UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA
) COMMUNITY IN-POWER AND ) DEVELOPMENT ASSOCIATION, INC., et al., ) ) Plaintiffs, ) ) v. ) No. 16-cv-1074 (KBJ) ) E. SCOTT PRUITT, in his official capacity as ) Administrator of the Environmental Protection ) Agency, ) ) Defendant. ) )
MEMORANDUM OPINION
Several environmental advocacy groups (“Plaintiffs”) have banded together to
file this action against the Environmental Protection Agency (“the EPA” or “the
agency”) pursuant to the Clean Air Act’s citizen suit provision, 42 U.S.C. § 7604(a)(2).
Plaintiffs seek to compel the EPA to perform obligatory and long-overdue rulemakings
to protect people and the environment from hazardous air pollution (see Compl., ECF
No. 1, ¶ 1), and the EPA admits that it has violated the Clean Air Act’s prescriptions,
insofar as the agency has failed to promulgate revised emission standards for the nine
source categories of pollutants at issue in this case in a timely fashion (see EPA’s Mem.
in Opp’n to Pls.’ Mot. for Summ. J. & in Supp. of Cross-Mot. for Summ. J. (“Def.’s
Mem.”), ECF No. 29-3, at 6). 1 Based on that admission, all that remains of this dispute
1 Page-number citations to the documents that the parties have filed refer to the page numbers that the Court’s electronic filing system automatically assigns. Any citations to the EPA’s cross-motion for summary judgment and its attachments thereto, including the memorandum in support, refer to the page numbers of the corrected cross-motion for summary judgment that this Court ordered the EPA to provide. (See Min. Order of June 15, 2017; see also General Order & Guidelines, ECF No. 8.)
1 is a determination of how quickly the EPA must act to issue the revised emission
standards.
Before this Court at present are Plaintiffs’ and the EPA’s motions for summary
judgment regarding that sole issue. (See Mot. of Pls. for Summ. J. (“Pls.’ Mot.”), ECF
No. 21; EPA’s Cross-Mot. for Summ. J. (“Def.’s Cross-Mot.”), ECF No. 23.) The
parties propose drastically different timelines for the EPA to act: Plaintiffs request a
completion schedule of no more than two years from the date of this Court’s Order (see
Pls.’ Mot. at 46), while the EPA asks for approximately seven years, until January of
2025, to complete the required rulemakings (see EPA’s Reply Mem. in Supp. of Cross-
Mot. for Summ. J. (“Def.’s Reply”), ECF No. 34, at 6). Having considered the parties’
briefs, their presentations at the motion hearing held on November 30, 2017 (see
generally Tr. of Mot. Hr’g (Nov. 30, 2017) (“Hr’g Tr.”), ECF No. 40), and the EPA’s
declarations, this Court will order the EPA to comply with its statutory obligations as
expeditiously as possible, although not on the extremely compressed timeline Plaintiffs
propose. Specifically, the EPA must complete all nine overdue rulemakings over the
next three and a half years, and no later than October 1, 2021. Accordingly, Plaintiffs’
motion for summary judgment will be GRANTED IN PART and DENIED IN PART,
and Defendant’s cross-motion for summary judgment will be DENIED. A separate
Order consistent with this Memorandum Opinion will follow.
I. BACKGROUND
A. The Clean Air Act
Congress enacted the Clean Air Act, 42 U.S.C. § 7401, et seq., in 1963, “to
protect and enhance the quality of the Nation’s air resources so as to promote the public
2 health and welfare and the productive capacity of its population.” Id. § 7401(b)(1). In
1990, Congress overhauled the Act and implemented “an aggressive regime of new
control requirements to address four crucially important air pollution problems: urban
smog, hazardous air pollution, acid rain, and depletion of the stratospheric ozone
layer.” Cal. Cmtys. Against Toxics v. Pruitt, 241 F. Supp. 3d 199, 200 (D.D.C. 2017);
see also Blue Ridge Envtl. Def. League v. Pruitt, 261 F. Supp. 3d 53, 55 (D.D.C. 2017)
(explaining that the overhaul resulted from Congress’s recognition that the Act had
“worked poorly” up to that point). Among many other changes, the 1990 amendments
created an initial list of hazardous air pollutants, such as cyanide, mercury, and
phosphorous, see 42 U.S.C. § 7412(b)(1), and imposed a series of deadlines by which
the EPA was required to promulgate, and periodically revise, emission standards for
sources that emit those pollutants, see id. § 7412(c)(1), (c)(2), (d).
