Order Code RL32203
CRS Report for Congress
Received through the CRS Web
Legal Analysis and Background on the EPA’s
Proposed Rules for Regulating Mercury
Emissions from Electric Utilities
Updated May 10, 2004
Michael John Garcia
American Law Division
Congressional Research Service ˜ The Library of Congress
Legal Analysis and Background on the EPA’s Proposed
Rules for Regulating Mercury Emissions from Electric
Mercury is a hazardous air pollutant (HAP) that can pose a serious public health
threat. In accordance with the authority provided by the Clean Air Act (CAA), the
EPA is considering regulation of mercury emissions from electric steam generating
units (electric utilities). Recently, the EPA proposed revising an earlier finding that
regulation of such emissions was warranted under section 112 of the CAA, because
the agency had failed to consider whether regulations promulgated under section 111
of the CAA would adequately address the health concerns associated with these
emissions. The EPA now proposes regulation of electric utilities’ mercury emissions
under section 111 through the adoption of a national “cap-and-trade” program.
A potential challenge to the EPA proposal could grow out of the statutory basis
for regulating electric utilities’ mercury emissions under section 111. If the EPA
implements the proposed revision and rule, it may argue that its interpretation should
be upheld given (1) the arguably conflicting amendments made to section 111 in
1990 and (2) the deference shown by the courts to agency interpretations. Opponents
may argue that the EPA’s interpretation is unreasonable because the conflicting
amendments do not represent a clear reversal of the previously explicit congressional
intent to prohibit an HAP listed under section 112 from being regulated under section
The EPA will also need to provide a substantive basis for revising its earlier
findings and adopting any proposed rule for mercury regulation, as the agency is
prohibited from executing its rulemaking authority in an arbitrary and capricious
manner. Accordingly the EPA would need to provide a reasoned analysis for
concluding that the public health threat posed by electric utilities’ mercury emissions
could be adequately addressed using section 111. The fact that the EPA previously
concluded that regulation was warranted under section 112 may impose a heightened
burden upon the EPA to prove that subsequent revisions do not violate CAA
Further, the agency may have difficulty “de-listing” electric utilities as a source
for regulation under section 112, which may require the EPA demonstrate that these
utilities are not a public health threat. The EPA may argue that this statutory
requirement does not apply because the agency listed electric utilities for regulation
only because it had previously failed to consider the possibility of regulating utilities
under section 111. Opponents of “de-listing” may argue that reinterpretation of
relevant CAA provisions should not be grounds to summarily rescind previous
One potential obstacle to EPA regulation under section 111 may be a 1998
settlement agreement requiring the EPA to take “final action” on a proposed rule to
regulate such emissions under section 112. Because this settlement decree was
apparently not incorporated into a judicial order, however, it does not appear that the
agreement could be used to restrict the EPA’s rulemaking discretion.
Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Rulemaking Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Regulation of Electric Utilities under Section 112 of the CAA . . . . . . . 4
Regulation under Section 111 of the CAA . . . . . . . . . . . . . . . . . . . . . . 5
Legal Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
EPA Authority to Regulate Electric Utilities under Section 111
of the CAA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
The EPA’s Discretion to Revise Previous Regulatory Findings
and Institute a “Cap-and-Trade Program” . . . . . . . . . . . . . . . . . . 14
Implications of the 1998 Settlement Agreement upon the Manner
in which the EPA Can Regulate Electric Utilities . . . . . . . . . . . . 16
Legal Analysis and Background on the
EPA’s Proposed Rules for Regulating
Mercury Emissions from Electric Utilities
Rulemaking Background. Mercury has long been recognized as a
hazardous air pollutant that can pose a significant threat to the public health. Indeed,
the Environmental Protection Agency (EPA) has recognized mercury emissions from
electric utility steam generating units (electric utilities) to be the “greatest concern to
public health from the [utilities] industry.”1 Accordingly, the EPA has recently
sought to regulate such emissions pursuant to the authority provided by the Clean Air
At least two provisions arguably provide the EPA with authority to regulate
mercury emissions from electric utilities. Section 112 of the CAA requires the EPA
to establish technology-based emission standards for listed hazardous air pollutants
(HAP), including mercury. Sources designated for regulation under section 112 are
required to adopt maximum achievable control technologies (MACT) to reduce their
hazardous emissions. Section 111, on the other hand, provides the EPA with the
authority to regulate designated pollutants under a broader and looser standard than
section 112. Under section 111, the EPA has authority to enact national “standards
of performance” for sources of pollutants that the Administrator of the EPA
(Administrator) concludes to pose a danger to the public health.3 What is at issue in
the present situation is whether mercury from electric utilities must be regulated
under 112 or whether the EPA may interpret the CAA to regulate such emissions
under the more flexible rules of section 111.
The EPA is only permitted to regulate HAP emissions from electric utilities
pursuant to section 112 if the Administrator determines that such regulation is
appropriate and necessary. On December 14, 2000, the Administrator of the EPA
concluded that, in accordance with section 112(n)(1)(A) of the CAA, the regulation
Regulatory Finding on the Emissions of Hazardous Air Pollutants from Electric Utility
Steam Generating Units, 65 Fed. Reg. 79,825 (Dec. 20, 2000) [hereinafter “December 2000
Regulatory Finding”] (quoting EPA, STUDY OF HAZARDOUS AIR POLLUTANT EMISSIONS
FROM ELECTRIC GENERATING UNITS — FINAL REPORT TO CONGRESS (Feb. 1998)).
