011

-0.011

-0.011

-0.011

2026

-

-

-

-

0.011

-0.011

-0.011

-0.011

2027

-

-

-

-

0.000070

-0.000070

-0.000070

-0.000070

2028

7.1

4.8

4.7

4.1

-1.1

13

13

12

2029

7.2

4.9

4.8

4.2

-1.1

13

13

12

2030

6.2

2.8

2.7

2.4

-0.72

9.7

9.6

9.3

2031

6.3

2.8

2.8

2.4

-0.72

9.9

9.8

9.5

2032

28

14

14

12

1.2

41

41

40

2033

29

15

14

13

1.2

42

42

40

2034

29

15

14

13

1.2

43

43

41

2035

30

15

15

13

1.2

44

43

42

2036

30

15

15

13

1.2

45

44

42

2037

31

16

15

14

1.2

45

45

43

2038

14

0.089

0.087

0.081

0.60

13

13

13

2039

14

0.088

0.086

0.080

0.60

13

13

13

2040

14

0.087

0.085

0.079

0.60

13

13

13

2041

14

0.085

0.083

0.077

0.60

14

14

14

2042

11

7.9

7.7

6.8

3.6

15

15

14

2043

11

8.0

7.8

6.9

3.6

16

15

14

2044

11

8.1

7.9

7.0

3.6

16

16

15

2045

11

8.2

8.0

7.1

3.6

16

16

15

2046

12

8.3

8.0

7.1

3.6

16

16

15

2047

12

8.4

8.1

7.2

3.6

17

16

15



Climate
Benefitsb

PM2.5 and 03-related
Health Benefits0

Compliance
Costs'1



Net
Benefits0













Discount Rate









2%

2%

3%

7%

2% 3% 7%

2%

3%

7%

PV

250

120

100

58

19 15 7.2

360

340

310

EAV

13

6.3

6.1

5.1

0.98 0.91 0.63

19

19

18

Non-Monetized Benefits'

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Annual values from 2024 to 2047 are not discounted. PV and EAV values discounted to 2024. Values have been
rounded to two significant figures. Rows may not appear to add correctly due to rounding.
b Monetized climate benefits are based on reductions in CO2 emissions and are calculated using three different
estimates of the social cost of CO2 (SC-CO2) (under 1.5 percent, 2.0 percent, and 2.5 percent near-term Ramsey
discount rates). For the presentational purposes of this table, we show the climate benefits associated with the SC-
CO2 at the 2 percent near-term Ramsey discount rate. Please see Table 4-3 for the full range of monetized climate
benefit estimates. See Section 4.2 for a discussion of the uncertainties associated with the climate benefit estimates.

7-13


-------
c The health benefits estimates use the larger of the two benefits estimates presented in Table 4-15 through Table 4-
19. Monetized benefits include those related to public health associated with reductions in PM2 5 and ozone
concentrations. The health benefits are associated with several point estimates.
d The discount rate in IPM is 3.76 percent, as described in Section 3.

e In this net benefits analysis, health benefits and costs are discounted at the rates shown in the table (i.e., two
percent, three percent, and seven percent). Climate benefits are discounted using a two percent discount rate only in
this net benefits analysis.

f Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

7-14


-------
APPENDIX A: CLIMATE BENEFITS
A.l Climate Benefits Estimated using the Interim SC-CO2 values used in the Proposal

This appendix presents the climate benefits of the final standards using the interim SC-
CO2 values used in the proposal of these rulemakings. The interim SC-CO2 values are presented
in Table A-l and the climate benefits using these values are presented in Table A-2.

Table A-l Interim SC-CO2 Values, 2028 to 2047 (2019 dollars per metric ton)

Discount Rate and Statistic

Emissions Year

5% Average

3% Average

2.5% Average

3% 95th Percentile

2028

$18

$59

$86

$178

2029

$19

$60

$87

$181

2030

$19

$61

$88

$184

2031

$20

$62

$90

$188

2032

$20

$63

$91

$192

2033

$21

$64

$92

$196

2034

$21

$66

$94

$200

2035

$22

$67

$95

$203

2036

$23

$68

$96

$207

2037

$23

$69

$98

$211

2038

$24

$70

$99

$215

2039

$24

$71

$101

$218

2040

$25

$72

$102

$222

2041

$26

$73

$103

$226

2042

$26

$75

$105

$229

2043

$27

$77

$107

$235

2044

$28

$78

$108

$239

2045

$28

$29

$110

$242

2046

$29

$80

$111

$246

2047

$30

$81

$112

$249

Note: The 2028 to 2047 SC-CO2 values are identical to those reported in the February 2021 SC-GHG TSD (IWG,
2021) adjusted to 2019 dollars using the annual GDP Implicit Price Deflator values in the U. S. Bureau of Economic
Analysis' (BEA) NIPA Table 1.1.9 (U.S. BEA, 2022). This table displays the values rounded to the nearest dollar;
the annual unrounded values used in the calculations in this analysis are available on OMB's website:
https://www.whitehouse.gOv/omb/information-regulatorv-affairs/regulatorv-matters/#scghgs.

A-l


-------
Table A-2 Stream of Projected Climate Benefits using Interim SC-CO2 values under the

Final Rules from 2028 to 2047 (millions of 2019 dollars, discounted to 2024)

SC-CO2 Discount Rate and Statistic



5%

3%

2.50%

3%

Emissions Year

Average

Average

Average

95th Percentile

2028

560

2,000

3,000

6,000

2029

570

2.000

2.900

5.900

2030

710

2.500

3.800

7,700

203 1

710

2.500

3.800

7.600

2032

1.700

6.100

9.200

19.000

2033

1.700

6.100

9.100

19.000

2034

1.600

6.100

9.100

18.000

2035

1.600

6,000

8.900

18.000

2036

1.600

5.900

8.800

18.000

2037

1.500

5.800

8.800

18.000

2038

650

2.500

3.800

7.600

2039

620

2.400

3,700

7,500

2040

610

2.400

3,700

7.400

2041

610

2.400

3.600

7,300

2042

450

1.800

2.800

5.600

2043

450

1.800

2.800

5.600

2044

440

1.800

2.800

5.600

2045

420

650

2.700

5.500

2046

420

1.800

2,700

5.400

2047

410

1,700

2,700

5,300

PV

17,000

64,000

99,000

200,000

EAV

1,200

3,800

5,500

12,000

Note: Climate benefits are based on reductions in CO2 emissions and are calculated using the IWG interim SC-CO2

estimates from IWG (2021).

A.2 References

IWG. (2021). Technical Support Document: Social Cost of Carbon, Methane, and Nitrous Oxide
Interim Estimates under Executive Order 13990. Washington DC: U.S. Government,
Interagency Working Group (IWG) on Social Cost of Greenhouse Gases.
https://www.whitehouse.gov/wp-

content/uploads/2021/02/TechnicalSupportDocument SocialCostofCarbonMethaneNitro
usOxide.pdf? source=email

U.S. BEA. (2022). Table 1.1.9. Implicit Price Deflators for Gross Domestic Product.
Washington, DC.

https://apps.bea. gov/iTable/?reqid=19&step=3&isuri=l&1921=survev& 1903=13

A-2


-------
APPENDIX B: AIR QUALITY MODELING

As noted in Section 4, EPA used photochemical modeling to create air quality surfaces198
that were then used in air pollution health benefits calculations of the three illustrative scenarios:
the final rules, the alternative 1, and the alternative 2 scenarios. The modeling-based surfaces
captured air pollution impacts resulting from changes in NOx, SO2 and direct PM2.5 emissions
from EGUs. This appendix describes the source apportionment modeling and associated methods
used to create air quality surfaces for the baseline scenario and three illustrative scenarios in four
snapshot years: 2028, 2030, 2035, 2040 and 2045. EPA created air quality surfaces for the
following pollutants and metrics: annual average PM2.5; April-September average of 8-hr daily
maximum (MDA8) ozone (AS-M03).

New ozone and PM source apportionment modeling outputs were created to support
analyses in the RIAs for multiple final EGU rulemaking efforts. The basic methodology for
determining air quality changes is the same as that used in the RIAs from multiple previous rules
(U.S. EPA, 2019, 2020a, 2020b, 2021b, 2022a). EPA calculated EGU emissions estimates of
NOx and SO2 for baseline and illustrative scenarios in all five snapshot years using the Integrated
Planning Model (IPM) (Section 3 of this RIA). EPA also used IPM outputs to estimate EGU
emissions of PM2.5 based on emission factors described in U.S. EPA (2021a).199 This appendix
provides additional details on the source apportionment modeling simulations and the associated
analysis used to create ozone and PM2.5 air quality surfaces.

B.l Air Quality Modeling Simulations

The air quality modeling utilized a 2016-based modeling platform which included
meteorology and base year emissions from 2016 and projected future-year emissions for 2026
for all sectors other than EGUs and 2030 for EGUs. The air quality modeling included
photochemical model simulations for a 2016 base year and a future year representing the
combined 2026/2030 emissions described above to provide hourly concentrations of ozone and
PM2.5 component species nationwide. In addition, source apportionment modeling was

198	The term "air quality surfaces" refers to continuous gridded spatial fields using a 12 km grid resolution.

199	For details, please see Flat File Generation Methodology and Post Processing Emissions Factors PM CO VOC
NH3 Updated Summer 2021 Reference Case, available at: https://www.epa.gov/power-sector-
modeling/supporting-documentation-2015 -ozone-naaqs-actions

B-l


-------
performed for the future year to quantify the contributions to ozone from NOx emissions and to
PM2.5 from NOx, SO2 and directly emitted PM2.5 emissions from EGUs on a state-by-state and
fuel-type basis. As described below, the modeling results for 2016 and the future year, in
conjunction with EGU emissions data for the baseline and three illustrative scenarios in 2028,
2030, 2035, 2040 and 2045 were used to construct the air quality surfaces that reflect the
influence of emissions changes between the baseline and the three illustrative scenarios in each
year.

The air quality model simulations (i.e., model runs) were performed using the
Comprehensive Air Quality Model with Extensions (CAMx) version 7.10200 (Ramboll Environ,
2021). The nationwide modeling domain (i.e., the geographic area included in the modeling)
covers all lower 48 states plus adjacent portions of Canada and Mexico using a horizontal grid
resolution of 12 x 12 km is shown in Figure B-l. CAMx requires a variety of input files that
contain information pertaining to the modeling domain and simulation period. These include
gridded, hourly emissions estimates and meteorological data, and initial and boundary
concentrations. The meteorological data and the initial and boundary concentrations were
identical to those described in U.S. EPA (2023a). Separate emissions inventories were prepared
for the 2016 base year and the projected future year. All other inputs (i.e., meteorological fields,
initial concentrations, ozone column, photolysis rates, and boundary concentrations) were
specified for the 2016 base year model application and remained unchanged for the projection-
year model simulation.

2016 base year emissions are described in detail in U.S. EPA (2023b). The types of
sources included in the emission inventory include stationary point sources such as EGUs and
non-EGUs; non-point emissions sources including those from oil and gas production and
distribution, agriculture, residential wood combustion, fugitive dust, and residential and
commercial heating and cooking; mobile source emissions from onroad and nonroad vehicles,
aircraft, commercial marine vessels, and locomotives; wild, prescribed, and agricultural fires;
and biogenic emissions from vegetation and soils. Future year emissions from all sources other
than EGUs were based on the 2026 emissions projections described in U.S. EPA (2023b). The

200 This CAMx simulation set the Rscale NH3 dry deposition parameter to 0 which resulted in more realistic model
predictions of PM2.5 nitrate concentrations than using a default Rscale parameter of 1.

B-2


-------
Post-IRA 2022 Reference Case of the EPA's Power Sector Platform v6 using Integrated
Planning Model (IPM), which includes the Final GNP, was also reflected.201 The EGU projected
inventory represents demand growth, fuel resource availability, generating technology cost and
performance, and other economic factors affecting power sector behavior. It also reflects
environmental rules and regulations, consent decrees and settlements, plant closures, and newly
built units for the calendar year 2030. In this analysis, the projected EGU emissions include
provisions of tax incentives impacting electricity supply in the Inflation Reduction Act of 2022
(IRA), Final GNP, 2021 Revised Cross-State Air Pollution Rule Update (RCU), the 2016
Standards of Performance for Greenhouse Gas Emissions from New, Modified, and
Reconstructed Stationary Sources, the Mercury and Air Toxics Rule (MATS) finalized in 2011,
and other finalized rules. Documentation and results of the Post-IRA 2022 Reference Case,
where the Final GNP was also included for EGUs, are available at (https://www.epa.gov/power-
sector-modeling/final-pm-naaqs).

Model predictions of ozone and PM2.5 concentrations were compared against ambient
measurements (U.S. EPA, 2023a, 2024). Ozone and PM2.5 model evaluations showed model
performance that was adequate for applying these model simulations for the purpose of creating
air quality surfaces to estimate ozone and PM2.5 benefits.

Hf

*•. >. 7





/ 1

/





J

Jfl







zC*

{sm\







V f> A

4 ) y\ |

> —fc—





13US3 OcmMi <

• ;<>:»> >
-------
The contributions to ozone and PM2.5 component species (e.g., sulfate, nitrate,
ammonium, elemental carbon (EC), organic aerosol (OA), and crustal material202) from EGU
emissions in individual states and from each EGU-fuel type were modeled using the "source
apportionment" tool approach. In general, source apportionment modeling quantifies the air
quality concentrations formed from individual, user-defined groups of emissions sources or
"tags". These source tags are tracked through the transport, dispersion, chemical transformation,
and deposition processes within the model to obtain hourly gridded203 contributions from the
emissions in each individual tag to hourly gridded modeled concentrations. For this RIA we used
the source apportionment contribution data to provide a means to estimate of the effect of
changes in emissions from each group of emissions sources (i.e., each tag) to changes in ozone
and PM2.5 concentrations. Specifically, we applied outputs from source apportionment modeling
for ozone and PM2.5 component species using the future year modeled case to obtain the
contributions from EGUs emissions in each state and fuel-type to ozone and PM2.5 component
species concentrations in each 12 km model grid resolution nationwide. Ozone contributions
were modeled using the Anthropogenic Precursor Culpability Assessment (APCA) tool and
PM2.5 contributions were modeled using the Particulate Matter Source Apportionment
Technology (PSAT) tool (Ramboll Environ, 2021). The ozone source apportionment modeling
was performed for the period April through September to provide data for developing spatial
fields for the April through September maximum daily eight hour (MDA8) (i.e., AS-M03)
average ozone concentration exposure metric. The PM2.5 source apportionment modeling was
performed for a full year to provide data for developing annual average PM2.5 spatial fields.

Table B-l, Table B-2, and Table B-3 provide emissions that were tracked for each source
apportionment tag.

Table B-l Future-year Emissions Allocated to Each Modeled Coal EGU State Source

Apportionment Tag	

Ozone Season Annual NOx Annual SO2 Annual PM2.5
	^	NOx (tons)	(tons)	(tons)	(tons)	

AL	2,537	5,046	1,929	700

AR4	NA	304	331	51

202	Crustal material refers to elements that are commonly found in the earth's crust such as Aluminum, Calcium,
Iron, Magnesium, Manganese, Potassium, Silicon, Titanium, and the associated oxygen atoms.

203	Hourly contribution information is provided for each grid cell to provide spatial patterns of the contributions from
each tag

B-4


-------
AZ

1,005

2,536

4,515

609

CA

222

511

99

27

CO

19

269

287

21

CT

0

0

0

0

DC

0

0

0

0

DE

0

0

0

0

FL

1,110

1,401

7,163

277

GA

1,654

2,534

3,247

159

IA

8,354

18,776

9,656

1,203

ID

0

0

0

0

IL

1,639

3,742

6,773

270

IN

4,886

18,146

26,584

2,252

KS1

NA

214

121

NA

KY

3,551

7,333

7,127

560

LA2-4

NA

47

NA

NA

MA

0

0

0

0

MD3

NA

139

272

31

MD + PA3

708

NA

NA

NA

ME

0

0

0

0

MI

1,532

4,071

12,478

380

MN

724

1,549

3,289

94

MO

2,947

23,480

38,989

853

MS4

NA

252

507

23

MT

3,771

8,842

4,056

1,252

NC

266

482

634

35

ND

8,583

19,562

25,398

1,923

NE1

7,817

17,507

43,858

NA

NE + KS1

NA

NA

NA

374

NH

0

0

0

0

NJ

0

0

0

0

NM

1,442

2,757

6,800

1,739

NV

0

1

1

0

NY

0

0

0

0

OH

3,152

10,485

21,721

901

OK4

NA

212

152

21

OR

0

0

0

0

PA3

NA

1,530

4,932

167

RI

0

0

0

0

SC

807

1,939

3,429

364

SD

418

1,100

1,022

27

TN

259

259

269

32

TX2-4

NA

7,031

NA

NA

TX + LA2

NA

NA

11,607

1,578

TX-reg4

2,698

NA

NA

NA


-------
UT	2,702	4,236	7,625	232

VA	466	1,124	259	445
VT 0 0 0 0
WA 0 0 0 0

WI	866	2,137	838	90

WV	6,824	16,358	17,631	1,753

WY	6,066	13,222	11,754	1,024

'KS and NE emissions grouped into multi-state tag for direct PM2 5
2LA and TX emissions grouped into multi-state tag for SO2 and direct PM2.5
3MD and PA emissions grouped into multi-state tag for ozone season NOx

4AR, KS, LA, MS, OK and TX emissions grouped into multi-state tag ("TX-reg") for ozone season NOx

Table B-2 Future-year Emissions Allocated to Each Modeled Natural Gas EGU State
Source Apportionment Tag	

State

Ozone Season NOx
(tons)

Annual NOx
(tons)

Annual SO2
(tons)

Annual PM2.5
(tons)

AL

2,833

5,132

0

1,979

AR

1,651

2,957

0

632

AZ

1,759

3,146

0

686

CA

1,960

5,773

0

1,964

CO

957

1,825

0

461

CT

461

778

0

160

DC

6

11

0

7

DE

383

502

0

134

FL

7,550

14,372

0

4,996

GA

2,279

4,182

0

1,740

IA

875

1,106

0

327

ID

336

513

0

185

IL

1,624

2,705

0

825

IN

1,180

2,166

0

955

KS

329

621

0

54

KY

980

2,806

0

699

LA

3,771

8,706

0

2,158

MA

482

725

0

244

MD

402

710

0

435

ME

232

273

0

21

MI

6,523

11,372

0

1,508

MN

661

928

0

87

MO

587

875

0

342

MS

1,926

3,860

0

1,140

MT

11

19

0

7

NC

1,803

3,426

0

1,213

ND

25

41

0

3

NE

13

47

0

4

NH

120

136

0

34

B-6


-------
NJ

1,024

1,910

0

608

NM

733

1,128

0

131

NV

1,693

2,471

0

648

NY

2,793

5,125

0

1,270

OH

1,838

3,824

0

1,617

OK

1,558

2,448

0

546

OR

5

188

0

87

PA

6,811

12,386

0

3,280

RI

115

153

0

73

SC

1,092

2,090

0

917

SD

93

105

0

11

TN

464

1,107

0

388

TX

7,652

14,715

0

3,567

UT

1,189

1,779

0

514

VA

1,836

3,409

0

1,087

VT

4

8

0

6

WA

485

1,311

0

464

WI

847

1,447

0

369

WV

109

180

0

50

WY

203

206

0

28

Table B-3

Future-year Emissions Allocated to the Modeled Other EGU Source

Apportionment Tag







4-/\

Ozone Season NOx

Annual NOx

Annual SO2

Annual PM2.5

dlalc

(tons)

(tons)

(tons)

(tons)

US3

20,611

48,619

9,631

7,915

aOnly includes US emissions from the contiguous 48 states

Examples of the magnitude and spatial extent of ozone and PM2.5 contributions are
provided in Figure B-2 through Figure B-5 for EGUs in California, Georgia, Iowa, and Ohio.
These figures show how the magnitude and the spatial patterns of contributions of EGU
emissions to ozone and PM2.5 component species depend on multiple factors including the
magnitude and location of emissions as well as the atmospheric conditions that influence the
formation and transport of these pollutants. For instance, NOx emissions are a precursor to both
ozone and PM2.5 nitrate. However, ozone and nitrate form under very different types of
atmospheric conditions, with ozone formation occurring in locations with ample sunlight and
ambient VOC concentrations while nitrate formation requires colder and drier conditions and the
presence of gas-phase ammonia. California's complex terrain that tends to trap air and allow
pollutant build-up combined with warm sunny summer and cooler dry winters and sources of

B-7


-------
both ammonia and VOCs make its atmosphere conducive to formation of both ozone and nitrate.
While the magnitude of EGU NOx emissions from gas plus coal EGUs is substantially larger in
Iowa than in California (Table B-l and Table B-2) the emissions from California lead to larger
maximum contributions to the formation of those pollutants due to the conducive conditions in
that state. Georgia and Ohio both had substantial NOx emissions. While maximum ozone
impacts shown for Georgia and Ohio EGUs are similar order of magnitude to maximum ozone
impacts from California EGUs, nitrate impacts are negligible in both Georgia and Ohio due to
less conducive atmospheric conditions for nitrate formation in those locations. California EGU
SO2 emissions in the future year source apportionment modeling are several orders of magnitude
smaller than SO2 emissions in Ohio and Georgia (Table B-l) leading to much smaller sulfate
contributions from California EGUs than from Ohio and Georgia EGUs. PM2.5 organic aerosol
EGU contributions in this modeling come from primary PM2.5 emissions rather than secondary
atmospheric formation. Consequently, the impacts of EGU emissions on this pollutant tend to
occur closer to the EGU sources than impacts of secondary pollutants (ozone, nitrate, and
sulfate) which have spatial patterns showing a broader regional impact. These patterns
demonstrate how the model captures important atmospheric processes which impact pollutant
formation and transport from emissions sources. Finally, Figure B-6 and Figure B-7 show EGU
ozone and PM2.5 contributions from all contiguous U.S. EGUs split out by fuel type. The spatial
differences between coal EGU, natural gas EGU, and other EGU contributions reflect the
varying location and magnitude of emissions from each type of EGU.

B-8


-------
n

a)Apr-Sep MDA8 03

) I

^ rxiA

yfl j Ml

f

a

f

MHY

V/

l\

\\
\

\j

1 \

c) Annual PM2.5 sulfate

1 • >

f W L

' /

< \LX

/
/

lW

w

>'-JZ

\j

b) Annual PM2.5 nitrate

f,'

7 \



/'

*



V

U*	Hi

cf) Annual PM2i5 OA

wl

W
\\

•JS



t

j

rA^I

£

/

/ v\

u V

Figure B-2 Maps of California EGU Tag contributions to a) April-September Seasonal
Average MDA8 Ozone (ppb); b) Annual Average PM2.5 Nitrate (jig/m3); c) Annual Average
PM2.5 sulfate (jLig/m3); d) Annual Average PM2.5 Organic Aerosol (ng/mJ)

B-9


-------
a) Apr-Sep MDA8 03

\A

f

rW

c) Annual PM2.5 sulfate

hi ,' V

r \ i 1

ET?¥)/^

ESSf



<=1 —f

f 	rtX J

L /Tw\ /

) V^-V (\

Vr ^

b) Annual PM2.5 nitrate

^ dA

In £&\

4LjL/T\iF

vr

(ML Wu * '

d) Annual PM2 5 OA

my

LLm
W

•w*i

V
\

v

J»	3»

0 oetwi al(UL w»' » <«'* •« l»+.e?

r^#ss

] *f<

i ;«

LH7I

Lj—Zl f

In

P
;

r

vT

\ 1

y

as	s»

kin • ii ue>o k P.1X Mm- a aaa «t

~0 at |1,U - a 099 «l C3I

Figure B-3 Maps of Georgia EGU Tag contributions to a) April-September Seasonal
Average MDA8 Ozone (ppb); b) Annual Average PM2.5 Nitrate (jig/m3); c) Annual Average
PM2.5 sulfate (fig/ni'); d) Annual Average PM2.5 Organic Aerosol (jig/m3)

B-10


-------


a) Apr-Sep MDA8 03

ion je

b) Annual PM2 5 nitrate

010



11

f —rX7\A

k Y 7 R	f

\3 PL \ rrfxY

M»

19

A/Ttt^

/ / { r^^k. rV\f\
I f w—	1 f Y/T] { UY

r->~J 1 1 A A Jf h r fa*

k YTti—

»\ / | I 1 1 nT\ y

00*

CUM
007
0M
0«



1X1 *»

00«

V

%

i o« a

DM

\)

000



« as m a» an i

Mm • a acr«o atcui «•< • m:m isaa.isTi

n ai w w w

Mln« a OM.O ldti.il, Mu • 0 03®1 |IOe.l4«.i

jo

c) Annual PM2.5 sulfate

#10 f*

d) Annual PM2 5 OA

¦»



k it"-rv\

on

/UrS-r ^ 3

/ m 	 jfefe frfm

I f U—	i { iBT1, ( \U

f S H h r<&

{vSv&i

0»

LM

MntwSf

\ \		1	 \J~

o« m

00?

006 l*>

007
009
oot

»

AlTnrfcf/

HJ U/]Ty

oot n

eat

hTM/W

DM

»

jirMk

\r >41

1 DO SO

v~\ / "i \

V V

¦ 0«



» W W M W
Min • a OM«Q M I1.1L Mw • OOlOtt I30«.«r)

as m m us »?

Mm • o oc*.a « (l.l|. MM • o 0 JS *(JJ7,»1)

Figure B-4 Maps of Iowa EGU Tag contributions to a) April-September Seasonal
Average MDA8 Ozone (ppb); b) Annual Average PM2.5 Nitrate (jig/nr'); c) Annual Average
PM2.5 sulfate (jtig/ni3); d) Annual Average PM2.5 Organic Aerosol (jig/m3)

B-l'l


-------
»«?

a) Apr-Sep MDA8 03

-

b) Annual PM2 5 nitrate



•«



k

J / i 	1

rJf\

'*> KM

P ^ T

h\| J 7^* Af





»

I(TV-tCnT>?



t« to





00.



\ \ J^jL 1

*

J

IV

\ \LMJ





-

\ > 1 n i *

..

\ i 1 Tn——?



00,

V

HAj

\ f V

VJ ^

j

1 am „

^ mi

V \)



001

s

a n at at

Mm- ao«.a siclU • 2 tw

» 1

91 Sffl 3U W





c) Annual PM2,5 sulfate

010 2*

d) Annual PM2 5 OA





Or	I	

rfyj-Tr^

I \ L i \ Lfli





)U\ ¦ ]	,

/ / v,		 /W

1 1 J) lf \ -J

tTrv-turnr^

l\ [ fl—

\ \J 1— 1 1

\ ) 7 n ru^^v

H I HirVY



Mr

»

Ja
>

001 1M

007

0« l*>

D«

00* »





M



/	

• 0.02 S3

ooi

— 000

\n







#" u* -M m

Mtn ¦ 12.1421

Figure B-5 Maps of Ohio EGU Tag contributions to a) April-September Seasonal
Average MDA8 Ozone (ppb); b) Annual Average PM2.5 Nitrate (jig/m3); c) Annual Average
PM2.5 sulfate (jiig/1113); d) Annual Average PM2.5 Organic Aerosol (jig/m3)

B-12


-------
a) Apr-Sep MDA8 Ozone contributions
from US Coal EGUs

:

I _ l

270

b) Apr-Sep MDA8 Ozone contributions
from US Natural Gas EGUs



c) Apr-Sep MDA8 Ozone contributions
from US Other EGUs

/Prr



inr—



-



i\Jxi 1



F \ / /



''1 % - 1 1 * >

Xd-J TV

"



HMj

	VI	





¦""v. m\

\f

060

030

Figure B-6 Maps of National EGU Tag contributions to April-September Seasonal
Average MDA8 ozone (ppb) by fuel for a) Coal EGUs; b) Natural Gas EGUs; c) All Other
EGUs

a) Annual PM2 5 contributions from US
Coal EGUs

b) Annual PM2.5 contributions from US
Natural Gas EGUs



—	I	1

^ \ / r r













VI V 1

c) Annual PM2.5 contributions from US
Other EGUs

[Pr



/

V \ 1 T



'1 i

%J ]

n ?tt

V



V\

J
'v\

y

0.4S

0.40

0.3S

0.30

0.25

0.20

015

0.10

005

Figure B-7 Maps of National EGU Tag contributions to Annual Average PM2.5 (fig/m3)
by fuel for a) Coal EGUs; b) Natural Gas EGUs; c) All Other EGUs

B.2 Applying Modeling Outputs to Create Spatial Fields

In this section we describe the method for creating spatial fields of AS-M03 and annual
average PM2.5 based on the 2016 and future year modeling. The foundational data include (1)
ozone and speciated PM2.5 concentrations in each model grid cell from the 2016 and the future

B-13


-------
year modeling, (2) ozone and speciated PM2.5 contributions in the future year of EGUs emissions
from each state in each model grid cell204, (3) future year emissions from EGUs that were input
to the contribution modeling (Table B-l, Table B-2 and Table B-3), and (4) the EGU emissions
from IPM for baseline and the three illustrative scenarios in each snapshot year. The method to
create spatial fields applies scaling factors to gridded source apportionment contributions based
on emissions changes between future year projections and the baseline and the three illustrative
scenarios to the modeled contributions. This method is described in detail below.

Spatial fields of ozone and PM2.5 in the future year were created based on "fusing"
modeled data with measured concentrations at air quality monitoring locations. To create the
spatial fields for each future emissions scenario, the fused future year model fields are used in
combination with the EGU source apportionment modeling and the EGU emissions for each
scenario and snapshot year. Contributions from each state and fuel EGU contribution "tag" were
scaled based on the ratio of emissions in the year/scenario being evaluated to the emissions in the
modeled future year scenario. Contributions from tags representing sources other than EGUs are
held constant at 2026 levels for each of the scenarios and years. For each scenario and year
analyzed, the scaled contributions from all sources were summed together to create a gridded
surface of total modeled ozone and PM2.5. The process is described in a step-by-step manner
below starting with the methodology for creating AS-M03 spatial fields followed by a
description of the steps for creating annual PM2.5 spatial fields.

Ozone:

1. Create fused spatial fields of future year AS-M03 incorporating information from the air
quality modeling and from ambient measured monitoring data. The enhanced Voronoi
Neighbor Average (eVNA) technique (Ding et al., 2016; Gold et al., 1997; U.S. EPA, 2007)
was applied to ozone model predictions in conjunction with measured data to create
modeled/measured fused surfaces that leverage measured concentrations at air quality
monitor locations and model predictions at locations with no monitoring data.

1.1. The AS-M03 eVNA spatial fields are created for the 2016 base year with EPA's
software package, Software for the Modeled Attainment Test - Community Edition

204 Contributions from EGUs were modeled using projected emissions for the future year modeled scenario. The
resulting contributions were used to construct spatial fields in 2028, 2030, 2035, 2040 and 2045.

B-14


-------
(SMAT-CE)205 (U.S. EPA, 2022b) using 3 years of monitoring data (2015-2017) and the
2016 modeled data.

1.2.	The model-predicted spatial fields (i.e., not the eVNA fields) of AS-M03 in 2016 were
paired with the corresponding model-predicted spatial fields in future year to calculate
the ratio of AS-M03 between 2016 and the future year in each model grid cell.

1.3.	To create a gridded future year eVNA surfaces, the spatial fields of 2016/future year
ratios created in step (1.2) were multiplied by the corresponding eVNA spatial fields for
2016 created in step (1.1) to produce an eVNA AS-M03 spatial field for the future year
using (Eq-1).

.... .	( .... . \ Modelg future	Eq-1

eVNAgJuture - (eVNAgi2016) x Model^^

•	eVNAgjuture is the eVNA concentration of AS-M03 or PM2.5 component species in
grid-cell, g, in the future year

•	eVNAg 2016 is the eVNA concentration of AS-M03 or PM2.5 component species in grid-
cell, g, in 2016

•	Modelgjuture is the CAMx modeled concentration of AS-M03 or PM2.5 component
species in grid-cell, g, in the future year

•	Modelg 2016 is the CAMx modeled concentration of AS-M03 or PM2.5 component in
grid-cell, g, in 2016

2. Create gridded spatial fields of total EGU AS-M03 contributions for each combination of

scenario and analysis year evaluated.

2.1. Use the EGU ozone season NOx emissions for the 2028 baseline and the corresponding
future year modeled EGU ozone season emissions (Table B-l, Table B-2, and Table
B-3) to calculate the ratio of 2028 baseline emissions to future year modeled emissions
for each EGU tag (i.e., an ozone scaling factor calculated for each state-fuel

205 SMAT-CE available for download at https://www.epa.gov/scram/photochemical-modeling-tools.

B-15


-------
combination)206. These scaling factors are provided in Table B-4, Table B-5, Table B-6,
Table B-7, Table B-18 and Table B-19.

2.2.	Calculate adjusted gridded AS-M03 EGU contributions that reflect differences in state-
fuel EGU NOx emissions between the modeled future year and the 2028 baseline by
multiplying the ozone season NOx scaling factors by the corresponding gridded AS-
M03 ozone contributions207 from each state-fuel EGU tag.

2.3.	Add together the adjusted AS-M03 contributions for each state-fuel EGU tag to produce
spatial fields of adjusted EGU totals for the 2028 baseline.208

2.4.	Repeat steps 2.1 through 2.3 for the three 2028 illustrative scenarios and for the baseline
and illustrative scenarios for each additional snapshot year. All scaling factors for the
baseline scenario and the three illustrative scenarios are provided in Table B-4, Table
B-5, Table B-6, Table B-7, Table B-18 and Table B-19.

3. Create a gridded spatial field of AS-M03 associated with IPM emissions for the 2028
baseline by combining the EGU AS-M03 contributions from step (2.3) with the
corresponding contributions to AS-M03 from all other sources. Repeat for each of the EGU
contributions created in step (2.4) to create separate gridded spatial fields for the baseline and
three illustrative scenarios for each snapshot year.

Steps 2 and 3 in combination can be represented by equation 2:

206	State-level tags were tracked for separately for coal EGUs and for natural gas EGUs. All other EGU emissions
were tracked using a single national tag. In addition, preliminary testing of this methodology showed unstable
results when very small magnitudes of emissions were tagged especially when being scaled by large factors. To
mitigate this issue, in cases where state-fuel EGU tags were associated with no or very small emissions, tags
were combined into multi-state regions.

207	The source apportionment modeling provided separate ozone contributions for ozone formed in VOC-limited
chemical regimes (03 V) and ozone formed in NOx-limited chemical regimes (03N). The emissions scaling
factors are multiplied by the corresponding 03N gridded contributions to MDA8 concentrations. Since there are
no predicted changes in VOC emissions in the control scenarios, the 03 V contributions remain unchanged.

208	The contributions from the unaltered 03 V tags are added to the summed adjusted 03N EGU tags.

B-16


-------
AS-M03g iy = eVNAgjfuture

(Cg.BC ^H,int ^H.bio fjres CgijSanthro

c+c+c+c — + r	

^g,Tot ^g,Tot ^g,Tot ^g.Tot ^g.Tot	Eq-2

T	T

Z^EGUVOC,g,t V"1 CeGUNOx,g,t SNOx,t,i,y

		-Y

t=1 g.Tot	frf	g.Tot

AS-M03g i y is the estimated fused model-obs AS-M03 for grid-cell, "g", scenario, ""i"2"9'. and

year, "y"210;

eVNAg juture is the future year eVNA future year AS-M03 concentration for grid-cell "g"
calculated using Eq-1.

Cgjot is the total modeled AS-M03 for grid-cell "g" from all sources in the future year source
apportionment modeling

Cg BC is the future year AS-M03 modeled contribution from the modeled boundary inflow;

Cg int is the future year AS-M03 modeled contribution from international emissions within the
modeling domain;

Cg,bio is the future year AS-M03 modeled contribution from biogenic emissions;

Cg,fires is the future year AS-M03 modeled contribution from fires;

Cg.usanthro is the total future year AS-M03 modeled contribution from U.S. anthropogenic
sources other than EGUs;

CEGuvoc,g,t is the future year AS-M03 modeled contribution from EGU emissions of VOCs from
state, "t";

CEGUNOx,g,t is the future year AS-M03 modeled contribution from EGU emissions ofNOx from
tag, "t"; and

SNOx,t,i,y is the EGU NOx scaling factor for tag, "t", scenario "i", and year, "y".

PM2.5

4. Create fused spatial fields of future year annual PM2.5 component species incorporating
information from the air quality modeling and from ambient measured monitoring data. The
eVNA technique was applied to PM2.5 component species model predictions in conjunction

209	Scenario "i" can represent either the baseline or one of the three illustrative scenarios

210	Snapshot year "y" can represent 2028, 2030, 2035, 2040 or 2045

B-17


-------
with measured data to create modeled/measured fused surfaces that leverage measured
concentrations at air quality monitor locations and model predictions at locations with no
monitoring data.

4.1.	The quarterly average PM2.5 component species eVNA spatial fields are created for the
2016 base year with EPA's SMAT-CE software package using 3 years of monitoring
data (2015-2017) and the 2016 modeled data.

4.2.	The model-predicted spatial fields (i.e., not the eVNA fields) of quarterly average PM2.5
component species in 2016 were paired with the corresponding model-predicted spatial
fields in the future year to calculate the ratio of PM2.5 component species between 2016
and the future year in each model grid cell.

4.3.	To create a gridded future year eVNA surfaces, the spatial fields of 2016/future year
ratios created in step (4.2) were multiplied by the corresponding eVNA spatial fields for
2016 created in step (4.1) to produce an eVNA annual average PM2.5 component species
spatial field for the future year using Eq-1.

5. Create gridded spatial fields of total EGU speciated PM2.5 contributions for each combination
of scenario and snapshot year.

5.1.	Use the EGU annual total NOx, SO2, and PM2.5 emissions for the 2028 baseline scenario
and the corresponding future year modeled EGU NOx, SO2, and PM2.5 emissions from
Table B-l, Table B-2, and Table B-3 to calculate the ratio of 2028 baseline emissions to
future year modeled emissions for each state-fuel EGU contribution tag (i.e., annual
nitrate, sulfate and directly emitted PM2.5 scaling factors calculated for each state-fuel
tag)211. These scaling factors are provided in Table B-8 through Table B-19.

5.2.	Calculate adjusted gridded annual PM2.5 component species EGU contributions that
reflect differences in state-fuel EGUNOx, SO2, and primary PM2.5 emissions between
the future modeled year and the 2028 baseline by multiplying the annual nitrate, sulfate

211 State-level tags were tracked for separately for coal EGUs and for natural gas EGUs. All other EGU emissions
were tracked using a single national tag. In addition, preliminary testing of this methodology showed unstable
results when very small magnitudes of emissions were tagged especially when being scaled by large factors. To
mitigate this issue, in cases where state-fuel EGU tags were associated with no or very small emissions, tags
were combined into multi-state regions.

B-18


-------
and directly emitted PM2.5 scaling factors by the corresponding annual gridded PM2.5
component species contributions from each state-fuel EGU tag212.

5.3.	Add together the adjusted PM2.5 contributions of for each EGU state-fuel tag to produce
spatial fields of adjusted EGU totals for each PM2.5 component species.

5.4.	Repeat steps 5.1 through 5.3 for the three illustrative scenarios in 2028 and for the
baseline and illustrative scenarios for each additional snapshot year. The scaling factors
for all PM2.5 component species for the baseline and illustrative scenarios are provided in
Table B-8 through Table B-19.

6.	Create gridded spatial fields of each PM2.5 component species for the 2028 baseline by
combining the EGU annual PM2.5 component species contributions from step (5.3) with the
corresponding contributions to annual PM2.5 component species from all other sources.

Repeat for each of the EGU contributions created in step (5.4) to create separate gridded
spatial fields for the baseline and three illustrative scenarios for all other snapshot years.

7.	Create gridded spatial fields of total PM2.5 mass by combining the component species
surfaces for sulfate, nitrate, organic aerosol, elemental carbon and crustal material with
ammonium, and particle-bound. Ammonium and particle-bound water concentrations are
calculated for each scenario based on nitrate and sulfate concentrations along with the
ammonium degree of neutralization in the base year modeling (2016) in accordance with
equations from the SMAT-CE modeling software (U.S. EPA, 2022b).

Steps 5 and 6 result in Eq-3 for PM2.5 component species: sulfate, nitrate, organic aerosol,
elemental carbon and crustal material.

212 Scaling factors for components that are formed through chemical reactions in the atmosphere were created as
follows: scaling factors for sulfate were based on relative changes in annual SO2 emissions; scaling factors for
nitrate were based on relative changes in annual NOx emissions. Scaling factors for PM2 5 components that are
emitted directly from the source (OA, EC, crustal) were based on the relative changes in annual primary PM2 5
emissions between the future year modeled emissions and the baseline and the three illustrative scenarios in each

PMs,g,i,y = eVNA,

s,g,future

Eq-3

snapshot year.

B-19


-------
•	PMs g i y is the estimated fused model-obs PM component species "s" for grid-cell, "g", scenario,
"i"213, and year, "y"214;

•	eVNAs gfUture is the future year eVNA PM concentration for component species "s" in grid-cell
"g" calculated using Eq-1.

•	Cs,g,Tot is the total modeled PM component species "s" for grid-cell "g" from all sources in the
future year source apportionment modeling

•	Cs,g,BC is the future year PM component species "s" modeled contribution from the modeled
boundary inflow;

•	Cs,g,int is the future year PM component species "s" modeled contribution from international
emissions within the modeling domain;

•	Cs,g,bio is the future year PM component species "s" modeled contribution from biogenic
emissions;

•	Cs,g,fires is the future year PM component species "s" modeled contribution from fires;

•	Cs,g,usanthro's the total future year PM component species "s" modeled contribution from U.S.
anthropogenic sources other than EGUs;

•	Cecus,g,t is the future year PM component species "s" modeled contribution from EGU
emissions ofNOx, SO2, or primary PM2.5 from tag, "t"; and

•	Ss,t,i,y is the EGU scaling factor for component species "s", tag, "t", scenario "i", and year, "y".
Scaling factors for nitrate are based on annual NOx emissions, scaling factors for sulfate are
based on annual SO2 emissions, scaling factors for primary PM2.5 components are based on
primary PM2.5 emissions

B.3 Scaling Factors Applied to Source Apportionment Tags

Table B-4 Baseline and Alternative 1 Scenario Ozone Scaling Factors for Coal EGU
Tags	

Baseline	Alternative 1

State Tag 2028 2030 2035 2040 2045 2028 2030 2035 2040 2045

AL	1.20 1.40 1.47 1.38 0.29 1.16 1.40 1.41 1.41 0.00

AZ	0.01 1.43 1.13 0.00 0.00 0.01 1.52 0.77 0.77 0.00

213	Scenario "i" can represent either baseline or one of the illustrative scenarios.

214	Snapshot year "y" can represent 2028, 2030, 2035, 2040, or 2045

B-20


-------
Baseline

Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

CA

0.00

0.00

0.00

0.00

0.00

0.00

0.11

0.00

0.00

0.00

CO

139.01

1.28

1.98

1.98

1.98

143.12

5.46

1.98

1.98

1.98

CT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DC

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DE

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

FL

0.47

1.24

0.10

0.10

0.03

0.47

1.16

0.12

0.12

0.03

GA

0.00

0.18

0.00

0.00

0.00

0.11

0.21

0.00

0.00

0.00

IA

1.17

1.18

0.77

0.46

0.42

1.15

0.88

0.04

0.00

0.00

ID

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

IL

0.97

0.96

0.81

0.14

0.00

0.97

0.96

1.00

1.00

0.00

IN

1.35

0.76

0.19

0.19

0.00

1.24

0.74

0.19

0.19

0.00

KY

0.79

0.95

0.97

0.83

0.06

0.80

0.92

0.90

0.90

0.00

MA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MDPA3

3.14

3.17

2.58

1.06

1.30

3.09

2.99

1.74

1.74

0.00

ME

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MI

0.75

0.00

0.00

0.00

0.00

0.75

0.00

0.00

0.00

0.00

MN

2.41

2.25

0.00

0.00

0.00

2.41

2.25

0.00

0.00

0.00

MO

2.72

1.57

0.67

0.31

0.27

2.57

1.37

0.52

0.52

0.01

MT

1.07

1.12

1.11

0.99

0.00

1.07

1.07

0.42

0.36

0.00

NC

9.89

6.41

2.86

1.50

2.86

8.81

10.07

0.00

0.13

0.00

ND

1.09

1.08

0.25

0.24

0.01

1.01

1.01

0.26

0.26

0.01

NE

1.16

1.18

0.73

0.55

0.41

1.15

1.15

0.12

0.10

0.02

NH

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NJ

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NM

0.98

0.98

0.01

0.01

0.01

0.98

0.98

0.01

0.01

0.01

NV

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NY

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

OH

0.58

1.07

0.00

0.00

0.00

0.35

1.07

0.08

0.08

0.00

OR

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

RI

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

SC

0.81

2.22

3.18

3.18

0.00

0.00

2.35

3.18

3.18

0.00

SD

0.87

1.33

0.00

0.00

0.00

0.87

1.33

0.00

0.00

0.00

TN

3.89

0.01

0.00

0.00

0.00

3.12

0.41

0.00

0.00

0.00

TX-regb

4.69

4.26

1.64

1.15

0.54

3.66

3.09

1.52

1.52

0.06

UT

1.00

0.06

0.06

0.06

0.04

1.00

0.06

0.00

0.00

0.00

VA

0.65

0.45

0.00

0.00

0.00

0.65

0.45

0.01

0.01

0.00

VT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WI

1.66

2.16

0.36

0.00

0.00

1.65

0.64

0.01

0.01

0.00

WV

0.92

1.16

0.92

0.27

0.10

0.93

1.20

0.25

0.25

0.00

WY

1.26

1.12

1.12

0.61

0.53

1.22

0.99

0.57

0.28

0.06

B-21


-------
Note: Emissions of Maryland, Arkansas, Kansas, Louisiana, Oklahoma, and Mississippi are less 10 tpy in the
original source apportionment modeling. Air quality impacts and emissions from those states were combined with
nearby states.

aMDPA: Maryland and Pennsylvania

bTX-reg: Arkansas, Kansas, Louisiana, Oklahoma, Mississippi, Texas

Table B-5 Alternative 2 and Final Rules Scenario Ozone Scaling Factors for Coal EGU
Tags	

Alternative 2	Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

1.17

1.40

1.41

1.41

0.00

1.17

1.40

1.41

1.41

0.00

AZ

0.01

1.60

0.78

0.78

0.00

0.01

1.52

0.77

0.77

0.00

CA

0.00

0.00

0.00

0.00

0.00

0.00

0.11

0.00

0.00

0.00

CO

139.01

1.28

1.98

1.98

1.98

146.51

6.20

1.98

1.98

1.98

CT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DC

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DE

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

FL

0.47

1.24

0.12

0.12

0.03

0.47

1.16

0.12

0.12

0.03

GA

0.00

0.17

0.00

0.00

0.00

0.02

0.18

0.00

0.00

0.00

IA

1.14

1.14

0.17

0.00

0.00

1.15

0.88

0.04

0.00

0.00

ID

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

IL

0.97

0.96

1.00

1.00

0.00

0.97

0.96

1.00

1.00

0.00

IN

1.29

0.76

0.19

0.19

0.00

1.23

0.75

0.19

0.19

0.00

KY

0.79

0.89

0.90

0.90

0.00

0.78

0.91

0.90

0.90

0.00

MA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MDPA3

3.12

3.00

1.74

1.74

0.00

3.09

3.00

1.74

1.74

0.00

ME

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MI

0.75

0.00

0.00

0.00

0.00

0.75

0.00

0.00

0.00

0.00

MN

2.41

2.25

0.00

0.00

0.00

2.41

2.25

0.00

0.00

0.00

MO

2.56

1.48

0.49

0.49

0.01

2.56

1.37

0.52

0.52

0.01

MT

1.07

1.12

0.42

0.36

0.00

1.07

1.07

0.42

0.36

0.00

NC

6.84

6.09

0.00

0.13

0.00

9.96

9.45

0.00

0.13

0.00

ND

1.01

1.08

0.26

0.26

0.01

1.06

1.01

0.26

0.26

0.01

NE

1.15

1.18

0.13

0.10

0.02

1.15

1.15

0.12

0.10

0.02

NH

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NJ

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NM

0.98

0.98

0.01

0.01

0.01

0.98

0.98

0.01

0.01

0.01

NV

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NY

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

OH

0.44

1.03

0.08

0.08

0.00

0.40

1.03

0.08

0.08

0.00

OR

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

RI

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

SC

0.32

2.19

3.18

3.18

0.00

0.11

2.35

3.18

3.18

0.00

B-22


-------
Alternative 2

Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

SD

0.87

1.33

0.00

0.00

0.00

0.87

1.33

0.00

0.00

0.00

TN

3.89

0.01

0.00

0.00

0.00

2.64

0.42

0.00

0.00

0.00

TX-regb

3.67

3.38

1.77

1.76

0.10

3.46

3.03

1.52

1.52

0.06

UT

1.00

0.06

0.00

0.00

0.00

1.00

0.06

0.00

0.00

0.00

VA

0.65

0.45

0.01

0.01

0.00

0.65

0.45

0.01

0.01

0.00

VT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WI

1.65

2.14

0.00

0.00

0.00

1.65

0.64

0.01

0.01

0.00

WV

0.93

1.19

0.25

0.25

0.00

0.93

1.20

0.25

0.25

0.00

WY

1.22

1.12

0.57

0.29

0.07

1.22

0.99

0.57

0.28

0.06

Note: Emissions of Maryland, Arkansas, Kansas, Louisiana, Oklahoma, and Mississippi are less 10 tpy in the
original source apportionment modeling. Air quality impacts and emissions from those states were combined with
nearby states.

aMDPA: Maryland and Pennsylvania

b TX-reg: Arkansas, Kansas, Louisiana, Oklahoma, Mississippi, Texas

Table B-6 Baseline and Alternative 1 Scenario Ozone Scaling Factors for Natural Gas
EGU Tags	

Baseline	Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

0.53

0.61

0.49

0.39

0.27

0.52

0.50

0.50

0.41

0.31

AR

0.65

0.68

0.43

0.20

0.10

0.68

0.77

0.47

0.21

0.10

AZ

0.69

0.68

0.67

0.68

0.45

0.69

0.69

0.71

0.68

0.40

CA

0.92

0.94

0.85

0.52

0.02

0.95

0.94

1.36

0.53

0.02

CO

3.26

0.63

0.50

0.48

0.12

3.47

0.61

0.76

0.44

0.16

CT

1.04

0.98

0.89

0.00

0.01

1.05

1.00

0.92

0.00

0.01

DC

0.86

0.59

0.33

0.21

0.16

0.86

0.59

0.30

0.19

0.12

DE

0.79

0.80

0.38

0.37

0.38

0.81

0.81

0.59

0.37

0.37

FL

1.08

1.03

1.04

0.89

0.66

1.08

1.05

1.04

0.89

0.64

GA

0.58

0.54

0.52

0.42

0.38

0.57

0.55

0.52

0.42

0.38

IA

0.53

0.42

0.16

0.04

0.01

0.55

0.54

0.21

0.04

0.01

ID

0.60

0.90

0.90

0.90

0.04

0.68

1.00

0.45

0.54

0.04

IL

0.69

0.61

0.42

0.21

0.00

0.71

0.66

0.45

0.21

0.00

IN

0.75

0.63

0.38

0.20

0.15

0.78

0.67

0.44

0.21

0.16

KS

1.38

1.32

0.25

0.14

0.10

1.36

1.36

0.52

0.50

0.48

KY

0.87

0.81

0.69

0.57

0.38

0.89

0.85

0.67

0.58

0.44

LA

1.04

1.00

0.72

0.45

0.41

1.02

0.95

0.74

0.48

0.41

MA

0.60

0.67

0.66

0.84

0.47

0.58

0.67

0.68

0.84

0.44

MD

1.51

1.33

1.12

0.84

0.79

1.51

1.29

1.07

0.79

0.73

ME

1.16

1.15

0.59

0.63

0.36

1.16

1.15

0.74

0.65

0.36

MI

0.68

0.70

0.55

0.41

0.23

0.77

0.82

0.64

0.47

0.22

B-23


-------
Baseline

Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

MN

0.92

0.84

0.34

0.17

0.13

0.92

0.93

0.38

0.20

0.08

MO

0.59

0.59

0.20

0.08

0.04

0.55

0.70

0.26

0.08

0.04

MS

0.64

0.62

0.50

0.45

0.29

0.64

0.62

0.50

0.45

0.28

MT

0.95

1.10

0.08

0.14

0.02

0.95

0.79

0.66

0.14

0.02

NC

0.77

0.59

0.68

0.63

0.51

0.77

0.58

0.71

0.73

0.61

ND

0.85

1.85

0.34

0.96

0.14

0.82

1.85

1.05

0.96

0.14

NE

5.91

5.92

0.28

0.87

0.02

5.90

6.46

8.17

6.07

4.85

NH

0.67

0.51

0.41

0.41

0.41

0.64

0.50

0.49

0.41

0.41

NJ

0.81

0.85

0.61

0.49

0.46

0.83

0.88

0.73

0.49

0.46

NM

1.00

0.84

0.77

0.35

0.47

0.98

0.82

0.66

0.29

0.61

NV

0.33

0.25

0.19

0.21

0.12

0.27

0.25

0.21

0.21

0.10

NY

1.03

0.99

0.65

0.28

0.28

1.04

0.99

0.63

0.28

0.28

OH

1.02

0.97

0.84

0.71

0.62

1.05

0.97

0.80

0.68

0.63

OK

1.69

1.57

0.48

0.33

0.32

1.63

1.60

0.66

0.49

0.48

OR

63.29

0.00

0.00

0.00

0.00

59.34

0.14

0.60

0.00

0.00

PA

0.79

0.69

0.34

0.24

0.23

0.91

0.69

0.36

0.23

0.23

RI

0.69

0.75

0.71

0.88

0.89

0.69

0.75

0.70

0.88

0.79

SC

0.93

0.96

0.59

0.59

0.56

0.94

0.96

0.58

0.54

0.53

SD

0.59

0.59

0.17

0.06

0.03

0.52

0.62

0.25

0.16

0.01

TN

1.12

1.09

1.07

0.90

0.51

1.18

1.08

1.02

0.91

0.50

TX

0.99

0.89

0.47

0.28

0.15

0.87

0.82

0.51

0.30

0.20

UT

0.50

0.43

0.34

0.37

0.31

0.50

0.42

0.32

0.35

0.29

VA

0.89

0.85

0.54

0.32

0.26

0.89

0.84

0.64

0.32

0.26

VT

0.00

0.37

3.53

3.99

0.00

0.00

0.37

3.53

3.99

0.00

WA

0.08

0.23

0.79

0.74

0.02

0.08

0.23

0.89

0.87

0.02

WI

0.74

0.70

0.58

0.30

0.14

0.73

0.86

0.68

0.31

0.18

WV

1.19

1.12

0.33

0.13

0.07

1.19

1.12

0.48

0.11

0.09

WY

0.01

0.04

0.06

0.06

0.00

0.01

0.03

0.35

0.17

0.00

Table B-7 Alternative 2 and Final Rules Scenario Ozone Scaling Factors for Natural
Gas EGU Tags	

Alternative 2	Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

0.53

0.61

0.50

0.41

0.31

0.53

0.50

0.50

0.41

0.31

AR

0.68

0.68

0.44

0.21

0.10

0.68

0.83

0.48

0.21

0.10

AZ

0.69

0.68

0.71

0.68

0.40

0.69

0.69

0.71

0.69

0.39

CA

0.95

0.94

1.39

0.53

0.02

0.94

0.94

1.58

0.54

0.02

CO

3.47

0.63

0.82

0.41

0.16

2.78

0.62

0.76

0.47

0.10

B-24


-------
Alternative 2

Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

CT

1.04

0.99

0.92

0.00

0.01

1.05

1.00

0.92

0.00

0.01

DC

0.86

0.59

0.30

0.19

0.12

0.81

0.59

0.28

0.15

0.11

DE

0.81

0.80

0.61

0.37

0.37

0.80

0.81

0.68

0.37

0.37

FL

1.08

1.03

1.04

0.89

0.64

1.08

1.05

1.05

0.89

0.63

GA

0.58

0.54

0.52

0.43

0.38

0.58

0.55

0.52

0.43

0.38

IA

0.55

0.39

0.20

0.04

0.01

0.55

0.54

0.21

0.04

0.01

ID

0.67

0.98

0.45

0.53

0.04

0.65

0.95

0.22

0.29

0.04

IL

0.69

0.62

0.45

0.21

0.00

0.74

0.68

0.45

0.22

0.00

IN

0.76

0.63

0.45

0.21

0.16

0.78

0.68

0.48

0.21

0.16

KS

1.37

1.34

0.56

0.43

0.41

1.30

1.36

0.61

0.50

0.48

KY

0.89

0.83

0.67

0.58

0.43

0.92

0.86

0.65

0.59

0.44

LA

1.03

1.02

0.75

0.48

0.42

1.02

0.94

0.74

0.48

0.42

MA

0.57

0.67

0.67

0.84

0.44

0.58

0.67

0.66

0.85

0.44

MD

1.35

1.28

1.07

0.79

0.73

1.51

1.32

1.14

0.68

0.71

ME

1.16

1.15

0.76

0.65

0.36

1.16

1.15

0.83

0.64

0.36

MI

0.77

0.71

0.64

0.47

0.22

0.78

0.82

0.67

0.49

0.22

MN

0.85

0.76

0.37

0.20

0.08

1.00

1.00

0.37

0.20

0.08

MO

0.55

0.60

0.28

0.08

0.04

0.55

0.72

0.30

0.08

0.04

MS

0.75

0.61

0.50

0.46

0.28

0.64

0.61

0.50

0.46

0.28

MT

0.95

0.79

2.04

0.14

0.02

0.95

0.79

0.83

0.14

0.02

NC

0.77

0.60

0.71

0.73

0.61

0.77

0.60

0.69

0.69

0.68

ND

0.82

1.84

1.05

0.96

0.14

0.90

1.85

1.05

0.96

0.14

NE

5.91

5.95

8.85

6.09

5.12

5.90

6.55

8.24

6.07

4.85

NH

0.64

0.50

0.49

0.41

0.41

0.63

0.50

0.49

0.41

0.41

NJ

0.83

0.85

0.72

0.49

0.46

0.84

0.95

0.73

0.57

0.46

NM

0.98

0.82

0.66

0.29

0.61

0.98

0.82

0.66

0.27

0.50

NV

0.30

0.24

0.21

0.21

0.10

0.24

0.25

0.21

0.21

0.10

NY

1.03

0.98

0.64

0.28

0.28

1.05

1.01

0.64

0.28

0.28

OH

1.03

0.96

0.80

0.68

0.62

1.03

0.96

0.77

0.67

0.61

OK

1.64

1.54

0.66

0.48

0.47

1.63

1.58

0.67

0.49

0.48

OR

59.93

0.14

0.59

0.00

0.00

61.93

0.00

2.45

0.00

0.00

PA

0.91

0.69

0.36

0.23

0.23

0.91

0.69

0.36

0.24

0.23

RI

0.69

0.75

0.70

0.88

0.79

0.69

0.75

0.70

0.88

0.79

SC

0.94

0.97

0.58

0.55

0.53

0.95

0.95

0.59

0.55

0.54

SD

0.52

0.56

0.25

0.15

0.01

0.52

0.62

0.29

0.16

0.01

TN

1.16

1.09

1.02

0.91

0.50

1.21

1.09

1.02

0.93

0.50

TX

0.88

0.79

0.51

0.29

0.18

0.85

0.82

0.51

0.30

0.20

UT

0.50

0.42

0.32

0.35

0.29

0.50

0.41

0.32

0.35

0.28

VA

0.89

0.83

0.64

0.33

0.26

0.89

0.85

0.64

0.33

0.28

VT

0.00

0.37

3.53

3.99

0.00

0.00

0.37

3.53

3.99

0.00

WA

0.08

0.20

0.89

0.87

0.02

0.10

0.22

0.90

0.88

0.02

B-25


-------
Alternative 2

Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

WI

0.73

0.64

0.68

0.31

0.18

0.74

0.87

0.68

0.31

0.18

wv

1.19

1.12

0.48

0.11

0.09

1.19

1.13

0.53

0.11

0.09

WY

0.01

0.01

0.35

0.15

0.00

0.02

0.03

0.33

0.17

0.00

Table B-8 Baseline and Alternative 1 Nitrate Scaling Factors for Coal EGU tags

Baseline	Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

1.33

1.45

1.65

1.54

0.14

1.31

1.45

1.60

1.60

0.00

AR

39.93

8.30

3.83

0.71

0.28

37.13

3.65

0.61

0.49

0.06

AZ

0.47

0.97

0.59

0.20

0.15

0.45

0.98

0.69

0.69

0.00

CA

0.24

0.36

0.16

0.13

0.00

0.24

0.43

0.02

0.01

0.00

CO

25.56

0.97

0.37

0.41

0.37

26.51

0.97

0.37

0.41

0.37

CT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DC

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DE

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

FL

0.89

1.20

0.26

0.26

0.14

0.89

1.38

0.18

0.18

0.02

GA

0.23

0.12

0.00

0.00

0.00

0.22

0.26

0.00

0.00

0.00

IA

1.20

1.16

0.68

0.28

0.19

1.19

0.89

0.03

0.00

0.00

ID

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

IL

0.98

0.92

0.62

0.14

0.00

0.97

0.93

0.99

0.99

0.00

IN

1.29

0.64

0.11

0.11

0.00

1.20

0.59

0.11

0.11

0.00

KS

45.15

46.03

3.08

3.08

0.00

28.46

30.56

3.08

3.08

0.00

KY

1.38

1.12

1.15

1.00

0.07

1.25

1.10

1.11

1.11

0.00

LA

24.63

16.33

25.37

13.43

2.22

20.11

20.90

13.43

13.43

2.28

MA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MD

3.54

3.54

3.54

3.54

2.97

3.54

3.57

3.57

3.57

0.00

ME

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MI

0.74

0.00

0.00

0.00

0.00

0.74

0.00

0.00

0.00

0.00

MN

2.97

2.31

0.00

0.00

0.00

2.93

2.34

0.00

0.00

0.00

MO

1.41

1.06

0.43

0.04

0.03

1.29

0.93

0.20

0.20

0.00

MS

4.02

3.60

1.06

1.00

1.00

1.94

3.60

8.15

8.15

0.00

MT

1.07

1.09

1.08

1.02

0.38

1.07

1.07

0.39

0.35

0.00

NC

19.19

11.95

3.66

3.51

3.84

15.57

12.43

0.07

0.07

0.07

ND

1.03

1.03

0.25

0.25

0.01

0.99

0.98

0.27

0.27

0.01

NE

1.14

1.13

0.61

0.37

0.18

1.12

1.11

0.11

0.09

0.01

NH

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NJ

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

B-26


-------
Baseline

Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

NM

0.99

0.99

0.01

0.01

0.01

0.99

1.00

0.01

0.01

0.01

NV

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NY

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

OH

0.90

0.94

0.19

0.00

0.00

0.79

0.97

0.07

0.06

0.00

OK

12.10

5.08

3.11

3.11

1.03

9.02

4.41

0.14

0.14

0.00

OR

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

PA

3.05

2.94

2.61

1.19

1.16

2.96

2.90

1.60

1.60

0.00

RI

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

SC

1.15

1.92

2.98

2.98

0.00

0.78

1.90

2.98

2.98

0.00

SD

0.93

1.11

0.00

0.00

0.00

0.91

1.11

0.00

0.00

0.00

TN

7.49

1.00

0.00

0.00

0.00

6.21

1.10

0.00

0.00

0.00

TX

1.02

1.13

0.87

0.47

0.12

0.77

0.73

0.81

0.81

0.01

UT

3.50

0.09

0.09

0.09

0.06

3.50

0.09

0.00

0.00

0.00

VA

0.67

0.41

0.12

0.00

0.00

0.67

0.41

0.01

0.01

0.00

VT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WI

1.84

2.07

0.38

0.00

0.00

1.77

0.56

0.01

0.01

0.00

WV

1.25

1.30

0.97

0.27

0.09

1.23

1.34

0.26

0.26

0.00

WY

1.32

1.15

1.14

0.61

0.48

1.29

1.03

0.56

0.29

0.06

Table B-9 Alternative 2 and Final Rules Nitrate Scaling Factors for Coal EGU Tags

Alternative 2	Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

1.32

1.45

1.60

1.60

0.00

1.32

1.45

1.60

1.60

0.00

AR

37.33

8.25

0.61

0.49

0.06

36.19

3.65

0.61

0.49

0.06

AZ

0.45

0.97

0.70

0.70

0.00

0.46

0.98

0.69

0.69

0.00

CA

0.24

0.36

0.02

0.01

0.00

0.24

0.43

0.02

0.01

0.00

CO

26.20

0.97

0.37

0.41

0.37

26.75

0.97

0.40

0.40

0.37

CT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DC

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DE

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

FL

0.89

1.20

0.18

0.18

0.02

0.89

1.38

0.18

0.18

0.02

GA

0.22

0.11

0.00

0.00

0.00

0.17

0.21

0.00

0.00

0.00

IA

1.19

1.15

0.13

0.00

0.00

1.19

0.89

0.03

0.00

0.00

ID

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

IL

0.97

0.93

0.99

0.99

0.00

0.97

0.93

0.99

0.99

0.00

IN

1.23

0.65

0.11

0.11

0.00

1.20

0.59

0.11

0.11

0.00

KS

28.47

29.33

3.08

3.08

0.00

27.03

29.44

3.08

3.08

0.00

KY

1.25

1.05

1.11

1.11

0.00

1.22

1.09

1.11

1.11

0.00

B-27


-------
Alternative 2

Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

LA

20.21

14.07

13.43

13.43

2.28

20.85

20.42

13.43

13.43

2.28

MA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MD

3.54

3.54

3.54

3.54

0.00

3.54

3.57

3.57

3.57

0.00

ME

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MI

0.74

0.00

0.00

0.00

0.00

0.74

0.00

0.00

0.00

0.00

MN

2.95

2.29

0.00

0.00

0.00

2.91

2.36

0.00

0.00

0.00

MO

1.29

0.96

0.19

0.19

0.00

1.29

0.93

0.20

0.20

0.00

MS

1.94

3.60

8.15

8.15

0.00

1.94

3.60

8.15

8.15

0.00

MT

1.07

1.09

0.39

0.35

0.00

1.07

1.07

0.39

0.35

0.00

NC

13.93

10.44

0.07

0.07

0.07

17.53

12.83

0.07

0.07

0.07

ND

1.00

1.02

0.27

0.27

0.01

1.01

0.98

0.27

0.27

0.01

NE

1.12

1.13

0.12

0.09

0.01

1.12

1.10

0.11

0.09

0.01

NH

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NJ

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NM

0.99

0.99

0.01

0.01

0.01

0.99

1.00

0.01

0.01

0.01

NV

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NY

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

OH

0.83

0.94

0.07

0.06

0.00

0.81

0.95

0.07

0.06

0.00

OK

9.55

5.09

2.08

2.08

0.00

5.42

4.41

0.14

0.14

0.00

OR

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

PA

2.96

2.90

1.60

1.60

0.00

2.96

2.90

1.60

1.60

0.00

RI

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

SC

0.92

1.79

2.98

2.98

0.00

0.78

1.90

2.98

2.98

0.00

SD

0.91

1.06

0.00

0.00

0.00

0.91

1.07

0.00

0.00

0.00

TN

7.49

1.00

0.00

0.00

0.00

5.40

1.00

0.00

0.00

0.00

TX

0.91

0.98

0.96

0.96

0.03

0.76

0.73

0.81

0.81

0.01

UT

3.50

0.09

0.00

0.00

0.00

3.50

0.09

0.00

0.00

0.00

VA

0.67

0.48

0.01

0.01

0.00

0.67

0.34

0.01

0.01

0.00

VT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WI

1.77

2.03

0.00

0.00

0.00

1.77

0.56

0.01

0.01

0.00

WV

1.24

1.32

0.26

0.26

0.00

1.22

1.34

0.26

0.26

0.00

WY

1.29

1.14

0.56

0.30

0.07

1.29

1.04

0.56

0.29

0.06

Table B-10 Baseline and Alternative 1 Nitrate Scaling Factors for Natural Gas EGU
Tags	

Baseline	Alternative 1

State Tag 2028 2030 2035 2040 2045 2028 2030 2035 2040 2045

AL 0.59 0.60 0.45 0.27 0.16 0.58 0.54 0.46 0.28 0.19

B-28


-------
Baseline

Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AR

0.56

0.68

0.38

0.13

0.06

0.58

0.78

0.41

0.14

0.06

AZ

0.73

0.85

0.83

0.75

0.37

0.72

0.82

0.86

0.74

0.35

CA

0.76

0.88

0.97

0.67

0.16

0.77

0.89

1.15

0.67

0.16

CO

2.02

0.71

0.72

0.76

0.30

2.10

0.73

0.93

0.72

0.42

CT

0.92

0.81

0.66

0.00

0.01

0.91

0.82

0.66

0.00

0.01

DC

0.63

0.47

0.26

0.18

0.13

0.62

0.47

0.19

0.13

0.10

DE

0.79

0.76

0.33

0.29

0.30

0.79

0.78

0.53

0.29

0.30

FL

1.11

1.06

1.01

0.73

0.49

1.11

1.06

1.00

0.77

0.48

GA

0.68

0.63

0.54

0.29

0.22

0.67

0.63

0.54

0.30

0.23

IA

0.49

0.42

0.13

0.03

0.01

0.50

0.55

0.19

0.03

0.01

ID

1.02

1.36

1.39

1.24

0.60

1.17

1.53

1.10

1.06

0.71

IL

0.54

0.54

0.29

0.12

0.00

0.55

0.61

0.34

0.13

0.00

IN

0.67

0.59

0.34

0.12

0.08

0.67

0.61

0.39

0.12

0.09

KS

0.96

0.87

0.20

0.07

0.05

0.95

0.96

0.37

0.27

0.26

KY

0.81

0.76

0.46

0.25

0.15

0.84

0.78

0.42

0.24

0.17

LA

0.96

0.94

0.61

0.27

0.24

0.94

0.98

0.62

0.28

0.24

MA

0.64

0.66

0.54

0.61

0.33

0.62

0.66

0.54

0.60

0.32

MD

1.47

1.35

1.05

0.72

0.66

1.46

1.33

0.93

0.59

0.61

ME

1.64

1.34

0.63

0.58

0.34

1.56

1.31

0.80

0.58

0.34

MI

0.65

0.71

0.43

0.30

0.15

0.71

0.80

0.56

0.32

0.15

MN

1.02

0.95

0.36

0.15

0.09

1.01

1.07

0.38

0.16

0.06

MO

0.52

0.52

0.19

0.06

0.03

0.49

0.63

0.26

0.06

0.03

MS

0.61

0.56

0.36

0.24

0.15

0.61

0.56

0.41

0.24

0.14

MT

0.66

0.80

0.05

0.08

0.01

0.66

0.64

0.41

0.08

0.01

NC

0.89

0.67

0.72

0.55

0.47

0.89

0.69

0.77

0.64

0.54

ND

0.66

1.32

0.26

0.60

0.09

0.65

1.47

1.01

0.60

0.09

NE

2.05

1.80

0.13

0.31

0.01

2.06

2.01

2.46

1.72

1.31

NH

0.78

0.59

0.44

0.38

0.36

0.76

0.58

0.49

0.38

0.36

NJ

0.82

0.83

0.51

0.34

0.39

0.83

0.85

0.61

0.35

0.39

NM

0.74

0.66

0.64

0.33

0.39

0.73

0.65

0.58

0.29

0.41

NV

0.50

0.39

0.44

0.40

0.23

0.46

0.35

0.46

0.39

0.23

NY

0.91

0.89

0.55

0.16

0.16

0.93

0.88

0.56

0.16

0.16

OH

1.00

0.98

0.87

0.59

0.42

1.01

0.98

0.80

0.55

0.40

OK

1.43

1.20

0.34

0.21

0.20

1.32

1.27

0.49

0.31

0.31

OR

5.58

0.96

0.50

0.00

0.00

5.45

0.95

0.48

0.00

0.00

PA

0.69

0.61

0.35

0.21

0.18

0.75

0.61

0.36

0.21

0.18

RI

0.76

0.76

0.64

0.71

0.68

0.76

0.76

0.61

0.71

0.61

SC

0.94

0.96

0.67

0.56

0.55

0.92

1.02

0.68

0.55

0.55

SD

0.55

0.55

0.16

0.06

0.04

0.50

0.59

0.29

0.16

0.01

TN

1.02

0.97

0.79

0.41

0.23

1.04

0.96

0.77

0.42

0.22

TX

0.97

0.88

0.42

0.17

0.08

0.84

0.79

0.47

0.19

0.11

B-29


-------
Baseline

Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

UT

0.52

0.62

0.56

0.58

0.46

0.52

0.59

0.53

0.53

0.43

VA

0.84

0.80

0.43

0.20

0.15

0.84

0.77

0.53

0.20

0.15

VT

0.10

0.16

1.53

1.73

0.00

0.10

0.16

1.53

1.73

0.00

WA

0.43

0.36

0.72

0.97

0.44

0.44

0.35

0.71

0.75

0.46

WI

0.66

0.67

0.45

0.18

0.08

0.65

0.81

0.56

0.19

0.11

WV

1.02

0.89

0.22

0.08

0.04

1.03

0.91

0.31

0.07

0.06

WY

0.01

0.04

0.06

0.06

0.00

0.01

0.03

0.35

0.16

0.00

B-30


-------
Table B-ll Alternative 2 and Final Rules Nitrate Scaling Factors for Natural Gas EGU
Tags	

Alternative 2	Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

0.59

0.60

0.46

0.28

0.19

0.59

0.54

0.47

0.29

0.19

AR

0.58

0.67

0.39

0.14

0.06

0.59

0.82

0.41

0.14

0.06

AZ

0.72

0.81

0.86

0.74

0.35

0.73

0.85

0.86

0.76

0.35

CA

0.77

0.89

1.17

0.67

0.16

0.76

0.89

1.40

0.66

0.16

CO

2.10

0.73

0.96

0.71

0.42

1.75

0.74

0.93

0.80

0.39

CT

0.91

0.81

0.66

0.00

0.01

0.91

0.83

0.66

0.00

0.01

DC

0.62

0.47

0.19

0.13

0.10

0.60

0.47

0.17

0.11

0.08

DE

0.79

0.77

0.54

0.29

0.30

0.80

0.80

0.60

0.29

0.30

FL

1.11

1.06

1.00

0.77

0.48

1.11

1.06

1.00

0.78

0.48

GA

0.68

0.63

0.55

0.31

0.23

0.68

0.62

0.55

0.31

0.23

IA

0.50

0.40

0.18

0.03

0.01

0.51

0.55

0.19

0.03

0.01

ID

1.17

1.51

1.11

1.05

0.71

1.12

1.42

0.93

0.80

0.63

IL

0.54

0.55

0.34

0.13

0.00

0.58

0.62

0.34

0.13

0.00

IN

0.67

0.58

0.40

0.12

0.09

0.68

0.62

0.43

0.13

0.09

KS

0.96

0.92

0.39

0.23

0.22

0.92

0.97

0.41

0.27

0.26

KY

0.84

0.77

0.42

0.24

0.17

0.85

0.78

0.40

0.25

0.17

LA

0.96

0.95

0.62

0.29

0.24

0.95

0.98

0.62

0.28

0.24

MA

0.61

0.66

0.54

0.60

0.32

0.62

0.66

0.53

0.61

0.32

MD

1.37

1.33

0.93

0.59

0.61

1.47

1.35

0.94

0.53

0.56

ME

1.56

1.29

0.81

0.58

0.34

1.56

1.30

0.88

0.59

0.34

MI

0.71

0.73

0.56

0.32

0.15

0.72

0.79

0.59

0.33

0.15

MN

0.96

0.89

0.38

0.16

0.06

1.08

1.14

0.38

0.16

0.06

MO

0.49

0.53

0.28

0.06

0.03

0.50

0.65

0.29

0.06

0.03

MS

0.66

0.55

0.41

0.24

0.14

0.62

0.56

0.41

0.24

0.14

MT

0.66

0.62

1.24

0.08

0.01

0.68

0.64

0.50

0.08

0.01

NC

0.90

0.68

0.77

0.63

0.54

0.89

0.70

0.74

0.63

0.57

ND

0.65

1.33

1.11

0.60

0.09

0.83

1.47

0.89

0.60

0.09

NE

2.06

1.82

2.65

1.73

1.39

2.06

2.09

2.48

1.72

1.31

NH

0.75

0.58

0.49

0.38

0.36

0.74

0.58

0.50

0.38

0.36

NJ

0.83

0.83

0.61

0.35

0.39

0.84

0.89

0.62

0.39

0.39

NM

0.73

0.64

0.58

0.29

0.41

0.72

0.65

0.58

0.28

0.41

NV

0.48

0.35

0.46

0.39

0.23

0.44

0.40

0.46

0.39

0.23

NY

0.92

0.88

0.57

0.16

0.16

0.92

0.90

0.57

0.16

0.16

OH

0.99

0.97

0.80

0.55

0.40

1.00

0.96

0.77

0.53

0.38

OK

1.34

1.18

0.49

0.30

0.30

1.30

1.28

0.50

0.31

0.31

OR

5.46

0.96

0.50

0.00

0.00

5.53

0.96

0.50

0.00

0.00

PA

0.76

0.61

0.36

0.21

0.18

0.77

0.61

0.36

0.22

0.18

RI

0.76

0.76

0.61

0.71

0.61

0.76

0.76

0.61

0.71

0.61

SC

0.93

0.98

0.67

0.55

0.55

0.92

1.02

0.68

0.53

0.55

B-31


-------
Alternative 2

Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

SD

0.50

0.53

0.29

0.16

0.01

0.49

0.59

0.32

0.16

0.01

TN

1.03

0.96

0.78

0.42

0.22

1.06

0.96

0.78

0.43

0.22

TX

0.85

0.76

0.46

0.18

0.10

0.83

0.79

0.47

0.19

0.11

UT

0.52

0.59

0.53

0.53

0.43

0.52

0.59

0.53

0.53

0.42

VA

0.84

0.77

0.54

0.21

0.15

0.85

0.80

0.55

0.21

0.16

VT

0.10

0.16

1.53

1.73

0.00

0.10

0.16

1.53

1.73

0.00

WA

0.44

0.36

0.72

0.75

0.46

0.45

0.35

0.68

0.75

0.41

WI

0.64

0.63

0.57

0.18

0.11

0.66

0.81

0.57

0.19

0.11

WV

1.03

0.90

0.31

0.07

0.06

1.04

0.92

0.34

0.07

0.06

WY

0.01

0.01

0.34

0.14

0.00

0.02

0.03

0.35

0.17

0.00

Table B-12 Baseline and Alternative 1 Sulfate Scaling Factors for Coal EGU Tags

Baseline	Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

4.96

5.39

7.07

5.96

0.34

4.92

5.39

6.66

6.43

0.00

AR

118.10

7.02

4.45

1.09

0.42

110.71

0.84

1.04

0.66

0.00

AZ

0.48

1.42

1.16

0.32

0.31

0.50

1.44

0.95

0.41

0.00

CA

0.33

0.50

0.26

0.19

0.00

0.33

0.57

0.11

0.07

0.00

CO

14.31

0.98

0.20

0.22

0.21

15.56

0.98

0.20

0.22

0.21

CT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DC

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DE

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

FL

0.98

1.16

0.50

0.50

0.38

0.98

1.27

0.50

0.50

0.00

GA

0.04

0.09

0.00

0.00

0.00

0.14

0.19

0.00

0.00

0.00

IA

1.31

1.25

0.78

0.32

0.21

1.29

0.81

0.03

0.00

0.00

ID

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

IL

1.01

0.73

0.48

0.10

0.00

1.01

0.74

0.78

0.78

0.00

IN

0.89

0.56

0.12

0.13

0.00

0.82

0.60

0.12

0.14

0.00

KS

52.35

51.92

11.39

11.39

0.00

35.24

37.81

11.39

11.39

0.00

KY

2.68

2.12

1.88

1.71

0.09

2.43

2.11

1.82

1.82

0.00

MA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MD

3.54

3.54

3.54

3.54

2.97

3.54

3.57

3.57

3.57

0.00

ME

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MI

0.85

0.00

0.00

0.00

0.00

0.85

0.00

0.00

0.00

0.00

MN

1.68

1.47

0.00

0.00

0.00

1.66

1.49

0.00

0.00

0.00

MO

2.20

1.08

0.71

0.10

0.12

2.07

0.98

0.07

0.07

0.00

MS

4.02

3.60

1.06

1.00

1.00

1.94

3.60

8.15

8.15

0.00

MT

1.85

2.06

1.92

1.30

0.39

1.85

1.72

1.29

0.91

0.00

NC

7.31

5.14

1.88

1.67

2.03

6.20

4.30

0.00

0.00

0.00

B-32


-------
Baseline

Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

ND

0.94

1.00

0.94

0.93

0.03

0.91

0.98

1.16

1.16

0.03

NE

0.96

0.95

0.58

0.35

0.18

0.95

0.94

0.05

0.04

0.00

NH

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NJ

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NM

1.00

1.00

0.01

0.01

0.01

1.00

1.00

0.01

0.01

0.01

NV

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NY

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

OH

0.78

0.61

0.29

0.00

0.00

0.68

0.70

0.04

0.02

0.00

OK

37.84

4.77

2.54

2.54

1.68

36.50

4.32

0.11

0.11

0.00

OR

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

PA

4.25

4.06

3.94

1.63

1.83

4.20

4.11

1.79

1.79

0.00

RI

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

SC

0.73

1.22

1.76

1.76

0.00

0.54

1.26

1.76

1.76

0.00

SD

1.05

1.27

0.00

0.00

0.00

1.08

1.27

0.00

0.00

0.00

TN

20.55

1.57

0.00

0.00

0.00

15.91

1.72

0.00

0.00

0.00

TXLA3

1.86

2.39

2.25

1.61

0.42

1.23

1.96

1.76

1.71

0.24

UT

0.93

0.06

0.06

0.05

0.04

0.95

0.06

0.00

0.00

0.00

VA

0.11

0.07

0.02

0.00

0.00

0.11

0.07

0.00

0.00

0.00

VT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WI

3.50

3.83

1.15

0.00

0.00

3.37

0.95

0.01

0.01

0.00

WV

1.40

1.39

1.08

0.36

0.12

1.30

1.52

0.31

0.31

0.00

WY

1.26

0.98

0.97

0.49

0.37

1.22

0.89

0.85

0.45

0.11

Note: Emissions of Louisiana are less 10 tpy in the original source apportionment modeling. Air quality impacts an
emissions from Texas and Louisiana were combined.