Two of those deadlines are relevant here. First, the Act states that the EPA
“shall review, and revise as necessary (taking into account developments in practices,
processes, and control technologies), emission standards promulgated under this section
no less often than every 8 years[.]” Id. § 7412(d)(6) (emphasis added). This means
that, within that timeframe, the EPA must conduct what is known as a “technology
review” to determine whether the agency should modify current emission standards in
light of any improvements in pollution control technology. (Decl. of Panagiotis
Tsirigotis (“June Tsirigotis Decl.”), ECF No. 29-5, ¶ 4.) Second, the EPA must
consider any risks to public health or the environment that remain despite the agency’s
previously-implemented emission standards, and develop additional standards to
mitigate any residual risks. (See id.) This “residual risk review” must occur “within 8
3 years after promulgation of standards for each category or subcategory of sources[.]”
42 U.S.C. § 7412(f)(2)(A) (emphasis added).
The EPA division that is primarily responsible for performing these reviews, as
well as any associated rulemakings, is the Sector Policies and Programs Division
(“SPPD”) within the Office of Air Quality Planning and Standards, Office of Air and
Radiation. (See June Tsirigotis Decl. ¶¶ 2–3; Decl. of Panagiotis Tsirigotis, Attach. 1
to June Tsirigotis Decl., ECF No. 29-5, ¶ 4.) That division generally performs both the
technology review and the residual risk review at the same time, through a combined
“risk and technology review” (“RTR”) assessment, which is supposed to take place
within the eight-year window set forth in the Clean Air Act. (June Tsirigotis Decl. ¶ 4.)
B. Procedural History
Plaintiffs filed the complaint in the instant case on June 8, 2016, alleging that the
EPA has failed to complete timely either the mandatory technology review or the
mandatory residual risk review for the emission standards associated with nine source
categories of hazardous air pollutants. (See Compl. ¶¶ 54–57.) 2 The applicable due
dates for the mandated reviews with respect to each of the nine pollutants are as
follows:
2 Plaintiffs’ complaint names former EPA Administrator Gina McCarthy as the defendant in this action (see Compl. ¶ 11); the current EPA Administrator, Edward Scott Pruitt, has been automatically substituted as the defendant pursuant to Federal Rule of Civil Procedure 25(d). Because Administrator Pruitt is being sued in his official capacity only, this suit functions as an action against the EPA, and will be treated as such for purposes of this Memorandum Opinion. See Cty. Bd. of Arlington v. U.S. Dep’t of Transp., 705 F. Supp. 2d 25, 28 (D.D.C. 2010) (“[A]n official-capacity suit is a way of pleading an action against the agency which the official heads.”).
4 Deadline for Date of Original Action Pursuant Source Category Promulgation to § 7412(d)(6) and § 7412(f)(2) Primary Copper Smelting June 12, 2002 June 12, 2010 Generic MACT II – Carbon Black July 12, 2002 July 12, 2010 Production Generic MACT II – Cyanide Chemicals July 12, 2002 July 12, 2010 Manufacturing Generic MACT II – Spandex Production July 12, 2002 July 12, 2010 Flexible Polyurethane Foam Fabrication Apr. 14, 2003 Apr. 14, 2011 Operations Refractory Products Manufacturing Apr. 16, 2003 Apr. 16, 2011 Semiconductor Manufacturing May 22, 2003 May 22, 2011 Primary Magnesium Refining Oct. 10, 2003 Oct. 10, 2011 Mercury Emissions from Mercury Cell Dec. 19, 2003 Dec. 19, 2011 Chlor-Alkali Plants
(See id. ¶ 1, Table A.)
Because the required reviews are long overdue at this point, Plaintiffs’ complaint
asks this Court to declare that the EPA has violated the Clean Air Act, and order the
agency to perform the mandated rulemakings for the nine source categories “in
accordance with an expeditious deadline specified by this Court[.]” (Id. ¶ 58; see also
Pls.’ Mot. at 45 (arguing that the Court should compel the EPA to perform the nine
overdue rulemakings “as expeditiously as possible” and “within the fastest possible
timeframe”).) For five of the source categories, which Plaintiffs do not specify,
Plaintiffs propose that the EPA be required to issue notices of proposed rules within
eight months of this Court’s Order, and to promulgate final rules within one year. (See
Pls.’ Mot. at 46.) For the remaining four source categories, which again Plaintiffs do
not specify, Plaintiffs request that notices of proposed rules be issued within twenty
months of this Court’s Order, and final rules be promulgated within two years. (See id.)