42 U.S.C. §§ 7401 et seq.
Section 111 requires the agency to (1) establish nationally uniform performance standards
for designated air pollutants from categories of new industrial facilities and (2) require states
to implement and enforce separate standards of performance for existing sources of
designated air pollutants. 42 U.S.C. § 7411.
of mercury emissions from electric utilities was “appropriate and necessary,” and the
Administrator therefore added these units to the list of source categories to be
regulated under section 112.4
This regulatory finding had an additional implication relating to a 1998 legal
settlement agreement between the EPA and the National Resources Defense Council
(NRDC), which was made to avoid litigation over the agency’s regulation of mercury
emissions from electric utilities. The agreement was a modification of an earlier
settlement agreement between the EPA and NRDC, and provided that if the EPA
found that regulation of mercury emissions from electric utilities was appropriate and
necessary on or before December 15, 2000, the EPA would be required to “sign” a
notice of proposed rulemaking to regulate such admissions, pursuant to section 112
of the CAA, by December 15, 2003.5 Further, the 1998 agreement required the EPA
to take “final action”on the proposed rule by December 15, 2004.6 Because the EPA
determined that regulation of mercury emissions was appropriate and necessary
within the time period specified by the settlement agreement, pursuant to the
agency’s above-mentioned December 2000 regulatory finding, the 1998 settlement
agreement’s subsequent provisions were triggered.
On December 15, 2003, the EPA proposed to revise its December 2000
regulatory finding that the regulation of mercury emissions from electric utilities was
“appropriate and necessary” under section 112.7 The EPA argued that regulations
promulgated under section 111 of the CAA could effectively address the public
health threat posed by mercury emissions, and that regulation under section 112 was
therefore unnecessary.8 Further, the EPA proposed two alternative rules aimed at
significantly reducing mercury emissions from electric utilities.9 One proposal would
regulate mercury emissions under section 112 and require new and existing electric
utilities to adopt MACT controls to reduce their mercury emissions. The EPA
estimates that this proposal would reduce mercury emissions by 14 tons, or an
estimated 29 percent, by the end of 2007.10 This proposal also fulfills the EPA’s
December 2000 Regulatory Finding, supra note 1.
Natural Resources Defense Council v. EPA, Stipulation for Modification of Settlement
Agreement, No. 92-1415 (D.C. Cir. Nov. 17, 1998).
EPA, Proposed National Emission Standards for Hazardous Air Pollutants; and, in the
Alternative, Proposed Standards of Performance for New and Existing Stationary Sources:
Electric Utility Steam Generating Units, Dec. 15, 2003, available at
[http://www.epa.gov/mercury/1880338.pdf] [hereinafter “Proposed Rule and Revision”]
(published in 69 Fed. Reg. 4652 (Jan. 30, 2004)).
See id. at 39-40.
A more detailed, substantive discussion of these proposals can be found in CRS Report
RL31881: Mercury Emissions to the Air: Regulatory and Legislative Proposals (Feb. 10,
See EPA, Fact Sheet, EPA Proposes Options for Significantly Reducing Mercury
Emissions from Electric Utilities, Dec. 15, 2003, available at
obligation under the 1998 settlement agreement to propose a rule under section 112
for the regulation of electric utilities’ mercury emissions by December 15, 2003.
The second proposed rule announced by the EPA would instead establish
standards of performance for new and existing electric utilities under section 111 of
the CAA. With limited exceptions, this proposal would establish performance
standards for new sources which would be identical to the proposed MACT standards
for units selling more than 25 megawatts of electrical output.11 Both existing and
new electric utilities would be regulated through the creation of a market-based “capand-trade” program,12 which would “cap” overall mercury emissions and permit
utilities to purchase emissions credits from cleaner-operating utilities. The proposal
would be implemented in two phases. Pursuant to the first phase, due by 2010, a
national cap for mercury emissions would be established at 34 tons.13 The second
phase, due by 2018, would lower the mercury cap to 15 tons.14 Caps would also be
set at the state level, with states having authority to determine how to control
emissions from individual electric utilities.15 States would also required to amend
their respective State Implementation Plans (“SIPs”), used to verify compliance with
the CAA, so as to describe how they would reduce mercury emissions to levels
permitted under the cap. On February 24, 2004, the EPA proposed a model “capand-trade” program.16 States adopting the EPA’s model program would have their
SIP approved by the EPA automatically.17
The EPA is also co-proposing and soliciting comment on implementing a
“cap-and-trade program” under section 112, similar to that being proposed under
section 111 of the CAA.18 However, the EPA has discussed a “cap-and-trade”
program primarily in reference to section 111(d), and this report will deal with the
legality of regulating mercury emissions from electric utilities under that section.
On April 28, 2004, the EPA extended the public comment period concerning
the proposed mercury regulations until June 29, 2004, two months longer than
originally announced, and stated that it would push back final action in the
rulemaking proceeding until March 15, 2005. Although this extension would appear
to conflict with the EPA’s legal settlement agreement with the NRDC requiring final
agency action on a proposed mercury rule by December 15, 2004, the NRDC claimed
Proposed Rule and Revision, supra note 7, at 193.
A similar model program used to control the interstate transport of nitrogen oxides was
adopted by all 21 states (and the District of Columbia) to which the rule applied.
Id. at 256.
Id. at 257.
Id. at 265-66.
Supplemental Notice for the Proposed National Emission Standards for Hazardous Air
Pollutants; and, in the Alternative, Proposed Standards of Performance for New and Existing
Stationary Sources: Electric Utility Steam Generating Units; Proposed Rule, 69 Fed. Reg.
12398 (signed Feb. 24, 2004).
See Proposed Rule and Revision, supra note 7, at 265-66.