3 TXLA: Louisiana and Texas

Table B-13

Alternative 2 and Final Rules Sulfate Scaling Factors for Coal EGU Tags





Alternative 2









Final Rules





State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

4.94

5.39

6.66

6.63

0.00

4.94

5.39

6.66

6.63

0.00

AR

111.33

6.92

1.04

0.66

0.00

107.76

0.84

1.04

0.66

0.00

AZ

0.50

1.42

0.96

0.41

0.00

0.50

1.44

0.95

0.41

0.00

CA

0.33

0.50

0.11

0.07

0.00

0.33

0.57

0.11

0.07

0.00

CO

15.29

0.98

0.20

0.22

0.21

15.78

0.98

0.21

0.21

0.21

CT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DC

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DE

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

FL

0.98

1.15

0.50

0.50

0.00

0.98

1.26

0.50

0.50

0.00

GA

0.04

0.09

0.00

0.00

0.00

0.05

0.12

0.00

0.00

0.00

B-33


-------
Alternative 2

Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

IA

1.29

1.23

0.09

0.00

0.00

1.29

0.81

0.03

0.00

0.00

ID

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

IL

1.01

0.74

0.78

0.78

0.00

1.01

0.74

0.78

0.78

0.00

IN

0.85

0.56

0.12

0.14

0.00

0.82

0.60

0.12

0.14

0.00

KS

35.29

36.29

11.39

11.39

0.00

33.52

36.70

11.39

11.39

0.00

KY

2.44

2.06

1.82

1.82

0.00

2.39

2.10

1.82

1.82

0.00

MA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MD

3.54

3.54

3.54

3.54

0.00

3.54

3.57

3.57

3.57

0.00

ME

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MI

0.85

0.00

0.00

0.00

0.00

0.85

0.00

0.00

0.00

0.00

MN

1.67

1.45

0.00

0.00

0.00

1.64

1.50

0.00

0.00

0.00

MO

2.07

0.99

0.06

0.06

0.00

2.07

0.98

0.07

0.07

0.00

MS

1.94

3.60

8.15

8.15

0.00

1.94

3.60

8.15

8.15

0.00

MT

1.85

2.06

1.29

0.91

0.00

1.81

1.73

1.29

0.91

0.00

NC

5.56

4.46

0.00

0.00

0.00

6.82

4.32

0.00

0.00

0.00

ND

0.91

1.00

0.95

0.95

0.02

0.93

0.97

1.16

1.16

0.03

NE

0.95

0.95

0.05

0.04

0.00

0.94

0.94

0.05

0.04

0.00

NH

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NJ

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NM

1.00

1.00

0.01

0.01

0.01

1.00

1.00

0.01

0.01

0.01

NV

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NY

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

OH

0.71

0.65

0.04

0.02

0.00

0.73

0.59

0.04

0.02

0.00

OK

27.70

4.78

1.51

1.51

0.00

18.72

4.32

0.11

0.11

0.00

OR

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

PA

4.20

4.11

1.79

1.79

0.00

4.21

4.11

1.79

1.79

0.00

RI

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

SC

0.61

1.15

1.76

1.76

0.00

0.54

1.26

1.76

1.76

0.00

SD

1.08

1.21

0.00

0.00

0.00

1.08

1.22

0.00

0.00

0.00

TN

20.55

1.57

0.00

0.00

0.00

13.24

1.57

0.00

0.00

0.00

TXLA3

1.25

1.46

1.82

1.77

0.24

1.21

1.92

1.76

1.71

0.24

UT

0.95

0.06

0.00

0.00

0.00

0.95

0.06

0.00

0.00

0.00

VA

0.11

0.08

0.00

0.00

0.00

0.11

0.06

0.00

0.00

0.00

VT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WI

3.37

3.77

0.00

0.00

0.00

3.37

0.95

0.01

0.01

0.00

WV

1.31

1.49

0.31

0.31

0.00

1.29

1.50

0.31

0.31

0.00

WY

1.22

0.97

0.83

0.47

0.13

1.22

0.90

0.85

0.45

0.12

Note: Emissions of Louisiana are less 10 tpy in the original source apportionment modeling. Air quality impacts and
emissions from Texas and Louisiana were combined.

3 TXLA: Louisiana and Texas

B-34


-------
Table B-14 Baseline and Alternative 1 Primary PM2.5 Scaling Factors for Coal EGU
Tags

Baseline	Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

1.20

1.31

1.43

1.33

0.14

1.18

1.31

1.38

1.38

0.00

AR

20.02

7.10

3.14

0.08

0.03

18.77

3.64

0.11

0.05

0.01

AZ

0.38

1.17

0.61

0.18

0.16

0.36

1.20

0.77

0.77

0.00

CA

0.24

0.36

0.16

0.13

0.00

0.24

0.43

0.01

0.01

0.00

CO

13.37

1.19

0.51

0.54

0.51

15.57

1.19

0.51

0.54

0.51

CT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DC

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DE

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

FL

1.40

1.84

0.25

0.25

0.13

1.40

2.10

0.17

0.17

0.02

GA

0.03

0.06

0.00

0.00

0.00

0.13

0.16

0.00

0.00

0.00

IA

1.17

1.14

0.67

0.28

0.19

1.16

0.79

0.04

0.00

0.00

ID

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

IL

1.17

0.95

0.57

0.03

0.00

1.16

0.96

0.98

0.98

0.00

IN

1.28

0.60

0.20

0.20

0.00

1.17

0.62

0.20

0.20

0.00

KY

1.30

1.19

0.77

0.36

0.16

1.25

1.17

0.50

0.50

0.00

MA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MD

3.54

3.54

3.54

3.54

2.97

3.54

3.57

3.57

3.57

0.00

ME

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MI

0.83

0.00

0.00

0.00

0.00

0.83

0.00

0.00

0.00

0.00

MN

3.50

2.70

0.00

0.00

0.00

3.47

2.73

0.00

0.00

0.00

MO

3.04

1.33

0.54

0.11

0.10

2.85

1.17

0.51

0.51

0.00

MS

4.02

3.60

1.06

1.00

1.00

1.94

3.60

8.15

8.15

0.00

MT

0.98

0.98

0.98

0.98

0.38

0.98

0.98

0.99

0.99

0.00

NC

21.57

17.32

6.08

6.14

6.26

18.03

16.39

0.08

0.08

0.08

ND

0.94

0.98

0.78

0.72

0.04

0.92

0.95

0.86

0.86

0.04

NEKS3

3.70

3.68

0.80

0.50

0.15

3.03

3.03

0.35

0.33

0.01

NH

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NJ

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NM

0.98

0.99

0.01

0.01

0.01

0.98

0.99

0.01

0.01

0.01

NV

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NY

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

OH

0.83

1.08

0.19

0.00

0.00

0.71

1.10

0.21

0.21

0.00

OK

14.75

8.14

8.94

8.94

1.00

10.35

4.39

0.55

0.55

0.00

OR

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

PA

3.12

3.04

2.28

1.14

1.14

3.12

2.89

0.91

0.91

0.00

RI

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

SC

1.03

2.17

3.78

3.78

0.00

0.76

2.10

3.78

3.78

0.00

B-35


-------
Baseline

Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

SD

0.93

1.11

0.00

0.00

0.00

0.91

1.11

0.00

0.00

0.00

TN

16.88

1.00

0.00

0.00

0.00

13.75

1.10

0.00

0.00

0.00

TXLAb

1.10

1.30

1.15

0.65

0.14

0.88

0.88

0.92

0.92

0.02

UT

2.92

0.06

0.06

0.06

0.04

2.86

0.06

0.00

0.00

0.00

VA

0.46

0.29

0.08

0.00

0.00

0.46

0.29

0.00

0.00

0.00

VT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WI

2.11

2.36

0.46

0.00

0.00

2.04

0.75

0.01

0.01

0.00

WV

1.29

1.45

1.23

0.56

0.06

1.29

1.50

0.53

0.53

0.00

WY

1.03

1.10

1.08

0.54

0.44

0.96

1.01

0.86

0.42

0.20

Note: Emissions of Louisiana and Kansas are less 10 tpy in the original source apportionment modeling. Air quality
impacts and emissions from those states were combined with nearby states.

3NEKS: Nebraska and Kansas
b TXLA: Louisiana and Texas

Table B-15 Alternative 2 and Final Rules Primary PM2.5 Scaling Factors for Coal EGU
Tags	

Alternative 2	Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

1.19

1.31

1.38

1.38

0.00

1.19

1.31

1.38

1.38

0.00

AR

18.82

7.05

0.11

0.05

0.01

18.31

3.64

0.11

0.05

0.01

AZ

0.36

1.17

0.78

0.78

0.00

0.36

1.20

0.77

0.77

0.00

CA

0.24

0.36

0.01

0.01

0.00

0.24

0.43

0.01

0.01

0.00

CO

15.10

1.19

0.51

0.54

0.51

15.92

1.19

0.53

0.53

0.51

CT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DC

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

DE

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

FL

1.40

1.84

0.17

0.17

0.02

1.40

2.09

0.17

0.17

0.02

GA

0.02

0.05

0.00

0.00

0.00

0.03

0.06

0.00

0.00

0.00

IA

1.16

1.13

0.12

0.00

0.00

1.16

0.79

0.04

0.00

0.00

ID

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

IL

1.16

0.96

0.98

0.98

0.00

1.16

0.96

0.98

0.98

0.00

IN

1.22

0.60

0.20

0.20

0.00

1.17

0.62

0.20

0.20

0.00

KY

1.23

1.11

0.49

0.49

0.00

1.20

1.16

0.50

0.50

0.00

MA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MD

3.54

3.54

3.54

3.54

0.00

3.54

3.57

3.57

3.57

0.00

ME

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

MI

0.83

0.00

0.00

0.00

0.00

0.83

0.00

0.00

0.00

0.00

MN

3.48

2.66

0.00

0.00

0.00

3.44

2.75

0.00

0.00

0.00

MO

2.85

1.23

0.49

0.49

0.01

2.85

1.16

0.51

0.51

0.00

B-36


-------
Alternative 2

Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

MS

1.94

3.60

8.15

8.15

0.00

1.94

3.60

8.15

8.15

0.00

MT

0.98

0.98

0.99

0.99

0.00

0.98

0.98

0.99

0.99

0.00

NC

18.20

15.92

0.08

0.08

0.08

19.37

16.16

0.08

0.08

0.08

ND

0.92

0.97

0.77

0.77

0.04

0.94

0.94

0.86

0.86

0.04

NEKS3

2.99

3.02

0.36

0.33

0.01

2.89

2.92

0.35

0.33

0.01

NH

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NJ

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NM

0.98

0.99

0.01

0.01

0.01

0.98

0.99

0.01

0.01

0.01

NV

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

NY

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

OH

0.76

1.05

0.21

0.21

0.00

0.74

1.05

0.21

0.21

0.00

OK

11.88

8.19

7.94

7.94

0.00

6.11

4.39

0.55

0.55

0.00

OR

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

PA

3.06

2.92

0.91

0.91

0.00

2.97

2.93

0.91

0.91

0.00

RI

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

SC

0.86

2.06

3.78

3.78

0.00

0.76

2.10

3.78

3.78

0.00

SD

0.91

1.06

0.00

0.00

0.00

0.91

1.07

0.00

0.00

0.00

TN

16.88

1.00

0.00

0.00

0.00

11.76

1.00

0.00

0.00

0.00

TXLAb

0.96

1.02

1.20

1.20

0.03

0.88

0.88

0.92

0.92

0.02

UT

2.86

0.06

0.00

0.00

0.00

2.86

0.06

0.00

0.00

0.00

VA

0.46

0.33

0.00

0.00

0.00

0.46

0.23

0.00

0.00

0.00

VT

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WA

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

WI

2.04

2.30

0.00

0.00

0.00

2.04

0.75

0.01

0.01

0.00

WV

1.29

1.47

0.52

0.52

0.00

1.26

1.49

0.53

0.53

0.00

WY

0.96

1.09

0.87

0.46

0.24

0.97

1.02

0.86

0.42

0.20

Note: Emissions of Louisiana and Kansas are less 10 tpy in the original source apportionment modeling. Air quality
impacts and emissions from those states were combined with nearby states.

3NEKS: Nebraska and Kansas
b TXLA: Louisiana and Texas

Table B-16 Baseline and Alternative 1 Primary PM2.5 Scaling Factors for Natural Gas
EGU Tags

Baseline	Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

0.85

0.84

0.71

0.46

0.31

0.85

0.82

0.73

0.47

0.34

AR

0.63

0.82

0.43

0.10

0.07

0.66

0.86

0.47

0.10

0.07

AZ

0.70

0.85

0.86

0.74

0.39

0.69

0.82

0.92

0.73

0.34

CA

0.96

1.06

0.98

0.77

0.20

0.97

1.08

1.01

0.77

0.20

B-37


-------
Baseline

Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

CO

1.23

0.74

0.77

0.75

0.32

1.21

0.75

0.88

0.67

0.39

CT

0.78

0.67

0.60

0.00

0.00

0.77

0.68

0.57

0.00

0.00

DC

0.15

0.13

0.11

0.10

0.08

0.15

0.13

0.07

0.07

0.07

DE

0.62

0.64

0.31

0.27

0.30

0.65

0.65

0.47

0.26

0.29

FL

0.97

0.98

0.95

0.77

0.55

0.97

0.98

0.94

0.78

0.54

GA

0.84

0.81

0.72

0.41

0.30

0.84

0.80

0.74

0.43

0.32

IA

0.50

0.48

0.20

0.06

0.01

0.52

0.65

0.38

0.07

0.02

ID

1.22

1.65

1.68

1.49

0.76

1.39

1.84

1.42

1.35

0.88

IL

0.49

0.55

0.28

0.13

0.00

0.50

0.61

0.37

0.14

0.00

IN

0.67

0.67

0.44

0.15

0.10

0.68

0.68

0.50

0.16

0.11

KS

1.11

1.01

0.19

0.08

0.04

1.10

1.09

0.45

0.33

0.31

KY

0.75

0.72

0.49

0.34

0.18

0.79

0.77

0.44

0.34

0.21

LA

0.79

0.80

0.64

0.29

0.19

0.78

0.79

0.68

0.30

0.19

MA

0.48

0.46

0.34

0.28

0.19

0.47

0.46

0.34

0.28

0.18

MD

1.05

1.08

0.85

0.63

0.61

1.05

1.04

0.86

0.54

0.59

ME

1.75

1.44

0.51

0.50

0.29

1.68

1.43

0.74

0.51

0.29

MI

0.75

0.87

0.63

0.48

0.28

0.79

0.86

0.68

0.46

0.27

MN

0.57

0.52

0.21

0.08

0.05

0.57

0.61

0.21

0.08

0.03

MO

0.30

0.33

0.10

0.03

0.01

0.27

0.44

0.18

0.02

0.01

MS

0.88

0.84

0.51

0.32

0.18

0.88

0.83

0.60

0.32

0.18

MT

0.17

0.21

0.03

0.03

0.00

0.17

0.18

0.47

0.03

0.00

NC

0.87

0.70

0.76

0.60

0.55

0.89

0.73

0.77

0.67

0.64

ND

0.47

0.92

0.19

0.43

0.06

0.46

1.02

0.68

0.43

0.06

NE

2.35

2.21

0.30

0.78

0.01

2.36

2.66

2.64

2.31

1.49

NH

0.59

0.43

0.31

0.27

0.25

0.57

0.42

0.36

0.27

0.25

NJ

0.82

0.84

0.52

0.40

0.42

0.82

0.85

0.67

0.41

0.41

NM

0.52

0.52

0.89

0.99

0.86

0.52

0.52

0.83

0.87

0.99

NV

0.72

0.84

0.83

0.85

0.36

0.70

0.80

0.87

0.84

0.37

NY

0.86

0.85

0.59

0.26

0.27

0.87

0.85

0.60

0.26

0.27

OH

0.95

0.95

0.89

0.63

0.42

0.96

0.95

0.85

0.59

0.41

OK

1.00

0.79

0.22

0.07

0.06

0.93

0.89

0.30

0.07

0.07

OR

3.29

0.74

0.39

0.00

0.00

3.24

0.72

0.37

0.00

0.00

PA

0.83

0.80

0.60

0.37

0.33

0.85

0.81

0.63

0.37

0.33

RI

0.83

0.78

0.65

0.38

0.35

0.83

0.79

0.61

0.38

0.33

SC

0.80

0.86

0.64

0.51

0.53

0.81

0.90

0.67

0.52

0.54

SD

0.73

0.73

0.25

0.13

0.11

0.70

0.89

0.41

0.26

0.02

TN

1.08

1.05

0.88

0.46

0.26

1.10

1.05

0.87

0.47

0.26

TX

0.90

0.83

0.45

0.19

0.09

0.81

0.77

0.50

0.20

0.10

UT

0.66

0.87

0.84

0.88

0.69

0.66

0.84

0.79

0.85

0.65

VA

0.81

0.73

0.47

0.26

0.17

0.82

0.72

0.60

0.27

0.18

VT

0.00

0.00

0.03

0.03

0.00

0.00

0.00

0.03

0.03

0.00

B-38


-------
Baseline	Alternative 1

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

WA

0.44

0.48

0.58

0.59

0.39

0.44

0.48

0.56

0.61

0.39

WI

0.56

0.66

0.43

0.18

0.08

0.55

0.78

0.52

0.19

0.11

WV

0.51

0.38

0.10

0.12

0.09

0.50

0.38

0.33

0.10

0.12

WY

0.01

0.04

0.03

0.03

0.00

0.01

0.01

1.32

0.49

0.00

Table B-17 Alternative 2 and Final Rules Primary PM2.5 Scaling Factors for Natural Gas
EGU Tags	

Alternative 2	Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

AL

0.85

0.84

0.73

0.47

0.34

0.85

0.82

0.74

0.47

0.34

AR

0.66

0.82

0.47

0.10

0.07

0.68

0.87

0.48

0.10

0.08

AZ

0.69

0.81

0.92

0.74

0.34

0.70

0.84

0.93

0.76

0.34

CA

0.97

1.07

1.01

0.77

0.20

0.97

1.08

1.07

0.78

0.20

CO

1.22

0.76

0.90

0.67

0.39

1.06

0.75

0.86

0.67

0.36

CT

0.76

0.67

0.57

0.00

0.00

0.77

0.69

0.57

0.00

0.00

DC

0.15

0.13

0.07

0.07

0.07

0.14

0.13

0.06

0.05

0.05

DE

0.65

0.64

0.48

0.26

0.29

0.65

0.66

0.50

0.27

0.29

FL

0.97

0.98

0.94

0.78

0.54

0.97

0.98

0.94

0.79

0.54

GA

0.84

0.80

0.74

0.43

0.32

0.84

0.80

0.74

0.43

0.32

IA

0.52

0.47

0.34

0.07

0.02

0.52

0.65

0.38

0.07

0.02

ID

1.39

1.81

1.44

1.34

0.87

1.35

1.71

1.27

1.06

0.79

IL

0.49

0.56

0.36

0.14

0.00

0.52

0.61

0.37

0.14

0.00

IN

0.67

0.66

0.52

0.16

0.11

0.69

0.70

0.55

0.17

0.11

KS

1.10

1.06

0.38

0.23

0.20

1.05

1.09

0.47

0.33

0.31

KY

0.80

0.76

0.44

0.34

0.20

0.82

0.78

0.40

0.30

0.21

LA

0.79

0.80

0.67

0.30

0.20

0.78

0.79

0.67

0.30

0.20

MA

0.46

0.46

0.34

0.28

0.18

0.47

0.45

0.33

0.29

0.18

MD

1.03

1.04

0.86

0.54

0.59

1.10

1.09

0.84

0.51

0.56

ME

1.68

1.40

0.76

0.51

0.29

1.68

1.42

0.83

0.51

0.29

MI

0.79

0.84

0.68

0.46

0.27

0.79

0.85

0.69

0.46

0.26

MN

0.53

0.48

0.21

0.08

0.03

0.62

0.66

0.21

0.08

0.03

MO

0.26

0.38

0.19

0.02

0.01

0.27

0.44

0.21

0.02

0.01

MS

0.89

0.83

0.59

0.33

0.17

0.88

0.83

0.60

0.33

0.17

MT

0.17

0.17

0.76

0.03

0.00

0.17

0.18

0.35

0.03

0.00

NC

0.90

0.72

0.77

0.67

0.64

0.89

0.74

0.74

0.65

0.64

ND

0.46

0.86

0.75

0.43

0.06

0.56

1.04

0.61

0.43

0.06

NE

2.37

2.23

2.72

2.45

1.65

2.37

2.83

2.69

2.31

1.48

B-39


-------
Alternative 2

Final Rules

State Tag

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

NH

0.57

0.42

0.36

0.27

0.25

0.56

0.42

0.36

0.27

0.25

NJ

0.82

0.84

0.67

0.41

0.41

0.83

0.86

0.68

0.42

0.41

NM

0.52

0.52

0.83

0.86

0.98

0.52

0.52

0.83

0.79

1.00

NV

0.71

0.79

0.87

0.84

0.37

0.69

0.84

0.87

0.85

0.38

NY

0.87

0.84

0.61

0.26

0.27

0.87

0.86

0.61

0.27

0.27

OH

0.94

0.94

0.85

0.59

0.41

0.94

0.93

0.83

0.57

0.40

OK

0.95

0.81

0.30

0.07

0.07

0.93

0.90

0.31

0.07

0.07

OR

3.24

0.73

0.38

0.00

0.00

3.28

0.73

0.38

0.00

0.00

PA

0.85

0.81

0.63

0.38

0.33

0.85

0.81

0.65

0.42

0.33

RI

0.83

0.79

0.61

0.38

0.33

0.83

0.79

0.61

0.38

0.34

SC

0.81

0.89

0.67

0.52

0.54

0.81

0.90

0.67

0.50

0.55

SD

0.70

0.65

0.41

0.25

0.02

0.61

0.89

0.44

0.26

0.02

TN

1.09

1.05

0.87

0.47

0.26

1.11

1.05

0.87

0.48

0.26

TX

0.81

0.75

0.49

0.19

0.10

0.80

0.77

0.50

0.20

0.10

UT

0.66

0.83

0.79

0.84

0.65

0.67

0.83

0.78

0.84

0.65

VA

0.82

0.72

0.60

0.28

0.18

0.82

0.73

0.61

0.30

0.19

VT

0.00

0.00

0.03

0.03

0.00

0.00

0.00

0.03

0.03

0.00

WA

0.44

0.49

0.56

0.61

0.39

0.45

0.48

0.56

0.61

0.39

WI

0.55

0.63

0.52

0.19

0.11

0.57

0.79

0.52

0.19

0.11

WV

0.50

0.38

0.32

0.10

0.12

0.50

0.41

0.40

0.11

0.12

WY

0.01

0.01

1.30

0.41

0.00

0.03

0.02

1.32

0.53

0.00

Table B-18 Baseline and Alternative 1 Scaling Factors for Other EGU Tags

Baseline	Alternative 1

Pollutants

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

Seasonal NOx

1.16

1.16

1.10

1.04

1.03

1.16

1.17

1.11

1.04

1.03

Annual NOx

1.17

1.17

1.11

1.03

1.00

1.17

1.17

1.12

1.03

1.00

Annual SO2

1.00

1.01

1.00

0.90

0.87

0.99

1.01

1.00

0.90

0.87

Annual PM2 5

1.37

1.37

1.32

1.27

1.20

1.38

1.39

1.34

1.28

1.20

Table B-19

Alternative 2 and Final Rules Scaling Factors for Other EGU Tags









Alternative 2









Final Rules





Pollutants

2028

2030

2035

2040

2045

2028

2030

2035

2040

2045

Seasonal NOx

1.16

1.17

1.11

1.04

1.03

1.16

1.17

1.12

1.05

1.03

Annual NOx

1.17

1.17

1.12

1.03

1.00

1.17

1.18

1.12

1.04

1

Annual SO2

0.99

1.01

1.00

0.90

0.87

0.99

1.01

1

0.9

0.87

Annual PM2 5

1.38

1.39

1.34

1.28

1.20

1.41

1.41

1.37

1.3

1.22

B-40


-------
B.4 Air Quality Surface Results

The spatial fields of baseline AS-M03 and Annual Average PM2.5 in 2028, 2030, 2035,
2040 and 2045 are presented in Figure B-8 through Figure B-12. It is important to recognize that
ozone is a secondary pollutant, meaning that it is formed through chemical reactions of precursor
emissions in the atmosphere. As a result of the time necessary for precursors to mix in the
atmosphere and for these reactions to occur, ozone can either be highest at the location of the
precursor emissions or peak at some distance downwind of those emissions sources. The spatial
gradients of ozone depend on a multitude of factors including the spatial patterns of NOx and
VOC emissions and the meteorological conditions on a particular day. Thus, on any individual
day, high ozone concentrations may be found in narrow plumes downwind of specific point
sources, may appear as urban outflow with large concentrations downwind of urban source
locations or may have a more regional signal. However, in general, because the AS-M03 metric
is based on the average of concentrations over more than 180 days in the spring and summer, the
resulting spatial fields are rather smooth without sharp gradients, compared to what might be
expected when looking at the spatial patterns of MDA8 ozone concentrations on specific high
ozone episode days. PM2.5 is made up of both primary and secondary components. Secondary
PM2.5 species sulfate and nitrate often demonstrate regional signals without large local gradients
while primary PM2.5 components often have heterogenous spatial patterns with larger gradients
near emissions sources. Both secondary and primary PM2.5 contribute to the spatial patterns
shown in Figure B-13 through Figure B-17 as demonstrated by the extensive areas of elevated
concentrations over much of the Eastern U.S. which have large secondary components and
hotspots in urban areas which are impacted by primary PM emissions.

Figure B-8 through Figure B-17 also present the model-predicted air quality changes
between the baseline and the three illustrative scenarios in 2028, 2030, 2035, 2040 and 2045 for
AS-M03 and PM2.5. Difference in these figures are calculated as the illustrative scenario minus
the baseline. The spatial patterns shown in the figures are a result of (1) of the spatial distribution
of EGU sources that are predicted to have changes in emissions and (2) of the physical or
chemical processing that the model simulates in the atmosphere. While SO2, NOx, and primary
PM2.5 emissions changes all contributed to the PM2.5 changes depicted in Figure B-13 through

B-41


-------
Figure B-17, the PM2.5 component species with the largest changes on average was sulfate and
consequently the SO2 emissions changes have the largest impact on predicted changes in PM2.5
concentrations in most locations through sulfate, ammonium and particle-bound water impacts.
The spatial fields used to create these maps serve as an input to the benefits analysis and the
environmental justice analysis.

Figure B-18 though Figure B-21 show changes in AS-M03 in 2030, 2035, 2040, and
2045 relative to 2028 baseline conditions. Figure B-22 through Figure B-25 show changes in
PM2.5 in 2030, 2035, 2040, and 2045 relative to 2028 baseline conditions. Relative to 2028
baseline conditions, these figures indicate that ozone and PM2.5 concentration will decline in
virtually all areas of the country for both baseline and final rules scenarios in each further out
snapshot year. However, some areas of the country may experience slower or faster rates of
decline in ozone and PM2.5 over time as a result of the modeled changes resulting from this rule.
Our comparison of air quality conditions with and without the rule suggests that for all snapshot
years the final rules will result in widespread reductions in PM2.5 concentrations. In all years, the
final rules are expected to result in reductions in ozone concentrations over many areas of the
US, although some areas may experience increases in ozone concentrations relative to forecasted
conditions without the rule. The extent of areas experiencing ozone increases varies among
snapshot years.

B-42


-------
2028 Baseline MDA8 O,

2028 MDA8 03 Policy- Baseline:
Alternative 2

1/ ~"7r	^	A

r' ***

75 ppb

I

25 ppb

0.2 ppb



2028 MDA8 03 Policy- Baseline:
Alternative 1



2028 MDA8 03 Policy- Baseline:
Final Rules

II ~fi—	 „	n

* »

+Jt > \

V i



0.2 ppb

I

f

-0.2 ppb

0.2 ppb

¦0.2 ppb

-0.2 ppb

Figure B-8 Maps of ASM-03 in 2028

Note: Baseline ozone concentrations (ppb) shown in upper left. Change in ozone in the alternative 1 scenario
compared to baseline values (ppb) shown in upper right. Change in ozone in the alternative 2 scenario compared to
baseline values (ppb) shown in lower left. Change in ozone in the final rules scenario compared to baseline values
shown in lower right (ppb).

B-43


-------
2030 Baseline MDA8 O,

2030 MDA8 03 Policy- Baseline:
Alternative 2

1/ ~)7 - • —	n

mkm

i

J

75 ppb

25 ppb

I

..

I

-0.2 ppb

2030 MDA8 03 Policy- Baseline:
Alternative 1

/

2030 IV1DA8 03 Policy- Baseline:

Final Rules

tf T[ —r——

1



: %
*





¥

Hi'I

rrr5\/

r

uir^ u

\ i



y

» m ih

I

0.2 ppb

I

I

-0.2 ppb

-0.2 ppb

Figure B-9 Maps of ASM-03 in 2030

Note: Baseline ozone concentrations (ppb) shown in upper left. Change in ozone in the alternative 1 scenario
compared to baseline values (ppb) shown in upper right. Change in ozone in the alternative 2 scenario compared to
baseline values (ppb) shown in lower left. Change in ozone in the final rules scenario compared to baseline values
shown in lower right (ppb).

B-44


-------
2035 Baseline MDA8 O,

2035 MDA8 03 Policy- Baseline:
Alternative 2

\* It

75 ppb



25 ppb

0.2 ppb

2035 MDA8 03 Policy- Baseline:
Alternative 1

2035 MDA8 03 Policy- Baseline:
Final Rules

I

0.2 ppb

-0.2 ppb

I

0.2 ppb

-0.2 ppb

V\

\ 1
v

-0.2 ppb

Figure B-10 Maps of ASM-03 in 2035

Note: Baseline ozone concentrations (ppb) shown in upper left. Change in ozone in the alternative 1 scenario
compared to baseline values (ppb) shown in upper right. Change in ozone in the alternative 2 scenario compared to
baseline values (ppb) shown in lower left. Change in ozone in the final rules scenario compared to baseline values
shown in lower right (ppb).

B-45


-------
2040 Baseline MDA8 0,

2040 MDA8 03 Policy-

Baseline:

Alternative 2



If K k-r

-Jl JT|



"~r \ *> yj

f r\4/i"

r"{ J

zZfc j

*

i, j.

t « T» ~m

\ j

75 ppb

25 ppb

0.2 ppb

2040 MDA8 03 Policy-

Baseline:

Alternative 1



1? Tf ^

/S\ jjLj ;-•••



%k



"¦I!

) I * *,/



X7l

-A

k Y J —J	J-"

> /

r '

V \ J

i

Xt

i -jf

YV

1 M 1» »

ia» »J

2040 MDA8 03 Policy- Baseline:
Final Rules

il	..



*

%

I

0.2 ppb

I

-0.2 ppb

0.2 ppb

/

-0.2 ppb	„	,* 1 »	' ta m "0.2 ppb

Figure B-ll Maps of ASM-03 in 2040

Note: Baseline ozone concentrations (ppb) shown in upper left. Change in ozone in the alternative 1 scenario
compared to baseline values (ppb) shown in upper right. Change in ozone in the alternative 2 scenario compared to
baseline values (ppb) shown in lower left. Change in ozone in the final rules scenario compared to baseline values
shown in lower right (ppb).

B-46


-------
2045 Baseline MDA8 O,

2045 MDA8 03 Policy- Baseline:
Alternative 2

**!

' J

V tSX

' i ?



75 ppb



25 ppb

0.2 ppb



2045 MDA8 03 Policy- Baseline:
Alternative 1

K

)



/

f

#

\

2045 MDA8 03 Policy- Baseline:
Final Rules



a



i

} j

i , ¦ JE

'

0.2 ppb





-0.2 ppb

0.2 ppb

-0.2 ppb ij	 ^	^ 	^-T	J ¦ -0.2 ppb

Figure B-12 Maps of ASM-OS in 2045

Note: Baseline ozone concentrations (ppb) shown in upper left. Change in ozone in the alternative 1 scenario
compared to baseline values (ppb) shown in upper right. Change in ozone in the alternative 2 scenario compared to
baseline values (ppb) shown in lower left. Change in ozone in the final rules scenario compared to baseline values
shown in lower right (ppb).

B-47


-------
2028 Baseline PM2.5

I'lawS

I

LivB

«lfa

¦a

' A

/

¦ mm -jC1,'".

1 ff w

» 20 n* w

2028 PM2.5 Policy-Baseline:
Alternative 2

/ / J I j\



SIM

pr	i_ -J-" / ^



7 : (

j- ' ¦ \

I

15ng/m3

0.1 ng/m3

2028 PM2 5 Policy- Baseline:
Alternative 1

r

> J,

2028 PM25 Policy-Baseline:
Final Rules

I? Tr-—

I

w

t

\\

\ 1
V

I

0.1 |ig/m3

-0.1 ng/m3

I

0.1 ng/m3

-O.lng/m3	„	„ ' J " -0.1 ng/m3

Figure B-13 Maps of PM2.5 in 2028

Note: Baseline PM2.5 concentrations (jig/m3) shown in upper left. Cliange InPMis in the alternative 1 scenario
compared to baseline values (jig/m3) shown in upper right. Cliange in PM2.5 in the alternative 2 scenario compared
to baseline values (iig/m3) shown in lower left. Cliange in PM2.5 in the final rales scenario compared to baseline
values shown in lower right (ng/m3).

B-48


-------
2030 Baseline PM

2.5

2030 PM2.5 Policy- Baseline:
Alternative 2

4

y

*vA

15 ng/m3

I

0.1 ng/m3

2030 PM2.5 Policy-Baseline:
Alternative 1

0.1 ng/m3





W

| 1

^ %

n

,	f \ \ xf

/¦#
75#

Z^J

\f

L LU

\ tv
\ \

2030 PM2,5 Policy-Baseline:
Final Rules

I

I

¦0.1 ng/m3

0.1 ng/m3

¦0.1 ng/m3	—J ¦ -0.1ng/m3

Figure B-14 Maps of PM2.5 in 2030

Note: Baseline PM25 concentrations (jig/m3) shown in upper left. Cliange U1PM2 5 in the alternative 1 scenario
compared to baseline values (jig/m3) shown in upper right. Cliange in PM2.5 in the alternative 2 scenario compared
to baseline values (iig/m3) shown in lower left. Cliange in PM2 5 in the final rales scenario compared to baseline
values shown in lower right (ng/m3).

B-49


-------
2035 Baseline PM

2.5

2035 PM2.5 Policy-Baseline:
Alternative 2

I? "Tr-—

w

15 ng/m3

0.1 ng/m3

2035 PM2.5 Policy- Baseline:
Alternative 1

2035 PM25 Policy-Baseline:
Final Rules

I

0.1 ng/m3

I

-0.1 ng/m3

0.1 ng/m3

I

-0.1 ng/m3 11

Figure B-15 Maps of PM2.5 in 2035

Note: Baseline PM25 concentrations (jig/m3) shown in upper left. Change U1PM2 5 in the alternative 1 scenario
compared to baseline values (ng/m3) shown in upper right. Change in PM2.5 in the alternative 2 scenario compared
to baseline values (iig/m3) shown in lower left. Change in PM2 5 in the final rales scenario compared to baseline
values shown in lower right (ng/m3).

0.1 ng/m3

B-50


-------
15ng/m3

0.1 ng/m3

-0.1 ng/m3

0.1 ng/m3

-0.1 ng/m3

0.1 ng/m3

-0.1 ng/m3

2040 PM2.5 Policy-Baseline:
Alternative 1

2040 Baseline PM25

2040 PM25 Policy-Baseline:
Alternative 2

2040 PM25 Policy-Baseline:
Final Rules

Figure B-16 Maps of PM2.5 in 2040

Note: Baseline PM.: s concentrations (ng/m3) shown in upper left. Cliange in PM2.5 in the alternative 1 scenario
compared to baseline values (^g/m3) shown in upper right. Cliange in PM2 5 in the alternative 2 scenario compared
to baseline values (|ig/m3) shown in lower left. Cliange in PM2 5 in the final rules scenario compared to baseline
values shown in lower right (|ig/m3).

B-51


-------
2045 Baseline PM

2.5

2045 PM2.s Policy- Baseline:
Alternative 2

11

%

f

•J:

15 ng/m3

I

0.1 ng/n

2045 PM2.5 Policy-Baseline:

Alternative 1



I \ * 		

L it ^	



y* \ 1 j I ^ . 1

n\

ZC, J

N —LJ C J k I )
m m m

4

I

0.1 ng/m3

2045 PM25 Policy-Baseline:
Final Rules





C / 1"" rfj*1 i



\a

1 a a» x*

1 \

\\

\ 1

a* w

I

I

-0.1 ng/m3

0.1 ng/m3



-0.1ng/m3 J	 . 	„			^-	J " -0.1 ng/m3

Figure B-l 7 Maps of PM2.5 in 2045

Note: Baseline PM2.5 concentrations (jig/m3) shown in upper left. Cliange InPMis in the alternative 1 scenario
compared to baseline values (jig/m3) shown in upper right. Cliange in PM2.5 in the alternative 2 scenario compared
to baseline values (iig/m3) shown in lower left. Cliange in PM2.5 in the final rales scenario compared to baseline
values shown in lower right (ng/m3).

B-52


-------
Apr-Sep MDA8 Ozone
2030 Baseline- 2028 Baseline

Apr-Sep MDA8 Ozone
2030 Final Rules-2028 Baseline

Mln= -2.S60 at (127,144). Max = 1.103 at (331,86)

n = -2.858 at (206.S5), Max =1.202 at (331,86)

Figure B-18 Maps of changes in 2030 ASM-03 from 2028 baseline conditions

Note: Baseline 2030 ozone concentrations compared to 2028 baseline ozone concentrations (ppb) shown on left.
Final Rules 2030 ozone concentrations compared to 2028 baseline ozone concentrations (ppb) shown on right. Color
bars for Figure B-18 through Figure B-21 differ in scale (±2ppb) from color bars used in Figure B-8 through Figure
13-12 (±0.2ppb).

Figure B-19 Maps of changes in 2035 ASM-03 from 2028 baseline conditions

B-53


-------
Note: Baseline 2035 ozone concentrations compared to 2028 baseline ozone concentrations (ppb) shown on left.
Final Rules 2035 ozone concentrations compared to 2028 baseline ozone concentrations (ppb) shown on right. Color
bars for Figure B-18 through Figure B-21 differ in scale (±2ppb) from color bars used in Figure B-8 through Figure
B-12 (±0.2ppb).

Apr-Sep MDA8 Ozone
2040 Baseline - 2028 Baseline

/ \

/ /V

V \ I , I *,.

\ 1

V 7s—

f. .

11

r\ f

\ \ 'J



Apr-Sep MDA8 Ozone
2040 Final Rules-2028 Baseline

rf ~~n

Wl



u

Tka

lT

J*

Mr-IT

"A

V

*\ L



\



Min =-6.222 at (20«.S5), Max = 1 651 at (331,86)

Min = -6.473 at (316.101), Max = 1 652 at (331,86)

Figure B-20 Maps of changes in 2040 ASM-03 from 2028 baseline conditions

Note: Baseline 2040 ozone concentrations compared to 2028 baseline ozone concentrations (ppb) shown on left.
Final Rules 2040 ozone concentrations compared to 2028 baseline ozone concentrations (ppb) shown on right. Color
bars for Figure B-18 through Figure B-21 differ in scale (±2ppb) from color bars used in Figure B-8 through Figure
B-12 (±0.2ppb).

B-54


-------
Apr-Sep MDA8 Ozone
2045 Baseline - 2028 Baseline

-dk.

' . -* f. . • tf

,

I

Apr-Sep MDA8 Ozone
2045 Final Rules-2028 Baseline

I

1-1.20
~
-2.00

PPb

lvtir> = -7.321 at (206,55), Ma* = OOOEfO al (1,1)

Vlln = -8.129 at (206.S5), Max = 0 OOE»0 at (1,1)

Figure B-21 Maps of changes in 2045 ASM-03 from 2028 baseline conditions

Note: Baseline 2045 ozone concentrations compared to 2028 baseline ozone concentrations (ppb) shown on left.
Final Rules 2045 ozone concentrations compared to 2028 baseline ozone concentrations (ppb) shown on right. Color
bars for Figure B-18 through Figure B-21 differ in scale (±2ppb) from color bars used in Figure B-8 through Figure
B-12 (±0.2ppb).

Annual Average PM2 5
2030 Baseline- 2028 Baseline

Annual Average PM2 5
2030 Final Rules-2028 Baseline

- -0 376 at (254,98), Ma* = 0 110 at (109.89)

Mln = -0 409 at (254,98), Ma* = 0 112 at (109,89)

Figure B-22 Maps of changes in 2030 PM2.5 from 2028 baseline conditions

B-55


-------
Note: Baseline 2030 PM2.5 concentrations compared to 2028 baseline PM2.5 concentrations (j.ig/m3) shown on left.
Final Rules 2030 PM2.5 concentrations compared to 2028 baseline PM2 5 concentrations ((.ig/1113) shown on right.
Color bars for Figure B-22 through Figure B-25 (+0.3 |ig/m3) differ from color bars used in Figure B-13 through
Figure B-17 (±0.1 (xg/m3)

Annual Average PM2.5
2035 Baseline - 2028 Baseline

Annual Average PM2 5
2035 Final Rules-2028 Baseline

Figure B-23 Maps of changes in 2035 PM2.5 from 2028 baseline conditions

Note: Baseline 2035 PM2.5 concentrations compared to 2028 baseline PM2.5 concentrations (j.ig/m3) shown on left.
Final Rules 2035 PM2.5 concentrations compared to 2028 baseline PM2 5 concentrations (pg/'m3) shown on right.
Color bars for Figure B-22 through Figure B-25 (+0.3 |ig/m3) differ from color bars used in Figure B-13 through
Figure B-17 (±0.1 (xg/m3)

B-56


-------
Annual Average PM2 5
2040 Baseline- 2028 Baseline

Annual Average PM2 5
2040 Final Rules-2028 Baseline

Figure B-24 Maps of changes in 2040 PM2.5 from 2028 baseline conditions

Note: Baseline 2040 PM2.5 concentrations compared to 2028 baseline Pftfc concentrations (ng/m3) shown on left.
Final Rules 2040 P.M2.5 concentrations compared to 2028 baseline PIVh 5 concentrations (ng/m3) shown on right.
Color bars for Figure B-22 through Figure B-25 (±0.3 ng/m3) differ from color bars used in Figure B-13 tlirough
Figure B-17 (±0.1 (xg/m3)

Annual Average PM25
2045 Final Rules-2028 Baseline

Annual Average PM2 5
2045 Baseline- 2028 Baseline

Figure B-25 Maps of changes in 2045 PM2.5 from 2028 baseline conditions

B-57


-------
Note: Baseline 2045 PM2 5 concentrations compared to 2028 baseline PM2.5 concentrations (ng/m3) shown on left.
Final Rules 2045 PM2 5 concentrations compared to 2028 baseline PM2.5 concentrations (ng/m3) shown on right.
Color bars for Figure B-22 through Figure B-25 (±0.3 ng/m3) differ from color bars used in Figure B-13 through
Figure B-17 (±0.1 ng/m3)

B.5 Uncertainties and Limitations of the Air Quality Methodology

One limitation of the scaling methodology for creating ozone and PM2.5 surfaces
associated with the baseline or illustrative scenarios described above is that the methodology
treats air quality changes from the tagged sources as linear and additive. It therefore does not
account for nonlinear atmospheric chemistry and does not account for interactions between
emissions of different pollutants and between emissions from different tagged sources. The
method applied in this analysis is consistent with how air quality estimations have been made in
several prior regulatory analyses (U.S. EPA, 2012, 2019, 2020a). We note that air quality is
calculated in the same manner for the baseline and for the illustrative scenarios, so any
uncertainties associated with these assumptions is propagated through results for both the
baseline and the illustrative scenarios in the same manner. In addition, emissions changes
between baseline and illustrative scenarios are relatively small compared to modeled future year
emissions that form the basis of the source apportionment approach described in this appendix.
Previous studies have shown that air pollutant concentrations generally respond linearly to small
emissions changes of up to 30 percent (Cohan et al., 2005; Cohan and Napelenok, 2011; Dunker
et al., 2002; Koo et al., 2007; Napelenok et al., 2006; Zavala et al., 2009). A second limitation is
that the source apportionment contributions are informed by the spatial and temporal distribution
of the emissions from each source tag as they occur in the future year modeled case. Thus, the
contribution modeling results do not allow us to consider the effects of any changes to spatial
distribution of EGU emissions within a state-fuel tag between the future year modeled case and
the baseline and illustrative scenarios analyzed in this RIA. Finally, the future year CAMx-
modeled concentrations themselves have some uncertainty. While all models have some level of
inherent uncertainty in their formulation and inputs, the base-year 2016 model outputs have been
evaluated against ambient measurements and have been shown to adequately reproduce spatially
and temporally varying concentrations (U.S. EPA, 2023a, 2024).

B-58


-------
B.6 References

Cohan, D. S., Hakami, A., Hu, Y., & Russell, A. G. (2005). Nonlinear Response of Ozone to
Emissions: Source Apportionment and Sensitivity Analysis. Environmental Science &
Technology, 39(17), 6739-6748. doi: 10.1021/es048664m

Cohan, D. S., & Napelenok, S. L. (2011). Air Quality Response Modeling for Decision Support.
Atmosphere, 2(3), 407-425. Retrieved from https://www.mdpi.eom/2073-4433/2/3/407

Ding, D., Zhu, Y., Jang, C., Lin, C.-J., Wang, S., Fu, J., . . . Qiu, X. (2016). Evaluation of health
benefit using BenMAP-CE with an integrated scheme of model and monitor data during
Guangzhou Asian Games. Journal of Environmental Sciences, 42, 9-18.
doi:https://doi.org/10.1016/i.ies.2015.06.003

Dunker, A. M., Yarwood, G., Ortmann, J. P., & Wilson, G. M. (2002). The Decoupled Direct
Method for Sensitivity Analysis in a Three-Dimensional Air Quality Model
Implementation, Accuracy, and Efficiency. Environmental Science & Technology,
36(13), 2965-2976. doi: 10.1021/esOl 12691

Gold, C. M., Remmele, P. R., & Roos, T. (1997). Voronoi methods in GIS. In M. van Kreveld, J.
Nievergelt, T. Roos, & P. Widmayer (Eds.), Algorithmic Foundations of Geographic
Information Systems (pp. 21-35). Berlin, Heidelberg: Springer Berlin Heidelberg.

Koo, B., Dunker, A. M., & Yarwood, G. (2007). Implementing the Decoupled Direct Method for
Sensitivity Analysis in a Particulate Matter Air Quality Model. Environmental Science &
Technology, ¥7(8), 2847-2854. doi: 10.1021/es0619962

Napelenok, S. L., Cohan, D. S., Hu, Y., & Russell, A. G. (2006). Decoupled direct 3D sensitivity
analysis for particulate matter (DDM-3D/PM). Atmospheric Environment, 40(32), 6112-
6121. doi:https://doi.org/10.1016/i.atmosenv.2006.05.039

Ramboll Environ. (2021). User's Guide Comprehensive Air Quality Model with Extensions
version 7.10. Retrieved from Novato, CA:

U.S. EPA. (2007). Technical Report on Ozone Exposure, Risk, and Impact Assessments for
Vegetation. (EPA 452/R-07-002). Research Triangle Park, NC: Office of Air Quality
Planning and Standards. https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=P 100PVGI.txt

U.S. EPA. (2012). Regulatory Impact Analysis for the Final Revisions to the National Ambient
Air Quality Standards for Particulate Matter. (EPA-452/R-12-005). Research Triangle
Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, Health and Environmental Impact Division.
https://www3.epa.gov/ttnecasl/regdata/RIAs/finalria.pdf

U. S. EPA. (2019). Regulatory Impact Analysis for the Repeal of the Clean Power Plan, and the
Emission Guidelines for Greenhouse Gas Emissions from Existing Electric Utility
Generating Units. (EPA-452/R-19-003). Research Triangle Park, NC: U.S.

B-59


-------
Environmental Protection Agency, Office of Air Quality Planning and Standards, Health
and Environmental Impact Division, https://www.epa.gov/sites/production/files/2Q19-
06/documents/utilities ria final cpp repeal and ace 2019-06.pdf

U.S. EPA. (2020a). Analysis of Potential Costs and Benefits for the National Emission Standards
for Hazardous Air Pollutants: Coal- and Oil-FiredElectric Utility Steam Generating
Units - Subcategory of Certain Existing Electric Utility Steam Generating Units Firing
Eastern Bituminous Coal Refuse for Emissions of Acid Gas Flazardous Air Pollutants.
Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Health and Environmental Impact Division.
https://www.epa.gov/sites/default/files/2020-Q4/documents/mats coal refuse cost-
benefit memo.pdf

U.S. EPA. (2020b). Benefit and Cost Analysis for Revisions to the Effluent Limitations
Guidelines and Standards for the Steam Electric Power Generating Point Source
Category. (EPA-821-R-20-003). Washington DC: U.S. Environmental Protection
Agency. https://www.epa.gov/sites/default/files/202Q-

08/documents/steam electric elg 2020 final reconsideration rule benefit and cost an
alvsis.pdf

U.S. EPA. (2021a). Flat File Generation Methodology: Version: Summer 2021 Reference Case
using EPA Platform v6. U.S. Environmental Protection Agency.
https://www.epa.gOv/svstem/files/documents/2021-09/flat-file-methodologv-epa-
platform-v6-summer-2021 -reference-case.pdf

U.S. EPA. (2021b). Regulatory Impact Analysis for the Final Revised Cross-State Air Pollution
Rule (CSAPR) Update for the 2008 Ozone NAAQS. (EPA-452/R-21-002). Research
Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Health and Environmental Impact Division.
https://www.epa. gov/ sites/default/files/2021 -
03/documents/revised csapr update ria final.pdf

U.S. EPA. (2022a). Regulatory Impact Analysis for Proposed Federal Implementation Plan

Addressing Regional Ozone Transport for the 2015 Ozone National Ambient Air Quality
Standard. (EPA-452/D-22-001). Research Triangle Park, NC: U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Health and
Environmental Impact Division, https://www.epa.gov/svstem/files/documents/2022-
03/transport ria proposal fip 2015 ozone naaqs 2022-02.pdf

U.S. EPA. (2022b). Software for Model Attainment Test - Community Edition (SMAT-CE) User's
Guide Software version 2.1. (EPA-454/B-22-013). Research Triangle Park, NC.
https://www.epa.gov/system/files/documents/2022-

1 l/User%27s%20Manual%20for%20SMAT-CE%202.1 EPA Report 11 30 2022.pdf

U.S. EPA. (2023a). Air Quality Modeling Final Rule Technical Support Document: 2015 Ozone
NAAQS Good Neighbor Plan. Research Triangle Park, NC: U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards.

B-60


-------
https://www.epa.gov/svstem/files/documents/2023-
03/AO%20Modeling%20Final%20Rule%2QT SD. pdf

U.S. EPA. (2023b). Technical Support Document (TSD): Preparation of Emissions Inventories
for the 2016v3 North American Emissions Modeling Platform. (EPA-454/B-23-002).
Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, https://www.epa.gov/svstem/files/documents/2023-
03/2016v3 EmisMod TSD Januarv2023 l.pdf

U.S. EPA. (2024). Air Quality Modeling Technical Support Document: PM2.5 Model Evaluation
for 2016 CAMx Modeling to Support Multiple 2024 EGURulemakings. Research
Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards

Zavala, M., Lei, W., Molina, M. J., & Molina, L. T. (2009). Modeled and observed ozone

sensitivity to mobile-source emissions in Mexico City. Atmos. Chem. Phys., 9( 1), 39-55.
doi: 10.5194/acp-9-3 9-2009

B-61


-------
APPENDIX C: ENVIRONMENTAL JUSTICE ANALYSIS
C.l Exposure Analysis Figures for the Alternative Scenarios

This appendix provides additional figures to complement the analysis in Section 6.

Year Group

—i CC M < O >- UJ
<. < < U u U Q

u.

<

S 9

=i 5

&

£



< Q
mZ S

LU

s

si

o
2

yi

2

2

u
z

Q
Z

LU
Z

1 =

11

>
z

ON
OK

a
o

<
a



a

U"l

Z

P



5

¦s

5

1

A AA
AAA

202S Reference

American Indian
Asian

Black [ mhi
Hispanic -omsi tins
Less educates
Unemployed
Uninsured

Bottom 2S94 hfe expectancy
English < well

-------
2030

Year Group
2028 Rtftr««e»

Arw»ean Incvtm
A*i»n PMt
$\Kk

WiJJJfV £ "d OS	5 li

.t»

wr^mpioyM
Upwnf-^Kl

3-3vtom 25% I ft i*p*ct»!*cy
Srtglisn < well
$lijft < w*ll
h£

Ltil tC .»!»S

Uretmploytd
Uninsuftd

S-snon 29% i *« «ip
snjliifi < w»ll
«cci!ney
£nglijih< vv#«l
>iJea:<-uo2xH>$S;





Figure C-2 Heat Map of the State Average PM2.5 Concentration Reductions (Blue) and
Increases (Red) Due to the Alternative 2 Scenario Across Demographic Groups in 2028,
2030, 2035, 2040, and 2045 (jig/m3)

C-2


-------
Population 2028

2030

2035

2040

2045

& "
a.

Race ¦fc 50^t-
	096.

r

r







r

& ~

Q.

Ethnicity ^ 50*t-
#

r

r







/

100%"

g-

Educational a.
Attainment

r

i

r







I

Employment
Status

r

r







/

insurance a
Status

r

i

r







r

& *
a

Life Expectancy ^

#

r





J

r

/

% '

Linguistic fi. _
Isolation

S

r

r







r

I '

Poverty Status •% S0as-

9

r

r







I

& " "
a.

Redlining -g SCc-
#

r

1

r



J

n

f

1

Tribal Land 5^0-

9





r



I

J

j

¦0.04 0.00 0.04 0.08
PM

•0.04 0.00 0.04 0.08
PM

*0.04 O.OO 0.04 0.08
PM

•0.04 0.00 0.04 0.08
PM

•0.04 0.00 0.04 0.08
PM

I Employed (0-99)

| Unemployed (0-99)

, Not in the tabor force (0-99)
| >Poverty line (0-99)

| wrea i'0-64}

| Uninsured (0-€4)

| More educated {>24. HS or more )
I Less educated {>24, no HS)

| Wh»te (0*99)

American Indian (0-99)

| Asian (0-99)

| Black (0-99)

| Top 75% life expectancy (0-99)
I Bottom 25% life expectancy (0-99)
1 No life expectancy (0-99)

| English well or better (0-99)
| English < well (0*99)

| HOLC Grades A-C (0-99)

| HOLC Grade D (0-99)

I Not Graded by HOLC (0-99)
| Not Tribal land (0-99)

| Tribal land (0-99)

Figure C-3 Distribution of PM2.5 Concentration (jig/m3) Reductions Across Populations,
Future Years for the Alternative 1 Scenario

C-3


-------
Population

Insurance	a.

Status	1

Linguistic	a-

Isolation

Poverty Status -5

#

Redlining	^

#

r

r
r
r
r

r
r
r

r

r
r

Z04S

r

r

f
f

r
r
r

| Employed (0-99)

I Unemployed (0-99)

Not in the labor force (0-99)
| >Poverty lire (0^99)

| 24 HS or more)
I Less educated (>24; no HS)
| White (0-99)

I American Indian (0-99)

| Asian (0-99)

| Black (0-99)

| top 75% life expectancy 10-99
| Bottom 2S% life expectancy (0-S9)
| No life expectancy (0-99)

| English well or better (0-99)
| English < well (0-99)

| HOLC Grades A-C (0-99)

| HOLC Grade 0(0-99)

I Not Graded by HOLC (0-99)
| Not Tribal land (0-99)

| Tribal land (0-99)

•0.04 0.00 0.04 0 08 -0.04 0.00 0.04 0.08 -0.04 0.00 0 04 0.08 -0.04 0.00 0.04 0.08 -0.04 0 00 0.04 0.08

Figure C-4 Distribution of PM2.5 Concentration (fig/m3) Reductions Across Populations,
Future Years for the Alternative 2 Scenario

C-4


-------
Year Group
2028 fteference

American Indian

Asian Qipnvto*)

Black	, -9HHHI

Hispanic -o&b

Less educated

Unemployed

Uninjured

Bottom 25% I ife expectancy
English* well
:
z z z z z :

i sc a. < - v
O O O o. a "•

• < u	^

555555









Figure C-5 Heat Map of the State Average Ozone Concentrations Reductions (Green)
and Increases (Red) Due to the Alternative 1 Scenario Across Demographic Groups in
2028, 2030, 2035, 2040, and 2045 (ppb)

C-5


-------
Year

2028

2040

Group

P.*far«ri«

Am#ne»r In0i»«

Asian Oio«e (ppb)
Slack
Hispanic ^ 59
L«33 K-CltK

Unamptoyad
bnmsgrae

SottC-rr. 2S% Irfa axp*Ct*nCy
E*ghjh < wal i

pacnney

Engl-sh< wan

<°3»arty '>ri»

K>iC<5ra«0

Tribal lane

R*f«r«rc*

Aenancan JnSian

Asian

Black

Hispanic

lass t-.:a*«c

Unamptoyad

Lfninswad

Better 25% I if* a*p*cta r«y
English < wall

bna

HCiCGradaQ
Trifc»i lac.c

:aa

u V Q «• o ¦

52S52Z22Z222000fi.i



Figure C-6 Heat Map of the State Average Ozone Concentrations Reductions (Green)
and Increases (Red) Due to the Alternative 2 Scenario Across Demographic Groups in
2028, 2030, 2035, 2040, and 2045 (ppb)

C-6


-------
Population

Race

Ethnicity <5 5(Wt
£ _

	w

a. i00*1

Educational a

,«s so^
Attainment °

o. 10m
Employment a
Status

Insurance	2

Status	|

Life	J"

Expectancy	°

, 100^

0%.

Linguistic
Isolation

Poverty
Status

Redlining

Tribal Land

5024: HS or more)

¦	Less educated (>24, no HS)

¦	White (0-99)

¦1 American Indian (0-99)

¦	Asian (0-99)

¦	Black (0-99)

¦	Top 75% life expectancy (0-99)

¦	Bottom 25% life expectancy (0-99)

¦	Life expectancy data unavailable (0-99)

¦	English well or better (0-99)

¦	English < well (0-99)

¦	HOLC Grades A-C (0-99)

¦	HOLC Grade D (0-99)

¦	Not Graded by HOLC (0-99)

¦	Not Tribal land (0-99)

¦	Tribal land (0-99)

-0.3 -0 1 0.1 0 3 O S -0 3 -0 1 0.1 0.3 O S -0.3 -0.1 0 1 0 3 0.5 -0.3 -01 0 1 0.3 0 5 -0.3 -0.1 01 0 3 0.5
03	03	03	03	03

Figure C-7 Distributions of Ozone Concentration Changes (ppb) Across Populations,
Future Years for the Alternative 1 Scenario

C-7


-------
Population

*	K

*

&
G.

O

I

o

r

r

r

r r
r
r
r
r
r
r
r

rj

I Employed (0-95)

I Unemployed (0*99)

Not In the labor force (0-99)

| > Poverty line (0-99)

| 24: HS or more}

I Less educated (>24. no HS)

| White (0-99)

I American Indian (0-99)

| Asian (0-99)

I Black (0-99)

| Top 75% life expectancy (0-99)
I Bottom 25% life expectancy (0-99)
I Life expectancy data unavailable (0-99)
| English well or better (0-99)

| English < well (0-99)

| HOLC Graces A-C (0-99)

| HOLC Grade D (0-99)

I Not Graded by HOIC (0-99)

| Not Tribal land (0-99)

| Trtpal land (0-99)

•0.3 -0.1 0.1 03 0.5 *0.3 -0.1 0.1 0.3 0.5 -0.3 -0.1 0.1 03 0.5 -0.3 -0.1 0.1 0.3 05 -0.3 -0.1
03	03	03	03

0.1 0.3 0 5
03

Figure C-8 Distributions of Ozone Concentration Changes (ppb) Across Populations,
Future Years for the Alternative 2 Scenario

C-8


-------
APPENDIX D: ASSESSMENT OF POTENTIAL COSTS AND EMISSIONS
IMPACTS OF FINAL NEW AND EXISTING SOURCE STANDARDS ANALYZED

SEPARATELY

D.l Modeling the Rules Independently

In this appendix, we describe the projected EGU compliance behavior, costs, and
emissions impacts for the final Emission Guidelines and final NSPS when modeled
independently. We also compare the results from each rule modeled individually with the results
presented elsewhere in the RIA that shows the final rules combined effects. This supplementary
analysis quantifies the climate benefits of these rules but does not quantify any additional
benefits, for instance health benefits from reductions in other pollutants, because of time and
resource constraints. The GHG mitigation measures modeled under each of these scenarios are
consistent with those applicable to each source category under the final rules, as outlined in
Table D-l and Table D-2.

Table D-l Summary of GHG Mitigation Measures for Existing Sources by Source

Category under the Final E

ailesa'b'c

Affected EGUs

Subcategory Definition

GHG Mitigation
Measure

Long-term existing coal-fired
steam generating units

Coal-fired steam generating units that have not
elected to commit to permanently cease
operations by 2040

CCS with 90% capture of
CO2, starting in 2035

Medium-term existing coal-
fired steam generating units

Coal-fired steam generating units that have not

elected to commit to permanently cease
operations prior to 2035, but have committed to
permanently ceasing operations prior to 2040

Natural gas co-firing at 40
percent of the heat input
to the unit, starting in
2030

3 All years shown in this table reflect IPM run years. Note that IPM run years encompass the specific calendar year
requirements of BSER, details of which are available in Section VII of the preamble.
b Coal units that lack existing SCR controls must install these controls in addition to CCS to comply.
c Coal-fired EGUs that convert entirely to burn natural gas by 2030 are no longer subject to coal-fired EGU
mitigation measures outlined above.

D-l


-------
Table D-2 Summary of Modeled GHG Mitigation Measures for New Sources by Source

Category under the Fina

Ruleabc





Modeled
Requirements
during 1st Phase

Modeled



Affected EGUs

Subcategory
Definition

Requirements
During 2nd Phase
(2035)

Baseload
Definition



NGCC units that









commence



CCS or co-fire



Baseload Economic NGCC
Additions

construction after
2023 and operate at
greater than
baseload annual
capacity factor

Efficient
generation

hydrogen at
sufficient level to
meet CCS
emission rate





NGCC units that









commence







Intermediate Load Economic
NGCC Additions

construction after
2023 and operate at
an annual capacity
factor of less than

Efficient generation





baseload





40%



NGCT units that







commence







Intermediate load Economic
NGCT Additions

construction after
2023 and operate at
an annual capacity
factor of more than

Emission rate consistent with NGCC
operation





40%









NGCT units that









commence







Peaking Economic NGCT
Additions

construction after
2023 and operate at
an annual capacity
factor of less than

Efficient generation





40%







3 All years shown in this table reflect IPM run years. Note that IPM run years encompass the specific calendar year
requirements of BSER, details of which are available in Section VII of the preamble.
b Delivered hydrogen price is assumed to be $1.15/kg in all years.

c The modeling does not reflect the requirements of the variable subcategory. We estimate this would have a limited
impact on the results.

D.2 Compliance Cost Assessment

The estimates of incremental costs of supplying electricity under the final rules and under
the final Emission Guidelines and final NSPS when modeled separately are presented in Table
D-3. Estimates for additional recordkeeping, monitoring, and reporting requirements for EGUs
are also included within the estimates in this table.

D-2


-------
Table D-3 National Power Sector Compliance Cost Estimates for the Illustrative
Scenarios (billions of 2019 dollars)	



Final

New Source
Rule Only

Existing Source
Rule Only

2024 to 2042 (Annualized)

0.43

0.14

0.52

2024 to 2047 (Annualized)

0.86

0.14

0.94

2028 (Annual)

-1.30

-0.05

-0.80

2030 (Annual)

-0.22

0.06

0.41

2035 (Annual)

1.28

-0.35

1.03

2040 (Annual)

0.59

-0.17

0.74

2045 (Annual)

3.34

-0.13

3.34

"2024 to 2042 (Annualized)" reflects total estimated annual compliance costs levelized over the period 2024 through
2042 and discounted using a 3.76 real discount rate.215 This does not include compliance costs beyond 2042. "2024
to 2047 (Annualized)" reflects total estimated annual compliance costs levelized over the period 2024 through 2047
and discounted using a 3.76 real discount rate. This does not include compliance costs beyond 2047. "2028
(Annual)" through "2045 (Annual)" costs reflect annual estimates in each of those run years.216

Existing coal-fired EGUs represent the largest share of affected resources within the final
rules. Hence the existing source rule is responsible for the majority of cost increases projected
under the final (combined effect) rule. New sources represent a smaller total share of the affected
sources under this rule, and hence cost increases projected under the final NSPS alone are
smaller than under the existing source rule.

Under the baseline, the proposed GNP rule results in installation of SCR controls in the
2030 run year on some coal-fired EGUs that currently lack them. Under the scenarios modeled, a
subset of these facilities retires rather than retrofit, since they would face additional requirements
under the GHG regulations modeled. This in turn results in lower capital costs in the first run
year and is balanced by higher costs in later years. Additionally, renewable costs are assumed to
decline over the forecast period. Given IPM's perfect foresight, the model choses to wait to build
incremental RE until later in the period when costs are lower. Under the illustrative policy

215	This table reports compliance costs consistent with expected electricity sector economic conditions. The PV of
costs was calculated using a 3.76 percent real discount rate consistent with the rate used in IPM's objective
function for cost-minimization. The PV of costs was then used to calculate the levelized annual value over a 19-
year period (2024 to 2042) and a 23-year period (2024 to 2047) using the 3.76 percent rate as well. Tables ES-19
and 8-4 report the PV of the annual stream of costs from 2024 to 2047 using 3 percent and 7 percent consistent
with OMB guidance.

216	Cost estimates include financing charges on capital expenditures that would reflect a transfer and would not
typically be considered part of total social costs.

D-3


-------
scenarios the model builds this capacity sooner, which results in lower costs in the years built,
but higher costs in future years.

D.3 Emissions Reduction Assessment

As indicated in Section 3, the CO2 emissions reductions are presented in this RIA from
2028 through 2045 and are based on IPM projections. Table D-4 presents the estimated reduction
in power sector CO2 emissions resulting from compliance with the final rule requirements, as
well as the estimated emissions from the final Emission Guidelines and final NSPS
independently.

The CO2 emission reductions follow an expected pattern: the existing source rule is
responsible for the majority of reductions under the final rules modeling presented in the RIA,
and these reductions occur primarily in the first half of the forecast period. The new source rule
is responsible for a smaller share of reductions, and these reductions occur more towards the
latter half of the forecast period. Cumulative CO2 reductions between 2028-47 under the final
rules (1.382 billion metric tons) are greater than under the existing source rule only (1.332 billion
metric tons) and under the final NSPS only (an increase of 15 million metric tons). Under the
New Source Rule only, CO2 emissions at new sources declines, but these are offset by increases
at existing sources, particularly in 2035. By 2040 reductions at new sources begin to match
increases in emissions at existing sources, and in 2045 reductions at new sources outweigh
increases at existing sources. Under the Existing Source Rule only, emissions from existing
sources are lower, and only partially offset by increases in emissions from new sources, resulting
in net emission decreases over the forecast period.

D-4


-------
Table D-4 EGU Annual CO2 Emissions and Emissions Changes (million metric tons)
for the Baseline and the Illustrative Scenarios from 2028 to 2045217

Annual
CO2



Total Emissions



Change from Baseline

(million
metric
tons)

Baseline

Final

New
Source
Rule Only

Existing
Source
Rule Only

Final

New
Source
Rule Only

Existing
Source
Rule Only

2028

1,159

1,121

1,162

1,121

-38

3

-38

2030

1,098

1,048

1,100

1,050

-50

2

-49

2035

724

601

729

600

-123

5

-125

2040

459

406

459

411

-54

0

-48

2045

307

265

303

271

-42

-4

-36

Cumulative
(2028-47)

12,538

11,156

12,553

11,206

-1,382

15

-1,332

There will also be impacts on non-CCh air emissions associated with EGUs burning fossil
fuels that result from compliance strategies modeled to meet the requirements of the final rules.
These other emissions include changes in emissions of NOx, SO2, and direct PM2.5 emissions
changes, as well as changes in ozone season NOx emissions. The emissions impacts are
presented in Table D-5.

217 This analysis is limited to the geographically contiguous lower 48 states.

D-5


-------
Table D-5 EGU Annual Emissions and Emissions Changes for Annual NOx, Ozone
Season (April to September) NOx, SO2, and Direct PM2.5 for the Baseline and Illustrative
Scenarios for 2028 to 2040

Annual

NOx



Total Emissions



Change from Baseline

(Thousand
Tons)

Baseline

Final

New
Source
Rule Only

Existing
Source
Rule Only

New

Final Source
Rule Only

Existing
Source
Rule Only

2028

461

441

465

439

-20

5

-22

2030

393

374

396

372

-20

3

-22

2035

259

210

266

204

-49

7

-55

2040

173

166

174

165

-6

2

-7

2045

107

83

107

83

-24

1

-24

Ozone
Season
NOxa



Total Emissions



Change from Baseline

(Thousand
Tons)

Baseline

Final

New
Source
Rule Only

Existing
Source
Rule Only

New

Final Source
Rule Only

Existing
Source
Rule Only

2028

189

183

191

182

-6

2

-7

2030

175

168

176

167

-7

1

-8

2035

119

100

122

96

-19

3

-23

2040

88

82

89

81

-6

2

-7

2045

59

45

59

45

-14

0

-14

Annual SO2



Total Emissions



Change from Baseline

(Thousand
Tons)

Baseline

Final

New
Source
Rule Only

Existing
Source
Rule Only

New

Final Source
Rule Only

Existing
Source
Rule Only

2028

454

420

461

424

-34

7

-30

2030

334

313

335

318

-20

2

-15

2035

240

150

244

150

-90

4

-90

2040

143

139

143

139

-4

1

-4

2045

55

13

53

13

-41

-1

-41

Direct PM2.5



Total Emissions



Change from Baseline

(Tons)

Baseline

Final

New
Source
Rule Only

Existing
Source
Rule Only

New

Final Source
Rule Only

Existing
Source
Rule Only

2028

71

69

71

69

-2

0

-2

2030

66

65

66

65

-2

0

-1

2035

51

49

52

49

-1

1

-2

2040

37

39

38

39

2

0

1

2045

24

22

24

22

-2

0

-2

D-6


-------
3 Ozone season is the May through September period in this analysis.

D.4 Impacts on Fuel Use and Generation Mix

The final NSPS and final Emission Guidelines expected to result in significant GHG
emissions reductions. They are also expected to have impacts on the power sector. Consideration
of these potential impacts is an important component of assessing the relative impact of the
illustrative scenarios. In this section we discuss the estimated changes in fuel use, fuel prices,
generation by fuel type, and capacity by fuel type for the 2030, 2035, 2040 and 2045 IPM model
run years under the final rules and under the final Emission Guidelines and final NSPS
independently.

As outlined in Table D-6, under the final existing source rule only, coal consumption falls
more than under the final rules, while coal consumption falls least under the final new source
rule only. Under the existing source rule only, GHG mitigation measures apply to existing coal-
fired EGUs as outlined in Table D-l. Hence coal capacity reductions are offset by increases in
new source NGCC generation. Under the new source rule-only modeling, the GHG mitigation
measures apply only to new fossil-fuel fired sources, as outlined in Table D-2. Hence generation
and emissions from these sources falls and are compensated for by increases in generation and
emissions from existing sources.

D-7


-------
Table D-6 2028, 2030, 2035, 2040 and 2045 Projected U.S. Power Sector Coal Use for

the Baseline and the Illustrative Scenarios

Million Tons

Percent Change from Baseline



Year

Baseline

Final

New
Source
Rule
Only

Existing
Source
Rule
Only

Final

New
Source
Rule
Only

Existing
Source
Rule
Only

Appalachia



40

37

42

35

-7%

6%

-11%

Interior



38

35

39

35

-7%

4%

-6%

Waste Coal

2028

7

7

7

7

0%

0%

0%

West



166

155

164

157

-7%

-1%

-6%

Total



251

234

253

235

-7%

1%

-6%

Appalachia



39

39

39

39

0%

0%

0%

Interior



35

36

38

34

1%

7%

-3%

Waste Coal

2030

7

7

7

7

0%

0%

0%

West



141

113

139

115

-20%

-1%

-18%

Total



222

194

223

195

-13%

0%

-12%

Appalachia



32

19

34

19

-40%

7%

-40%

Interior



19

25

21

25

30%

6%

30%

Waste Coal

2035

7

3

7

3

-53%

0%

-53%

West



89

63

88

63

-29%

-1%

-29%

Total



147

111

150

111

-25%

2%

-25%

Appalachia



19

19

20

19

1%

3%

1%

Interior



10

25

10

25

150%

0%

150%

Waste Coal

2040

3

3

3

3

0%

0%

0%

West



61

56

60

56

-8%

0%

-8%

Total



93

103

93

103

11%

0%

11%

Appalachia



4

0

4

0

-100%

9%

-100%

Interior



1

0

1

0

-100%

0%

-100%

Waste Coal

2045

3

0

3

0

-100%

0%

-100%

West



20

3

20

3

-85%

0%

-85%

Total



28

3

28

3

-89%

1%

-90%

As outlined in Table D-7 gas consumption follows the opposite trend to coal consumption
under the three scenarios shown. Under the existing source rule, gas consumption remains at
similar levels to the final rules (gas generation compensates for declining coal generation), while
under the new source rule, gas generation is moderately lower as a result of GHG mitigation
measures applied to new fossil-fuel fired sources, while similar measures are not applied to
existing coal-fired sources.

D-8


-------
Table D-7 2028, 2030, 2035, 2040 and 2045 Projected Power Sector Natural Gas Use for

the Baseline and the Illustrative Scenarios

Trillion Cubic Feet

Percent Change from Baseline

Year

Baseline

Final

New
Source
Rule
Only

Existing
Source

Final

New
Source
Rule
Only

Existing
Source







Rule Only



Rule Only

2028

11.6

11.5

11.5

11.5

-1.0%

-0.6%

-1.0%

2030

11.7

11.7

11.7

11.7

0.0%

0.0%

0.0%

2035

9.3

9.7

9.3

9.6

4.3%

-0.1%

4.1%

2040

6.4

6.4

6.4

6.5

-0.1%

-0.1%

1.0%

2045

4.2

4.3

4.2

4.4

1.1%

-1.8%

3.0%

As outlined in Table D-8 and Table D-9 coal and gas prices are similar under the final
rules and Existing Source rules, while changes are smaller under the final NSPS.