Thus, Plaintiffs would have the agency promulgate final rules for five source categories
5 by March 31, 2019, and promulgate final rules for the remaining four source categories
by March 31, 2020.
Although the EPA admits “that it has not yet completed its duty to conduct the
technology and residual risk reviews pursuant to 42 U.S.C. § 7412(d)(6) and (f)(2) for
the nine [source] categories” at issue in this case (Def.’s Mem. at 6), it argues that
“[t]he issue . . . is not the shortest time period in which [the agency] can issue a rule,
but rather the time that is needed for [the agency] to complete a rulemaking that is
sufficiently thorough to meet the purpose intended by Congress” (id. at 28). The EPA
further maintains that Plaintiffs’ proposed schedule “would undermine the soundness of
the [emissions] rules by not allowing enough time for the [a]gency to gather necessary
emissions data or to properly allow for public participation” (id. at 16), and thus,
Plaintiffs’ schedule “would require a significant compromise on the quality of the rule
at issue” (id. at 28). Instead, the EPA requests significantly more time to complete the
mandated rulemakings, as indicated in the schedule below:
EPA Proposed EPA Final Source Category Rule Date Rule Date Mercury Emissions from Mercury Cell July 23, 2021 July 22, 2022 Chlor-Alkali Plants Semiconductor Manufacturing Nov. 11, 2021 Nov. 10, 2022 Generic MACT II – Cyanide Chemicals Dec. 9, 2021 Dec. 8, 2022 Manufacturing Generic MACT II – Spandex Production Dec. 9, 2021 Dec. 8, 2022 Generic MACT II – Carbon Black Feb. 24, 2022 Feb. 23, 2023 Production Primary Copper Smelting Nov. 17, 2022 Feb. 8, 2024 Flexible Polyurethane Foam Fabrication July 20, 2023 Oct. 10, 2024 Operations Refractory Products Manufacturing Aug. 24, 2023 Nov. 14, 2024 Primary Magnesium Refining Oct. 19, 2023 Jan. 16, 2025
(See Def.’s Reply at 6.)
6 The EPA makes two primary arguments in support of its proposed schedule.
First, the EPA claims that all available agency resources to perform the nine RTRs at
issue are already “fully utilized” because the agency is in the process of completing
several dozen RTRs for other source categories within schedules mandated by other
court orders and consent decrees. (Def.’s Reply at 12; see also Def.’s Mem. at 13–14
(listing the other schedules); June Tsirigotis Decl. ¶ 12 (describing the other court-
mandated rulemakings as “[c]ritical to the schedules” proposed by the EPA to complete
the nine RTR rulemakings in the instant case).) Indeed, in March of last year, two
courts from this District ordered the EPA to complete the rulemakings for seven source
categories by December of 2018, another twenty source categories by March of 2020,
and the remaining six source categories at issue in those cases by June of 2020. See
Blue Ridge, 261 F. Supp. 3d at 61; Cal. Cmtys., 241 F. Supp. 3d at 207. 3
Because of these other court-ordered deadlines, the EPA says it will not have the
necessary staff and resources available to even begin the nine RTRs at issue here until
March of 2020—after the agency has completed most of the rulemakings already
ordered by other courts. (See Def.’s Mem. at 13–15; see also June Tsirigotis Decl. ¶ 12
(“All available SPPD staff are now assigned to RTR projects subject to court-ordered
deadlines established in these other cases, and the workload for these 33 other RTRs
3 The EPA also identifies several other RTRs that are currently underway. Those RTRs include rulemakings for the Pulp and Paper Combustion Sources source category and the Nutritional Yeast Manufacturing source category, which, pursuant to the order issued in Sierra Club v. McCarthy, No. 15-cv-1165, 2016 WL 1055120 (N.D. Cal. Mar. 15, 2016), were due by October 1, 2017. See id. at *7. While the instant case was still pending, the EPA successfully completed the RTRs and associated rulemakings for those two source categories within the deadline imposed by the Northern District of California. (See Pls.’ Notice of Additional Evid. (“Pls.’ Notice”), ECF No. 38, at 1.) The agency has also continued to engage in a number of discretionary rulemaking activities relating to the Clean Air Act (see Pls.’ 2d Notice of Additional Evid. (“Pls.’ 2d Notice”), ECF No. 41, at 1–4), and Plaintiffs point out that the agency has also recently requested through the President’s 2019 Proposed Budget that Congress reduce the funding dedicated to the agency’s air quality efforts (see id. at 4).