Id. at 2, 51-52, 56.
prior to the extension that it had formally notified the EPA that it would agree to
extend the settlement deadline until March 15 to allow the agency to extend the
public comment period and undertake further analysis of the proposed rules’ effect.19
Regulation of Electric Utilities under Section 112 of the CAA. Section
112 of the CAA was amended in 1990 to establish National Emissions Standards for
Hazardous Air Pollutants (HAP). Pursuant to this section, the EPA is generally
required to list categories and subcategories of “major sources” and certain “area
sources” of 188 listed HAP, including mercury.20 Upon listing a source category for
HAP, the EPA must establish emission standards that require designated sources to
install MACT that achieve “the maximum degree of reduction in emissions,” taking
into consideration the costs of compliance and non-air quality health and
environmental impacts.21 Standards for new sources of HAP “shall not be less
stringent than the most stringent emissions level that is achieved in practice by the
best controlled similar source.”22 Although standards promulgated for existing
sources may be less stringent than those for new sources, they can be no less
stringent than (1) the average emission limitation achieved by the best performing
12 percent of existing sources, if there are 30 or more sources in the category or
subcategory or (2) the best performing 5 similar sources in categories or
subcategories with less than 30 sources.23
If a source category or subcategory is properly designated as being subject to
emission standards promulgated under section 112, it can only be “de-listed” in
limited circumstances pursuant to section 112(c)(9) of the CAA. For the EPA to “delist” a listed source determined to emit HAP that may result in cancer in humans, a
determination must be made that no source in the category emits HAP in such
quantities as may cause a lifetime risk of cancer greater than one in one million to the
individual in the population that is most exposed to such pollutants from the source.24
In the case of a listed source determined to emit HAP that may result in noncancerous adverse health effects in humans or adverse environmental effects, such
as mercury, a determination must be made that (1) no source in the category emits
HAP at a level exceeding that which is adequate to protect public health with an
Press Release, Natural Resources Defense Council, NRDC Extends Deadline for Final
Rule on Power Plant Pollution (Apr. 27, 2004), avai l able at
42 U.S.C. § 7412(c)(1). A “major source” is any stationary source or group of stationary
sources location within a contiguous area under common control that emits or has the
potential to emit at least 10 tons per year of any HAP or 25 tons per year of any combination
of HAP. 42 U.S.C. § 7412(a)(1). An “area source” is any stationary source of HAP that is
not a major source. 42 U.S.C. § 7412(a)(2).
42 U.S.C. § 7412(d)(2).
42 U.S.C. § 7412(d)(3).
42 U.S.C. § 7412(c)(9)(B)(i).
ample margin of safety and (2) no adverse environmental effects will result from the
“de-listing” the source.25
Electric utilities are regulated under a separate provision of section 112 than
other potential sources of HAP. The 1990 CAA amendments required the EPA to
study the hazards to public health reasonably anticipated to be caused by continuing
HAP emissions from electric utilities.26 On the basis of the study’s results, the EPA
was required to regulate such utilities under section 112 if the Administrator found
that such regulation was “appropriate and necessary” given the public health effects
of the utilities’ HAP emissions.27
Regulation under Section 111 of the CAA. Section 111 of the CAA
provides for a somewhat looser and broader standard for regulation of pollutants than
section 112. Section 111(b) requires the Administrator of the EPA to establish
Federal “standards of performance” directly regulating new sources that the
Administrator determines to either cause or contribute significantly to air pollution
that may reasonably be anticipated to endanger public health or welfare.28 With
respect to existing sources, pursuant to section 111(d), each state must submit to the
Administrator for approval an SIP that imposes “standards of performance” and
provides for the implementation and enforcement of such standards for existing
sources of pollutants.29 Section 111(a)(1) defines a “standard of performance” as an
emissions standard for air pollutants that “reflects the degree of emission limitation
achievable through the application of the best system of emission reduction” that the
Administrator determines has been adequately demonstrated, taking into account
costs of achievement and non-air quality health and environmental impacts.30
The precise relationship between sections 111 and 112 of the CAA, and whether
the EPA has authority under section 111 to regulate sources of HAP listed under
section 112, is not expressly clear from the sections’ language. Section 111(b) makes
no mention of section 112. More problematic is the fact that while section 112 does
expressly limit the EPA’s ability to regulate new sources of HAP under section
111(d), the breadth of this limitation is not entirely clear because of conflicting
statutory language. Prior to the 1990 CAA amendments, the express language of
section 111(d) prevented the EPA Administrator from establishing standards of
performance pursuant to the section for any existing source of an emission listed
under section 112.31 In amending the CAA in 1990, however, Congress
simultaneously enacted two amendments to section 111(d) which are arguably
conflicting. The amendment to section 111(d) introduced by the Senate and
described as a “conforming amendment” provides that section 111(d) cannot be used
42 U.S.C. § 7412(c)(9)(B)(ii).
42 U.S.C. § 7412(n)(1).
42 U.S.C. § 7412(n)(1)(A).
42 U.S.C. § 7411(b).
42 U.S.C. § 7411(d).
42 U.S.C. § 7411(a)(1).
See Proposed Rule and Revision, supra note 7, at 181.
to regulate an existing source for any pollutant listed under section 112(b).32 In
contrast, the House amendment to section 111(d) only prevents the provision from
being used to regulate air pollutants “emitted from a source category which is
regulated under section 112.”33
This difference in language may significantly alter the scope of section 111(d).
Under the Senate amendment, the EPA could not regulate mercury emissions from
electric utilities pursuant to section 111(d), because mercury is listed as an HAP
under section 112(b). In contrast, under one interpretation of the House amendment,
the EPA would be permitted to regulate mercury emissions from electric utilities
pursuant to section 111(d) so long as the EPA had yet to promulgate regulations for
HAP emissions by electric utilities under section112.
