January 18, 2019Updated May 19, 2020
EPA Reconsiders Basis forBenefits of Mercury and Air Toxics Standards
In late 2018Limits
In April 2020, the U.S. Environmental Protection Agency
(EPA) proposed to reverse its previous determinations that
concluded that limits on hazardous air pollutants
(HAPs) from coal- and oil-fired
power plants are not
“appropriate and necessary” (A&N) under Clean
Air Act
(CAA) Section 112(n) (EPA, (1) (“Reconsideration of
Supplemental Finding and Residual Risk and Technology
Review,” December 27, 2018) (hereinafter, “A&N
proposal”). While the A&N proposal would not revoke the
mercury and acid gas emissions limits established in the
2012 Mercury and Air Toxics Standards (MATS) Rule, it
has raised questions about whether EPA will take additional
action on MATS.
The A&N proposal also reveals changes in EPA’s
interpretation of the CAA and use of benefit-cost analysis.
EPA’s analysis for the 2018 A&N proposal excludes cobenefits—the human health benefits from reductions in
pollutants not targeted by MATS—from its consideration of
whether MATS is “appropriate and necessary” under CAA
Section 112(n). With this exclusion, the 2018 analysis finds
that monetized costs outweigh monetized benefit estimates
by several orders of magnitude. EPA previously
determined—in 2000, 2012, and 2016—that it was
appropriate and necessary to regulate hazardous air
pollutants from power plants. EPA’s 2011 MATS analysis
counted co-benefits and concluded that the rule’s benefits
outweighed the costs. This In Focus provides background
on MATS, discusses EPA’s reconsideration of benefits and
costs, and concludes with potential issues for Congress. A
detailed summary of the proposal and legal issues is beyond
the scope of this product.
Background
Hazardous air pollutants (HAPs), also known as air toxics,
are pollutants known or suspected to cause cancer or other
serious health effects, such as reproductive issues or birth
defects. Among the HAPs emitted by power plants,
mercury has been the principal HAP of concern. Mercury is
a neurotoxin that travels through the air to water, where it is
converted to methylmercury, which moves through the food
chain as larger organisms consume smaller ones.
Consumption of fish and shellfish contaminated with
methylmercury is the primary source of mercury exposure
for humans. Fetuses and children are particularly vulnerable
to methylmercury, which may impair neurological
development. Mercury exposure at high levels may also
harm the brain, heart, kidneys, lungs, and immune system
(EPA, “Basic Information about Mercury”).
The CAA Amendments of 1990 required EPA to study the
“hazards to public health reasonably anticipated to occur”
from HAPs emitted by coal- and oil-fired power plants,
after imposition of other CAA requirements (42 U.S.C.
§7412(n)). It also required EPA to examine the health and
environmental effects of mercury emissions from these
sources and available control technologies and their costs,
and to determine whether regulation of power plant HAPs
was “appropriate and necessary” (42 U.S.C. §7412(n)).
In 2000, EPA determined that it was “appropriate and
necessary” to regulate coal- and oil-fired power plants
under CAA Section 112, and it added them to the Section
112 list of source categories.
In 2005, EPA changed course. EPA withdrew the 2000
“appropriate and necessary” finding and finalized a rule to
remove coal- and oil-fired power plants from the Section
112 list. Rather than establishing maximum achievable
control technology (MACT) standards to control mercury
emissions under Section 112, EPA promulgated a cap-andtrade program to limit power plant mercury emissions under
Section 111. These 2005 actions, however, were vacated by
the U.S. Court of Appeals for the DC Circuit in 2008 before
EPA could implement the cap-and-trade program. The DC
Circuit ruled that EPA unlawfully delisted coal- and oilfired power plants from the Section 112 list because EPA
failed to comply with the statutory delisting criteria.
In 2012, EPA reaffirmed the 2000 “appropriate and
necessary” finding and promulgated MATS. The MATS
Rule, which remains in effect today, established MACT
standards to reduce mercury and acid gases from most
existing coal- and oil-fired power plants.
EPA’s accompanying analysis, published in late 2011,
estimated that the annual benefits would be between $37
billion and $90 billion in 2016. Nearly all of the monetized
benefits were from the rule’s particulate matter co-benefits.