Table D-8 2028, 2030, 2035 and 2040 Projected Minemouth and Power Sector Delivered
Coal Price (2019 dollars) for the Baseline and the Illustrative Scenarios	

$/MMBtu

Percent Change from Baseline





Baseline

Final

New
Source
Rule
Only

Existing
Source
Rule
Only

Final

New
Source
Rule
Only

Existing
Source
Rule
Only

Minemouth
Delivered

2028

0.98
1.54

0.97
1.52

0.99
1.56

0.96
1.52

-1%
-1%

1%
1%

-1%

-2%

Minemouth
Delivered

2030

1.02
1.56

1.05
1.53

1.03
1.57

1.04
1.53

3%
-2%

1%
1%

2%
-2%

Minemouth
Delivered

2035

1.07
1.55

1.10
1.55

1.08
1.56

1.10
1.55

3%
0%

1%
1%

3%
0%

Minemouth
Delivered

2040

1.17
1.59

1.22
1.60

1.17
1.60

1.21
1.60

4%
1%

0%
0%

3%
0%

Minemouth
Delivered

2045

1.37

1.38

1.50
0.94

1.37
1.39

1.50
0.95

9%
-32%

0%
1%

9%
-31%

D-9


-------
Table D-9 2028, 2030, 2035 and 2040 Projected Henry Hub and Power Sector Delivered
Natural Gas Price (2016 dollars) for the Baseline and the Illustrative Scenarios	

$/MMBtu

Percent Change from Baseline





Baseline

Final

New
Source
Rule
Only

Existing
Source
Rule
Only

Final

New
Source
Rule
Only

Existing
Source
Rule
Only

Henry Hub
Delivered

2028

2.78
2.84

2.72
2.78

2.78
2.84

2.76
2.82

-2%
-2%

0%
0%

-1%
-1%

Henry Hub
Delivered

2030

2.89
2.95

2.90
2.97

2.89
2.96

2.95
3.01

0%
1%

0%
0%

2%
2%

Henry Hub
Delivered

2035

2.87

2.88

2.95
2.97

2.82

2.83

2.95
2.97

3%
3%

-2%
-2%

3%
3%

Henry Hub
Delivered

2040

2.82
2.79

2.79
2.77

2.77
2.75

2.83
2.81

-1%
-1%

-2%
-2%

1%
1%

Henry Hub
Delivered

2045

2.95
2.94

2.95
2.94

2.91
2.89

3.00
2.98

0%
0%

-1%

-2%

2%
2%

As outlined in Table D-10 the generation mix remains generally similar under the final and
existing source rules, but the non-imposition of GHG mitigation measures on new fossil-fired
sources under the existing source rule only scenario results in some increase in generation from
new NGCC capacity relative to the final. Under the new source only scenario, the overall
generation mix is similar to the baseline, with the exception of higher coal dispatch driven by the
GHG mitigation measures on new fossil-fired sources reducing the total dispatch of new NGCC
units.

Table D-10 2028, 2030, 2035 and 2040 Projected U.S. Generation by Fuel Type for the

Baseline and the Illustrative Scenarios

Generation (TWh)

Percent Change from Baseline



Year

Baseline

Final

New
Source
Rule
Only

Existing
Source
Rule
Only

Final

New
Source
Rule
Only

Existing
Source
Rule
Only

Unabated Coal



472

441

480

441

-7%

2%

-7%

Coal & CCS



0

0

0

0

-

-

-

Coal & Nat. Gas co-



0

0

0

0







firing

2028







Unabated Nat. Gas

1,652

1,631

1,636

1,642

-1%

-1%

-1%

Nat. Gas & CCS



0

0

0

0

-

-

-

Nuclear



751

751

751

751

0%

0%

0%

Hydro



293

293

293

292

0%

0%

0%

D-10


-------
Non-Hydro RE



1,141

1,191

1,148

1,182

4%

1%

4%

Oil/Gas Steam



26

28

27

25

8%

7%

-1%

Other



31

31

31

31

0%

0%

0%

Grand Total



4,365

4,366

4,366

4,364

0%

0%

0%

Unabated Coal



407

355

409

356

-13%

1%

-12%

Coal & CCS



3

5

3

5

71%

0%

76%

Coal & Nat. Gas co-
firing

Unabated Nat. Gas



0

1,670

2

1,660

0

1,660

2

1,674

0%

0%

1%

Nat. Gas & CCS



0

0

0

0

-

-

-

Nuclear

2030

729

729

729

729

0%

0%

0%

Hydro



298

299

298

298

0%

0%

0%

Non-Hydro RE



1,329

1,381

1,335

1,370

4%

0%

3%

Oil/Gas Steam



25

28

26

25

12%

8%

0%

Other



31

31

31

31

0%

0%

0%

Grand Total



4,491

4,491

4,491

4,490

0%

0%

0%

Unabated Coal



160

0

166

0

-100%

4%

-100%

Coal & CCS



76

133

76

133

74%

0%

74%

Coal & Nat. Gas co-
firing

Unabated Nat. Gas



0

1,341

4

1,379

0

1,321

4

1,398

4%

-1%

5%

Nat. Gas & CCS



3

7

7

2

105%

105%

-37%

Nuclear

2035

667

666

667

666

0%

0%

0%

Hydro



319

317

318

317

-1%

0%

-1%

Non-Hydro RE



2,229

2,286

2,238

2,273

3%

0%

2%

Oil/Gas Steam



8

9

8

9

21%

4%

14%

Other



31

30

31

30

0%

0%

0%

Grand Total



4,834

4,831

4,831

4,833

0%

0%

0%

Unabated Coal



61

0

62

0

-100%

2%

-100%

Coal & CCS



76

128

76

128

68%

0%

68%

Coal & Nat. Gas co-
firing

Unabated Nat. Gas



0

933

0

919

0

918

0

947

0%

-1%

3%

Nat. Gas & CCS



3

7

7

2

105%

105%

-37%

Nuclear

2040

614

613

614

613

0%

0%

0%

Hydro



336

336

337

335

0%

0%

0%

Non-Hydro RE



3,097

3,119

3,108

3,095

1%

0%

0%

Oil/Gas Steam



5

6

5

6

28%

2%

26%

Other



29

29

29

29

0%

0%

0%

Grand Total



5,154

5,157

5,155

5,155

0%

0%

0%

Unabated Coal

2045

45

0

46

0

-100%

2%

-100%

D-ll


-------
Coal & CCS



4

3

4

3

-7%

0%

-8%

Coal & Nat. Gas co-
firing



0

0

0

0

-

-

-

Unabated Nat. Gas



614

612

595

635

1%

-2%

5%

Nat. Gas & CCS



3

6

6

2

103%

100%

-42%

Nuclear



471

472

471

474

0%

0%

1%

Hydro



343

342

343

342

0%

0%

0%

Non-Hydro RE



4,032

4,089

4,048

4,066

1%

0%

1%

Oil/Gas Steam



4

6

5

6

25%

1%

24%

Other



28

27

28

27

0%

0%

0%

Grand Total



5,544

5,557

5,544

5,555

0%

0%

0%

As outlined in Table D-l 1 the capacity mix follows similar trends to those seen under the
generation mix table. The capacity mix under the final and existing source rule scenarios are
similar, while the capacity mix under the baseline and new source rule only scenarios are similar.
The new source rule only is projected to result in less new NGCC and more existing coal
capacity relative to the baseline, while the existing source rule only is projected to result in less
coal capacity and more new NGCC capacity relative to the projected final results.

Table D-ll 2028, 2030, 2035, 2040 and 2045 Projected U.S. Capacity by Fuel Type for

the Baseline and the Illustrative Scenarios

Capacity (GW)

Percent Change from Baseline



Year

Baseline

Final

New
Source
Rule
Only

Existing
Source
Rule
Only

Final

New
Source
Rule
Only

Existing
Source
Rule
Only

Unabated Coal



106

101

108

98

-4%

2%

-7%

Coal & CCS



0

0

0

0

-

-

-

Coal & Nat. Gas co-



0

0

0

0







firing









Unabated Nat. Gas



471

472

467

476

0%

-1%

1%

Nat. Gas & CCS



0

0

0

0

-

-

-

Nuclear

2028

94

94

94

94

0%

0%

0%

Hydro



102

102

102

102

0%

0%

0%

Non-Hydro RE



394

407

396

404

3%

1%

3%

Oil/Gas Steam



63

64

62

64

2%

0%

2%

Other



7

7

7

7

0%

0%

0%

Grand Total



1,236

1,246

1,236

1,245

1%

0%

1%

D-12


-------
Unabated Coal



85

72

85

70

-15%

1%

-18%

Coal & CCS



0

1

0

1

72%

0%

77%

Coal & Nat. Gas co-
firing



0

1

0

1

-

-

-

Unabated Nat. Gas



479

480

475

484

1%

0%

2%

Nat. Gas & CCS



0

0

0

0

-

-

-

Nuclear

2030

91

91

91

91

0%

0%

0%

Hydro



104

104

104

104

0%

0%

0%

Non-Hydro RE



440

454

443

450

3%

1%

2%

Oil/Gas Steam



64

73

64

73

13%

0%

13%

Other



7

7

7

7

0%

0%

0%

Grand Total



1,269

1,281

1,269

1,279

1%

0%

1%

Unabated Coal



41

0

42

0

-100%

4%

-100%

Coal & CCS



11

19

11

19

74%

0%

74%

Coal & Nat. Gas co-
firing



0

1

0

1

-

-

-

Unabated Nat. Gas



476

484

472

486

2%

0%

2%

Nat. Gas & CCS



0

1

1

0

104%

104%

-36%

Nuclear

2035

84

84

84

84

0%

0%

0%

Hydro



107

107

107

107

0%

0%

0%

Non-Hydro RE



699

714

701

710

2%

0%

2%

Oil/Gas Steam



55

66

55

65

19%

0%

18%

Other



7

7

7

7

0%

0%

0%

Grand Total



1,479

1,482

1,480

1,479

0%

0%

0%

Unabated Coal



31

0

31

0

-99%

0%

-99%

Coal & CCS



11

18

11

18

68%

0%

68%

Coal & Nat. Gas co-
firing



0

0

0

0

-

-

-

Unabated Nat. Gas



516

525

515

526

2%

0%

2%

Nat. Gas & CCS



0

1

1

0

104%

104%

-36%

Nuclear

2040

79

79

79

79

0%

0%

0%

Hydro



112

112

112

112

0%

0%

0%

Non-Hydro RE



943

952

947

944

1%

0%

0%

Oil/Gas Steam



54

65

54

64

19%

0%

19%

Other



7

7

7

7

0%

0%

0%

Grand Total



1,753

1,759

1,758

1,751

0%

0%

0%

Unabated Coal



29

0

29

0

-99%

1%

-99%

Coal & CCS



1

1

1

1

-5%

0%

-7%

Coal & Nat. Gas co-

2045

0

0

0

0







firing









Unabated Nat. Gas



565

581

563

583

3%

0%

4%

D-13


-------
Nat. Gas & CCS



0

1

1

0

104%

104%

-36%

Nuclear



65

65

65

65

0%

0%

0%

Hydro



112

112

112

112

0%

0%

0%

Non-Hydro RE



1,232

1,250

1,238

1,242

1%

0%

1%

Oil/Gas Steam



54

64

54

64

19%

0%

19%

Other



7

7

7

7

0%

0%

0%

Grand Total



2,065

2,080

2,069

2,073

1%

0%

0%

D-14


-------
United States	Office of Air Quality Planning and Standards	Publication No. EPA-452/R-24-009

Environmental Protection	Health and Environmental Impacts Division	April 2024

Agency	Research Triangle Park, NC


-------
I Q \
I#

xSr/

Regulatory Impact Analysis for the New Source
Performance Standards for Greenhouse Gas
Emissions from New, Modified, and
Reconstructed Fossil Fuel-Fired Electric
Generating Units; Emission Guidelines for
Greenhouse Gas Emissions from Existing Fossil
Fuel-Fired Electric Generating Units; and
Repeal of the Affordable Clean Energy Rule


-------

-------
EPA-452/R-24-009
April 2024

Regulatory Impact Analysis for the New Source Performance Standards for Greenhouse Gas
Emissions from New, Modified, and Reconstructed Fossil Fuel-Fired Electric Generating Units;
Emission Guidelines for Greenhouse Gas Emissions from Existing Fossil Fuel-Fired Electric
Generating Units; and Repeal of the Affordable Clean Energy Rule

U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Health and Environmental Impacts Division
Research Triangle Park, NC


-------
CONTACT INFORMATION

This document has been prepared by staff from the Office of Air and Radiation of the U.S.
Environmental Protection Agency. Questions related to this document should be addressed to the
Air Economics Group in the Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Office of Air and Radiation, Research Triangle Park, North Carolina 27711
(email: OAQPSeconomics@epa.gov).

ACKNOWLEDGEMENTS

In addition to U.S. EPA staff from the Office of Air and Radiation, personnel from the Office of
Policy of the U.S. Environmental Protection Agency contributed data and analysis to this
document.


-------
TABLE OF CONTENTS

Table of Contents	i

Table of Tables	iv

Table of Figures	viii

Executive Summary	1

ES. 1 Introduction	ES-1

ES.2 Regulatory Requirements	ES-2

ES.3 Baseline and Analysis Years	ES-3

ES.4 Emissions Impacts	ES-5

ES.5 Compliance Costs	ES-6

ES.6 Benefits	ES-8

ES.6.1 Climate Benefits	ES-8

ES.6.2 Health Benefits	ES-9

ES.6.3 Additional Unqualified Benefits	ES-9

ES.6.4 Total Climate and Health Benefits	ES-10

ES.7 Economic Impacts	ES-12

ES.8 Environmental Justice Impacts	ES-14

ES.9 Comparison of Benefits and Costs	ES-16

ES. 10 References	ES-20

1	Introduction and Background	1-1

1.1	Introduction	1-1

1.2	Legal and Economic Basis for Rulemaking	1-3

1.2.1	Statutory Requirement	1-3

1.2.2	The Need for Air Emissions Regulation	1-5

1.3	Overview of Regulatory Impact Analysis	1-5

1.3.1	Repeal of Affordable Clean Energy (ACE) Rule	1-5

1.3.2	Baseline and Analysis Years	1-7

1.3.3	Best System of Emission Reduction (BSER)	1-8

1.3.4	Illustrative Scenarios	1-9

1.4	Organization of the Regulatory Impact Analysis	1-9

1.5	References	1-11

2	Industry Profile	2-1

2.1	Background	2-1

2.2	Power Sector Overview	2-1

2.2.1	Generation	2-1

2.2.2	Transmission	2-13

2.2.3	Distribution	2-14

2.3	Sales, Expenses, and Prices	2-14

2.3.1	Electricity Prices	2-15

2.3.2	Prices of Fossil Fuel Used for Generating Electricity	2-17

2.3.3	Changes in Electricity Intensity of the U.S. Economy from 2010 to 2022	2-18

3	Compliance Costs, Emissions, and Energy Impacts	3-1

3.1	Overview	3-1

3.2	Illustrative Scenarios	3-1

3.3	Monitoring, Reporting, and Recordkeeping Costs	3-5

3.4	Power Sector Modeling Framework	3-7

3.5	EPA's Power Sector Modeling of the Baseline Run and Three Illustrative Scenarios	3-10

3.5.1 EPA's IPM Baseline Run \ 7.23	3-11

l


-------
3.5.2	Methodology for Evaluating the Illustrative Scenarios	3-12

3.5.3	Methodology for Estimating Compliance Costs	3-14

3.6	Estimated Impacts of the Illustrative Scenarios	3-14

3.6.1	Emissions Reduction Assessment	3-14

3.6.2	Compliance Cost Assessment	3-18

3.6.3	Impacts on Fuel Use, Prices, and Generation Mix	3-20

3.7	Limitations	3-38

3.8	References	3-42

4	Benefits Analysis	4-1

4.1	Introduction	4-1

4.2	Climate Benefits	4-2

4.3	Human Health Benefits	4-22

4.3.1	Air Quality Modeling Methodology and Results	4-23

4.3.2	Selecting Air Pollution Health Endpoints to Quantify	4-25

4.3.3	Calculating Counts of Air Pollution Effects Using the Health Impact Function	4-30

4.3.4	Calculating the Economic Valuation of Health Impacts	4-32

4.3.5	Benefits Analysis Data Inputs	4-35

4.3.6	Quantifying Cases of Ozone-Attributable Premature Death	4-38

4.3.7	Quantifying Cases of PM2s-Attributable Premature Death	4-40

4.3.8	Characterizing Uncertainty in the Estimated Benefits	4-43

4.3.9	Estimated Number and Economic Value of Health Benefits	4-47

4.4	Additional Unquantified Benefits	4-67

4.4.1	Hazardous Air Pollutant Impacts	4-70

4.4.2	NO Health Benefits	4-72

4.4.3	SO2 Health Benefits	4-72

4.4.4	Ozone Welfare Benefits	4-73

4.4.5	NO2 and SO2 Welfare Benefits	4-74

4.4.6	Visibility Impairment Benefits	4-75

4.4.7	Water Quality and Availability Benefits	4-75

4.5	Total Benefits	4-81

4.6	References	4-93

5	Social Costs and Economic Impacts	5-1

5.1	Energy Market Impacts	5-1

5.2	Economy-wide Social Costs and Economic Impacts	5-3

5.2.1	Economy-wide Modelling	5-3

5.2.2	Overview of the SAGE CGE Model	5-4

5.2.3	Linking IPM PE Model to SAGE CGE Model	5-9

5.2.4	Results	5-17

5.2.5	Limitations to Analysis	5-31

5.3	Small Entity Analysis	5-33

5.3.1	Overview	5-33

5.3.2	EGU Small Entity Analysis and Results	5-34

5.4	Labor Impacts	5-41

5.4.1	Overview of Methodology	5-43

5.4.2	Overview of Power Sector Employment	5-44

5.4.3	Projected Sectoral Employment Changes due to the Final Rules	5-45

5.4.4	Conclusions	5-47

5.5	References	5-49

6	Environmental Justice Impacts	6-1

6.1	Introduction	6-1

6.2	Analyzing EJ Impacts in These Final Rules	6-3

6.3	GHG Impacts on Environmental Justice and other Populations of Concern	6-4

it


-------
6.4	Demographic Proximity Analyses of Existing Facilities	6-10

6.5	EJ PM2.5 and Ozone Exposure Impacts	6-17

6.5.1	Populations Predicted to Experience PM2.5 and Ozone Air Quality Changes	6-21

6.5.2	PM2 5 EJ Exposure Analysis	6-23

6.5.3	Ozone EJ Exposure Analysis	6-30

6.6	Qualitative Discussion of EJ PM2 5 Health Impacts	6-39

6.7	Qualitative Discussion of New Source EJ Impacts	6-40

6.8	Summary	6-40

6.9	References	6-44

7 Comparison of Benefits and Costs	7-1

7.1	Introduction	7-1

7.2	Methods	7-1

7.3	Results	7-2

Appendix A: Climate Benefits	A-l

A. 1 Climate Benefits Estimated using the Interim SC-CO2 values used in the Proposal	A-l

A.	2 References	A-2

Appendix B: Air Quality Modeling	B-l

B.	1	Air Quality Modeling Simulations	B-l

B.2	Applying Modeling Outputs to Create Spatial Fields	B-13

B.3	Scaling Factors Applied to Source Apportionment Tags	B-20

B.4	Air Quality Surface Results	B-41

B. 5	Uncertainties and Limitations of the Air Quality Methodology	B -5 8

B.6	References	B-59

Appendix C: Environmental Justice Analysis	C-l

C.	1 Exposure Analysis Figures for the Alternative Scenarios	C-l

Appendix D: Assessment of Potential Costs and Emissions Impacts of Final New
and Existing Source Standards Analyzed Separately	D-l

D.	1	Modeling the Rules Independently	D-l

D.2	Compliance Cost Assessment	D-2

D.3	Emissions Reduction Assessment	D-4

D.4	Impacts on Fuel Use and Generation Mix	D-7

111


-------
TABLE OF TABLES

Table

ES-1

Table

ES-2

Table

ES-3

Table

ES-4

Table

ES-5

Table

2-1

Table

2-2

Table

2-3

Table

2-4

Table

2-5

Table

3-1

Table

3-2

Table

3-3

Table

3-4

Table

3-5

Table

3-6

Table

3-7

Table

3-8

Table

3-9

Table

3-10

Table

3-11

Table

3-12

Table

3-13

Table

3-14

Table

3-15

Table

3-16

Table

3-17

Table

3-18

Table

4-1

Table

4-2

Table

4-3

Table

4-4

Table

4-5

Projected EGU Emissions and Emissions Changes for the Three Illustrative Scenarios for 2028, 2030,

2035, 2040, and 2045 3	ES-6

Total National Compliance Cost Estimates for the Three Illustrative Scenarios (discounted to 2024,

billion 2019 dollars)	ES-7

Benefits for the Three Illustrative Scenarios, (discounted to 2024, billion 2019 dollars)3	ES-11

Summary of Certain Energy Market Impacts for the Illustrative Final Rules Scenario Relative to the

Baseline	ES-12

Net Benefits of the Illustrative Scenarios (billions of 2019 dollars, discounted to 2024)3	ES-18

Total Net Summer Electricity Generating Capacity by Energy Source, 2010-22 and 2015-22	2-4

Net Generation by Energy Source, 2010-22 and 2015 - 22 (Trillion kWh = TWh)	2-6

Net Generation in 2015 and 2022 (Trillion kWh = TWh)	2-6

Coal and Natural Gas Generating Units, by Size, Age, Capacity, and Average Heat Rate in 2023.... 2-9

Total U.S. Electric Power Industry Retail Sales, 2010-22 and 2014-22 (billion kWh)	2-15

Summary of Modeled GHG Mitigation Measures for Existing Sources by Source Category under the

Illustrative Final Rules and Alternative 1 Scenario3,bc	3-2

Summary of Modeled GHG Mitigation Measures for Existing Sources by Source Category under the

Illustrative Alternative 2 Scenario3,bc	3-3

Summary of GHG Mitigation Measures for New Sources by Source Category under the Illustrative

Final Rules, Alternative 1 and Alternative 2 Scenarios3,13	3-4

Summary of State and Industry Annual Respondent Cost of Reporting and Recordkeeping

Requirements (million 2019 dollars)	3-7

EGU Annual CO2 Emissions and Emissions Changes (million metric tons) for the Baseline and the

Illustrative Scenarios from 2028 through 2045 	3-15

EGU Annual Emissions and Emissions Changes for NOx, SO2, PM2.5, Hg and Ozone NOx for the

Illustrative Scenarios for 2028 to 2045 	3-17

National Power Sector Compliance Cost Estimates (billions of 2019 dollars) for the Illustrative

Scenarios	3-18

2028, 2030, 2035, 2040 and 2045 Projected U.S. Power Sector Coal Use for the Baseline and the

Illustrative Scenarios	3-22

2028, 2030, 2035, 2040 and 2045 Projected U.S. Power Sector Natural Gas Use for the Baseline and

the Illustrative Scenarios	3-23

2028, 2030, 2035, 2040 and 2045 Projected U.S. Power Sector Hydrogen Use for the Baseline and the

Illustrative Scenarios	3-23

2028, 2030, 2035, 2040 and 2045 Projected Minemouth and Power Sector Delivered Coal Price (2019

dollars) for the Baseline and the Illustrative Scenarios	3-24

2028, 2030, 2035, 2040 and 2045 Projected Henry Hub and Power Sector Delivered Natural Gas

Price (2019 dollars) for the Baseline and the Illustrative Scenarios	3-25

2028, 2030, 2035, 2040 and 2045 Projected U.S. Generation by Fuel Type for the Baseline and the

Illustrative Scenarios	3-26

2028, 2030, 2035, 2040 and 2045 Projected U.S. Capacity by Fuel Type for the Baseline and the

Illustrative Scenarios	3-31

Average Retail Electricity Price by Region for the Baseline and the Illustrative Scenarios, 2030... 3-34
Average Retail Electricity Price by Region for the Baseline and the Illustrative Scenarios, 2035... 3-35
Average Retail Electricity Price by Region for the Baseline and the Illustrative Scenarios, 2040... 3-36
Average Retail Electricity Price by Region for the Baseline and the Illustrative Scenarios, 2045... 3-37

Estimates of the Social Cost of CO2 Values, 2028-2047 (2019 dollars per Metric Ton CO2)3	4-14

Annual CO2 Emissions Reductions (million metric tons) for the Illustrative	4-16

Estimated Climate Benefits of Reduced CO2 Emissions from the Illustrative Scenarios, 2028 to 2047

(billions of 2019 dollars)	4-17

Health Effects of Ambient Ozone and PM2.5 and Climate Effects	4-29

Estimated Avoided Ozone-Related Premature Respiratory Mortalities and Illnesses for the Illustrative
Scenarios for 2028 (95 percent confidence interval)3	4-49


-------
Table 4-6 Estimated Avoided Ozone-Related Premature Respiratory Mortalities and Illnesses for the Illustrative

Scenarios for 2030 (95 percent confidence interval)3	4-50

Table 4-7 Estimated Avoided Ozone-Related Premature Respiratory Mortalities and Illnesses for the Illustrative

Scenarios for 2035 (95 percent confidence interval)3	4-51

Table 4-8 Estimated Avoided Ozone-Related Premature Respiratory Mortalities and Illnesses for the Illustrative

Scenarios for 2040 (95 percent confidence interval)3	4-52

Table 4-9 Estimated Avoided Ozone-Related Premature Respiratory Mortalities and Illnesses for the Illustrative

Scenarios in 2045 (95 percent confidence interval)3'13	4-53

Table 4-10 Estimated Avoided PM-Related Premature Mortalities and Illnesses for the Illustrative Scenarios in

2028 (95 percent confidence interval)	4-54

Table 4-11 Estimated Avoided PM-Related Premature Mortalities and Illnesses for the Illustrative Scenarios in

2030 (95 percent confidence interval)	4-55

Table 4-12 Estimated Avoided PM-Related Premature Mortalities and Illnesses for the Illustrative Scenarios in

2035 (95 percent confidence interval)	4-56

Table 4-13 Estimated Avoided PM-Related Premature Mortalities and Illnesses for the Illustrative Scenarios in

2040 (95 percent confidence interval)	4-57

Table 4-14 Estimated Avoided PM-Related Premature Mortalities and Illnesses for the Illustrative Scenarios in

2045 (95 percent confidence interval)	4-58

Table 4-15 Estimated Discounted Economic Value of Avoided Ozone and PM2 5-Attributable Premature

Mortality and Illness for the Illustrative Scenarios in 2028 (95 percent confidence interval; billions of

2019 dollars) d	4-59

Table 4-16 Estimated Discounted Economic Value of Avoided Ozone and PM2 5-Attributable Premature

Mortality and Illness for the Illustrative Scenarios in 2030 (95 percent confidence interval; billions of

2019 dollars);d	4-60

Table 4-17 Estimated Discounted Economic Value of Avoided Ozone and PM2 5-Attributable Premature

Mortality and Illness for the Illustrative Scenarios in 2035 (95 percent confidence interval; billions of

2019 dollars);d	4-61

Table 4-18 Estimated Discounted Economic Value of Avoided Ozone and PM2 5-Attributable Premature

Mortality and Illness for the Illustrative Scenarios in 2040 (95 percent confidence interval; billions of

2019 dollars);d	4-62

Table 4-19 Estimated Discounted Economic Value of Avoided Ozone and PM2 5-Attributable Premature

Mortality and Illness for the Illustrative Scenarios in 2045 (95 percent confidence interval; billions of

2019 dollars);d	4-63

Table 4-20 Estimated Discounted Economic Value of Avoided Ozone and PM2 5-Attributable Premature

Mortality and Illness for the Illustrative Scenarios in 2028, 2030, 2035, 2040 and 2045 (billions of

2019 dollars)"'"	4-64

Table 4-21 Stream of Human Health Benefits from 2028 through 2047: Monetized Benefits Quantified as Sum of
Long-Term Ozone Mortality and Illness and Long-Term PM2 5 Mortality and Illness for EGUs

(discounted at 2 percent; billions of 2019 dollars)3	4-65

Table 4-22 Stream of Human Health Benefits from 2028 through 2047: Monetized Benefits Quantified as Sum of
Long-Term Ozone Mortality and Illness and Long-Term PM2 5 Mortality and Illness for EGUs

(discounted at 3 percent; billions of 2019 dollars)3	4-66

Table 4-23 Stream of Human Health Benefits from 2028 through 2047: Monetized Benefits Quantified as Sum of
Long-Term Ozone Mortality and Illness and Long-Term PM2 5 Mortality and Illness for EGUs

(discounted at 7 percent; billions of 2019 dollars)3	4-67

Table 4-24 Unqualified Health and Welfare Benefits Categories	4-68

Table 4-25 Total Benefits for the Illustrative Scenarios for 2028 (billions of 2019 dollars)3	4-82

Table 4-26 Total Benefits for the Illustrative Scenarios for 2030 (billions of 2019 dollars)3	4-83

Table 4-27 Total Benefits for the Illustrative Scenarios for 2035 (billions of 2019 dollars)3	4-84

Table 4-28 Total Benefits for the Illustrative Scenarios for 2040 (billions of 2019 dollars)3	4-85

Table 4-29 Total Benefits for the Illustrative Scenarios for 2045 (billions of 2019 dollars)3	4-86

Table 4-30 Benefits for the Final Rules Illustrative Scenario from 2024 through 2047 (billions of 2019 dollars)3 4-
87

Table 4-31 Benefits for the Alternative 1 Illustrative Scenario from 2028 through 2047 (billions of 2019

dollars)1'	4-89

v


-------
Table 4-32 Benefits for the Alternative 2 Illustrative Scenario from 2024 through 2047 (billions of 2019 dollars)3

	4-91

Table 5-1 Summary of Certain Energy Market Impacts (percent change)	5-2

Table 5-2 SAGE Dimensional Details	5-7

Table 5-3 IPM Cost Outputs	5-15

Table 5-4 Compliance Costs, Transfers, and Social Costs (billions of 2019 dollars)	5-20

Table 5-5 SBA Size Standards by NAICS Code	5-37

Table 5-6 Historical NGCC and NGCT Additions (2017-present)	5-38

Table 5-7 Projected Impact of the Final Rule on Small Entities in 2035	5-40

Table 5-8 Changes in Labor Utilization: Construction-Related (number of job-years of employment in a single

year)	5-46

Table 5-9 Changes in Labor Utilization: Recurring Non-Construction (number of job-years of employment in a

single year)	5-47

Table 6-1 Proximity Demographic Assessment Results Within 5 km of Coal-Fired Units Greater than 25 MW

Affected by these Final Rules a b c	6-14

Table 6-2 Proximity Demographic Assessment Results Within 10 km of Coal-Fired Units Greater than 25 MW

Affected by these Final Rules a,b	6-15

Table 6-3 Proximity Demographic Assessment Results Within 50 km of Coal-Fired Units Greater than 25 MW

Affected by these Final Rules a,b	6-16

Table 6-4 Demographic Populations Included in the Ozone and PM2.5 EJ Exposure Analysis	6-21

Table 7-1 Net Benefits of the Three Illustrative Scenarios in 2028 (billion 2019 dollars)ab	7-4

Table 7-2 Net Benefits of the Three Illustrative Scenarios in 2030 (billion 2019 dollars)a b	7-5

Table 7-3 Net Benefits of the Three Illustrative Scenarios in 2035 (billion 2019 dollars)a b	7-6

Table 7-4 Net Benefits of the Three Illustrative Scenarios in 2040 (billion 2019 dollars)a b	7-7

Table 7-5 Net Benefits of the Three Illustrative Scenarios in 2045 (billion 2019 dollars)a b	7-8

Table 7-6 Net Benefits of the Final Rules Illustrative Scenario for 2024 to 2047 (billion 2019 dollars)a	7-9

Table 7-7 Net Benefits of the Alternative 1 Illustrative Scenario for 2024 to 2047 (billion 2019 dollars)a	7-11

Table 7-8 Net Benefits of the Alternative 2 Illustrative Scenario for 2024 to 2047 (billion 2019 dollars)a	7-13

Table A-l Interim SC-CO2 Values, 2028 to 2047 (2019 dollars per metric ton)	A-l

Table A-2 Stream of Projected Climate Benefits using Interim SC-CO2 values under the Final Rules from 2028

to 2047 (millions of 2019 dollars, discounted to 2024)	A-2

Table B-l Future-year Emissions Allocated to Each Modeled Coal EGU State Source Apportionment Tag.... B-4
Table B-2 Future-year Emissions Allocated to Each Modeled Natural Gas EGU State Source Apportionment

Tag	B-6

Table B-3 Future-year Emissions Allocated to the Modeled Other EGU Source Apportionment Tag	B-7

Table B-4 Baseline and Alternative 1 Scenario Ozone Scaling Factors for Coal EGU Tags	B-20

Table B-5 Alternative 2 and Final Rules Scenario Ozone Scaling Factors for Coal EGU Tags	B-22

Table B-6 Baseline and Alternative 1 Scenario Ozone Scaling Factors for Natural Gas EGU Tags	B-23

Table B-7 Alternative 2 and Final Rules Scenario Ozone Scaling Factors for Natural Gas EGU Tags	B-24

Table B-8 Baseline and Alternative 1 Nitrate Scaling Factors for Coal EGU tags	B-26

Table B-9 Alternative 2 and Final Rules Nitrate Scaling Factors for Coal EGU Tags	B-27

Table B-10 Baseline and Alternative 1 Nitrate Scaling Factors for Natural Gas EGU Tags	B-28

Table B-l 1 Alternative 2 and Final Rules Nitrate Scaling Factors for Natural Gas EGU Tags	B-31

Table B-12 Baseline and Alternative 1 Sulfate Scaling Factors for Coal EGU Tags	B-32

Table B-13 Alternative 2 and Final Rules Sulfate Scaling Factors for Coal EGU Tags	B-33

Table B-14 Baseline and Alternative 1 Primary PM2 5 Scaling Factors for Coal EGU Tags	B-35

Table B-15 Alternative 2 and Final Rules Primary PM2 5 Scaling Factors for Coal EGU Tags	B-36

Table B-16 Baseline and Alternative 1 Primary PM2 5 Scaling Factors for Natural Gas EGU Tags	B-37

Table B-17 Alternative 2 and Final Rules Primary PM2 5 Scaling Factors for Natural Gas EGU Tags	B-39

Table B-18 Baseline and Alternative 1 Scaling Factors for Other EGU Tags	B-40

Table B-19 Alternative 2 and Final Rules Scaling Factors for Other EGU Tags	B-40

Table D-l Summary of GHG Mitigation Measures for Existing Sources by Source Category under the Final

Rules "•	D-l

Table D-2 Summary of Modeled GHG Mitigation Measures for New Sources by Source Category under the

Final Rule "•	D-2

vi


-------
Table

D-3

Table

D-4

Table

D-5

Table

D-6

Table

D-7

Table

D-8

Table

D-9

Table

D-10

Table

D-ll

National Power Sector Compliance Cost Estimates for the Illustrative Scenarios (billions of 2019

dollars)	D-3

EGU Annual CO2 Emissions and Emissions Changes (million metric tons) for the Baseline and the

Illustrative Scenarios from 2028 to 2045	D-5

EGU Annual Emissions and Emissions Changes for Annual NOx, Ozone Season (April to September)

NOx, SO2, and Direct PM2 5 for the Baseline and Illustrative Scenarios for 2028 to 2040	D-6

2028, 2030, 2035, 2040 and 2045 Projected U.S. Power Sector Coal Use for the Baseline and the

Illustrative Scenarios	D-8

2028, 2030, 2035, 2040 and 2045 Projected Power Sector Natural Gas Use for the Baseline and the

Illustrative Scenarios	D-9

2028, 2030, 2035 and 2040 Projected Minemouth and Power Sector Delivered Coal Price (2019

dollars) for the Baseline and the Illustrative Scenarios	D-9

2028, 2030, 2035 and 2040 Projected Henry Hub and Power Sector Delivered Natural Gas Price

(2016 dollars) for the Baseline and the Illustrative Scenarios	D-10

2028, 2030, 2035 and 2040 Projected U.S. Generation by Fuel Type for the Baseline and the

Illustrative Scenarios	D-10

2028, 2030, 2035, 2040 and 2045 Projected U.S. Capacity by Fuel Type for the Baseline and the
Illustrative Scenarios	D-12

vii


-------
TABLE OF FIGURES

Figure 2-1 National Coal-fired Capacity (GW) by Age of EGU, 2023 	2-5

Figure 2-2 Average Annual Capacity Factor by Energy Source	2-8

Figure 2-3 Cumulative Distribution in 2021 of Coal and Natural Gas Electricity Capacity and Generation, by Age

	2-10

Figure 2-4 Fossil Fuel-Fired Electricity Generating Facilities, by Size	2-11

Figure 2-5 Selected Historical Mean LCOE Values	2-12

Figure 2-6 Real National Average Electricity Prices (including taxes) for Three Major End-Use Categories... 2-16
Figure 2-7 Relative Real Prices of Fossil Fuels for Electricity Generation; Change in National Average Real

Price per MMBtu Delivered to EGU	2-17

Figure 2-8 Relative Growth of Electricity Generation, Population and Real GDP Since 2010	2-18

Figure 2-9 Relative Change of Real GDP, Population and Electricity Generation Intensity Since 2010	2-19

Figure 3-1 Historical and Projected Capacity Mix (GW)	3-30

Figure 3-2 Historical and Projected Generation Mix (GW)	3-30

Figure 3-3 Electricity Market Module Regions	3-38

Figure 4-1 Data Inputs and Outputs for the BenMAP-CE Model Using PM2 5 as an Example	4-35

Figure 5-1 Depiction of the Circular Flow of the Economy	5-6

Figure 5-2 Hybrid Linkage Approach for IPM and SAGE	5-13

Figure 5-3 Percent Change in Real GDP and Components	5-21

Figure 5-4 Percent Change in Sectoral Output (Electricity, Coal, Natural Gas)	5-23

Figure 5-5 Percent Change in Sectoral Output (Rest of Economy)	5-24

Figure 5-6 Percent Change in Economy-wide Sectoral Output (All Sectors)	5-25

Figure 5-7 Percent Change in Real Output Prices	5-26

Figure 5-8 Percent Change in Economy-wide Labor Demand (All Sectors)	5-28

Figure 5-9 Percent Change in Labor Demand (Electricity, Coal, Natural Gas)	5-28

Figure 5-10 Percent Change in Labor Demand (Rest of Economy)	5-29

Figure 5-11 Distribution of General Equilibrium Social Costs	5-30

Figure 6-1 Number of People Residing in the Contiguous U. S. Areas Improving or Not Changing (Blue) or
Worsening (Orange) in 2028, 2030, 2035, 2040, and 2045 for PM2 5 and Ozone and the National
Average Magnitude of Pollutant Concentration Reductions (ng/m3 and ppb) for the 3 Regulatory

Options	6-23

Figure 6-2 Heat Map of the National Average PM2.5 Concentrations in the Baseline Across Demographic Groups

in 2028, 2030, 2035, 2040, and 2045 (ug/nf)	6-25

Figure 6-3 Heat Map of the Reductions in National Average PM2.5 Concentrations Due to the Three Illustrative

Scenarios Across Demographic Groups in 2028, 2030, 2035, 2040, and 2045 (ng/m3)	6-26

Figure 6-4 Heat Map of the State Average PM2.5 Concentration Reductions (Blue) and Increases (Red) Due to the
Final Rules Scenario Across Demographic Groups in 2028, 2030, 2035, 2040, and 2045 (ng/m3)

(Alternative Scenarios are shown in Appendix C)	6-28

Figure 6-5 Distribution of PM2.5 Concentration (ng/m3) Reductions Across Populations, Future Years for the

Final Rules Scenario (Alternative scenarios are shown in Appendix C)	6-30

Figure 6-6 Heat Map of the National Average Ozone Concentrations in the Baseline Across Demographic

Groups in 2028, 2030, 2035, 2040, and 2045 (ppb)	6-34

Figure 6-7 Heat Map of Reductions (Green) and Increases (Red) in National Average Ozone Concentrations Due
to the Three Regulatory Options Across Demographic Groups in 2028, 2030, 2035, 2040, and 2045

(ppb)	6-35

Figure 6-8 Heat Map of the State Average Ozone Concentrations Reductions (Green) and Increases (Red) Due to
the Final Rules Scenario Across Demographic Groups in 2028, 2030, 2035, 2040, and 2045 (ppb)

(Alternative Scenarios are shown in Appendix C)	6-37

Figure 6-9 Distributions of Ozone Concentration Changes (ppb) Across Populations, Future Years for the Final

Rules Scenario (Alternative Scenarios are shown in Appendix C)	6-39

Figure B-l Air Quality Modeling Domain	B-3

Vlll


-------
Figure B-2 Maps of California EGU Tag contributions to a) April-September Seasonal Average MDA8 Ozone

(ppb); b) Annual Average PM2.5 Nitrate (ng/m3); c) Annual Average PM2.5 sulfate (ng/m3); d) Annual

Average PM2.5 Organic Aerosol (ng/m3)	B-9

Figure B-3 Maps of Georgia EGU Tag contributions to a) April-September Seasonal Average MDA8 Ozone

(ppb); b) Annual Average PM2.5 Nitrate (ng/m3); c) Annual Average PM2.5 sulfate (ng/m3); d) Annual

Average PM2.5 Organic Aerosol (ng/m3)	B-10

Figure B-4 Maps of Iowa EGU Tag contributions to a) April-September Seasonal Average MDA8 Ozone (ppb);
b) Annual Average PM2.5 Nitrate (|ig/m3): c) Annual Average PM2.5 sulfate (|ig/m3): d) Annual

Average PM2.5 Organic Aerosol (ng/m3)	B-ll

Figure B-5 Maps of Ohio EGU Tag contributions to a) April-September Seasonal Average MDA8 Ozone (ppb);
b) Annual Average PM2 5 Nitrate (|ig/m3): c) Annual Average PM2.5 sulfate (ng/m3); d) Annual

Average PM2.5 Organic Aerosol (ng/m3)	B-12

Figure B-6 Maps of National EGU Tag contributions to April-September Seasonal Average MDA8 ozone (ppb)

by fuel for a) Coal EGUs; b) Natural Gas EGUs; c) All Other EGUs	B-13

Figure B-7 Maps of National EGU Tag contributions to Annual Average PM2.5 (ng/m3) by fuel for a) Coal EGUs;

b) Natural Gas EGUs; c) All Other EGUs	B-13

Figure B-8 Maps of ASM-03 in 2028	B-43

Figure B-9 Maps of ASM-03 in 2030	B-44

Figure B-10 Maps of ASM-03 in 2035	B-45

Figure B-l 1 Maps of ASM-03 in 2040	B-46

Figure B-12 Maps of ASM-03 in 2045	B-47

Figure B-13 Maps of PM2 5 in 2028	B-48

Figure B-14 Maps of PM2 5 in 2030	B-49

Figure B-15 Maps of PM2 5 in 2035	B-50

Figure B-16 Maps of PM2 5 in 2040	B-51

Figure B-17 Maps of PM2 5 in 2045	B-52

Figure B-18 Maps of changes in 2030 ASM-03 from 2028 baseline conditions	B-53

Figure B-19 Maps of changes in 2035 ASM-03 from 2028 baseline conditions	B-53

Figure B-20 Maps of changes in 2040 ASM-03 from 2028 baseline conditions	B-54

Figure B-21 Maps of changes in 2045 ASM-03 from 2028 baseline conditions	B-55

Figure B-22 Maps of changes in 2030 PM2.5 from 2028 baseline conditions	B-55

Figure B-23 Maps of changes in 2035 PM2.5 from 2028 baseline conditions	B-56

Figure B-24 Maps of changes in 2040 PM2.5 from 2028 baseline conditions	B-57

Figure B-25 Maps of changes in 2045 PM2.5 from 2028 baseline conditions	B-57

Figure C-l Heat Map of the State Average PM2 5 Concentration Reductions (Blue) and Increases (Red) Due to the
Alternative 1 Scenario Across Demographic Groups in 2028, 2030, 2035, 2040, and 2045 (|ig/m3) C-l
Figure C-2 Heat Map of the State Average PM2.5 Concentration Reductions (Blue) and Increases (Red) Due to the
Alternative 2 Scenario Across Demographic Groups in 2028, 2030, 2035, 2040, and 2045 (|ig/m3) C-2
Figure C-3 Distribution of PM2.5 Concentration (|ig/m3) Reductions Across Populations, Future Years for the

Alternative 1 Scenario	C-3

Figure C-4 Distribution of PM2.5 Concentration (|ig/m3) Reductions Across Populations, Future Years for the

Alternative 2 Scenario	C-4

Figure C-5 Heat Map of the State Average Ozone Concentrations Reductions (Green) and Increases (Red) Due to
the Alternative 1 Scenario Across Demographic Groups in 2028, 2030, 2035, 2040, and 2045 (ppb) C-

5

Figure C-6 Heat Map of the State Average Ozone Concentrations Reductions (Green) and Increases (Red) Due to
the Alternative 2 Scenario Across Demographic Groups in 2028, 2030, 2035, 2040, and 2045 (ppb) C-

6

Figure C-l Distributions of Ozone Concentration Changes (ppb) Across Populations, Future Years for the

Alternative 1 Scenario	C-l

Figure C-8 Distributions of Ozone Concentration Changes (ppb) Across Populations, Future Years for the

Alternative 2 Scenario	C-8

IX


-------
EXECUTIVE SUMMARY

ES.l Introduction

In 2009, the EPA concluded that GHG emissions endanger our nation's public health and
welfare.1 Since that time, the evidence of the harms posed by GHG emissions has only grown,
and Americans experience the destructive and worsening effects of climate change every day.2
Fossil fuel-fired EGUs are the nation's largest stationary source of GHG emissions, representing
25 percent of the United States' total GHG emissions in 2021.3 At the same time, a range of cost-
effective technologies and approaches to reduce GHG emissions from these sources is available
to the power sector—including carbon capture and seque strati on/storage (CCS), co-firing with
less GHG-intensive fuels, and more efficient generation. Congress has also acted to provide
funding and other incentives to encourage the deployment of various technologies, including
CCS, to achieve reductions in GHG emissions from the power sector.

In this notice, the EPA is finalizing several actions under section 111 of the Clean Air
Act (CAA) to reduce the significant quantity of GHG emissions from fossil fuel-fired EGUs by
establishing emission guidelines and new source performance standards (NSPS) that are based
on available and cost-effective technologies that directly reduce GHG emissions from these
sources. Consistent with the statutory command of CAA section 111, the final NSPS and
emission guidelines reflect the application of the best system of emission reduction (BSER) that,
taking into account costs, energy requirements, and other statutory factors, is adequately
demonstrated.

Specifically, the EPA is first finalizing the repeal of the Affordable Clean Energy (ACE)
Rule. Second, the EPA is finalizing emission guidelines for GHG emissions from existing fossil
fuel-fired steam generating EGUs, which include both coal-fired and oil/gas-fired steam
generating EGUs. Third, the EPA is finalizing revisions to the NSPS for GHG emissions from
new and reconstructed fossil fuel-fired stationary combustion turbine EGUs. Fourth, the EPA is
finalizing revisions to the NSPS for GHG emissions from fossil fuel-fired steam generating units

1	74 FR 66496 (December 15, 2009).

2	The 5th National Climate Assessment (NCA5) states that the effects of human-caused climate change are already

far-reaching and worsening across every region of the United States and that climate change affects all aspects of
the energy system-supply, delivery, and demand-through the increased frequency, intensity, and duration of
extreme events and through changing climate trends.

3	https://www.epa.gov/ghgemissions/sources-greenhouse-gas-emissions.

ES-1


-------
that undertake a large modification, based upon the 8-year review required by the CAA. The
EPA is not finalizing emission guidelines for GHG emissions from existing fossil fuel-fired
combustion turbines at this time and plans to expeditiously issue an additional proposal that more
comprehensively addresses GHG emissions from this portion of the fleet. The EPA
acknowledges that the share of GHG emissions from existing fossil fuel-fired combustion
turbines has been growing and is projected to continue to do so, particularly as emissions from
other portions of the fleet decline, and that it is vital to regulate the GHG emissions from these
sources consistent with CAA section 111.

These final actions ensure that the new and existing fossil fuel-fired EGUs that are
subject to these rules reduce their GHG emissions in a manner that is cost-effective and improves
the emissions performance of the sources, consistent with the applicable CAA requirements and
caselaw. These standards and emission guidelines will significantly decrease GHG emissions
from fossil fuel-fired EGUs and the associated harms to human health and welfare. Further, the
EPA has designed these standards and emission guidelines in a way that is compatible with the
nation's overall need for a reliable supply of affordable electricity.

In accordance with Executive Order (E.O). 12866 and 13563, the guidelines of OMB
Circular A-4 and EPA's Guidelines for Preparing Economic Analyses (U.S. EPA, 2014), the
RIA analyzes the benefits and costs associated with the projected emissions reductions under the
requirements of the final rules, and two alternative sets of requirements to inform EPA and the
public about these projected impacts.4

ES.2 Regulatory Requirements

These final actions include the repeal of the ACE Rule, BSER determinations and
emission guidelines for existing fossil fuel-fired steam generating units, and BSER
determinations and accompanying standards of performance for GHG emissions from new and
reconstructed fossil fuel-fired stationary combustion turbines and modified fossil fuel-fired steam

4 Circular A-4 was recently revised. The effective date of the revised Circular A-4 (2023) is March 1, 2024, for
regulatory analyses received by OMB in support of proposed rules, interim final rules, and direct final rules, and
January 1, 2025, for regulatory analyses received by OMB in support of other final rules. For all other rules,
Circular A-4 (2003) is applicable until those dates.

ES-2


-------
generating units. See Section I.C of the final rule preamble for a summary of the major
provisions of these regulations.

ES.3 Baseline and Analysis Years

The impacts of final regulatory actions are evaluated relative to a modeled baseline that
represents expected behavior in the electricity sector under market and regulatory conditions in
the absence of a regulatory action. EPA used the Integrated Planning Model (IPM)5 to conduct
the electric generating units (EGU) analysis discussed in this section, relying on the EPA's
Power Sector Platform 2023 Using IPM to establish the baseline for this analysis. For a detailed
description please see Section 3 of this document. EPA frequently updates the power sector
modeling baseline to reflect the latest available electricity demand forecasts from the U.S.
Energy Information Administration (EIA) at the time the modeling was completed as well as
expected costs and availability of new and existing generating resources, fuels, emission control
technologies, and regulatory requirements. The electricity supply baseline includes the final
Good Neighbor Plan (GNP), the Revised Cross-State Air Pollution Rule (CSAPR) Update,
CSAPR Update, and CSAPR, as well as the 2012 Mercury and Air Toxics Standards. The power
sector baseline also includes the 2015 Effluent Limitation Guidelines (ELG) and the 2015 Coal
Combustion Residuals (CCR), and the recently finalized 2020 ELG and CCR rules. This version
of the model ("EPA's Power Sector Platform 2023 Using IPM") also includes recent updates to
state and federal legislation affecting the power sector, including Public Law 117-169, 136 Stat.
1818 (August 16, 2022), commonly known as the Inflation Reduction Act of 2022 (IRA)6. The
modeling documentation, available in the docket, includes a summary of all legislation reflected
in this version of the model as well as a description of how that legislation is implemented in the

5	Information on IPM can be found at the following link: https://www.epa.gov/airmarkets/power-sector-modeling.

6	The Inflation Reduction Act (IRA) contains tax credit provisions that affect power sector operations, which are

incorporated into the IPM modeling. Details are included the IPM documentation. The Clean Electricity
Investment and Production Tax Credits (provisions 48E and 45Y of the IRA) are described in more detail in
Section 4. The credit for Carbon Capture and Sequestration (provision 45Q) is described in Section 3. The
impacts of the Zero-Emission Nuclear Power Production Credit (provision 45U) are reflected through modifying
nuclear retirement limits, as described in Section 4. The Credit for the Production of Clean Hydrogen (provision
45V) is reflected through the inclusion of an exogenously delivered price of hydrogen fuel, see Section 9. The
Advanced Manufacturing Production Tax Credit (45X) was reflected through adjustments to the short-term
capital cost added for renewable technologies, see Section 4. Documentation available at:
https://www.epa.gov/airmarkets/power-sector-modeling

ES-3


-------
model.7 Also, see Section 3 for additional detail about the power sector baseline for this RIA.
Additionally, EPA conducted sensitivity analyses that examined the impacts of several recently
finalized EPA rulemakings (including MATS RTR, ELG, and EPA's LDV, MDV and HDV
vehicle rules) in addition to the 111 rulemaking, as well as exploring alternative assumptions
around gas prices and demand.

This RIA evaluates the benefits, costs, and certain impacts of compliance with three
illustrative scenarios: the final rules and two alternative regulatory option scenarios which
assume both existing and new source GHG mitigation requirements. For details of the controls
modeled for each of the source categories under the three illustrative scenarios, please see
Section 3.2 of this RIA.

We evaluated the potential benefits, costs, and net benefits of the three illustrative
scenarios for the years 2024 to 2047 from the perspective of 2024, using the discount rates of
two percent, three percent, and seven percent.8 In addition, the Agency presents the assessment
of costs, benefits, and net benefits for specific snapshot years, consistent with historic practice.
These snapshot years are 2028, 2030, 2035, 2040 and 2045. The Agency believes that these
specific years are each representative of several surrounding years, which enables the analysis of
costs and benefits over the timeframe of 2024 to 2047. The year 2028 is the first year of detailed
power sector modeling for this RIA and approximates when the regulatory impacts of the final
111(b) new source performance standards on the power sector will begin. However, because the
Agency estimates that some monitoring, reporting, and recordkeeping (MR&R) costs may be
incurred in 2024, we analyze compliance costs in years before 2028. Therefore, while MR&R
costs analysis is presented beginning in the year 2024, the detailed assessment of costs,
emissions impacts, and benefits begins in the year 2028. The analysis timeframe concludes in
2047, as this is the last year that may be represented with the analysis conducted for the specific

7	For a discussion on the impacts of the IRA from a range of models including EPA IPM, please see: Bistline, J., et

al., Emissions and energy impacts of the Inflation Reduction Act. Science, 2023. 380(6652): p. 1324-1327. DOI:
10.1126/science. adg3781. Available from: https://www.science.org/doi/10.1126/science.adg3781

8	Results using the 2 percent discount rate were not included in the proposals for these actions. The 2003 version of

OMB 's Circular A-4 had generally recommended 3 percent and 7 percent as default rates to discount social costs
and benefits. The analysis of the proposed rule used these two recommended rates. In November 2023, OMB
finalized an update to Circular A-4, in which it recommended the general application of a 2 percent rate to
discount social costs and benefits (subject to regular updates), which is an estimate of consumption-based
discount rate. We include cost and benefits results calculated using a 2 percent discount rate consistent with the
update to Circular A-4 (OMB, 2023).

ES-4


-------
year of 2045. While the results are described and presented in more detail later in this executive
summary and throughout the RIA, we present the high-level results of the analysis here.

ES.4 Emissions Impacts

Table ES-1 presents the estimated impact on power sector emissions in the contiguous
U.S. resulting from compliance with the final rules as modeled by the illustrative final rules
scenario. The projections indicate that the illustrative final rules scenario results in national,
annual emission reductions of CO2, direct PM2.5, NOx, and SO2 for each of the snapshot years
analyzed. The illustrative alternative 1 and alternative 2 scenarios result in national emission
reductions of CO2, NOx, and SO2 throughout the year for each of the snapshot years analyzed
but increases in 2040 in direct PM2.5 and mercury. Under the illustrative final rules scenario, the
cumulative CO2 emission reductions over the 2028 to 2047 timeframe are estimated to be 1,382
million metric tons. Under the alternative 1 and alternative 2 illustrative scenarios, cumulative
CO2 emission reductions over the 2028 to 2047 timeframe are estimated to be 1,365 million
metric tons and 1,303 million metric tons, respectively.9

9 See Table 4-2 for annual CO2 emission reductions.

ES-5


-------
Table ES-1 Projected EGU Emissions and Emissions Changes for the Three Illustrative
Scenarios for 2028, 2030, 2035, 2040, and 2045 a	



CO2 (million
metric tons)

Annual NOx
(thousand
short tons)

Ozone
Season NOx
(thousand
short tons)b

Annual SO2
(thousand
short tons)

Direct PM2.5
(thousand
short tons)

Mercury
(tons)

Final Rules



2028

-38

-20

-6

-34

-2

-0.1

2030

-50

-20

-7

-20

-2

-0.1

2035

-123

-49

-19

-90

-1

-0.1

2040

-54

-6

-6

-4

2

0.2

2045

-42

-24

-14

-41

-2

-0.2

Alternative 1



2028

-36

-19

-6

-30

-2

-0.1

2030

-48

-20

-7

-16

-2

-0.1

2035

-124

-51

-20

-90

-2

-0.1

2040

-53

-7

-6

-4

1

0.2

2045

-40

-24

-14

-41

-2

-0.2

Alternative 2



2028

-32

-17

-5

-28

-1

-0.1

2030

-27

-11

-4

-15

-1

0.0

2035

-122

-48

-18

-94

-1

-0.1

2040

-53

-5

-6

-8

2

0.3

2045

-40

-24

-14

-41

-2

-0.1

3 This analysis is limited to the geographically contiguous lower 48 states.
b Ozone season is the May through September period in this analysis.

ES.5 Compliance Costs

The compliance cost estimates presented in this RIA are based on IPM projection
supplemented with cost estimates for MR&R. As described previously, this RIA evaluates three
illustrative scenarios: the final rules, alternative 1, and alternative 2. The alternative 1 and
alternative 2 scenarios assume the definition of annual capacity factor for baseload operation for
new turbines is 50 percent, whereas under the final rules scenario baseload is defined as 40
percent annual capacity factor. The final rules and alternative 1 scenarios assume all medium-
term existing coal-fired steam generating units must co-fire at least 40 percent natural gas by
2030, while the alternative 2 scenario assumes that all medium-term existing coal fired steam
generating units must co-fire at least 40 percent natural gas by 2035.

Table ES-2 below summarizes the present value (PV) and equivalent annualized value
(EAV) of the total national compliance cost estimates10 for the illustrative final rules scenario

10 Compliance costs refer to the difference between policy and baseline IPM projected capital, O&M, fuel,

transmission, and CO2 storage and transportation costs. Other costs are not accounted for. Please see Sections 3.5
and 5.2 for further discussion of the differences between compliance costs and social costs.

ES-6


-------
and the alternative scenarios. We present the PV of the costs over the 24-year period of 2024 to
2047. We also present the equivalent annualized value (EAV), which represents a flow of
constant annual values that, had they occurred annually, would yield a sum equivalent to the PV.
The EAV represents the value of a typical cost for each year of the analysis. Section 3 reports
how annual power costs are projected to change over the time period of analysis.11

Table ES-2 Total National Compliance Cost Estimates for the Three Illustrative
Scenarios (discounted to 2024, billion 2019 dollars)	



2% Discount Rate

3% Discount Rate

7% Discount Rate

PV

EAV

PV

EAV

PV

EAV

Final Rules

19

0.98

15

0.91

7.5

0.65

Alternative 1

19

0.99

16

0.93

7.8

0.68

Alternative 2

19

0.98

15

0.91

7.2

0.63

Note: Values have been rounded to two significant figures.

Projected compliance costs are similar across the scenarios. Costs peak in 2035 across all
scenarios, reflecting the date of imposition of the final Emission Guidelines for coal-fired steam
generating units and tightening NSPS requirements. The final rules scenario results in the
greatest early buildout of renewable energy, resulting in the lowest near-term costs and higher
longer-term costs. As a result, over the 2024 - 2047 time period, the final rules scenario shows
lower costs than alternative 1. However, over the entire forecast period of 2024 - 2055, costs are
higher under the final rules.

Tax credits under the IRA directly subsidize lower emitting generation (i.e., CCS, RE
etc.). The baseline as a result continues to show reductions in thermal generation and capacity
share over the forecast period. The addition of the policy further accentuates these trends, but the
IRA continues to play an important role. As a result, costs remain lower than they would absent
the IRA.

11 Results using the 2 percent discount rate were not included in the proposal for this action. The 2003 version of
OMB 's Circular A-4 had generally recommended 3 percent and 7 percent as default rates to discount social costs
and benefits. The analysis of the proposed rule used these two recommended rates. In November 2023, OMB
finalized an update to Circular A-4, in which it recommended the general application of a 2 percent rate to
discount social costs and benefits (subject to regular updates), which is an estimate of consumption-based
discount rate. We include cost and benefits results calculated using a 2 percent discount rate consistent with the
update to Circular A-4 (OMB, 2023).

ES-7


-------
ES.6 Benefits

This section reports the estimated monetized climate and health benefits associated with
emission reductions for each of the three illustrative scenarios described in prior sections and
discusses other unquantified benefits.

ES.6.1 Climate Benefits

Elevated concentrations of GHGs in the atmosphere have been warming the planet,
leading to changes in the Earth's climate including changes in the frequency and intensity of heat
waves, precipitation, and extreme weather events, rising seas, and retreating snow and ice. The
well-documented atmospheric changes due to anthropogenic GHG emissions are changing the
climate at a pace and in a way that threatens human health, society, and the natural environment.
Climate change touches nearly every aspect of public welfare in the U.S. with resulting
economic costs, including: changes in water supply and quality due to changes in drought and
extreme rainfall events; increased risk of storm surge and flooding in coastal areas and land loss
due to inundation; increases in peak electricity demand and risks to electricity infrastructure; and
the potential for significant agricultural disruptions and crop failures (though offset to some
extent by carbon fertilization).

There will be important climate benefits associated with the CO2 emissions reductions
expected from these final rules. Climate benefits from reducing emissions of CO2 are monetized
using estimates of the social cost of carbon (SC-CO2) that reflect recent advances in the scientific
literature on climate change and its economic impacts and incorporate recommendations made
by the National Academies of Science, Engineering, and Medicine (National Academies,
2017). As noted in the proposal for these rulemakings, the EPA presented these updated SC-CO2
estimates in sensitivity analysis in the RIA for the agency's December 2022 Oil and Gas
NSPS/EG Supplemental Proposed Rulemaking, "Standards of Performance for New,
Reconstructed, and Modified Sources and Emissions Guidelines for Existing Sources: Oil and
Natural Gas Sector Climate Review". The EPA solicited public comment and conducted an
external peer review of these estimates, and the Agency has used the estimates in the RIA for the
December 2023 Final Oil and Gas NSPS/EG Rulemaking (U.S. EPA, 2023b). See Section 4.2 of

ES-8


-------
this RIA for more discussion of the approach to monetization of the climate benefits associated
with these rules.

ES. 6.2 Health Benefits

These rules are expected to reduce annual, national emissions of direct PM2.5, NOx, and
SO2. Because NOx and SO2 are also precursors to secondary formation of ambient PM2.5,
reducing these emissions would reduce human exposure to ambient PM2.5 throughout the year
and would reduce the incidence of PIVh.s-attributable health effects. These final rules are
expected to reduce ozone season NOx emissions. In the presence of sunlight, NOx, and volatile
organic compounds (VOCs) undergo chemical reactions in the atmosphere resulting in ozone
formation. Reducing NOx emissions reduces human exposure to ozone and the incidence of
ozone-related health effects in most locations, though ozone response to NOx emissions
reductions depends on local conditions.

In this RIA, EPA estimates national-level health benefits resulting from the changes in
PM2.5 and ozone concentrations expected to occur with these final rules. The health effect
endpoints, effect estimates, and benefit unit-values, and how they were selected, are described in
the Technical Support Document (TSD) titled Estimating PM2.5- and Ozone-Attributable Health
Benefits (U.S. EPA, 2023a). Our approach for updating the endpoints and to identify suitable
epidemiological studies, baseline incidence rates, population demographics, and valuation
estimates is summarized in Section 4.3.

ES. 6.3 Additional Unquantified Benefits

Data, time, and resource limitations prevented EPA from quantifying the estimated health
impacts or monetizing estimated benefits associated with direct exposure to hazardous air
pollutants (HAPs), NO2, and SO2, independent of the role NO2 and SO2 play as precursors to
PM2.5 and ozone. In addition, these limitations prevented quantification of welfare benefits
accrued due to reduced pollutant impacts on ecosystem and reductions in visibility impairment.
While all health benefits and welfare benefits were not able to be quantified, it does not imply
that there are not additional benefits associated with reductions in exposures to HAPs, ozone,
PM2.5, NO2, or SO2. For a qualitative description of these and potential water quality benefits,
please see Section 4.4 of this RIA.

ES-9


-------
ES.6.4 Total Climate and Health Benefits

Table ES-3 presents the total monetized climate and health benefits for the illustrative
final rules scenario and the alternative scenarios.12

12 Monetized climate benefits are discounted using a 2 percent discount rate, consistent with EPA's updated

estimates of the SC-CO2. OMB has long recognized that climate effects should be discounted only at appropriate
consumption-based discount rates. Because the SC-CO2 estimates reflect net climate change damages in terms of
reduced consumption (or monetary consumption equivalents), the use of the social rate of return on capital (7
percent under OMB Circular A-4 (2003)) to discount damages estimated in terms of reduced consumption would
inappropriately underestimate the impacts of climate change for the purposes of estimating the SC-CO2. See
Section 4.2 for more discussion.

ES-10


-------
Table ES-3 Benefits for the Three Illustrative Scenarios, (discounted to 2024, billion 2019
dollars)"	

All Benefits Calculated using 2% Discount Rate



Climate Benefits b

PM2.5 and 03-related
Health Benefits c

Total Benefits d



PV

EAV

PV EAV

PV

EAV

Final Rules

270

14

120 6.30

390

21

Alternative 1

270

14

120 6.50

390

21

Alternative 2

250

13

120 6.30

370

20





Climate Benefits Calculated using 2% Discount Rate,
Health Benefits Calculated using 3% Discount Rate





Climate Benefits b

PM2.5 and 03-related
Health Benefits c

Total Benefits d



PV

EAV

PV EAV

PV

EAV

Final Rules

270

14

100 6.1

370

20

Alternative 1

270

14

110 6.2

370

20

Alternative 2

250

13

100 6.1

360

20





Climate Benefits Calculated using 2% Discount Rate,
Health Benefits Calculated using 7% Discount Rate





Climate Benefits b

PM2.5 and 03-related
Health Benefits c

Total Benefits d



PV

EAV

PV EAV

PV

EAV

Final Rules

270

14

59 5.2

330

19

Alternative 1

270

14

60 5.2

330

19

Alternative 2

250

13

58 5.1

310

19

Non-Monetized Benefits d

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
b Monetized climate benefits are based on reductions in CO2 emissions and are calculated using three different
estimates of the social cost of CO2 (SC-CO2) (under 1.5 percent, 2.0 percent, and 2.5 percent near-term Ramsey
discount rates). For the presentational purposes of this table, we show the climate benefits associated with the SC-
CO2 at the 2 percent near-term Ramsey discount rate. Please see Table 4-5 for the full range of monetized climate
benefit estimates. See Section 4.2 for a discussion of the uncertainties associated with the climate benefit estimates.
c For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates. For discussions of the uncertainty associated with these health benefits estimates, see Section 4.3.8.
d Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

ES-11


-------
ES.7 Economic Impacts

As a result of the compliance costs incurred by the regulated sector, these final actions
have economic and energy market implications. The energy impact estimates presented here
reflect EPA's illustrative analysis of the final rules. States are afforded flexibility to implement
the final rules, and thus the impacts could be different to the extent states make different choices
than those assumed in the illustrative analysis. Table ES-4 presents a variety of energy market
impact estimates for 2028, 2030, 2035, 2040, and 2045 for the illustrative final rules scenario,
relative to the baseline. These results are EPA's best estimate of possible compliance pathways
with the final rules. However, there are several key areas of uncertainty inherent in these
projections as outlined in Section 3.7.

Table ES-4 Summary of Certain Energy Market Impacts for the Illustrative Final Rules
Scenario Relative to the Baseline



2028

2030

2035

2040

2045

Retail electricity prices

-1%

0%

1%

0%

1%

Average price of coal delivered to the power sector

-1%

-1%

0%

0%

-32%

Coal production for power sector use

-6%

-4%

-21%

15%

-84%

Price of natural gas delivered to power sector

-2%

0%

3%

0%

0%

Price of average Henry Hub (spot)

-2%

-1%

3%

0%

0%

Natural gas use for electricity generation

-1%

-2%

4%

0%

2%

These and other energy market impacts are discussed more extensively in Section 3 of the
RIA, and a more detailed version of the table is available in Section 5 of the RIA.

More broadly, changes in production in a directly regulated sector may have effects on
other markets when output from that sector - for these final rules, electricity - is used as an input
in the production of other goods. The final rules may affect upstream industries that supply
goods and services to the sector, along with labor and capital markets, as well as changes in
household consumption patterns due to changes in the price of electricity and other final goods
prices. Changes in firm and household behavior in response to the final rules could also interact
with pre-existing distortions, such as taxes, resulting in additional social costs. Computable
general equilibrium (CGE) models are analytical tools that can be used to evaluate the broad
economy-wide impacts of a regulatory action and its social cost by including interactions and
feedbacks between them. In response to a Science Advisory Board recommendation (U.S. EPA

ES-12


-------
Science Advisory Board, 2017), EPA developed a new CGE model for the U.S. economy called
SAGE designed for use in regulatory analysis, which was peer reviewed (U.S. EPA Science
Advisory Board, 2020).

EPA used SAGE to evaluate the economy-wide social costs and economic impacts of
these final rules. Note that SAGE does not currently estimate changes in emissions nor account
for environmental benefits. The annualized social cost estimated in SAGE for the finalized rules
is approximately $1.32 billion (2019 dollars) between 2024 and 2047 using a 4.5 percent
discount rate that is consistent with the internal discount rate in the model. Under the assumption
that compliance costs from IPM in 2056 continue until 2081, the equivalent annualized value for
social costs in the SAGE model is $1.51 billion (2019 dollars) over the period from 2024 to
2081, again using a 4.5 percent discount rate that is consistent with the internal discount rate of
the model. The social cost estimate reflects the combined effect of the finalized rules'
requirements and interactions with IRA subsidies for specific technologies that are expected to
see increased use in response to the finalized rules. We are not able to identify their relative roles
at this time. The social cost estimates in the economy-wide analysis discussed in Section 5.2 are
substantially lower than the projected benefits of the final rules. The economy-wide analysis is
considered a complement to the more detailed evaluation of sector costs produced by IPM. A
detailed discussion of the social costs and distributional impacts of the final rules is contained in
Section 5.2 of this RIA.13

Environmental regulation may affect groups of workers differently, as changes in
abatement and other compliance activities cause labor and other resources to shift. An
employment impact analysis describes the characteristics of groups of workers potentially
affected by a regulation, as well as labor market conditions in affected occupations, industries,
and geographic areas. Employment impacts of these final actions are discussed more extensively
in Section 5 of the RIA.

13 Section 5.2 also discusses the differences between social costs estimated by SAGE and the compliance costs
estimated by IPM. For example, IPM estimates compliance costs incurred by firms in the electricity sector, but
because of the availability of subsidy payments, there are additional real resource costs to the economy outside of
the regulated sector. To estimate the social costs for the economy as a whole, EPA has used information from
IPM as an input into SAGE. The economy-wide analysis is considered a complement to the more detailed
evaluation of sector costs produced by IPM.

ES-13


-------
ES.8 Environmental Justice Impacts

Environmental justice (EJ) concerns for each rulemaking are unique and should be
considered on a case-by-case basis, and EPA's EJ Technical Guidance (2015) states that "[t]he
analysis of potential EJ concerns for regulatory actions should address three questions:

1.	Are there potential EJ concerns associated with environmental stressors affected by the
regulatory action for population groups of concern in the baseline?

2.	Are there potential EJ concerns associated with environmental stressors affected by the
regulatory action for population groups of concern for the regulatory option(s) under
consideration?

3.	For the regulatory option(s) under consideration, are potential EJ concerns created or
mitigated compared to the baseline?"

To address these questions, EPA developed an analytical approach that considers the
purpose and specifics of the rulemaking, as well as the nature of known and potential exposures
and impacts. For the final rules, we quantitatively evaluate 1) the proximity of affected facilities
to communities with EJ concerns for consideration of local pollutants impacted by these rules but
not modeled here (Section 6.4), and 2) the distribution of ozone and PM2.5 concentrations in the
baseline and changes due to the final rulemaking across different demographic groups on the
basis of race, ethnicity, poverty status, employment status, health insurance status, life
expectancy, redlining, Tribal land, age, sex, educational attainment, and degree of linguistic
isolation (Section 6.5). While these analyses assess the distribution of non-climate impacts at
more near-term and local spatial scales, we also discuss potential EJ climate impacts from
projected long-term climate change (Section 6.3). Each of these analyses was performed to
answer separate questions and is associated with unique limitations and uncertainties.

Baseline demographic proximity analyses provide information as to whether there may
be potential EJ concerns associated with environmental stressors, such as local HAP, emitted
from sources affected by the regulatory action for certain population groups of concern (Section
6.4). The baseline demographic proximity analyses examined the demographics of populations
living within 5 km, 10 km, and 50 km of the following three sets of sources: 1) 114 coal plants
with units subject to the 111 final rules, 2) 23 coal plants (a subset of the 114) with known

ES-14


-------
retirement plans between 2033 and 2040, with units subject to the 111 final rules, and 3) 94 coal
plants (a subset of the 114) without known retirement plans before 2040, with units subject to the
final rules. See Section 6.4 regarding data limitations for the 5 km proximity analysis. The
proximity analysis of the full population of potentially affected units greater than 25 MW (114
facilities) indicated that the demographic percentages of the population within 5 km and 10 km
of the facilities are relatively similar to the national averages with the exception of the American
Indian population (1 percent and 0.8 percent, respectively) that is higher than the national
average (0.6 percent). This higher percentage is driven mostly by 7 facilities that have an
American Indian percentage within a 5 km and 10 km radius that ranges from 10 percent to just
over 40 percent which is substantially above the national average (0.6 percent). Also within a 5
km and 10 km radius, the population living below the federal poverty line (14 percent for both
distances) as well as the population living below 2x the federal poverty line (34 percent and 33
percent, respectively) are both higher than the national averages (13 percent and 29 percent,
respectively). The proximity analysis of the 23 units with known retirement plans between 2033
and 2040 (a subset of the total 114 units) found that the percentages of the population within 5
km and 10 km that is below the poverty line (14 percent for both distances) and below 2x the
federal poverty line (33 percent and 31 percent, respectively) are both higher than the national
averages (13 percent and 29 percent, respectively). The proximity analysis for the 94 units
without known retirement plans before 2040 (a subset of the total 114 units) shows
demographics similar to the 114 facilities' proximity analysis.

Baseline ozone and PM2.5 exposure analyses show that certain populations, such as
Hispanic populations, Asian populations, those linguistically isolated, and those less educated
will experience disproportionately higher ozone and PM2.5 exposures as compared to the national
average (Section 6.5). Black populations will also experience disproportionately higher PM2.5
concentrations than the reference group, and American Indian populations and children will
experience disproportionately higher ozone concentrations than the reference group. Therefore,
there likely are potential EJ concerns for population groups of concern in the baseline.

Finally, we evaluate how these final rulemakings are expected to differentially impact
demographic populations with regard to ozone and PM2.5 exposure changes. Our analysis
indicates that the final rules will secure modest widespread reductions in ozone and PM2.5
pollution. Relative to 2028 baseline conditions, our analysis indicates that ozone and PM2.5

ES-15


-------
concentrations will decline in virtually all areas of the country. However, some areas of the
country may experience slower or faster rates of decline in ozone and PM2.5 pollution over time
as a result of the changes in generation and utilization resulting from the rule. In all years, the
final rules are expected to result in reductions in ozone concentrations over many areas of the
US, although some areas, in some analysis periods, may experience increases in ozone
concentrations relative to forecasted conditions without the rule. The extent of areas
experiencing ozone increases varies among snapshot years. As a general rule, however, these
changes in PM2.5 and ozone concentrations are small (approximately 0.4 percent relative to the
baseline) such that baseline disparities in the ozone and PM2.5 concentration burdens are likely to
remain after implementation of the regulatory action. This EJ assessment also suggests that this
action is unlikely to mitigate or exacerbate PM2.5 exposures disparities across populations of EJ
concern analyzed. Regarding ozone exposures, the final rules will not likely mitigate or
exacerbate ozone exposure disparities for the population groups evaluated in most years;
however, ozone exposure disparities may be slightly exacerbated for some population groups
analyzed in 2035 and those living on Tribal lands in 2040, as well as slightly mitigated for those
living on Tribal lands in 2028 and 2030.

ES.9 Comparison of Benefits and Costs

In this RIA, the regulatory impacts are evaluated for the specific snapshot years of 2028,
2030, 2035, 2040, and 2045, and MR&R costs are estimated for all years in the 2024 to 2047
timeframe. Comparisons of benefits to costs for the snapshot years of 2028, 2030, 2035, 2040,
and 2045 are presented in Section 7 of this RIA. Here we present the PV and EAV of costs,
benefits, and net benefits, calculated for the years 2024 to 2047 from the perspective of 2024,
using the discount rates of two percent, three percent, and seven percent. All dollars are in 2019
dollars. The compliance cost estimates are net of changes in renewable energy, hydrogen, and
CCS subsidies.

We also present the EAV, which represents a flow of constant annual values that, had
they occurred in each year from 2024 to 2047, would yield a sum equivalent to the PV. The EAV
represents the value of a typical cost or benefit for each year of the analysis, in contrast to the
year-specific estimates reported in the costs and benefits sections of this RIA.

ES-16


-------
The comparison of benefits and costs in PV and EAV terms for the illustrative final rules
scenario and alternative 1 and 2 scenarios can be found in Table ES-5. Estimates in the tables are
presented as rounded values.

ES-17


-------
Table ES-5 Net Benefits of the Illustrative Scenarios (billions of 2019 dollars, discounted
to 2024)a	

All Values Calculated using 2% Discount Rate



Climate

PM2.5 and 03-related

Compliance

Net Benefits



Benefits b

Health Benefits c



Costs

d



PV

EAV

PV EAV

PV

EAV

PV EAV

Final Rules

270

14

120 6.3

19

0.98

370 20

Alternative 1

270

14

120 6.5

19

0.99

370 20

Alternative 2

250

13

120 6.3

19

0.98

360 19

Compliance Cost and Health Benefits Calculated using 3% Discount Rate,
Climate Benefits Calculated using 2% Discount Rate

Climate	PM2 s and 03-related	Compliance Net Benefits

Benefitsb	Health Benefitsc	Costs	d



PV

EAV

PV

EAV

PV

EAV

PV

EAV

Final Rules

270

14

100

6.1

15

0.91

360

19

Alternative 1

270

14

110

6.2

16

0.93

360

19

Alternative 2

250

13

100

6.1

15

0.91

340

19

Compliance Cost and Health Benefits Calculated using 7% Discount Rate,
Climate Benefits Calculated using 2% Discount Rate

Climate	PM2.5 and 03-related	Compliance Net Benefits

Benefitsb	Health Benefitsc	Costs	d



PV

EAV

PV

EAV

PV

EAV

PV

EAV

Final Rules

270

14

59

5.2

7.5

0.65

320

19

Alternative 1

270

14

60

5.2

7.8

0.68

320

19

Alternative 2

250

13

58

5.1

7.2

0.63

310

18

Non-Monetized Benefitse

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
b Monetized climate benefits are based on reductions in CO2 emissions and are calculated using three different
estimates of the social cost of CO2 (SC-CO2) (under 1.5 percent, 2.0 percent, and 2.5 percent near-term Ramsey
discount rates). For the presentational purposes of this table, we show the climate benefits associated with the SC-
CO2 at the 2 percent near-term Ramsey discount rate. Please see Table 4-5 for the full range of monetized climate
benefit estimates. See Section 4.2 for a discussion of the uncertainties associated with the climate benefit estimates.
c For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates. For discussions of the uncertainty associated with these health benefits estimates, see Section 4.3.8.
d In this net benefits analysis, health benefits and costs are discounted at the rates shown in the table (i.e., two
percent, three percent, and seven percent). Climate benefits are discounted using a two percent discount rate only in
this net benefits analysis.