It should be noted that while the Statutes at Large includes both the House and
Senate amendments to section 111(d), the United States Code only reflects the
language of the House amendment.34 However, Title 42 of the U.S. Code, which
contains the CAA, has yet to be codified into positive law.35 When the language of
the Statutes at Large conflicts with a corresponding section in the U.S. Code that has
not been codified, the language of the Statutes at Large is controlling.36
A decision by the EPA to regulate mercury emissions from electric utilities
pursuant to section 111 is likely to raise three legal concerns. First, does the CAA in
certain circumstances permit the EPA to regulate such emissions under section 111?
Second, even if such authority exists, can the EPA revise its December 2000
regulatory finding and establish a national “cap-and-trade” program? Third, does the
1998 settlement agreement obligate the EPA to regulate mercury emissions from
electric utilities pursuant to section 112 of the CAA? This section will provide
analysis of these issues.
EPA Authority to Regulate Electric Utilities under Section 111 of the
CAA. An agency’s authority to enact regulations is determined by the authority
provided to it by statute. However, the scope of this authority is not always clear,
raising questions as to the ability of an agency to act in a particular manner. Such is
the case in the present situation, where the EPA is considering regulating mercury
emissions from electric utilities under section 111 rather than section 112.
Depending upon how these statutory provisions are interpreted, the EPA may be
Pub. Law 101-549, § 302(a) (Nov. 15, 1990).
Pub. Law 101-549, § 108(g) (Nov. 15, 1990).
42 U.S.C. § 7411(d)(1).
See United States v. Ward, 131 F.3d 335, 339-40 (3d. 1997).
See, e.g. American Bank & Trust Co. v. Dallas County, 463 U.S. 855, 864 (1983); United
States v. Welden, 377 U.S. 95, 98 n.4 (1964); Cheney R.R. Co., Inc. v. Railroad Ret. Bd.,
50 F.3d 1071, 1076 (D.C. Cir. 1995).
either permitted or prohibited from instituting a “cap-and-trade” program for mercury
emissions from electric utilities under section 111.
In Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc.,37 the
Supreme Court announced the primary criteria the courts would use in reviewing a
legal challenge to an agency’s interpretation of its statutory authority. Although
Chevron has generally been the guiding standard that courts use to determine the
permissibility of an agency’s interpretation of its statutory authority, recent Supreme
Court cases have suggested that Chevron analysis may not always be the appropriate
method with which to determine the permissibility of an agency’s interpretation. It
is possible that the EPA’s interpretation of section 111(d) will not be subjected to
Chevron analysis, but rather a different standard. Accordingly, this section will
analyze the EPA’s interpretation of relevant provisions of the CAA under the
framework of Chevron and, with respect to section 111(d), a potentially relevant nonChevron standard.
Analysis of the EPA’s Interpretation of CAA sections 112 and 111
under Chevron. The Chevron case concerned the EPA’s implementation of the
1977 amendments to the CAA, which in relevant part required the establishment of
a permit program regulating “new or modified major stationary sources” of air
pollution in states that had failed to meet national air quality standards. In
promulgating regulations for the permit program, the EPA interpreted the
legislation’s use of the term “stationary source” so as to allow states to adopt a plantwide definition of the term, meaning that a plant’s multiple emission sources could
be considered within a single “bubble” for purposes of calculating emission rates and
changes thereto. NRDC challenged this action, alleging that the EPA’s
implementation of the term “stationary source” was an impermissible interpretation
of the CAA amendments and was therefore contrary to law.
In resolving the legality of EPA’s interpretation of the CAA, the Court
announced a two-prong test for determining the legitimacy of an agency’s
interpretation of its statutory authority. The first prong of the Chevron test is “the
question whether Congress has directly spoken to the precise question at issue. If the
intent of Congress is clear, that is the end of the matter; for the court, as well as the
agency, must give effect to the unambiguously expressed intent of Congress.”38 A
court may attempt to ascertain congressional intent following “traditional tools of
statutory construction.”39 If congressional intent behind a statute is unclear, the
second prong of the Chevron test is considered: “[W]hether the agency’s
467 U.S. 837 (1984).
Id. at 842-43.
Id. at 843 n.9. Although the Court did not note what these “traditional tools” were, it did
look to the express language of the 1977 CAA amendments and the legislative history of
relevant provisions of the CAA when applying the first prong of this test in Chevron. See
id. at 851-53. Subsequently, the Supreme Court has held that a court’s determination of
congressional intent “must be guided to a degree by common sense as to the manner in
which Congress is likely to delegate a policy decision of....[significant] economic and
political magnitude to an administrative agency.” Food and Drug Admin. v. Brown &
Williamson Tobacco Corp., 529 U.S. 120, 133 (2000).
[intepretation] is based on a permissible construction of the statute.”40 A reviewing
court does not impose its own interpretation of the statute for that of the agency.
Rather, it assesses whether the agency’s interpretation is a reasonable one.41 In
making this assessment, the court gives deference to the agency’s interpretation.42
The Supreme Court in Chevron noted that the district court had made a proper
determination that the CAA was ambiguous as to the meaning of “stationary source,”
because neither the express language of the Act nor the relevant legislative history
contained any specific comment on the “bubble” concept or addressed whether a
plant-wide definition of a “stationary source” would be permissible.43 Accordingly,
the Court in Chevron found that congressional intent could not be clearly determined
and moved to the second prong of the test. Using this second prong, the Court
determined in Chevron that the EPA’s implementation of the CAA relied on a
reasonable interpretation of “stationary source,” and the Court therefore upheld the
The statutory interpretations that the EPA would rely upon to implement its
proposed revision to its December 2000 regulatory finding and its proposed “capand-trade” program under section 111 might be subjected to legal challenges by
opponents of these actions. If challenged, the EPA will almost certainly argue that
the interpretations of sections 111 and 112 that form the legal basis behind the
agency’s proposed revision and rule are legally valid under the Chevron test.