EPA monetized one of the expected mercury impacts—IQ
loss to children exposed to mercury from recreationally
caught freshwater fish—but could not monetize other
mercury impacts. Such non-monetized impacts may
include, according to EPA, other neurologic effects (e.g.,
memory and behavior), cardiovascular effects, and effects
on wildlife. Factors that precluded comprehensively
monetizing mercury and other HAP benefits from the
MATS rule included gaps in toxicological data,
uncertainties in estimating human effects based on animal
experiments, and insufficient economic research to translate
the health and environmental effects to dollar value terms.
Regulatory impact analyses dating back to 2000 have
acknowledged the difficulty in monetizing HAP reduction
benefits, emphasizing that the lack of monetized estimates
does not mean the benefits lack value. Previous
administrations have concluded that such benefits justify
emission standards, albeit under different authorities of the
CAA. For example, during the George W. Bush
https://crsreports.congress.gov
EPA Reconsiders Basis for Mercury and Air Toxics Standards
Administration, the EPA’s 2004 analysis of a proposed
action to reduce power plant mercury emissions concluded
that the non-monetized benefits were “large enough to
justify substantial investment in emission reductions”
(EPA, “Benefit Analysis for the Section 112 Utility Rule,”
p. 49).
Numerous parties petitioned the courts to review MATS.
Among other things, some petitioners disagreed with EPA’s
conclusion that it was not appropriate to consider costs
when making an “appropriate and necessary” finding under
CAA Section 112. In 2015, the Supreme Court agreed with
the petitioners and remanded the rule for further
consideration (Michigan v. EPA, 135 S. Ct. 2699 (2015)).
In 2016, EPA finalized a supplemental “appropriate and
necessary” finding based on its review of the 2012 rule’s
estimated costs. EPA used two approaches. The agency’s
first and preferred approach evaluated whether compliance
costs were reasonable based on the industry’s historical
annual revenues and capital expenditures, retail electricity
rates, and potential impacts on reliability. The second
approach involved a direct comparison of the estimated
compliance costs and the estimated benefits, which
included co-benefits. The 2016 Supplemental Finding
concluded that under both approaches, it is appropriate and
necessary to regulate HAPs, including mercury, from power
plants after considering regulatory costs.
2018 Reconsideration
In late 2018, EPA proposed to reverse the 2016
Supplemental Finding based on its new conclusion that the
monetized compliance costs greatly outweigh the
monetized benefits of HAPs reductions.
The Trump Administration determined that EPA’s benefitcost comparison for the 2016 Supplemental Finding was
flawed because it included co-benefits from non-HAP
pollutants. While EPA acknowledged that estimation of all
benefits and costs, including ancillary impacts, is consistent
with federal guidance and standard economic practice, the
agency concluded that it erred when it gave benefits (HAP
reductions) and co-benefits (non-HAP reductions) equal
consideration when making its 2016 “appropriate and
necessary” finding under Section 112(n). The 2018 A&N
proposal describes CAA Section 112(n)(1)(A) as “focused
on hazards resulting from HAP-specific emissions” and
concludes “it is not proper to place much weight” on nonHAP co-benefits (p. 27).
This interpretation marks a change from the Obama
Administration, which concluded that nothing in the CAA
prohibits EPA from considering co-benefits in a benefitcost analysis for an “appropriate and necessary” finding.
The 2016 Supplemental Finding characterized the non-HAP
reductions as a “direct result of achieving the HAP
emission limits under MATS” and included these
monetized co-benefits in the total benefits estimate. EPA’s
2016 Supplemental Finding also pointed to the legislative
history, noting that Senate Report 101-228 “recognized that
MACT standards would have a collateral benefit of
controlling criteria pollutants as well and viewed this as an
important benefit of the air toxics program” (81 Federal
Register 24439, April 25, 2016).
EPA’s 2018 proposal revised the 2016 benefit-cost
comparison by excluding the monetized co-benefits. This
resulted in the estimated compliance costs—$9.6 billion in
2015—outweighing the monetized HAP benefit
estimates—$0.5 million to $6 million, depending on the
discount rate, in 2016. (As in previous analyses, compliance
cost projections for 2016 were not available due to model
configuration.) EPA concluded that HAPs regulation is not
appropriate and necessary under Section 112(n) because
monetized costs exceed monetized HAP benefits.
EPA also asserted that the non-monetized HAP benefits are
less than the monetized compliance costs but did not
explain the basis for this conclusion. EPA stated, without
further elaboration, that identifying the unquantified HAP
benefits “is not sufficient, in light of the gross imbalance of
monetized costs and HAP benefits, to support a finding that
it is appropriate and necessary to regulate” power plants
under CAA Section 112 (2018 proposal, p. 27).