ES-18


-------
e Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

As discussed in Section 4 of this RIA, the monetized benefits estimates provide an
incomplete overview of the beneficial impacts of the final rules. In particular, the monetized
climate benefits are incomplete and an underestimate as explained in Section 4.2. In addition,
important health, welfare, and water quality benefits anticipated under these final rules are not
quantified or monetized. EPA anticipates that taking non-monetized effects into account would
show the final rules to have greater benefit than the tables in this section reflect. Simultaneously,
the estimates of compliance costs used in the net benefits analysis may provide an incomplete
characterization of the social costs of the rule. See Section 5.2 for a discussion of why
compliance costs from IPM may differ from social costs estimated in the SAGE model using a
general equilibrium framework. The balance of unquantified benefits and costs is ambiguous but
is unlikely to change the result that the benefits of the final rules exceed the costs by billions of
dollars annually.

We also note that the RIA follows EPA's historical practice of using a detailed
technology-rich partial equilibrium model of the electricity and related fuel sectors to estimate
the incremental costs of producing electricity under the requirements of proposed and final major
EPA power sector rules. In Section 5.2 of this RIA, EPA has also included an economy-wide
analysis that considers additional facets of the economic response to the final rules, including the
full resource requirements of the expected compliance pathways, some of which are paid for
through subsidies. The social cost estimates in the economy-wide analysis and discussed in
Section 5.2 are still far below the projected benefits of the final rules.

ES-19


-------
ES.10 References

National Academies. (2017). Valuing Climate Damages: Updating Estimation of the Social Cost
of Carbon Dioxide. Washington DC: The National Academies Press.

U.S. EPA. (2014). Guidelines for Preparing Economic Analyses. (EPA 240-R-10-001).

Washington DC: U.S. Environmental Protection Agency, Office of Policy, National
Center for Environmental Economics, https://www.epa.gov/environmental-
economics/guidelines-preparing-economic-analvses

U.S. EPA. (2015). Guidance on Considering Environmental Justice During the Development of
Regulatory Actions, https://www.epa.gov/sites/default/files/2015-
06/documents/considering-ei-in-rulemaking-guide-final.pdf

U.S. EPA. (2023a). Estimating PM2.5- and Ozone-Attributable Health Benefits. Research
Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Health and Environmental Impact Division.
https://www.epa.gov/svstern/files/documents/2023-01/Estirnating%20PM2.5-
%20and%200zone-Attributable%20Health%20Benefits%20TSD O.pdf

U.S. EPA. (2023b). Regulatory Impact Analysis of the Standards of Performance for New,

Reconstructed, and Modified Sources and Emissions Guidelines for Existing Sources: Oil
and Natural Gas Sector Climate Review. (EPA-452/R-23-013). Research Triangle Park,
NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, Health and Environmental Impact Division.

https://www.epa.gov/svstem/files/documents/2023-12/eol2866 oil-and-gas-nsps-eg-
climate-review-2060-avl6-ria-20231130.pdf

U.S. EPA Science Advisory Board. (2017). SAB Advice on the Use of Economy-Wide Models in
Evaluating the Social Costs, Benefits, and Economic Impacts of Air Regulations. (EPA-
SAB-17-012). Washington DC

U.S. EPA Science Advisory Board. (2020). Technical Review ofEPA's Computable General
Equilibrium Model, SAGE. (EPA-SAB-20-010). Washington DC

ES-20


-------
1 INTRODUCTION AND BACKGROUND

1.1 Introduction

In 2009, the EPA concluded that GHG emissions endanger our nation's public health and
welfare.14 Since that time, the evidence of the harms posed by GHG emissions has only grown,
and Americans experience the destructive and worsening effects of climate change every day.15
Fossil fuel-fired EGUs are the nation's largest stationary source of GHG emissions, representing
25 percent of the United States' total GHG emissions in 2021.16 At the same time, a range of
cost-effective technologies and approaches to reduce GHG emissions from these sources is
available to the power sector—including carbon capture and sequestration/storage (CCS), co-
firing with less GHG-intensive fuels, and more efficient generation. Congress has also acted to
provide funding and other incentives to encourage the deployment of various technologies,
including CCS, to achieve reductions in GHG emissions from the power sector.

In this notice, the EPA is finalizing several actions under section 111 of the Clean Air
Act (CAA) to reduce the significant quantity of GHG emissions from fossil fuel-fired EGUs by
establishing emission guidelines and new source performance standards (NSPS) that are based
on available and cost-effective technologies that directly reduce GHG emissions from these
sources. Consistent with the statutory command of CAA section 111, the final NSPS and
emission guidelines reflect the application of the best system of emission reduction (BSER) that,
taking into account costs, energy requirements, and other statutory factors, is adequately
demonstrated.

Specifically, the EPA is first finalizing the repeal of the Affordable Clean Energy (ACE)
Rule. Second, the EPA is finalizing emission guidelines for GHG emissions from existing fossil
fuel-fired steam generating EGUs, which include both coal-fired and oil/gas-fired steam
generating EGUs. Third, the EPA is finalizing revisions to the NSPS for GHG emissions from
new and reconstructed fossil fuel-fired stationary combustion turbine EGUs. Fourth, the EPA is

14	74 FR 66496 (December 15, 2009).

15	The 5th National Climate Assessment (NCA5) states that the effects of human-caused climate change are already

far-reaching and worsening across every region of the United States and that climate change affects all aspects of
the energy system-supply, delivery, and demand-through the increased frequency, intensity, and duration of
extreme events and through changing climate trends.

16	https://www.epa.gov/ghgemissions/sources-greenhouse-gas-emissions.

1-1


-------
finalizing revisions to the NSPS for GHG emissions from fossil fuel-fired steam generating units
that undertake a large modification, based upon the 8-year review required by the CAA. The
EPA is not finalizing emission guidelines for GHG emissions from existing fossil fuel-fired
combustion turbines at this time and plans to expeditiously issue an additional proposal that more
comprehensively addresses GHG emissions from this portion of the fleet. The EPA
acknowledges that the share of GHG emissions from existing fossil fuel-fired combustion
turbines has been growing and is projected to continue to do so, particularly as emissions from
other portions of the fleet decline, and that it is vital to regulate the GHG emissions from these
sources consistent with CAA section 111.

These final actions ensure that the new and existing fossil fuel-fired EGUs that are
subject to these rules reduce their GHG emissions in a manner that is cost-effective and improves
the emissions performance of the sources, consistent with the applicable CAA requirements and
caselaw. These standards and emission guidelines will significantly decrease GHG emissions
from fossil fuel-fired EGUs and the associated harms to human health and welfare. Further, the
EPA has designed these standards and emission guidelines in a way that is compatible with the
nation's overall need for a reliable supply of affordable electricity.

In accordance with Executive Order (E.O). 12866 and 13563, the guidelines of OMB
Circular A-4 and EPA's Guidelines for Preparing Economic Analyses (U.S. EPA, 2014), the
RIA analyzes the benefits and costs associated with the projected emissions reductions under the
requirements of the final rules, and two alternative sets of requirements to inform EPA and the
public about these projected impacts.17

We evaluated the potential impacts of the three illustrative scenarios using the present
value (PV) of costs, benefits, and net benefits, calculated for the years 2024 to 2047, discounted
to 2024. In addition, the Agency presents the assessment of costs, benefits, and net benefits for
specific snapshot years, consistent with historic practice. These snapshot years are 2028, 2030,
2035,2040 and 2045.

17 Circular A-4 was recently revised. The effective date of the revised Circular A-4 (2023) is March 1, 2024, for
regulatory analyses received by OMB in support of proposed rules, interim final rules, and direct final rules, and
January 1, 2025, for regulatory analyses received by OMB in support of other final rules. For all other rules,
Circular A-4 (2003) is applicable until those dates.

1-2


-------
1.2 Legal and Economic Basis for Rulemaking

In this section, we summarize the statutory requirements in the CAA that serve as the
legal basis for the final rules and the economic theory that supports environmental regulation as a
mechanism to enhance social welfare. The CAA requires EPA to prescribe regulations for new
and existing sources of air pollution. In turn, those regulations attempt to address negative
externalities created when private entities fail to internalize the social costs of air pollution.

1.2.1 Statutory Requirement

EPA's authority for and obligation to issue these final rules is CAA section 111, which
establishes mechanisms for controlling emissions of air pollutants from new and existing
stationary sources. This provision requires the EPA Administrator 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."18 EPA has listed more than 60 stationary source categories under this
provision.19 EPA has the authority to define the scope of the source categories, determine the
pollutants for which standards should be developed, and distinguish among classes, types, and
sizes within categories in establishing the standards.

Once EPA lists a source category, EPA must, under CAA section 111(b)(1)(B), establish
"standards of performance" for emissions of air pollutants from new sources (including modified
and reconstructed sources) in the source categories.20 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 EPA establishes NSPS for sources in a source category under CAA section 111(b),
EPA is also required, under CAA section 111(d)(1), 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 111(d)'s

18	CAA §111(b)(1)(A).

19	See 40 CFR 60 subparts Cb - OOOO.

20	CAA §111(b)(1)(B), 111(a)(1).

1-3


-------
mechanism for regulating existing sources differs from the one that CAA section 111(b) provides
for new sources because CAA section 111(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 111(a)(1) 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 111(d)(1) grants the authority, in applying a standard of
performance, to take into account the source's remaining useful life and other factors.

Under CAA section 111(d), a state must submit its plan to EPA for approval, and EPA
must approve the state plan if it is "satisfactory."21 If a state does not submit a plan, or if EPA
does not approve a state's plan, then EPA must establish a plan for that state.22 Once a state
receives 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. See section V of the preamble to the
final rules for more detailed statutory background and regulatory history for CAA Section 111.

1.2.1.1	Regulated Pollutant

In 2009, the EPA concluded that GHG emissions endanger our nation's public health and
welfare.23 Since that time, the evidence of the harms posed by GHG emissions has only grown,
and Americans experience the destructive and worsening effects of climate change every day.24

1.2.1.2	Definition of Affected Sources

These rules establish GHG mitigation measures on certain fossil fuel-fired electric
generating units. For details on the source categories and the mitigation measures considered
please see sections VII, VIII, and IX of the preamble.

21	CAA section 111(d)(2)(A).

22	CAA section 111(d)(2)(A).

23	74 FR 66496 (December 15, 2009).

24	The 5th National Climate Assessment (NCA5) states that the effects of human-caused climate change are already

far-reaching and worsening across every region of the United States and that climate change affects all aspects of
the energy system-supply, delivery, and demand-through the increased frequency, intensity, and duration of
extreme events and through changing climate trends.

1-4


-------
1.2.2 The Need for Air Emissions Regulation

OMB Circular A-4 indicates that one of the reasons a regulation may be issued is to
address a market failure. The major types of market failure include externalities, market power,
and inadequate or asymmetric information. Correcting market failures is one reason for
regulation; it is not the only reason. Other possible justifications include improving the function
of government, correcting distributional unfairness, or securing privacy or personal freedom.

Environmental problems are classic examples of externalities - uncompensated benefits
or costs of one's action imposed on another party. For example, the smoke from a factory may
adversely affect the health of exposed individuals and soil the property in nearby neighborhoods.
For the final regulatory actions analyzed in this RIA, the good produced is electricity from fossil
fuel-fired EGUs. If these electricity producers pollute the atmosphere when generating power,
the social costs will not be borne exclusively by the polluting firm but rather by society as a
whole. Thus, the producer is imposing a negative externality, or a social cost of emissions, on
society. The equilibrium market price of electricity may fail to incorporate the full opportunity
cost to society of these products. Consequently, absent a regulation on emissions, producers may
not internalize the social cost of emissions and social costs will be higher as a result. The final
rules will work towards addressing this market failure by causing affected producers to more
fully internalize the negative externality associated with GHG emissions from electricity
generation by new fossil fuel-fired combustion turbines and existing fossil fuel-fired steam-
generating EGUs.

1.3 Overview of Regulatory Impact Analysis
1.3.1 Repeal of Affordable Clean Energy (ACE) Rule

Section VI of the preamble explains that EPA is repealing the Affordable Clean Energy
(ACE) Rule. The RIA for the ACE Rule presented the projected impacts of an illustrative policy
scenario that modeled heat rate improvements (HRI) at coal-fired EGUs (U.S. EPA, 2019). In the

1-5


-------
ACE RIA, EPA projected the ACE Rule would have compliance costs in 2030 of about $280
million and CO2 emissions reductions of about 11 million short tons in 2030.25

As explained in the preamble, EPA concludes based on new information including
experience implementing the ACE Rule that the suite of HRI set forth in the rule, at best, would
provide negligible CO2 reductions. The ACE Rule's projected benefits were premised in part on
a 2009 technical report by Sargent & Lundy that evaluated the effects of HRI technologies. In
2023, Sargent & Lundy issued an updated report which details that the HRI selected as the BSER
in the ACE Rule would bring fewer emissions reductions than estimated in 2009.26 The 2023
report concludes that, with few exceptions, HRI technologies are less effective at reducing CO2
emissions than assumed in 2009. Also, most sources had already optimized application of HRI,
and so there are fewer opportunities to reduce emissions than previously anticipated.
Additionally, for a subset of sources, HRI are likely to cause a rebound effect leading to an
increase in GHG emissions for those sources for the reasons explained in section X.D.5.a. of the
preamble. The estimate of the rebound effect was quite pronounced in the ACE Rule's own
analysis - the rule projected that it would increase CO2 emissions from power plants in 15 states.
Accordingly, EPA no longer believes that the suite of HRI the ACE Rule selected as the BSER is
an appropriate BSER for existing coal-fired EGUs.

Consequently, EPA has determined it is appropriate to repeal the ACE Rule and to
reevaluate whether other technologies constitute the BSER. EPA now concludes that different,
more effective technologies like co-firing of natural gas and CCS are now cost reasonable for
designated facilities with longer operating horizons. Since the ACE Rule was promulgated,
changes in the power industry, developments in the costs of controls, and new federal subsidies
have made these other more effective technologies more broadly available and less costly.

As noted in the ACE RIA, the ACE Rule itself required no specified degree of emission
limitation or standards of performance. States were given only general criteria to inform their
efforts to design standards for sources. After the ACE Rule was promulgated, early efforts at
implementation of the rule underscored that the rule did not include enough specificity to ensure
GHG reductions consistent with the RIA. Furthermore, even if we assumed the same degree of

25	In comparison, these final rules are projected to reduce approximately 50 million metric tons of CO2 in 2030, and

approximately 123 million metric tons of CO2 in 2035 (see Table 3-5).

26	See Heat Rate Improvement Method Costs and Limitations Memo, which is available in the docket for this action.

1-6


-------
effectiveness as was assumed in the ACE Rule RIA, the number of units that would be covered if
the ACE Rule were implemented today would be much lower because of declines in coal-fired
generation since the ACE Rule was promulgated as well as increases in the number of facilities
who have elected to commit to permanently cease operations in the coming years. Because of
these factors, the expected costs and associated emissions changes of the ACE Rule are likely
much less than projected in the 2019 ACE Rule RIA.27

Accordingly, based on reconsideration of the emissions impact of HRI and new
information gained from early implementation of the ACE Rule, among other factors, EPA
anticipates that the implementation of the ACE Rule would likely produce negligible, if any,
change in costs or emissions relative to a world without the rule. In addition, the final 111(b) and
111(d) actions only occur after the repeal of the ACE Rule. As such, it is EPA's finding and
conclusion that there is likely to be no difference in the baseline between a world where ACE is
implemented and one where it is not.

1.3.2 Baseline and Analysis Years

The impacts of final regulatory actions are evaluated relative to a modeled baseline that
represents expected behavior in the electricity sector under market and regulatory conditions in
the absence of a regulatory action. EPA used the Integrated Planning Model (IPM)28 to conduct
the electric generating units (EGU) analysis discussed in this section, relying on the EPA's
Power Sector Platform 2023 Using IPM to establish the baseline for this analysis. For a detailed
description please see Section 3 of this document. EPA frequently updates the power sector
modeling baseline to reflect the latest available electricity demand forecasts from the U.S.

Energy Information Administration (EIA) as well as expected costs and availability of new and
existing generating resources, fuels, emission control technologies, and regulatory requirements.
The baseline includes the final Good Neighbor Plan (GNP), the Revised Cross-State Air
Pollution Rule (CSAPR) Update, CSAPR Update, and CSAPR, as well as the 2012 Mercury and
Air Toxics Standards. The power sector baseline also includes the 2015 Effluent Limitation
Guidelines (ELG) and the 2015 Coal Combustion Residuals (CCR), and the recently finalized

27	For details on historical coal retirements, please see the Power Sector Trends - TSD available in the docket for

this rulemaking. For details on projected coal capacity under the baseline, please see Table 3-14.

28	Information on IPM can be found at the following link: https://www.epa.gov/airmarkets/power-sector-modeling.

1-7


-------
2020 ELG and CCR rules. This version of the model ("EPA's Power Sector Platform 2023 using
IPM 2022") also includes recent updates to state and federal legislation affecting the power
sector, including Public Law 117-169, 136 Stat. 1818 (August 16, 2022), commonly known as
the Inflation Reduction Act of 2022 (IRA). The modeling documentation, available in the docket,
includes a summary of all legislation reflected in this version of the model as well as a
description of how that legislation is implemented in the model. Also, see Section 3 for
additional detail about the power sector baseline for this RIA.

We evaluated the potential impacts of the three illustrative scenarios for the years 2024 to
2047 from the perspective of 2024, using discount rates of two percent, three percent, and seven
percent. In addition, the Agency presents the assessment of costs, benefits, and net benefits for
specific snapshot years, consistent with historic practice. These snapshot years are 2028, 2030,
2035, 2040, and 2045. The Agency believes that these specific years are each representative of
several surrounding years, which enables the analysis of costs and benefits over the timeframe of
2024 to 2047. The year 2028 is the first year of detailed power sector modeling for this RIA and
approximates when the regulatory impacts of the final 111(b) new source performance standards
on the power sector will begin. However, because the Agency estimates that some monitoring,
reporting, and recordkeeping (MR&R) costs may be incurred in 2024, we analyze compliance
costs in years before 2028. Therefore, while MR&R costs analysis is presented beginning in the
year 2024, the detailed assessment of costs, emissions impacts, and benefits begins in the year
2028. The analysis timeframe concludes in 2047, as this is the last year that may be represented
with the analysis conducted for the specific year of 2045.

1.3.3 Best System of Emission Reduction (BSER)

These final actions include the repeal of the ACE Rule, BSER determinations and
emission guidelines for existing fossil fuel-fired steam generating units, and BSER
determinations and accompanying standards of performance for GHG emissions from new and
reconstructed fossil fuel-fired stationary combustion turbines and modified fossil fuel-fired steam
generating units. See Section I.C of the final rule preamble for a summary of the major
provisions of these regulations. Related information can also be found in Technical Support
Documents (TSDs) available in the rulemaking docket.

1-8


-------
1.3.4 Illustrative Scenarios

This RIA evaluates the potential benefits, costs, and other impacts associated with
compliance actions projected under three illustrative scenarios: one scenario representing the
final rules, and two scenarios representing alternative sets of requirements. The modeling of the
illustrative final rules scenario that is discussed in Sections 3 through 7 of this RIA includes the
final 111(b) requirements for new and reconstructed stationary combustion turbines and final
111(d) requirements for existing fossil fuel-fired steam generating units.

The GHG mitigation measures in this RIA are illustrative since States are afforded
flexibility to implement the final rules, and thus the impacts could be different to the extent states
make different choices than those assumed in the illustrative analysis. Additionally, the way that
EGUs comply with the GHG mitigation measures may differ from the methods forecast in the
modeling for this RIA. See Section 3.4 for further discussion of the modeling approach used in
the analysis presented below. For details of the controls modeled for each of the existing source
categories starting in run year 2030 under the three illustrative scenarios please see Section 3.2 of
this document.

1.4 Organization of the Regulatory Impact Analysis

This RIA is organized into the following remaining sections:

•	Section 2: Industry Profile. This section describes the electric power sector in detail.

•	Section 3: Cost, Emissions, and Energy Impacts. This section summarizes the
projected compliance costs and other energy impacts associated with the regulatory
options.

•	Section 4: Benefits Analysis. This section presents the projected climate benefits of CO2
emissions reductions, and the health and environmental benefits of reductions in
emissions of nitrogen oxides (NOx), fine particulate matter (PM2.5) and sulfur dioxide
(SO2). Potential benefits to drinking water quality and quantity are also discussed.

1-9


-------
Section 5: Social Costs and Economic Impacts. This section includes a discussion of
energy market impacts, economy-wide social costs and economic impacts, potential small
entity impacts, and labor impacts.

Section 6: Environmental Justice Impacts. This section includes an assessment of
potential impacts to potential EJ populations.

Section 7: Comparison of Benefits and Costs. This section compares the total projected
benefits with total projected costs and summarizes the projected net benefits of the three
illustrative scenarios examined. The section also includes a discussion of potential
benefits that EPA is unable to quantify and monetize.

Appendix A: Climate Benefits. This section presents the climate benefits of the final
standards using the interim SC-CO2 values used in the proposals of these rulemakings.
Appendix B: Air Quality Modeling. This section describes the air quality modeling
simulations, provides details on the methodology to apply the air quality modeling to
estimate ozone and PM2.5 impacts of the illustrative final rules scenario and presents
resulting surfaces that represent air quality changes associated with the illustrative
scenarios.

Appendix C: Environmental Justice Analysis. This section presents additional figures
associated with the alternative 1 scenario and alternative 2 scenario.

Appendix D: Assessment of Potential Costs and Emissions Impacts of Final New
and Existing Source Standards Analyzed Separately. This section summarizes the
projected compliance costs and other energy impacts associated with the imposition of
new source standards independently from existing source standards.

1-10


-------
1.5 References

OMB. (2003). Circular A-4: Regulatory Analysis. Washington DC.
https://www.whitehouse.gov/wp-

content uploads legacy drupal files omb circulars A4 a-4.pdf

OMB. (2023). Circular A-4: Regulatory Analysis. Washington DC.

https://www.whitehouse.gov/wp-content/uploads/2023/ll/CircularA-4.pdf

U.S. EPA. (2014). Guidelines for Preparing Economic Analyses. (EPA 240-R-10-001).

Washington DC: U.S. Environmental Protection Agency, Office of Policy, National
Center for Environmental Economics, https://www.epa.gov/environmental-
economics/guidelines-preparing-economic-analyses

U. S. EPA. (2019). Regulatory Impact Analysis for the Repeal of the Clean Power Plan, and the
Emission Guidelines for Greenhouse Gas Emissions from Existing Electric Utility
Generating Units. (EPA-452/R-19-003). Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Health
and Environmental Impact Division, https://www.epa.gov/sites/production/files/2019-
06/documents/utilities ria final cpp repeal and ace 2019-06.pdf

1-11


-------
2 INDUSTRY PROFILE

2.1	Background

In the past decade, there have been substantial structural changes in both the mix of
generating capacity and in the share of electricity generation supplied by different types of
generation. These changes are the result of multiple factors in the power sector, including
replacements of older generating units with new units, changes in the electricity intensity of the
U.S. economy, growth and regional changes in the U.S. population, technological improvements
in electricity generation from both existing and new units, changes in the prices and availability
of different fuels, and substantial growth in electricity generation from renewable energy
sources. Many of these trends will likely continue to contribute to the evolution of the power
sector.29 The evolving economics of the power sector, specifically the increased natural gas
supply and subsequent relatively low natural gas prices, have resulted in more natural gas being
used to produce both base and peak load electricity. Additionally, rapid growth in the
deployment of wind and solar technologies has led to their now constituting a significant share of
generation. The combination of these factors has led to a decline in the share of electricity
generated from coal.30 This section presents data on the evolution of the power sector over the
past two decades from 2010 through 2022, as well as a focus on the period 2015 through 2022.
Projections of future power sector behavior and the impact of the final rules are discussed in
more detail in Section 3 of this RIA.

2.2	Power Sector Overview

The production and delivery of electricity to customers relies on of three distinct stages:
the generation, transmission, and distribution of electricity.

2.2.1 Generation

Electricity generation is the first process in the delivery of electricity to consumers. There
are two important aspects of electricity generation: capacity and net generation. Generating

29	For details on the evolution of EPA's power sector projections, please see archive of IPM outputs available at:
epa.gov/power-sector-modeling

30	For details, please see "Power Sector Trends Technical Support Document" available in the docket for this

rulemaking.

2-1


-------
Capacity refers to the maximum amount of production an EGU is capable of producing in a
typical hour, typically measured in megawatts (MW) for individual units, or gigawatts (1 GW =
1,000 MW) for multiple EGUs. Electricity Generation refers to the amount of electricity actually
produced by an EGU over some period of time, measured in kilowatt-hours (kWh) or gigawatt-
hours (1 GWh = 1 million kWh). Net Generation is the amount of electricity that is available to
the grid from the EGU (i.e., excluding the amount of electricity generated but used within the
generating station for operations). Electricity generation is most often reported as the total annual
generation (or some other period, such as seasonal). In addition to producing electricity for sale
to the grid, EGUs perform other services important to reliable electricity supply, such as
providing backup generating capacity in the event of unexpected changes in demand or
unexpected changes in the availability of other generators. Other important services provided by
generators include facilitating the regulation of the voltage of supplied generation.

Individual EGUs are not used to generate electricity 100 percent of the time. Individual
EGUs are periodically not needed to meet the regular daily and seasonal fluctuations of
electricity demand. Units are also unavailable during routine and unanticipated outages for
maintenance. 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. These factors result in the share
of potential generating capacity being substantially different from the share of actual 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 combustion 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 most common 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 2010 to 2022 and 2015 to 2022.

2-2


-------
In 2022 the power sector comprised a total capacity31 of 1,201 GW, an increase of 162
GW (or 16 percent) from the capacity in 2010 (1,039 GW). The largest change over this period
was the decline of 127 GW of coal capacity, reflecting the retirement/rerating of close to 40
percent of the coal fleet. This reduction in coal capacity was offset by increases in natural gas,
solar, and wind capacities of 95 GW, 72 GW, and 102 GW respectively. Substantial amounts of
distributed solar (40 GW) were also added.

These trends persist over the shorter 2015-21 period as well; total capacity in 2022 (1,201
GW) increased by 127 GW (or 12 percent). The largest change in capacity was driven by a
reduction of 90 GW of coal capacity. This was offset by a net increase of 63 GW of natural gas
capacity, an increase of 69 GW of wind, and an increase of 59 GW of solar. Additionally, 30
GW of distributed solar were also added over 2015-22.

31 This includes generating capacity at EGUs primarily operated to supply electricity to the grid and combined heat
and power facilities classified as Independent Power Producers (IPP) and excludes generating capacity at
commercial and industrial facilities that does not operate primarily as an EGU. Natural Gas information in this
section (unless otherwise stated) reflects data for all generating units using natural gas as the primary fossil heat
source. This includes Combined Cycle Combustion Turbine, Gas Turbine, steam, and miscellaneous (< 1
percent).

2-3


-------
Table 2-1 Total Net Summer Electricity Generating Capacity by Energy Source,

2010-22 and 2015-22



2010

2022

Change Between '10
and '22



Net



Net

Net



Net

Energy Source

Summer

% Total

Summer

Summer

% Total

Summer

Capacity
(GW)

Capacity

Capacity
(GW)

Capacity
(GW)

Capacity

Capacity
(GW)

Coal

317

30%

189

16%

-40%

-127

Natural Gas

407

39%

502

42%

23%

95

Nuclear

101

10%

95

8%

-6%

-7

Hydro

101

10%

103

9%

2%

2

Petroleum

56

5%

31

3%

-45%

-25

Wind

39

4%

141

12%

261%

102

Solar

1

0%

73

6%

8310%

72

Distributed Solar

0

0%

40

3%



40

Other Renewable

14

1%

15

1%

7%

1

Misc

4

0%

12

1%

239%

9

Total

1,039

100%

1,201

100%

16%

162





2015

2022

Change Between '15
and '22

Energy Source

Net
Summer
Capacity
(GW)

% Total
Capacity

Net
Summer
Capacity
(GW)

% Total
Capacity

%
Increase

Capacity
Change
(GW)

Coal

280

26%

189

16%

-32%

-90

Natural Gas

439

41%

502

42%

14%

63

Nuclear

99

9%

95

8%

-4%

-4

Hydro

102

10%

103

9%

1%

1

Petroleum

37

3%

31

3%

-16%

-6

Wind

73

7%

141

12%

95%

69

Solar

14

1%

73

6%

433%

59

Distributed Solar

10

1%

40

3%

307%

30

Other Renewable

17

2%

15

1%

-11%

-2

Misc

4

0%

12

1%

182%

8

Total

1,074

100%

1,201

100%

12%

127

Source: EIA. Electric Power Annual 2020 and 2022, Table 4.2. A and 4.2.B

The average age of coal-fired power plants that retired between 2015 and 2023 was over
50 years. Older power plants tend to become uneconomic over time as they become more costly
to maintain and operate, and as newer and more efficient alternative generating technologies are

2-4


-------
built. As a result, coal's share of total U.S. electricity generation has been declining for over a
decade, while generation from natural gas and renewables has increased significantly.32 As
shown in Figure 2-1 below, 70 percent of the coal fleet in 2023 had an average age of over 40
years.

0-10yrs 10- 20yrs 20-30yrs 30-40yrs 40-50yrs 50-60yrs 60 + yrs

Figure 2-1 National Coal-fired Capacity (GW) by Age of EGU, 2023

Source: NEEDS v7

In 2022, electric generating sources produced a net 4,292 TWh to meet national
electricity demand, which was around 4 percent higher than 2010. As presented in Table 2-2, 60
percent of electricity in 2022 was produced through the combustion of fossil fuels, primarily coal
and natural gas, with natural gas accounting for the largest single share. The total generation
share from fossil fuels in 2022 (60 percent) was 10 percent less than the share in 2010 (70
percent). Moreover, the share of fossil generation supplied by coal fell from 65 percent in 2010
to 33 percent by 2022, while the share of fossil generation supplied by natural gas rose from 35
percent to 67 percent over the same period. In absolute terms, coal generation declined by 55
percent, while natural gas generation increased by 71 percent. This reflects both the increase in
natural gas capacity during that period as well as an increase in the utilization of new and

32 EIA, Today in Energy (April 17, 2017) available at https://www.eia.gov/todayinenergy/detail.php?id=30812

2-5


-------
existing gas EGUs during that period. The combination of wind and solar generation also grew
from 2 percent of the mix in 2010 to 14 percent in 2022.

Table 2-2 Net Generation by Energy Source, 2010 - 22 and 2015 - 22 (Trillion kWh =

TWh)



2010

2022

Change Between '10
and '22

Energy Source

Net
Generation
(TWh)

Fuel
Source
Share

Net
Generation
(TWh)

Fuel
Source
Share

%
Increase

Generation
Change
(TWh)

Coal

1,847

45%

832

19%

-55%

-1,016

Natural Gas

988

24%

1,687

39%

71%

699

Nuclear

807

20%

772

18%

-4%

-35

Hydro

255

6%

249

6%

-2%

-6

Petroleum

37

1%

23

1%

-38%

-14

Wind

95

2%

434

10%

359%

340

Solar

1

0%

144

3%

11764%

143

Distributed Solar

0

0%

61

1%



61

Other Renewable

71

2%

68

2%

-5%

-3

Misc

24

1%

23

1%

-6%

-1

Total

4,125

100%

4,292

100%

4%

167

Table 2-3 Net Generation in 2015 and 2022 (Trillion kWh = TWh)



2015

2022

Change Between '15
and '22

Energy Source

Net
Generation
(TWh)

Fuel
Source
Share

Net
Generation
(TWh)

Fuel
Source
Share

%
Increase

Generation
Change
(TWh)

Coal

1,352

33%

832

19%

-39%

-521

Natural Gas

1,335

33%

1,687

39%

27%

354

Nuclear

797

19%

772

18%

-3%

-26

Hydro

244

6%

249

6%

2%

5

Petroleum

28

1%

23

1%

-19%

-5

Wind

191

5%

434

10%

128%

244

Solar

25

1%

144

3%

478%

119

Distributed Solar

14

0%

61

1%

333%

47

Other Renewable

80

2%

68

2%

-15%

-12

Misc

27

1%

23

1%

-16%

-4

Total

4,092

100%

4,292

100%

5%

200

Source: EIA. Electric Power Annual 2020 and 2022, Table 3. l.A and 3. l.B

2-6


-------
Coal-fired and nuclear generating units have historically supplied "base load" electricity,
meaning that these units operate through most hours of the year and serve the portion of
electricity load that is continually present. Although much of the coal fleet has historically
operated as base load, there can be notable differences in the design of various facilities (see
Table 2-3) which, along with relative fuel prices, can impact the operation of coal-fired power
plants. As one example of design variations, coal-fired units less than 100 megawatts (MW) in
size comprise 17 percent of the total number of coal-fired units, but only 2 percent of total coal-
fired capacity, and they tend to have higher heat rates. Gas-fired generation is generally better
able to vary output, is a 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. Over the last decade, however, the
generally low price of natural gas and the growing age of the coal fleet has resulted in increasing
capacity factors for many gas-fired plants and decreasing capacity factors for many coal-fired
plants. As shown in Figure 2-2, average annual coal capacity factors have declined from 67
percent to 50 percent over the 2010 to 2022 period, indicating that a larger share of units are
operating in non-baseload fashion. Over the same period, natural gas combined cycle capacity
factors have risen from an annual average of 44 percent to 57 percent.

2-7


-------


70%





v. - &



uU/O

\ 	\

	

1—

50%
40%
30%
20%

S

o
¦*->

u
TO
LL

>-

\ /"
X''

u
re

Q_

re
u

«

1
\
i
i
i
i
i
i
i
i

\

i
t
i
i
i
i
\

\
\

i
i
i
/
t
t
\

>

"re
d
c
c
<

_ _ _







10%









0%





2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 202 1 2022
— — —Coal — — —Natural Gas

Figure 2-2 Average Annual Capacity Factor by Energy Source

Source: EIA. Electric Power Annual 2020 and 2022, Table 4.8. A

Table 2-4 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 69 percent of the coal EGU fleet capacity is over 500 MW per unit, 82 percent
of the gas fleet is between 50 and 500 MW per unit.

2-8


-------
Table 2-4 Coal and Natural Gas Generating Units, by Size, Age, Capacity, and Average
Heat Rate in 2023

Unit Size
Grouping
(MW)

No.
Units

% of All
Units

Avg. Age

Avg. Net
Summer
Capacity
(MW)

Total Net
Summer
Capacity
(MW)

% Total
Capacity

Avg. Heat

Rate
(Btu/kWh)

COAL

0-24

23

5%

54

12

266

0%

11,174

25-49

32

7%

37

35

1,124

1%

11,541

50-99

23

5%

35

75

1,720

1%

11,807

100 - 149

28

6%

54

121

3,397

2%

11,198

150 - 249

50

11%

51

193

9,643

5%

10,844

250 - 499

110

25%

43

373

40,997

22%

10,674

500 - 749

122

27%

42

605

73,849

40%

10,298

750 - 999

50

11%

40

824

41,221

22%

10,158

1000 - 1500

10

2%

46

1,261

12,611

7%

9,841

Total Coal

448

100%

45

413

184,828

100%

10,693

NATURAL GAS

0-24

4,679

56%

30

4

20,963

4%

13,006

25-49

899

11%

26

41

36,619

7%

11,545

50-99

1,000

12%

29

72

71,611

14%

12,194

100 - 149

391

5%

26

125

48,863

10%

9,548

150 - 249

1,037

12%

20

180

186,503

37%

8,194

250 - 499

309

4%

21

330

101,969

20%

8,072

500 - 749

47

1%

30

585

27,495

5%

9,374

750 - 999

8

0%

47

838

6,706

1%

11,366

1000 - 1500

0

0%





0

0%



Total Gas

8,362

100%

27

60

500,730

100%

11,790

Source: National Electric Energy Data System (NEEDS) v.7

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.

In terms of the age of the generating units, almost 69 percent of the total coal generating
capacity has been in service for more than 40 years, while nearly 81 percent of the natural gas
capacity has been in service less than 40 years. Figure 2-3 presents the cumulative age
distributions of the coal and gas fleets, highlighting the pronounced differences in the ages of the
fleets of these two types of fossil-fuel generating capacity. Figure 2-3 also includes the
distribution of generation, which is similar to the distribution of capacity.

2-9


-------
0	10	20	30	40	50	60	70

Age of EGU (years)

Gas Cap — — —Gas Gen	Coal Cap — — —Coal Gen

Figure 2-3 Cumulative Distribution in 2021 of Coal and Natural Gas Electricity
Capacity and Generation, by Age

Source: eGRID 2021 (November 2023 release from EPA eGRID website). Figure presents data from generators that
came online between 1950 and 2021 (inclusive); a 71-year period. Full eGRID data includes generators that came
online as far back as 1915. Full data from 1915 onward is used in calculating cumulative distributions; figure
truncation at 70 years is merely to improve visibility of diagram.

The locations of existing fossil units in EPA's National Electric Energy Data System
(NEEDS) v.6 are shown in Figure 2-4.

2-10


-------
Figure 2-4 Fossil Fuel-Fired Electricity Generating Facilities, by Size

Source: National Electric Energy Data System (NEEDS) v.6

Note: This map displays fossil capacity at facilities in the NEEDS v.6 IPM frame. NEEDS v.6 reflects generating
capacity expected to be on-line at the end of 2023. This includes planned new builds already under construction and
planned retirements. In areas with a dense concentration of facilities, some facilities may be obscured.

The costs of renewable generation have fallen significantly due to technological
advances, improvements in performance, and local, state, and federal incentives such as the
recent extension of federal tax credits. According to Lazard, a financial advisoiy and asset
management firm, the current unsubsidized levelized cost of electricity for wind and solar energy
technologies is lower than the cost of technologies like coal, natural gas or nuclear, and in some
cases even lower than just the operating cost, which is expected to lead to ongoing and
significant deployment of renewable energy. Levelized cost of electricity is only one metric used
to compare the cost of different generating technologies. It contains a number of uncertainties
including utilization and regional factors.33 While this chart illustrates general trends, unit
specific build decisions will incorporate many other variables. These trends of declining costs
and cost projections for renewable resources are borne out by a range of other studies including
the NREL Annual Technology Baseline34, DOE's Land-Based Wind Market Report35, LBNL's

33	Lazard, Levelized Cost of Energy Analysis-Version 16.0,2023. https://www.lazard.com/media/typdgxmm/lazards-

lcoeplus-april-2023.pdf

34	Available at: https://atb.nrel.gov/

m Available at: https://www.energj .gov/eere/wind/articles/land-based-wind-market-report-2022-edition

2-11


-------
Utility Scale solar report36, EIA's Annual Energy Outlook37, and DOE's 2022 Grid Energy
Storage Technology Cost and Performance Assessment.

Selected Historical Mean Unsubsidized LCOE Values'1'

Mean LCOE
(VMWh)

$380

320

260

200

140

80

20

2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2023
LCOE Version 3.0	4.0	5.0	6.0	7.0	8.0	9.0	10.0	11.0	12.0	13.0	14.0	15.0	16.0

Figure 2-5 Selected Historical Mean LCOE Values

Source: Lazard, Levelized Cost of Energy Analysis-Version 16.0, April 2023

Notes: (1) Reflects the average of the high and low LCOE for each respective technology in each respective year.
Percentages represent the total decrease in the average LCOE since Lazard's LCOE \ 3.0. (2) The LCOE no longer
analyzes solar thermal costs; percent decrease is as of Lazard's LCOE v 13.0. (3) Prior versions of Lazard's LCOE
divided Utility-Scale Solar PV into Thin Film and Crystalline subcategories. All values before Lazard's LCOE
vl6.0 reflect those of the Solar PV—Crystalline technology.

The broad trends away from coal-fired generation and toward lower-emitting generation
are reflected in the recent actions and recently announced plans of many power plants across the
industry — spanning all types of companies in all locations. Furthermore, as detailed below,
many utilities have made commitments to move toward cleaner energy. Throughout the country,
utilities have included commitments towards cleaner energy in public releases, planning
documents, and integrated resource plans (IRPs). For strategic business reasons and driven by
the economics of different supply options, many major utilities plan to increase their renewable
energy holdings and continue reducing GFIG emissions, regardless of what federal regulatory
requirements might exist. The Edison Electric Institute (EEI) has confirmed these developments:
"While the CPP was stayed by the Supreme Court in 2016, the power sector will have complied

36	Available at: https://emp.lbl.gov/utility-scale-solar/

37	Available at: https://www.eia.gov/outlooks/aeo/pdf/electricity_generation.pdf

38	Available at: https://www.energy.gov/eere/analysis/2022-grid-energy -storage-teclinologj -cost-and-perfonnance-

assessment

$24*
$243

Nuclear
47%

Gas Peaking

(39%)

Solar Thermal
Tower*
(16%)

Gas Combined
Cycle
(15%)

Solar PV-
Utillty-Scale131
(83%)

Wind—Onshore
(63%)

2-12


-------
with the final 2030 goals of the rule—in terms of gross emissions reductions—before the 2022
start date included in that program."39 This trend is not unique to the largest owner-operators of
coal-fired generation; smaller utilities, public power, cooperatives, and municipal entities are
also contributing to these changes.

While EPA does not account for future planning statements from utility providers in the
economic modeling since they are not legally enforceable, the number and scale of these
announcements is significant on a systemic level. These statements are part of long-term
planning processes that cannot be easily revoked due to considerable stakeholder involvement in
the planning process, including the involvement of regulators. The direction to which these
utility providers have publicly stated they are moving is consistent across the sector and
undergirded by market fundamentals lending economic credibility to these commitments and
confidence that that most plans will be implemented.

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,40 each operating synchronously. Within each of these
transmission networks, there are multiple areas where the operation of power plants is monitored
and controlled by regional organizations to ensure that electricity generation and load are kept in
balance. In some areas, the operation of the transmission system is under the control of a single
regional operator;41 in others, individual utilities42 coordinate the operations of their generation,
transmission, and distribution systems to balance the system across their respective service
territories.

39	EEI Comments on ACE, at 4 (Oct. 31, 2018)

40	These three network interconnections are the Western Interconnection, comprising the western parts of the U.S.

and Canada, the Eastern Interconnection, comprising the eastern parts of the U.S. and Canada except parts of
eastern Canada in the Quebec Interconnection, and the Texas Interconnection, encompassing the portion of the
Texas electricity system commonly known as the Electric Reliability Council of Texas (ERCOT). See map of all
NERC interconnections at

https://www.nerc.com/AboutNERC/keyplayers/PublishingImages/NERC%20Interconnections.pdf.

41	For example, PJM Interconnection, LLC., New York Independent System Operator (NYISO), Midwest

Independent System Operator (MISO), California Independent System Operator (CAISO), etc.

42	For example, Los Angeles Department of Water and Power, Florida Power and Light, etc.

2-13


-------
2.2.3 Distribution

Distribution of electricity involves networks of lower voltage lines and substations that
take the higher voltage power from the transmission system and step it down to lower voltage
levels to match the needs of customers. The transmission and distribution system is the classic
example of a natural monopoly, in part because it is not practical to have more than one set of
lines running from the electricity generating sources to substations or from substations to
residences and businesses.

Over the last few decades, several jurisdictions in the United States began restructuring
the power industry to separate transmission and distribution from generation, ownership, and
operation. Historically, vertically integrated utilities established much of the existing
transmission infrastructure. However, as parts of the country have restructured the industry,
transmission infrastructure has also been developed by transmission utilities, electric
cooperatives, and merchant transmission companies, among others. Distribution, also historically
developed by vertically integrated utilities, is now often managed by a number of utilities that
purchase and sell electricity, but do not generate it. Electricity restructuring has focused
primarily on efforts to reorganize the industry to encourage competition in the generation
segment of the industry, including ensuring open access of generation to the transmission and
distribution services needed to deliver power to consumers. In many states, such efforts have also
included separating generation assets from transmission and distribution assets to form distinct
economic entities. Transmission and distribution remain price-regulated throughout the country
based on the cost of service.

2.3 Sales, Expenses, and Prices

Electric generating sources provide electricity for ultimate commercial, industrial and
residential customers. Each of the three major ultimate categories consume roughly a quarter to a
third of the total electricity produced (see Table 2-5).43 Some of these uses are highly variable,
such as heating and air conditioning in residential and commercial buildings, while others are

43 Transportation (primarily urban and regional electrical trains) is a fourth ultimate customer category which
accounts less than one percent of electricity consumption.

2-14


-------
relatively constant, such as industrial processes that operate 24 hours a day. The distribution
between the end use categories changed very little between 2010 and 2022.

Table 2-5 Total U.S. Electric Power Industry Retail Sales, 2010-22 and 2014-22 (billion
kWh)



2010

2022



Sales/Direct
Use (Billion
kWh)

Share of Total
End Use

Sales/Direct
Use (Billion
kWh)

Share of Total
End Use



Residential

1,446

37%

1,509

37%

Sales

Commercial
Industrial

1,330
971

34%
25%

1,391
1,020

34%
25%



Transportation

8

0%

7

0%

Total

3,755

97%

3,927

97%

Direct Use

132

140

Total End Use

3,887

4,067



2015

2022



Sales/Direct
Use (Billion
kWh)

Share of Total
End Use

Sales/Direct
Use (Billion
kWh)

Share of Total
End Use



Residential

1,404

36%

1,509

37%

Sales

Commercial
Industrial

1,361
987

35%
25%

1,391
1,020

34%
25%



Transportation

8

0%

7

0%

Total

3,759

96%

3,927

97%

Direct Use

141

140

Total End Use

3,900

4,067

Source: Table 2.2, EIA Electric Power Annual, 2020 and 2022 (October 19 2023 release)

Notes: Retail sales are not equal to net generation (Table 2-2) because net generation includes net imported
electricity and loss of electricity that occurs through transmission and distribution, along with data collection frame
differences and non-sampling error. Direct Use represents commercial and industrial facility use of onsite net
electricity generation; electricity sales or transfers to adjacent or co-located facilities; and barter transactions.

2.3.1 Electricity Prices

Electricity prices vary substantially across the United States, differing both between the
ultimate customer categories and by state and region of the country. Electricity prices are
typically highest for residential and commercial customers because of the relatively high costs of
distributing electricity to individual homes and commercial establishments. The higher prices for
residential and commercial customers are the result of the extensive distribution network
reaching to virtually every building in every part of the country and the fact that generating
stations are increasingly located relatively far from population centers, increasing transmission

2-15


-------
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 have historically been less variable. Overall, industrial
customer prices are usually considerably closer to the wholesale marginal cost of generating
electricity than residential and commercial prices.

On a state-by-state basis, all retail electricity prices vary considerably. In 2022, the
national average retail electricity price (all sectors) was 12.4 cents/kWh, with a range from 8.2
cents (Wyoming) to 39.72 cents (Hawaii).44

The real year prices for 2010 through 2022 are shown in Figure 2-6. Average national
retail electricity prices decreased between 2010 and 2022 by 4 percent in real terms (2022
dollars), and 2 percent between 2015-22.45 The amount of decrease differed for the three major
end use categories (residential, commercial and industrial). National average commercial prices
decreased the most (4 percent), and industrial prices decreased the least (1 percent) between
2015-21.



18.0







Ib.U

rH

o





140

XT



5

12.0

t.A*



r-

10.0

03









Kfi

Oi



o



Sw

Q»

6.0

>•



.tr

u

4,0

3«».



TS

jj

2.0



0.0

2010 2011 2012 2013 2014 2015 2016 201? 2018 2019 2020 2021
—•Residential - commercial • 'industrial «•» «¦«¦—« Total

Figure 2-6 Real National Average Electricity Prices (including taxes) for Three Major
End-Use Categories

Source: EIA. Electric Power Annual 2020 and 2022, Table 2.4.

44	EIA State Electricity Profiles with Data for 2022 (http://www.eia.gov/electricity/state/)

45	All prices in this section are estimated as real 2022 prices adjusted using the GDP implicit price deflator unless

otherwise indicated.

2-16


-------
2.3.2 Prices of Fossil Fuel Used for Generating Electricity

Another important factor in the changes in electricity prices are the changes in delivered
fuel prices46 for the three major fossil fuels used in electricity generation: coal, natural gas, and
petroleum products. Relative to real prices in 2015, the national average real price (in 2022
dollars) of coal delivered to EGUs in 2022 had decreased by 12 percent, while the real price of
natural gas increased by 84 percent. The real price of delivered petroleum products also
increased by 102 percent, and petroleum products declined as an EGU fuel (in 2022 petroleum
products generated 1 percent of electricity). The combined real delivered price of all fossil fuels
(weighted by heat input) in 2022 increased by 62 percent over 2015 prices. Figure 2-7 shows the
relative changes in real price of all 3 fossil fuels between 2010 and 2022.

-80%

Coal	Petroleum	Natural Gas

-60%

a>
HD
TO

c


-------
2.3.3 Changes in Electricity Intensity of the U.S. Economy from 2010 to 2022

An important aspect of the changes in electricity generation (i.e., electricity demand)
between 2010 and 2022 is that while total net generation increased by 4 percent over that period,
the demand growth for generation was lower than both the population growth (8 percent) and
real GDP growth (30 percent). Figure 2-8 shows the growth of electricity generation, population,
and real GDP during this period.

35%

Real GDP	Generation	Population

Figure 2-8 Relative Growth of Electricity Generation, Population and Real GDP Since
2010

Sources: Generation: U.S. EIA Electric Power Annual 2021 and 2022. Population: U.S. Census. Real GDP: U.S.
Bureau of Economic Analysis

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 2010 to 2022. On a per capita basis, real GDP per
capita grew by 20 percent between 2010 and 2022. At the same time electricity generation per
capita decreased by 3 percent. The combined effect of these two changes improved the overall
electricity generation efficiency in the U.S. market economy. Electricity generation per dollar of
real GDP decreased 20 percent. These relative changes are shown in Figure 2-9.

2-18


-------
25%

-15%

-20%	^

-25%

Real GDP / Capita	Generation / Capita	Generation / Real GDP

Figure 2-9 Relative Change of Real GDP, Population and Electricity Generation
Intensity Since 2010

Sources: Generation: U.S. EIA Electric Power Annual 2020 and 2022. Population: U.S. Census. Real GDP: U.S.
Bureau of Economic Analysis

2-19


-------
3 COMPLIANCE COSTS, EMISSIONS, AND ENERGY IMPACTS

3.1	Overview

This section reports the compliance costs, emissions, and energy analyses performed for
the final NSPS and final Emission Guidelines. EPA used the Integrated Planning Model (IPM)47
to conduct the electric generating units (EGU) analysis discussed in this section. As explained in
detail below, this section presents analysis for three illustrative scenarios that differ in the level
of EGU greenhouse gas (GHG) mitigation measures, and timing thereof in the lower 48 states
subject to this action. The analysis for EGUs in the section includes effects from certain
provisions of the Inflation Reduction Act (IRA) of 2022 in the baseline.48 The analysis presented
in this section reflects the combined effects of the final rules on new and existing sources. The
impacts of each action independently are presented in Appendix D.

The section is organized as follows: following a summary of the illustrative scenarios
analyzed and a summary of EPA's methodologies, we present estimates of compliance costs for
EGUs, as well as estimated impacts on emissions, generation, capacity, fuel use, fuel price, and
retail electricity price for select run years.49

3.2	Illustrative Scenarios

These rules establish GHG mitigation measures on certain fossil fuel-fired electric
generating units. The EGUs covered by these rules are existing fossil fuel-fired steam generating

47	Information on IPM can be found at the following link: https://www.epa.gov/airmarkets/power-sector-modeling.

48	The Inflation Reduction Act (IRA) contains tax credit provisions that affect power sector operations, details of

which are incorporated into the IPM modeling. Details are included in the IPM documentation. The Clean
Electricity Investment and Production Tax Credits (provisions 48E and 45Y of the IRA) are described in more
detail in Section 4. The credit for Carbon Capture and Sequestration (provision 45Q) is described in Section 3.
The impacts of the Zero-Emission Nuclear Power Production Credit (provision 45U) are reflected through
modifying nuclear retirement limits, as described in Section 4. The Credit for the Production of Clean Hydrogen
(provision 45 V) is reflected through the inclusion of an exogenously delivered price of hydrogen fuel, see
Section 9. The Advanced Manufacturing Production Tax Credit (45X) was reflected through adjustments to the
short-term capital cost added for renewable technologies, see Section 4. Documentation available at:
https://www.epa.gov/power-sector-modeling

49	IPM uses model years to represent the full planning horizon being modeled. By mapping multiple calendar years

to a run year, the model size is kept manageable. IPM considers the costs in all years in the planning horizon
while reporting results only for model run years. For this analysis, IPM maps the calendar year 2028 to run year
2028, calendar years 2029-31 to run year 2030, calendar years 2032-37 to run year 2035, calendar years 2038-42
to run year 2040, calendar years 2043-47 to run year 2045 and calendar years 2048-52 to run year 2050. For
model details, please see Chapter 2 of the IPM documentation, available at:
https://www.epa.gov/airmarkets/power-sector-modeling

3-1


-------
units greater than 25 MW, and new and reconstructed fossil fuel-fired combustion turbines that
commence construction or reconstruction after the publication of this final regulation. For details
on the source categories and the mitigation measures considered please see sections VII, VIII,
and IX of the preamble.

This RIA evaluates the benefits, costs, and certain impacts of compliance with three
illustrative scenarios: one scenario representing the final rules, and two scenarios representing
alternative sets of requirements. To the extent possible, EPA evaluated the 111(b) final rule for
new natural-gas fired EGUs and 111(d) final rule for existing coal fired EGUs in combination to
better analyze the interactive effects of the final rules. For details of the controls modeled for
each of the existing source categories starting in run year 2030 under the three illustrative
scenarios please see Table 3-1 and Table 3-2 below.

Table 3-1 Summary of Modeled GHG Mitigation Measures for Existing Sources by
Source Category under the Illustrative Final Rules and Alternative 1 Scenarioa,b,c	

Affected EGUs

Subcategory Definition

GHG Mitigation Measure

Long-term existing coal-
fired steam generating units

Coal-fired steam generating units that have
not elected to commit to permanently cease
operations by 2040

CCS with 90% capture of CO2,
starting in 2035

Medium-term existing coal-
fired steam generating units

Coal-fired steam generating units that have
not elected to commit to permanently cease
operations prior to 2035 but have committed
to permanently ceasing operations by 2040

Natural gas co-firing at 40 percent
of the heat input to the unit,
starting in 2030

3 All years shown in this table reflect IPM run years. Note that IPM run years encompass the specific calendar year
requirements of BSER, details of which are available in Section VII of the preamble.
b Coal units that lack existing SCR controls must install these controls in addition to CCS to comply.
c Coal-fired EGUs that convert entirely to burn natural gas by 2030 are no longer subject to coal-fired EGU
mitigation measures outlined above.

3-2


-------
Table 3-2 Summary of Modeled GHG Mitigation Measures for Existing Sources by
Source Category under the Illustrative Alternative 2 Scenarioa,b,c	

Affected EGUs

Subcategory Definition

GHG Mitigation
Measure

Long-term existing coal-fired
steam generating units

Coal-fired steam generating units that have not
elected to commit to permanently cease
operations by 2040

CCS with 90% capture of
CO2, starting in 2035

Medium-term existing coal-
fired steam generating units

Coal-fired steam generating units that have not

elected to commit to permanently cease
operations prior to 2035 but have committed to
permanently ceasing operations by 2040

Natural gas co-firing at 40
percent of the heat input
to the unit, starting in
2035

3 All years shown in this table reflect IPM run years. Note that IPM run years encompass the specific calendar year
requirements of BSER, details of which are available in Section VII of the preamble.
b Coal units that lack existing SCR controls must install these controls in addition to CCS to comply.
c Coal-fired EGUs that convert entirely to burn natural gas by 2035 are no longer subject to coal-fired EGU
mitigation measures outlined above.

3-3


-------
Table 3-3 Summary of GHG Mitigation Measures for New Sources by Source Category
under the II ustrative Final Rules, Alternative 1 and Alternative 2 Scenariosa,b	

Affected
EGUs

Subcategory
Definition

Modeled
Requirements
During 1st
Phase

Modeled
Requirements
During 2nd Phase
(2035)

Baseload
Definition:
Alternative 1
and Alternative
2 Scenarios

Baseload
Definition: Final
Rules Scenario

Baseload
Economic
NGCC
Additions

NGCC units

that
commence
construction
after 2023 and
operate at
greater than
baseload
annual
capacity
factor

Efficient
generation

CCS or co-fire
hydrogen at
sufficient level to
meet CCS emission
rate

Intermediate

Load
Economic
NGCC
Additions

NGCC units

that
commence
construction
after 2023 and
operate at an
annual
capacity
factor of less
than baseload

Efficient generation

50%

40%

Intermediate

load
Economic

NGCT
Additions

NGCT units

that
commence
construction
after 2023 and
operate at an
annual
capacity
factor of more
than 20%

Emission rate consistent with NGCC
operation

Peaking
Economic

NGCT
Additions

NGCT units

that
commence
construction
after 2023 and
operate at an
annual
capacity
factor of less
than 20%

Efficient generation

3 All years shown in this table reflect IPM run years. Note that IPM run years encompass the specific calendar year
requirements of BSER, details of which are available in Section VII of the preamble.
b Delivered hydrogen price is assumed to be $1.15/kg in all years.

c The modeling does not reflect the requirements of the variable subcategory. We estimate this would have a limited
impact on the results.

3-4


-------
The illustrative compliance outcomes in this RIA represent EGU behavior in response to
GHG mitigation measures applied to affected source categories in given IPM run years.50 This
RIA analyzes the final rules, as well as two alternative scenarios. The alternative 1 and
alternative 2 scenarios assume the definition of annual capacity factor for baseload operation for
new turbines is 50 percent, whereas under the final rules scenario baseload is defined as 40
percent annual capacity factor. The final rules and alternative 1 scenarios assume all medium-
term existing coal-fired steam generating units must co-fire at least 40 percent natural gas by
203051, while the alternative 2 scenario assumes that all medium-term existing coal fired steam
generating units must co-fire at least 40 percent natural gas by 2035.

The GHG mitigation measures in this RIA are illustrative since States are afforded
flexibility to implement the final rules, and thus the impacts could be different to the extent states
make different choices than those assumed in the illustrative analysis. Additionally, the way that
EGUs comply with the GHG mitigation measures may differ from the methods forecast in the
modeling for this RIA. See Section 3.4 for further discussion of the modeling approach used in
the analysis presented below.

3.3 Monitoring, Reporting, and Recordkeeping Costs

EPA projected monitoring, reporting and recordkeeping (MR&R) costs for both state
entities and affected EGUs for the years 2024 onwards. The MR&R cost estimates presented
below apply to the three illustrative scenarios.

EPA estimates that industry will incur MR&R costs due to the New Source Performance
Standards for Greenhouse Gas Emissions from New, Modified, and Reconstructed Fossil Fuel-
Fired Electric Generating Units. More specifically, we estimate costs associated with 40 CFR

50	IPM uses model years to represent the full planning horizon being modeled. By mapping multiple calendar years

to a run year, the model size is kept manageable. IPM considers the costs in all years in the planning horizon
while reporting results only for model run years. For this analysis, IPM maps the calendar year 2028 to run year
2028, calendar years 2029-31 to run year 2030, calendar years 2032-37 to run year 2035, calendar years 2038-42
to run year 2040, calendar years 2043-47 to run year 2045 and calendar years 2048-52 to run year 2050. For
model details, please see Chapter 2 of the IPM documentation, available at:
https://www.epa.gov/airmarkets/power-sector-modeling

51	CCS costs used in this analysis are developed by Sargent & Lundy and are outlined in Chapter 6 of the IPM

documentation. These costs do not include the solvent acid or water washing costs. For details, please see:
https://www.epa.gov/power-sector-modeling.

3-5


-------
Part 60, Subpart TTTTa, as described in the supporting statement found in the docket. For
purposes of RIA analysis, we assume that national costs in 2026 are approximately $35,000 in
2019 dollars, and then increase by approximately $35,000 in 2019 dollars each year thereafter to
reflect costs associated with additional respondents.52 We estimate that states will not incur
MR&R costs associated with the Final New Source Performance Standards.

EPA estimates that industry will not incur incremental MR&R costs due to the Emission
Guidelines for Greenhouse Gas Emissions from Existing Fossil Fuel-Fired Electric Generating
Units. We estimate that states will incur MR&R costs associated with this final rule. We estimate
that this may affect 43 states, resulting in a total national annual burden of approximately 89,400
hours of labor, or approximately $11 million in 2019 dollars. For detailed information, see the
Information Collection Request Support Statement for the Emission Guidelines for Greenhouse
Gas Emissions from Existing Fossil Fuel-Fired Electric Generating Units available in the docket
for these actions. For purposes of this analysis, we estimate that these MR&R costs will be
incurred over the three-year period of 2024 through 2026.

52 For purposes of this regulatory impact analysis: (1) As described in the TTTTa supporting statement in the docket,
we estimate there to be six new respondents in 2026; (2) We assume that these six respondents are simple cycle
units, and that NGCC units would not incur MR&R costs incremental to existing TTTT requirements; (3) We
assume that the number of new respondents would increase by six new respondents per year for each year over
this analysis timeframe through 2047.

3-6


-------
Table 3-4 Summary of State and Industry Annual Respondent Cost of Reporting and

Recordkeeping Requirements (million 2019 dollars)



Final NSPS for New, Modified, and
Reconstructed Fossil Fuel-Fired
Electric Generating Units

Final EGs for Existing Fossil Fuel-
Fired Electric Generating Units

Total



Industry

State3

Industryb

State

Total

2024

-

-

-

11

11

2025

-

-

-

11

11

2026

0.035

-

-

11

11

2027

0.07

-

-

-

0.070

2028

0.11

-

-

-

0.11

2029

0.14

-

-

-

0.14

2030

0.18

-

-

-

0.18

2031

0.21

-

-

-

0.21

2032

0.25

-

-

-

0.25

2033

0.28

-

-

-

0.28

2034

0.32

-

-

-

0.32

2035

0.35

-

-

-

0.35

2036

0.39

-

-

-

0.39

2037

0.42

-

-

-

0.42

2038

0.46

-

-

-

0.46

2039

0.49

-

-

-

0.49

2040

0.53

-

-

-

0.53

2041

0.56

-

-

-

0.56

2042

0.60

-

-

-

0.60

2043

0.63

-

-

-

0.63

2044

0.67

-

-

-

0.67

2045

0.70

-

-

-

0.70

2046

0.74

-

-

-

0.74

2047

0.77

-

-

-

0.77

3 EPA estimates that states will not incur MR&R costs for the Final NSPS for New, Modified, and Reconstructed
Fossil Fuel-Fired Electric Generating Units.

b EPA estimates that industry will not incur MR&R costs for the Final EGs for Existing Fossil Fuel-Fired Electric
Generating Units.

3.4 Power Sector Modeling Framework

IPM is a state-of-the-art, peer-reviewed, dynamic linear programming model that can be
used to project power sector behavior under future business-as-usual conditions and to examine
prospective air pollution control policies throughout the contiguous United States for the entire
electric power system. EPA used IPM to project likely future electricity market conditions with
and without the final NSPS and Emission Guidelines.

IPM, developed by the consultancy ICF, is a multi-regional, dynamic, deterministic linear
programming model of the contiguous U.S. electric power sector. It provides estimates of least
cost capacity expansion, electricity dispatch, and emissions control strategies while meeting

3-7


-------
energy demand and environmental, transmission, dispatch, and reliability constraints. The model
accounts for all major electric regions throughout the country, including transmission capabilities
and constraints between them. This ensures that key transmission constraints are represented in
IPM and that each individual IPM region has less internal transmission congestion based on
today's loads and resource mix.

EPA has used IPM for almost three decades to better understand power sector behavior
under future business-as-usual conditions and to evaluate the economic and emissions impacts of
prospective environmental policies. The model is designed to reflect electricity markets as
accurately as possible. EPA uses the best available information from utilities, industry experts,
gas and coal market experts, financial institutions, and government statistics as the basis for the
detailed power sector modeling in IPM. The model documentation provides additional
information on the assumptions discussed here as well as all other model assumptions and
inputs.53

The model incorporates a detailed representation of the fossil-fuel supply system that is
used to estimate equilibrium fuel prices. The model uses natural gas fuel supply curves and
regional gas delivery costs (basis differentials) to simulate the fuel price associated with a given
level of gas consumption within the system. These inputs are derived using ICF's Gas Market
Model (GMM), a supply/demand equilibrium model of the North American gas market.54

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

53	Detailed information and documentation of EPA's Baseline run using IPM (v6), including all the underlying

assumptions, data sources, and architecture parameters can be found on EPA's website at:
https://www.epa. gov/power-sector-modeling.

54	See Chapter 8 of EPA's Baseline run using IPM v6 documentation, available at:
https://www.epa. gov/power-sector-modeling

3-8


-------
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.55

To estimate the annualized costs of additional capital investments in the power sector,
EPA uses a conventional and widely accepted approach that applies a capital recovery factor
(CRF) multiplier to capital investments and adds that to the annual incremental operating
expenses. The CRF is derived from estimates of the power sector's cost of capital (i.e., private
discount rate), the amount of insurance coverage required, local property taxes, and the life of
capital.56 It is important to note that there is no single CRF factor applied in the model; rather,
the CRF varies across technologies, book life of the capital investments, and regions in the
model in order to better simulate power sector decision-making.57

EPA has used IPM extensively over the past three decades to analyze options for reducing
power sector emissions. Previously, the model has been used to estimate the costs, emission
changes, and power sector impacts for the Clean Air Interstate Rule (U.S. EPA, 2005), the Cross-
State Air Pollution Rule (U.S. EPA, 201 la), the Mercury and Air Toxics Standards (U.S. EPA,
201 lb), the Clean Power Plan for Existing Power Plants (U.S. EPA, 2015b), the Cross-State Air
Pollution Update Rule (U.S. EPA, 2016), the Repeal of the Clean Power Plan, and the Emission
Guidelines for Greenhouse Gas Emissions from Existing Electric Utility Generating Units (U.S.
EPA, 2019), and the Revised Cross-State Air Pollution Update Rule (U.S. EPA, 2021), and the
Federal Good Neighbor Plan Addressing Regional Ozone Transport for the 2015 Ozone National
Ambient Air Quality Standards (U.S. EPA, 2023). EPA has also used IPM to estimate the air
pollution reductions and power sector impacts of water and waste regulations affecting EGUs,
including contributing to RIAs for the Cooling Water Intakes (316(b)) Rule (U.S. EPA, 2014a),
the Disposal of Coal Combustion Residuals from Electric Utilities rule (U.S. EPA, 2015c), the
Steam Electric Effluent Limitation Guidelines (U.S. EPA, 2015a), and the Steam Electric
Reconsideration Rule (U.S. EPA, 2020)

The model and EPA's input assumptions undergo periodic formal peer review. The
rulemaking process also provides opportunity for expert review and comment by a variety of

55	See Chapter 7 of the IPM documentation, available at: https://www.epa.gov/power-sector-modeling

56	See Chapter 10 of the IPM documentation, available at: https://www.epa.gov/airmarkets/power-sector-modeling

57	Costs modeled in IPM reflect the costs faced by industry, and therefore are net of subsidies included in the IRA

3-9


-------
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 September
2019 U.S. EPA commissioned a peer review of EPA Baseline version 6, and in October 2014
U.S. EPA commissioned a peer review of EPA Baseline version 5.13 using the Integrated
Planning Model.58 Additionally, and in the late 1990s, the Science Advisory Board reviewed
IPM as part of the CAA Amendments Section 812 prospective studies.59 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 20 years. IPM has also been employed by
states (e.g., for the Regional Greenhouse Gas Initiative, the Western Regional Air Partnership,
Ozone Transport Assessment Group), other Federal and state agencies, environmental groups,
and industry.

3.5 EPA's Power Sector Modeling of the Baseline Run and Three Illustrative Scenarios

The IPM "baseline" for any regulatory impact analysis is a business-as-usual scenario
that represents expected behavior in the electricity sector under market and regulatory conditions
in the absence of a regulatory action. As such, an IPM baseline represents an element of the
baseline for this RIA.60 EPA frequently updates the IPM baseline to reflect the latest available
electricity demand forecasts from the U.S. Energy Information Administration (EIA) as well as
expected costs and availability of new and existing generating resources, fuels, emission control
technologies, and regulatory requirements. The IPM baseline also includes power-sector related
provisions from the IRA.61

58	See Response and Peer Review Reports, available at:
https://www.epa.gov/power-sector-modeling/ipm-peer-reviews.

59	http ://www2. epa. gov/clean-air-act-overview/benefits-and-costs-clean-air-act

60	As described in Chapter 5 of EPA's Guidelines for Preparing Economic Analyses, the baseline "should

incorporate assumptions about exogenous changes in the economy that may affect relevant benefits and costs
(e.g., changes in demographics, economic activity, consumer preferences, and technology), industry compliance
rates, other regulations promulgated by EPA or other government entities, and behavioral responses to the
proposed rule by firms and the public" (U.S. EPA, 2014b).

61	A wide variety of modeling teams have assessed baselines with IRA. The baseline estimated here is generally in

line with these other estimates. See Bistline, et al. (2023). "Power Sector Impacts of the Inflation Reduction Act
of 2022," In Preparation.

3-10


-------
3.5.1 EPA's IPMBaseline Run v7.23

For our analysis of the final NSPS, and the final Emissions Guidelines, EPA used EPA's
Power Sector Platform 2023 using IPM, as well as a companion updated database of EGU units
(the National Electricity Energy Data System or NEEDS 12-04-23) that is used in EPA's
modeling applications of IPM.62 The IPM Baseline includes the CSAPR (201 la), CSAPR Update
(2016), the Revised CSAPR Update (2021), and the proposed Good Neighbor Plan for 2015
Ozone NAAQS (2023), as well as the Mercury and Air Toxics Standards (2020). The baseline
also includes the 2015 Effluent Limitation Guidelines (ELG) and the 2015 Coal Combustion
Residuals (CCR), and the finalized 2020 ELG and CCR rules.63 Finalized in December 2021, the
impacts of the 2023 and Later Model Year Light-Duty Vehicle GHG Emissions Standards are
also captured in the baseline; the rule includes requirements for model years 2023 through 2026.
The impacts of the Standards of Performance for New, Reconstructed, and Modified Sources and
Emissions Guidelines for Existing Sources: Oil and Natural Gas Sector Climate Review are not
captured in the baseline.64 The proposed GNP Supplemental Rule (2023), the proposed Multi-
Pollutant Emissions Standards for Model Years 2027 and Later Light-Duty and Medium-Duty
Vehicles (2023), the proposed Heavy-duty Greenhouse Gas "Phase 3" for Model Years 2027 and
Later (2023), the proposed National Emission Standards for Hazardous Air Pollutants: Coal- and
Oil-Fired Electric Utility Steam Generating Units Review of the Residual Risk and Technology
Review (2023), and the proposed Steam Electric Power Generating Effluent Guidelines (2023)
were not included. Additionally, the model was also updated to account for recent updates to
state and federal legislation affecting the power sector, including Public Law 117-169, 136 Stat.
1818 (August 16, 2022), commonly known as the Inflation Reduction Act of 2022 (IRA). The
Integrated Planning Model (IPM) Documentation includes a summary of all legislation reflected
in this version of the model as well as a description of how that legislation is implemented in the
model. The IPM documentation provides details on the provisions of the IRA that were
incorporated into this analysis, including provisions relating to tax subsidies for non-emitting

62	https://www.epa.gov/power-sector-modeling

63	For a full list of modeled policy parameters, please see:
https://www.epa.gov/airmarkets/power-sector-modeling

64	Available at: https://www.federalregister.gOv/documents/2021/l 1/15/2021-24202/standards-of-performance-for-

new-reconstructed-and-modified-sources-and-emissions-guidelines-for

3-11


-------
generation, energy storage, and CCS.65 The model runs for the main RIA analysis examine the
combined effects of the final NSPS, and the final Emissions Guidelines. Appendix C examines
the impact of the two rules independently. The analysis of power sector cost and impacts
presented in this section is based on a single IPM Baseline run, and represents incremental
impacts projected solely as a result of compliance with the GHG mitigation measures presented
in Table 3-1, Table 3-2, and Table 3-3.

3.5.2 Methodology for Evaluating the Illustrative Scenarios

To estimate the costs, benefits, and economic and energy market impacts of the final
NSPS, and the final Emissions Guidelines, EPA conducted quantitative analysis of the three
illustrative scenarios: one scenario representing the final rules, and two scenarios representing
alternative sets of requirements. Details about these illustrative scenarios as analyzed in this RIA,
are provided above in Section 3.2.

Before undertaking power sector analysis to evaluate compliance with the illustrative
scenarios, EPA first considered available GHG mitigation strategies that could be implemented
by the 2035 run year. EPA considered the following GHG control strategies: Carbon Capture and
Storage (CCS), efficient generation practices, natural gas co-firing at existing coal-fired EGUs
and hydrogen co-firing at new combined cycle and combustion turbine EGUs. EPA then
developed subcategory definitions that assigned GHG mitigation measures to the appropriate
affected sources.66 This RIA projects the system-wide least-cost strategies for complying with the
assigned GHG mitigation measures. Least-cost compliance may lead to the application of
different control strategies at a given source, which is in keeping with the cost-saving
compliance flexibility afforded by this rulemaking.

65	The Inflation Reduction Act (IRA) contains a number of tax credit provisions that affect power sector operations.

The Clean Electricity Investment and Production Tax Credits (provisions 48E and 45Y of the IRA) are described
in more detail in Section 4. The credit for Carbon Capture and Sequestration (provision 45Q) is described in
Section 3. The impacts of the Zero-Emission Nuclear Power Production Credit (provision 45U) are reflected
through modifying nuclear retirement limits, as described in Section 4. The Credit for the Production of Clean
Hydrogen (provision 45 V) is reflected through the inclusion of an exogenously delivered price of hydrogen fuel,
see Section 9. The Advanced Manufacturing Production Tax Credit (45X) was reflected through adjustments to
the short-term capital cost added for renewable technologies, see Section 4. For a discussion of the uncertainties
around the modeling of the impacts of the IRA including CCS and market conditions, please see the Limitations
Discussion in Section 3.7. Documentation is available at: https://www.epa.gov/power-sector-modeling

66	For details, please see sections VII, VIII and X of the preamble.

3-12


-------
While CCS at new and existing sources and co-firing natural gas at existing coal
facilities67 are captured endogenously within IPM v6.21, hydrogen co-firing at new gas EGUs is
at present represented exogenously, but alternative representations are likely to be considered in
future modeling.

Hydrogen is an exogenous input to the model, represented as a fuel that is available at
affected sources at a delivered cost of $1.15/kg, inclusive of $3/kg subsidies under the IRA.

These costs are consistent with DOE projections of 2030 for delivered costs of electrolytic low-
GHGhydrogen in the range of $0.70/kg to $1.15/kg for power sector applications, given R&D
advancements and economies of scale.68 A growing number of studies are demonstrating more
efficient and less expensive techniques to produce low-GHG electrolytic hydrogen; and, tax
credits and market forces are expected to accelerate innovation and drive down costs even further
over the next decade.69 70 71

We also note the model does not track upstream emissions associated with the production
of the hydrogen (or any other modeled fuels such as coal and natural gas), nor any incremental
electricity demand associated with its production. Under the illustrative Final Rules scenario,
incremental electricity demand from hydrogen production in 2035 is estimated at about 0.1
GWh, or less than 0.001 percent of the total projected nationwide generation.

As noted in Section 5.2, IPM estimates compliance costs incurred by regulated firms, but
because of the availability of subsidy payments, there are also real resource costs to the economy
outside of the regulated sector. IPM provides EPA's best estimate of the costs of the final rules to
the electricity sector and related energy sectors (i.e., natural gas, coal mining). To estimate the
social costs for the economy as a whole, EPA has used information from IPM as an input into the

67	For details on CCS modeling in IPM, please see Chapter 6 of the documentation, available at:

https://www.epa.gov/power-sector-modeling. Additionally, EPA has summarized the CCS costs for affected
existing coal-fired steam generating units in the "GHG Mitigation Measures for Steam EGUs" Technical Support
Document. For the universe of coal-fired steam generating units that have not committed to retirement or convert
to gas by 2039, assuming a 12 year amortization period and an 80% capacity factor, EPA estimates the average
abatement cost to be -$5/ton, inclusive of 45Q tax subsidies.

68	DOE Pathways to Commercial Liftoff: Clean Hydrogen, March 2023 See: https://liftoff.energy.gov/wp-

content/uploads/2023/03/20230320-Liftoff-Clean-H2-vPUB-0329-update.pdf

69	"Sound waves boost green hydrogen production," Power Engineering, January 4, 2023.

70	"Direct seawater electrolysis by adjusting the local reaction environment of a catalyst," Nature Energy, January

30, 2023.

71	Hydrogen from Next-generation Electrolyzers of Water (H2NEW) | H2NEW (energy.gov)

3-13


-------
Agency's computable general equilibrium model, SAGE. The economy-wide analysis is
considered a complement to the more detailed evaluation of sector costs produced by IPM.

The annualized social cost estimated in SAGE for the finalized rules is approximately
$1.32 billion (2019 dollars) between 2024 and 2047 using a 4.5 percent discount rate that is
consistent with the internal discount rate in the model. Under the assumption that compliance
costs from IPM in 2056 continue until 2081, the equivalent annualized value for social costs in
the SAGE model is $1.51 billion (2019 dollars) over the period from 2024 to 2081, again using a

4.5	percent discount rate that is consistent with the internal discount rate of the model. The social
cost estimate reflects the combined effect of the finalized rules' requirements and interactions
with IRA subsidies for specific technologies that are expected to see increased use in response to
the finalized rules. We are not able to identify their relative roles at this time. Note that SAGE
does not currently estimate changes in emissions nor account for environmental benefits. See
Section 5.2 for more discussion on the economy-wide analysis with SAGE and estimates of
private and social costs.

3.5.3 Methodology for Estimating Compliance Costs

This section describes EPA's approach to quantify estimated compliance costs in the
power sector associated with the three illustrative scenarios, which include estimates projected
directly by the model, and costs estimated outside the model framework. The model projections
capture the costs associated with installation of GHG mitigation measures at affected sources as
well as the resulting effects on dispatch as the relative operating costs for units are affected.
Additionally, EPA estimates monitoring, reporting and recordkeeping (MR&R) costs for affected
EGUs for the timeframe of 2024 to 2047, and these costs are added to the estimated change in
the total system production cost projected by IPM.

3.6	Estimated Impacts of the Illustrative Scenarios

3.6.1 Emissions Reduction Assessment

As indicated in Section 3.2, the EGU CO2 emissions reductions are presented in this RIA
from 2028 through 2045 and are based on IPM projections. Table 3-5 presents the estimated
reduction in power sector CO2 emissions resulting from compliance with the evaluated

3-14


-------
illustrative scenarios. The alternative scenarios produce smaller emissions reductions than the
final rules.