The first question that the EPA would likely need to address concerns the
reinterpretation of section 112(n)(1)(A) that forms the basis for the revision of its
December 2000 regulatory finding. Although the EPA initially found that regulation
of electric utilities was both appropriate and necessary, it now concludes that while
such regulation would be appropriate under section 112, it is not “necessary” because
such power plants could be adequately regulated under section 111. Accordingly,
the EPA argues, it does not have legal authority to regulate mercury emissions under
section 112, because section 112(n)(1)(A) only permits such regulation when it is
both appropriate and necessary.44
If challenged, the EPA will likely be able to argue that its interpretation of
section 112(n)(1)(A) fulfills the Chevron test. Indeed, the EPA might argue that its
interpretation of section 112(n)(1)(A) is in accordance with express statutory
language and there is no need for the court to even proceed to the second prong of
Chevron.45 Section 112(n)(1)(A) does not say that the EPA can regulate electric
Chevron, 467 U.S. at 843.
Id. at 844.
See Proposed Rule and Revision, supra note 7, at 19.
This argument might be further bolstered by portions of the limited legislative history
behind the 1990 amendments to the CAA. The EPA quotes Congressman Michael Oxley,
utilities under section 112 if it finds such regulation to be either appropriate or
necessary; rather, it permits such regulation only if the EPA finds such regulation to
be “appropriate and necessary.”46 Even assuming that a court finds that the
congressional intent behind section 112(n)(1)(A) is ambiguous and proceeds to the
second prong of the Chevron test, there does not appear to be any obvious basis to
argue that the EPA’s interpretation of the phrase “appropriate and necessary” is
unreasonable. Accordingly, the EPA has a strong argument that its interpretation of
its statutory authority under section 112(n)(1)(A) is permissible, and that its
December 2000 regulatory finding that electric utilities should be regulated under
section 112 can be revised if the EPA no longer has reason to believe such regulation
is both appropriate and necessary.
The second question of statutory interpretation likely to be raised, assuming that
the EPA adopts the proposed “cap-and-trade” program under section 111, is whether
HAP emissions from electric utilities can be regulated under that section. Assuming
that the EPA’s interpretation of section 111 is challenged, there is reason to believe
that a court could conclude that congressional intent is ambiguous as to whether HAP
emissions from electric utilities can be regulated under section 111. As discussed
previously, section 111(b) makes no reference to section 112, and section 111(d)
contains apparently conflicting amendments that would either prohibit the EPA from
regulating existing sources for HAP listed under section 112 or, alternatively, permit
such regulation if the HAP source was not otherwise regulated under section 112.47
It therefore appears that a court hearing a challenge to the EPA’s authority under
section 111 to regulate mercury emissions from electric utilities could move to the
second prong of Chevron analysis and assess whether the EPA’s interpretation of the
section was reasonable. The EPA may argue that its interpretation is legally
permissible because the Senate and House amendments represent conflicting
congressional intent as to the scope of the EPA’s authority to regulate under section
111, and the EPA’s reinterpretation of section 111(d) harmonizes these conflicting
amendments. The Supreme Court has previously recognized that when an agency
interpretation of its statutory authority “represents a reasonable accommodation of
conflicting policies that were committed to the agency’s care by the statute, [the
court] should not disturb it unless it appears from the statute or its legislative history
that the accommodation is not one that Congress would have sanctioned.”48
The EPA might argue that its interpretation of section 111 is a “reasonable
accommodation” given the conflicting language of the amendments to section 111(d).
a member of the conference committee on the 1990 CAA amendments, as stating that
Congress intended section 112 to regulate “only those units that...[the Administrator]
determines — after taking into account compliance with all other provisions of the
act...-have been demonstrated to cause a significant threat of serious adverse affects on
public health.”Id. at 47.
42 U.S.C. § 7412(n)(1)(A) (emphasis added).
See supra pp. 4-5.
United States v. Shimer, 367 U.S. 374, 383 (1961) (quoted by Chevron, 467 U.S. at 845).
Further, the EPA may argue that it is unlikely that Congress would have been
opposed to such an accommodation, given that Congress’s stated purpose in passing
the CAA was “to protect and enhance the quality of the Nation’s air resources so as
to promote the public health and welfare....”49 Permitting the EPA to regulate HAP
emissions from electric power plants exclusively under section112 would limit such
regulation only to circumstances where the imposition of section 112 MACT controls
was “appropriate and necessary” for the public health. The EPA might argue that
permitting the regulation of HAP from electric utilities under section 111 would
enable the EPA to address public health concerns caused by such emissions even in
situations where such concern would not be sufficient to warrant regulation under
On the other hand, opponents of the EPA’s proposed “cap-and-trade” program
might dispute the reasonableness of the EPA’s interpretation of its statutory authority
under section 111. They might argue that the fact that Congress chose to list specific
HAP under section 112 indicated that Congress believed that these pollutants
required stricter measures than those permitted under section 111, and the EPA is
now acting contrary to this intent by attempting to regulate an HAP source under a
less stringent standard than Congress desired.
Opponents may also note that prior to the 1990 Amendments, the explicit
language of section 111(d) prevented it from being used to regulate existing sources
of HAP listed under section 112, and the Senate amendment reflects this continued
understanding. In the limited legislative history on the 1990 amendments to section
111(d) of the CAA, there does not appear to be any significant, specific discussion
of the House amendment. Opponents of the EPA’s “cap-and-trade” program might
therefore argue that this indicates that Congress did not intend the amendment to
permit the EPA to regulate HAP emissions under a different section of the CAA than
it had previously. Further, CRS has been unable to find an example of the EPA
acting pursuant to section 111(d) to regulate an existing source for an HAP listed
under section 112. Given these factors, opponents of the EPA’s proposed “cap-andtrade” program may argue that it is unreasonable for the EPA to interpret section 111
as permitting the regulation of sources of HAP listed under section 112.