Available rulemaking documents do not explain why EPA
disagrees with its previous conclusions about nonmonetized HAP benefits. EPA’s 2011 MATS analysis
stated that non-monetized benefits “could be substantial,
including the overall value associated with HAP reductions,
value of increased agricultural crop and commercial forest
yields, visibility improvements, and reductions in nitrogen
and acid deposition and the resulting changes in ecosystem
functions” (EPA, Regulatory Impact Analysis for the Final
Mercury and Air Toxics Standards, 2011, p. 28).
Potential Issues for Congress
In addition to uncertainty about whether EPA will take
additional action on the MATS emission standards, the
2018 proposal raises questions about how EPA should
factor benefits and costs into regulatory decisions. As
discussed above, it raises questions about EPA’s
consideration of non-monetized HAP benefits and whether
excluding co-benefits is consistent with the CAA. Such
questions are relevant in light of the Administration’s
reconsideration of existing CAA regulations.
Federal guidance directs agencies to assess whether the
benefits of a proposal justify the costs but does not require
monetized benefits to outweigh monetized costs. The
guidance recognizes that quantified benefit and cost
estimates may not capture all anticipated benefits and costs
and directs analysts to identify non-quantified impacts “of
sufficient importance to justify consideration in the
regulatory decision” (OMB Circular A-4, p. 10).
Determining whether non-monetized health and
environmental benefits and co-benefits (monetized or not)
justify monetized compliance costs is inherently difficult
and may continue to spark debate. In 2015, the Supreme
Court rejected arguments that costs are irrelevant to an
“appropriate and necessary” finding under CAA Section
112(n), but it did not address whether EPA has authority to
consider monetized co-benefits in evaluating the cost of
MATS (Michigan v. EPA, 135 S. Ct. 2699, 2711 (2015)).
https://crsreports.congress.gov
EPA Reconsiders Basis for Mercury and Air Toxics Standards
IF11078
Kate C. Shouse, Analyst in Environmental PolicyApril 16, 2020). The 2020 A&N rule reversed
prior A&N determinations, which led to the 2000 listing of
coal- and oil-fired power plants as a major source of HAPs
and the 2012 Mercury and Air Toxics Standards (MATS)
limiting those HAPs. Notwithstanding the 2020 A&N rule,
the 2012 MATS limits remain in effect for power plants
because EPA determined that it could not meet the criteria
under CAA 112(c)(9) to delist them. Furthermore, the A&N
finding does not change the regulatory status of other
pollution sources because CAA Section 112(n)(1) applies
only to power plants. Some have raised questions about
why EPA reversed the A&N finding and how it might
affect regulated entities. For example, some power plant
owners are concerned the A&N reversal may compromise
their ability to recover from ratepayers the costs of
installing MATS pollution controls. Others find this
unlikely, but legal challenges to the 2020 A&N rule are
expected.
The 2020 A&N rule reveals a change in EPA’s
interpretation of a unique statutory provision—Section
112(n)(1)—which may nonetheless set a precedent for
EPA’s consideration of benefits under other CAA
authorities. EPA stated that the 2020 A&N rule corrects
errors in the agency’s consideration of benefits in a prior
A&N finding. In its determination for the 2020 A&N rule,
EPA excluded from consideration any co-benefits to human
health from reductions in pollutants not targeted by MATS.
This In Focus discusses EPA’s reconsideration of benefits
and costs and potential issues for Congress. Section 112(c)
delistings and legal issues are beyond this product’s scope.
Historical EPA Actions
Hazardous air pollutants (HAPs) are pollutants known or
suspected to cause cancer or other serious health effects,
such as reproductive problems or birth defects. Among the
HAPs emitted by power plants, mercury has been of
principal concern. Mercury, which occurs naturally in coal,
travels through the air to water, where it is converted to
methylmercury and moves up the food chain. Consumption
of fish and shellfish contaminated with methylmercury is
the primary source of human mercury exposure. Fetuses
and children are particularly vulnerable to methylmercury
exposure, which may impair neurological development.
Methylmercury exposure at high levels may harm the brain,
heart, kidneys, lungs, and immune system.