Table 3-5 EGU Annual CO2 Emissions and Emissions Changes (million metric tons)

for the Baseline and the Illustrative Scenarios from 2028 t

irough 2045 72



Annual CO2



Total Emissions

Change from Baseline

(million
metric
tons)

Baseline

Final

Alternative

Alternative

Final Alternative

Alternative

Rules

1

2

Rules 1

2

2028

1,159

1,121

1,123

1,127

-38 -36

-32

2030

1,098

1,048

1,050

1,071

-50 -48

-27

2035

724

601

601

603

-123 -124

-122

2040

459

406

406

406

-54 -53

-53

2045

307

265

267

267

-42 -40

-40

Within the compliance modeling, sources within each subcategory are subject to GHG
mitigation measures beginning in 2030. Since IPM is forward looking, investment decisions
prior to the start of the program are influenced by how those assets would fare under the policy
assumed. Hence, we see small reductions in 2028, prior to the imposition of the policy in 2030.
Emission reductions peak in 2035 across all scenarios, reflective of the start of the requirements
on existing coal-fired EGUs. Under the alternative 1 and alternative 2 scenarios, the baseload
definition is assumed to be 50 percent under the NSPS, while the final rules scenario assumes a
40 percent baseload definition. The final rules and alternative 1 scenarios assume all medium-
term existing coal-fired steam generating units must co-fire at least 40 percent natural gas by
2030, while the alternative 2 scenario assumes that all medium-term existing coal fired steam
generating units must co-fire at least 40 percent natural gas by 2035.

The impact of the IRA is to increase the cost-competitiveness of low-emitting
technology, with the result that emissions are projected to fall significantly over the forecast
period under the baseline. Hence reductions from the rules are highest in 2035 relative to the
baseline and also decline overtime. For details on the EGU emissions controls assumed in each
of the illustrative scenarios, please see Table 3-1, Table 3-2, and Table 3-3.

72 This analysis is limited to the geographically contiguous lower 48 states.

3-15


-------
In addition to the annual CO2 reductions, there will also be reductions of other air
emissions associated with EGUs burning fossil fuels that result from compliance strategies to
reduce annual CO2 emissions. These other emissions include the annual total changes in
emissions ofNOx, SO2, direct PM2.5, and ozone season NOx emissions changes. The emissions
reductions are presented in Table 3-6.

3-16


-------
Table 3-6 EGU Annual Emissions and Emissions Changes for NOx, SO2, PM2.5, Hg and
Ozone NOx for the Illustrative Scenarios for 2028 to 2045

Annual NOx



Total Emissions



Change from Baseline

(Thousand Tons)

Baseline

Final Rule

Alt 1

Alt 2

Final Rule

Alt 1

Alt 2

2028

461

441

442

444

-20

-19

-17

2030

393

374

374

382

-20

-20

-11

2035

259

210

207

211

-49

-51

-48

2040

173

166

166

167

-6

-7

-5

2045

107

83

83

83

-24

-24

-24

Ozone Season NOxa



Total Emissions



Change from Baseline

(Thousand Tons)

Baseline

Final Rule

Alt 1

Alt 2

Final Rule

Alt 1

Alt 2

2028

189

183

183

184

-6

-6

-5

2030

175

168

168

171

-7

-7

-4

2035

119

100

99

101

-19

-20

-18

2040

88

82

82

82

-6

-6

-6

2045

59

45

45

45

-14

-14

-14

Annual SO2



Total Emissions



Change from Baseline

(Thousand Tons)

Baseline

Final Rule

Alt 1

Alt 2

Final Rule

Alt 1

Alt 2

2028

454

420

424

426

-34

-30

-28

2030

334

313

317

319

-20

-16

-15

2035

240

150

150

146

-90

-90

-94

2040

143

139

139

135

-4

-4

-8

2045

55

13

13

14

-41

-41

-41

Annual
Mercury



Total Emissions



Change from Baseline

(Tons)

Baseline

Final Rule

Alt 1

Alt 2

Final Rule

Alt 1

Alt 2

2028

3.1

3.0

3.0

3.0

-0.1

-0.1

-0.1

2030

2.9

2.8

2.8

2.9

-0.1

-0.1

0.0

2035

2.5

2.4

2.4

2.4

-0.1

-0.1

-0.1

2040

2.0

2.3

2.3

2.3

0.2

0.2

0.3

2045

1.4

1.2

1.2

1.2

-0.2

-0.2

-0.1

Direct PM2.5



Total Emissions



Change from Baseline

(Thousand Tons)

Baseline

Final Rule

Alt 1

Alt 2

Final Rule

Alt 1

Alt 2

2028

71

69

69

69

-2

-2

-1

2030

66

65

65

65

-2

-2

-1

2035

51

49

49

49

-1

-2

-1

2040

37

39

39

39

2

1

2

2045

24

22

22

22

-2

-2

-2

3 Ozone season is the May through September period in this analysis.

3-17


-------
3.6.2 Compliance Cost Assessment

The estimates of the changes in the cost of supplying electricity for the illustrative
scenarios presented in Table 3-7.73 Since the rules are estimated to result in additional
recordkeeping, monitoring or reporting requirements, the costs associated with compliance,
monitoring, recordkeeping, and reporting requirements are included within the estimates in this
table.

Table 3-7 National Power Sector Compliance Cost Estimates (billions of 2019 dollars)
for the Illustrative Scenarios



Final Rules

Alternative 1

Alternative 2

2024 to 2042 (Annualized)

0.43

0.46

0.38

2024 to 2047 (Annualized)

0.86

0.88

0.85

2028 (Annual)

-1.30

-1.08

-1.06

2030 (Annual)

-0.22

-0.05

-0.72

2035 (Annual)

1.28

1.21

1.16

2040 (Annual)

0.59

0.64

0.60

2045 (Annual)

3.34

3.26

3.59

"2024 to 2042 (Annualized)" reflects total estimated annual compliance costs levelized over the period 2024 through
2042 and discounted using a 3.76 real discount rate.74 This does not include compliance costs beyond 2042. "2024
to 2047 (Annualized)" reflects total estimated annual compliance costs levelized over the period 2024 through 2047
and discounted using a 3.76 real discount rate. This does not include compliance costs beyond 2047. "2028
(Annual)" through "2045 (Annual)" costs reflect annual estimates in each of those run years.75

There are several notable aspects of the results presented in Table 3-7. One notable result
in Table 3-7 is that the estimated annual compliance costs for the three scenarios are negative
(i.e., a cost reduction) in 2028 and 2030, although these illustrative scenarios reduce C02
emissions as shown in Table 3-5. While seemingly counterintuitive, estimating negative
compliance costs in a single year is possible given the assumption of perfect foresight. IPM's

73	Reported yearly costs reflect costs incurred in IPM run year mapped to respective calendar year. For details,

please see Chapter 2 of the IPM documentation.

74	This table reports compliance costs consistent with expected electricity sector economic conditions. The PV of

costs was calculated using a 3.76 percent real discount rate consistent with the rate used in IPM's objective
function for cost-minimization. This discount rate is meant to capture the observed equilibrium market rate at
which investors are willing to sacrifice present consumption for future consumption and is based on a Weighted
Average Cost of Capital (WACC). The PV of costs was then used to calculate the levelized annual value over a
19-year period (2024 to 2042) and a 24-year period (2024 to 2047) using the 3.76 percent rate as well. Table 3-7
reports the PV of the annual stream of costs from 2024 to 2047 using 3 percent and 7 percent consistent with
OMB guidance.

75	Cost estimates include financing charges on capital expenditures that would reflect a transfer and would not

typically be considered part of total social costs.

3-18


-------
objective function is to minimize the discounted present value (PV) of a stream of annual total
cost of generation over a multi-decadal time period.76 Under the baseline, the proposed GNP rule
results in installation of SCR controls in the 2030 run year on some coal-fired EGUs that
currently lack them. Under the scenarios modeled, a subset of these facilities retires rather than
retrofit, since they would face additional requirements under the GHG regulations modeled. This
in turn results in lower capital costs in the first run year and is balanced by higher costs in later
years. Additionally, renewable costs are assumed to decline over the forecast period. Given
IPM's perfect foresight, the model choses to wait to build incremental RE until later in the period
when costs are lower. Under the illustrative policy scenarios the model builds this capacity
sooner, which results in lower costs in the years built, but higher costs in future years.

Costs peak in 2035 across all scenarios, reflecting the date of imposition of the final
Emission Guidelines for coal-fired steam generating units and tightening NSPS requirements.
The final rules scenario results in the greatest early buildout of RE, resulting in the lowest near-
term costs and higher longer-term costs. As a result, over the 2024 - 2047 time period, the final
rules scenario shows slightly lower costs than alternative 1 and alternative 2. However, over the
entire forecast period, costs are higher under the final rules.77

In addition to evaluating annual compliance cost impacts, EPA believes that a full
understanding of these three illustrative scenarios benefits from an evaluation of annualized costs
over the 2028 to 2045 timeframe. Starting with the estimated annual cost time series, it is
possible to estimate the net present value of that stream, and then estimate a levelized annual cost
associated with compliance with each illustrative scenario.78 For this analysis we first calculated
the PV of the stream of costs from 2024 through 204579 using a 3.76 percent discount rate. In
this cost annualization, we use a 3.76 percent discount rate, which is consistent with the rate used
in IPM's objective function for minimizing the PV of the stream of total costs of electricity
generation. This discount rate is meant to capture the observed equilibrium market rate at which

76	For more information, please see Chapter 2 of the IPM documentation.

77	The present value of costs over the 2024-57 time period using a 3.76 percent discount rate are $18.6 billion for the

alternative 1, $18.8 billion for the final rules, and $18.1 billion for the alternative 2.

78	The XNPV() function in Microsoft Excel for Windows 365 was used to calculate the PV of the variable stream of

costs, and the PMT() function in Microsoft Excel for Windows 365 is used to calculate the level annualized cost
from the estimated PV.

79	Consistent with the relationship between IPM run years and calendar years, EPA assigned run year compliance

cost estimates to all calendar years mapped to that run year. For more information, see Chapter 7 of the IPM
Documentation.

3-19


-------
investors are willing to sacrifice present consumption for future consumption and is based on a
Weighted Average Cost of Capital (WACC).80 After calculating the PV of the cost streams, the
same 3.76 percent discount rate and 2024 to 2047 time period are used to calculate the levelized
annual (i.e., annualized) cost estimates shown in Table 3-7.81 The same approach was used to
develop the annualized cost estimates for the 2024 to 2047 timeframe.

3.6.3 Impacts on Fuel Use, Prices, and Generation Mix

The final NSPS, and the final Emissions Guidelines are expected to result in significant
GHG emissions reductions. The rules are also expected to have some impacts to the economics
of the power sector. Consideration of these potential impacts is an important component of
assessing the relative impact of the illustrative scenarios. In this section we discuss the estimated
changes in fuel use, fuel prices, generation by fuel type, capacity by fuel type, and retail
electricity prices for the 2028, 2030, 2035, 2040, and 2045 IPM model run years.

Table 3-8 and Table 3-9 present the percentage changes in national coal and natural gas
usage by EGUs in the 2028, 2030, 2035, 2040 and 2045 run years. These fuel use estimates
reflect some power companies choosing natural gas and renewables over coal in 2030 rather than
implement available cost-reasonable controls as a result of the imposition of GHG mitigation
measures under the final Emissions Guidelines for coal-fired steam generating units.

Under the baseline, current market trends persist and are accentuated by the IRA. Hence
coal capacity continues to decline over the forecast period, and there is continued penetration of
non-emitting resources such as wind and solar.

Of the 181 GW of coal-fired capacity active in 2023, only 80 GW have not announced
retirement or coal to gas conversion by 2040. Furthermore, of these 80 GW, by 2040 56 GW will
be 53 years or older (which is the average retirement age for coal EGUs over the 2015-22 time

80	The IPM Baseline run documentation (Appendix B.4.1 Introduction to Discount Rate Calculations) states "The

real discount rate for all expenditures (capital, fuel, variable operations and maintenance, and fixed operations
and maintenance costs) in the EPA Platform v6 is 3.76 percent."

81	The PMT() function in Microsoft Excel for Windows 365 is used to calculate the level annualized cost from the

estimated PV.

3-20


-------
period).82 EPA projects that under the baseline, 42 GW of coal are projected to be active in 2040.
Coal consumption declines consistent with this decrease in capacity over the forecast period.

At the same time, tighter natural gas markets as a result of increased LNG exports results
in declining gas consumption over the forecast period, particularly after 2035, when improved
renewable cost and performance consistent with NREL ATB 2023 further erode gas generation
share. In the baseline, increases in LNG exports reach their highest levels by 2040, resulting in
tighter natural gas markets. At the same time, RE cost and performance improvements mean that
RE becomes more competitive. This results in less gas and more RE deployment, driving down
emissions. In other words, emissions decline over the forecast period in the baseline, and decline
faster in 2040 and 2045. Steam retirements continue over the forecast period. As a result in the
policy scenario, requirements on existing steam generation result in large reductions in 2035
driven by lower thermal generation and increased adoption of BSER technology but then
emissions begin to converge back to baseline levels.

Under the illustrative scenarios, increases in gas demand are highest in 2035, driven by
reductions in coal-fired generation as a result of the existing source standards. After 2035, the
absolute increases in gas consumption are smaller, consistent with baseline trends towards
declining gas consumption and higher levels of RE deployment. In 2030, increases in gas
consumption are lowest under the alternative 2 scenario, consistent with shifting the
requirements on medium-term coal fired electricity generating steam units assumed from 2030
under the final rules and alternative 1 to 2035 in the alternative 2 scenario.

To put these reductions into context, under the Baseline, power sector coal consumption is
projected to decrease from 251 million tons in 2028 to 222 million tons in 2030 (5 percent
annually between 2028-2030), and to 147 million tons in 2035 (7 percent annually between
2030-2035). Under the final rules, coal consumption is projected to decrease from 234 million
tons in 2028 to 194 million tons in 2030 (8 percent annually between 2028-2030), and 111
million tons in 2035 (9 percent annually between 2030-2035). Between 2015 and 2020, annual
coal consumption in the electric power sector fell between 8 and 19 percent annually.83 Coal

82	The annual average retirement age for coal-fired EGUs between 2000-2022 ranged between 47 and 61 years old,

and the average retirement age over that period was 53 years. Similarly, the average age for retiring coal-fired
EGUs between 2015-2022 was 53 years, demonstrating the consistency of retirement ages throughout the years.

83	U.S. EIA Monthly Energy Review, Table 6.2, January 2022.

3-21


-------
consumption falls by the greatest amount in 2035, consistent with the imposition of the
requirements on existing coal-fired steam generating units. For units that adopt CCS, 45Q tax
credits result in higher levels of dispatch and therefore coal consumption at those sources relative
to the baseline. These sources consume different types of coal depending on location and relative
cost, resulting in non-uniform subnational coal consumption impacts (i.e. production declines in
some regions and increases in others).

Table 3-10 presents the projected hydrogen power sector consumption under the Baseline
and the Illustrative Scenarios.84

Table 3-8 2028, 2030, 2035, 2040 and 2045 Projected U.S. Power Sector Coal Use for
the Baseline and the Illustrative Scenarios

Million Tons



Year

Baseline

Final

Alt. 1

Alt. 2

Final

Alt. 1

Alt. 2

Appalachia



40

37

36

37

-7%

-8%

-7%

Interior



38

35

36

36

-7%

-5%

-4%

Waste Coal

2028

7

7

7

7

0%

0%

0%

West



166

155

156

156

-7%

-6%

-6%

Total



251

234

235

237

-7%

-6%

-6%

Appalachia



39

39

39

39

0%

1%

0%

Interior



35

36

36

34

1%

2%

-2%

Waste Coal

2030

7

7

7

7

0%

0%

0%

West



141

113

113

133

-20%

-20%

-6%

Total



222

194

195

214

-13%

-12%

-4%

Appalachia



32

19

19

19

-40%

-40%

-40%

Interior



19

25

25

25

30%

30%

30%

Waste Coal

2035

7

3

3

3

-53%

-53%

-53%

West



89

63

63

67

-29%

-29%

-25%

Total



147

111

111

114

-25%

-25%

-22%

Appalachia



19

19

19

19

1%

1%

0%

Interior



10

25

25

25

150%

150%

150%

Waste Coal

2040

3

3

3

3

0%

0%

0%

West



61

56

56

59

-8%

-8%

-3%

Total



93

103

103

106

11%

11%

14%

Appalachia



4

0

0

0

-100%

-100%

-100%

Interior



1

0

0

0

-100%

-100%

-85%

Waste Coal

2045

3

0

0

0

-100%

-100%

-100%

West



20

3

3

3

-85%

-85%

-84%

Total



28

3

3

3

-89%

-90%

-88%

Percent Change from Baseline

84 Please note that hydrogen consumption is rounded to the nearest trillion Btu.

3-22


-------
Table 3-9 2028, 2030, 2035, 2040 and 2045 Projected U.S. Power Sector Natural Gas

Use for the Baseline and the Illustrative Scenarios

Trillion Cubic Feet

Percent Change from Baseline

Year

Baseline

Final

Alt. 1

Alt. 2

Final

Alt. 1

Alt. 2

2028

11.6

11.5

11.5

11.5

-1.0%

-1.0%

-1.0%

2030

11.7

11.7

11.7

11.5

0.0%

0.0%

-1.7%

2035

9.3

9.7

9.7

9.7

4.3%

4.4%

4.4%

2040

6.4

6.4

6.4

6.4

-0.1%

0.0%

0.0%

2045

4.2

4.3

4.3

4.3

1.1%

1.9%

1.8%

Table 3-10 2028, 2030, 2035, 2040 and 2045 Projected U.S. Power Sector Hydrogen Use
for the Baseline and the Illustrative Scenarios

Trillion Btu

Year	Baseline	Final	Alt. 1	Alt. 2

2028	0.00	0.00	0.00	0.00

2030	0.00	0.00	0.00	0.00

2035	0.00	0.00	0.00	0.00

2040	0.00	0.00	0.00	0.00

2045	0.22	0.45	0.49	0.44

Table 3-11 and Table 3-12 present the projected coal and natural gas prices in 2028, 2030,
2035, 2040 and 2045, as well as the percent change from the baseline projected due to the
illustrative scenarios. In 2028, earlier RE builds result in lower gas consumption and higher gas
prices. By 2030, reductions in coal generation stemming from requirements on medium-term
coal fired electricity generating steam units result in higher gas consumption and prices. In 2035,
increases in gas consumption are highest relative to the baseline, stemming from the
requirements on long-term coal fired electricity generating steam units and the NSPS. Impacts
lessen in 2040 onwards as the system approaches baseline levels with higher levels of RE
generation, and less gas and coal generation.

Under the alternative 2, gas prices remain similar to baseline levels in 2030 as a result of
the requirements on medium-term coal fired electricity generating steam units assumed to take

3-23


-------
place in 2035. Under the final rules scenario, the lower baseload threshold results in lower
amounts of generation from new gas, resulting in smaller increases in gas generation relative to
baseline levels than under the alternative 1 and alternative 2 illustrative scenarios.

Growing LNG exports result in tighter natural gas markets, particularly in 2035 and
beyond, while RE cost and performance continues to improve. At the same time, requirements
on new combustion turbines result in fewer new NGCCs that run at baseload levels. This means
that as steam generation falls, it is filled by a combination of higher existing gas and new RE
generation, tamping down on the increase in gas consumption.

Table 3-11 2028, 2030, 2035, 2040 and 2045 Projected Minemouth and Power Sector
Delivered Coal Price (2019 dollars) for the Baseline and the Illustrative Scenarios	

$/MMBtu

Percent Change from Baseline





Baseline

Final

Alt. 1

Alt. 2

Final

Alt. 1

Alt. 2

Minemouth
Delivered

2028

0.98
1.54

0.97
1.52

0.97
1.52

0.97
1.52

-1%
-1%

-1%
-1%

-1%
-1%

Minemouth
Delivered

2030

1.02
1.56

1.05
1.53

1.05
1.53

1.02
1.54

3%
-2%

3%
-2%

0%
-1%

Minemouth
Delivered

2035

1.07
1.55

1.10
1.55

1.10
1.55

1.09
1.54

3%
0%

3%
0%

2%
0%

Minemouth
Delivered

2040

1.17
1.59

1.22
1.60

1.22
1.60

1.21
1.60

4%
1%

4%
1%

3%
0%

Minemouth
Delivered

2045

1.37

1.38

1.50
0.94

1.50
0.94

1.50
0.94

9%
-32%

9%
-32%

9%
-32%

3-24


-------
Table 3-12 2028, 2030, 2035, 2040 and 2045 Projected Henry Hub and Power Sector

Delivered Natural Gas Price (2019 dollars) for the Baseline and the Illustrative Scenarios

$/MMBtu

Percent Change from Baseline





Baseline

Final

Alt. 1

Alt. 2

Final

Alt. 1

Alt. 2

Henry Hub
Delivered

2028

2.78
2.84

2.72
2.78

2.74
2.80

2.74
2.80

-2%
-2%

-2%
-2%

-2%
-2%

Henry Hub
Delivered

2030

2.89
2.95

2.90
2.97

2.91
2.98

2.87
2.93

0%
1%

1%
1%

-1%

0%

Henry Hub
Delivered

2035

2.87

2.88

2.95
2.97

2.95
2.97

2.95
2.97

3%
3%

3%
3%

3%
3%

Henry Hub
Delivered

2040

2.82
2.79

2.79
2.77

2.81
2.79

2.81
2.79

-1%
-1%

0%
0%

0%
0%

Henry Hub
Delivered

2045

2.95
2.94

2.95
2.94

2.95
2.94

2.95
2.94

0%
0%

0%
0%

0%
0%

Gas capacity is higher as a result of greater NGCT buildout. These NGCT units operate at
low capacity factors, which means gas consumption is similar between the two scenarios, as is
natural gas price.

Table 3-13 presents the projected percentage changes in the amount of electricity
generation in 2028, 2030, 2035 and 2040 by fuel type. Consistent with the fuel use projections
and emissions trends above, EPA projects an overall shift from coal to gas and renewables under
the baseline, and these trends persist under the illustrative scenarios analyzed. The projected
impacts are highest in 2035 reflecting the imposition of the final Emissions Guidelines and are
smaller thereafter. 45(q) is available for 12 years within the modeling,85 after which point units
no longer receive tax credits and must dispatch based on unsubsidized operating costs.

85 EPA assumes a 12-year booklife for CCS consistent with the duration of the 45(q) tax credit

3-25


-------
Table 3-13 2028, 2030, 2035, 2040 and 2045 Projected U.S. Generation by Fuel Type for

the Baseline and the Illustrative Scenarios

Generation (TWh)

Percent Change from Baseline



Year

Baseline

Baseline

Final

Alt. 1

Final

Alt. 1

Alt. 2

Unabated Coal



472

441

443

447

-7%

-6%

-5%

Coal & CCS



0

0

0

0

-

-

-

Coal with Nat. Gas co-firing



0

0

0

0

-

-

-

Unabated Nat. Gas



1,652

1,631

1,634

1,633

-1%

-1%

-1%

Nat. Gas & CCS



0

0

0

0

-

-

-

Nuclear

2028

751

751

751

751

0%

0%

0%

Hydro



293

293

293

293

0%

0%

0%

Non-Hydro RE



1,141

1,191

1,186

1,182

4%

4%

4%

Oil/Gas Steam



26

28

27

27

8%

7%

7%

Other



31

31

31

31

0%

0%

0%

Grand Total



4,365

4,366

4,365

4,365

0%

0%

0%

Unabated Coal



407

355

357

391

-13%

-12%

-4%

Coal & CCS



3

5

5

3

71%

76%

0%

Coal with Nat. Gas co-firing



0

2

2

0

-

-

-

Unabated Nat. Gas



1,670

1,660

1,664

1,642

0%

0%

-1%

Nat. Gas & CCS



0

0

0

0

-

-

-

Nuclear

2030

729

729

729

729

0%

0%

0%

Hydro



298

299

298

298

0%

0%

0%

Non-Hydro RE



1,329

1,381

1,377

1,373

4%

4%

3%

Oil/Gas Steam



25

28

27

25

12%

11%

3%

Other



31

31

31

31

0%

0%

0%

Grand Total



4,491

4,491

4,491

4,491

0%

0%

0%

Unabated Coal



160

0

0

0

-100%

-100%

-100%

Coal & CCS



76

133

133

136

74%

74%

78%

Nat. Gas co-firing



0

4

4

6

-

-

-

Unabated Nat. Gas



1,341

1,379

1,386

1,384

4%

4%

4%

Nat. Gas & CCS



3

7

6

6

105%

64%

64%

Nuclear

2035

667

666

666

666

0%

0%

0%

Hydro



319

317

317

317

-1%

-1%

-1%

Non-Hydro RE



2,229

2,286

2,281

2,278

3%

2%

2%

Oil/Gas Steam



8

9

9

9

21%

16%

17%

Other



31

30

30

30

0%

0%

0%

Grand Total



4,834

4,831

4,833

4,832

0%

0%

0%

Unabated Coal
Coal & CCS

2040

61
76

0

128

0

128

0

131

-100%
68%

-100%
68%

-100%
73%

3-26


-------
Coal with Nat. Gas co-firing



0

0

0

0

-

-

-

Unabated Nat. Gas



933

919

924

924

0%

0%

0%

Nat. Gas & CCS



3

7

6

6

105%

64%

64%

Nuclear



614

613

613

613

0%

0%

0%

Hydro



336

336

336

336

0%

0%

0%

Non-Hydro RE



3,097

3,119

3,114

3,111

1%

1%

0%

Oil/Gas Steam



5

6

6

6

28%

27%

27%

Other



29

29

29

29

0%

0%

0%

Grand Total



5,154

5,157

5,156

5,155

0%

0%

0%

Unabated Coal



45

0

0

0

-100%

-100%

-100%

Coal & CCS



4

3

3

4

-7%

-9%

4%

Coal with Nat. Gas co-firing



0

0

0

0

-

-

-

Unabated Nat. Gas



614

612

620

619

1%

2%

2%

Nat. Gas & CCS



3

6

5

5

103%

65%

65%

Nuclear

2045

471

472

473

473

0%

0%

0%

Hydro



343

342

342

342

0%

0%

0%

Non-Hydro RE



4,032

4,089

4,081

4,081

1%

1%

1%

Oil/Gas Steam



4

6

6

6

25%

25%

25%

Other



28

27

27

27

0%

0%

0%

Grand Total



5,544

5,557

5,557

5,556

0%

0%

0%

Note: In this table, "Non-Hydro RE" includes biomass, geothermal, landfill gas, solar, and wind. Oil/Gas steam
category includes coal to gas conversions.

Table 3-14 presents the projected percentage changes in the amount of generating capacity
in 2028, 2030, 2035, 2040 and 2045 by primary fuel type. In 2035, the final Emission Guidelines
is assumed to be in effect under all three scenarios. Under the final rules, 104 GW of coal-fired
EGUs have committed retirements by 2035 (21 GW incremental to baseline). One GW of coal-
fired EGUs who have committed to retirement by 2040 are medium-term existing coal-fired
steam generating units and, as such, install 40 percent natural gas co-firing requirement. 19 GW
of coal-fired EGUs who plan to operate past 2040 are subject to the long-term existing coal-fired
steam generating unit subcategory and, as such, install CCS (reflecting 8 GW incremental to the
baseline). Finally, 19 GW of coal-fired EGUs undertake coal to gas conversion (6 GW
incremental to the baseline).

Under the baseline, total coal retirements between 2028 and 2035 are projected to be 84
GW (or 12 GW annually). Under the final rules, total coal retirements between 2028 and 2035

3-27


-------
are projected to be 104 GW (or 15 GW annually). This is compared to an average recent
historical retirement rate of 11 GW per year from 2015 - 2020.86

By 2030 the final rules are projected to result in an additional 5 GW of coal retirements, by
2035 an incremental 21 GW of coal retirements and by 2040 an incremental 14 GW of coal
retirements relative to the baseline. These compliance decisions reflect EGU operators making
least-cost decisions on how to achieve efficient compliance with the rules while maintaining
sufficient generating capacity to maintain resource adequacy.87

IPM endogenously estimates the capacity credit (i.e. the accredited capacity that can count
towards meeting the resource adequacy constraints within the model) for wind, solar, and storage
as a function of penetration.88 Additionally, IPM models operating reserves at the regional level,
and can account for the impact of solar and wind on operating reserves requirements.89

An incremental 15 GW of renewable capacity additions (consisting of an incremental 3
GW of solar and 12 GW of wind builds) and 9 GW of storage is projected by 2035 in the
illustrative final rule. Under the final rules, 18 GW of economic NGCC additions occur by 2035
(1 GW less than the baseline), and 24 GW of economic NGCT additions occur by 2035 (10 GW
incremental to the baseline). These builds partially reflect early action, i.e., builds that would
otherwise have occurred later in the forecast period under the baseline. Of these units, 870 MW
of NGCCs install CCS in 2035.

Under the baseline, the reduction in generation from natural-gas and coal fired facilities is
greater than the reduction in their capacities over time. Hence thermal resources tend to be
operated less frequently over time, due to the increase in low-emitting generation. These trends
persist under the illustrative scenarios.

As shown in Figure 3.2 below, The coal-fired generation share was 49 percent in 2007 and
20 percent in 2022, and is projected to fall to 3 percent in 2040 under the baseline and 1 percent

86	See EIA's Today in Energy: https://www.eia.gov/todayinenergy/detail.php?id=50838.

87	For further discussion of how the rule is anticipated to integrate into the ongoing power sector transition while not

impacting resource adequacy, see section XIV(F) of the preamble, and the Resource Adequacy Assessment TSD
included in the docket.

88	For details, please see chapter 4 of the IPM documentation, available at: https://www.epa.gov/power-sector-

modeling

89	For details, please see chapter 3 of the IPM documentation, available at: https://www.epa.gov/power-sector-

modeling

3-28


-------
by 2045. Under the final rules scenario, coal-fired generation share is projected to fall to 2
percent by 2040 and less than 1 percent in 2045. The natural gas-fired generation share was 22
percent in 2007 and 39 percent in 2022 and is projected to fall to 18 percent in 2040 under the
baseline and 11 percent by 2045. Under the final rules scenario, natural gas-fired generation
share is projected to fall to 17 percent by 2040 and 11 percent in 2045. The wind and solar
generation share was 1 percent in 2007 and 13 percent in 2022 and is projected to grow to 57
percent in 2040 under the baseline and 69 percent by 2045. Under the final rules scenario, wind
and solar generation share is projected to grow to 58 percent by 2040 and 69 percent in 2045.

3-29


-------
Figure 3-1 Historical and Projected Capacity Mix (GW)

2,500

¦ Coal ¦ Natural Gas Nuclear ¦ Hydro Wind Solar Distributed Solar ¦ Other Renewable

2,000

1,500

1,000

500

I i i 111 i 11 i I i 1111

r"-»ooOHrMro"d-ir><£>r~xOOCT>OHr\i
OOOrHrHTHrHrHrHrH^H^HrHrslfNfM

oooooooooooooooo

rM(NfN(N(Nrj(N(NrM(NrM(N(N(N(N(N

— 13 —

Historical

2028 . 2030

2035

2040

Sources: EI A Power Annual and EPA projections

Figure 3-2 Historical and Projected Generation Mix (GW)

7,000

I Coal ¦ Natural Gas Nuclear ¦ Hydro

Wind

Solar

Distributed Solar ¦ Other Renewable

6,000
5,000
4,000 |
3,000
2,000
1,000

II

h in m i^i io

oo at

O O O t—It—I*—I*—It—It—I t—I t—I t—I H OJ (N . .
OOOOOOOOOOOOOOOO
lN(N(NIN(NfNfNlNfMfM(N(NfM(N(N(N

Historical

¦

n

i

I

1



¦



1

1

1

1

0;

i/i

QJ



c

_oj

C

_aj

ID

Cd

"a>

Z3

cc

nj

—

03

—

CO

03
C

CO

ro
C



Ll.



"u_

2028 .

2030

2035

2040

Sources: EIA Power Annual and EPA projections

3-30


-------
Table 3-14 2028, 2030, 2035, 2040 and 2045 Projected U.S. Capacity by Fuel Type for

the Baseline and the Illustrative Scenarios

Capacity (GW)

Percent Change from
Baseline



Year

Bas
elin
e

Final
Rules

Alt.
1

Alt.

2

Final Rules

Alt. 1

Alt. 2

Unabated Coal



106

101

101

102

-4%

-4%

-4%

Coal & CCS



0

0

0

0

-

-

-

Coal with Nat. Gas co-firing



0

0

0

0

-

-

-

Unabated Nat. Gas



471

472

473

472

0%

0%

0%

Nat. Gas & CCS



0

0

0

0

-

-

-

Nuclear

2028

94

94

94

94

0%

0%

0%

Hydro



102

102

102

102

0%

0%

0%

Non-Hydro RE



394

407

406

405

3%

3%

3%

Oil/Gas Steam



63

64

64

64

2%

2%

2%

Other



7

7

7

7

0%

0%

0%

Grand Total



1,2

36

1,246

1,246

1,245

1%

1%

1%

Unabated Coal



85

72

72

80

-15%

-15%

-5%

Coal & CCS



0

1

1

0

72%

77%

0%

Coal with Nat. Gas co-firing



0

1

1

0

-

-

-

Unabated Nat. Gas



479

480

481

480

1%

1%

1%

Nat. Gas & CCS



0

0

0

0

-

-

-

Nuclear

2030

91

91

91

91

0%

0%

0%

Hydro



104

104

104

104

0%

0%

0%

Non-Hydro RE



440

454

453

452

3%

3%

3%

Oil/Gas Steam



64

73

73

66

13%

13%

3%

Other



7

7

7

7

0%

0%

0%

Grand Total



1,2

69

1,281

1,281

1,280

1%

1%

1%

Unabated Coal



41

0

0

0

-100%

-100%

-100%

Coal & CCS



11

19

19

20

74%

74%

78%

Coal with Nat. Gas co-firing



0

1

1

1

-

-

-

Unabated Nat. Gas



476

484

484

484

2%

2%

2%

Nat. Gas & CCS



0

1

1

1

104%

63%

63%

Nuclear

2035

84

84

84

84

0%

0%

0%

Hydro



107

107

107

107

0%

0%

0%

Non-Hydro RE



699

714

713

711

2%

2%

2%

Oil/Gas Steam



55

66

66

66

19%

19%

20%

Other



7

7

7

7

0%

0%

0%

Grand Total



1,4

79

1,482

1,481

1,481

0%

0%

0%

3-31


-------
Unabated Coal



31

0

0

0

-99%

-99%

-99%

Coal & CCS



11

18

18

19

68%

68%

73%

Coal with Nat. Gas co-firing



0

0

0

0

-

-

-

Unabated Nat. Gas



516

525

525

525

2%

2%

2%

Nat. Gas & CCS



0

1

1

1

104%

63%

63%

Nuclear

2040

79

79

79

79

0%

0%

0%

Hydro



112

112

112

112

0%

0%

0%

Non-Hydro RE



943

952

951

950

1%

1%

1%

Oil/Gas Steam



54

65

65

65

19%

19%

20%

Other



7

7

7

7

0%

0%

0%

Grand Total



1,7

53

1,759

1,757

1,757

0%

0%

0%

Unabated Coal



29

0

0

0

-99%

-99%

-99%

Coal & CCS



1

1

1

1

-5%

-7%

13%

Coal with Nat. Gas co-firing



0

0

0

0

-

-

-

Unabated Nat. Gas



565

581

581

580

3%

3%

3%

Nat. Gas & CCS



0

1

1

1

104%

63%

63%

Nuclear



65

65

65

65

0%

0%

0%



2045















Hydro



112

112

112

112

0%

0%

0%

Non-Hydro RE



1,2

32

1,250

1,248

1,248

1%

1%

1%

Oil/Gas Steam



54

64

64

65

19%

19%

20%

Other



7

7

7

7

0%

0%

0%

Grand Total



2,0
65

2,080

2,078

2,078

1%

1%

1%

Note: In this table, "Non-Hydro RE" includes biomass, geothermal, landfill gas, solar, and wind

EPA estimated the change in the retail price of electricity (2019 dollars) using the Retail
Price Model (RPM).90 The RPM was developed by ICF for EPA and uses the IPM estimates of
changes in the cost of generating electricity to estimate the changes in average retail electricity
prices. The prices are average prices over consumer classes (i.e., consumer, commercial, and
industrial) and regions, weighted by the amount of electricity used by each class and in each
region. The RPM combines the IPM annual cost estimates in each of the 64 IPM regions with

90 See documentation available at: https://www.epa.gov/airmarkets/retail-price-model

3-32


-------
EIA electricity market data for each of the 25 electricity supply regions in the electricity market
module of the National Energy Modeling System (NEMS).91

Table 3-15, Table 3-16, Table 3-17, and Table 3-18 present the projected percentage
changes in the retail price of electricity for the three illustrative scenarios in 2030, 2035, 2040
and 2045, respectively. Consistent with other projected impacts presented above, average retail
electricity prices at both the national and regional level are projected to experience the largest
impacts in 2035. Consistent with the decline in total production cost in 203092 National
electricity rates are projected to fall 0.5 percent below baseline levels in 2030, or a decrease of
0.47 mills/kWh (2019 dollars). In 2035, EPA estimates that these rules will result in a 1 percent
increase in national average retail electricity price, or by about 1.33 mills/kWh (2019 dollars). In
2040, EPA estimates that these rules will result in a 0.2 percent increase in national average retail
electricity price, or by about 0.15 mills/kWh. In 2045, EPA estimates that these rules will result
in a 0.7 percent increase in national average retail electricity price, or by about 0.63 mills/kWh.

91	See documentation available at:

https://www.eia.gov/outlooks/aeo/nems/documentation/electricity/pdf/EMM_2022.pdf

92	Under the baseline, the proposed GNP rule results in installation of SCR controls in the 2030 run year on some

coal-fired EGUs that currently lack them. Under the scenarios modeled, a subset of these facilities retires rather
than retrofit, since they would face additional requirements under the GHG regulations modeled. This in turn
results in lower capital costs in the first run year and is balanced by higher costs in later years. Additionally,
renewable costs are assumed to decline over the forecast period. Given IPM's perfect foresight, the model choses
to wait to build incremental RE until later in the period when costs are lower. Under the illustrative policy
scenarios the model builds this capacity sooner, which results in lower costs in the years built, but higher costs in
future years.

3-33


-------
Table 3-15 Average Retail Electricity Price by Region for the Baseline and the
Illustrative Scenarios, 2030	

All Sector

2030 Average Retail Electricity Price
(2019 mills/kWh)

Percent Change from Baseline

Region

Baseline

Final
Rules

Alt. 1

Alt. 2

Final
Rules

Alt. 1

Alt. 2

TRE

73

73

73

73

-1%

-1%

-1%

FRCC

98

98

98

97

0%

0%

0%

MISW

93

93

93

93

0%

0%

-1%

MISC

91

91

91

91

0%

0%

-1%

MISE

109

107

107

107

-2%

-2%

-2%

MISS

86

83

83

83

-3%

-3%

-3%

ISNE

157

157

156

156

0%

0%

0%

NYCW

210

211

211

211

0%

0%

0%

NYUP

126

126

126

126

0%

0%

0%

PJME

110

107

107

106

-3%

-3%

-3%

PJMW

97

97

96

96

0%

-1%

-1%

PJMC

89

87

87

87

-3%

-3%

-3%

PJMD

76

77

76

76

0%

-1%

-1%

SRCA

92

92

92

92

0%

0%

0%

SRSE

95

95

95

95

0%

0%

0%

SRCE

71

71

71

71

0%

0%

0%

SPPS

78

78

78

77

0%

1%

-1%

SPPC

97

97

96

97

-1%

-1%

-1%

SPPN

65

66

66

65

1%

1%

0%

SRSG

102

102

102

102

0%

0%

0%

CANO

143

142

142

142

0%

-1%

-1%

CASO

174

173

173

173

0%

0%

-1%

NWPP

82

81

81

81

-1%

-1%

-1%

RMRG

101

100

100

100

0%

-1%

-1%

BASN

96

97

97

97

1%

1%

1%

NATIONAL

100

99

99

99

-0.5%

-1%

-1%

3-34


-------
Table 3-16 Average Retail Electricity Price by Region for the Baseline and the
Illustrative Scenarios, 2035	

All Sector

2035 Average Retail Electricity Price
(2019 mills/kWh)

Percent Change from Baseline

Region

Baseline

Final
Rules

Alt. 1

Alt. 2

Final
Rules

Alt. 1

Alt. 2

TRE

78

80

80

80

2%

2%

2%

FRCC

92

92

92

92

1%

1%

1%

MISW

84

85

85

85

1%

1%

1%

MISC

81

82

82

82

1%

1%

1%

MISE

96

99

98

98

3%

3%

3%

MISS

79

81

81

81

2%

2%

2%

ISNE

156

156

156

156

0%

0%

0%

NYCW

209

210

210

210

0%

0%

0%

NYUP

125

126

125

125

1%

1%

1%

PJME

108

113

112

112

4%

3%

3%

PJMW

92

95

94

94

3%

3%

3%

PJMC

75

79

79

79

6%

5%

5%

PJMD

71

74

74

74

4%

3%

3%

SRCA

89

90

90

90

0%

0%

0%

SRSE

90

91

91

91

1%

1%

1%

SRCE

67

67

67

67

0%

0%

0%

SPPS

69

70

70

70

1%

1%

1%

SPPC

80

80

80

80

-1%

-1%

0%

SPPN

63

64

64

64

1%

1%

1%

SRSG

103

104

104

104

0%

0%

0%

CANO

140

141

141

141

1%

1%

1%

CASO

173

173

173

173

0%

0%

0%

NWPP

79

79

79

79

0%

0%

0%

RMRG

93

95

95

95

2%

2%

2%

BASN

97

96

96

96

-1%

-1%

-1%

NATIONAL

96

97

97

97

1%

1%

1%

3-35


-------
Table 3-17 Average Retail Electricity Price by Region for the Baseline and the
Illustrative Scenarios, 2040	

All Sector

2040 Average Retail Electricity Price
(2019 mills/kWh)

Percent Change from Baseline

Region

Baseline

Final
Rules

Alt. 1

Alt. 2

Final
Rules

Alt. 1

Alt. 2

TRE

74

73

73

73

0%

0%

0%

FRCC

88

89

89

89

0%

0%

0%

MISW

79

80

80

80

1%

1%

1%

MISC

73

73

73

73

0%

0%

0%

MISE

98

98

98

98

0%

0%

0%

MISS

74

75

75

74

1%

1%

0%

ISNE

167

168

168

168

1%

1%

1%

NYCW

236

235

235

235

0%

0%

0%

NYUP

138

138

138

138

0%

0%

0%

PJME

117

117

117

117

0%

0%

0%

PJMW

89

90

89

90

0%

0%

0%

PJMC

78

78

78

78

0%

0%

0%

PJMD

74

74

74

74

0%

0%

0%

SRCA

87

87

87

87

0%

0%

0%

SRSE

84

84

84

84

0%

0%

0%

SRCE

66

66

66

66

0%

0%

0%

SPPS

66

66

66

66

1%

1%

1%

SPPC

76

76

76

76

0%

0%

0%

SPPN

62

62

62

62

0%

0%

0%

SRSG

98

98

98

98

0%

0%

0%

CANO

144

145

145

145

0%

0%

0%

CASO

172

173

173

173

0%

0%

0%

NWPP

81

80

80

80

-1%

-1%

-1%

RMRG

88

88

88

88

1%

1%

0%

BASN

96

95

95

95

-1%

-1%

-1%

NATIONAL

95

95

95

95

0.2%

0.2%

0.1%

3-36


-------
Table 3-18 Average Retail Electricity Price by Region for the Baseline and the
Illustrative Scenarios, 2045	

All Sector

2045 Average Retail Electricity Price
(2019 mills/kWh)

Percent Change from Baseline

Region

Baseline

Final
Rules

Alt. 1

Alt. 2

Final
Rules

Alt. 1

Alt. 2

TRE

66

66

66

66

1%

1%

1%

FRCC

86

86

86

86

0%

0%

0%

MISW

77

78

78

78

1%

1%

1%

MISC

70

71

71

71

1%

1%

1%

MISE

94

95

95

95

1%

1%

1%

MISS

69

68

69

69

0%

0%

0%

ISNE

161

161

161

161

0%

0%

0%

NYCW

227

229

229

229

1%

1%

1%

NYUP

128

129

129

129

1%

1%

1%

PJME

116

116

116

116

0%

0%

0%

PJMW

87

87

87

87

1%

1%

1%

PJMC

76

77

77

77

0%

0%

0%

PJMD

74

74

74

74

0%

0%

0%

SRCA

87

87

87

87

0%

0%

0%

SRSE

84

85

85

85

1%

1%

1%

SRCE

64

65

65

65

1%

1%

1%

SPPS

65

66

66

66

1%

1%

2%

SPPC

70

71

71

71

1%

1%

1%

SPPN

63

65

65

66

3%

3%

3%

SRSG

94

95

95

95

1%

1%

1%

CANO

141

141

141

141

0%

0%

0%

CASO

171

171

171

171

0%

0%

0%

NWPP

82

84

84

84

3%

3%

3%

RMRG

83

85

85

85

2%

2%

2%

BASN

94

95

95

95

2%

1%

2%

NATIONAL

92

93

93

93

0.7%

0.6%

0.7%

3-37


-------
1231.
NWPP

[MISWj

19
SPPN

NYUP

*12 .
gjMCl

¦2ll

[CANO]

25
BASN

m24*-

RMRG

MISC

|(13K
PJMD

22
CASO

¦14J|

[srca!

NYCW

Figure 3-3 Electricity Market Module Regions

Source: EIA (http://www. eia. gov/forecasts/aeo/pdf/nerc_map.pdf)

3.7 Limitations

EPA's modeling is based on expert judgment of various input assumptions for variables
whose outcomes are uncertain. As a general matter, the Agency reviews the best available
information from engineering studies of air pollution controls and new capacity construction
costs to support a reasonable modeling framework for analyzing the cost, emission changes, and
other impacts of regulatory actions for EGUs. The annualized cost of the rules for EGUs, as
quantified here, is EPA's best assessment of the cost of implementing the rules for the power
sector. These costs are generated from rigorous economic modeling of anticipated changes in the
power sector due to implementation of the rule.

There are several key areas of uncertainty related to the electric power sector that are worth
noting, including:

• Electric demand: The analysis includes an assumption for future electric demand. This is
based on AEO 2023 reference case with incremental demand from EPA's OTAQ's on the books

3-38


-------
rules that are not captured in AEO 2023 reference case projections.93 To the extent electric
demand is higher or lower, it may increase/decrease the projected future thermal/renewable
composition of the fleet. Hence higher demand, all else equal, may result in fewer baseline
retirements, while a different load shape could incentivize different levels of RE penetration.

•	Natural gas supply and demand: The baseline includes significant growth in LNG
exports, driving tighter natural gas prices relative to the forecast used to estimate the impacts of
the final rules. To the extent prices are higher or lower, it would influence the use of natural gas
for electricity generation and overall competitiveness of other EGUs (e.g., coal, RE and nuclear
units).

•	Longer-term planning by utilities: Many utilities have announced long-term clean energy
and/or climate commitments, with a phasing out of large amounts of coal capacity by 2030 and
continuing through 2050. These announcements, some of which are not legally binding, are not
necessarily reflected in the baseline, and may alter the amount of coal capacity projected in the
baseline that would be covered under this rule.

•	Inflation Reduction Act (IRA): The IRA was passed in August of 2022. In order to
illustrate the impact of the IRA on this rulemaking, EPA included a baseline that incorporates
key provisions of the IRA as well as imposing the final rules as modeled in this RIA on that
baseline. However, additional effects of the IRA beyond those modeled in this RIA could result
in a change in projected system compliance costs and emissions outcomes.94

•	Hydrogen production: Currently, hydrogen is an exogenous input to the model,
represented as a fuel that is available at affected sources at a delivered cost of $1.15/kg. The
model does not track any upstream emissions95 associated with the production of the hydrogen,
nor any incremental electricity demand associated with its production.96 The incorporation of
these effects could change the amount of hydrogen selected as a compliance measure. The model

93	For details, see chapter 3 of the IPM documentation available at: https://www.epa.gov/power-sector-modeling

94	For details of IRA representation in this analysis please see IPM documentation, available at:

https://www.epa.gov/power-sector-modeling

95	IPM does not track upstream emissions for any modeled fuels.

96	Potential impacts associated with hydrogen production and utilization are discussed in preamble Sections

VII(F)(3), and XIV(E)(3). These include water use in hydrogen production, combustibility, and potential
increased NOx emissions from combustion of higher percentages of hydrogen in natural gas blends. Analysis in
this RIA does not assess these potential impacts, nor the potential impacts of hydrogen gas release on climate or
air quality through atmospheric chemical reactions.

3-39


-------
also does not account for any possible increases in NOx emission rates at higher levels of
hydrogen blending.97 For details on hydrogen modeling assumptions, please see Section 3.5.2.

The baseline includes modeling to capture the finalized 2020 Effluent Limitation
Guidelines (ELG), and it also incorporates information provided by owners of affected facilities
to state permitting authorities in October 2021 that indicate their likely compliance pathway,
including retirement by 2028. Potential future incorporation of this information may result in
additional coal plant retirements in an updated baseline scenario, which could affect modeled
costs and benefits of the rules depending on the extent that these retirements occur before
compliance deadlines for this action. Similarly, the baseline accounts for the effect of expected
compliance methods for the 2020 CCR Rule. It is possible that the waste streams of coal plants
are subject to multiple rules listed above, and that the interactions between these requirements
may alter compliance behavior that would likely occur under any of the rules in isolation, i.e.
plants may adopt compliance methods that are different than those represented in the baseline. In
order to estimate the impact of recently finalized EPA regulations, sensitivity analysis was
performed using IPM and included in the docket for this rulemaking that included a
characterization of the LDV, MDV and HDV (2024) vehicle rules, ELG (2024), and MATS
(2024) rules in addition to the final carbon rules presented in this RIA.

The impact of the Standards of Performance for New, Reconstructed, and Modified
Sources and Emissions Guidelines for Existing Sources: Oil and Natural Gas Sector Climate
Review98 are also not included in this analysis. Inclusion of these standards would likely increase
the price of natural gas modestly as a result of limitations on the usage of reciprocating internal
combustion engines in the pipeline transportation of natural gas. All else equal, inclusion of this
program would likely result in a modest increase in the total cost of compliance for this rule. The
proposed GNP Supplemental Rule (2023), the proposed Multi-Pollutant Emissions Standards for
Model Years 2027 and Later Light-Duty and Medium-Duty Vehicles (2023), the proposed
Heavy-duty Greenhouse Gas "Phase 3" for Model Years 2027 and Later (2023), the proposed
National Emission Standards for Hazardous Air Pollutants: Coal- and Oil-Fired Electric Utility

97	For details on the possible increases in NOx emission rates at higher levels of hydrogen blending, please see the

Hydrogen in Combustion Turbine Electricity Generating Units TSD, available in the docket for this rulemaking.

98	Available at: https://www.federalregister.gOv/documents/2021/l 1/15/2021-24202/standards-of-performance-for-

new-reconstructed-and-modified-sources-and-emissions-guidelines-for

3-40


-------
Steam Generating Units Review of the Residual Risk and Technology Review (2023), and the
proposed Steam Electric Power Generating Effluent Guidelines (2023) were not included.
Inclusion of these rules may result in changes projected compliance outcomes. Additionally,
EPA performed a variety of sensitivity analysis looking at lower natural gas prices, higher
electricity demand and also higher electricity demand coupled with EPA's additional Power
Sector Rules (MATS, ELG and the Final Rules). These sensitivity analyses continue to show that
the Final Rules, in the context of higher demand and other pending power sector rules, still
demonstrate compliance pathways that respect these NERC reliability considerations and
constraints, while achieving significant emissions reductions at reasonable costs. These results
are discussed in "IPM Sensitivities Technical MEMO" and "The Resource Adequacy Analysis:
Vehicle Rules, Final 111 EGU Rules, ELG, and MATS Technical MEMO" in the docket for this
rulemaking.

The IPM modeling of CCS is inclusive of the cost of installation and operating capture
technology and includes heat rate and capacity penalties to account for the parasitic load of the
capture equipment. The costs also reflect the cost of transport and storage of captured CO2 based
on the distance between CO2 production and storage sites. One possible area of uncertainty is
delays in the time taken to receive necessary permits, which are not modeled in IPM. As laid out
in the preamble, EPA has provided flexibilities in order to manage any delays in the process.

These are key uncertainties that may affect the overall composition of electric power
generation fleet and could thus have an effect on the estimated costs and impacts of this action.
However, these uncertainties would largely affect the modeling of the baseline and illustrative
scenarios similarly, and therefore, the impact on the incremental projections (reflecting the
potential costs/benefits of the regulatory alternatives) would be more limited and are not likely to
result in notable changes to the assessment of the final NSPS and Emission Guidelines found in
this section. While it is important to recognize these key areas of uncertainty, they do not change
EPA's overall confidence in the estimated impacts of the illustrative regulatory alternatives
presented in this section. EPA continues to monitor industry developments and makes
appropriate updates to the modeling platforms in order to reflect the best and most current data
available.

3-41


-------
3.8 References

Bistline, J., Mehrota, N., & Wolfram, C. (2023). Economic Implications of the Climate

Provisions of the Inflation Reduction Act. Brookings Papers on Economic Activity.
Retrieved from https://www.brookings.edu/wp-

content/uploads/2023/03/BPEA Spring2023 Bistline-et-al unembargoedUpdated.pdf

U.S. EPA. (2005). Regulatory Impact Analysis for the Final Clean Air Interstate Rule. Research
Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Health and Environmental Impact Division.
https://www.epa.gov/sites/default/files/2020-07/documents/transport ria final-clean-air-
interstate-rule 2005-03.pdf

U.S. EPA. (201 la). Regulatory Impact Analysis for the Federal Implementation Plans to Reduce
Interstate Transport of Fine Particulate Matter and Ozone in 27 States; Correction of
SIP Approvals for 22 States. Research Triangle Park, NC: U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards, Health and Environmental Impact
Division, https://www3.epa.gov/ttn/ecas/docs/ria/transport ria final-csapr 2011-06.pdf

U.S. EPA. (2011b). Regulatory Impact Analysis for the Final Mercury and Air Toxics Standards.
(EPA-452/R-11-011). Research Triangle Park, NC: U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards, Health and Environmental Impact
Division. http://www.epa.gov/ttn/ecas/regdata/RIAs/matsriafinal.pdf

U.S. EPA. (2014a). Economic Analysis for the Final Section 316(b) Existing Facilities Rule.
(EPA-821-R-14-001). Washington DC: U.S. Environmental Protection Agency.
https://www.epa.gov/sites/default/files/2015-05/documents/cooling-water phase-
4 economics 2014.pdf

U.S. EPA. (2014b). Guidelines for Preparing Economic Analyses. (EPA 240-R-10-001).

Washington DC: U.S. Environmental Protection Agency, Office of Policy, National
Center for Environmental Economics, https://www.epa.gov/environmental-
economics/guidelines-preparing-economic-analvses

U.S. EPA. (2015a). Benefit and Cost Analysis for the Effluent Limitations Guidelines and

Standards for the Steam Electric Power Generating Point Source Category. (EPA-821-
R-15-005). Washington DC: U.S. Environmental Protection Agency.
https://www.epa.gov/sites/default/files/2015-10/documents/steam-electric benefit-cost-
analysis 09-29-2015.pdf

U.S. EPA. (2015b). Regulatory Impact Analysis for the Clean Power Plan Final Rule. (EPA-
452/R-l5-003). Research Triangle Park, NC: U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards, Health and Environmental Impact
Division, https://www.epa.gov/sites/default/files/2020-07/documents/utilities ria final-
clean-power-plan-existing-units 2015-08.pdf

3-42


-------
U.S. EPA. (2015c). Regulatory Impact Analysis: EPA's 2015 RCRA Final Rule Regulating Coal
Combustion Residual (CCR) Landfills and Surface Impoundments At Coal-Fired Electric
Utility Power Plants. (EPA-821-R-20-003). Washington DC: U.S. Environmental
Protection Agency. https://www.regulations.gOv/document/EPA-HO-RCRA-2009-0640-
12034

U.S. EPA. (2016). Regulatory Impact Analysis of the Cross-State Air Pollution Rule (CSAPR)
Update for the 2008 National Ambient Air Quality Standards for Ground-Level Ozone.
(EPA-452/R-16-004). Research Triangle Park, NC: U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards, Health and Environmental Impact
Division, https://www.epa.gov/sites/default/files/2020-07/documents/transport ria final -
csapr-update 2016-09.pdf

U. S. EPA. (2019). Regulatory Impact Analysis for the Repeal of the Clean Power Plan, and the
Emission Guidelines for Greenhouse Gas Emissions from Existing Electric Utility
Generating Units. (EPA-452/R-19-003). Research Triangle Park, NC: U.S.

Environmental Protection Agency, Office of Air Quality Planning and Standards, Health
and Environmental Impact Division. https://www.epa.gov/sites/production/files/2Q19-
06/documents/utilities ria final cpp repeal and ace 2019-06.pdf

U.S. EPA. (2020). Benefit and Cost Analysis for Revisions to the Effluent Limitations Guidelines
and Standards for the Steam Electric Power Generating Point Source Category. (EPA-
821-R-20-003). Washington DC: U.S. Environmental Protection Agency.
https://www.epa.gov/sites/default/files/202Q-

08/documents/steam electric elg 2020 final reconsideration rule benefit and cost an
alvsis.pdf

U.S. EPA. (2021). Regulatory Impact Analysis for the Final Revised Cross-State Air Pollution
Rule (CSAPR) Update for the 2008 Ozone NAAQS. (EPA-452/R-21-002). Research
Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Health and Environmental Impact Division.
https://www.epa. gov/ sites/default/files/2021 -
03/documents/revised csapr update ria final.pdf

U.S. EPA. (2023). Regulatory Impact Analysis for the Final Federal Good Neighbor Plan

Addressing Regional Ozone Transport for the 2015 Ozone National Ambient Air Quality
Standards. (EPA-452/R-23-001). Research Triangle Park, NC: U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Health and
Environmental Impact Division, https://www.epa.gov/system/files/documents/2023-
03/SAN%208670%20Federal%20Good%20Neighbor%20Plan%2020230315%20RIA Fi
nal.pdf

3-43


-------
4 BENEFITS ANALYSIS

4.1 Introduction

The final rules are expected to reduce emissions of carbon dioxide (CO2), nitrogen oxides
(NOx), fine particulate matter (PM2.5), sulfur dioxide (SO2), and hazardous air pollutants (HAP)
including mercury (Hg) nationally over the period of the analysis. While there are national
reductions in NOx, SO2, and PM2.5 emissions that generally leads to improved air quality, the
emissions changes for these pollutants are heterogeneous in space and time leading to
heterogeneous impacts for ozone and to a lesser extent for PM2.5 concentrations. This section
reports the estimated monetized climate and health benefits associated with emission changes for
each of the three illustrative scenarios described in prior sections and discusses other
unquantified benefits.

This section describes the methods used to estimate the climate benefits of GHG
emissions reductions expected from the final rules using estimates of the social cost of
greenhouse gases (SC-CO2) that reflect recent advances in the scientific literature on climate
change and its economic impacts and incorporate recommendations made by the National
Academies of Science, Engineering, and Medicine (National Academies, 2017). The SC-CO2 is
the monetary value of the net harm to society associated with a marginal increase in CO2
emissions in a given year, or the benefit of avoiding that increase. In principle, SC-CO2 includes
the value of all climate change impacts (both negative and positive), including (but not limited
to) changes in net agricultural productivity, human health effects, property damage from
increased flood risk and natural disasters, disruption of energy systems, risk of conflict,
environmental migration, and the value of ecosystem services. The SC-CO2, therefore, reflects
the societal value of reducing emissions of CO2 by one metric ton and is the theoretically
appropriate value to use in conducting benefit-cost analyses of policies that affect CO2
emissions.

This section also describes the methods used to estimate the benefits to human health of
the changes in concentrations of ozone and PM2.5 from EGUs. This analysis uses methodology
for determining air quality changes that has been used in the RIAs from multiple previous
proposed and final rules (U.S. EPA, 2019b, 2020a, 2020b, 2021, 2022c). The approach involves
two major steps: (1) developing spatial fields of air quality across the U.S. for baseline and three

4-1


-------
illustrative scenarios for 2028, 2030, 2035, 2040 and 2045 using nationwide photochemical
modeling and related analyses; and (2) using these spatial fields in BenMAP-CE to quantify the
benefits under each scenario and each year as compared to the baseline in that year. Health
benefit analyses were also run for each year between 2028 and 2047, using the model surfaces
for 2028, 2030, 2035, 2040 and 2045 as described in Section 4.3.1, but accounting for the change
in population size in each year, income growth and baseline mortality incidence rates at five-year
increments. Specifically, the analysis quantifies health benefits resulting from changes in ozone
and PM2.5 concentrations in 2028, 2030, 2035, 2040 and 2045 for each of the three illustrative
scenarios (i.e., final rules, alternative 1 scenario, and alternative 2 scenario). The methods for
quantifying the number and value of air pollution-attributable premature deaths and illnesses are
described in the Technical Support Document (TSD) titled Estimating PM2.5- and Ozone-
Attributable Health Benefits (U.S. EPA, 2023b) and further referred to as the Health Benefits
TSD in this RIA.

Though the final rules are likely to also yield positive benefits associated with reducing
pollutants other than CO2, ozone, and PM2.5, time, resource, and data limitations prevented us
from characterizing the value of those reductions. Specifically, in this RIA, EPA does not
monetize health benefits of reducing direct exposure to NO2, SO2 or hazardous air pollutants nor
ecosystem effects and visibility impairment associated with changes in air quality. In addition,
this RIA does not include monetized benefits from reductions in pollutants in other media, such
as water effluents. We qualitatively discuss these unquantified benefits in this section. This RIA
also does not quantify impacts of the CCS and hydrogen compliance technologies beyond the
direct compliance cost and emissions impacts reflected in Section 3, which is discussed in more
detail in Sections 3.7 and 6.

4.2 Climate Benefits

The EPA estimates the climate benefits of CO2 emissions reductions expected from the
final rules using estimates of the social cost of carbon (SC-CO2) that reflect recent advances in
the scientific literature on climate change and its economic impacts and incorporate
recommendations made by the National Academies of Science, Engineering, and Medicine
(National Academies, 2017). The EPA published and used these estimates in the RIA for the
December 2023 Final Oil and Gas NSPS/EG Rulemaking, "Standards of Performance for New,

4-2


-------
Reconstructed, and Modified Sources and Emissions Guidelines for Existing Sources: Oil and
Natural Gas Sector Climate Review" (U.S. EPA, 2023e). The EPA solicited public comment on
the methodology and use of these estimates in the RIA for the agency's December 2022 Oil and
Gas NSPS/EG Supplemental Proposal" and has conducted an external peer review of these
estimates, as described further below.

The SC-CO2 is the monetary value of the net harm to society associated with a marginal
increase in CO2 emissions in a given year, or the benefit of avoiding that increase. In principle,
SC-CO2 includes the value of all climate change impacts (both negative and positive), including
(but not limited to) changes in net agricultural productivity, human health effects, property
damage from increased flood risk and natural disasters, disruption of energy systems, risk of
conflict, environmental migration, and the value of ecosystem services. The SC-CO2, therefore,
reflects the societal value of reducing emissions of CO2 by one metric ton and is the theoretically
appropriate value to use in conducting benefit-cost analyses of policies that affect CO2
emissions. In practice, data and modeling limitations restrain the ability of SC-CO2 estimates to
include all physical, ecological, and economic impacts of climate change, implicitly assigning a
value of zero to the omitted climate damages. The estimates are, therefore, a partial accounting
of climate change impacts and likely underestimate the marginal benefits of abatement.

Since 2008, the EPA has used estimates of the social cost of various greenhouse gases
(i.e., SC-CO2, SC-CH4, and SC-N2O), collectively referred to as the "social cost of greenhouse
gases" (SC-GHG), in analyses of actions that affect GHG emissions. The values used by the
EPA from 2009 to 2016, and since 2021 - including in the proposal for this rulemaking - have
been consistent with those developed and recommended by the Interagency Working Group
(IWG) on the SC-GHG; and the values used from 2017 to 2020 were consistent with those
required by E.O. 13783, which disbanded the IWG. During 2015-2017, the National Academies
conducted a comprehensive review of the SC-CO2 and issued a final report in 2017
recommending specific criteria for future updates to the SC-CO2 estimates, a modeling
framework to satisfy the specified criteria, and both near-term updates and longer-term research
needs pertaining to various components of the estimation process (National Academies, 2017).

99 See https://www.epa.gov/environmental-economics/scghg for a copy of the final report and other related
materials.

4-3


-------
The IWG was reconstituted in 2021 and E.O. 13990 directed it to develop a comprehensive
update of its SC-GHG estimates, recommendations regarding areas of decision-making to which
SC-GHG should be applied, and a standardized review and updating process to ensure that the
recommended estimates continue to be based on the best available economics and science going
forward.

The EPA is a member of the IWG and is participating in the IWG's work under E.O.
13990. As noted in previous EPA RIAs, including in the proposal RIA for this rulemaking, while
that process continues, the EPA is continuously reviewing developments in the scientific
literature on the SC-GHG, including more robust methodologies for estimating damages from
emissions, and looking for opportunities to further improve SC-GHG estimation 10°. In the
December 2022 Oil and Gas NSPS/EG Supplemental Proposal RIA, the Agency included a
sensitivity analysis of the climate benefits of the Supplemental Proposal using a new set of SC-
GHG estimates that incorporates recent research addressing recommendations of the National
Academies (2017) in addition to using the interim SC-GHG estimates presented in the Technical
Support Document: Social Cost of Carbon, Methane, and Nitrous Oxide Interim Estimates under
Executive Order 13990 (2021) that the IWG recommended for use until updated estimates that
address the National Academies' recommendations are available.

The EPA solicited public comment on the sensitivity analysis and the accompanying draft
technical report, External Review Draft of Report on the Social Cost of Greenhouse Gases:
Estimates Incorporating Recent Scientific Advances, which explains the methodology underlying
the new set of estimates, in the December 2022 Supplemental Oil and Gas Proposal. The
response to comments document can be found in the docket for that action.

To ensure that the methodological updates adopted in the technical report are consistent
with economic theory and reflect the latest science, the EPA also initiated an external peer
review panel to conduct a high-quality review of the technical report, completed in May 2023.
See 88 FR at 26075/2 noting this peer review process. The peer reviewers commended the
agency on its development of the draft update, calling it a much-needed improvement in
estimating the SC-GHG and a significant step towards addressing the National Academies'

100 EPA strives to base its analyses on the best available science and economics, consistent with its responsibilities,
for example, under the Information Quality Act.

4-4


-------
recommendations with defensible modeling choices based on current science. The peer reviewers
provided numerous recommendations for refining the presentation and for future modeling
improvements, especially with respect to climate change impacts and associated damages that
are not currently included in the analysis. Additional discussion of omitted impacts and other
updates have been incorporated in the technical report to address peer reviewer
recommendations. Complete information about the external peer review, including the peer
reviewer selection process, the final report with individual recommendations from peer
reviewers, and the EPA's response to each recommendation is available on EPA's website.101

The remainder of this section provides an overview of the methodological updates
incorporated into the SC-GHG estimates used in this final RIA. A more detailed explanation of
each input and the modeling process is provided in the final technical report, EPA Report on the
Social Cost of Greenhouse Gases: Estimates Incorporating Recent Scientific Advances.

Appendix A shows the benefits of the final rules using the interim SC-GHG (IWG, 2021)
estimates presented in the proposal RIA for comparison purposes.

The steps necessary to estimate the SC-GHG with a climate change integrated assessment
model (IAM) can generally be grouped into four modules: socioeconomics and emissions,
climate, damages, and discounting. The emissions trajectories from the socioeconomic module
are used to project future temperatures in the climate module. The damage module then
translates the temperature and other climate endpoints (along with the projections of
socioeconomic variables) into physical impacts and associated monetized economic damages,
where the damages are calculated as the amount of money the individuals experiencing the
climate change impacts would be willing to pay to avoid them. To calculate the marginal effect
of emissions, i.e., the SC-GHG in year t, the entire model is run twice - first as a baseline and
second with an additional pulse of emissions in year t. After recalculating the temperature effects
and damages expected in all years beyond t resulting from the adjusted path of emissions, the
losses are discounted to a present value in the discounting module. Many sources of uncertainty
in the estimation process are incorporated using Monte Carlo techniques by taking draws from
probability distributions that reflect the uncertainty in parameters.

101 https://www.epa.gov/environmental-economics/scghg-tsd-peer-review.

4-5


-------
The SC-GHG estimates used by the EPA and many other federal agencies since 2009
have relied on an ensemble of three widely used IAMs: Dynamic Integrated Climate and
Economy (DICE) (Nordhaus, 2010); Climate Framework for Uncertainty, Negotiation, and
Distribution (FUND) (Anthoff and Tol, 2013a, 2013b); and Policy Analysis of the Greenhouse
Gas Effect (PAGE) (Hope, 2013). In 2010, the IWG harmonized key inputs across the IAMs, but
all other model features were left unchanged, relying on the model developers' best estimates
and judgments. That is, the representation of climate dynamics and damage functions included in
the default version of each IAM as used in the published literature was retained.

The SC-GHG estimates in this RIA no longer rely on the three IAMs (i.e., DICE, FUND,
and PAGE) used in previous SC-GHG estimates. As explained previously, EPA uses a modular
approach to estimate the SC-GHG, consistent with the National Academies' (2017) near-term
recommendations. That is, the methodology underlying each component, or module, of the SC-
GHG estimation process is developed by drawing on the latest research and expertise from the
scientific disciplines relevant to that component. Under this approach, each step in the SC-GHG
estimation improves consistency with the current state of scientific knowledge, enhances
transparency, and allows for more explicit representation of uncertainty.

The socioeconomic and emissions module relies on a new set of probabilistic projections
for population, income, and GHG emissions developed under the Resources for the Future (RFF)
Social Cost of Carbon Initiative (Rennert, Prest, et al., 2022). These socioeconomic projections
(hereafter collectively referred to as the RFF-SPs) are an internally consistent set of probabilistic
projections of population, GDP, and GHG emissions (CO2, CH4, and N2O) to 2300. Based on a
review of available sources of long-run projections necessary for damage calculations, the RFF-
SPs stand out as being most consistent with the National Academies' recommendations.
Consistent with the National Academies' recommendation, the RFF-SPs were developed using a
mix of statistical and expert elicitation techniques to capture uncertainty in a single probabilistic
approach, taking into account the likelihood of future emissions mitigation policies and
technological developments, and provide the level of disaggregation necessary for damage
calculations. Unlike other sources of projections, they provide inputs for estimation out to 2300
without further extrapolation assumptions. Conditional on the modeling conducted for the SC-
GHG estimates, this time horizon is far enough in the future to capture the majority of
discounted climate damages. Including damages beyond 2300 would increase the estimates of

4-6


-------
the SC-GHG. As discussed in U.S. EPA (2023d), the use of the RFF-SPs allows for capturing
economic growth uncertainty within the discounting module.

The climate module relies on the Finite Amplitude Impulse Response (FaIR) model,
(IPCC, 2021b; Millar et al., 2017; C. J. Smith et al., 2018) a widely used Earth system model
which captures the relationships between GHG emissions, atmospheric GHG concentrations, and
global mean surface temperature. The FaIR model was originally developed by Richard Millar,
Zeb Nicholls, and Myles Allen at Oxford University, as a modification of the approach used in
IPCC AR5 to assess the GWP and GTP (Global Temperature Potential) of different gases. It is
open source, widely used (e.g., IPCC (2018, 2021a) and was highlighted by the National
Academies (2017) as a model that satisfies their recommendations for a near-term update of the
climate module in SC-GHG estimation. Specifically, it translates GHG emissions into mean
surface temperature response and represents the current understanding of the climate and GHG
cycle systems and associated uncertainties within a probabilistic framework. The SC-GHG
estimates used in this RIA rely on FaIR version 1.6.2 as used by the IPCC (2021a, 2021b). It
provides, with high confidence, an accurate representation of the latest scientific consensus on
the relationship between global emissions and global mean surface temperature and offers a code
base that is fully transparent and available online. The uncertainty capabilities in FaIR 1.6.2 have
been calibrated to the most recent assessment of the IPCC (which importantly narrowed the
range of likely climate sensitivities relative to prior assessments). See U.S. EPA (2022f) for more
details.

The socioeconomic projections and outputs of the climate module are inputs into the
damage module to estimate monetized future damages from climate change102. The National
Academies' recommendations for the damage module, scientific literature on climate damages,
updates to models that have been developed since 2010, as well as the public comments received
on individual EPA rulemakings and the IWG's February 2021 TSD, have all helped to identify
available sources of improved damage functions. The IWG (e.g., IWG (2010, 2016a, 2021)), the

102 In addition to temperature change, two of the three damage modules used in the SC-GHG estimation require
global mean sea level (GMSL) projections as an input to estimate coastal damages. Those two damage modules
use different models for generating estimates of GMSL. Both are based off reduced complexity models that can
use the FaIR temperature outputs as inputs to the model and generate projections of GMSL accounting for the
contributions of thermal expansion and glacial and ice sheet melting based on recent scientific research. Absent
clear evidence on a preferred model, the SC-GHG estimates presented in this RIA retain both methods used by
the damage module developers. See U.S. EPA (2023d) for more details.

4-7


-------
National Academies (2017), comprehensive studies (e.g., Rose et al. (2014)), and public
comments have all recognized that the damages functions underlying the IWG SC-GHG
estimates used since 2013 (taken from DICE 2010 (Nordhaus, 2010); FUND 3.8 (Anthoff and
Tol, 2013b); and PAGE 2009 (Hope, 2013)) do not include all the important physical, ecological,
and economic impacts of climate change. The climate change literature and the science
underlying the economic damage functions have evolved, and DICE 2010, FUND 3.8, and
PAGE 2009 now lag behind the most recent research.

The challenges involved with updating damage functions have been widely recognized.
Functional forms and calibrations are constrained by the available literature and need to
extrapolate beyond warming levels or locations studied in that literature. Research and public
resources focused on understanding how these physical changes translate into economic impacts
have been significantly less than the resources focused on modeling and improving our
understanding of climate system dynamics and the physical impacts from climate change
(Auffhammer, 2018). Even so, there has been a large increase in research on climate impacts and
damages in the time since DICE 2010, FUND 3.8, and PAGE 2009 were published. Along with
this growth, there continues to be wide variation in methodologies and scope of studies, such that
care is required when synthesizing the current understanding of impacts or damages. Based on a
review of available studies and approaches to damage function estimation, the EPA uses three
separate damage functions to form the damage module. They are:

A subnational-scale, sectoral damage function (based on the Data-driven Spatial Climate
Impact Model (DSCIM) developed by the Climate Impact Lab (Carleton, 2022; CIL, 2023; Rode
et al., 2021), a country-scale, sectoral damage function (based on the Greenhouse Gas Impact
Value Estimator (GIVE) model developed under RFF's Social Cost of Carbon Initiative
(Rennert, Prest, et al., 2022), and a meta-analysis-based damage function (based on Howard and
Sterner (2017)). The damage functions in DSCIM and GIVE represent substantial improvements
relative to the damage functions underlying the SC-GHG estimates used by the EPA to date and
reflect the forefront of scientific understanding about how temperature change and SLR lead to
monetized net (market and nonmarket) damages for several categories of climate impacts. The
models' spatially explicit and impact-specific modeling of relevant processes allow for improved
understanding and transparency about mechanisms through which climate impacts are occurring
and how each damage component contributes to the overall results, consistent with the National

4-8


-------
Academies' recommendations. DSCIM addresses common criticisms related to the damage
functions underlying current SC-GHG estimates (e.g., (Pindyck, 2017)) by developing multi-
sector, empirically grounded damage functions. The damage functions in the GIVE model offer a
direct implementation of the National Academies' near-term recommendation to develop
updated sectoral damage functions that are based on recently published work and reflective of
the current state of knowledge about damages in each sector. Specifically, the National
Academies noted that "[t]he literature on agriculture, mortality, coastal damages, and energy
demand provide immediate opportunities to update the [models]" (National Academies, 2017),
which are the four damage categories currently in GIVE. A limitation of both models is that the
sectoral coverage is still limited, and even the categories that are represented are incomplete.
Neither DSCIM nor GIVE yet accommodate estimation of several categories of temperature
driven climate impacts (e.g., morbidity, conflict, migration, biodiversity loss) and only represent
a limited subset of damages from changes in precipitation. For example, while precipitation is
considered in the agriculture sectors in both DSCIM and GIVE, neither model takes into account
impacts of flooding, changes in rainfall from tropical storms, and other precipitation related
impacts. As another example, the coastal damage estimates in both models do not fully reflect
the consequences of SLR-driven salt-water intrusion and erosion, or SLR damages to coastal
tourism and recreation. Other missing elements are damages that result from other physical
impacts (e.g., ocean acidification, non-temperature-related mortality such as diarrheal disease
and malaria) and the many feedbacks and interactions across sectors and regions that can lead to
additional damages103. See U.S. EPA (2023c) for more discussion of omitted damage categories
and other modeling limitations. DSCIM and GIVE do account for the most commonly cited
benefits associated with CO2 emissions and climate change - CO2 crop fertilization and declines
in cold related mortality. As such, while the GIVE- and DSCIM-based results provide state-of-
the-science assessments of key climate change impacts, they remain partial estimates of future
climate damages resulting from incremental changes in C02, CH4, and N20104.

103	The one exception is that the agricultural damage function in DSCIM and GIVE reflects the ways that trade can
help mitigate damages arising from crop yield impacts.

104	One advantage of the modular approach used by these models is that future research on new or alternative
damage functions can be incorporated in a relatively straightforward way. DSCIM and GIVE developers have
work underway on other impact categories that may be ready for consideration in future updates (e.g., morbidity
and biodiversity loss).

4-9


-------
Finally, given the still relatively narrow sectoral scope of the recently developed DSCIM
and GIVE models, the damage module includes a third damage function that reflects a synthesis
of the state of knowledge in other published climate damages literature. Studies that employ
meta-analytic techniques105 offer a tractable and straightforward way to combine the results of
multiple studies into a single damage function that represents the body of evidence on climate
damages that pre-date CIL and RFF's research initiatives. The first use of meta-analysis to
combine multiple climate damage studies was done by Tol (2009) and included 14 studies. The
studies in Tol (2009) served as the basis for the global damage function in DICE starting in
version 2013R (Nordhaus, 2014) The damage function in the most recent published version of
DICE, DICE 2016, is from an updated meta-analysis based on a review of existing damage
studies and included 26 studies published over 1994-2013 (Nordhaus and Moffat, 2017). Howard
and Sterner (2017) provide a more recent published peer-reviewed meta-analysis of existing
damage studies (published through 2016) and account for additional features of the underlying
studies. This study address differences in measurement across studies by adjusting estimates
such that the data are relative to the same base period. They also eliminate double counting by
removing duplicative estimates. Howard and Sterner's final sample is drawn from 20 studies that
were published through 2015. Howard and Sterner (2017) present results under several
specifications and shows that the estimates are somewhat sensitive to defensible alternative
modeling choices. As discussed in detail in U.S. EPA (2023d), the damage module underlying
the SC-GHG estimates in this RIA includes the damage function specification (that excludes
duplicate studies) from Howard and Sterner (2017) that leads to the lowest SC-GHG estimates,
all else equal.