The third question of statutory interpretation that may be raised is whether the
language of section 111(d), which requires the EPA to establish standards of
performance for existing sources that represent “the best system of emission
reduction...which the Administrator determines has been adequately demonstrated,”50
provides the EPA with authority to establish a “cap-and-trade” program for mercury
emissions from electric utilities. Assuming for purposes of discussion that a
reviewing court finds that section 111(d) can be used to regulate mercury emissions
from electric utilities, a straightforward application of the Chevron test might suggest
that the EPA could establish a “cap-and-trade” program for these emissions if the
Administrator concluded that it was the “best system” for reducing such emissions.
42 U.S.C. § 7401(b)(1).
42 U.S.C. § 7411(a)(1) (defining the meaning of “standards of performance” as used under
Alternative Analysis of the EPA’s Interpretation of section 111(d)
Not Relying on Chevron. While application of the Chevron test has been the
primary means for the courts to determine the legality of an agency’s statutory
interpretation for almost two decades, recent Supreme Court cases indicate that the
Chevron test may not always be the appropriate means to evaluate agency
interpretations. In 2000, for example, the Court held that Chevron deference does not
apply to interpretations contained in opinion letters, agency manuals, and similar
informal issuances,51 though it stressed that Chevron deference “does apply to an
agency interpretation contained in a regulation.”52 More relevant to the present case,
the Court recognized in the same term that while Chevron deference is appropriate
when a “statute’s ambiguity constitutes an implicit delegation from Congress to the
agency to fill in the statutory gaps,”53 in a limited number of cases “there may be
reason to hesitate before concluding that Congress has intended such an implicit
The issue of agency interpretation in the Chevron line of cases is arguably
different from the issue concerning the EPA’s reinterpretation of section 111(d). The
Chevron test may be understood to apply in instances where the scope of a
congressional delegation of authority is ambiguous because the legislation at issue
does not speak directly to the matter that the agency is attempting to regulate.55 In
the present case, however, Congress did speak directly to the question of whether the
EPA could regulate mercury emissions from electric utilities under section 111, but
did so in a manner that was self-contradicting. Accordingly, a court hearing a
challenge to the EPA’s interpretation of section 111 may decide that Chevron is an
inappropriate starting point for assessing the legitimacy of the agency’s
Although it was decided prior to Chevron, the D.C. Circuit’s decision in
Citizens to Save Spencer County v. Environmental Protection Agency56 dealt with a
situation very similar to the present case and might be used to resolve the legality of
the EPA’s interpretation of section 111. The Spencer County case concerned the
EPA’s implementation of conflicting congressional directives resulting from the
1977 CAA amendments. Whereas one provision of the amended CAA prohibited
Christenson v. Harris County, 529 U.S. 576, 587 (2000).
Id. See also United States v. Mead Corp., 533 U.S. 218 (2001) (holding that Chevron
deference is typically appropriate only when evaluating agency actions having the force of
law, and not when evaluating agency rulings that are only binding on the agency and/or a
Brown & Williamson Tobacco Corp., 529 U.S. at 159 (2000).
See id. at 133 (“[I]f Congress has not specifically addressed the question, a reviewing court
must respect the agency’s construction of the statute so long as it is permissible”) (internal
citations omitted); Auer v. Robbins, 519 U.S. 452, 457 (1997) (“Because Congress has not
directly spoken to the precise question at issue, we must sustain...[the agency’s] approach
so long as it is based on a permissible construction of the statute”) (internal citations
600 F.2d 844 (D.C. Cir. 1979).
construction of any major pollution-emitting facility unless certain substantive
requirements had been met, another provision suggested that less stringent
construction requirements contained in state plans made pursuant to preexisting
federal regulations remained in effect until revised state plans had been approved.57
In regulations implementing the 1977 CAA amendments, the EPA pursued a “middle
path” that attempted to harmonize these conflicting congressional directives.58 In
upholding the EPA’s regulations harmonizing the conflicting CAA provisions, the
D.C. Circuit stated that it was “guided by the rule that the maximum possible effect
should be afforded to all statutory provisions, and, whenever possible, none of those
provisions rendered null or void.”59 To appropriately harmonize conflicting
provisions, the court declared that an agency should:
[L]ook for guidance to the statute as a whole and to consider the
underlying goals and purposes of the legislature in enacting the statute,
while avoiding unnecessary hardship or surprise to affected parties and
remaining within the general statutory bounds prescribed. Only by this
approach can legislative purposes and statutory instructions be given the
greatest possible practical effect.60
The Spencer County court concluded that if an agency acts “in a reasonable and
responsible manner” in exercising the perhaps inadvertent discretion provided by
conflicting statutory provisions, the result it reaches should be upheld.61 However,
the court noted that it is ultimately the reviewing court’s responsibility “to determine
whether the administering agency in the present case had authority to devise its
‘middle path,’ whether it pursued the appropriate procedural route in so doing, and
whether its results are substantively reasonable.”62
Opponents of the EPA’s reinterpretation of section 111 are likely to argue that
the agency’s interpretation of the conflicting House and Senate amendments is not
in accordance with the harmonization principle detailed in Spencer County.
Opponents will likely argue that rather than attempting to harmonize the
amendments, the EPA has instead chosen to exclusively adopt the House amendment
permitting regulation under section 111(d) of emission sources not regulated under
section 112, while ignoring the Senate amendment’s prohibition on the application
of section 111(d) to regulate any HAP listed under section 112. Opponents would
likely argue that to the extent that the amendments should be harmonized, it should
be in a manner giving preference to the Senate amendment, because the Senate
amendment reflects Congress’s longstanding intent to have HAP that are listed under
Id. at 851-54.