CAA Section 112(n)(1) required EPA to study the “hazards
to public health reasonably anticipated to occur” from
HAPs emitted by power plants after imposition of other
CAA requirements. It also required EPA to examine the
health and environmental effects of mercury emissions
from these sources, available control technologies and their
costs, and whether regulation of power plant HAPs was
“appropriate and necessary” (42 U.S.C. §7412(n)).
In 2000, EPA determined that it was appropriate and
necessary to regulate hazardous air pollutants from coaland oil-fired power plants. This determination required
EPA to take additional steps to regulate HAPs. EPA added
coal- and oil-fired power plants to the Section 112 list of
source categories in 2000.
In 2005, EPA changed course. EPA withdrew the 2000
A&N finding and finalized a rule to remove coal- and oilfired power plants from the Section 112 list. Instead, EPA
promulgated a cap-and-trade program to limit power plant
mercury emissions under Section 111. The U.S. Court of
Appeals for the D.C. Circuit vacated these 2005 actions,
however, and ruled that EPA unlawfully delisted coal- and
oil-fired power plants from the Section 112 list because
EPA failed to comply with the statutory delisting criteria.
In 2012, EPA reaffirmed the 2000 A&N finding and
promulgated the Mercury and Air Toxics Standards Rule.
The rule, which remains in effect, established emissions
standards to reduce mercury and acid gases from most
existing coal- and oil-fired power plants.
EPA’s accompanying analysis, published in 2011, projected
annual benefits between $37 billion and $90 billion in 2016.
Nearly all of the monetized benefits were from the rule’s
particulate matter co-benefits. EPA monetized one of the
expected mercury impacts—intelligence quotient loss to
children exposed to mercury from recreationally caught
freshwater fish—but could not monetize other mercury
impacts. Such non-monetized impacts may include other
neurologic effects (e.g., memory and behavior),
cardiovascular effects, and effects on wildlife.
EPA’s regulatory impact analyses have historically reported
difficulty in monetizing HAP reduction benefits but have
also noted that the lack of monetized estimates does not
mean the benefits lack value. Previous Administrations
concluded that such benefits justify emission standards,
albeit under different CAA authorities. For example, EPA’s
2004 analysis of a rule to reduce power plant mercury
emissions concluded that non-monetized benefits were
“large enough to justify substantial investment in emission
reductions” (“Benefit Analysis for the Section 112 Utility
Rule”).
https://crsreports.congress.gov
EPA Reconsiders Benefits of Mercury and Air Toxics Limits
Numerous parties petitioned the courts to review MATS.
Among other things, some petitioners disagreed with EPA’s
conclusion that it was not appropriate to consider costs
when making an A&N finding under CAA Section 112. In
2015, the Supreme Court agreed with the petitioners and
remanded the rule for further consideration, but it did not
address whether EPA has authority to consider monetized
co-benefits in evaluating the cost of MATS (Michigan v.
EPA, 135 S. Ct. 2699 (2015)).
In 2016, EPA finalized a supplemental A&N finding based
on its review of the 2012 rule’s estimated costs. EPA
evaluated whether compliance costs were “reasonable” and
compared the estimated compliance costs to the estimated
benefits, including co-benefits. EPA concluded that it was
appropriate and necessary to regulate mercury and other
HAPs from power plants after considering regulatory costs.
2020 Appropriate and Necessary Finding
In 2020, EPA reversed the 2016 supplemental finding,
concluding that HAPs regulation is not appropriate and
necessary under Section 112(n) because monetized costs
exceed monetized HAP reduction benefits. The 2020 A&N
rule revised the 2016 benefit-cost comparison by excluding
the monetized co-benefits. This exclusion resulted in the
estimated compliance costs ($9.6 billion in 2015),
outweighing the monetized HAP benefits ($0.5 million to
$6 million, depending on the discount rate, in 2016).
The 2020 A&N rule concluded that EPA’s benefit-cost
comparison for the 2016 supplemental finding was flawed
because it included co-benefits from non-HAP pollutants.
While EPA acknowledged that estimation of all benefits
and costs, including ancillary impacts, is consistent with
federal guidance, the agency concluded that it had erred
when it gave equal consideration to benefits (HAP
reductions) and co-benefits (non-HAP reductions) when
making its 2016 A&N finding under Section 112(n). The
2020 A&N rule concluded that an A&N finding under
Section 112(n)(1) must instead be justified
“overwhelmingly” by HAP reduction benefits.