The discounting module discounts the stream of future net climate damages to its present
value in the year when the additional unit of emissions was released. Given the long-time
horizon over which the damages are expected to occur, the discount rate has a large influence on
the present value of future damages. Consistent with the findings of (National Academies, 2017),
the economic literature, OMB Circular A-4's guidance for regulatory analysis, and IWG
recommendations to date (IWG, 2010, 2013, 2016a, 2016b, 2021; Technical Support Document:

105 Meta-analysis is a statistical method of pooling data and/or results from a set of comparable studies of a problem.
Pooling in this way provides a larger sample size for evaluation and allows for a stronger conclusion than can be
provided by any single study. Meta-analysis yields a quantitative summary of the combined results and current
state of the literature.

4-10


-------
Social Cost of Carbon for Regulatory Impact Analysis under Executive Order 12866, 2010), the
EPA continues to conclude that the consumption rate of interest is the theoretically appropriate
discount rate to discount the future benefits of reducing GHG emissions and that discount rate
uncertainty should be accounted for in selecting future discount rates in this intergenerational
context. OMB 's Circular A-4 points out that "the analytically preferred method of handling
temporal differences between benefits and costs is to adjust all the benefits and costs to reflect
their value in equivalent units of consumption and to discount them at the rate consumers and
savers would normally use in discounting future consumption benefits" (OMB, 2003).106 The
damage module described above calculates future net damages in terms of reduced consumption
(or monetary consumption equivalents), and so an application of this guidance is to use the
consumption discount rate to calculate the SC-GHG. Thus, EPA concludes that the use of the
social rate of return on capital (7 percent under the 2003 OMB Circular A-4 guidance), which
does not reflect the consumption rate, to discount damages estimated in terms of reduced
consumption would inappropriately underestimate the impacts of climate change for the
purposes of estimating the SC-GHG107.

For the SC-GHG estimates used in this RIA, EPA relies on a dynamic discounting
approach that more fully captures the role of uncertainty in the discount rate in a manner
consistent with the other modules. Based on a review of the literature and data on consumption
discount rates, the public comments received on individual EPA rulemakings, and the February
2021 TSD (IWG, 2021), and the National Academies (2017) recommendations for updating the
discounting module, the SC-GHG estimates rely on discount rates that reflect more recent data
on the consumption interest rate and uncertainty in future rates. Specifically, rather than using a
constant discount rate, the evolution of the discount rate over time is defined following the latest
empirical evidence on interest rate uncertainty and using a framework originally developed by
Ramsey (1928) that connects economic growth and interest rates. The Ramsey approach
explicitly reflects (1) preferences for utility in one period relative to utility in a later period and
(2) the value of additional consumption as income changes. The dynamic discount rates used to

106	Similarly, OMB's Circular A-4 (2023) points out that "The analytically preferred method of handling temporal
differences between benefits and costs is to adjust all the benefits and costs to reflect their value in equivalent
units of consumption before discounting them".

107	See also the discussion of the inappropriateness of discounting consumption-equivalent measures of benefits
and costs using a rate of return on capital in Circular A-4 (OMB, 2003).

4-11


-------
develop the SC-GHG estimates applied in this RIA have been calibrated following the Newell et
al. (2022) approach, as applied in Rennert, Errickson, et al. (2022); Rennert, Prest, et al. (2022).
This approach uses the discounting formula Ramsey (1928) in which the parameters are
calibrated such that (1) the decline in the certainty-equivalent discount rate matches the latest
empirical evidence on interest rate uncertainty estimated by Bauer and Rudebusch (2020, 2023)
and (2) the average of the certainty-equivalent discount rate over the first decade matches a near-
term consumption rate of interest. Uncertainty in the starting rate is addressed by using three
near-term target rates (1.5, 2.0, and 2.5 percent) based on multiple lines of evidence on observed
market interest rates.

The resulting dynamic discount rate provides a notable improvement over the constant
discount rate framework used for SC-GHG estimation in previous EPA RIAs. Specifically, it
provides internal consistency within the modeling and a more complete accounting of
uncertainty consistent with economic theory (Arrow et al., 2013; Cropper et al., 2014) and the
National Academies (2017) recommendation to employ a more structural, Ramsey-like approach
to discounting that explicitly recognizes the relationship between economic growth and
discounting uncertainty. This approach is also consistent with the National Academies (2017)
recommendation to use three sets of Ramsey parameters that reflect a range of near-term
certainty-equivalent discount rates and are consistent with theory and empirical evidence on
consumption rate uncertainty. Finally, the value of aversion to risk associated with net damages
from GHG emissions is explicitly incorporated into the modeling framework following the
economic literature. See U.S. EPA (2022f) for a more detailed discussion of the entire
discounting module and methodology used to value risk aversion in the SC-GHG estimates.

Taken together, the methodologies adopted in this SC-GHG estimation process allow for
a more holistic treatment of uncertainty than past estimates used by the EPA. The updates
incorporate a quantitative consideration of uncertainty into all modules and use a Monte Carlo
approach that captures the compounding uncertainties across modules. The estimation process
generates nine separate distributions of discounted marginal damages per metric ton - the
product of using three damage modules and three near-term target discount rates - for each gas
in each emissions year. These distributions have long right tails reflecting the extensive evidence
in the scientific and economic literature that shows the potential for lower-probability but higher-
impact outcomes from climate change, which would be particularly harmful to society. The

4-12


-------
uncertainty grows over the modeled time horizon. Therefore, under cases with a lower near-term
target discount rate - that give relatively more weight to impacts in the future - the distribution
of results is wider. To produce a range of estimates that reflects the uncertainty in the estimation
exercise while also providing a manageable number of estimates for policy analysis, the EPA
combines the multiple lines of evidence on damage modules by averaging the results across the
three damage module specifications. The full results generated from the updated methodology
for methane and other greenhouse gases (SC-CO2, SC-CH4, and SC-N2O) for emissions years
2020 through 2080 are provided in U.S. EPA (2023d).

Table 4-1 summarizes the resulting averaged certainty-equivalent SC-CO2 estimates
under each near-term discount rate that are used to estimate the climate benefits of the CO2
emission reductions expected from the final rules. These estimates are reported in 2019 dollars
but are otherwise identical to those presented in U.S. EPA (2023e). The SC-CO2 increases over
time within the models — i.e., the societal harm from one metric ton emitted in 2030 is higher
than the harm caused by one metric ton emitted in 2028 — because future emissions produce
larger incremental damages as physical and economic systems become more stressed in response
to greater climatic change, and because GDP is growing over time and many damage categories
are modeled as proportional to GDP.

4-13


-------
Table 4-1 Estimates of the Social Cost of CO2 Values, 2028-2047 (2019 dollars per
Metric Ton CO2) a	

Emission Year



Near-term Ramsey Discount Rate



2.5%

2%

1.5%

2028

140

220

370

2029

140

220

380

2030

140

230

380

2031

150

230

380

2032

150

230

390

2033

150

240

390

2034

150

240

400

2035

160

240

400

2036

160

250

410

2037

160

250

410

2038

160

260

420

2039

170

260

420

2040

170

260

430

2041

170

270

430

2042

180

270

440

2043

180

280

440

2044

180

280

450

2045

190

280

450

2046

190

290

460

2047

190

290

460

3 Source: U.S. EPA (2023e).

Note: These SC-CO2 values are identical to those reported in the technical report (U.S. EPA, 2023e) adjusted to
2019 dollars using the annual GDP Implicit Price Deflator values in the U.S. Bureau of Economic Analysis' (BEA)
NIPA Table 1.1.9 (U.S. BEA, 2022) rounded to 2 significant figures. The values are stated in $/metric ton CO2 and
vary depending on the year of CO2 emissions. This table displays the values rounded to two significant figures. The
annual unrounded values used in the calculations in this RIA are available in Appendix A.4 of U.S. EPA (2023e)
and at: www.epa.gov/environmental-economics/scghg.

The methodological updates described above represent a major step forward in bringing
SC-GHG estimation closer to the frontier of climate science and economics and address many of
the National Academies' (2017) near-term recommendations. Nevertheless, the resulting SC-
CO2 estimates presented in Table 4-1, still have several limitations, as would be expected for
any modeling exercise that covers such a broad scope of scientific and economic issues across a
complex global landscape. There are still many categories of climate impacts and associated
damages that are only partially or not reflected yet in these estimates and sources of uncertainty
that have not been fully characterized due to data and modeling limitations. For example, the
modeling omits most of the consequences of changes in precipitation, damages from extreme

4-14


-------
weather events, the potential for nongradual damages from passing critical thresholds (e.g.,
tipping elements) in natural or socioeconomic systems, and non-climate mediated effects of
GHG emissions. Importantly, the updated SC-GHG methodology does not yet reflect interactions
and feedback effects within, and across, Earth and human systems. For example, it does not
explicitly reflect potential interactions among damage categories, such as those stemming from
the interdependencies of energy, water, and land use. These, and other, interactions and
feedbacks were highlighted by the National Academies as an important area of future research
for longer-term enhancements in the SC-GHG estimation framework.

Table 4-2 presents the estimated reductions in CO2 emissions from the final rules, and
Table 4-3 presents the associated estimated annual, undiscounted climate benefits using the SC-
CO2 estimates presented in Table 4-1 for the stream of years beginning in 2028 through 2047.
Also shown in Table 4-3 are the present value (PV) of monetized climate benefits discounted
back to 2024 and equivalent annualized values (AV) associated with each of the three SC-CO2
values. In this analysis, to calculate the present and annualized values of climate benefits, EPA
uses the same discount rate as the near-term target Ramsey rate used to discount the climate
benefits from future CO2 reductions. To calculate the present and annualized values of climate
benefits in Table 4-3, EPA uses the same discount rate as the near-term target Ramsey rate used
to discount the climate benefits from future CO2 reductions.108 That is, future climate benefits
estimated with the SC-CO2 at the near-term 2.5, 2 , and 1.5 percent Ramsey rate are discounted
to the base year of the analysis using a constant 2.5, 2, and 1.5 percent rate, respectively.

108 As discussed in U.S. EPA (2023d) the error associated with using a constant discount rate rather than the
certainty-equivalent rate path to calculate the present value of a future stream of monetized climate benefits is
small for analyses with moderate time frames (e.g., 30 years or less). EPA (2023d) also provides an illustration
of the amount that climate benefits from reductions in future emissions will be underestimated by using a
constant discount rate relative to the more complicated certainty-equivalent rate path.

4-15


-------
Table 4-2 Annual CO2 Emissions Reductions (million metric tons) for the Illustrative
Scenarios from 2028 through 2047	

Million Metric Tons of CO2

Emission Year

Final Rules

Alternative 1

Alternative 2

2028

38

36

32

2029

38

36

32

2030

50

48

27

2031

50

48

27

2032

123

124

122

2033

123

124

122

2034

123

124

122

2035

123

124

122

2036

123

124

122

2037

123

124

122

2038

54

53

53

2039

54

53

53

2040

54

53

53

2041

54

53

53

2042

42

40

40

2043

42

40

40

2044

42

40

40

2045

42

40

40

2046

42

40

40

2047

42

40

40

Total

1,382

1,365

1,303

4-16


-------
Table 4-3 Estimated Climate Benefits of Reduced CO2 Emissions from the Illustrative
Scenarios, 2028 to 2047 (billions of 2019 dollars)	

Final Rules

Alternative 1

Alternative 2





Near-term Ramsey
Discount Rate



Near-term Ramsey
Discount Rate

Near-term Ramsey
Discount Rate



2.5%

2.0%

1.5%

2.5%

2.0%

1.5%

2.5%

2.0%

1.5%

2028

5.2

8.4

14

4.9

7.9

13

4.4

7.1

12

2029

5.3

8.5

14

5

8

14

4.5

7.2

12

2030

7.1

11

19

6.8

11

18

3.9

6.2

10

2031

7.2

12

19

7

11

18

4

6.3

11

2032

18

29

48

18

29

48

18

28

47

2033

19

29

48

19

29

49

18

29

48

2034

19

30

49

19

30

49

19

29

48

2035

19

30

50

19

30

50

19

30

49

2036

20

31

50

20

31

50

19

30

49

2037

20

31

51

20

31

51

20

31

50

2038

00
00

14

22

00
00

14

22

8.7

14

22

2039

9

14

23

9

14

22

8.9

14

22

2040

9.1

14

23

9.1

14

23

9

14

23

2041

9.3

14

23

9.3

14

23

9.2

14

23

2042

7.4

11

18

7.1

11

17

7.1

11

18

2043

7.5

12

18

7.2

11

18

7.2

11

18

2044

7.7

12

19

7.4

11

18

7.4

11

18

2045

7.8

12

19

7.5

11

18

7.5

11

18

2046

8

12

19

7.6

12

18

7.6

12

18

2047

8.1

12

19

7.7

12

19

7.8

12

19

PV

160

270

470

160

270

460

150

250

440

EAV

9

14

23

9

14

23

8.6

13

22

Note: Values have been rounded to two significant figures.

Unlike many environmental problems where the causes and impacts are distributed more
locally, GHG emissions are a global externality making climate change a true global challenge.
GHG emissions contribute to damages around the world regardless of where they are emitted.
Because of the distinctive global nature of climate change, in the RIA for these final rules the
EPA centers attention on a global measure of climate benefits from GHG reductions. Consistent
with all IWG recommended SC-GHG estimates to date, the SC-CO2 values presented Table 4-1
provide a global measure of monetized damages from CO2 and Table 4-3 presents the monetized
global climate benefits of the CO2 emission reductions expected from the final rules. This

4-17


-------
approach is the same as that taken in EPA regulatory analyses from 2009 through 2016 and since
2021. It is also consistent with guidance in OMB Circular A-4 (OMB 2003, 2023) that
recommends reporting of important international effects.109 EPA also notes that EPA's cost
estimates in RIAs, including the cost estimates contained in this RIA, regularly do not
differentiate between the share of compliance costs expected to accrue to U.S. firms versus
foreign interests, such as to foreign investors in regulated entities.110 A global perspective on
climate effects is therefore consistent with the approach EPA takes on costs. There are many
reasons, as summarized in this section - and as articulated by OMB and in IWG assessments
(2010, 2013, 2016a, 2016b, 2021), the 2015 Response to Comments (IWG, 2015), and in detail
in U.S. EPA (2023 c), and in Appendix A of the Response to Comments document for the
December 2023 Final Oil and Gas NSPS/EG Rulemaking - why the EPA focuses on the global
value of climate change impacts when analyzing policies that affect GHG emissions.

International cooperation and reciprocity are essential to successfully addressing climate
change, as the global nature of greenhouse gases means that a ton of GHGs emitted in any other
country harms those in the U.S. just as much as a ton emitted within the territorial U.S.

109	The 2003 version of OMB Circular A-4 states when a regulation is likely to have international effects, "these
effects should be reported"; while OMB Circular A-4 recommends that international effects we reported
separately, the guidance also explains that "[d]ifferent regulations may call for different emphases in the
analysis, depending on the nature and complexity of the regulatory issues." (OMB, 2003).

The 2023 update to Circular A-4 states that "In certain contexts, it may be particularly appropriate to include effects
experienced by noncitizens residing abroad in your primary analysis. Such contexts include, for example, when:

•	assessing effects on noncitizens residing abroad provides a useful proxy for effects on U.S. citizens and residents

that are difficult to otherwise estimate;

•	assessing effects on noncitizens residing abroad provides a useful proxy for effects on U.S. national interests that

are not otherwise fully captured by effects experienced by particular U. S. citizens and residents (e.g., national
security interests, diplomatic interests, etc.);

•	regulating an externality on the basis of its global effects supports a cooperative international approach to the

regulation of the externality by potentially inducing other countries to follow suit or maintain existing efforts; or

•	international or domestic legal obligations require or support a global calculation of regulatory effects" (OMB

2023). Due to the global nature of the climate change problem, the OMB recommendations of appropriate
contexts for considering international effects are relevant to the CO2 emission reductions expected from the final
rule. For example, as discussed in this RIA, a global focus in evaluating the climate impacts of changes in CO2
emissions supports a cooperative international approach to GHG mitigation by potentially inducing other
countries to follow suit or maintain existing efforts, and the global SC-CO2 estimates better capture effects on
U.S. citizens and residents and U.S. national interests that are difficult to estimate and not otherwise fully
captured.

110	For example, in the RIA for the 2018 Proposed Reconsideration of the Oil and Natural Gas Sector Emission
Standards for New, Reconstructed, and Modified Sources, EPA acknowledged that some portion of regulatory
costs will likely "accru[e] to entities outside U.S. borders" through foreign ownership, employment, or
consumption (EPA 2018, p. 3-13). In general, a significant share of U.S. corporate debt and equities are foreign-
owned, including in the oil and gas industry.

4-18


-------
Assessing the benefits of U.S. GHG mitigation activities requires consideration of how those
actions may affect mitigation activities by other countries, as those international mitigation
actions will provide a benefit to U.S. citizens and residents by mitigating climate impacts that
affect U.S. citizens and residents. This is 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. The only way to achieve an efficient allocation of resources for
emissions reduction on a global basis — and so benefit the U.S. and its citizens and residents —
is for all countries to base their policies on global estimates of damages. A wide range of
scientific and economic experts have emphasized the issue of international cooperation and
reciprocity as support for assessing global damages of GHG emission in domestic policy
analysis. Using a global estimate of damages in U.S. analyses of regulatory actions allows the
U.S. to continue to actively encourage other nations, including emerging major economies, to
also assess global climate damages of their policies and to take steps to reduce emissions. For
example, many countries and international institutions have already explicitly adapted the global
SC-GHG estimates used by EPA in their domestic analyses (e.g., Canada, Israel) or developed
their own estimates of global damages (e.g., Germany), and recently, there has been renewed
interest by other countries to update their estimates since the draft release of the updated SC-
GHG estimates presented in the December 2022 Oil and Gas NSPS/EG Supplemental Proposal
RIA.111 Several recent studies have empirically examined the evidence on international GHG
mitigation reciprocity, through both policy diffusion and technology diffusion effects. See U.S.
EPA (2022f) for more discussion.

For all of these reasons, the EPA believes that a global metric is appropriate for assessing
the climate benefits of avoided GHG emissions in this final RIA. In addition, as emphasized in
the National Academies (2017) recommendations, "[i]t is important to consider what constitutes
a domestic impact in the case of a global pollutant that could have international implications that
impact the United States." The global nature of GHG pollution and its impacts means that U.S.
interests are affected by climate change impacts through a multitude of pathways and these need

111 In April 2023, the government of Canada announced the publication of an interim update to their SC-GHG
guidance, recommending SC-GHG estimates identical to EPA's updated estimates presented in the December
2022 Supplemental Proposal RIA. The Canadian interim guidance will be used across all Canadian federal
departments and agencies, with the values expected to be finalized by the end of the year.
https://www.canada.ca/en/environment-climate-change/services/climate-change/science-research-data/social-
cost-ghg.html.

4-19


-------
to be considered when evaluating the benefits of GHG mitigation to U.S. citizens and residents.
The increasing interconnectedness of global economy and populations means that impacts
occuring outside of U.S. borders can have significant impacts on U.S. interests. Examples of
affected interests include direct effects on U.S. citizens and assets located abroad, international
trade, and tourism, and spillover pathways such as economic and political destabilization and
global migration that can lead to adverse impacts on U.S. national security, public health, and
humanitarian concerns. Those impacts point to the global nature of the climate change problem
and are better captured within global measures of the social cost of greenhouse gases.

In the case of these global pollutants, for the reasons articulated in this section, the
assessment of global net damages of GHG emissions allows EPA to fully disclose and
contextualize the net climate benefits of CO2 emission reductions expected from these final rules.
The EPA disagrees with public comments received on the December 2022 Oil and Gas
NSPS/EG Supplemental Proposal that suggested that the EPA can or should use a metric focused
on benefits resulting solely from changes in climate impacts occuring within U.S. borders. The
global models used in the SC-GHG modeling described above do not lend themselves to be
disaggregated in a way that could provide sufficiently robust information about the distribution
of the rule's climate benefits to citizens and residents of particular countries, or population
groups across the globe and within the U.S. Two of the models used to inform the damage
module, the GIVE and DSCIM models, have spatial resolution that allows for some geographic
disaggregation of future climate impacts across the world. This permits the calculation of a
partial GIVE and DSCIM-based SC-GHG measuring the damages from four or five climate
impact categories projected to physically occur within the U.S., respectively, subject to caveats.
As discussed at length in U.S. EPA (2023c), these damage modules are only a partial accounting
and do not capture all of the pathways through which climate change affects public health and
welfare. For example, this modeling omits most of the consequences of changes in precipitation,
damages from extreme weather events (e.g., wildfires), the potential for nongradual damages
from passing critical thresholds (e.g., tipping elements) in natural or socioeconomic systems, and
non-climate mediated effects of GHG emissions other than CO2 fertilization (e.g., tropospheric
ozone formation due to CH4 emissions). Thus, they only cover a subset of potential climate
change impacts. Furthermore, as discussed at length in U.S. EPA (2023d), the damage modules
do not capture spillover or indirect effects whereby climate impacts in one country or region can

4-20


-------
affect the welfare of residents in other countries or regions—such as how economic and health
conditions across countries will impact U.S. business, investments, and travel abroad.

Additional modeling efforts can and have shed further light on some omitted damage
categories. For example, the Framework for Evaluating Damages and Impacts (FrEDI) is an
open-source modeling framework developed by the EPA112 to facilitate the characterization of
net annual climate change impacts in numerous impact categories within the contiguous U.S. and
monetize the associated distribution of modeled damages (Sarofim et al., 2021; U.S. EPA, 2021).
The additional impact categories included in FrEDI reflect the availability of U.S.-specific data
and research on climate change effects. As discussed in U.S. EPA (2023c) results from FrEDI
show that annual damages resulting from climate change impacts within the contiguous U.S.
(CONUS) (i.e., excluding Hawaii, Alaska, and U.S. territories) and for impact categories not
represented in GIVE and DSCIM are expected to be substantial. As discussed in U.S. EPA
(2021), results from FrEDI show that annual damages resulting from climate change impacts
within the contiguous U.S. (CONUS) (i.e., excluding Hawaii, Alaska, and U.S. territories) and
for impact categories not represented in GIVE and DSCIM are expected to be substantial. For
example, FrEDI estimates a partial SC-CO2 of $36/mtC02 for damages physically occurring
within CONUS for 2030 emissions (under a 2 percent near-term Ramsey discount rate) (Hartin,
et al. 2023), compared to a GIVE and DSCIM-based U.S.-specific SC-CO2 of $16/mtC02 and
$14/mtC02, respectively, for 2030 emissions (2019 dollars). While the FrEDI results help to
illustrate how monetized damages physically occurring within CONUS increase as more impacts
are reflected in the modeling framework, they are still subject to many of the same limitations
associated with the DSCIM and GIVE damage modules, including the omission or partial
modeling of important damage categories.113 Finally, none of these modeling efforts - GIVE,

112	The FrEDI framework and Technical Documentation have been subject to a public review comment period and
an independent external peer review, following guidance in the EPA Peer-Review Handbook for Influential
Scientific Information (ISI). Information on the FrEDI peer-review is available at the EPA Science Inventory
(EPA Science Inventory, 2021).

113	Another method that has produced estimates of the effect of climate change on U.S.-specific outcomes uses a top-
down approach to estimate aggregate damage functions. Published research using this approach include total-
economy empirical studies that econometrically estimate the relationship between GDP and a climate variable,
usually temperature. As discussed in U.S. EPA (2023d) the modeling framework used in the existing published
studies using this approach differ in important ways from the inputs underlying the SC-GHG estimates described
above (e.g., discounting, risk aversion, and scenario uncertainty). Hence, we do not consider this line of evidence
in the analysis for this RIA. Updating the framework of total-economy empirical damage functions to be
consistent with the methods described in this RIA and U.S. EPA (2023d) would require new analysis. Finally,

4-21


-------
DSCIM, and FrEDI - reflect non-climate mediated effects of GHG emissions experienced by
U.S. populations (other than CO2 fertilization effects on agriculture).

Applying the U.S.-specific partial SC-GHG estimates derived from the multiple lines of
evidence described above to the GHG emissions reduction expected under the final rules would
yield substantial benefits. For example, the present value of the climate benefits of the final rules
over 2024 - 2047 as measured by FrEDI from climate change impacts in CONUS are estimated
to be $42 billion (under a 2 percent near-term Ramsey discount rate). However, the numerous
explicitly omitted damage categories and other modeling limitations discussed above and
throughout U.S. EPA (2023c) make it likely that these estimates underestimate the benefits to
U.S. citizens and residents of the GHG reductions from the final rule; the limitations in
developing a U.S.-specific estimate that accurately captures direct and spillover effects on U.S.
citizens and residents further demonstrates that it is more appropriate to use a global measure of
climate benefits from GHG reductions. The EPA will continue to review developments in the
literature, including more robust methodologies for estimating the magnitude of the various
damages to U.S. populations from climate impacts and reciprocal international mitigation
activities, and explore ways to better inform the public of the full range of GHG impacts.

4.3 Human Health Benefits

Estimating the health benefits of reductions in ozone and PM2.5 exposure begins with
estimating the change in exposure for each individual and then estimating the change in each
individual's risks for health outcomes affected by exposure. The benefit of the reduction in each
health risk is based on the exposed individual's willingness to pay (WTP) for the risk change,
assuming that each outcome is independent of one another. The greater the magnitude of the risk
reduction from a given change in concentration, the greater the individual's WTP, all else equal.
The social benefit of the change in health risks equals the sum of the individual WTP estimates
across all of the affected individuals residing in the U.S.114 We conduct this analysis by adapting

because total-economy empirical studies estimate market impacts, they do not include any non-market impacts of
climate change (e.g., heat related mortality) and therefore are also only a partial estimate. EPA will continue to
review developments in the literature and explore ways to better inform the public of the full range of GHG
impacts.

114 This RIA also reports the change in the sum of the risk, or the change in the total incidence, of a health outcome
across the population. If the benefit per unit of risk is invariant across individuals, the total expected change in

4-22


-------
primary research—specifically, air pollution epidemiology studies and economic value studies—
from similar contexts. This approach is sometimes referred to as "benefits transfer." Below we
describe the procedure we follow for: (1) developing spatial fields of air quality for baseline and
three illustrative scenarios (2) selecting air pollution health endpoints to quantify; (3) calculating
counts of air pollution effects using a health impact function; (4) specifying the health impact
function with concentration-response parameters drawn from the epidemiological literature to
calculate the economic value of the health impacts. We estimate the quantity and economic value
of air pollution-related effects using a "damage-function." This approach quantifies counts of air
pollution-attributable cases of adverse health outcomes and assigns dollar values to those counts,
while assuming that each outcome is independent of one another.

As structured, these final rules would affect the distribution of ozone and PM2.5
concentrations in much of the U.S. This RIA estimates ozone- and PIVh.s-related health impacts
that are distinct from those reported in the RIAs for both ozone and PM NAAQS (U.S. EPA,
2012, 2015c, 2022d). The ozone and PM NAAQS RIAs illustrate, but do not predict, the benefits
and costs of strategies that States may choose to enact when implementing a revised NAAQS;
these costs and benefits are illustrative and cannot be added to the costs and benefits of policies
that prescribe specific emission control measures. This RIA estimates the benefits (and costs) of
specific emissions control measures. The benefit estimates are based on these modeled changes
in PM2.5 and summer season average ozone concentrations for each of the years 2028, 2030,
2035, 2040 and 2045.

4.3.1 Air Quality Modeling Methodology and Results

The final rules influence the level of pollutants emitted in the atmosphere that adversely
affect human health, including directly emitted PM2.5, as well as SO2 and NOx, which are both
precursors to ambient PM2.5. NOx emissions are also a precursor to ambient ground-level ozone.
EPA used air quality modeling to estimate changes in ozone and PM2.5 concentrations that may
occur as a result of the three illustrative scenarios for the final rules relative to the baseline.

the incidence of the health outcome across the population can be multiplied by the benefit per unit of risk to
estimate the social benefit of the total expected change in the incidence of the health outcome.

4-23


-------
As described in the Air Quality Modeling Appendix (Appendix B), gridded spatial fields
of ozone and PM2.5 concentrations representing the baseline and three illustrative scenarios were
derived from CAMx source apportionment modeling in combination with NOx, SO2, and
primary PM2.5 EGU emissions obtained from the outputs of the IPM runs described in Section 3
of this RIA. The air quality modeling includes all inventoried pollution sources in the contiguous
U.S., contributions from all sources other than EGUs are held constant at projected 2026 levels
in this analysis, and the only changes quantified between the baseline and three illustrative
scenarios are those associated with the projected impacts of the final rules on EGU emissions.
EPA prepared gridded spatial fields of air quality for the baseline and the three illustrative
scenarios for two health-impact metrics: annual mean PM2.5 and April through September
seasonal average 8-hour daily maximum (MDA8) ozone (AS-M03). These ozone and PM2.5
gridded spatial fields cover all locations in the contiguous U.S. and were used as inputs to
BenMAP-CE115 which, in turn, was used to quantify the benefits from the final rules.

The basic methodology for determining air quality changes is the same as that used in the
RIAs from multiple previous rules (U.S. EPA, 2019b, 2020a, 2020b, 2021, 2022c). The Air
Quality Modeling Appendix (Appendix B) provides additional details on the air quality modeling
and the methodologies EPA used to develop gridded spatial fields of summertime ozone and
annual PM2.5 concentrations.

The Air Quality Modeling Appendix also provides figures showing the geographical
distribution of air quality changes in the illustrative scenarios relative to the baseline. The spatial
fields of baseline AS-M03 and Annual Average PM2.5 and changes in the illustrative policy
scenarios relative to the baseline in 2028, 2030, 2035, 2040 and 2045 are presented in Figure B-8
through Figure B-17. The spatial patterns shown in the figures are a result of (1) of the spatial
distribution of EGU sources that are predicted to have changes in emissions and (2) of the
physical or chemical processing that the model simulates in the atmosphere. In all years, the final
rules are expected to result in reductions in ozone concentrations over many areas of the US,

115 EPA recently convened a Scientific Advisory Board (SAB) Panel to consider EPA's methods for particulate
matter and ozone air quality benefit analysis. The SAB report, issued in January 2024, endorsed the current
methods as "scientifically robust and appropriate for regulatory analyses". The report also offered
recommendations for future improvements. These include combining estimates of mortality outcomes, shifting
towards more transparent data sources, improving estimates of the cost of health injuries, improving the analysis
of uncertainty, integrating labor productivity and human capital impacts of air pollution, and improving the user
experience. EPA is working to improve its methods in response to these recommendations.

4-24


-------
although some areas may experience increases in ozone concentrations relative to forecasted
conditions without the rule. The extent of areas experiencing ozone increases varies among
snapshot years. Our comparison of air quality conditions with and without the rule suggests that
for all snapshot years the final rules will result in widespread reductions in PM2.5 concentrations.

Figure B-18 though Figure B-21 show changes in AS-M03 in 2030, 2035, 2040, and
2045 relative to 2028 baseline conditions. Figure B-22 through Figure B-25 show changes in
PM2.5 in 2030, 2035, 2040, and 2045 relative to 2028 baseline conditions. Relative to 2028
baseline conditions, these figures indicate that ozone and PM2.5 concentration will decline in
virtually all areas of the country for both baseline and final rules scenarios in each further out
snapshot year. However, some areas of the country may experience slower or faster rates of
decline in ozone and PM2.5 over time as a result of the modeled changes resulting from this rule.

4.3.2 Selecting Air Pollution Health Endpoints to Quantify

As a first step in quantifying ozone and PIVh.s-related human health impacts, the Agency
consults the Integrated Science Assessment for Ozone and Related Photochemical Oxidants
(Ozone ISA) (U.S. EPA, 2020d), the Integrated Science Assessment for Particulate Matter (PM
ISA) (U.S. EPA, 2019a), and the Supplement to the ISA for Particulate Matter (U.S. EPA,
2022f). These documents synthesize the toxicological, clinical, and epidemiological evidence to
determine whether PM is causally related to an array of adverse human health outcomes
associated with either acute (i.e., hours or days-long) or chronic (i.e., years-long) exposure; for
each outcome, the ISA reports this relationship to be causal, likely to be causal, suggestive of a
causal relationship, inadequate to infer a causal relationship or not likely to be a causal
relationship. Historically, the Agency estimates the incidence of air pollution effects for those
health endpoints that the ISA classified as either causal or likely-to-be-causal. The analysis also
accounts for recommendations from the Science Advisory Board (U.S. EPA Science Advisory
Board, 2019, 2020a). When updating each health endpoint EPA considered: (1) the extent to
which there exists a causal relationship between that pollutant and the adverse effect; (2) whether
suitable epidemiologic studies exist to support quantifying health impacts; (3) and whether
robust economic approaches are available for estimating the value of the impact of reducing
human exposure to the pollutant. Our approach for updating the endpoints and to identify
suitable epidemiologic studies, baseline incidence rates, population demographics, and valuation

4-25


-------
estimates is summarized below. The Health Benefits TSD (U.S. EPA, 2023b) fully describes the
Agency's approach for quantifying the number and value of estimated air pollution-related
impacts. In this document the reader can find the rationale for selecting health endpoints to
quantify; the demographic, health and economic data used; modeling assumptions; and our
techniques for quantifying uncertainty116.

In brief, the ISA for ozone found short-term (less than one month) exposures to ozone to
be causally related to respiratory effects, a "likely to be causal" relationship with metabolic
effects and a "suggestive of, but not sufficient to infer, a causal relationship" for central nervous
system effects, cardiovascular effects, and total mortality. The ISA reported that long-term
exposures (one month or longer) to ozone are "likely to be causal" for respiratory effects
including respiratory mortality, and a "suggestive of, but not sufficient to infer, a causal
relationship" for cardiovascular effects, reproductive effects, central nervous system effects,
metabolic effects, and total mortality. The PM ISA found short-term exposure to PM2.5 to be
causally related to cardiovascular effects and mortality, respiratory effects as likely-to-be-
causally related, and a suggestive relationship for metabolic effects and nervous system effects.
The ISA identified cardiovascular effects and total mortality as being causally related to long-
term exposure to PM2.5. A likely-to-be-causal relationship was determined between long-term
PM2.5 exposures and respiratory effects, nervous system effects, and cancer effects; and the
evidence was suggestive of a causal relationship for male and female reproduction and fertility
effects, pregnancy and birth outcomes, and metabolic effects. Table 4-4 reports the ozone and
PM2.5-related human health impacts effects we quantified and those we did not quantify in this
RIA. The list of benefit categories not quantified is not exhaustive. And, among the effects
quantified, it might not have been possible to quantify completely either the full range of human
health impacts or economic values. Section 4.4 and Table 4-4 below report other omitted health
and environmental benefits expected from the emissions and effluent changes as a result of this
final rule, such as health effects associated with NO2 and SO2, and any welfare effects such as
acidification and nutrient enrichment.

116

The analysis was completed using BenMAP-CE version 1.5.8, which is a variant of the current publicly available
version.

4-26


-------
Consistent with economic theory, the willingness-to-pay (WTP) for reductions in
exposure to environmental hazards will depend on the expected impact of those reductions on
human health and other outcomes. All else equal, WTP is expected to be higher when there is
stronger evidence of a causal relationship between exposure to the contaminant and changes in a
health outcome (McGartland et al., 2017). For example, in the case where there is no evidence of
a potential relationship the WTP would be expected to be zero and the effect should be excluded
from the analysis. Alternatively, when there is some evidence of a relationship between exposure
and the health outcome, but that evidence is insufficient to definitively conclude that there is a
causal relationship, individuals may have a positive WTP for a reduction in exposure to that
hazard (Kivi and Shogren, 2010; U.S. EPA Science Advisory Board, 2020b). Lastly, the WTP
for reductions in exposure to pollutants with strong evidence of a relationship between exposure
and effect are likely positive and larger than for endpoints where evidence is weak, all else equal.
Unfortunately, the economic literature currently lacks a settled approach for accounting for how
WTP may vary with uncertainty about causal relationships.

Given this challenge, the Agency draws its assessment of the strength of evidence on the
relationship between exposure to PM2.5 or ozone and potential health endpoints from the ISAs
that are developed for the NAAQS process as discussed above. The focus on categories
identified as having a "causal" or "likely to be causal" relationship with the pollutant of interest
is to estimate the pollutant-attributable human health benefits in which we are most confident.117
All else equal, this approach may underestimate the benefits of PM2.5 and ozone exposure
reductions as individuals may be WTP to avoid specific risks where the evidence is insufficient
to conclude they are "likely to be caus[ed]" by exposure to these pollutants.118 At the same time,
WTP may be lower for those health outcomes for which causality has not been definitively

117	This decision criterion for selecting health effects to quantify and monetize PM2.5 and ozone is only applicable to
estimating the benefits of exposure of these two pollutants. This is also the approach used for identifying the
unqualified benefit categories for criteria pollutants. This decision criterion may not be applicable or suitable
for quantifying and monetizing health and ecological effects of other pollutants. The approach used to determine
whether there is sufficient evidence of a relationship between an endpoint affected by non-criteria pollutants, and
consequently a positive WTP for reductions in those pollutants, for other unqualified benefits described in this
section can be found in the source documentation for each of these pollutants (see relevant sections below). The
conceptual framework for estimating benefits when there is uncertainty in the causal relationship between a
hazard and the endpoints it potentially affects described here applies to these other pollutants.

118	EPA includes risk estimates for an example health endpoint with a causality determination of "suggestive, but not
sufficient to infer" that is associated with a potentially substantial economic value in the quantitative uncertainty
characterization (Health Benefits TSD section 6.2.3).

4-27


-------
established. This approach treats relationships with ISA causality determinations of "likely to be
causal" as if they were known to be causal, and therefore benefits could be overestimated. Table
4-4 reports the effects we quantified and those we did not quantify in this RIA. The list of benefit
categories not quantified is not exhaustive. The table below omits welfare effects such as
acidification and nutrient enrichment.

4-28


-------
Table 4-4 Health Effects of Ambient Ozone and PM2.5 and Climate Effects

Category

Effect

Effect
Quantified

Effect
Monetized

More
Information

Premature
mortality from

Adult premature mortality based on cohort study
estimates and expert elicitation estimates (age 65-99
or age 30-99)

~

~

PM ISA

exposure to PM2.5

Infant mortality (age <1)

V

V

PM ISA



Heart attacks (age >18)

S



PM ISA



Hospital admissions—cardiovascular (ages 65-99)

~

~

PM ISA



Emergency department visits— cardiovascular (age
0-99)





PM ISA



Hospital admissions—respiratory (ages 0-18 and 65-
99)

Emergency room visits—respiratory (all ages)
Cardiac arrest (ages 0-99; excludes initial hospital
and/or emergency department visits)

Stroke (ages 65-99)

Asthma onset (ages 0-17)

~

~

PM ISA



~

V
S

V

	71'	

PM ISA

PM ISA

PM ISA
PM ISA



Asthma symptoms/exacerbation (6-17)

S

~

PM ISA

Nonfatal
morbidity from

Lung cancer (ages 30-99)

Allergic rhinitis (hay fever) symptoms (ages 3-17)
Lost work days (age 18-65)

~
~
V

~
~
V

PM ISA
PM ISA
PM ISA

exposure to PM2.5

Minor restricted-activity days (age 18-65)





PM ISA



Hospital admissions—Alzheimer's disease (ages 65-
99)

~

~

PM ISA



Hospital admissions—Parkinson's disease (ages 65-
99)





PM ISA



Other cardiovascular effects (e.g., other ages)

—

	

PM ISA2



Other respiratory effects (e.g., pulmonary function,
non-asthma ER visits, non-bronchitis chronic
diseases, other ages and populations)

—

	

PM ISA2



Other nervous system effects (e.g., autism, cognitive
decline, dementia)

—

	

PM ISA2



Metabolic effects (e.g., diabetes)

—

	

PM ISA2



Reproductive and developmental effects (e.g., low
birth weight, pre-term births, etc.)

—

	

PM ISA2



Cancer, mutagenicity, and genotoxicity effects

—

	

PM ISA2

Mortality from

Premature respiratory mortality based on short-term
study estimates (0-99)





Ozone ISA

exposure to ozone

Premature respiratory mortality based on long-term
study estimates (age 30-99)

V

V

Ozone ISA



Hospital admissions—respiratory (ages 0-99)

S

s

Ozone ISA



Emergency department visits—respiratory (ages 0-
99)

~

~

Ozone ISA



Asthma onset (0-17)





Ozone ISA

Nonfatal
morbidity from
exposure to ozone

Asthma symptoms/exacerbation (asthmatics age 2-

	17)	

Allergic rhinitis (hay fever) symptoms (ages 3-17)
Minor restricted-activity days (age 18-65)

~
~

~
~

Ozone ISA

Ozone ISA
Ozone ISA



School absence days (age 5-17)

~

~

Ozone ISA



Decreased outdoor worker productivity (age 18-65)
Metabolic effects (e.g., diabetes)

Other respiratory effects (e.g., premature aging of
lungs)

—

—

Ozone ISA2
Ozone ISA2

Ozone ISA2

4-29


-------
Cardiovascular and nervous system effects
Reproductive and developmental effects

O/.onc ISA2
Ozone ISA2

Climate impacts from carbon dioxide (CO2)

•S Section 4.2

Climate
Effects

Other climate impacts (e.g., ozone, black carbon,
aerosols, other impacts)

IPCC,
Ozone ISA,
PM ISA

1	Valuation estimate excludes initial hospital and/or emergency department visits.

2	Not quantified due to data availability limitations and/or because current evidence is only suggestive of causality.

4.3.3 Calculating Counts of Air Pollution Effects Using the Health Impact Function

We use the environmental Benefits Mapping and Analysis Program—Community
Edition (BenMAP-CE) software program to quantify counts of premature deaths and illnesses
attributable to photochemical modeled changes in annual mean PM2.5 and summer season
average ozone concentrations for the years 2028, 2030, 2035, 2040 and 2045 using health impact
functions (Sacks et al., 2020). A health impact function combines information regarding: the
concentration-response relationship between air quality changes and the risk of a given adverse
outcome; the population exposed to the air quality change; the baseline rate of death or disease in
that population; and the air pollution concentration to which the population is exposed.

BenMAP quantifies counts of attributable effects using health impact functions, which
combine information regarding the: concentration-response relationship between air quality
changes and the risk of a given adverse outcome; population exposed to the air quality change;
baseline rate of death or disease in that population; and air pollution concentration to which the
population is exposed.

The following provides an example of a health impact function, in this case for PM2.5
mortality risk. We estimate counts of PIVh.s-related total deaths (yij) during each year i among
adults aged 18 and older (a) in each county in the contiguous U.S. j (j=l,...,J where J is the total
number of counties) as

where moija is the baseline total mortality rate for adults aged a=18-99 in county j in year
i stratified in 10-year age groups, e is Euler's Number, P is the risk coefficient for total mortality
for adults associated with annual average PM2.5 exposure, Cij is the annual mean PM2.5

yij Ea yija

yija = moija xeA(p-ACij-l) x pija, Eq[l]

4-30


-------
concentration in county j in year i, and Pija is the number of county adult residents aged a= 18-99
in county j in year i stratified into 5-year age groups.119

The BenMAP-CE tool is pre-loaded with projected population from the Woods & Poole
company; cause-specific and age-stratified death rates from the Centers for Disease Control and
Prevention, projected to future years; recent-year baseline rates of hospital admissions,
emergency department visits and other morbidity outcomes from the Healthcare Cost and
Utilization Program and other sources; concentration-response parameters from the published
epidemiologic literature cited in the Integrated Science Assessments for particulate matter and
ground-level ozone; and cost of illness or willingness to pay economic unit values for each
endpoint.

To assess economic value in a damage-function framework, the changes in environmental
quality must be translated into effects on people or on the things that people value. In some
cases, the changes in environmental quality can be directly valued. In other cases, such as for
changes in ozone and PM, a health and welfare impact analysis must first be conducted to
convert air quality changes into effects that can be assigned dollar values.

We note at the outset that EPA rarely has the time or resources to perform extensive new
research to measure directly either the health outcomes or their values for regulatory analyses.
Thus, similar to Kiinzli et al. (2000) and other, more recent health impact analyses, our estimates
are based on the best available methods of benefits transfer. Benefits transfer adapts primary
research from similar contexts to obtain the most accurate measure of benefits for the
environmental quality change under analysis. Adjustments are made for the level of
environmental quality change, the socio-demographic and economic characteristics of the
affected population, and other factors to improve the accuracy and robustness of benefits
estimates.

119 In this illustrative example, the air quality is resolved at the county level. For this RIA, we simulate air quality
concentrations at 12 km grid resolution. The BenMAP-CE tool assigns the rates of baseline death and disease
stored at the county level to the grid cell level using an area-weighted algorithm. This approach is described in
greater detail in the appendices to the BenMAP-CE user manual.

4-31


-------
4.3.4 Calculating the Economic Valuation of Health Impacts

After quantifying the change in adverse health impacts, the final step is to estimate the
economic value of these avoided impacts. The appropriate economic value for a change in a
health effect depends on whether the health effect is viewed ex ante (before the effect has
occurred) or ex post (after the effect has occurred). Reductions in ambient concentrations of air
pollution generally lower the risk of future adverse health effects by a small amount for a large
population. The appropriate economic measure is therefore ex ante WTP for changes in risk.
However, epidemiological studies generally provide estimates of the relative risks of a particular
health effect avoided due to a reduction in air pollution. A convenient way to use these data in a
consistent framework is to convert probabilities to units of avoided statistical incidences. This
measure is calculated by dividing individual WTP for a risk reduction by the related observed
change in risk. For example, suppose a regulation reduces the risk of premature mortality from 2
in 10,000 to 1 in 10,000 (a reduction of 1 in 10,000). If individual WTP for this risk reduction is
$1000, then the WTP for an avoided statistical premature mortality amounts to $10 million
($1000/0.0001 change in risk). Hence, this value is population-normalized, as it accounts for the
size of the population and the percentage of that population experiencing the risk. The same type
of calculation can produce values for statistical incidences of other health endpoints.

For some health effects, such as hospital admissions, WTP estimates are generally not
available. In these cases, we instead use the cost of treating or mitigating the effect to
economically value the health impact. For example, for the valuation of hospital admissions, we
use the avoided medical costs as an estimate of the value of avoiding the health effects causing
the admission. These cost-of-illness (COI) estimates generally (although not 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.

This analysis uses several recent improvements in health endpoint valuation relative to
the methods described in the Health Benefits TSD (U.S. EPA, 2023b). School loss days now
account for lost human capital formation, as was discussed in the Health Benefits TSD which
was reviewed by the EPA Scientific Advisory Board's Review of BenMAP and Benefits

4-32


-------
Methods. We include new estimates of the cost asthma onset and stroke beyond those described
in the Health Benefits TSD.

The new valuation estimate for school loss days is described in the Health Benefits TSD
in Section 5.3.8. We include two costs of school loss days: caregiver costs and loss of learning.
We calculate each separately and then sum. Caregiver costs are valued at their employers'
average cost for employed caregivers. For unemployed caregivers, the opportunity cost of their
time is calculated as the average take-home pay. The loss of learning is calculated based on the
impact of absences on learning multiplied by the impact of school learning on adult earnings.
The loss of learning estimate is currently only available for middle and high school students. The
two costs are summed.

The caregiver costs assumes that an adult caregiver stays home with the child and loses
any wage income they would have earned that day. For working caregivers, we follow EPA
guidance and value their time at the average wage including fringe benefits and overhead costs.
The average daily wage in 2021 was $195 (2015 dollars, assumed to be the average weekly wage
divided by 5),120 which yields an average daily labor cost of $340 for employed parents after
applying average multipliers of 1.46 for fringe benefits and 1.2 for overhead. For nonworking
caregivers, we assume that the opportunity cost of time is the average after-tax earnings. We
estimate the income tax rate for a median household to be 7 percent, yielding net earnings of
$195 multiplied by 0.93 or $181 (2015 dollars). The income tax rate of 7 percent is the
percentage difference in median post-tax income and median income from Tables A1 and CI in
Shrider et al. (2021).

The probability that a parent is working is measured with the employment population
ratio among people with their own children under 18 and is 77.2 percent.121 Combining the cost
of working and nonworking caregivers yields a caregiver cost of $305 per school loss day.

To measure the loss of learning, we update the Liu et al. (2021) estimate. Liu et al. (2021)
estimated the impact of a school absence on learnings as measured by an end-of-course test
score. We multiply by an estimate of the impact of learning as measured by end-of-course test

120	U.S. Bureau of Labor Statistics (2022), series Employment, Hours, and Earnings from the Current Employment
Statistics (Series ID CES0500000011).

121	U.S. Bureau of Labor Statistics Employment Characteristics of Families, 2021, Table 5.

4-33


-------
scores on adult income from Chetty et al. (2014). This approach yields an estimated learning loss
of $2,842 per school absence (discounted at 2 percent), $2,230 per school absence (discounted at
3 percent) and $975 per school absence (discounted at 7 percent).

We updated the Chetty et al. (2014) estimate to use 2010 income and to estimate lifetime
incomes discounted at 3 percent and 7 percent. Liu et al. (2021) estimate that a school absence
leads to a $1,200 reduction in lifetime earnings, based on the Chetty et al. (2014) estimate that
lifetime earnings are $522,000 (2010 dollars). We use 2010 ACS data from IPUMS to calculate
expected lifetime earnings of $1,137,732 (discounting at 2 percent), $892,579 (discounting at 3
percent) and $390,393 (discounting at 7 percent). We then multiply the Liu et al. (2021) estimate
of $1,200 by ($1,137,732 divided by $522,000) and ($892,579 divided by $522,000) and
($390,393 divided by $522,000) and convert from 2010 dollars to 2015 dollars based on the
Consumer Price Index for All Urban Consumers.

We use caregiver costs for preschool and elementary school children and the sum of
caregiver costs and loss of learning for middle school and high school students. We calculate that
31 percent of children under 18 are middle school and high school ages 13-18, assuming each
bin is distributed equally, so the combined average effect is $1,186 ($305 plus $2,842 multiplied
by 0.31) with 2 percent discounting, $1,000 ($305 plus $2,230 multiplied by 0.31) with 3 percent
discounting, and $610 ($305 plus $975 multiplied by 0.31) with 7 percent discounting in 2015
dollars (U.S. Census Bureau, 2010).122

We include a new estimate of the cost of illness of asthma onset based on Fann and
Maniloff (2023) since that described in the Health Benefits TSD (U.S. EPA, 2023b). These
estimates are $181,249 with a 2 percent discount rate, $146,370 with a 3 percent discount rate,
and $76,629 in 2015$. We include a new estimate of the cost of illness of stroke onset based on
Fann and Maniloff (2023). These estimates are $158,763 with a 2 percent discount rate,

$150,675 with a 3 percent discount rate, and $123,984 in 2015$.

122 U.S. Census Bureau, Age and Sex Composition in the United States: 2010, Table 1,

https ://www. census, gov/data/tables/2010/demo/age-and-sex/2010-age-sex-compo sition. html

4-34


-------
4.3.5 Benefits Analysis Data Inputs

In Figure 4-1, we summarize the key data inputs to the health impact and economic
valuation estimates using PM2.5 inputs as an example, which were calculated using BenMAP-CE
model version 1.5.1 (Sacks et al., 2020). In the sections below we summarize the data sources for
each of these inputs, including demographic projections, incidence and prevalence rates, effect
coefficients, and economic valuation.

Figure 4-1 Data Inputs and Outputs for the BenMAP-CE Model Using PM2.5 as an
Example

4.3.5.1 Demographic Data

Quantified and monetized human health impacts depend on the demographic
characteristics of the population, including age, location, and income. We use projections based
on economic forecasting models developed by Woods & Poole, Inc. (2015). The Woods & Poole
database contains county4evel projections of population by age, sex, and race to 2060, relative to
a baseline using the 2010 Census data. Projections in each county are determined simultaneously
with every other county in the U.S. to consider patterns of economic growth and migration. The
sum of growth in county4evel populations is constrained to equal a previously determined
national population growth, based on Bureau of Census estimates (Hollmann et al., 2000).

4-35


-------
According to Woods & Poole, linking county-level growth projections together and constraining
the projected population to a national-level total growth avoids potential errors introduced by
forecasting each county independently (for example, the projected sum of county-level
populations cannot exceed the national total). County projections are developed in a four-stage
process:

First, national-level variables such as income, employment, and populations are
forecasted.

Second, employment projections are made for 179 economic areas defined by the
Bureau of Economic Analysis (U.S. BEA, 2004), using an "export-base"
approach, which relies on linking industrial-sector production of non-locally
consumed production items, such as outputs from mining, agriculture, and
manufacturing with the national economy. The export-based approach requires
estimation of demand equations or calculation of historical growth rates for output
and employment by sector.

Third, population is projected for each economic area based on net migration rates
derived from employment opportunities and following a cohort-component
method based on fertility and mortality in each area.

Fourth, employment and population projections are repeated for counties, using
the economic region totals as bounds. The age, sex, and race distributions for each
region or county are determined by aging the population by single year by sex and
race for each year through 2060 based on historical rates of mortality, fertility,
and migration.

4.3.5.2 Baseline Incidence and Prevalence Estimates

Epidemiological studies of the association between pollution levels and adverse health
effects generally provide a direct estimate of the relationship of air quality changes to the relative
risk of a health effect, rather than estimating the absolute number of avoided cases. For example,
a typical result might be that a 5 |ig/m3 decrease in daily PM2.5 levels is associated with a
decrease in hospital admissions of 3 percent. A baseline incidence rate, necessary to convert this
relative change into a number of cases, is the estimate of the number of cases of the health effect

4-36


-------
per year in the assessment location, as it corresponds to baseline pollutant levels in that location.
To derive the total baseline incidence per year, this rate must be multiplied by the corresponding
population number. For example, if the baseline incidence rate is the number of cases per year
per million people, that number must be multiplied by the millions of people in the total
population.

The Health Benefits TSD (U.S. EPA, 2023b) (Table 12) summarizes the sources of
baseline incidence rates and reports average incidence rates for the endpoints included in the
analysis. For both baseline incidence and prevalence data, we used age-specific rates where
available. We applied concentration-response functions to individual age groups and then
summed over the relevant age range to provide an estimate of total population benefits. National-
level incidence rates were used for most morbidity endpoints, whereas county-level data are
available for premature mortality. Whenever possible, the national rates used are national
averages, because these data are most applicable to a national assessment of benefits. For some
studies, however, the only available incidence information comes from the studies themselves; in
these cases, incidence in the study population is assumed to represent typical incidence at the
national level.

We projected mortality rates such that future mortality rates are consistent with our
projections of population growth (U.S. EPA, 2023b). To perform this calculation, we began first
with an average of 2007-2016 cause-specific mortality rates. Using Census Bureau projected
national-level annual mortality rates stratified by age range, we projected these mortality rates to
2060 in 5-year increments (U.S. Census Bureau). Further information regarding this procedure
may be found in the Health Benefits TSD and the appendices to the BenMAP user manual (U.S.
EPA, 2022a, 2023b).

The baseline incidence rates for hospital admissions and emergency department visits
reflect the revised rates first applied in the Revised Cross-State Air Pollution Rule Update (U.S.
EPA, 2021). In addition, we revised the baseline incidence rates for acute myocardial infarction.
These revised rates are more recent than the rates they replace and more accurately represent the
rates at which populations of different ages, and in different locations, visit the hospital and
emergency department for air pollution-related illnesses (AHRQ, 2016). Lastly, these rates
reflect unscheduled hospital admissions only, which represents a conservative assumption that

4-37


-------
most air pollution-related visits are likely to be unscheduled. If air pollution-related hospital
admissions are scheduled, this assumption would underestimate these benefits.

4.3.5.3 Effect Coefficients

Our approach for selecting and parametrizing effect coefficients for the benefits analysis
is described fully in the Health Benefits TSD. Because of the substantial economic value
associated with estimated counts of PIVh.s-attributable deaths, we describe our rationale for
selecting among long-term exposure epidemiologic studies below; a detailed description of all
remaining endpoints may be found in the Health Benefits TSD.

A substantial body of published scientific literature documents the association between
PM2.5 concentrations and the risk of premature death (U.S. EPA, 2019a, 2022f). This body of
literature reflects thousands of epidemiology, toxicology, and clinical studies. The PM ISA,
completed as part of this review of the PM standards and reviewed by the Clean Air Scientific
Advisory Committee (CASAC) (U.S. EPA Science Advisory Board, 2022) concluded that there
is a causal relationship between mortality and both long-term and short-term exposure to PM2.5
based on the full body of scientific evidence (U.S. EPA, 2019a, 2022f). The size of the mortality
effect estimates from epidemiologic studies, the serious nature of the effect itself, and the high
monetary value ascribed to prolonging life make mortality risk reduction the most significant
health endpoint quantified in this analysis.

EPA selects hazard ratios from cohort studies to estimate counts of PM-related premature
death, following a systematic approach detailed in the Health Benefits TSD accompanying this
RIA that is generally consistent with previous RIAs (e.g., (U.S. EPA, 2019b, 2020a, 2020b,
2021, 2022c)). Briefly, clinically significant epidemiologic studies of health endpoints for which
ISAs report strong evidence are evaluated using established minimum and preferred criteria for
identifying studies and hazard ratios best characterizing risk. Further discussion of the cohort
studies and hazard ratios for quantifying ozone- and PIVh.s-attributable premature death can be
found below in Sections 4.3.6 and 4.3.7.

4.3.6 Quantifying Cases of Ozone-Attributable Premature Death

Mortality risk reductions account for the majority of monetized ozone-related and PM2.5-
related benefits. For this reason, this subsection and the following provide a brief background of

4-38


-------
the scientific assessments that underly the quantification of these mortality risks and identifies
the risk studies used to quantify them in this RIA, for ozone and PM2.5, respectively. As noted
above, the Health Benefits TSD describes fully the Agency's approach for quantifying the
number and value of ozone and PM2.5 air pollution-related impacts, including additional
discussion of how the Agency selected the risk studies used to quantify them in this RIA. The
Health Benefits TSD also includes additional discussion of the assessments that support
quantification of these mortality risk than provide here.

In 2008, the National Academies of Science (National Research Council, 2008) issued a
series of recommendations to EPA regarding the procedure for quantifying and valuing ozone-
related mortality due to short-term exposures. Chief among these was that "... short-term
exposure to ambient ozone is likely to contribute to premature deaths" and the committee
recommended that "ozone-related mortality be included in future estimates of the health benefits
of reducing ozone exposures..The NAS also recommended that".. .the greatest emphasis be
placed on the multicity and [National Mortality and Morbidity Air Pollution Studies
(NMMAPS)] ... studies without exclusion of the meta-analyses" (National Research Council,
2008). Prior to the 2015 Ozone NAAQS RIA, the Agency estimated ozone-attributable
premature deaths using an NMMAPS-based analysis of total mortality (Bell et al., 2004), two
multi-city studies of cardiopulmonary and total mortality (Huang et al., 2005; Schwartz, 2005)
and effect estimates from three meta-analyses of non-accidental mortality (Bell et al., 2005; Ito et
al., 2005; Levy et al., 2005). Beginning with the 2015 Ozone NAAQS RIA, the Agency began
quantifying ozone-attributable premature deaths using two newer multi-city studies of non-
accidental mortality (R. L. Smith et al., 2009; Zanobetti and Schwartz, 2008) and one long-term
cohort study of respiratory mortality (Jerrett et al., 2009). The 2020 Ozone ISA included changes
to the causality relationship determinations between short-term exposures and total mortality, as
well as including more recent epidemiologic analyses of long-term exposure effects on
respiratory mortality (U.S. EPA, 2020d).

EPA quantifies and monetizes effects the Integrated Science Assessment (ISA) identifies
as having either a causal or likely-to-be-causal relationship with the pollutant. Relative to the
2015 ISA, the 2020 ISA for Ozone reclassified the casual relationship between short-term ozone
exposure and total mortality, changing it from "likely to be causal" to "suggestive of, but not
sufficient to infer, a causal relationship." The 2020 Ozone ISA separately classified short-term

4-39


-------
O3 exposure and respiratory outcomes as being "causal" and long-term exposure as being "likely
to be causal." When determining whether there existed a causal relationship between short- or
long-term ozone exposure and respiratory effects, EPA evaluated the evidence for both
morbidity and mortality effects. The ISA identified evidence in the epidemiologic literature of an
association between ozone exposure and respiratory mortality, finding that the evidence was not
entirely consistent and there remained uncertainties in the evidence base.

EPA continues to quantify premature respiratory mortality attributable to both short- and
long-term exposure to ozone because doing so is consistent with: (1) the evaluation of causality
noted above; and (2) EPA's approach for selecting and quantifying endpoints described in the
Technical Support Document (TSD) "Estimating PM2.5- and Ozone-Attributable Health
Benefits," which was recently reviewed by the U.S. EPA Science Advisory Board (U.S. EPA,
2023b; U.S. EPA Science Advisory Board, 2024).

Beginning with the RCU analysis we use two estimates of ozone-attributable respiratory
deaths from short-term exposures estimated using the risk estimate parameters from Zanobetti
and Schwartz (2008) and Katsouyanni et al. (2009). Ozone-attributable respiratory deaths from
long-term exposures are estimated using Turner et al. (2016). Due to time and resource
limitations, we were unable to reflect the warm season defined by Zanobetti and Schwartz (2008)
as June-August. Instead, we apply this risk estimate to our standard warm season of April-
September.(R. L. Smith et al., 2009; Zanobetti and Schwartz, 2008) and one long-term cohort
study of respiratory mortality (Jerrett et al., 2009).

Table 11 in the Health Benefits TSD lists the ozone risk estimates used in benefits analysis.
4.3.7 Quantifying Cases of PM2 5-A ttributable Premature Death

The PM ISA, which was reviewed by the Clean Air Scientific Advisory Committee of
EPA's Science Advisory Board (SAB-CASAC), concluded that there is a causal relationship
between mortality and both long-term and short-term exposure to PM2.5 based on the entire body
of scientific evidence (U.S. EPA, 2022e; U.S. EPA Science Advisory Board, 2019, 2022). The
PM ISA also concluded that the scientific literature supports the use of a no-threshold log-linear
model to portray the PM-mortality concentration-response relationship while recognizing
potential uncertainty about the exact shape of the concentration-response relationship. The 2019

4-40


-------
PM ISA, which informed the setting of the 2020 PM NAAQS, reviewed available studies that
examined the potential for a population-level threshold to exist in the concentration-response
relationship. Based on such studies, the ISA concluded that the evidence supports the use of a
"no-threshold" model and that "little evidence was observed to suggest that a threshold exists"
(U.S. EPA, 2009a) (pp. 2-25 to 2-26). Consistent with this evidence, the Agency historically has
estimated health impacts above and below the prevailing NAAQS (U.S. EPA, 2010, 201 la,
2011b, 2012, 2015a, 2015b, 2015c, 2016b).

Following this systematic approach led to the identification of three studies best
characterizing the risk of premature death associated with long-term exposure to PM2.5 in the
U.S. (Pope et al., 2019; Turner et al., 2016; Wu et al., 2020). The PM ISA, Supplement to the
ISA, and 2022 Policy Assessment also identified these three studies as providing key evidence of
the association between long-term PM2.5 exposure and mortality (U.S. EPA, 2019a, 2022b,
2022f). These studies used data from three U.S. cohorts: (1) an analysis of Medicare
beneficiaries (Medicare); (2) the American Cancer Society (ACS); and (3) the National Health
Interview Survey (NHIS). As premature mortality typically constitutes the vast majority of
monetized benefits in a PM2.5 benefits assessment, quantifying effects using risk estimates
reported from multiple long-term exposure studies using different cohorts helps account for
uncertainty in the estimated number of PM-related premature deaths. Below we summarize the
three identified studies and hazard ratios and then describe our rationale for quantifying
premature PM-attributable deaths using two of these studies.

Wu et al. (2020) evaluated the relationship between long-term PM2.5 exposure and all-
cause mortality in more than 68.5 million Medicare enrollees (over the age of 64), using
Medicare claims data from 2000-2016 representing over 573 million person-years of follow up
and over 27 million deaths. This cohort included over 20 percent of the U.S. population and was,
at the time of publishing, the largest air pollution study cohort to date. The authors modeled
PM2.5 exposure at a 12 km grid resolution using a hybrid ensemble-based prediction model that
combined three machine learning models and relied on satellite data, land-use information,
weather variables, chemical transport model simulation outputs, and monitor data. Wu et al.,
2020 fit five different statistical models: a Cox proportional hazards model, a Poisson regression
model, and three causal inference approaches (GPS estimation, GPS matching, and GPS
weighting). All five statistical approaches provided consistent results; we report the results of the

4-41


-------
Cox proportional hazards model here. The authors adjusted for numerous individual-level and
community-level confounders, and sensitivity analyses suggest that the results are robust to
unmeasured confounding bias. In a single-pollutant model, the coefficient and standard error for
PM2.5 are estimated from the hazard ratio (1.066) and 95 percent confidence interval (1.058-
1.074) associated with a change in annual mean PM2.5 exposure of 10.0 |ig/m3 (Wu et al., 2020,
Table S3, Main analysis, 2000-2016 Cohort, Cox PH). We use a risk estimate from this study in
place of the risk estimate from Di et al. (2017). These two epidemiologic studies share many
attributes, including the Medicare cohort and statistical model used to characterize population
exposure to PM2.5. As compared to Di et al. (2017), Wu et al. (2020) includes a longer follow-up
period and reflects more recent PM2.5 concentrations.

Pope et al. (2019) examined the relationship between long-term PM2.5 exposure and all-
cause mortality in a cohort of 1,599,329 U.S. adults (aged 18-84 years) who were interviewed in
the National Health Interview Surveys (NHIS) between 1986 and 2014 and linked to the
National Death Index (NDI) through 2015. The authors also constructed a sub-cohort of 635,539
adults from the full cohort for whom body mass index (BMI) and smoking status data were
available. The authors employed a hybrid modeling technique to estimate annual-average PM2.5
concentrations derived from regulatory monitoring data and constructed in a universal kriging
framework using geographic variables including land use, population, and satellite estimates.
Pope et al. (2019) assigned annual-average PM2.5 exposure from 1999-2015 to each individual by
census tract and used complex (accounting for NHIS's sample design) and simple Cox
proportional hazards models for the full cohort and the sub-cohort. We select the Hazard Ratio
calculated using the complex model for the sub-cohort, which controls for individual-level
covariates including age, sex, race-ethnicity, inflation-adjusted income, education level, marital
status, rural versus urban, region, survey year, BMI, and smoking status. In a single-pollutant
model, the coefficient and standard error for PM2.5 are estimated from the hazard ratio (1.12) and
95 percent confidence interval (1.08-1.15) associated with a change in annual mean PM2.5
exposure of 10.0 |ig/m3 (Pope et al., 2019) (Table 2, Subcohort). This study exhibits two key
strengths that makes it particularly well suited for a benefits analysis: (1) it includes a long
follow-up period with recent (and thus relatively low) PM2.5 concentrations; (2) the NHIS cohort
is representative of the U.S. population, especially with respect to the distribution of individuals
by race, ethnicity, income, and education.

4-42


-------
EPA has historically used estimated Hazard Ratios from extended analyses of the ACS
cohort (Krewski et al., 2009; Pope et al., 2002; Pope et al., 1995) to estimate PM-related risk of
premature death. More recent ACS analyses (Pope et al., 2015; Turner et al., 2016):

extended the follow-up period of the ACS CSP-II to 22 years (1982-2004),

evaluated 669,046 participants over 12,662,562 person-years of follow up and
237,201 observed deaths, and

applied a more advanced exposure estimation approach than had previously been
used when analyzing the ACS cohort, combining the geostatistical Bayesian Maximum Entropy
framework with national-level land use regression models.

The total mortality hazard ratio best estimating risk from these ACS cohort studies was
based on a random-effects Cox proportional hazard model incorporating multiple individual and
ecological covariates (relative risk =1.06, 95 percent confidence intervals 1.04-1.08 per 10|ig/m3
increase in PM2.5) from Turner et al., 2016. The relative risk estimate is identical to a risk
estimate drawn from earlier ACS analysis of all-cause long-term exposure PM2.5-attributable
mortality (Krewski et al., 2009). However, as the ACS hazard ratio is quite similar to the
Medicare estimate of (1.066, 1.058-1.074), especially when considering the broader age range
(>29 vs >64), only the Wu et al. (2020) and Pope et al. (2019) are included in the main benefits
assessments, with Wu et al. (2020) representing results from both the Medicare and ACS
cohorts.

Table 10 in the Health Benefits TSD lists the PM2.5 risk estimates used in benefits
analysis.

4.3.8 Characterizing Uncertainty in the Estimated 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. The Health
Benefits TSD details our approach to characterizing uncertainty in both quantitative and
qualitative terms (U.S. EPA, 2023b). That Health Benefits TSD describes the sources of
uncertainty associated with key input parameters including emissions 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 benefits, and

4-43


-------
assumptions regarding the future state of the country (i.e., regulations, technology, and human
behavior). Each of these inputs is uncertain and affects the size and distribution of the estimated
benefits. When the uncertainties from each stage of the analysis are compounded, even small
uncertainties can have large effects on the total quantified benefits.

To characterize uncertainty and variability into this assessment, we incorporate three
quantitative analyses described below and in greater detail within the Health Benefits TSD
(Section 7.1):

1.	A Monte Carlo assessment that accounts for random sampling error and between study
variability in the epidemiological and economic valuation studies;

2.	The quantification of PM-related mortality using alternative PM2.5 mortality effect
estimates drawn from two long-term cohort studies; and

3.	Presentation of 95th percentile confidence interval around each risk estimate.

Quantitative characterization of other sources of PM2.5 uncertainties are discussed only in
Section 7.1 of the Health Benefits TSD:

1.	For adult all-cause mortality:

a.	The distributions of air quality concentrations experienced by the original

cohort population (Health Benefits TSD Section 7.1.2.1);

b.	Methods of estimating and assigning exposures in epidemiologic studies

(Health Benefits TSD Section 7.1.2.2);

c.	Confounding by ozone (Health Benefits TSD Section 7.1.2.3); and

d.	The statistical technique used to generate hazard ratios in the epidemiologic

study (Health Benefits TSD Section 7.1.2.4).

2.	Plausible alternative risk estimates for asthma onset in children (Health Benefits TSD
Section 7.1.3), cardiovascular hospital admissions (Health Benefits TSD Section 7.1.4,), and
respiratory hospital admissions (Health Benefits TSD Section 7.1.5);

3.	Effect modification of PM2.5-attributable health effects in at-risk populations (Health
Benefits TSD Section 7.1.6).

4-44


-------
Quantitative consideration of baseline incidence rates and economic valuation estimates
are provided in Section 7.3 and 7.4 of the TSD, respectively. Qualitative discussions of various
sources of uncertainty can be found in Section 7.5 of the TSD.

4.3.8.1	Monte Carlo Assessment

Similar to recent RIAs that monetize PM2.5 and ozone-related health benefits, we used
Monte Carlo methods for characterizing random sampling error associated with the
concentration response functions from epidemiological studies and random effects modeling to
characterize both sampling error and variability across the economic valuation functions. The
Monte Carlo simulation in the BenMAP-CE software randomly samples from a distribution of
incidence and valuation estimates to characterize the effects of uncertainty on output variables.
Specifically, we used Monte Carlo methods to generate confidence intervals around the
estimated health impact and monetized benefits. The reported standard errors in the
epidemiological studies determined the distributions for individual effect estimates for endpoints
estimated using a single study. For endpoints estimated using a pooled estimate of multiple
studies, the confidence intervals reflect both the standard errors and the variance across studies.
The confidence intervals around the monetized benefits incorporate the epidemiology standard
errors as well as the distribution of the valuation function. 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.

4.3.8.2	Sources of Uncertainty Treated Qualitatively

Although we strive to incorporate as many quantitative assessments of uncertainty as
possible, there are several aspects we are only able to address qualitatively. These attributes are
summarized below and described more fully in the Health Benefits TSD.

Key assumptions underlying the estimates for premature mortality, which account for
over 98 percent of the total monetized benefits in this analysis, include the following:

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

4-45


-------
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, which was reviewed by
CASAC, concluded that "across exposure durations and health effects categories ... the evidence
does not indicate that any one source or component is consistently more strongly related with
health effects than PM2.5 mass" (U.S. EPA, 2019a).

2.	We assume that the health impact function for fine particles is log-linear down to
the lowest air quality levels modeled in this analysis. Thus, the estimates include health benefits
from reducing fine particles in areas with varied concentrations of PM2.5, including both regions
that are in attainment with the particulate matter NAAQS and those that do not meet the standard
down to the lowest modeled concentrations. The PM ISA concluded that "the majority of
evidence continues to indicate a linear, no-threshold concentration-response relationship for
long-term exposure to PM2.5 and total (nonaccidental) mortality" (U.S. EPA, 2019a). The
Supplement to the 2019 Integrated Science Assessment for Particulate Matter continues to
support a no-threshold concentration-response relationship.123

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 Science
Advisory Board, 2004), which affects the valuation of mortality benefits at different discount
rates. Similarly, we assume there is a cessation lag between the change in PM exposures and
both the development and diagnosis of lung cancer.

4.	Uncertainties associated with the IPM projections used to derive the inputs for the
air quality modeling in this analysis are outlined in Section 3.8. IPM is a system-wide least-cost
optimization model that projects EGU behavior across the geographically contiguous U.S., and
projects one possible combination of compliance outcomes under a given policy scenario. The
GHG mitigation measures in this RIA are illustrative since States are afforded flexibility to
implement the final rules, and thus the impacts could be different to the extent states make
different choices than those assumed in the illustrative analysis. Additionally, the way that EGUs

123 https://assessments.epa.gov/isa/document/&deid=354490

4-46


-------
comply with the GHG mitigation measures may differ from the methods forecast in the modeling
for this RIA.

5. Uncertainties associated with applying air quality modeling to create ozone and
PM2.5 surfaces are discussed in Appendix B.

4.3.9 Estimated Number and Economic Value of Health Benefits

Table 4-5 through Table 4-14 report the estimated number of reduced premature deaths
and illnesses in each year relative to the baseline along with the 95 percent confidence interval.
Table 4-5 through report the ozone-related health benefits for each scenario and year, and Table
4-10 through Table 4-14 report the PM-related health benefits for each scenario and year. The
number of reduced estimated deaths and illnesses from the three illustrative scenarios are
calculated from the sum of individual reduced mortality and illness risk across the population.

Table 4-15 through Table 4-19 report the estimated economic value of avoided premature
deaths and illness in each year relative to the baseline along with the 95 percent confidence
interval. Table 4-20 summarizes the monetized benefits for all illustrative scenarios and the five
analysis years. We also report the stream of benefits from 2028 through 2047 for the final rules
and alternatives, using the monetized sums of long-term ozone and PM2.5 mortality and
morbidity impacts (Table 4-15 through Table 4-19).124 When estimating the value of improved
air quality over a multi-year time horizon, the analysis applies population growth and income
growth projections for each future year through 2047 and estimates of baseline mortality
incidence rates at five-year increments.

Table 4-21 through Table 4-23 include two estimates for each scenario at a 2, 3, and 7
percent discount rate. These estimates were quantified using two different epidemiological
estimates for the mortality impact of ozone and two different epidemiological estimates for the
mortality impact of PM, as well as their sum. For ozone, one estimate reflects the impacts
associated with short-term exposure on mortality impacts while the other reflects long-term
exposure on mortality. For PM, one estimate reflects impacts associated mortality estimated
based on Pope et al. (2019), while the other reflects impacts associated with mortality estimated

124 EPA continues to refine its approach for estimating and reporting PM-related effects at lower concentrations. The
Agency acknowledges the additional uncertainty associated with effects estimated at these lower levels and seeks
to develop quantitative approaches for reflecting this uncertainty in the estimated PM benefits.

4-47


-------
based on Wu et al. (2020). These estimates should not be thought of as representing low and high
bounds.

As shown in Tables 4-21 through 4-23, the present value of the monetized human health
benefits from 2028 through 2047 is as high as $120 billion (2 percent discount rate) and the
equivalent annualized values (EAVs) are $6.3 billion at a 2 percent discount rate. Although there
are disbenefits for ozone in 2040 and a mix of benefits and disbenefits for PM2.5 in 2040, we
expect that most locations with disbenefits from the final rules will still show positive benefits
when accounting for impacts on ozone and PM2.5 concentrations over the entire period that
health benefits are analyzed (i.e., 2028 to 2047).

4-48


-------
Table 4-5 Estimated Avoided Ozone-Related Premature Respiratory Mortalities and
Illnesses for the Illustrative Scenarios for 2028 (95 percent confidence interval)"	





Final Rules

Alternative 1

Alternative 2

Avoided premature respiratory mortalities

Long-

term

exposure

Turner et al. (2016)b

60

(42 to 78)

56

(39 to 73)

54

(37 to 69)

Short-

term

exposure

Katsouyanni et al. (2009)b c and
Zanobetti et al. (2008)c pooled

2.7
(1.1 to 4.3)

2.5
(1.0 to 4.0)

2.4
(0.98 to 3.8)

Morbidity
effects

Long-

Asthma onsetd

510
(430 to 580)

450
(390 to 510)

430
(370 to 490)

term
exposure

Allergic rhinitis symptomsf

2,900
(1,500 to 4,200)

2,600
(1,400 to 3,800)

2,500
(1,300 to 3,600)



Hospital admissions—respiratory0

7.9
(-2.1 to 17)

7.3
(-1.9 to 16)

6.6
(-1.7 to 15)



ED visits—respiratory6

160
(45 to 340)

150
(41 to 310)

140
(40 to 300)

Short-
term

Asthma symptoms

93,000 (-12,000
to 190,000)

84,000 (-10,000
to 170,000)

80,000 (-9,900
to 170,000)

exposure

Minor restricted-activity daysc e

41,000
(16,000 to
64,000)

36,000
(15,000 to
57,000)

35,000
(14,000 to
55,000)



School absence days

33,000
(-4,600 to 69,000)

29,000
(-4,200 to
62,000)

28,000
(-4,000 to
59,000)

3 Values rounded to two significant figures.

b Applied risk estimate derived from April-September exposures to estimates of ozone across the May-September
warm season.

c Converted ozone risk estimate metric from maximum daily 1-hour average (MDA1) to maximum daily 8-hour
average (MDA8).

d Applied risk estimate derived from June-August exposures to estimates of ozone across the May-September warm
season.

e Applied risk estimate derived from full year exposures to estimates of ozone across the May-September warm
season.

f Converted ozone risk estimate metric from daily 24-hour average (DA24) to MDA8.