See id. at 859.
Id. at 870.
Id. at 871.
Id. at 872.
Citizens of Spencer County, 600 F.2d at 873.
section 112 regulated exclusively under that provision, and nothing in the legislative
history of the House amendment reveals an intent to reverse this position.63
On the other hand, the EPA may counter that its reinterpretation of section 111
is in accordance with the harmonization principle detailed in Spencer County and
gives an appropriate degree of preference to the Senate amendment. The EPA
concedes that application of the Senate amendment alone would prevent the agency
from using section 111 to regulate any HAP emission listed under section 112,
including mercury emissions from electric utilities,64 and that the Senate amendment
“reflects the Senate’s intent to retain the pre-1990 approach of precluding regulation
under CAA section 111(d) for any HAP that is listed under section 112(b).65 The
EPA notes, however, that a literal reading of the House amendment, which prohibits
section 111(d) from applying to any pollutant “emitted from a source category which
is regulated under section 112,” would prevent the EPA from regulating both HAP
and non-HAP emissions from sources regulated under section 112, even though
section 112 is only used to regulate HAP emissions.66 The EPA therefore argues that
its reinterpretation of section 111(d) gives some effect to both amendments:
First, [the reinterpretation] gives effect to the Senate’s desire to focus on
HAP listed under section 112(b), rather than applying the section 111(d)
exclusion to non-HAP emitted from a source category regulated under
section 112, which a literal reading of the House amendment would do.
Second, it gives effect to the House’s apparent desire to increase the scope
of the EPA’s authority under section 111(d) and to avoid duplicative
regulation of HAP for a particular source category.67
The EPA alleges that its reinterpretation of section 111(d) is a reasonable one that
gives more effect to the Senate amendment than to the House amendment; a
preference the EPA deems appropriate because a literal reading of the House
amendment “would be inconsistent with the general thrust of the 1990 [CAA]
amendments.”68 Using this analysis, the EPA might argue that its reinterpretation of
section 111(d) is a reasonable effort to harmonize explicit, conflicting congressional
directives in a manner that gives weight to the overall purpose behind the CAA, and
Opponents might also argue that the House amendment conflicts with the purposes behind
the CAA, for reasons discussed at supra p. 10, and that the EPA should therefore ignore it
for purposes of interpreting section 111(d) of the CAA. Both the D.C. Circuit and Supreme
Court have recognized that, in rare circumstances, the plain meaning of a statute does not
control its interpretation by the courts when “the literal application of [the] statute will
produce a result demonstrably at odds with the intentions of its drafters.” Environmental
Defense Fund v. E.P.A., 82 F.3d 451, 468 (D.C. Cir. 1996) (quoting United States v. Ron
Pair Enterprises, Inc., 489 U.S. 235, 242 (1989)).
Proposed Rule and Revision, supra note 7 at 184-85.
Id. at 185.
Id. at 186.
that this harmonization effort accordingly should be upheld by a reviewing court
under either Chevron or Spencer County analysis.
The EPA’s Discretion to Revise Previous Regulatory Findings and
Institute a “Cap-and-Trade Program”. Although the EPA’s interpretation of
its statutory authority might be legally permissible, this does not necessarily mean
that every proposed regulation issued in accordance with this interpretation would be
upheld by the courts. A court reviewing actions taken by the agency may find them
to be unlawful if they are “arbitrary, capricious, an abuse of discretion, or otherwise
not in accordance with law.”69 By statute, the EPA is required to engage in a “noticeand-comment” rulemaking procedure concerning any proposed emission regulation
under CAA sections 111 or 112, whereby the EPA must provide notice of the
proposed action so that affected parties can comment on it.70 Courts have generally
required agencies that engage in “notice-and-comment” rulemaking to provide
detailed explanations of final rules which respond to notable comments and
arguments submitted by interested parties.71
In the present case, the EPA’s proposed “cap-and-trade” program and revision
of its December 2000 regulatory finding are based on the premise that regulation
under section 112 is not appropriate and necessary because the public health hazards
of mercury emissions from electric utilities can be adequately addressed through
regulation under section 111. Accordingly, the EPA has initiated a notice and
comment rulemaking procedure in which interested parties are provided with the
opportunity to comment on the proposed rule and regulatory revision prior to the
EPA issuing a final decision. If the EPA ultimately decides to revise its December
2000 finding and implement a “cap-and-trade” program, it will likely need to
adequately address the comments and data it receives during the notice and comment
proceeding for its final action to be upheld by a reviewing court.
The fact that the EPA previously concluded in December 2000 that the
regulation of mercury emissions from electric utilities was appropriate and necessary
under section 112 might increase the agency’s burden in proving that its subsequent
revisions are not arbitrary and capricious. The Supreme Court has long recognized
that “regulatory agencies do not establish rules of conduct to last forever.”72
Changing circumstances and the availability of new information requires that an
42 U.S.C. § 7607(d)(9)(A).
See 42 U.S.C. §§ 7607(d)(1),(5)-(6).
See, e.g., Action on Smoking & Health v. Civil Aeronautics Bd., 699 F.2d 1209, 1217-19
(D.C. Cir. 1983) (setting aside an agency rule in part because the agency failed to consider
several regulations suggested by commenters); National Tour Brokers Ass’n v. U. S., 591
F.2d 896 (D.C. Cir. 1978) ( recognizing that the purpose of comments is to enable an agency
to consider and benefit from the expertise of commenters when making rules); Automotive
Parts & Accessories Ass’n v. Boyd, 407 F.2d 330, 338 (D.C. Cir. 1968) (requiring agencies
to respond to significant issues raised in the rulemaking comments); see also 42 U.S.C. §
7607(d)(6)(B) (“The promulgated rule shall also be accompanied by a response to each of
the significant comments, criticisms, and new data submitted in written or oral presentations
during the comment period”).