7 does not allow us to begin work on the 9 RTRs at issue here until most of the work is
complete on the 33 RTRs already subject to a court-ordered deadline.”).) But Plaintiffs
reject the EPA’s claimed staffing shortage, arguing, inter alia, that “[the] EPA has not
shown that it cannot supplement the currently-assigned SPPD staff with other EPA
staff[,]” or “that it would be impossible to hire more contractors to act more quickly[,]”
or “that no SPPD or otherwise qualified agency staff-time is assigned to purely
‘discretionary’ activities[.]” (Pls.’ Combined Reply in Supp. of Pls.’ Mot. for Summ. J.
& Opp’n to EPA’s Cross-Mot. (“Pls.’ Reply”), ECF No. 31, at 27–28 (citations
omitted).) Plaintiffs further insist that the overdue rulemakings must be prioritized, and
that, if necessary, the agency can “redirect resources from other regulatory initiatives to
ensure the fullest use of resources to fulfill its obligations.” (Pls.’ Mot. at 42–43
(internal quotation marks and citation omitted).)
The EPA’s other main argument is that Plaintiffs’ schedule fails to acknowledge
the complexity of the rulemakings at issue, and that the RTR rulemaking process
involves nine distinct phases that must be completed for each individual source
category. (See Def.’s Mem. at 16–24; see also June Tsirigotis Decl. ¶¶ 13–21.)
According to the EPA, it will take approximately twenty-eight months to complete the
fastest rulemaking—for the Mercury Emissions from Mercury Cell Chlor-Alkali Plants
source category—and approximately fifty-eight months to complete the longest—for the
Primary Magnesium Refining source category. (See June Tsirigotis Decl. ¶ 12, Tables 2
& 3 (attached hereto as Appendix A); see also Def.’s Reply at 6.) Plaintiffs counter
that the EPA’s proposed timeline to complete each phase not only includes
“discretionary” time that is not required to complete the rulemakings, but is also
8 characterized by “speculative and equivocal language” about what may (as opposed to
what will) be required at each phase, and adds “extra, hypothetical time” to complete
tasks that may not ultimately be required. (Pls.’ Reply at 35.)
The parties’ cross-motions for summary judgment are ripe for this Court’s
review (see Pls.’ Mot.; Def.’s Cross-Mot.; Pls.’ Reply; Def.’s Reply; Pls.’ Surreply
Opposing Def.’s Cross-Mot. for Summ. J. (“Pls.’ Surreply”), ECF No. 37), and the
Court heard the parties’ arguments on the cross-motions during a motion hearing held
on November 30, 2017 (see generally Hr’g Tr.).
II. LEGAL STANDARDS FOR REMEDYING A CLEAN AIR ACT VIOLATION
The Clean Air Act permits “any person [to] commence a civil action on his own
behalf” against the EPA “where there is alleged a failure of [the EPA] to perform any
act or duty . . . which is not discretionary with [the agency].” 42 U.S.C. § 7604(a). The
Act also empowers district courts to “order [the EPA] to perform” a mandated act or
duty and to “compel [non-discretionary] agency action unreasonably delayed[.]” Id.
The D.C. Circuit has interpreted this provision as allowing district courts to exercise
their equity powers “to set enforceable deadlines both of an ultimate and an
intermediate nature[.]” Nat. Res. Def. Council, Inc. v. Train, 510 F.2d 692, 705 (D.C.
Cir. 1974).
While a district court possesses broad discretion to set deadlines for compliance,
it may not, of course, order the EPA “to do an impossibility.” Id. at 713 (internal
quotation marks and citation omitted); see also Cal. Cmtys., 241 F. Supp. 3d at 207
(rejecting plaintiffs’ proposed timeline as being “simply too compressed . . . to afford
any reasonable possibility of compliance” (internal quotation marks and citation
9 omitted)); Sierra Club, 2016 WL 1055120, at *3 (“[C]ourts should not demand a
deadline for agency compliance that is impossible or infeasible.”). But an agency has a
“heavy burden to demonstrate” that the ordered requirements are impossible to meet, or
that it is unable to comply with a particular remedial timeline. Ala. Power Co. v.