4-49


-------
Table 4-6 Estimated Avoided Ozone-Related Premature Respiratory Mortalities and
Illnesses for the Illustrative Scenarios for 2030 (95 percent confidence interval)"	





Final Rules

Alternative 1

Alternative 2

Avoided premature respiratory mortalities

Long-

term

exposure

Turner et al. (2016)b

60

(42 to 78)

58

(41 to 76)

47

(32 to 61)

Short-

term

exposure

Katsouyanni et al. (2009)b c and
Zanobetti et al. (2008)c pooled

2.7
(1.1 to 4.3)

2.6
(1.1 to 4.2)

2.1

(0.95 to 3.2)

Morbidity
effects

Long-

Asthma onsetd

520
(440 to 590)

510
(430 to 580)

390
(330 to 440)

term
exposure

Allergic rhinitis symptomsf

3,000
(1,600 to 4,300)

2,900
(1,500 to 4,300)

2,200
(1,200 to 3,300)



Hospital admissions—respiratory0

7.7
(-2.0 to 17)

7.5
(-2.0 to 17)

5.9
(-1.5 to 13)



ED visits—respiratory6

160
(44 to 340)

160
(43 to 330)

130
(35 to 270)

Short-
term

Asthma symptoms

97,000 (-12,000
to 200,000)

95,000 (-12,000
to 200,000)

73,000 (-8,900
to 150,000)

exposure

Minor restricted-activity daysc e

41,000
(16,000 to
65,000)

33,000
(-4,700 to
70,000)

32,000
(13,000 to
50,000)



School absence days

34,000
(-4,800 to 71,000)

41,000
(16,000 to
64,000)

26,000
(-3,600 to
54,000)

3 Values rounded to two significant figures.

b Applied risk estimate derived from April-September exposures to estimates of ozone across the May-September
warm season.

c Converted ozone risk estimate metric from maximum daily 1-hour average (MDA1) to maximum daily 8-hour
average (MDA8).

d Applied risk estimate derived from June-August exposures to estimates of ozone across the May-September warm
season.

e Applied risk estimate derived from full year exposures to estimates of ozone across the May-September warm
season.

f Converted ozone risk estimate metric from daily 24-hour average (DA24) to MDA8.

4-50


-------
Table 4-7 Estimated Avoided Ozone-Related Premature Respiratory Mortalities and
Illnesses for the Illustrative Scenarios for 2035 (95 percent confidence interval)"	





Final Rules

Alternative 1

Alternative 2

Avoided premature respiratory mortalities

Long-

term

exposure

Turner et al. (2016)b

120
(85 to 160)

140
(98 to 180)

120
(86 to 160)

Short-

term

exposure

Katsouyanni et al. (2009)b c and
Zanobetti et al. (2008)c pooled

5.6
(2.2 to 8.8)

6.4
(2.6 to 10)

5.6
(2.3 to 8.9)

Morbidity
effects

Long-

Asthma onsetd

710
(610 to 810)

850
(730 to 970)

720
(620 to 820)

term
exposure

Allergic rhinitis symptomsf

4,200
(2,200 to 6,100)

5,000
(2,700 to 7,300)

4,300
(2,300 to 6,200)



Hospital admissions—respiratory0

15

(-3.9 to 33)

17

(-4.5 to 38)

15

(-4.0 to 34)



ED visits—respiratory6

250
(70 to 540)

300
(81 to 620)

260
(70 to 540)

Short-
term

Asthma symptoms

130,000
(-16,000 to
280,000)

160,000
(-20,000 to
330,000)

140,000
(-17,000 to
280,000)

exposure

Minor restricted-activity daysc e

61,000
(24,000 to
96,000)

73,000
(29,000 to
120,000)

63,000
(25,000 to
99,000)



School absence days

48,000
(-6,800 to
100,000)

58,000
(-8,200 to
120,000)

49,000
(-7,000 to
100,000)

3 Values rounded to two significant figures.

b Applied risk estimate derived from April-September exposures to estimates of ozone across the May-September
warm season.

c Converted ozone risk estimate metric from maximum daily 1-hour average (MDA1) to maximum daily 8-hour
average (MDA8).

d Applied risk estimate derived from June-August exposures to estimates of ozone across the May-September warm
season.

e Applied risk estimate derived from full year exposures to estimates of ozone across the May-September warm
season.

f Converted ozone risk estimate metric from daily 24-hour average (DA24) to MDA8.

4-51


-------
Table 4-8 Estimated Avoided Ozone-Related Premature Respiratory Mortalities and
Illnesses for the Illustrative Scenarios for 2040 (95 percent confidence interval)"	





Final Rules

Alternative 1

Alternative 2

Avoided premature respiratory mortalities

Long-

term

exposure

Turner et al. (2016)b

-7.0
(-4.9 to -9.1)

-1.1
(-0.73 to -1.4)

-8.8
(-6.1 to -11)

Short-

term

exposure

Katsouyanni et al. (2009)b c and
Zanobetti et al. (2008)c pooled

-0.32
(-0.50 to -0.13)

-0.047
(-0.074 to -
0.019)

-0.40
(-0.62 to -0.16)

Morbidity
effects

Long-

Asthma onsetd

-39
(-33 to -44)

4.2
(3.7 to 4.7)

-58
(-50 to -66)

term



-240
(-120 to -350)

14

(7.8 to 21)

-350
(-180 to -500)

exposure

Allergic rhinitis symptomsf



Hospital admissions—respiratory0

-1.4
(0.36 to-3.1)

-0.61
(0.16 to-1.3)

-1.6
(0.42 to-3.6)



ED visits—respiratory6

-25
(-53 to -7.0)

-11

(-24 to -3.1)

-33
(-68 to -9.0)

Short-
term



-7,400
(900 to -15,000)

590
(-76 to 1,200)

-11,000

Asthma symptoms

(1,300 to-
23,000)

exposure



-4,800
(-1,900 to -7,600)

-860
(-340 to -1,400)

-6,100



Minor restricted-activity daysc e

(-2,400 to -
9,600)



School absence days

-2,800
(390 to -5,800)

130
(-20 to 250)

-4,000
(560 to -8,400)

3 Values rounded to two significant figures.

b Applied risk estimate derived from April-September exposures to estimates of ozone across the May-September
warm season.

c Converted ozone risk estimate metric from maximum daily 1-hour average (MDA1) to maximum daily 8-hour
average (MDA8).

d Applied risk estimate derived from June-August exposures to estimates of ozone across the May-September warm
season.

e Applied risk estimate derived from full year exposures to estimates of ozone across the May-September warm
season.

f Converted ozone risk estimate metric from daily 24-hour average (DA24) to MDA8.

4-52


-------
Table 4-9 Estimated Avoided Ozone-Related Premature Respiratory Mortalities and
Illnesses for the Illustrative Scenarios in 2045 (95 percent confidence interval)a'b





Final Rules

Alternative 1

Alternative 2

Avoided premature respiratory mortalities

Long-

term

exposure

Turner et al. (2016)b

130
(89 to 170)

130
(90 to 170)

130
(90 to 170)

Short-

Katsouyanni et al.

5 8

5 9

5 8

term
exposure

(2009)b c and Zanobetti
et al. (2008)c pooled

(2.3 to 9.2)

(2.4 to 9.2)

(2.4 to 9.2)

Morbidity effects

Long-

term

exposure

Asthma onsetd

860
(730 to 970)

860
(740 to 980)

860
(740 to 980)

Allergic rhinitis
symptomsf

5,000
(2,600 to 7,300)

5,000
(2,600 to 7,300)

5,000
(2,600 to 7,300)



Hospital admissions—

16

17

17



respiratory0

(-4.3 to 37)

(-4.3 to 37)

(-4.3 to 37)



ED visits—respiratory6

290
(79 to 600)

290
(80 to 610)

290
(80 to 610)

Short-
term

Asthma symptoms

160,000 (-20,000 to
330,000)

160,000 (-20,000 to
330,000)

160,000 (-20,000 to
330,000)

exposure

Minor restricted-

76,000

76,000

76,000



activity daysce

(30,000 to 120,000)

(30,000 to 120,000)

(30,000 to 120,000)



School absence days

57,000
(-8,100 to 120,000)

58,000
(-8,100 to 120,000)

58,000
(-8,100 to 120,000)

3 Values rounded to two significant figures.

b Applied risk estimate derived from April-September exposures to estimates of ozone across the May-September
warm season.

c Converted ozone risk estimate metric from MDA1 to MDA8.

d Applied risk estimate derived from June-August exposures to estimates of ozone across the May-September warm
season.

e Applied risk estimate derived from full year exposures to estimates of ozone across the May-September warm
season.

f Converted ozone risk estimate metric from DA24 to MDA8.

4-53


-------
Table 4-10 Estimated Avoided PM-Related Premature Mortalities and Illnesses for the
Illustrative Scenarios in 2028 (95 percent confidence interval)	

Avoided Mortality

Final Rules

Alternative 1

Alternative 2

(Pope et al., 2019) (adult
mortality ages 18-99 years)

450
(320 to 570)

400
(290 to 510)

370
(260 to 470)

(Wu et al., 2020) (adult mortality
ages 65-99 years)

210
(180 to 230)

190
(170 to 210)

170
(150 to 190)

(Woodruff et al., 2008) (infant
mortality)

0.59
(-0.37 to 1.5)

0.51
(-0.32 to 1.3)

0.48
(-0.30 to 1.2)

Avoided Morbidity

Hospital admissions—
cardiovascular (age >18)

31

(23 to 39)

28

(20 to 35)

56

(-22 to 130)

Hospital admissions—respiratory

23

(8.2 to 38)

21

(7.2 to 33)

19

(6.5 to 30)

ED visits-cardiovascular

68

(-26 to 160)

61

(-23 to 140)

56

(-22 to 130)

ED visits—respiratory

140

(27 to 290)

120
(24 to 250)

110
(22 to 240)

Acute Myocardial Infarction

7.4
(4.3 to 10)

6.6
(3.8 to 9.2)

5.9
(3.5 to 8.3)

Cardiac arrest

3.3

(-1.4 to 7.5)

3.0
(-1.2 to 6.7)

2.7
(-1.1 to 6.2)

Hospital admissions-
Alzheimer's Disease

120
(88 to 150)

100
(78 to 130)

95

(71 to 120)

Hospital admissions-
Parkinson's Disease

14

(7.0 to 20)

12

(6.3 to 18)

11

(5.8 to 17)

Stroke

13

(3.4 to 22)

12

(3.0 to 20)

11

(2.8 to 18)

Lung cancer

15

(4.5 to 25)

13

(4.0 to 22)

12

(3.7 to 20)

Hay Fever/Rhinitis

3.400
(830 to 5,900)

3,000
(730 to 5,200)

2,800
(680 to 4,900)

Asthma Onset

530
(510 to 550)

470
(450 to 490)

440

(420 to 460)

Asthma symptoms - Albuterol
use

100,000
(-48,000 to 240,000)

88,000
(-43,000 to 210,000)

82,000
(-40,000 to 200,000)

Lost work days

26,000
(22,000 to 29,000)

23,000
(19,000 to 26,000)

21.000
(18,000 to 24,000)

Minor restricted-activity daysdf

150.000
(120,000 to 180,000)

130,000
(110,000 to 160,000)

120.000
(100,000 to 150,000)

Note: Values rounded to two significant figures.

4-54


-------
Table 4-11 Estimated Avoided PM-Related Premature Mortalities and Illnesses for the
Illustrative Scenarios in 2030 (95 percent confidence interval)	

Avoided Mortality

Final Rules

Alternative 1

Alternative 2

(Pope et al., 2019) (adult
mortality ages 18-99 years)

290
(200 to 360)

250
(180 to 310)

190
(140 to 240)

(Wu et al., 2020) (adult mortality
ages 65-99 years)

140
(120 to 150)

120
(100 to 130)

89

(79 to 100)

(Woodruff et al., 2008) (infant
mortality)

0.36
(-0.23 to 0.93)

0.32
(-0.20 to 0.82)

0.27
(-0.17 to 0.70)

Avoided Morbidity

Hospital admissions—
cardiovascular (age >18)

20

(14 to 25)

17

(12 to 21)

13

(9.6 to 17)

Hospital admissions—respiraton

14

(4.9 to 23)

12

(4.3 to 20)

7.7
(3.4 to 12)

ED visits-cardiovascular

43

(-16 to 99)

37

(-14 to 86)

31

(-12 to 71)

ED visits—respiratory

85

(17 to 180)

74

(15 to 150)

65

(13 to 130)

Acute Myocardial Infarction

4.6
(2.6 to 6.4)

3.9
(2.3 to 5.5)

3.0
(1.7 to 4.2)

Cardiac arrest

2.1

(-0.86 to 4.8)

1.8

(-0.74 to 4.1)

1.5

(-0.60 to 3.3)

Hospital admissions-
Alzheimer's Disease

75

(56 to 94)

65

(49 to 81)

46

(34 to 57)

Hospital admissions-
Parkinson's Disease

9.0
(4.6 to 13)

7.8
(3.9 to 11)

5.7
(2.9 to 8.4)

Stroke

8.4
(2.2 to 14)

7.2
(1.9 to 12)

5.7
(1.5 to 9.7)

Lung cancer

9.5
(2.9 to 16)

8.2
(2.5 to 14)

6.5
(2.0 to 11)

Hay Fever/Rhinitis

2,200
(530 to 3,800)

1,900
(470 to 3,400)

1,600
(390 to 2,800)

Asthma Onset

340
(330 to 360)

300
(290 to 310)

250
(240 to 260)

Asthma symptoms - Albuterol
use

64,000
(-31,000 to 160,000)

57,000
(-28,000 to 140,000)

47,000
(-23,000 to 110,000)

Lost work days

16,000
(13,000 to 18,000)

14,000
(12,000 to 16,000)

12,000
(9,800 to 13,000)

Minor restricted-activity daysdf

94,000
(76,000 to 110,000)

82,000
(67,000 to 97,000)

69,000
(56,000 to 81,000)

Note: Values rounded to two significant figures.

4-55


-------
Table 4-12 Estimated Avoided PM-Related Premature Mortalities and Illnesses for the
Illustrative Scenarios in 2035 (95 percent confidence interval)	

Avoided Mortality

Final Rules

Alternative 1

Alternative 2

(Pope et al., 2019) (adult
mortality ages 18-99 years)

1,100
(820 to 1,400)

1,200
(840 to 1,500)

1,200
(840 to 1,500)

(Wu et al., 2020) (adult mortality
ages 65-99 years)

560
(490 to 620)

580
(510 to 640)

580
(510 to 650)

(Woodruff et al., 2008) (infant
mortality)

1.2

(-0.73 to 3.0)

1.2

(-0.75 to 3.1)

1.2

(-0.75 to 3.1)

Avoided Morbidity

Hospital admissions—
cardiovascular (age >18)

81

(59 to 100)

84

(61 to 110)

84

(61 to 110)

Hospital admissions—respiratory

40

(17 to 62)

55

(19 to 89)

55

(19 to 90)

ED visits-cardiovascular

170
(-64 to 390)

170
(-66 to 400)

170
(-66 to 400)

ED visits—respiratory

310
(62 to 650)

320
(63 to 670)

320
(63 to 670)

Acute Myocardial Infarction

19

(11 to 26)

19

(11 to 27)

19

(11 to 27)

Cardiac arrest

8.0
(-3.3 to 18)

8.3
(-3.4 to 19)

8.3
(-3.4 to 19)

Hospital admissions-
Alzheimer's Disease

320
(240 to 390)

320
(240 to 400)

330
(240 to 410)

Hospital admissions-
Parkinson's Disease

36

(18 to 53)

37

(19 to 55)

37

(19 to 55)

Stroke

33

(8.5 to 56)

34

(8.7 to 58)

34

(8.8 to 58)

Lung cancer

38

(12 to 64)

40

(12 to 66)

40

(12 to 66)

Hay Fever/Rhinitis

7,700
(1,900 to 13,000)

8,000
(1,900 to 14,000)

8,000
(1,900 to 14,000)

Asthma Onset

1,200
(1,100 to 1,200)

1,200
(1,200 to 1,300)

1,200
(1,200 to 1,300)

Asthma symptoms - Albuterol
use

230,000
(-110,000 to 550,000)

240,000
(-110,000 to 570,000)

240,000
(-110,000 to 570,000)

Lost work days

57,000
(48,000 to 66,000)

60,000
(50,000 to 69,000)

60,000
(50,000 to 69,000)

Minor restricted-activity daysdf

340,000
(270,000 to 400,000)

350,000
(280,000 to 420,000)

350,000
(280,000 to 420,000)

Note: Values rounded to two significant figures.

4-56


-------
Table 4-13 Estimated Avoided PM-Related Premature Mortalities and Illnesses for the
Illustrative Scenarios in 2040 (95 percent confidence interval)	

Avoided Mortality

Final Rules

Alternative 1

Alternative 2

(Pope et al., 2019) (adult
mortality ages 18-99 years)

-22
(-16 to -28)

0.92
(0.66 to 1.2)

17

(13 to 22)

(Wu et al., 2020) (adult mortality
ages 65-99 years)

-11
(-9.2 to -12)

0.92
(0.81 to 1.0)

9.5
(8.3 to 11)

(Woodruff et al., 2008) (infant
mortality)

-0.023
(0.015 to -0.060)

-0.0039
(0.0025 to -0.010)

0.0057
(-0.0036 to 0.015)

Avoided Morbidity

Hospital admissions—
cardiovascular (age >18)

-2.1
(-1.6 to -2.7)

-0.37
(-0.27 to -0.47)

0.80
(0.58 to 1.0)

Hospital admissions—respiraton

-1.6
(-0.56 to -2.6)

-0.50
(-0.19 to -0.79)

0.15
(0.019 to 0.27)

ED visits-cardiovascular

-4.0
(1.5 to -9.3)

-0.61
(0.23 to -1.4)

1.2

(-0.48 to 2.9)

ED visits—respiratory

-5.9
(-12 to -1.2)

0.36
(0.071 to 0.75)

3.0

(0.60 to 6.3)

Acute Myocardial Infarction

-0.60
(-0.35 to -0.84)

-0.21
(-0.12 to -0.29)

0.065
(0.038 to 0.092)

Cardiac arrest

-0.13
(-0.30 to 0.053)

0.039
(-0.016 to 0.086)

0.14
(-0.058 to 0.32)

Hospital admissions-
Alzheimer's Disease

-14
(-10 to -18)

-7.8
(-5.8 to -9.7)

-2.2
(-1.6 to -2.8)

Hospital admissions-
Parkinson's Disease

-0.56
(-0.29 to -0.83)

0.18
(0.090 to 0.26)

0.77
(0.39 to 1.1)

Stroke

-0.53
(-0.14 to -0.90)

0.15
(0.038 to 0.25)

0.59
(0.15 to 1.0)

Lung cancer

-0.62
(-0.19 to -1.0)

0.22
(0.066 to 0.36)

0.77
(0.24 to 1.3)

Hay Fever/Rhinitis

-78

(-19 to -140)

79

(19 to 140)

160
(39 to 280)

Asthma Onset

-11
(-11 to-12)

13

(12 to 13)

25

(24 to 26)

Asthma symptoms - Albuterol
use

-2,400
(1,100 to-5,700)

2,300
(-1,100 to 5,500)

4,700
(-2,300 to 11,000)

Lost work days

-870
(-740 to -1,000)

360
(310 to 420)

990
(840 to 1,100)

Minor restricted-activity daysdf

-5,200
(-4,200 to -6,100)

2,100
(1,700 to 2,500)

5,900
(4,800 to 7,000)

Note: Values rounded to two significant figures.

4-57


-------
Table 4-14 Estimated Avoided PM-Related Premature Mortalities and Illnesses for the
Illustrative Scenarios in 2045 (95 percent confidence interval)	

Avoided Mortality

Final Rules

Alternative 1

Alternative 22

(Pope et al., 2019) (adult
mortality ages 18-99 years)

530
(380 to 670)

530
(380 to 680)

530
(380 to 680)

(Wu et al., 2020) (adult mortality
ages 65-99 years)

270
(240 to 300)

270
(240 to 300)

270
(240 to 300)

(Woodruff et al., 2008) (infant
mortality)

0.47
(-0.30 to 1.2)

0.47
(-0.30 to 1.2)

0.47
(-0.30 to 1.2)

Avoided Morbidity

Hospital admissions—
cardiovascular (age >18)

39

(28 to 49)

39

(28 to 49)

39

(28 to 49)

Hospital admissions—respiraton

24

(8.0 to 39)

24

(8.0 to 39)

24

(8.0 to 39)

ED visits-cardiovascular

78

(-30 to 180)

79

(-30 to 180)

79

(-30 to 180)

ED visits—respiratory

150
(29 to 300)

150
(29 to 310)

150
(29 to 300)

Acute Myocardial Infarction

8.7
(5.0 to 12)

8.8
(5.1 to 12)

8.7
(5.1 to 12)

Cardiac arrest

3.7

(-1.5 to 8.4)

3.7
(-1.5 to 8.5)

3.7

(-1.5 to 8.4)

Hospital admissions-
Alzheimer's Disease

150
(110 to 190)

150
(110 to 190)

150
(110 to 190)

Hospital admissions-
Parkinson's Disease

16

(8.2 to 24)

16

(8.3 to 24)

16

(8.3 to 24)

Stroke

15

(3.8 to 25)

15

(3.9 to 26)

15

(3.9 to 26)

Lung cancer

19

(5.7 to 31)

19

(5.7 to 31)

19

(5.7 to 31)

Hay Fever/Rhinitis

3,500
(840 to 6,000)

3,500
(840 to 6,100)

3,500
(840 to 6,000)

Asthma Onset

530
(510 to 550)

540
(510 to 560)

530
(510 to 550)

Asthma symptoms - Albuterol
use

100,000
(-49,000 to 250,000)

100,000
(-50,000 to 250,000)

100,000
(-50,000 to 250,000)

Lost work days

27,000
(23,000 to 31,000)

27,000
(23,000 to 31,000)

27,000
(23,000 to 31,000)

Minor restricted-activity daysdf

160,000
(130,000 to 190,000)

160,000
(130,000 to 190,000)

160,000
(130,000 to 190,000)

Note: Values rounded to two significant figures.

4-58


-------
Table 4-15 Estimated Discounted Economic Value of Avoided Ozone and PM2.5-
Attributable Premature Mortality and Illness for the Illustrative Scenarios in 2028 (95
percent confidence interval; billions of 2019 dollars)"'*1	

Disc.
Rate

Pollutant

Ozone Benefits

PM Benefits

Ozone plus PM Benefits

2% Final
Rules
i Alternative

$0.19 and

$0.81

$2.4

and

$5.0

$2.6b

and

$5i

$0.17

and

$0.75

$2.2

and

$4.4

$2.3b

and

$5.2C

Alternative
2

Final
Rules

$0.16 and

$0.71

$0.18
($0.68 to
$0.32)

and

$0.78
($0.12 to
$1.9)

$2.0

and

$4.1

$2.3
($0.32
to $6.0)

and

$4.8
($0.54 to
$13)

$2.1b

and

$4i

$2.5b
($0.38
to $6.3)

and

$5.6C
($0.66 to
$15)

Alternative
1

$0.16
($0,061
to $0.29)

and

$0.72
($0.11 to
$1.8)

$2.1
($0.28
to $5.4)

and

$4.3
($0.48 to
$11)

$2.3b
($0.34
to $5.7)

and

$5.0C
($0.59 to
$13)

Alternative
2

$0.15
($0.58 to
$0.28)

and

$0.69
($0.11 to
$1.7)

$1.9
($0.26
to $4.9)

and

$4.0
($0.44 to
$11)

$2.1b
($0.32
to $5.2)

and

$4.7C
($0.55 to
$12)

7%

Final
Rules

$0.13
($0,036
to $0.24)

and

$0.66
($0,085 to
	$1,7)	

$2.1
($0.26
to $5.4)

and

$4.3
($0.46 to
$12)

$2.2b
($0.29
to $5.6)

and

$2.0"
($0.26
to $5.0)

$5.0C
($0.54 to
$13)

Alternative
1

$0.11
($0,033
to $0.22)

and

$0.61
($0,078 to
$1.6)

$1.9
($0.23
to $4.8)

and

$3.9
($0.41 to
$10)

and

$4.5C
($0.49 to
$12)

Alternative
2

$0.11
($0,031
to $0.21)

and

$0.59
($0,075 to
$1.5)

$1.7
($0.21
to $4.4)

and

$3.5
($0.37 to
$9.5)

$1.8b
($0.24
to $4.6)

and

$4.1C
($0.45 to
$11)

3 Values rounded to two significant figures. The two benefits estimates are separated by the word "and" to signify
that they are two separate estimates. The estimates do not represent lower- and upper-bound estimates and should
not be summed.

b Sum of ozone mortality estimated using the pooled short-term ozone exposure risk estimate and the Wu et al.
(2020) long-term PM2 5 exposure mortality risk estimate.

c Sum of the Turner et al. (2016) long-term ozone exposure risk estimate and the Pope et al. (2019) long-term PM2
exposure mortality risk estimate.

d EPA is unable to provide confidence intervals for 2 percent-based estimates currently.

4-59


-------
Table 4-16 Estimated Discounted Economic Value of Avoided Ozone and PM2.5-
Attributable Premature Mortality and Illness for the Illustrative Scenarios in 2030 (95

percent confidence interval; billions of 201

19 dollars)3'11

Disc.
Rate

Pollutant

Ozone Benefits

PM Benefits

Ozone plus PM Benefits

2%

Final Rules

$0.19 and $0.82

$1.6 and $3.2

$1.8b and $4.0C

Alternative
1

$0.19 and $0.79

$1.4 and $2.8

$1.5b and $3.6C

Alternative ; S() |4 m({ S() 63 $!.<) (,nd $2.1 S 1.2 and $2.8C

3%

Final Rules

so IS so "<>

($0,069 and ($0.12 to
to 0.33) $2.0)

$1.5 $3.1
($0.20 to and ($0.35 to
$3.9) $8.3)

$1.7b $3.9C
($0.27 to and ($0.47 to
$4.3) $10)

Alternative
1

$0.18 $0.77
($0,068 and ($0.12 to
to 0.32) $1.9)

$1.3 $2.7
($0.18 to and ($0.30 to
$3.4) $7.1)

$1.5b $3.4C
($0.24 to and ($0.42 to
$3.7) $9.0)

Alternative
2

$014

($0t052 and ($0,095
$0.25) t0 $L5)

$1.0 $2.1
($0.14 to and ($0.23 to
$2.6) $5.5)

$l.lb $2.7C
($0.19 to and ($0.33 to
$2.8) $7.1)

7%

Final Rules

$0.13 $0.67
($0,037 and ($0,086
to 0.25) to $1.7)

$1.4 $2.8
($0.17 to and ($0.29 to
$3.5) $7.4)

$1.5b $3.5C
($0.20 to and ($0.38 to
$3.8) $9.2)

Alternative
1

$013 tO 65
($0t036 and ($0,084

$0.24) t0$L7>

$1.2 $2.4
($0.14 to and ($0.25 to
$3.0) $6.4)

$1.3b $3.0C
($0.18 to and ($0.34 to
$3.2) $8.0)

Alternative
2

$0,097

($0t028 and ($0,066
$0.19) to $1.3)

$0.89 $1.9
($0.11 to and ($0.20 to
$2.3) $5.0)

$0.99b $2.4C
($0.14 to and ($0.26 to
$2.5) $6.3)

3 Values rounded to two significant figures. The two benefits estimates are separated by the word "and" to signify
that they are two separate estimates. The estimates do not represent lower- and upper-bound estimates and should not
be summed.

b Sum of ozone mortality estimated using the pooled short-term ozone exposure risk estimate and the Wu et al.
(2020) long-term PM2 5 exposure mortality risk estimate.

c Sum of the Turner et al. (2016) long-term ozone exposure risk estimate and the Pope et al. (2019) long-term PM2 5
exposure mortality risk estimate.

d EPA is unable to provide confidence intervals for 2 percent-based estimates currently.

4-60


-------
Table 4-17 Estimated Discounted Economic Value of Avoided Ozone and PM2.5-
Attributable Premature Mortality and Illness for the Illustrative Scenarios in 2035 (95

percent confidence interval; billions of 201

9 dollars)

i,d









Disc.
Rate

Pollutant

Ozone Benefits

PM Benefits

Ozone plus PM Benefits

2%

Final Rules

$0.29

and

$1.6

$6.6

and

$13

$6.9b

and

$15c



Alternative
1

$0.34

and

$1.8

$6.8

and

$13

$7.1b

and

$15c



Alternative
2

$0.29

and

$1.6

$6.8

and

$14

$7.1b

and

$15c

3%

Final Rules

$0.28
($0,096
to 0.52)

and

$1.5
($0.21 to
$3.9)

S(i 4

($0.82 to
Slo

and

$13
($1.4 to
$34)

$6.7b
($0.91 to
$17)

and

$14c
($1.6 to
$38)



Alternative
1

$0.33
($0.12

to
$0.61)

and

$1.8
($0.25 to
$4.5)

$6.6
($0.84 to
$17)

and

$13
($1.4 to
$35)

$6.9b
($0.96 to
$18)

and

$15c
($1.7 to
$39)



Alternative
2

$0.28
($0,098
to 0.53)

and

$1.6
($0.21 to
$4.0)

S(| (¦

($0.85 to
$17)

and

$13
($1.4 to
$35)

$6.9b
($0.95 to
$18)

and

$15c
($1.6 to
$39)

7%

Final Rules

$0.20
($0,053
to 0.41)

and

$1.3
($0.16 to
$3.5)

$5.7
($0.68 to
Sl^i

and

$11
($1.2 to
$30)

$5.9b
($0.73 to
$15)

and

$13c
($1.3 to
$34)



Alternative
1

$0.24
($0,063

to
$0.49)

and

$1.5
($0.18 to
$4.0)

$5.9
($0.70 to
$15)

and

$12
($1.2 to
$31)

$6.1b
($0.76 to
$16)

and

$13c
($1.4 to
$35)



Alternative
2

$0.20
($0,054

to
$0.42)

and

$1.4
($0.16 to
$3.5)

$5.9
($0.70 to
$15)

and

$12
($1.2 to
$31)

$6.1b
($0.76 to
$16)

and

$13c
($1.4 to
$35)

3 Values rounded to two significant figures. The two benefits estimates are separated by the word "and" to signify
that they are two separate estimates. The estimates do not represent lower- and upper-bound estimates and should not
be summed.

b Sum of ozone mortality estimated using the pooled short-term ozone exposure risk estimate and the Wu et al.
(2020) long-term PM2 5 exposure mortality risk estimate.

c Sum of the Turner et al. (2016) long-term ozone exposure risk estimate and the Pope et al. (2019) long-term PM2 5
exposure mortality risk estimate.

d EPA is unable to provide confidence intervals for 2 percent-based estimates currently.

4-61


-------
Table 4-18 Estimated Discounted Economic Value of Avoided Ozone and PM2.5-
Attributable Premature Mortality and Illness for the Illustrative Scenarios in 2040 (95
percent confidence interval; billions of 2019 dollars)"'*1	

Disc.
Rate

Pollutant

Ozone Benefits

PM Benefits

Ozone plus PM Benefits

2%

Final Rules

-$0,093 and -$0,016

-$0.26 and -$0.13

-$0.35b and -$0.14c

Alternative
1

-$0,011 and $0.00033

$0,012 and $0,012

$0-°b043 and $0.012c

Alternative ; _$Q n ^ -$0,023 i $0.11 and $0.21 i $0.087b and $0.09lc

3%

Final Rules

-$0 090 -$0,016
(-$0 23 ,0 and <-$0 03°
¦$0012> -$OM53)

-$0.25

("$0-67 and (-$0.32 to

$0,026) -$0°15)

-$0 34b

* ' _«o 14c
("$0-90 and (-$0.35 to

$0,038) -$0-02°)

Alternative
1

¦$0()" $0 0003
<"$»032 and (-$0013

$0.00027) t0 $-0013)

$0.0 u s

and ($0.0012
$0,029) t0 $0 °29)

$0.00045 $0.012c

b(- (-

fl fid

$0,030 to $0.000077
$0,030) to $0,030)

Alternative
2

. -$0,022
(-$0.29 to and ^°41
-$0016) -$0,078)

$0." $020

($0t014 and ($0,022 to
$0.29) $°-53)

$0.085b $0.088c
(-$0.27 and (-$0,027 tc
to $0.52) $0.28)

7%

Final Rules

-HO 078 ¦$0 011
(-$0.20 to and (~$°024
-$0.0089) -$0.0029)

-$0.22

("$0-60 and (-$0.28 to
$0,023) -$0-°13)

-$0.30b

(-$0.80 -$0.12c

to- and (-$0.31 to-
$0,032) $0,016)

Alternative
1

-$0,010

(-$0,029 $0.000078
to - and (-$0.0016
$0.000089 to
) $0.00081)

$0,094 ^

^U;UU and ($0.00049
$0,025) t0 $0 °26)

$0.00065 $0.0094c
b(- and (-$0.0011
$0,028 to to $0,026)
$0,026)

Alternative
2

-$0,099 -$0,016
(-$0.26 to and ("$^032
-$0012) -$0.0043)

$0,098

($0 011 and ($0,018 to

$0.26) $048)

$0.079b $0.082c
( $0.24 and (-$0,021 tc
10 $0.47) $0.25)

3 Values rounded to two significant figures. The two benefits estimates are separated by ilie word "and" to signify
that they are two separate estimates. The estimates do not represent lower- and upper-bound estimates and should not
be summed.

b Sum of ozone mortality estimated using the pooled short-term ozone exposure risk estimate and the Wu et al.
(2020) long-term PM2 5 exposure mortality risk estimate.

c Sum of the Turner et al. (2016) long-term ozone exposure risk estimate and the Pope et al. (2019) long-term PM2 5
exposure mortality risk estimate.

d EPA is unable to provide confidence intervals for 2 percent-based estimates currently.

4-62


-------
Table 4-19 Estimated Discounted Economic Value of Avoided Ozone and PM2.5-
Attributable Premature Mortality and Illness for the Illustrative Scenarios in 2045 (95

percent confidence interval; billions of 2

019 dollars

a,d









Disc.
Rate

Pollutant

Ozone Benefits

PM Benefits

Ozone plus PM Benefits

2%

Final Rules

$0.34

and

$1.8

$3.3

and

$6.4

$3.6b

and

$8.2C



Alternative
1

$0.34

and

$1.8

$3.3

and

$6.5

$3.7b

and

$8.2C



Alternative
2

$0.34

and

$1.8

$3.3

and

$6.4

$3.7b

and

$8.2C

3%

Final Rules

$0.33
($0.12

to
$0.60)

and

$1.7
($0.24
to $4.4)

$3.2
($0.40 to
$8.3)

and

$6.3
($0.67 to
$17)

$3.5b
($0.52 to
$8.9)

and

$7.9C
($0.91 to
$21)



Alternative
1

$0.33
($0.12

to
$0.61)

and

$1.7
($0.24
to $4.4)

$3.2
($0.40 to
$8.4)

and

$6.3
($0.67 to
$17)

$3.6b
($0.52 to
$9.0)

and

$8.0C
($0.92 to
$21)



Alternative
2

$0.33
($0.12

to
$0.61)

and

$1,700
($0.24
to $4.4)

$3.2
($0.40 to
$8.3)

and

$6.3
($0.67 to
$17)

$3.6b
($0.52 to
$8.9)

and

$8.0C
($0.91 to
$21)

7%

Final Rules

$0.24
($0,063

to
$0.48)

and

$1.5
($0.18
to $3.9)

$2.9
($0.33 to
$7.4)

and

$5.5
($0.57 to
$15)

$3.1b
($0.40 to
$7.9)

and

$7.0C
($0.75 to
$19)



Alternative
1

$0.24
($0,063

to
$0.48)

and

$1.5
($0.18
to $3.9)

$2.9
($0.34 to
$7.5)

and

$5.6
($0.58 to
$15)

$3.1b
($0.40 to
$7.9)

and

$7.r
($0.76 to
$19)



Alternative
2

$0.24
($0,063

to
$0.48)

and

$1.5
($0.18
to $3.9)

$2.9
($0.34 to
$7.5)

and

$5.6
($0.58 to
$15)

$3.1b
($0.40 to
$7.9)

and

$7.r
($0.75 to
$19)

3 Values rounded to two significant figures. The two benefits estimates are separated by the word "and" to signify
that they are two separate estimates. The estimates do not represent lower- and upper-bound estimates and should
not be summed.

b Sum of ozone mortality estimated using the pooled short-term ozone exposure risk estimate and the Wu et al.
(2020) long-term PM2 5 exposure mortality risk estimate.

c Sum of the Turner et al. (2016) long-term ozone exposure risk estimate and the Pope et al. (2019) long-term PM2 5
exposure mortality risk estimate.

d EPA is unable to provide confidence intervals for 2 percent-based estimates currently.

4-63


-------
Table 4-20 Estimated Discounted Economic Value of Avoided Ozone and PM2.5-
Attributable Premature Mortality and Illness for the Illustrative Scenarios in 2028, 2030,

2035

, 2040 and 2045 (bil

ions of 2019 dollars)a'b





2% Discount Rate

3% Discount Rate

7% Discount Rate





Ozone and PM Benefits

Ozone and PM Benefits

Ozone and PM Benefits

2028

Final Rules

$5.8

$5.6

$5.0



Alternative 1

$5.2

$5.0

$4.5



Alternative 2

$4.8

$4.7

$4.1



Final Rules

$4.0

$3.9

$3.5

2030

Alternative 1

$3.6

$3.4

$3.0



Alternative 2

$2.8

$2.7

$2.4



Final Rules

$15

$14

$13

2035

Alternative 1

$15

$15

$13



Alternative 2

$15

$15

$13



Final Rules

-$0.35

-$0.34

-$0.30

2040

Alternative 1

$0.00043

$0.00045

-$0.00065



Alternative 2

$0,087

$0,085

$0,079



Final Rules

$8.2

$7.9

$7.0

2045

Alternative 1

$8.2

$8.0

$7.1



Alternative 2

$8.2

$8.0

$7.1

3 Values rounded to two significant figures. The two benefits estimates are separated by the word "and" to signify
that they are two separate estimates. The estimates do not represent lower- and upper-bound estimates and should
not be summed.

b Values are the monetized benefits of the mortality and illnesses included in Tables 4-5 through 4-14.

4-64


-------
Table 4-21 Stream of Human Health Benefits from 2028 through 2047: Monetized
Benefits Quantified as Sum of Long-Term Ozone Mortality and Illness and Long-
Term PM2.5 Mortality and Illness for EGUs (discounted at 2 percent; billions of 2019
dollars)"	



Final Rules

Alternative 1

Alternative 2

2028*

$5.8

$5.2

$4.8

2029

$5.9

$5.3

$4.9

2030*

$4.0

$3.6

$2.8

203 1

$4.1

$3.6

$2.8

2032

$14

$14

$14

2033

$14

$15

$15

2034

$14

$15

$15

2035*

$15

$15

$15

2036

$15

$16

$15

2037

$15

$16

$16

2038

-$0.34

$0.0026

$0,089

2039

-$0.34

$0.0016

$0,088

2040*

-$0.35

$0.00043

$0,087

204 lb

-$0.36

-$0.0011

$0,085

2042

$7.9

$8.0

$7.9

2043

$8.0

$8.1

$8.0

2044

$8.1

$8.1

$8.1

2045*

$8.2

$8.2

$8.2

2046

$8.3

$8.3

$8.3

2047

$8.3

$8.4

$8.4

PV

$120

$120

$120

EAV

$6.3

$6.5

$6.3

*Year in which air quality models were run. Benefits for all other years were extrapolated from years with
model-based air quality estimates. Benefits calculated as value of avoided: PM2 5-attributable deaths (quantified
using a concentration-response relationship from the Pope et al. 2019 study); Ozone-attributable deaths
(quantified using a concentration-response relationship from the Turner et al. 2016 study); and PM2 5 and ozone-
related morbidity effects.

3 For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with
several point estimates.

b The 2040 analysis year is applied to the years 2038-2041. As the population grows each year, the ozone
disbenefits in the 2040 analysis run are applied to a larger population each year leading to declining benefits
during this period.

4-65


-------
Table 4-22 Stream of Human Health Benefits from 2028 through 2047: Monetized
Benefits Quantified as Sum of Long-Term Ozone Mortality and Illness and Long-Term
PM2.5 Mortality and Illness for EGUs (discounted at 3 percent; billions of 2019 dollars)"



Final Rules

Alternative 1

Alternative 2

2028*

$5.6

$5.0

$4.7

2029

$5.8

$5.2

$4.8

2030*

$3.9

$3.4

$2.7

203 1

$4.0

$3.5

$2.8

2032

$13

$14

$14

2033

$14

$14

$14

2034

$14

$15

$14

2035*

$14

$15

$15

2036

$15

$15

$15

2037

$15

$15

$15

2038

-$0.33

$0.0025

$0,087

2039

-$0.33

$0.0016

$0,086

2040*

-$0.34

$0.00045

$0,085

204 lb

-$0.35

-$0.00097

$0,083

2042

$7.7

$7.7

$7.7

2043

$7.7

$7.8

$7.8

2044

$7.8

$7.9

$7.9

2045*

$7.9

$8.0

$8.0

2046

$8.0

$8.1

$8.0

2047

$8.1

$8.1

$8.1

PV

$100

$110

$100

EAV

$6.1

$6.2

$6.1

*Year in which air quality models were run. Benefits for all other years were extrapolated from years with model-
based air quality estimates. Benefits calculated as value of avoided: PM2 5-attributable deaths (quantified using a
concentration-response relationship from the Pope et al. 2019 study); Ozone-attributable deaths (quantified using a
concentration-response relationship from the Turner et al. 2016 study); and PM25 and ozone-related morbidity
effects.

3 For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates.

b The 2040 analysis year is applied to the years 2038-2041. As the population grows each year, the ozone
disbenefits in the 2040 analysis run are applied to a larger population each year leading to declining benefits during
this period.

4-66


-------
Table 4-23 Stream of Human Health Benefits from 2028 through 2047: Monetized
Benefits Quantified as Sum of Long-Term Ozone Mortality and Illness and Long-Term
PM2.5 Mortality and Illness for EGUs (discounted at 7 percent; billions of 2019 dollars)"



Final Rules

Alternative 1

Alternative 2

2028*

$5.0

$4.5

$4.1

2029

$5.1

$4.6

$4.2

2030*

$3.5

$3.0

$2.4

203 1

$3.5

$3.1

$2.4

2032

$12

$12

$12

2033

$12

$13

$13

2034

$12

$13

$13

2035*

$13

$13

$13

2036

$13

$13

$13

2037

$13

$14

$14

2038

$-0.29

$0.0013

$0,081

2039

$-0.3

$0.00044

$0.08

2040*

$-0.3

$-0.00065

$0,079

204 lb

$-0.31

$-0.0019

$0,077

2042

$6.8

$6.9

$6.8

2043

$6.9

$6.9

$6.9

2044

$6.9

$7.0

$7.0

2045

$7.0

$7.1

$7.1

2046

$7.1

$7.2

$7.1

2047

$7.2

$7.2

$7.2

PV

$59

$60

$58

EAV

$5.2

$5.2

$5.1

*Year in which air quality models were run. Benefits for all other years were extrapolated from years with model-
based air quality estimates. Benefits calculated as value of avoided: PM2 5-attributable deaths (quantified using a
concentration-response relationship from the Pope et al. 2019 study); Ozone-attributable deaths (quantified using a
concentration-response relationship from the Turner et al. 2016 study); and PM25 and ozone-related morbidity
effects.

3 For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates.

b The 2040 analysis year is applied to the years 2038-2041. As the population grows each year, the ozone
disbenefits in the 2040 analysis run are applied to a larger population each year leading to declining benefits during
this period.

4.4 Additional Unquantified Benefits

Data, time, and resource limitations prevented EPA from quantifying the estimated health
impacts or monetizing estimated benefits associated with incremental changes in direct exposure
to NO2 and SO2, independent of the role NO2 and SO2 play as precursors to PM2.5 and ozone, as
well as ecosystem effects, and visibility impairment that might result from emissions changes

4-67


-------
associated with compliance with the final requirements. While all health benefits and welfare
benefits were not quantified, it does not imply that there are not additional benefits associated
with reductions in human exposures to NO2 or SO2 and ecosystem exposure to air pollutants
potentially resulting from emissions changes under this rule. In this section, we provide a
qualitative description of these and water quality benefits, which are listed in Table 4-24. Note
also that some pollutants from U.S. EGUs, such as NO2, SO2, and particulate matter, can be
transported downwind into foreign countries, in particular Canada and Mexico. Therefore,
reduced pollution from U.S. EGUs can lead to public health and welfare benefits in foreign
countries. EPA is currently unable to quantify or monetize these effects.

Table 4-24 Unquantified Health and Welfare Benefits Categories

Category

Effect

Effect
Quantified

Effect
Monetized

More
Information

Improved Human Health



Asthma hospital admissions

—

—

NO2 ISA1



Chronic lung disease hospital admissions

—

—

NO2 ISA1



Respiratory emergency department visits

—

—

NChISA1

Reduced incidence of

Asthma exacerbation

—

—

NO2 ISA1

morbidity from exposure
to NO2

Acute respiratory symptoms

—

—

NO2 ISA1

Premature mortality

—

—

NO2 ISA1-2-3



Other respiratory effects (e.g., airway
hyperresponsiveness and inflammation, lung
function, other ages and populations)

—

—

NO2 ISA2-3

Reduced incidence of
mortality and morbidity
through drinking water
from reduced effluent
discharges.

Bladder, colon, and rectal cancer from
halogenated disinfection byproducts
exposure.

—

—

SEELGBCA4

Reproductive and developmental effects
from halogenated disinfection byproducts
exposure.

—

—

SEELGBCA4



Neurological and cognitive effects to
children from lead exposure from fish
consumption (including need for specialized
education).

—

—

SEELGBCA4



Possible cardiovascular disease from lead
exposure

—

—

SEELGBCA4

Reduced incidence of
morbidity and mortality
from toxics through fish
consumption from reduced
effluent discharges.

Neurological and cognitive effects from in
in-utero mercury exposure from maternal
fish consumption

—

—

SEELGBCA4

Skin and gastrointestinal cancer incidence
from arsenic exposure

—

—

SEELGBCA4

Cancer and non-cancer incidence from
exposure to toxic pollutants (lead, cadmium,
thallium, hexavalent chromium etc.

—	— SEELGBCA4

Neurological, alopecia, gastrointestinal
effects, reproductive and developmental
damage from short-term thallium exposure.

4-68


-------
Reduced incidence of

morbidity and mortality
from recreational water
exposure from reduced
effluent discharges.

Cancer and Non-Cancer incidence from
exposure to toxic pollutants (methyl-
mercury, selenium, and thallium.)

—

—

SEELGBCA4

Improved Environment

Reduced visibility

Visibility in Class 1 areas

—

—

PMISA1

impairment

Visibility in residential areas

—

—

PMISA1

Reduced effects on
materials

Household soiling

—

—

PMISA1'2

Materials damage (e.g., corrosion, increased
wear)

—

—

PMISA2

Reduced effects from PM
deposition (metals and
organics)

Effects on individual organisms and
ecosystems

—

—

PMISA2



Visible foliar injury on vegetation

—

—

Ozone ISA1



Reduced vegetation growth and reproduction

—

—

Ozone ISA1



Yield and quality of commercial forest
products and crops

—

—

Ozone ISA1



Damage to urban ornamental plants

—

—

Ozone ISA2

Reduced vegetation and
ecosystem effects from

Carbon sequestration in terrestrial
ecosystems

—

—

Ozone ISA1

exposure to ozone

Recreational demand associated with forest
aesthetics

—

—

Ozone ISA2



Other non-use effects





Ozone ISA2



Ecosystem functions (e.g., water cycling,
biogeochemical cycles, net primary
productivity, leaf-gas exchange, community
composition)

—

—

Ozone ISA2



Recreational fishing

—

—

NOx SOxISA1



Tree mortality and decline

—

—

NOx SOxISA2

Reduced effects from acid
deposition

Commercial fishing and forestry effects

—

—

NOx SOxISA2

Recreational demand in terrestrial and
aquatic ecosystems

—

—

NOx SOxISA2



Other non-use effects





NOx SOxISA2



Ecosystem functions (e.g., biogeochemical
cycles)

—

—

NOx SOxISA2



Species composition and biodiversity in
terrestrial and estuarine ecosystems

—

—

NOx SOxISA2

Reduced effects from
nutrient enrichment from
deposition.

Coastal eutrophication

—

—

NOx SOxISA2

Recreational demand in terrestrial and
estuarine ecosystems

—

—

NOx SOxISA2

Other non-use effects





NOx SOxISA2



Ecosystem functions (e.g., biogeochemical
cycles, fire regulation)

—

—

NOx SOxISA2

Reduced vegetation effects
from ambient exposure to
SO2 andNOx

Injury to vegetation from SO2 exposure

—

—

NOx SOxISA2

Injury to vegetation from NOx exposure

—

—

NOx SOxISA2

Improved water aesthetics
from reduced effluent

Improvements in water clarity, color, odor in
residential, commercial and recreational





SEELGBCA4

discharges.

settings.







4-69


-------


Protection of Threatened and Endangered
(T&E) species from changes in habitat and
potential population effects.

—

—

SE ELG BCA4

Effects on aquatic
organisms and other
wildlife from reduced
effluent discharges

Other non-use effects

—

—

SE ELG BCA4

Changes in sediment contamination on
benthic communities and potential for re-
entrainment.

—

—

SE ELG BCA4

Quality of recreational fishing and other
recreational use values.

—

—

SE ELG BCA4



Commercial fishing yields and harvest
quality.

—

—

SE ELG BCA4

Reduced water treatment
costs from reduced
effluent discharges

Reduced drinking, irrigation, and other
agricultural use water treatment costs.

—

—

SE ELG BCA4



Increased storage availability in reservoirs

—

—

SE ELG BCA4

Reduced sedimentation
from effluent discharges

Improved functionality of navigable
waterways

—

—

SE ELG BCA4



Decreased cost of dredging

—

—

SE ELG BCA4

Benefits of reduced water
withdrawal

Benefits from effects aquatic and riparian
species from additional water availability.

—

—

SE ELG BCA4

Increased water availability in reservoirs
increasing hydropower supply, recreation,
and other services.

—

—

SE ELG BCA4

1	We assess these benefits qualitatively due to data and resource limitations for this RIA.

2	We assess these benefits qualitatively because we do not have sufficient confidence in available data or methods.

3	We assess these benefits qualitatively because current evidence is only suggestive of causality or there are other significant
concerns over the strength of the association.

4	Benefit and Cost Analysis (BCA) for Revisions to the Effluent Limitations Guidelines (ELG) and Standards for the Steam
Electric (SE) Power Generating Point Source Category.

4.4.1 Hazardous Air Pollutant Impacts

4.4.1.1 Mercury Air Pollutant Impacts

The final rules are expected to reduce fossil fuel-fired EGU generation and
consequentially is expected to lead to reduced HAP emissions. HAP emitted from EGUs can
cause premature mortality from heart attacks, cancer, and neurodevelopmental delays in children,
and detrimentally affect economically vital ecosystems used for recreational and commercial
purposes. Further, these public health effects have been particularly pronounced for certain
segments of the American population that are especially vulnerable (e.g., subsistence fishers and
their children) to impacts from EGU HAP emissions.

The final rules are expected to reduce emissions of mercury. Mercury is a persistent,
bioaccumulative toxic metal that is emitted from power plants in three forms: gaseous elemental
mercury (HgO), oxidized mercury compounds (Hg+2), and particle-bound mercury (HgP).
Elemental mercury does not quickly deposit or chemically react in the atmosphere, resulting in

4-70


-------
residence times that are long enough to contribute to global scale deposition. Oxidized mercury
and HgP deposit quickly from the atmosphere impacting local and regional areas in proximity to
sources. MeHg is formed by microbial action in the top layers of sediment and soils, after
mercury has precipitated from the air and deposited into waterbodies or land. Once formed,
MeHg is taken up by aquatic organisms and bioaccumulates up the aquatic food web. Larger
predatory fish may have MeHg concentrations many times, typically on the order of one million
times, that of the concentrations in the freshwater body in which they live. MeHg can adversely
impact ecosystems and wildlife. The projected reductions in mercury are expected to reduce the
bioconcentration of MeHg in fish. Subsistence fishing is associated with vulnerable populations,
including minorities and those of low socioeconomic status. Further reductions in mercury
emissions from lignite-fired facilities could help address exposure inequities for the subsistence
fisher sub-population.

Human exposure to MeHg is known to have several adverse neurodevelopmental
impacts, such as IQ loss measured by performance on neurobehavioral tests, particularly on tests
of attention, fine motor-function, language, and visual spatial ability. In addition, evidence in
humans and animals suggests that MeHg can have adverse effects on both the developing and the
adult cardiovascular system, including fatal and non-fatal ischemic heart disease (IHD). Further,
nephrotoxicity, immunotoxicity, reproductive effects (impaired fertility), and developmental
effects have been observed with MeHg exposure in animal studies disease (ATSDR, 2022).
MeHg has some genotoxic activity and is capable of causing chromosomal damage in a number
of experimental systems. EPA has classified MeHg as a "possible" human carcinogen.

4.4.1.2 Metal HAP

The projected reductions in emissions of non-mercury metal HAP are expected to reduce
exposure to carcinogens, such as nickel, arsenic, and hexavalent chromium, in the surrounding
areas. U.S. EGUs are the largest source of selenium (Se) emissions and a major source of
metallic HAP emissions including arsenic (As), chromium (Cr), nickel (Ni), and cobalt (Co).
Additionally, U.S. EGUs emit cadmium (Cd), beryllium (Be), lead (Pb), and manganese (Mn).
These emissions include metal HAPs that are persistent and bioaccumulative (Cd, As, and Pb)
and others have the potential to cause cancer (Ni, Cr, Cd, Be, Co, and Pb). PM controls are

4-71


-------
expected to reduce metal HAP emissions and therefore reduce the potential for adverse effects
from metal HAP exposure.

Exposure to these metal HAP, depending on exposure duration and levels of exposures, is
associated with a variety of adverse health effects. These adverse health effects may include
chronic health disorders (e.g., irritation of the lung, skin, and mucus membranes; decreased
pulmonary function, pneumonia, or lung damage; detrimental effects on the central nervous
system; damage to the kidneys; and alimentary effects such as nausea and vomiting). As of 2023,
three of the key metal HAP emitted by EGUs (As, Cr, and Ni) have been classified as human
carcinogens, while two others (Cd, and Se) are classified as probable human carcinogens.

4.4.2	NO2 Health Benefits

In addition to being a precursor to PM2.5 and ozone, NOx emissions are also linked to a
variety of adverse health effects associated with direct exposure. This analysis only quantifies
and monetizes the ozone PM2.5 benefits associated with the reductions in NOx emissions and
does not quantify the impacts of changing direct exposure to NO2. Following a comprehensive
review of health evidence from epidemiologic and laboratory studies, the Integrated Science
Assessment for Oxides of Nitrogen —Health Criteria (NOx ISA) concluded that there is a likely
causal relationship between respiratory health effects and short-term exposure to NO2 (U.S. EPA,
2016a). These epidemiologic and experimental studies encompass a number of endpoints
including emergency department visits and hospitalizations, respiratory symptoms, airway
hyperresponsiveness, airway inflammation, and lung function. The NOx ISA also concluded that
the relationship between short-term NO2 exposure and premature mortality was "suggestive but
not sufficient to infer a causal relationship," because it is difficult to attribute the mortality risk
effects to NO2 alone. Although the NOx ISA stated that studies consistently reported a
relationship between NO2 exposure and mortality, the effect was generally smaller than that for
other pollutants such as PM.

4.4.3	SO2 Health Benefits

In addition to being a precursor to PM2.5, SO2 emissions are also linked to a variety of
adverse health effects associated with direct exposure. This analysis only quantifies and
monetizes the PM2.5 benefits associated with the reductions in SO2 emissions and does not

4-72


-------
quantify the impacts of changing direct exposure to SO2. Following an extensive evaluation of
health evidence from epidemiologic and laboratory studies, the Integrated Science Assessment
for Oxides of Sulfur —Health Criteria (SO2 ISA) ISA concluded that there is a causal
relationship between respiratory health effects and short-term exposure to SO2 (U.S. EPA, 2017).
The immediate effect of SO2 on the respiratory system in humans is bronchoconstriction.
Asthmatics are more sensitive to the effects of SO2 likely resulting from pre-existing
inflammation associated with this disease. A clear concentration-response relationship has been
demonstrated in laboratory studies following exposures to SO2 at concentrations between 20 and
100 ppb, both in terms of increasing severity of effect and percentage of asthmatics adversely
affected. Based on our review of this information, we identified three short-term morbidity
endpoints that the SO2 ISA identified as a "causal relationship": asthma exacerbation,
respiratory-related emergency department visits, and respiratory-related hospitalizations. The
differing evidence and associated strength of the evidence for these different effects is described
in detail in the SO2 ISA. The SO2 ISA also concluded that the relationship between short-term
SO2 exposure and premature mortality was "suggestive of a causal relationship" because it is
difficult to attribute the mortality risk effects to SO2 alone. Although the SO2 ISA stated that
studies are generally consistent in reporting a relationship between SO2 exposure and mortality,
there was a lack of robustness of the observed associations to adjustment for other pollutants.

4.4.4 Ozone Welfare Benefits

Exposure to ozone has been associated with a wide array of vegetation and ecosystem
effects in the published literature (U.S. EPA, 2020d). Sensitivity to ozone is highly variable
across species, with over 65 plant species identified as "ozone-sensitive", many of which occur
in state and national parks and forests. These effects include those that damage or impair the
intended use of the plant or ecosystem. Such effects can include reduced growth and/or biomass
production in sensitive plant species, including forest trees, reduced yield and quality of crops,
visible foliar injury, species composition shift, and changes in ecosystems and associated
ecosystem services. See Section F of the Ozone Transport Policy Analysis Proposed Rule TSD
(U.S. EPA, 2022g) for a summary of an assessment of risk of ozone-related growth impacts on
selected forest tree species.

4-73


-------
4.4.5 NO2 and SO2 Welfare Benefits

As described in the Integrated Science Assessment (ISA) for Oxides of Nitrogen, Oxides
of Sulfur and Particulate Matter Ecological Criteria (NOx/SOx/PM ISA), NOx and SO2
emissions also contribute to a variety of adverse welfare effects, including those associated with
acidic deposition, visibility impairment, and nutrient enrichment (U.S. EPA, 2020c). Deposition
of nitrogen and sulfur causes acidification, which can cause a loss of biodiversity of fishes,
zooplankton, and macro invertebrates in aquatic ecosystems, as well as a decline in sensitive tree
species, such as red spruce (Picea rubens) and sugar maple (Acer saccharum) in terrestrial
ecosystems. In the northeastern U.S., the surface waters affected by acidification are a source of
food for some recreational and subsistence fishermen and for other consumers and support
several cultural services, including aesthetic and educational services and recreational fishing.
Biological effects of acidification in terrestrial ecosystems are generally linked to aluminum
toxicity, which can cause reduced root growth, restricting the ability of the plant to take up water
and nutrients. These direct effects can, in turn, increase the sensitivity of these plants to stresses,
such as droughts, cold temperatures, insect pests, and disease leading to increased mortality of
canopy trees. Terrestrial acidification affects several important ecological services, including
declines in habitat for threatened and endangered species (cultural), declines in forest aesthetics
(cultural), declines in forest productivity (provisioning), and increases in forest soil erosion and
reductions in water retention (cultural and regulating) (U.S. EPA, 2008).

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,
2008).

4-74


-------
4.4.6	Visibility Impairment Benefits

Reducing ambient PM2.5 levels would improve levels of 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).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 show that visibility benefits can be a significant
welfare benefit category. In this analysis we did not quantify visibility-related benefits and did
not 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 (U.S.
EPA, 2012).

Reductions in emissions of direct PM2.5, SO2, and NO2 will improve the level of visibility
throughout the United States because primary and secondary PM2.5 impairs visibility by
scattering and absorbing light (U.S. EPA, 2009b). 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, 2009b). 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, 2009b).

4.4.7	Water Quality and Availability Benefits

As described in Section 3, operators are expected to increase generation from lower-
emitting resources in the baseline, and these final rules are expected to continue this trend.
Operators may increase generation at some subset of fossil fuel units, particularly those that
install CCS. As described in Section 3, incremental adoption of CCS and hydrogen technologies
are expected under this rulemaking, and as noted in preamble sections VII(F)(3), X(D)(1), and

4-75


-------
XIV(E)(3), these technologies have water demands and may have implications for water
availability.

At coal units that decrease generation, there are several negative health, ecological, and
productivity effects associated with water effluent and intake that will be avoided. The impacts
of coal generation on water quality and availability are qualitatively described below. For
additional discussion of these impacts and welfare implications, see U.S. EPA (2020b) and U.S.
EPA (2023a). Coal units that increase generation, particularly those that install CCS, may have
associated water quality disbenefits if there is increased effluent related to wet-flue gas
desulfurization (FGD) controls and bottom ash (BA) transport. However, this concern would be
mitigated with the finalization of the 2023 Proposed Supplemental Effluent Limitations
Guidelines and Standards for the Steam Electric Power Generating Point Source Category,
which proposes zero-discharge effluent limitations for FGD wastewater and BA transport
water.125 Also, the proposed effluent limitation guidelines propose new numeric limits to
combustion residual leachate, which addresses concerns that FGD waste increases leachate of
mercury.

4.4.7.1 Potential Water Quality Benefits of Reduced Coal-Fired Power Generation

Discharges of wastewater from coal-fired power plants contain toxic and bioaccumulative
pollutants (e.g., selenium, mercury, arsenic, nickel), halogen compounds (containing bromide,
chloride, or iodide), nutrients, and total dissolved solids (TDS), which can cause human health
and environmental harm through surface water and fish tissue contamination. Pollutants in coal
combustion wastewater are of particular concern because they can occur in large quantities (i.e.,
total pounds) and at high concentrations (i.e., exceeding drinking water Maximum Contaminant
Levels (MCLs)) in discharges and leachate to groundwater and surface waters. These potential
beneficial effects follow directly from reductions in pollutant loadings to receiving waters, and
indirectly from other changes in plant operations. The potential benefits come in the form of
reduced morbidity, mortality, and on environmental quality and economic activities; reduction in
water use, which provides benefits in the form of increased availability of surface water and
groundwater; and reductions in the use of surface impoundments to manage Coal Combustion

125 https://www.epa.gov/eg/steam-electric-power-generating-effluent-guidelines-2023-proposed-rule

4-76


-------
Residual wastes, with benefits in the form of avoided cleanup and other costs associated with
impoundment releases.

Discharges of wastewater from coal-fired power plants affect human health risk by
changing exposure to pollutants in water via two principal exposure pathways: (1) treated water
sourced from surface waters affected by coal-fired power plant discharges and (2) fish and
shellfish taken from waterways affected by coal-fired power plant discharges. The human health
benefits from surface water quality improvements may include drinking water benefits, fish
consumption benefits, and other complimentary measures.

In addition, corresponding surface water quality changes can affect the ecological
condition and recreation use effects. EPA expects the ecological impacts from reduced coal-fired
power plant discharges could include habitat changes for fresh- and saltwater plants,
invertebrates, fish, and amphibians, as well as terrestrial wildlife and birds that prey on aquatic
organisms exposed to pollutants from coal combustion. The change in pollutant loadings has the
potential to result in changes in ecosystem productivity in waterways and the health of resident
species, including threatened and endangered (T&E) species. Loadings from coal-fired power
generation have the potential to impact the general health of fish and invertebrate populations,
their propagation to waters, and fisheries for both commercial and recreational purposes.

Changes in water quality also have the potential to impact recreational activities such as
swimming, boating, fishing, and water skiing.

Potential economic productivity effects may stem from changes in the quality of public
drinking water supplies and irrigation water; changes in sediment deposition in reservoirs and
navigational waterways; and changes in tourism, commercial fish harvests, and property values.

4.4.7.2 Drinking Water

Pollutants discharged by coal-fired power plants to surface waters may affect the quality
of water used for public drinking supplies. In turn these impacts to public water supplies have the
potential to affect the costs of drinking water treatment (e.g., filtration and chemical treatment)
by changing eutrophication levels and pollutant concentrations in source waters. Eutrophication
is one of the main causes of taste and odor impairment in drinking water, which has a major

4-77


-------
negative impact on public perceptions of drinking water safety. Additional treatment to address
foul tastes and odors can significantly increase the cost of public water supply.

Although public drinking water supplies are subject to legally enforceable maximum
contaminant levels (MCLs), established by EPA, pollutants discharged from coal-fired power
plants, particularly episodic releases, may not be removed adequately during treatment at a
drinking water treatment plant exposing consumers to these contaminants through ingestion,
inhalation, and skin absorption. The constituents found in the power plant discharge may also
interact with drinking water treatment processes and contribute to the formation of disinfection
byproducts that can have adverse human health impacts.

4.4.7.3	Fish Consumption

Recreational and subsistence fishers (and their household members) who consume fish
caught in the reaches downstream of coal-fired power plants may be affected by changes in
pollutant concentrations in fish tissue. See U.S. EPA (2020b) and U.S. EPA (2023a) for a
demonstration of the changes in risk to human health from exposure to contaminated fish tissue.
This document describes the neurological effects to children ages 0 to 7 from exposure to lead;
the neurological effects to infants from in-utero exposure to mercury; the incidence of skin
cancer from exposure to arsenic; and the reduced risk of other cancer and non-cancer toxic
effects.

4.4.7.4	Changes in Surface Water Quality

Reduced coal-fired power plant discharges may affect the value of ecosystem services
provided by surface waters through changes in the habitats or ecosystems (aquatic and
terrestrial). Society values changes in ecosystem services by a number of mechanisms, including
increased frequency of use and improved quality of the habitat for recreational activities (e.g.,
fishing, swimming, and boating). Individuals also value the protection of habitats and species
that may reside in waters that receive water discharges from coal-plants, even when those
individuals do not use or anticipate future use of such waters for recreational or other purposes,
resulting in nonuse values.

4-78


-------
4.4.7.5	Impacts on Threatened and Endangered Species

For T&E species, even minor changes to reproductive rates and mortality levels may
represent a substantial portion of annual population variation. Therefore, changing the discharge
of coal-fired power plant pollutants to aquatic habitats has the potential to impact the
survivability of some T&E species living in these habitats. The economic value for these T&E
species primarily comes from the nonuse values people hold for the survivorship of both
individual organisms and species survival.

4.4.7.6	Changes in Sediment Contamination

Water effluent discharges from coal-fired power plants can also contaminate waterbody
sediments. For example, sediment adsorption of arsenic, selenium, and other pollutants found in
water discharges can result in accumulation of contaminated sediment on stream and lake beds,
posing a particular threat to benthic (i.e., bottom-dwelling) organisms. These pollutants can later
be re-released into the water column and enter organisms at different trophic levels.
Concentrations of selenium and other pollutants in fish tissue of organisms of lower trophic
levels can bio-magnify through higher trophic levels, posing a threat to the food chain at large
(Ruhl et al., 2012).

4.4.7.7	Reservoir Capacity and Sedimentation Changes in Navigational Waterways

Reservoirs serve many functions, including storage of drinking and irrigation water
supplies, flood control, hydropower supply, and recreation. Streams can carry sediment into
reservoirs, where it can settle and cause buildup of sediment layers over time, reducing reservoir
capacity (Graf et al., 2010, 2011) and the useful life of reservoirs unless measures such as
dredging are taken to reclaim capacity (Hargrove et al., 2010; Miranda, 2017). Likewise,
navigable waterways, including rivers, lakes, bays, shipping channels and harbors, are prone to
reduced functionality due to sediment build-up, which can reduce the navigable depth and width
of the waterway (Ribaudo and Johansson, 2006). For many navigable waters, periodic dredging
is necessary to remove sediment and keep them passable. Dredging of reservoirs and navigable
waterways can be costly. EPA expects that changes in suspended solids effluent discharge from
coal-fired power plants could reduce sediment loadings to surface waters decreasing reservoir
and navigable waterway maintenance costs by changing the frequency or volume of dredging

4-79


-------
activity. (Hargrove et al., 2010; Miranda, 2017). Likewise, navigable waterways, including
rivers, lakes, bays, shipping channels and harbors, are prone to reduced functionality due to
sediment build-up, which can reduce the navigable depth and width of the waterway (Ribaudo
and Johansson, 2006). For many navigable waters, periodic dredging is necessary to remove
sediment and keep them passable. Dredging of reservoirs and navigable waterways can be costly.
EPA expects that changes in suspended solids effluent discharge from coal-fired power plants
could reduce sediment loadings to surface waters decreasing reservoir and navigable waterway
maintenance costs by changing the frequency or volume of dredging activity. (Graf et al., 2010,
2011) and the useful life of reservoirs unless measures such as dredging are taken to reclaim
capacity (Hargrove et al., 2010; Miranda, 2017). Likewise, navigable waterways, including
rivers, lakes, bays, shipping channels and harbors, are prone to reduced functionality due to
sediment build-up, which can reduce the navigable depth and width of the waterway (Ribaudo
and Johansson, 2006). For many navigable waters, periodic dredging is necessary to remove
sediment and keep them passable. Dredging of reservoirs and navigable waterways can be costly.
EPA expects that changes in suspended solids effluent discharge from coal-fired power plants
could reduce sediment loadings to surface waters decreasing reservoir and navigable waterway
maintenance costs by changing the frequency or volume of dredging activity. (Hargrove et al.,
2010; Miranda, 2017). Likewise, navigable waterways, including rivers, lakes, bays, shipping
channels and harbors, are prone to reduced functionality due to sediment build-up, which can
reduce the navigable depth and width of the waterway (Ribaudo and Johansson, 2006). For many
navigable waters, periodic dredging is necessary to remove sediment and keep them passable.
Dredging of reservoirs and navigable waterways can be costly. EPA expects that changes in
suspended solids effluent discharge from coal-fired power plants could reduce sediment loadings
to surface waters decreasing reservoir and navigable waterway maintenance costs by changing
the frequency or volume of dredging activity.

4.4.7.8 Changes in Water Withdrawals

A reduction in water withdrawals from coal-fired power plants may benefit aquatic and
riparian species downstream of the power plant intake through the provision of additional water
resources in the face of drying conditions and increased rainfall variability. Reductions in water

4-80


-------
withdrawals will also lower the number of aquatic organisms impinged and entrained by the
power plant's water filtration and cooling systems.

4.5 Total Benefits

Table 4-25 through Table 4-28 present the combined monetized climate benefits126 and
PM2.5 and ozone-related health benefits for the three illustrative scenarios for the five snapshot
years analyzed. Table 4-30 through Table 4-32 present the stream of annual monetized combined
climate benefits and PM2.5 and ozone-related health benefits for the three illustrative scenarios,
as well as the present values (PVs) and equivalent annualized values (EAVs), calculated for the
2024 to 2047 timeframe.

126 Monetized climate benefits are discounted using a 2 percent discount rate, consistent with EPA's updated

estimates of the SC-CO2. OMB has long recognized that climate effects should be discounted only at appropriate
consumption-based discount rates. Because the SC-CO2 estimates reflect net climate change damages in terms of
reduced consumption (or monetary consumption equivalents), the use of the social rate of return on capital (7
percent under OMB Circular A-4 (2003)) to discount damages estimated in terms of reduced consumption would
inappropriately underestimate the impacts of climate change for the purposes of estimating the SC-CO2. See
Section 4.2 for more discussion.

4-81


-------
Table 4-25 Total Benefits for the Illustrative Scenarios for 2028 (billions of 2019
dollars)"	

Climate Benefits and PM2.5 and
03-related Health Benefitsb

(Discount Rate Applied to Health
Benefits)



SC-CO2 Near-
term Ramsey
Discount Rate

Climate Benefits
Only

2%

3%

7%

Final Rules



1.5%

14

20

20

19



2.0%

8.4

14

14

13



2.5%

5.2

11

11

10

Alternative 1



1.5%

13

19

18

18



2.0%

7.9

13

13

12



2.5%

4.9

10

10

9.4

Alternative 2



1.5%

12

17

17

16



2.0%

7.1

12

12

11



2.5%

4.4

9.2

9.1

8.5

Non-Monetized Benefits0

Benefits from reductions in HAP emissions
Benefits from improved water quality and availability
Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP
Reductions in exposure to ambient NO2 and SO2
Improved visibility (reduced haze) from PM2 5 reductions

3 Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
b For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates. For discussions of the uncertainty associated with these health benefits estimates, see Section 4.3.8.
c Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

4-82


-------
Table 4-26 Total Benefits for the Illustrative Scenarios for 2030 (billions of 2019
dollars)"	

Climate Benefits and PM2.5 and
03-related Health Benefitsb

(Discount Rate Applied to Health
Benefits)



SC-CO2 Near-
term Ramsey
Discount Rate

Climate Benefits
Only

2%

3%

7%

Final Rules



1.5%

19

23

23

22



2.0%

11

15

15

15



2.5%

7.1

11

11

11

Alternative 1



1.5%

18

22

22

21



2.0%

11

14

14

14



2.5%

6.8

10

10

9.9

Alternative 2



1.5%

10

13

13

13



2.0%

6.2

9.0

8.9

8.6



2.5%

3.9

6.7

6.6

6.3

Non-Monetized Benefits0

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
b For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates. For discussions of the uncertainty associated with these health benefits estimates, see Section 4.3.8.
c Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

4-83


-------
Table 4-27 Total Benefits for the Illustrative Scenarios for 2035 (billions of 2019
dollars)"	

Climate Benefits and PM2.5 and
03-related Health Benefitsb

(Discount Rate Applied to Health
Benefits)



SC-CO2 Near-
term Ramsey
Discount Rate

Climate Benefits
Only

2%

3%

7%

Final Rules



1.5%

50

64

64

62



2.0%

30

45

44

43



2.5%

19

34

33

32

Alternative 1



1.5%

50

65

65

63



2.0%

30

46

45

43



2.5%

19

35

34

32

Alternative 2



1.5%

49

64

64

62



2.0%

30

45

44

43



2.5%

19

34

34

32

Non-Monetized Benefits0

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
b For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates. For discussions of the uncertainty associated with these health benefits estimates, see Section 4.3.8.
c Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

4-84


-------
Table 4-28 Total Benefits for the Illustrative Scenarios for 2040 (billions of 2019
dollars)"	

Climate Benefits and PM2.5 and
03-related Health Benefitsb

(Discount Rate Applied to Health
Benefits)



SC-CO2 Near-
term Ramsey
Discount Rate

Climate Benefits
Only

2%

3%

7%

Final Rules



1.5%

23

22

22

22



2.0%

14

14

14

14



2.5%

9.1

8.8

8.8

8.8

Alternative 1



1.5%

23

23

23

23



2.0%

14

14

14

14



2.5%

9.1

9.1

9.1

9.1

Alternative 2



1.5%

23

23

23

23



2.0%

14

14

14

14



2.5%

9.0

9.1

9.1

9.1

Non-Monetized Benefits0

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
b For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates. For discussions of the uncertainty associated with these health benefits estimates, see Section 4.3.8.
c Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

4-85


-------
Table 4-29 Total Benefits for the Illustrative Scenarios for 2045 (billions of 2019
dollars)"	

Climate Benefits and PM2.5 and
03-related Health Benefitsb

(Discount Rate Applied to Health
Benefits)



SC-CO2 Near-
term Ramsey
Discount Rate

Climate Benefits
Only

2%

3%

7%

Final Rules



1.5%

19

27

27

26



2.0%

12

20

20

19



2.5%

7.8

16

16

15

Alternative 1



1.5%

18

26

26

25



2.0%

11

20

19

18



2.5%

7.5

16

15

15

Alternative 2



1.5%

18

26

26

25



2.0%

11

20

19

18



2.5%

7.5

16

15

15

Non-Monetized Benefits0

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
b For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates. For discussions of the uncertainty associated with these health benefits estimates, see Section 4.3.8.
c Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

4-86


-------
Table 4-30 Benefits for the Final Rules Illustrative Scenario from 2024 through 2047
(billions of 2019 dollars)"	

All Values Calculated using
2% Discount Rate

Health Benefits Calculated
using 3% Discount Rate,
Climate Benefits Calculated
using 2% Discount Rate

Health Benefits Calculated
using 7% Discount Rate,
Climate Benefits Calculated
using 2% Discount Rate





PM25





PM25





PM25





Climate

and O3-
related
Health
Benefits0

Total

Climate

and O3-
related
Health
Benefits0

Total

Climate

and O3-
related
Health
Benefits0

Total



Benefits

Benefits

Benefits

Benefits

Benefits

Benefits

2028

8.4

5.8

14

8.4

5.6

14

8.4

5.0

13

2029

8.5

5.9

14

8.5

5.8

14

8.5

5.1

14

2030

11

4.0

15

11

3.9

15

11

3.5

15

2031

12

4.1

16

12

4.0

16

12

3.5

15

2032

29

14

43

29

13

42

29

12

41

2033

29

14

43

29

14

43

29

12

42

2034

30

14

44

30

14

44

30

12

42

2035

30

15

45

30

14

44

30

13

43

2036

31

15

46

31

15

45

31

13

44

2037

31

15

46

31

15

46

31

13

44

2038

14

-0.34

13

14

-0.33

13

14

-0.29

13

2039

14

-0.34

14

14

-0.33

14

14

-0.30

14

2040

14

-0.35

14

14

-0.34

14

14

-0.30

14

2041

14

-0.36

14

14

-0.35

14

14

-0.31

14

2042

11

7.9

19

11

7.7

19

11

6.8

18

2043

12

8.0

20

12

7.7

19

12

6.9

18

2044

12

8.1

20

12

7.8

20

12

6.9

19

2045

12

8.2

20

12

7.9

20

12

7.0

19

2046

12

8.3

20

12

8.0

20

12

7.1

19

2047

12

8.3

21

12

8.1

20

12

7.2

19

pyd

270

120

390

270

100

370

270

59

330

EAVd

14

6.3

21

14

6.1

20

14

5.2

19

Non-Monetized Benefits®

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Emissions impacts are not estimated for the years 2024 to 2027. As a result, the first year of benefits analysis is
2028.

4-87


-------
b Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
c For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates. For discussions of the uncertainty associated with these health benefits estimates, see Section 4.3.8.
d The PV and EAV values in this table are for the timeframe of 2024 to 2047, not 2028 to 2047.
e Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

4-88


-------
Table 4-31 Benefits for the Alternative 1 Illustrative Scenario from 2028 through 2047
(billions of 2019 dollars)a,b	

All Values Calculated using
2% Discount Rate

Health Benefits Calculated
using 3% Discount Rate,
Climate Benefits Calculated
using 2% Discount Rate

Health Benefits Calculated
using 7% Discount Rate,
Climate Benefits Calculated
using 2% Discount Rate





PM25





PM25





PM25





Climate

and O3-
related
Health
Benefits0

Total

Climate

and O3-
related
Health
Benefits0

Total

Climate

and O3-
related
Health
Benefits0

Total



Benefits

Benefits

Benefits

Benefits

Benefits

Benefits

2028

7.9

5.2

13

7.9

5.0

13

7.9

4.5

12

2029

8.0

5.3

13

8.0

5.2

13

8.0

4.6

13

2030

11

3.6

14

11

3.4

14

11

3.0

14

2031

11

3.6

15

11

3.5

15

11

3.1

14

2032

29

14

43

29

14

43

29

12

41

2033

29

15

44

29

14

44

29

13

42

2034

30

15

45

30

15

44

30

13

43

2035

30

15

46

30

15

45

30

13

43

2036

31

16

46

31

15

46

31

13

44

2037

31

16

47

31

15

47

31

14

45

2038

14

0.0026

14

14

0.0025

14

14

0.0013

14

2039

14

0.0016

14

14

0.0016

14

14

0.00044

14

2040

14

0.00043

14

14

0.00045

14

14

-0.00065

14

2041

14

-0.0011

14

14

-0.00097

14

14

-0.0019

14

2042

11

8.0

19

11

7.7

19

11

6.9

18

2043

11

8.1

19

11

7.8

19

11

6.9

18

2044

11

8.1

19

11

7.9

19

11

7.0

18

2045

11

8.2

20

11

8.0

19

11

7.1

18

2046

12

8.3

20

12

8.1

20

12

7.2

19

2047

12

8.4

20

12

8.1

20

12

7.2

19

PVd

270

120

390

270

110

370

270

60

330

EAVd

14

6.5

21

14

6.2

20

14

5.2

19

Non-Monetized Benefits®

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Emissions impacts are not estimated for the years 2024 to 2027. As a result, the first year of benefits analysis is
2028.

b Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.