American Trucking Assoc. v. Atchison, T. & S.F.R. Co., 387 U.S. 397 (1967).
agency “must be given ample latitude” to adapt and revise its rules and policies.73
Nevertheless, there is a presumption that a “settled course of behavior embodies the
agency’s informed judgment that, by pursuing that course, it will carry out the
policies committed to it by Congress...[and there is] a presumption that those policies
will be carried out best if the settled rule is adhered to.”74
Accordingly, an agency changing its regulatory course by rescinding a previous
rule is often required to supply “a reasoned analysis” for the change beyond that
which might have been required when it did not act in the first instance.75 In order
to regulate mercury emissions under section 111 of the CAA rather than section 112,
the EPA might need to overcome the presumption that the conclusions reached in its
December 2000 regulatory finding were accurate.
A further potential obstacle to EPA regulation of mercury emissions from
electric utilities under section 112 is the fact that the EPA’s earlier decision to list
such emissions for regulation might thereby prevent their subsequent regulation
under section 111. As discussed earlier, section 112(c)(9) requires that once an HAP
source is listed for regulation under section 112, it cannot be “de-listed” unless new
information essentially reveals that the source’s designated HAP emissions pose no
public health threat.76
The EPA argues that this requirement is inapposite in the present case, because
electric utilities were listed for regulation under section 112 only because the EPA
had failed to consider the possibility of regulating these sources under section 111.77
The EPA analogizes this situation with other times when it has “de-listed” a source
category without following the criteria of section 112(c)(9), after the agency had
determined that it lacked a factual predicate for initially listing the category.78 For
example, in 2002 the EPA removed asphalt concrete manufacturers from the list of
major sources to be regulated under section 112 after the agency concluded that these
sources did not emit or have the potential to emit a sufficient amount of HAP to
satisfy the statutory definition of a “major source.”79
Opponents might argue, however, that reinterpretation of section 112 should not
be grounds to summarily rescind previous designations that were based on an equally
reasonable yet contrary interpretation of the statute. Opponents might also argue that
it is inappropriate for the EPA to analogize instances where factual error was
responsible for misdesignation to the present case, where there is no dispute over the
Motor Vehicle Assoc. of U.S., Inc. v. State Farm Mutual Auto. Insurance Co., 463 U.S.
29, 42 (1984).
Id. (quoting Atchison, T. & S.F.R. Co. v. Wichita Bd. of Trade, 412 U.S. 800, 807-08
See supra p.3.
Proposed Rule and Revision, supra note 7 at 205-206.
public health threat posed by mercury emissions from electric utilities, but rather
whether the EPA can interpret section 112 differently than it had previously.
Implications of the 1998 Settlement Agreement upon the Manner in
which the EPA Can Regulate Electric Utilities. A legal settlement agreement
is generally used to end or avoid litigation between parties, by obligating one or more
of the parties to take a particular course of action. The remedy available when a
settlement agreement is breached is dependant upon the nature of the agreement and
its express provisions.80 For a settlement agreement to act as controlling law that
binds an agency to take a particular course of action with respect to rulemaking, it
generally needs to be incorporated into a judicial order as a consent decree.81 If a
settlement agreement between an agency and a private party concerning agency
rulemaking has not been incorporated into a court order, the only remedy available
to the non-breaching party might be to move for the original case to be reopened.82
In that case, “the agency is back in its presettlement position, and the agreement itself
should not be seen as a significant limitation on the agency’s ability to change its
The 1998 settlement agreement between the EPA and NRDC was a
modification of an earlier 1994 agreement between the two parties that was filed with
the D.C. Circuit. According to the NRDC, however, neither the 1994 nor 1998
settlement agreement was incorporated into a judicial order. It therefore seems that
even if it could be argued that the EPA would be in breach of the settlement
agreement if it chose to regulate mercury emissions from electric utilities under
section 111 rather than 112 of the CAA, the NRDC’s remedy might be limited to
moving to reopen the case forming the underlying basis behind the breached
agreement. Accordingly, it does not appear likely that the 1998 settlement agreement
could be used to restrict the EPA’s rulemaking discretion.
For a discussion of the diverse nature of settlement agreements and their potential effect
on EPA rulemaking, see Jeffrey M. Gaba, Informal Rulemaking by Settlement Agreement,
73 GEO L.J. 1241 (1985).
A consent decree is a legal document approved by a court that formalizes an agreement
between parties in a case so as to avoid further litigation. The consent decree itself is “an
agreement that the parties desire and expect will be ... enforceable as ... a judicial decree that
is subject to the rules generally applicable to other judgments and decrees.” Frew v.
Hawkins, 2004 WL 57266 (U.S. Sup. Ct. Jan. 14, 2004) (quoting Rufo v. Inmates of Suffolk
County Jail, 502 U.S. 367, 378 (1992)). See FED. R. CIV. P. 54(a) (defining a “judgment”
as including “a decree and any order from which an appeal lies”); Gorsuch, 718 F.2d at
1125 (declaring that a consent decree “must be treated ‘as a judicial act’”) (quoting United
States v. Swift & Co., 286 U.S. 106, 115 (1932)).
See Gaba, supra note 80, at 1265 (Agency compliance with a settlement agreement not
incorporated into a judicial order “would presumably only dissolve the stay and accelerate
the underlying lawsuit”); Natural Resources Defense Council, Inc. v. Reilly, 781 F.Supp.
806, 807 (D.D.C. 1992) (discussing NRDC moving to reopen case against EPA after EPA
breached settlement agreement not incorporated into a judicial order).
Gaba, supra note 80, at 1265.