This interpretation marks a change from that of the prior
Administration’s EPA, which concluded that nothing in the
CAA prohibits EPA from considering co-benefits in a
benefit-cost analysis for an A&N finding. The 2016
supplemental finding characterized the non-HAP reductions
as a “direct result of achieving the HAP emission limits
under MATS” and included these monetized co-benefits in
the total benefits estimate. EPA’s 2016 supplemental
finding also pointed to the CAA legislative history, noting
that Senate Report 101-228 expected that HAP limits
“would have a collateral benefit of controlling criteria
pollutants as well and viewed this as an important benefit of
the air toxics program” (81 Federal Register 24439, April
25, 2016).
The 2020 A&N rule also reveals a potential shift in EPA’s
assessment of non-monetized benefits. EPA’s 2011 MATS
analysis stated that non-monetized benefits “could be
substantial, including the overall value associated with HAP
reductions, value of increased agricultural crop and
commercial forest yields, visibility improvements, and
reductions in nitrogen and acid deposition and the resulting
changes in ecosystem functions” (“Regulatory Impact
Analysis for the Final Mercury and Air Toxics Standards,”
2011). The 2020 A&N rule acknowledges HAP reduction
benefits from MATS that cannot be monetized but finds
that the value of those benefits is unlikely to alter the
agency’s conclusion. Specifically, EPA determined that the
costs of the MATS would likely outweigh the HAP
reduction benefits even if the agency were able to monetize
all of them. EPA noted that many of the non-monetized
HAP reduction benefits relate to illnesses, which have had
lower economic values than mortality effects in its past
analyses.
Federal Guidance on Benefit-Cost
Analysis
Separate from the CAA, federal guidelines inform EPA’s
benefit-cost analyses. For example, Office of Management
and Budget Circular A-4 directs agencies to assess whether
the benefits of a proposal justify the costs. It does not
require monetized benefits to outweigh monetized costs.
Circular A-4 recognizes that quantified benefit and cost
estimates may not capture all anticipated benefits and costs
and directs analysts to identify non-quantified impacts “of
sufficient importance to justify consideration in the
regulatory decision.”
EPA has also developed its own guidance, Guidelines for
Preparing Economic Analyses, to complement Circular A-4
and other guidance. EPA has recently drafted updates to its
Guidelines, which its Science Advisory Board is reviewing.
Among other things, the draft update affirms that economic
analysis should account for all benefits and costs of a
proposal and advises distinguishing benefits from cobenefits. It also advises considering whether “more
economically efficient or appropriate ways” are available to
obtain co-benefits if the proposal is expected to “induce
large” co-benefits. In addition, EPA’s forthcoming
proposed rule is expected to provide guidance regarding the
agency’s approach to benefits assessment.
Potential Issues for Congress
EPA’s approach to benefits in the 2020 A&N rule may set a
precedent for future rulemakings. Although EPA linked the
2020 A&N analysis to its interpretation of CAA Section
112(n)(1), the EPA Administrator has said on the record
that the analysis foreshadows a more general analytical
approach in future air pollution rulemakings. EPA has not
specified whether that means it would exclude or give less
weight to co-benefits in other air rulemakings. Such
modification of co-benefit estimates would result in less
favorable assessments, on a benefit-cost basis, of the rules.
As EPA develops its benefits proposal and updates its
Guidelines, Congress may exercise oversight over how
EPA factors benefits and costs into regulatory decisions.
Issues include consideration of non-monetized benefits and
whether excluding co-benefits is consistent with the CAA.
Kate C. Shouse, Analyst in Environmental Policy
https://crsreports.congress.gov
IF11078
EPA Reconsiders Benefits of Mercury and Air Toxics Limits
Disclaimer
This document was prepared by the Congressional Research Service (CRS). CRS serves as nonpartisan shared staff to
congressional committees and Members of Congress. It operates solely at the behest of and under the direction of Congress.
Information in a CRS Report should not be relied upon for purposes other than public understanding of information that has
been provided by CRS to Members of Congress in connection with CRS’s institutional role. CRS Reports, as a work of the
United States Government, are not subject to copyright protection in the United States. Any CRS Report may be
reproduced and distributed in its entirety without permission from CRS. However, as a CRS Report may include
copyrighted images or material from a third party, you may need to obtain the permission of the copyright holder if you
wish to copy or otherwise use copyrighted material.
https://crsreports.congress.gov | IF11078 · VERSION 2 · NEW4 · UPDATED