4-89


-------
c For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates. For discussions of the uncertainty associated with these health benefits estimates, see Section 4.3.8.
d The PV and EAV values in this table are for the timeframe of 2024 to 2047, not 2028 to 2047.
e Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

4-90


-------
Table 4-32 Benefits for the Alternative 2 Illustrative Scenario from 2024 through 2047
(billions of 2019 dollars)"	

All Values Calculated using
2% Discount Rate

Health Benefits Calculated
using 3% Discount Rate,
Climate Benefits Calculated
using 2% Discount Rate

Health Benefits Calculated
using 7% Discount Rate,
Climate Benefits Calculated
using 2% Discount Rate





PM25





PM25





PM25





Climate

and O3-
related
Health
Benefits0

Total

Climate

and O3-
related
Health
Benefits0

Total

Climate

and O3-
related
Health
Benefits0

Total



Benefits

Benefits

Benefits

Benefits

Benefits

Benefits

2028

7.1

4.8

12

7.1

4.7

12

7.1

4.1

11

2029

7.2

4.9

12

7.2

4.8

12

7.2

4.2

11

2030

6.2

2.8

9.0

6.2

2.7

8.9

6.2

2.4

8.6

2031

6.3

2.8

9.2

6.3

2.8

9.1

6.3

2.4

OO
OO

2032

28

14

43

28

14

42

28

12

41

2033

29

15

43

29

14

43

29

13

41

2034

29

15

44

29

14

44

29

13

42

2035

30

15

45

30

15

44

30

13

43

2036

30

15

46

30

15

45

30

13

44

2037

31

16

46

31

15

46

31

14

44

2038

14

0.089

14

14

0.087

14

14

0.081

14

2039

14

0.088

14

14

0.086

14

14

0.080

14

2040

14

0.087

14

14

0.085

14

14

0.079

14

2041

14

0.085

14

14

0.083

14

14

0.077

14

2042

11

7.9

19

11

7.7

19

11

6.8

18

2043

11

8.0

19

11

7.8

19

11

6.9

18

2044

11

8.1

19

11

7.9

19

11

7.0

18

2045

11

8.2

20

11

8.0

19

11

7.1

18

2046

12

8.3

20

12

8.0

20

12

7.1

19

2047

12

8.4

20

12

8.1

20

12

7.2

19

pyd

250

120

370

250

100

360

250

58

310

EAVd

13

6.3

20

13

6.1

20

13

5.1

19

Non-Monetized Benefits®

Benefits from reductions in HAP emissions

Benefits from improved water quality and availability

Ecosystem benefits associated with reductions in emissions of CO2, NOx, SO2, PM, and HAP

Reductions in exposure to ambient NO2 and SO2

Improved visibility (reduced haze) from PM2 5 reductions

3 Emissions impacts are not estimated for the years 2024 to 2027. As a result, the first year of benefits analysis is
2028.

4-91


-------
b Values have been rounded to two significant figures. Rows may not appear to add correctly due to rounding.
c For simplicity of presentation, the estimated value of the health benefits reported here are the larger of the two
benefits estimates presented in Table 4-15 through Table 4-19. Monetized benefits include those related to public
health associated with reductions in PM2 5 and ozone concentrations. The health benefits are associated with several
point estimates. For discussions of the uncertainty associated with these health benefits estimates, see Section 4.3.8.
d The PV and EAV values in this table are for the timeframe of 2024 to 2047, not 2028 to 2047.
e Several categories of climate, human health, and welfare benefits from CO2, NOx, SO2, PM and HAP emissions
reductions remain unmonetized and are thus not directly reflected in the quantified benefit estimates in this table.
See Section 4.2 for a discussion of climate effects that are not yet reflected in the SC-CO2 and thus remain
unmonetized and Section 4.4 for a discussion of other non-monetized benefits.

4-92


-------
4.6 References

AHRQ. (2016). Healthcare Cost and Utilization Project (HCUP). Retrieved from:
https://www.ahrq. gov/data/hcup/index.html

Anthoff, D., & Tol, R. S. J. (2013a). Erratum to: The uncertainty about the social cost of carbon:
A decomposition analysis using FUND (vol 117, pg 515, 2013). Climatic Change,
727(2), 413-413. doi:10.1007/sl0584-013-0959-l

Anthoff, D., & Tol, R. S. J. (2013b). The uncertainty about the social cost of carbon: A
decomposition analysis using FUND. Climatic Change, 117(3), 515-530.
doi: 10.1007/sl0584-013-0706-7

Arrow, K., Cropper, M., Gollier, C., Groom, B., Heal, G., Newell, R., . . . Weitzman, M. (2013).
Determining Benefits and Costs for Future Generations. Science, 3¥7(6144), 349-350.
doi: doi: 10.1126/ science. 1235665

ATSDR. (2022). Toxicological Profile for Mercury (Draft for Public Comment). (CAS#: 7439-
97-6). U.S. Center for Desease Control.

https://wwwn.cdc.gov/TSP/ToxProfiles/ToxProfiles.aspx?id=115&tid=24

Auffhammer, M. (2018). Quantifying economic damages from climate change. Journal of
Economic Perspectives, 32(4), 33-52.

Bauer, M. D., & Rudebusch, G. D. (2020). Interest Rates under Falling Stars. American
Economic Review, 110(5), 1316-1354. doi:10.1257/aer.20171822

Bauer, M. D., & Rudebusch, G. D. (2023). The Rising Cost of Climate Change: Evidence from
the Bond Market. The Review of Economics and Statistics, 105(5), 1255-1270.
doi: 10.1162/rest_a_01109

Bell, M. L., Dominici, F., & Samet, J. M. (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), 436-445. doi:10.1097/01.ede.0000165817.40152.85

Bell, M. L., McDermott, A., Zeger, S. L., Samet, J. M., & Dominici, F. (2004). Ozone and short-
term mortality in 95 US urban communities, 1987-2000. Jama, 292(19), 2372-2378.
doi: 10.1001/jama.292.19.2372

Carleton, T., Jina, A., Delgado, M., Greenstone, M., Houser, T., Hsiang, S., Hultgren, A., Kopp,
R.E., McCusker, K.E., Nath, I., Rising, J., Ashwin, A., Seo, H., Viaene, A., Yaun, J., and
Zhang, A.,. (2022). Valuing the Global mortality Consequences of Climate Change
Accounting for Adaptation Costs and Benefits. The Quarterly Journal of Economics,
137(4), 2037-2105.

Chetty, R., Friedman, J. N., & Rockoff, J. E. (2014). Measuring the Impacts of Teachers II:

Teacher Value-Added and Student Outcomes in Adulthood. American Economic Review,
104(9), 2633-2679. doi: 10.1257/aer. 104.9.2633

4-93


-------
CIL, C. I. L. (2023). Documentation for Data-driven Spatial Climate Impact Model (DSCIM).
Retrieved from

Cropper, M. L., Freeman, M. C., Groom, B., & Pizer, W. A. (2014). Declining Discount Rates.
American Economic Review, 104(5), 538-543. doi:10.1257/aer,104.5.538

Di, Q., Wang, Y., Zanobetti, A., Wang, Y., Koutrakis, P., Choirat, C., . . . Schwartz, J. D. (2017).
Air pollution and mortality in the Medicare population. New England Journal of
Medicine, 376(26), 2513-2522. doi:10.1056/NEJMoal702747

Graf, W. L., Wohl, E., Sinha, T., & Sabo, J. L. (2010). Sedimentation and sustainability of
western American reservoirs. Water Resources Research, 46(12).
doi :https://doi.org/10.1029/2009WR008836

Graf, W. L., Wohl, E., Sinha, T., & Sabo, J. L. (2011). Correction to "Sedimentation and
sustainability of western American reservoirs". Water Resources Research, 47(8).
doi:https://doi.org/10.1029/2011WR011172

Hargrove, W. L., Johnson, D., Snethen, D., & Middendorf, J. (2010). From Dust Bowl to Mud

Bowl: Sedimentation, conservation measures, and the future of reservoirs. Journal of Soil
and Water Conservation, 65(1), 14A-17A. doi:10.2489/jswc.65.1.14A

Hollmann, F., Mulder, T., & Kalian, J. J. W., DC: US Bureau of the Census. (2000).

Methodology and assumptions for the population projections of the United States: 1999
to 2100 (Population Division Working Paper No. 38). 338.

Hope, C. (2013). Critical issues for the calculation of the social cost of C02: why the estimates
from PAGE09 are higher than those from PAGE2002. Climate Change, 117(113), 531 -
543.

Howard, P. H., & Sterner, T. (2017). Few and Not So Far Between: A Meta-analysis of Climate
Damage Estimates. Environmental and Resource Economics, 68( 1), 197-225.
doi: 10.1007/s 10640-017-0166-z

Huang, Y., Dominici, F., & Bell, M. L. (2005). Bayesian hierarchical distributed lag models for
summer ozone exposure and cardio-respiratory mortality. Environmetrics, 16(5), 547-
562. doi:10.1002/env.721

IPCC. (2018). Global Warming of 1.5°C. An IPCC Special Report on the impacts of global
warming of 1.5°C above pre-industrial levels and related global greenhouse gas
emission pathways, in the context of strengthening the global response to the threat of
climate change, sustainable development, and efforts to eradicate poverty (V. Masson-
Delmotte, P. Zhai, H.-O. Portner, D. Roberts, J. Skea, P. R. Shukla, A. Pirani, W.
Moufouma-Okia, C. Pean, R. Pidcock, S. Connors, J. B. R. Matthews, Y. Chen, X. Zhou,
M. I. Gomis, E. Lonnoy, T. Maycock, a. M. Tignor, & T. Waterfield Eds.).

IPCC. (2021a). The Earth's Energy Budget, Climate Feedbacks, and Climate Sensitivity:
Contribution of Working Group I to the Sixth Assessment Report of the

4-94


-------
Intergovernmental Panel on Climate Change. In V. Masson-Delmotte, P. Zhai, A. Pirani,
S.L. Connors, C. Pean, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M.

Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O.
Yelek
-------
Jerrett, M., Burnett, R. T., Pope, C. A., Ito, K., Thurston, G., Krewski, D., . . . Thun, M. (2009).
Long-Term Ozone Exposure and Mortality. New England Journal of Medicine, 360(11),
1085-1095. doi:10.1056/NEJMoa0803894

Katsouyanni, K., Samet, J. M., Anderson, H. R., Atkinson, R., Le Tertre, A., Medina, S., . . .
Committee, H. E. I. H. R. (2009). Air pollution and health: a European and North
American approach (APHENA). Res Rep Health Ejf Inst(\42), 5-90. Retrieved from
https://www.ncbi.nlm.nih.gov/pubmed/20Q73322

Kivi, P. A., & Shogren, J. F. (2010). Second-Order Ambiguity in Very Low Probability Risks:

Food Safety Valuation. Journal of Agricultural and Resource Economics, 35(3), 443-456.
Retrieved from http://www.istor.org/stable/23243065

Krewski, D., Jerrett, M., Burnett, R. T., Ma, R., Hughes, E., Shi, Y., . . . Calle, E. E. (2009).
Extended follow-up and spatial analysis of the American Cancer Society study linking
particulate air pollution and mortality. Health Effects Institute Boston, MA.

Kiinzli, N., Kaiser, R., Medina, S., Studnicka, M., Chanel, O., Filliger, P., . . . Quenel, P. (2000).
Public-health impact of outdoor and traffic-related air pollution: a European assessment.
The Lancet, 356(9232), 795-801.

Levy, J. I., Chemerynski, S. M., & Sarnat, J. A. (2005). Ozone exposure and mortality: an
empiric bayes metaregression analysis. Epidemiology, 16(4), 458-468.
doi:10.1097/01.ede.0000165820.08301.b3

Liu, J., Lee, M., & Gershenson, S. (2021). The short- and long-run impacts of secondary school
absences. Journal of Public Economics, 199, 104441.
doi :https:// doi .org/10.1016/i ,i pubeco.2021.104441

Maniloff, P., & Fann, N. (2023). Estimates of the cost of illness of myocardial infarction, asthma
and stroke. Applied Economics, 1-11. doi:10.1080/00036846.2023.2257935

McGartland, A., Revesz, R., Axelrad, D. A., Dockins, C., Sutton, P., & Woodruff, T. J. (2017).
Estimating the health benefits of environmental regulations. Science, 357(6350), 457-
458. doi:doi: 10.1126/science.aam8204

Millar, R. J., Nicholls, Z. R., Friedlingstein, P., & Allen, M. R. (2017). A modified impulse-
response representation of the global near-surface air temperature and atmospheric
concentration response to carbon dioxide emissions. Atmos. Chem. Phys., 17(11), 7213-
7228. doi: 10.5194/acp-17-7213 -2017

Miranda, L. E. (2017). Section 3: Sedimentation. In Reservoir Fish Habitat Management.
Totowa, New Jersey: Lightning Press.

National Academies. (2017). Valuing Climate Damages: Updating Estimation of the Social Cost
of Carbon Dioxide. Washington DC: The National Academies Press.

4-96


-------
National Research Council. (2008). Estimating Mortality Risk Reduction and Economic Benefits
from Controlling Ozone Air Pollution. In. Washington (DC): National Academies Press
(US).

Newell, R. G., Pizer, W. A., & Prest, B. C. (2022). A Discounting Rule for the Social Cost of
Carbon. Journal of the Association of Environmental and Resource Economists, 9(5),
1017-1046. doi: 10.1086/718145

Nordhaus, W. D. (2010). Economic aspects of global warming in a post-Copenhagen

environment. . Proceedings of the National Academy of Sciences of the United States of
America, 107(126), 11721-11726.

Nordhaus, W. D. (2014). Estimates of the Social Cost of Carbon: Concepts and Results from the
DICE-2013R Model and Alternative Approaches. Journal of the Association of
Environmental and Resource Economists, 7(1/2), 273-312. doi: 10.1086/676035

Nordhaus, W. D., & Moffat, A. (2017). A survey of global impacts of climate change:

replication, survey methods, and a statistical analysis. National Bureau of Economic
Research, Working Paper 23646. Retrieved from http://www.nber.org/papers/w23646

OMB. (2003). Circular A-4: Regulatory Analysis. Washington DC.
https://www.whitehouse.gov/wp-

content/uploads/legacv drupal files/omb/circulars/A4/a-4.pdf

OMB. (2023). Circular A-4: Regulatory Analysis. Washington DC.

https://www.whitehouse.gov/wp-content/uploads/2023/ll/CircularA-4.pdf

Pindyck, R. S. (2017). Comments on Proposed Rule and Regulatory Impact Analysis on the
Delay and Suspension of Certain Requirements for Waster Prevention and Resource
Conservation. Comment submitted on Nov. 6, 2017.

https://downloads.regulations.gOv/EPA-HO-OAR-2018-0283-6184/attachment 6.pdf

Pope, C. A., Burnett, R. T., Thun, M. J., Calle, E. E., Krewski, D., Ito, K., & Thurston, G. D.
(2002). Lung cancer, cardiopulmonary mortality, and long-term exposure to fine
particulate air pollution. Jama, 287(9), 1132-1141.

Pope, C. A., Lefler, J. S., Ezzati, M., Higbee, J. D., Marshall, J. D., Kim, S.-Y., . . . Robinson, A.
L. (2019). Mortality risk and fine particulate air pollution in a large, representative cohort
of US adults. Environmental Health Perspectives, 127(7), 077007.

Pope, C. A., Thun, M. J., Namboodiri, M. M., Dockery, D. W., Evans, J. S., Speizer, F. E., &

Heath, C. W. (1995). Particulate air pollution as a predictor of mortality in a prospective
study of US adults. American Journal of Respiratory Critical Care Medicine, 151(3),
669-674.

Pope, C. A., Turner, M. C., Burnett, R. T., Jerrett, M., Gapstur, S. M., Diver, W. R., . . . Brook,
R. D. (2015). Relationships Between Fine Particulate Air Pollution, Cardiometabolic

4-97


-------
Disorders, and Cardiovascular Mortality. Circulation Research, 77(5(1), 108-115.
doi:doi:10.1161/CIRCRESAHA. 116.305060

Ramsey, F. P. (1928). A Mathematical Theory of Saving. The Economic Journal, 35(152), 543-
559. doi: 10.2307/2224098

Rennert, K., Errickson, F., Prest, B. C., Rennels, L., Newell, R. G., Pizer, W., . . . Anthoff, D.
(2022). Comprehensive Evidence Implies a Higher Social Cost of C02. Nature,
610(1933), 687-692. doi:10.1038/s41586-022-05224-9

Rennert, K., Prest, B. C., Pizer, W. A., Newell, R. G., Anthoff, D., Kingdon, C., . . . Errickson, F.
(2022). The social cost of carbon: Advances in long-term probabilistic projections of
population, GDP, emissions, and discount rates. Brookings Papers on Economic Activity,
Fall 2021, 223-305.

Ribaudo, M., & Johansson, R. (2006). Water Quality: Impacts on Agriculture. In K. Wiebe & N.
Gollehon (Eds.), Agricultural Resources and Environmental Indicators, 2006 Edition
(EIB-16). Washington DC: Economic Research Service, U.S. Department of Agriculture.

Rode, A., Carleton, T., Delgado, M., Greenstone, M., Houser, T., Hsiang, S., . . . Nath, I. (2021).
Estimating a social cost of carbon for global energy consumption. Nature, 598(7880),
308-314.

Rose, S., Turner, D., Blanford, G., Bistline, J., de la Chesnaye, F., & Wilson, T. (2014).

Understanding the Social Cost of Carbon: A Technical Assessment. EPRI Technical
Update Report. Palo Alto, CA.

Ruhl, L., Vengosh, A., Dwyer, G. S., Hsu-Kim, H., Schwartz, G., Romanski, A., & Smith, S. D.
(2012). The Impact of Coal Combustion Residue Effluent on Water Resources: A North
Carolina Example. Environmental Science & Technology, 46(2.1), 12226-12233.
doi: 10.1021/es303263x

Sacks, J. D., Fann, N., Gumy, S., Kim, I., Ruggeri, G., & Mudu, P. (2020). Quantifying the

Public Health Benefits of Reducing Air Pollution: Critically Assessing the Features and
Capabilities of WHO's AirQ+ and U.S. EPA's Environmental Benefits Mapping and
Analysis Program—Community Edition (BenMAP—CE). Atmosphere, 77(5), 516.
Retrieved from https://www.mdpi.eom/2073-4433/ll/5/516

Sarofim, M. C., Martinich, J., Neumann, J. E., Willwerth, J., Kerrich, Z., Kolian, M., . . . Hartin,
C. (2021). A Temperature Binning Approach for Multi-sector Climate Impact Analysis.
Climatic Change, 165(1), 22. doi: 10.1007/sl0584-021-03048-6

Schwartz, J. (2005). How sensitive is the association between ozone and daily deaths to control
for temperature? Am JRespir Crit Care Med, 177(6), 627-631.
doi: 10.1164/rccm.200407-9330C

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 (ISSN

4-98


-------
0737-5352-32). Retrieved from Fort Collins, CO:

http://vista.cira.colostate.edu/Improve/spatial-and-seasonal-patterns-and-long-term-
variabilitv-of-the-composition-of-the-haze-in-the-united-states-an-analvsis-of-data-from-
the-improve-network-1996/

Smith, C. J., Forster, P. M., Allen, M., Leach, N, Millar, R. J., Passerello, G. A., & Regayre, L.
A. (2018). FAIR vl.3: a simple emissions-based impulse response and carbon cycle
model. Geosci. Model Dev., 77(6), 2273-2297. doi:10.5194/gmd-ll-2273-2018

Smith, R. L., Xu, B., & Switzer, P. (2009). Reassessing the relationship between ozone and

short-term mortality in US urban communities. Inhalation toxicology, 27(sup2), 37-61.

Technical Support Document: Social Cost of Carbon for Regulatory Impact Analysis under
Executive Order 12866. (2010).

Tol, R. (2009). An analysis of mitigation as a response to climate change. Copenhagen
Consensus on Climate. Copenhagen Consensus Center.

https://copenhagenconsensus.com/sites/default/files/ap mitigation tol v 3.0.pdf

Turner, M. C., Jerrett, M., Pope, A., Ill, Krewski, D., Gapstur, S. M., Diver, W. R., . . . Burnett,
R. T. (2016). Long-term ozone exposure and mortality in a large prospective study.
American Journal of Respiratory and Critical Care Medicine, 793(10), 1134-1142.
doi:10.1164/rccm.201508-16330C

U.S. BEA. (2004). New BEA Economic Areas For 2004. Washington DC.
https://www.bea.gov/news/2004/new-bea-economic-areas-20Q4

U.S. BEA. (2022). Table 1.1.9. Implicit Price Deflators for Gross Domestic Product.
Washington, DC.

https://apps.bea. gov/iTable/?reqid=19&step=3&isuri=l&1921=survev& 1903=13

U.S. EPA. (2008). Integrated Science Assessment for Oxides of Nitrogen and Sulfur-Ecological
Criteria National (Final Report). (EPA/600/R-08/082F). Research Triangle Park, NC.
http: // cfpub. epa. gov/ncea/cfm / recordi spl ay. cfim ? dei d=201485

U.S. EPA. (2009a). Integrated Science Assessment for Particulate Matter (Final Report). (EPA-
600/R-08-139F). Research Triangle Park, NC: Office of Research and Development,
National Center for Environmental Assessment.
https://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=216546

U.S. EPA. (2009b). Integrated Science Assessment for Particulate Matter (second external

review draft). (EPA-600/R-08-139B). Research Triangle Park, NC: Office of Research
and Development, National Center for Environmental Assessment.
http: //cfpub. epa. gov/ncea/cfm/recordi spl ay. cfim ? dei d=210586

U.S. EPA. (2010). Regulatory Impact Analysis (RIA) for Existing Stationary Compression
Ignition Engines NESHAP Final Draft. (EPA 452/R-10-002).

4-99


-------
https://www.epa.gov/sites/default/files/2020-Q7/documents/ic-engines ria final-existing-
ci-engines 2010-02.pdf

U.S. EPA. (201 la). Regulatory Impact Analysis for the Federal Implementation Plans to Reduce
Interstate Transport of Fine Particulate Matter and Ozone in 27 States; Correction of
SIP Approvals for 22 States. Research Triangle Park, NC: U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards, Health and Environmental Impact
Division, https://www3.epa.gov/ttn/ecas/docs/ria/transport ria final-csapr 2011-06.pdf

U.S. EPA. (2011b). Regulatory Impact Analysis for the Final Mercury and Air Toxics Standards.
(EPA-452/R-11-011). Research Triangle Park, NC: U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards, Health and Environmental Impact
Division. http://www.epa.gov/ttn/ecas/regdata/RIAs/matsriafinal.pdf

U.S. EPA. (2012). Regulatory Impact Analysis for the Final Revisions to the National Ambient
Air Quality Standards for Particulate Matter. (EPA-452/R-12-005). Research Triangle
Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, Health and Environmental Impact Division.
https://www3.epa.gov/ttnecasl/regdata/RIAs/finalria.pdf

U.S. EPA. (2015 a). Regulatory Impact Analysis for Residential Wood Heaters NSPS Revision:
Final Report. (EPA-452/R-15-001). https://www.epa.gov/sites/default/files/2020-
07/documents/wood-heaters ria final-nsps-revision 2015-02.pdf

U.S. EPA. (2015b). Regulatory Impact Analysis for the Clean Power Plan Final Rule. (EPA-
452/R-l5-003). Research Triangle Park, NC: U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards, Health and Environmental Impact
Division, https://www.epa.gov/sites/default/files/2020-07/documents/utilities ria final-
clean-power-plan-existing-units 2015-08.pdf

U.S. EPA. (2015 c). Regulatory Impact Analysis of the Final Revisions to the National Ambient
Air Quality Standards for Ground-Level Ozone. (EPA-452/R-15-007). Research Triangle
Park, NC: Office of Air Quality Planning and Standards, Health and Environmental
Impacts Division. https://www3.epa.gov/ttnecasl/docs/20151Q01ria.pdf

U.S. EPA. (2016a). Integrated Science Assessment for Oxides of Nitrogen - Health Criteria.

(EPA/600/R-15/068). Research Triangle Park, NC: Office of Research and Development,

National Center for Environmental Assessment.

http s: //cfpub. epa. gov/ ncea/i sa/ recordi spl ay. cfm? dei d=310879

U.S. EPA. (2016b). Regulatory Impact Analysis of the Cross-State Air Pollution Rule (CSAPR)
Update for the 2008 National Ambient Air Quality Standards for Ground-Level Ozone.
(EPA-452/R-16-004). Research Triangle Park, NC: U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards, Health and Environmental Impact
Division, https://www.epa.gov/sites/default/files/2020-07/documents/transport ria final -
csapr-update 2016-09.pdf

4-100


-------
U.S. EPA. (2017). Integrated Science Assessment for Sulfur Oxides - Health Criteria.

(EPA/600/R-17/451). Research Triangle Park, NC: Office of Research and Development,
National Center for Environmental Assessment.
https://cfpub.epa. gov/ncea/isa/recordisplav.cfm?deid=338596

U.S. EPA. (2019a). Integrated Science Assessment (ISA) for Particulate Matter (Final Report).
(EPA/600/R-19/188). Research Triangle Park, NC: U.S. Environmental Protection
Agency, Office of Research and Development, Center for Public Health and
Environmental Assessment, https://www.epa.gov/naaqs/particulate-matter-pm-standards-
integrated-science-assessments-current-review

U. S. EPA. (2019b). Regulatory Impact Analysis for the Repeal of the Clean Power Plan, and the
Emission Guidelines for Greenhouse Gas Emissions from Existing Electric Utility
Generating Units. (EPA-452/R-19-003). Research Triangle Park, NC: U.S.

Environmental Protection Agency, Office of Air Quality Planning and Standards, Health
and Environmental Impact Division, https://www.epa.gov/sites/production/files/2019-
06/documents/utilities ria final cpp repeal and ace 2019-06.pdf

U.S. EPA. (2020a). Analysis of Potential Costs and Benefits for the National Emission Standards
for Hazardous Air Pollutants: Coal- and Oil-FiredElectric Utility Steam Generating
Units - Subcategory of Certain Existing Electric Utility Steam Generating Units Firing
Eastern Bituminous Coal Refuse for Emissions of Acid Gas Hazardous Air Pollutants.
Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Health and Environmental Impact Division.
https://www.epa.gov/sites/default/files/2020-Q4/documents/mats coal refuse cost-
benefit memo.pdf

U.S. EPA. (2020b). Benefit and Cost Analysis for Revisions to the Effluent Limitations
Guidelines and Standards for the Steam Electric Power Generating Point Source
Category. (EPA-821-R-20-003). Washington DC: U.S. Environmental Protection
Agency. https://www.epa.gov/sites/default/files/202Q-

08/documents/steam electric elg 2020 final reconsideration rule benefit and cost an
alvsis.pdf

U.S. EPA. (2020c). Integrated Science Assessment (ISA) for Oxides of Nitrogen, Oxides of Sulfur
and Particulate Matter Ecological Criteria. (EPA/600/R-20/278). Washington DC: U.S.
Environmental Protection Agency.

https://cfpub.epa. gov/ncea/isa/recordisplav.cfm?deid=349473

U.S. EPA. (2020d). Integrated Science Assessment (ISA) for Ozone and Related Photochemical
Oxidants (FinalReport). (EPA/600/R-20/012). Washington DC: U.S. Environmental
Protection Agency, https://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=348522

U.S. EPA. (2021). Regulatory Impact Analysis for the Final Revised Cross-State Air Pollution
Rule (CSAPR) Update for the 2008 Ozone NAAQS. (EPA-452/R-21-002). Research
Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Health and Environmental Impact Division.

4-101


-------
https://www.epa. gov/ sites/default/files/2021 -
03/documents/revised csapr update ria final.pdf

U.S. EPA. (2022a). BenMAP-CE User Manual and Appendices. Research Triangle Park, NC:

Office of Air Quality Planning and Standards, https://www.epa.gov/benmap/benmap-ce-
manual-and-appendices

U.S. EPA. (2022b). Policy Assessment for the Reconsideration of the National Ambient Air

Quality Standards for Particulate Matter. (EPA-452/R-22-004). Research Triangle Park,
NC. https://www.epa.gov/svstem/files/documents/2Q22-

05/Final%20Policv%20Assessment%20for%20the%20Reconsideration%20of%20the%2
0PM%20NAAQS Mav2022 O.pdf

U.S. EPA. (2022c). Regulatory Impact Analysis for Proposed Federal Implementation Plan

Addressing Regional Ozone Transport for the 2015 Ozone National Ambient Air Quality
Standard. (EPA-452/D-22-001). Research Triangle Park, NC: U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Health and
Environmental Impact Division, https://www.epa.gov/svstem/files/documents/2022-
03/transport ria proposal fip 2015 ozone naaqs 2022-02.pdf

U.S. EPA. (2022d). Regulatory Impact Analysis for the Proposed Reconsideration of the

National Ambient Air Quality Standards for Particulate Matter. (EPA-452/P-22-001).
Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Health and Environmental Impact Division.
https://www.epa.gov/svstem/files/documents/2023-01/naaqs-pm ria proposed 2022-
12.pdf

U.S. EPA. (2022e). Software for Model Attainment Test - Community Edition (SMAT-CE) User's
Guide Software version 2.1. (EPA-454/B-22-013). Research Triangle Park, NC.
https://www.epa.gov/svstem/files/documents/2Q22-

1 l/User%27s%20Manual%20for%20SMAT-CE%202.1 EPA Report 11 30 2022.pdf

U.S. EPA. (2022f). Supplement to the 2019 Integrated Science Assessment for Particulate Matter
(FinalReport). (EPA/600/R-22/028). Research Triangle Park, NC: U.S. Environmental
Protection Agency, Office of Research and Development, Center for Public Health and
Environmental Assessment.

https://cfpub.epa. gov/ncea/isa/recordisplav.cfm?deid=354490

U.S. EPA. (2022g). Technical Support Document (TSD) for the Proposed Federal

Implementation Plan Addressing Regional Ozone Transport for the 2015 Ozone National
Ambient Air Quality Standard: Ozone Transport Policy Analysis Proposed Rule TSD.
U.S. Environmental Protection Agency, Office of Air and Radiation.
https://www.epa.gOv/svstem/files/documents/2022-03/ozone-transport-policv-analvsis-
proposed-rule-tsd.pdf

U.S. EPA. (2023a). Benefit and Cost Analysis for Proposed Supplemental Effluent Limitations
Guidelines and Standards for the Steam Electric Power Generating Point Source
Category. (EPA-821-R-23-003). Washington, D.C.

4-102


-------
https://www.epa.gov/svstem/files/documents/2023-Q3/steam-electric-benefit-cost-
analysis proposed feb-2023.pdf

U.S. EPA. (2023b). Estimating PM2.5- and Ozone-Attributable Health Benefits. Research
Triangle Park, NC: U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Health and Environmental Impact Division.
https://www.epa.gov/svstern/files/documents/2023-01/Estirnating%20PM2.5-
%20and%200zone-Attributable%20Health%20Benefits%20TSD O.pdf

U.S. EPA. (2023 c). Regulatory Impact Analysis for the Proposed New Source Performance

Standards for Greenhouse Gas Emissions from New, Modified, and Reconstructed Fossil
Fuel-Fired Electric Generating Units; Emission Guidelines for Greenhouse Gas
Emissions from Existing Fossil Fuel-Fired Electric Generating Units; and Repeal of the
Affordable Clean Energy Rule. (EPA-452/R-23-006). Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Health
and Environmental Impact Division, https://www.epa.gov/svstem/files/documents/2023-
05/utilities ria proposal 2023-05.pdf

U.S. EPA. (2023 d). Regulatory Impact Analysis of the Standards of Performance for New,

Reconstructed, and Modified Sources and Emissions Guidelines for Existing Sources: Oil
and Natural Gas Sector Climate Review. (EPA-452/R-23-013). Research Triangle Park,
NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, Health and Environmental Impact Division.

https://www.epa.gov/svstem/files/documents/2023-12/eol2866 oil-and-gas-nsps-eg-
climate-review-2060-avl6-ria-20231130.pdf

U.S. EPA. (2023 e). Supplementary Material for the Regulatory Impact Analysis for the Final

Rulemaking, "Standards of Performance for New, Reconstructed, and Modified Sources
and Emissions Guidelines for Existing Sources: Oil and Natural Gas Sector Climate
Review ": EPA Report on the Social Cost of Greenhouse Gases: Estimates Incorporating
Recent Scientific Advances. Washington, DC: U.S. EPA

U.S. EPA Science Advisory Board. (2004). Letter from Trudy Cameron, Ph.D., Chair, Clean Air
Scientific Advisory Committee, to Administrator Michael O. Leavitt Re: Advisory Council
on Clean Air Compliance Analysis Response to Agency Request on Cessation Lag. (EPA-
COUNCIL-LTR-05-001). Washington DC.
https://nepis.epa. gov/Exe/ZyPURL.cgi?Dockev=P100JMYX.txt

U.S. EPA Science Advisory Board. (2019). Letter from Louis Anthony Cox, Jr., Chair, Clean Air
Scientific Advisory Committee, to Administrator Andrew R. Wheeler. Re: CASAC Review
of the EPA's Integrated Science Assessment for Particulate Matter (External Review
Draft - October 2018). (EPA-CASAC-19-002). Washington DC

U.S. EPA Science Advisory Board. (2020a). Letter from Louis Anthony Cox, Jr., Chair, Clean
Air Scientific Advisory Committee, to Administrator Andrew R. Wheeler. Re: CASAC
Review of the EPA 's Integrated Science Assessment for Ozone and Related

4-103


-------
Photochemical Oxidants (External Review Draft - September 2019). (EPA-CASAC-20-
002). Washington DC

U.S. EPA Science Advisory Board. (2020b). Letter from Michael Honeycutt Chair, Scientific
Advisory, to Administrator Lisa Jackson. Re: Science Advisory Board (SAB)
Consideration of the Scientific and Technical Basis of EPA 's Proposed Rule titled
"Increasing Consistency and Transparency in Considering Benefits and Costs in the
Clean Air Act Rulemaking Process. ". (EPA-SAB-20-012). Washington DC

U.S. EPA Science Advisory Board. (2022). CASAC Review of the EPA's Policy Assessment for
the Reconsideration of the National Ambient Air Quality Standards for Particulate
Matter (External Review Draft - October 2021). (EPA-CASAC-22-002). Research
Triangle Park, NC: U.S. Environmental Protection Agency,.
https://casac.epa. gov/ords/sab/f?p=l 13:12:1342972375271::: 12

Woodruff, T. J., Darrow, L. A., & Parker, J. D. (2008). Air pollution and postneonatal infant
mortality in the United States, 1999-2002. Environmental Health Perspectives, 77(5(1),
110-115.

Woods & Poole. (2015). Complete Demographic Database. Retrieved from
https://www.woodsandpoole.com/

Wu, X., Braun, D., Schwartz, J., Kioumourtzoglou, M. A., & Dominici, F. (2020). Evaluating the
impact of long-term exposure to fine particulate matter on mortality among the elderly.
SciAdv, 6(29), eaba5692. doi:10.1126/sciadv.aba5692

Zanobetti, A., & Schwartz, J. (2008). Mortality displacement in the association of ozone with
mortality: an analysis of 48 cities in the United States. Am JRespir Crit Care Med,
777(2), 184-189. doi:10.1164/rccm.200706-8230C

4-104


-------
5 SOCIAL COSTS AND ECONOMIC IMPACTS

This section discusses potential energy market impacts, economy-wide social costs and
economic impacts, small entity impacts, and labor impacts associated with these final rules. The
social cost and economy-wide impacts are estimated using EPA's SAGE model. Note that SAGE
does not currently estimate changes in emissions nor account for environmental benefits. For
additional discussion of impacts on fuel use and electricity prices, see Section 3.

5.1 Energy Market Impacts

The energy sector impacts presented in Section 3 of this RIA include potential changes in
the prices for electricity, natural gas, and coal resulting from the requirements of the final rules.
This section addresses the impact of these potential changes on other markets and discusses some
of the determinants of the magnitude of these potential impacts. We refer to these changes as
secondary market impacts.

Under the final emission guidelines for existing fossil-fuel fired steam generating units,
coal-fired EGUs are not directly required to use any of the measures that EPA determines
constitute BSER. Rather, CAA section 111(d) allows each state in applying standards of
performance based on the BSER candidate technologies to take into account remaining useful
life and other factors. Given the flexibility afforded states in implementing the emission
guidelines under 111(d) and the flexibilities coal-fired EGUs have in complying with the
subsequent, state-established emission standards, the potential economic impacts of the
illustrative scenarios reported in this RIA are necessarily illustrative of actions that states and
affected EGUs 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.

To estimate the energy market impacts of the rules, EPA modeled an illustrative final
rules scenario using IPM, as described in Section 1 and Section 3. This section provides a
quantitative assessment of the energy price impacts for the illustrative final rules scenario and
qualitative assessment of the factors that will in part determine the timing and magnitude of

5-1


-------
potential effects in other markets. Table 5-1 summarizes projected changes in energy prices and
fuel use resulting from the illustrative final scenario.

Table 5-1 Summary of Certain Energy Market Impacts (percent change)





2028

2030

2035

2040

2045

Retail electricity prices
(2019 mills/kWh)

Baseline
Final Rules

Percentage Change (%)

110
109
-0.7%

113
112
-0.5%

109

110
1.4%

108
108
0.2%

105
105
0.7%

Average price of coal delivered to the
power sector

Baseline
Final Rules

1.7
1.7

1.8
1.7

1.8
1.7

1.8
1.8

1.6
1.1

(2019$/ton)

Percentage Change (%)

-1.4%

-1.1%

-0.5%

0.5%

-31.6%

Coal production for power sector use
(million tons)

Baseline
Final Rules

Percentage Change (%)

250
236
-6%

218
209
-4%

141
112
-21%

90
104
15%

26
4
-4%

Price of natural gas delivered to power
sector
(2019$/MMBtu)

Baseline
Final Rules

Percentage Change (%)

3.2
3.2
-1.5%

3.3
3.3
-0.5%

3.3

3.4
3.0%

3.2
3.2
0.0%

3.3
3.3
0.1%

Price of average Henry Hub (spot)
(2019$/MMBtu)

Baseline
Final Rules

Percentage Change (%)

3.1
3.1
-1.6%

3.3
3.3
-0.6%

3.3
3.3
2.9%

3.2
3.2
-0.1%

3.3
3.3
0.0%

Natural gas use for electricity generation
(Trillion Cubic Feet)

Baseline
Final Rules

Percentage Change (%)

12
11
-1.0%

12
12
-1.7%

9

10
4.4%

6
6

0.0%

4
4

1.8%

Note: Positive values indicate increases relative to the baseline.

To provide some historical context to Table 5-1, we present below recent trends observed
over the last decade (2012 to 2022) for the energy market impacts listed:127

•	The annual percent change in real electricity price over this period has been from -2.4
percent to 4 percent and averaged -0.3 percent.

•	The percent change to the real annual price of coal for electricity generation has ranged
from -7.2 percent to 11.4 percent over the past decade and averaged -2.3 percent.

•	The percent change to annual coal use for electricity plants has ranged from -19 percent
to 15 percent over the past decade and averaged -5.6 percent.

•	The percent change to the real annual average cost of natural gas for electricity
generation has ranged from -36 percent to 107 percent over the past decade and
averaged 7.1 percent.

127 EIA. Electric Power Annual 2021 and 2022, available at: https://www.eia.gov/electricity/annual/

5-2


-------
• The percent change to annual natural gas use for electricity plants has ranged from -10.2
percent to 21.0 percent over the past decade and averaged 4.9 percent.

Overall, these projected changes are largely within the range of recent historical changes.

The projected energy market and electricity retail rate impacts of the final rules are
discussed more extensively in Section 3, which also presents projections of power sector
generation and capacity changes by technology and fuel type. The change in retail electricity
prices reported in Chapter 3 is a national average across residential, commercial, and industrial
consumers.

5.2 Economy-wide Social Costs and Economic Impacts

This section analyzes the potential economy-wide impacts of the final rules using a
computable general equilibrium (CGE) model. CGE models are designed to capture substitution
possibilities between production, consumption, and trade; interactions between economic
sectors; and interactions between a policy shock and pre-existing market distortions, such as
taxes that have altered consumption, investment, and labor decisions. As such, CGE models can
provide insights into the effects of regulation that occur outside of the directly regulated sector
because they are able to represent the entire economy in equilibrium in the baseline and under a
regulatory or policy scenario. A CGE model can also be used to estimate the social cost of a
regulation.

5.2.1 Economy-wide Modelling

In 2015, EPA formed a Science Advisory Board (SAB) panel to explore the use of
general equilibrium approaches, and more specifically CGE models, to prospectively evaluate
the costs, benefits, and economic impacts of environmental regulation. In its final report, the
SAB recommended that the Agency enhance its regulatory analyses using CGE models "to offer
a more comprehensive assessment of the benefits and costs" of regulatory actions by capturing
important interactions between markets and that such efforts will be most informative when there
are both significant cross-price effects and pre-existing distortions in those markets (U.S. EPA

5-3


-------
Science Advisory Board, 2017).128 Given the typical level of aggregation in CGE models and
their focus on long run equilibria, the panel observed that CGE modeling results are
complements to, rather than substitutes for, the other types of detailed analysis EPA conducts for
its rulemakings. The report also noted that CGE frameworks offer valuable insights into the
social costs of regulation even when estimates of the benefits of the regulation are not
incorporated into the models, though it highlighted explicit treatment of benefits within a CGE
framework as a long-term research priority. In addition, the panel observed that CGE models
may also offer insights into the ways costs are distributed across regions, sectors, or households.

In response, EPA has invested in building capacity in this class of economy-wide
modeling. A key outcome of this effort is EPA's CGE model of the U.S. economy, called SAGE.
The SAGE model can provide an important complement to the analyses typically performed
during regulatory development by evaluating a broader set of economic impacts and offering an
economy-wide estimate of social costs.129 Note that SAGE does not currently estimate changes in
emissions nor account for environmental benefits. Model version v2.1.1 of SAGE is used in this
analysis.

5.2.2 Overview of the SA GE CGE Model

SAGE is a CGE model that provides a complete, but relatively aggregated, representation
of the entire U.S. economy. CGE models assume that for some discrete period of time an
economy can be characterized by a set of conditions in which supply equals demand in all
markets (referred to as equilibrium). When the imposition of a regulation alters conditions in one
or more markets, the CGE model estimates a new set of relative prices and quantities for all
markets that return the economy to a new equilibrium.130 For example, the model estimates

128	CGE models provide "a fiscally disciplined, consistent and comprehensive accounting framework. They can
ensure that projected behavior of firms and households in a regulated market is fully consistent with the behavior
of those agents in other markets. Consistent representation of behavior, in turn, leads to connections between
markets, allowing CGE models to pick up effects that spill over from one market to another" (SAB 2017).

129	CGE models may also be able to provide additional information on the benefits of regulatory interventions,
though this is a relatively new but active area of research. Note that until the benefits that accrue to society from
mitigating environmental externalities can be incorporated in a CGE model, the economic welfare measure from
the CGE model is incomplete and needs to be augmented with traditional benefits analysis to develop measures
of net benefits.

130	CGE models are generally focused on analyzing medium- or long-run policy effects since they characterize the
new equilibrium (i.e., when supply once again equals demand in all markets). Their ability to capture the

5-4


-------
changes in relative prices and quantities for sector outputs and household consumption of goods,
services, and leisure that allow the economy to return to equilibrium after the regulatory
intervention. In addition, the model estimates a new set of relative prices and demand for factors
of production (e.g., labor, capital, and land) consistent with the new equilibrium, which in turn
determines estimates of household income changes as a result of the regulation (Marten, 2023).
In CGE models, the social cost of the regulation is estimated as the change in economic welfare
in the post-regulation simulated equilibrium from the pre-regulation "baseline" equilibrium. As
discussed in EPA's Guidelines for Preparing Economic Analyses, social costs are the total
economic burden of a regulatory action (U.S. EPA, 2014). 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 because of
reallocating some resources towards pollution mitigation.

Unlike engineering cost or partial equilibrium approaches typically used to evaluate the
costs of regulations, CGE models account for how effects in directly regulated sectors interact
with and affect the behavior of other sectors and consumers. Figure 5-1 uses a simplified circular
flow diagram to depict how input and output markets are generally connected to each other in
CGE models. Following a standard assumption in economics, the model assumes that households
maximize their wellbeing, while firms maximize their profits. Households supply factors of
production to firms in exchange for income (e.g., wages, profits, and interest payments). Firms
use the available factors of production and materials to produce outputs that are then bought and
consumed by households.

transition path of the economy depends on the degree to which they include characteristics of the economy the
restrict its ability to adjust instantaneously (e.g., rigidities in capital markets).

5-5


-------
Figure 5-1 Depiction of the Circular Flow of the Economy

The SAGE model includes explicit subnational regional representation within the U.S. at
the Census Region level. Each region contains representative firms for each of the 23 sectors in
the model that vary by the commodity they produce and have region-specific production
technologies. Each region also has five representative households that vary by income level and
have region-specific preferences (see Table 5-2). Within the economy, households and firms are
assumed to interact in perfectly competitive markets. In addition to households and firms, there
is a single government in SAGE that represents all state, local and federal governments within
the U.S. The government imposes taxes on capital earnings, labor earnings, and production and
uses that revenue (in addition to deficit spending) to provide government services, make transfer
payments to households, and pay interest on government debt.

5-6


-------
Table 5-2 SAGE Dimensional Details

Time

Sectors

Census

Households

Capital

Periods

Regions

(income)

Vintage

2016-2081

Agriculture, forestry, fishing, and hunting

Northeast

<30k

Extant

(5-year

Crude oil

South

30-50k

New

time steps)

Coal mining

Midwest

50-70k





Metal ore and nonmetallic mineral mining

West

70-150k





Electric power



>150k





Natural gas









Water, sewage, and other utilities









Construction









Food and beverage manufacturing









Wood product manufacturing









Petroleum refineries









Chemical manufacturing









Plastics and rubber products manufacturing









Cement manufacturing









Primary metal manufacturing









Fabricated metal product manufacturing









Electronics and technology manufacturing









Transportation equipment manufacturing









Other manufacturing









Transportation









Truck transportation









Services









Healthcare services







Modeling domestic and international trade presents a unique challenge in that the model's
structure needs to account for the fact that the U.S. can be both an importer and an exporter of
the same good at both the national and regional level. SAGE addresses this issue through use of
the "Armington" approach, which assumes that imported and exported versions of the same good
are not perfect substitutes. In SAGE, this assumption is applied to both international and cross-
regional trade within the United States. In addition, SAGE recognizes that the U.S. is a relatively
large part of the global economy and shifts in its imports and exports have the potential to
influence world prices (i.e., the model assumes the United States is a large, open economy).

SAGE is a forward-looking intertemporal model, which means that households and firms
are assumed to make their decisions taking into account what is expected to occur in future years
and how current decisions will impact those outcomes. In an intertemporal model, care is needed

5-7


-------
to ensure that, in response to a new policy, the economy does not instantaneously jump to a new
equilibrium in a way that is inconsistent with the rate at which the economy can realistically
adjust. SAGE seeks to model a more realistic transition path, in part, by differentiating the
flexibility of physical capital by its age. Under this approach the model distinguishes between
existing capital constructed in response to previous investments and new capital constructed after
the start of the model's simulation. Existing capital is assumed to be relatively inflexible and is
used for its original purpose unless a relatively high cost is incurred to alter its functionality.
New capital is more flexible and easily adjusts to changes in the future. Independent of its
vintage, once capital has been constructed in a specific region it cannot be moved to another
region. While physical capital is not mobile, households can make investments in whatever
region of the country they desire.

The dynamics of the baseline economy in SAGE are informed through the calibration of
key exogenous parameters in the model. Most importantly are population and productivity
growth over time. The model reflects heterogeneity in productivity growth across sectors of the
economy consistent with trends that have been historically observed. In addition, the model
captures improvements in energy efficiency that are expected for firms and households going
forward. Additional baseline characteristics, such as changes to government spending and
deficits and changes to international flows of money and investments, are calibrated to key
government forecasts or informed by historical trends.

The SAGE model relies on many data sources to calibrate its parameters. The foundation
is a state-level dataset produced by IMPLAN that describes the interrelated flows of market
goods and factors of production over the course of a year with a high level of sectoral detail.131
This dataset is augmented by information from other sources, such as the Bureau of Economic
Analysis, Energy Information Administration, Federal Reserve, Internal Revenue Service,
Congressional Budget Office, and the National Bureau of Economic Research. The result is a
static dataset that describes the structure and behavior of the economy in a single year.132 These
data are combined with key behavioral parameters for firms and households that are adopted

131	While the underlying IMPLAN data are proprietary, EPA provides the social accounting matrix based on these
data in the publicly available version of SAGE. The data set for the model may also be built anew by following
the instructions in the model documentation along with a licensed version of IMPLAN (www.IMPLAN.com).

132	SAGE is solved using the General Algebraic Modeling System (GAMS) and PATH solver. The model's build
stream is written in both R and GAMS.

5-8


-------
from the published literature or econometrically estimated specifically for the purposes of
calibrating SAGE. To develop the forward-looking baseline for the model, additional
information on key parameters, such as productivity growth, future government spending, and
energy efficiency improvements are incorporated from sources including the Congressional
Budget Office and Energy Information Administration.

To ensure that SAGE is consistent with economic theory and reflects the latest science,
EPA initiated a separate SAB panel to conduct a technical review of SAGE, completed in August
2020 (U.S. EPA Science Advisory Board, 2020). Peer review of SAGE was in accordance with
requirements laid out for a Highly Influential Science Assessment (HISA) consistent with OMB
guidelines.133 The SAB report commended the agency on its development of SAGE, calling it a
well-designed open-source model. The report included recommendations for refining and
improving the model, including several changes that the SAB advised EPA to incorporate before
using the model in regulatory analysis (denoted as Tier 1 recommendations by the SAB). The
SAB's Tier 1 recommendations, including improving the calibration of government expenditures
and deficits and the foreign trade deficit; allowing for more flexibility in the consumer demand
system; and representing the United States as a large open economy, are incorporated into the
model version used in this analysis (v2.1.1), as are several of the SAB's other medium- and long-
run recommendations. For more details on the SAGE model, complete documentation, source
code and build stream are available on EPA's website.134

5.2.3 Linking IPM PE Model to SA GE CGE Model

For these rules, EPA has relied on the Integrated Planning Model (IPM), a partial
equilibrium large-scale unit-level linear programming model, to assess the costs of compliance in
the power sector and related energy markets (see Section 3.4 for more details on the use of IPM).
The SAB noted that electricity sector regulations seem a good candidate for economy-wide
modeling because of the many backward and forward linkages that may result in effects in other
sectors in the economy (SAB, 2017). For example, changes in the price of electricity can affect
its use in the production of other goods and services. There may also be impacts to upstream

133	Office of Management and Budget (2004). Issuance of OMB's 'Final Information Quality Bulletin for Peer

Review.' https://cfpub.epa.gov/si/m05-03.pdf

134	https://www.epa.gov/environmental-economics/cge-modeling-regulatory-analysis

5-9


-------
industries that supply goods and services to the electricity sector (e.g., energy commodities),
labor markets in response to changes in factor prices, and household demand due to changes in
the end-use price of electricity.

5.2.3.1 Compliance Costs and Social Costs

As described in Section 3, for each baseline and policy alternative, IPM solves for the
least-cost approach to meet fixed electricity demands based on highly detailed information about
electricity generation, air pollution control technologies, and primary energy sector market
conditions (coal and natural gas) while satisfying regulatory requirements, resource adequacy,
and other constraints in the electricity sector. Potential effects outside of the electricity, coal, and
natural gas sectors are not evaluated within IPM. The compliance cost estimates from IPM for
the finalized rules equals the estimated increase in expenditures by the power sector to achieve
and maintain compliance with the final rules relative to the baseline while meeting a fixed
electricity demand.

Specifically, IPM minimizes system cost, which is the sum of the total amortized
payments to electricity-generating, pollution control, and transmission investments, delivered
fuel costs, total variable and fixed operating and maintenance (O&M) costs, and expenditures on
pollution (i.e., CO2) transportation and storage, subject to regulatory and other constraints:135

system cost = amortized payments to capital + delivered fuel costs + O&M costs + expenditures

on pollution transport and storage.

Note that system costs include transfers. For example, amortized payments to capital are
inclusive of corporate, state, and local taxes, investment tax credits, and interest payments.
Similarly, expenditures on pollution transport and storage account for 45Q tax credit payments.
This allows IPM to appropriately account for transfers, including taxes and subsidies (e.g., IRA
tax credits) that may target specific technologies and influence their adoption when modeling
generation and investment.136

135	For further details on IPM's objective function and model formulation, see Chapter 2, and in particular Section
2.2, of the IPM documentation, available at: https://www.epa.gov/power-sector-modeling. IPM's objective
function also accounts for energy and capacity payments for transmission.

136	See Section 3 and IPM documentation for further discussion of the representation of the IRA in IPM, fuel and
technology cost assumptions, and related uncertainties.

5-10


-------
System costs can be expressed in an alternative but equivalent way as the sum of
expenditures on real resources plus taxes and interest payments less subsidies received:

system cost = real resource expenditures + taxes + interest payments - subsidies.

Real resource expenditures are the expenditures on inputs required to produce electricity
(e.g., labor, materials, fuel), less any transfer payments, as estimated by the IPM model:137

real resource expenditures = real capital expenditures + fuel expenditures + O&M expenditures +
pollution transportation and storage input expenditures.

As described below, SAGE estimates the economy-wide impacts of the final rules using
the estimated change in real resource expenditures by the electricity sector, which is the
compliance cost estimate from IPM excluding transfers. This allows SAGE to capture the
expected change in the electricity sector's demand for these real resources due to the policy. To
determine the real resource expenditures, the estimates of system costs are separated into their
constituent components, to the extent feasible. For example, the real capital expenditures are
calculated as the amortized payment on capital excluding corporate, state, and local taxes,
interest payments, and tax credits for renewables and battery storage.138 The input expenditures
for pollution transportation and storage are calculated as the expenditures on pollution transport
and storage excluding 45Q payments. As described below, SAGE also uses the estimated
incremental subsidy payments from IPM to capture their effect on electricity prices, which is
important for modeling the output margin.139

The estimated compliance costs from IPM differ from the social costs of these rules for
several reasons. First, the estimated compliance costs from IPM include changes beyond real
resources costs, specifically transfers that should be excluded from an estimate of social costs.
Second, the compliance cost estimates from IPM do not account for all relevant margins of
substitution that the economy may use to respond to the final rules (e.g., electricity demand).140

137	These expressions show that expenditures on real resources projected by IPM represent the combined influence
of various taxes and subsidies (as well as other market and regulatory factors identified above).

138	The labor share of O&M costs is inclusive of taxes on labor, which are accounted for in the translation of these
resource costs to SAGE to avoid double counting them.

139	Unlike the analysis supporting the proposal for these rules, there is a de minimis expected change in the total 45 V
tax credit payments due to the final rules relative to the baseline.

140	In comparing frameworks for estimating social costs, the economics literature usually compares partial to general
equilibrium measures of social costs, focusing in particular on the second and third differences in this list. In

5-11


-------
Third, the compliance cost estimates from IPM do not account for the possibility of significant
cross-price effects and interactions with other pre-existing market distortions elsewhere in the
economy. Fourth, the compliance costs estimates from IPM do not account for reallocation
across sectors, potential reductions in aggregate investment, or the resulting effects on economic
growth. By construction, SAGE explicitly allows for these possible responses and is therefore
used to estimate social costs, while leveraging the insights that the detailed IPM provides on
compliance behavior and costs to the power sector from these final rules.

5.2.3.2 Overview of Linking Methodology

To model the economy-wide effects of the final rules, we calibrate the SAGE model
inputs that represent the impact of the final rules such that sectoral costs in a corresponding
partial equilibrium sub-model of SAGE (called SAGE-PE) align with the compliance costs
(excluding transfers) derived from the technology-rich IPM. This approach of aligning
compliance costs between the two models allows us to avoid confounding the estimate of
economy-wide effects with differences in the models' representations of sectors shared by both
IPM and SAGE.141 Care is given in translating IPM outputs for use in SAGE so that the two
models adequately capture equivalent compliance costs.142

Figure 5-2 provides an overview of the approach leveraging the IPM results to introduce
the incremental costs of the final rules into the SAGE model. In the first step (characterized as
Step 0), model differences in structure and accounting are reconciled by translating IPM
incremental system costs to a format consistent with the SAGE framework. This includes
aligning model years, distributing IPM costs to SAGE model inputs (by fuel, other materials,

certain cases, such as when market prices are not expected to change meaningfully, a compliance cost estimate
may provide a sufficient approximation of social costs. For these rules, IPM estimates changes in prices in
electricity, coal, and natural gas markets, and therefore, the compliance cost estimate may also differ
meaningfully from a partial equilibrium estimate of social costs for these markets.

141	The SAB (2017) noted that it will "often be necessary and appropriate for EPA to link a GE [general equilibrium]
model having a modest degree of detail to one or more PE models having greater detail. Linked models will
usually involve some degree of inconsistency in the definitions of overlapping variables and parameters, but that
may be acceptable given the increased degree of detail that a linked analysis could provide."

142	There are several valid approaches for linking models (see SAB 2017). In developing a strategy for linking IPM
and SAGE, we adhere to the following criteria: it should be theoretically sensible and produce reasonable results;
it should incorporate identical partial equilibrium responses across both SAGE and IPM without iteratively
linking the models (since IPM is proprietary); it should be practically implementable in the development of a
regulatory analysis; and the outcomes should be available to the public for the purposes of comment and
transparency.

5-12


-------
labor, and capital), attributing costs to production vintages, and removing transfer payments that
may be important for IPM to capture investment behavior but inappropriate for inputs into SAGE
as they would result in double counting.

The reconciled incremental costs are used to calibrate a representation of the final rules in
SAGE-PE, which is a partial equilibrium representation of the electricity sector (and related
primary energy sectors, such as the coal mining and natural gas) as defined from SAGE that
mimics the sectoral behavior of IPM, to the degree that is possible. While SAGE-PE does not
have the technology detail of IPM, it captures aggregate endogenous responses in electricity and
primary energy sector prices, input requirements, trade, and asset values of existing capital
resources. SAGE-PE does not include aspects of the economy represented in the full SAGE
model but that are not captured in IPM. This means that market outcomes in sectors other than
the electricity, coal mining and natural gas sectors, electricity demand, factor prices, and
constraints on factor supply are all treated as exogenous in SAGE-PE.

Step 0

Translate
compliance
costs (i.e.,
incremental
cost of
supplying
electricity) to

SAGE
accounting
framework.

Model
Inputs

Reconciled
compliance
costs.

Figure 5-2 Hybrid Linkage Approach for IPM and SAGE

Because SAGE-PE is a sub-model of SAGE, most of its model equations are described in
Marten et al. (2023). The subset of SAGE equations and variables that comprise SAGE-PE
include conditional profit maximizing production behavior, sub-national and foreign trade, and
market clearing conditions that equate supply and demand in the electricity, coal mining and
natural gas sectors. As in SAGE, SAGE-PE models optimal behavior through a series of
equilibrium conditions formulated as a mixed complementarity problem. Production and trade
are characterized through zero profit conditions that require unit costs to be greater than or equal

5-13


-------
to unit revenues. Market clearing conditions that equate supply and demand for the electricity,
coal mining and natural gas sectors determine their prices. A second set of market clearing
conditions are used to determine prices in regional trade markets. SAGE-PE maintain an
endogenous rental rate on extant capital to model the changes in the shadow value on existing
capital stock.

A common way to represent an environmental regulation in a CGE model is through a
productivity shock. This can be interpreted as requiring more inputs (e.g., control technologies)
to produce the same amount of output but in compliance with the regulation. In the SAGE and
SAGE-PE models, this is implemented through augmenting the reference productivity indices
denominated by input (materials, fuels, labor, and capital) and is described in detail in the model
documentation (Marten, 2023). The productivity shock is differentiated across model year,
regions, sectors, and production vintages. In the baseline, all productivity indices are set to unity
with the exception of those assigned to labor inputs which reflect projections of sector-
differentiated labor productivity.143

To align SAGE with IPM, the productivity shock is calibrated so that the compliance
costs are aligned between SAGE-PE and the IPM solution. The incremental SAGE-PE costs are
defined as the difference in production costs between the policy equilibrium and the baseline.
The productivity shock is adjusted to equate SAGE-PE and IPM incremental costs. Because
prices for factors and non-energy inputs are not endogenously determined in SAGE-PE the
incremental input costs for factors and non-energy inputs are driven through quantity demand
changes for labor, new capital, and material inputs. Incremental costs for electricity, coal mining
and natural gas inputs incorporate both changes in prices as well as input demand quantities.
Electricity production in SAGE-PE is exogenous except for adjustments necessary to satisfy
reductions or increases in electricity input demands in the electricity sector and primary energy
sectors in response to the final rules. The calibrated productivity shock is then passed to the full
SAGE model to generate social cost, distributional, and indirect impacts of the modeled policy,

143 The SAGE model was modified to allow production with extant capital to require incremental new capital for
compliance (e.g., pollution control retrofits, fuel switching). This modification is implemented by defining an
additional productivity index associated with new capital demands in production with extant capital.

5-14


-------
where model years 2026 and beyond are endogenously determined. See Schreiber et al. (2023)
for more details on the linking approach.

5.2.3.3 Translating IPM Outputs into SA GE Inputs

IPM produces detailed cost and emissions outputs by model plant (or aggregate
representations of unit-level information of existing generators, or characterizations of new or
retrofit/retire options) and wholesale electricity price impacts by IPM region. This detailed
information is important for quantifying the sectoral compliance behavior attributed to a
regulatory shock. However, to link IPM and SAGE to capture the broader economy-wide
impacts, IPM costs need to be translated to SAGE factors and commodities. Table 5-3
summarizes the key dimensions of IPM used to calibrate the inputs for the SAGE model. Key
variables include capital costs, fuel costs, and fixed and variable operations and maintenance
costs. Capital costs are reported both as overnight capital costs and amortized capital payments.
Overnight capital costs reflect the total value of the resources used to install a piece of capital
"overnight," or without any financing costs associated with loan repayment. In reality, these
expenditures are not paid immediately but rather spread out over a fixed time period with interest
via amortized capital payments. The "cost" of capital in IPM is a combination of a rate of return,
tax payments, and financing charges (embodied in the capital charge rate) and is used to amortize
payments over the lifetime of the capital investment. Costs are further denominated by IPM
region, fuel type, and generator vintage.

Table 5-3 IPM Cost Outputs

Time Periods

Cost Categories

IPM Regions

Generator Vintage

2028-2055

Overnight capital costs
Amortized capital
payments
Fuel costs

Fixed operations and
maintenance costs
Variable operations and
maintenance costs

67 IPM Regions

Existing
New

5-15


-------
IPM incremental costs are translated into the SAGE framework by: (1) mapping IPM
model years to SAGE model years;144 (2) mapping IPM regions to SAGE regions; (3) splitting
delivered fuel costs to separate transportation costs; (4) mapping variable operations and
maintenance costs to specific inputs in SAGE according to the reference cost structure in the
model; (5) attributing fixed operations and maintenance costs to labor; (6) attributing incremental
costs on existing and new generation to production with extant and new capital, respectively;145
and (7) removing taxes and transfers from capital payments using the difference between the
capital charge rate and the capital recovery factor to recover the real resource costs. Unlike at
proposal, the IPM analysis for the final rules projects very little incremental hydrogen demand,
so EPA did not need to make assumptions about hydrogen input requirements for modeling it in
SAGE.

Aligning the SAGE model with IPM is complicated by the difference in how each model
accounts for capital payments. First, taxes and transfers (e.g., finance payments) need to be
removed from capital costs to recover the real resource requirements for inputs to SAGE.

Second, differences in representation of capital between the two models needs to be reconciled;
SAGE accounts for capital as a cumulatively depreciated asset that represents the aggregate
physical capital stock in the U.S., whereas IPM defines capital more specifically with
heterogeneous terms and costs by technology. The models can be aligned by either targeting
incremental overnight capital costs (e.g., the magnitude and timing of the resource change) or
through targeting amortized capital payments. Because the accounting for capital is different
between models, the former approach can lead to significant differences in capital payments
between models. Therefore, the second approach is used to align incremental amortized
payments to capital that exclude tax payments when calibrating the productivity shock. Because
the representation of capital is different between the models, differences in induced investment in

144	IPM year 2028 is mapped to SAGE model year 2026. Subsequent IPM years (2030-2055) are mapped to the
SAGE model year that is one year later (2031-2056). Because SAGE has a longer time horizon than IPM (to
2081), IPM incremental costs in 2055 are expected to continue into the future and are mapped to SAGE model
years 2061-2081.

145	Production with extant and new capital is not equivalent to differentiating existing and new generation in the IPM
modeling framework. For example, the lifespan of existing generators in IPM can be extended through
investments in ways that are not directly comparable to production with extant capital in the SAGE model. In
this analysis, we attribute all incremental costs associated with existing generation to production with extant
capital until 2051. Incremental costs on existing generation in model years after 2051 are levied on production
with new capital.

5-16


-------
the capital stock from targeting consistent amortized payments can be thought of as a translation
of payments (e.g., a means to translate a fixed term investment into a cumulatively depreciated
asset).

Because SAGE does not include an explicit representation of the Inflation Reduction Act
(IRA) in the baseline, the model linkage methodology must be adjusted to account for IRA
investment and production subsidies (i.e., ITC/PTC and 45Q). The SAGE-PE model is calibrated
to match both the real resource requirements for the expected compliance pathway and the
impact of the IRA subsidies on the compliance expenditure for the electricity sector. To
accomplish this, the real resource requirements represented by the IRA subsidies are included in
the incremental costs of the final rules (i.e., the incremental costs exclude subsidy payments).146
To avoid overstating electricity price impacts and the social costs of higher electricity prices, the
net tax rate on electricity sector production is also adjusted within the calibration of the SAGE-
PE model to reflect the IRA subsidies that offset a portion of the compliance expenditures for the
electricity sector. This approach allows the model to explicitly capture the private costs faced by
the electricity sector, the upstream and downstream impacts of the resource requirements for the
subsidized technologies and fuels, and changes to government budgets associated with the use of
subsidies. The SAGE model is closed by assuming the government budget is balanced through
lump sum transfers with households. Aggregate changes in government budgets can occur in
model simulations due to changes in the use of the IRA subsidies and changes in revenues from
other taxes (e.g., output, capital, and labor) as the economy adjusts in response to the final rules.
Additional features of the IRA are not explicitly represented in SAGE at this time.

5.2.4 Results

This section summarizes the estimated economy-wide impacts of the final rules. We
report the SAGE model outcomes from implementing the described framework for linking
SAGE with IPM. Results include aggregate social costs of the final rules, changes to gross
domestic product (GDP) and its components, national sectoral output, national sectoral labor

146 ITC/PTC subsidies are levied on capital whereas the 45Q subsidy is shared across inputs according to an assumed
cost structure for carbon capture and storage based on a combination of both the natural gas extraction sector in
the SAGE model and the cost structure of pipeline transportation from the Bureau of Economic Analysis. The
adopted approach for modeling the costs of carbon capture and storage approximately align with information
found in Ortiz et al. (2013) and McFarland and Herzog (2006).

5-17


-------
demand changes, and distributional impacts across regions and households. Note that SAGE
does not currently estimate changes in emissions nor account for environmental benefits.

5.2.4.1 Economy-wide Social Costs

Table 5-4 presents the economy-wide, general equilibrium social costs of the final rules,
calculated as equivalent variation. In this context, equivalent variation is an estimate of the
amount of money that society would be willing to pay to avoid the compliance requirements of
the final rules, setting aside health, climate, and other benefits (quantified or described
qualitatively elsewhere in the RIA). For comparison, Table 5-4 also presents the compliance
costs estimated by IPM to be paid by the electricity sector for real resources - and which exclude
all transfer payments - mapped to the SAGE model years. For both the compliance costs and the
general equilibrium social costs, Table 5-4 presents the present value and annualized costs using
a discount rate of 4.5 percent, which is consistent with the internal discount rates in the SAGE
model.147 Compliance costs and transfer changes are presented as they are input into the SAGE
model. Section 5.2.3.3 discusses our assumptions for mapping IPM model years to SAGE model
years. Therefore, present value and equivalent annualized value estimates of the IPM inputs to
SAGE reported in Table 5-4 are not comparable to those reported in Sections 3 or 7 and are
provided here for transparency and as a point of comparison for the social cost estimates.

The annualized social cost estimated in SAGE for the finalized rules is approximately
$1.32 billion (2019 dollars) between 2024 and 2047 using the 4.5 percent discount rate that is
consistent with the internal discount rates in the model. Under the assumption that compliance
costs from IPM in 2056 continue until 2081, the equivalent annualized value for social costs in
the SAGE model is $1.51 billion (2019 dollars) over the period from 2024 to 2081, again using a
4.5 percent discount rate. This social cost estimate reflects the combined effects of the final
rules' requirements and interactions with IRA subsidies for specific technologies that are

147 The SAGE model estimates the present value of costs (i.e., equivalent variation) for each representative

household in the model and sums those estimates to calculate the present value of social costs. The present value
of costs for a representative household is based on its calibrated intertemporal utility function and the
equilibrium solution. Implicit in those estimates are endogenous discount rates that vary by household and over
time. The intertemporal preferences of households are calibrated such that their average discount rate over the
first 20 years of the model is consistent with a discount rate of 4.5 percent, which based on the effective marginal
capital tax rate in the model, is consistent with a 7.0 percent social rate of return to capital. See Section 3.4 of the
SAGE model documentation at https://www.epa.gov/environmental-economics/cge-modeling-regulatory-
analysis.

5-18


-------
expected to see increased use in response to the final rules. We are not able to identify their
relative roles at this time.

The general equilibrium social costs from SAGE differ from the compliance costs
excluding transfers from IPM for several reasons. First, the general equilibrium costs reflect
demand responses for electricity and energy inputs as the economy (inclusive of firms and
households) respond to the impacts of the final rules and shift production and consumption
behavior. Second, the general equilibrium costs account for interactions with pre-existing
distortions in the economy, mainly taxes and subsidies. Third, the general equilibrium costs
account for effects of reallocation, potential reductions in aggregate investment, and the resulting
effects on economic growth.

The compliance costs from IPM peak in the 2036 and 2041 SAGE model years. The
estimated social costs are spread out more evenly over the model time horizon as the economy
smooths out the impact by reallocating investment and consumption decisions. For the period
from 2024 to 2047, the equivalent annualized value for social costs in the SAGE model is
smaller than the annualized estimates of the compliance costs. However, the reported costs
between 2024 and 2047 represent a truncated estimate of the total social costs estimated in the
SAGE model, which accounts for changes in the economy after 2047 due to the forward-looking
nature of the model out to the end of the model horizon, 2081.148

148 The estimated social costs are about 20 percent lower (through 2047) or 6 percent higher (through 2081) than the
compliance costs from IPM (excluding transfers). The empirical literature finds that social costs for a set of
generic, illustrative single sector environmental regulation scenarios are 6 percent to 33 percent larger than
engineering-based compliance expenditures over the entirety of the model time horizon, noting that the specific
details of the individual regulation can significantly affect the social cost estimates (A. L. Marten et al., 2019).

5-19


-------
Table 5-4 Compliance Costs, Transfers, and Social Costs (billions of 2019 dollars)

SAGE Model Year

Compliance Costs - Input
to SAGE (Excluding
Transfers)

Change in Transfers -
Input to SAGE

General Equilibrium
Social Costs

2026

-1.46

0.10

1.14

2031

-0.32

0.05

1.27

2036

6.13

-4.79

1.37

2041

5.02

-4.44

1.48

2046

1.94

1.36

1.59

2051

0.94

0.49

1.73

2056

0.74

0.97

1.86

Present Value

25.17
1.66

-21.89
-1.44

19.94
1.32

(2024 to 2047, 4.5%)
Equivalent Annualized Value
(2024 to 2047, 4.5%)

Present Value

30.51

-16.41

32.36

(2024 to 2081, 4.5%)

Equivalent Annualized Value
(2024 to 2081, 4.5%)

1.42

-0.77

1.51

Notes: Social costs are calculated as equivalent variation. Present value and annualized cost estimates are calculated
by interpolating between SAGE model years and use a discount rate of 4.5 percent, which is consistent with the
internal discount rate in SAGE. Compliance costs and the change in the transfer amounts are calculated from the
IPM outputs. Transfers include changes in tax payments on capital, production and investment tax credits (e.g., the
45Q tax credit), and interest payments. Negative transfer values reflect decreases in net additional payments out of
the sector or increases in payments into the sector (e.g., subsidies) due to the final rules. Incremental monitoring and
reporting costs are not accounted for in this analysis. Compliance costs and transfer changes are reported as they are
input into the SAGE model. Section 5.2.3.3 discusses assumptions on mapping IPM model years to SAGE model
years. Present value and equivalent annualized value estimates are based on this mapping and are therefore not
directly comparable to estimates in Sections 3 and 7.

5.2.4.2 Impacts on GDP

The estimated percent change in real gross domestic product (GDP), or the real value of
the goods and services produced by the U.S. economy, and its components are presented in
Figure 5-3. GDP is defined as the sum of the value (price times quantity) of all market goods and
services produced in the economy and is equal to Consumption (C) + Investment (I) +
Government (G) + (Exports (X) - Imports (M)). The final rules are estimated to increase GDP in
2026 and 2031 by 0.015 percent and 0.020 percent due to increases in investment, but
subsequently result in a modest decrease in GDP with a peak reduction of 0.017 percent in 2036.
GDP is a measure of economic output and not a measure of social welfare. Thus, the expected

5-20


-------
social cost of a regulation will generally not be the same as the expected change in GDP (U.S.
EPA, 2015).149

Year

Figure 5-3 Percent Change in Real GDP and Components

Figure 5-3 also reports changes in the components of GDP from the expenditure side.
The final rules are expected to accelerate investments in the electricity sector, leading to a net
increase in aggregate investment in 2026 and 2031 (0.09 percent and 0.13 percent, respectively)
to augment the capital stock for compliance with the rule. Increased investment reallocates
resources away from consumption and as a result, consumption falls throughout the model time
horizon. Aggregate investment is expected to fall in later model years. The net trade balance is
expected to show modest declines in the initial years as relative prices change domestically due
to compliance with the final rules, shifting some purchases towards imports, though the effect is
expected to dissipate over time.

149 U.S. EPA Science Advisory Board (2017) notes: "GE models are strongly grounded in economic theory, which
allows social costs to be evaluated using equivalent variation or other economically-rigorous approaches.
Simpler measures, such as changes in gross domestic product or in household consumption, do not measure
welfare accurately and are inappropriate for evaluating social costs."

5-21


-------
5.2.4.3 Impacts on Output

SAGE endogenously models production for every sector in the economy, the final
demand for goods by households, and household behavior regarding savings and labor supply.
Therefore, the general equilibrium solution incorporates estimates of how changes in the prices
for electricity, coal mining, and natural gas inputs due to the final rules affect input demand in
other sectors of the economy. The general equilibrium solution also estimates changes in final
demand from households, the reallocation of resources across sectors and time, and changes in
household investment and labor choices as relative prices change (including wages, rental rates
on capital, and returns on natural resources).

Figure 5-4 presents the percent change in national output for the electricity, coal mining,
and natural gas extraction and distribution sectors in model years 2026, 2031, 2036, and 2041.
These output changes are based on what is expected to occur in the electricity sector as well as
changes elsewhere in the economy. As expected, the largest economy-wide changes,
denominated in percent change, are concentrated in these sectors. These changes reflect the
estimated shifts in generation sources in addition to an economy-wide demand response to
increases in electricity prices. As the price of electricity rises, the economy is expected to reduce
demand for electricity through a variety of pathways. Similarly, output changes in the coal
mining and natural gas reflect changes in both the electricity sector and the broader economy
(inclusive of import and export changes).

5-22


-------
Electric power¦

Coal mining-

Natural gas

	1	1		1	1	 	1	1		1	1	 	1	1		1	1	 	1	1		1	1	

-3% -2% -1% 0% 1% 2% -3% -2% -1% 0% 1% 2% -3% -2% -1% 0% 1% 2% -3% -2% -1% 0% 1% 2%

% Change

Figure 5-4 Percent Change in Sectoral Output (Electricity, Coal, Natural Gas)

Measured in terms of percent change from the baseline, output changes in other sectors of
the economy are expected to be smaller relative to the electricity, coal mining, and natural gas
sectors. Figure 5-5 presents the percent change in output for the remaining sectors of the
economy as reflected in the SAGE model for 2026, 2031, 2036, and 2041 (note the axis scale is
different than in Figure 5-4). Modest output reductions are estimated in some relatively more
energy intensive sectors (e.g., chemical manufacturing) and those that support coal use in the
electricity sector (e.g., transportation) whereas output increases in sectors associated with capital
formation in 2026 to support investments needed to comply with final rules.

Combining output impacts across all sectors in the economy, Figure 5-6 presents the
estimated net economy-wide percent changes in output in 2026, 2031, 2036, and 2041.

Aggregate U.S. production is expected to increase by 0.01 percent in 2026 and by 0.02 percent in
2031, with declines of similar magnitude in subsequent years. The model suggests modest
increases in production in 2026 and 2031 in capital forming sectors in anticipation of rule
requirements, resulting in an overall increase in output. In later model years, output reductions in
the electricity sector, primary energy sectors, and energy-intensive sectors slightly outweigh
output increases elsewhere in the economy.

5-23


-------
T ransportation
Chemical manufacturi

ng

Water, sewage, and o
ther utilities

Petroleum refineries
Healthcare services

Food and beverage ma
nufacturing

Crude oil

Metal ore and nonmet
alic mineral mining

j_ Agriculture, forestr
O yt fishing and hunting
o

Q> Wood product manufac
co .

turing

Truck transportation

Plastics and rubber
products manufacturing

Fabricated metal pro
duct manufacturing

Other manufacturing

Electronics and tech
nology manufacturing

Primary metal manufa
cturing

Transportation equip
ment manufacturing

Cement manufacturing

Construction

~

]

*