Costle, 636 F.2d 323, 359 (D.C. Cir. 1979). That burden “serves to prevent an agency
from shirking its duties by reason of mere difficulty or inconvenience[,]” Am. Hosp.
Ass’n v. Price, 867 F.3d 160, 168 (D.C. Cir. 2017), and it “is especially heavy where
the agency has failed to demonstrate any diligence whatever in discharging its statutory
duty to promulgate regulations and has in fact ignored that duty for several years[,]”
Sierra Club v. Johnson, 444 F. Supp. 2d 46, 53 (D.D.C. 2006) (internal quotation marks
and citation omitted). Thus, courts “must scrutinize carefully claims of impossibility,
and ‘separate justifications grounded in the purposes of the Act from the footdragging
efforts of a delinquent agency.’” Id. (quoting Train, 510 F.2d at 713). Courts should
also be wary of agency arguments “that additional time is needed simply to improve the
quality or soundness of the regulations to be enacted[,]” id., or “that competing
regulatory priorities preclude compliance with statutorily-mandated deadlines[,]” id. at
54.
Notwithstanding the heavy burden that an agency bears to prove its inability to
comply with deadlines imposed by a statute that mandates certain agency obligations,
the D.C. Circuit has “recognized two possible legitimate constraints on the agency’s
ability to meet a statutory deadline[.]” Id. at 53. First, a court must be mindful of the
“budgetary commitments and manpower demands [that are] required[,]” and thus avoid
imposing deadlines that “are beyond the agency’s capacity or would unduly jeopardize
10 the implementation of other essential programs.” Train, 510 F.2d at 712. Second, it is
generally acknowledged that, at times, “[the] EPA may be unable to conduct sufficient
evaluation of available control technology to determine which is the best practicable[,]
or may confront problems in determining the components of particular industrial
discharges.” Id. In such circumstances, “additional time [may be] necessary to
[e]nsure that the guidelines are rooted in an understanding of the relative merits of
available control technologies” and “to give meaningful consideration of the technical
intricacies of promising control mechanisms[,]” which “may well speed achievement of
the goal of pollution abatement by obviating the need for time-consuming corrective
measures at a later date.” Id.
III. ANALYSIS
It is clear to this Court that the EPA has failed to satisfy its “heavy burden to
demonstrate . . . impossibility,” Ala. Power Co., 636 F.2d at 359, either with respect to
the commencement of its work on the nine overdue RTRs at issue in this case, or with
respect to the completion of the review processes that Congress has required.
Therefore, as explained below, this Court finds that the EPA’s proposed schedule—
which requests another seven years to complete rulemakings that are already between
six and eight years late—is not the fastest possible schedule, and the Court will give the
EPA until October 1, 2021, to complete its RTRs and any associated rulemaking
activities.
A. The EPA Has Not Demonstrated That It Is Impossible To Begin Work Before March Of 2020
The Court’s first concern arises with respect to the EPA’s contention that it will
not be able to begin the RTRs at issue in this case, or any associated rulemakings, until
11 March of 2020 (see Def.’s Mem. at 9), because of “39 other RTR rulemakings in which
[the] EPA is either currently engaged or will begin shortly” (id. at 13). The agency
points to the “limited number of personnel with the expertise needed to conduct the
necessary reviews and develop additional regulations” (id. at 14–15) as the primary
reason for its proposed delay in starting the work at issue. But the EPA’s assertion that
it needs more than two years to even begin working on the instant RTR rulemakings is
unsatisfactory, for at least two reasons.
First, the agency has admitted that it is possible to “utilize contractors for certain
tasks” and to “use personnel from areas of [the] EPA other than SPPD to work on the
RTR rulemakings[.]” (Id. at 29–30.) The agency’s declarant acknowledges that the
EPA has “re-programmed resources [from other agency divisions and offices] in order
to accelerate” the completion of court-ordered projects in the past, albeit at the expense
of other, unidentified projects. (June Tsirigotis Decl. ¶ 12.) See also Sierra Club, 444
F. Supp. 2d at 54 (“[S]hifting resources in response to statutory requirements and court
orders is commonplace for [the] EPA.” (internal quotation marks and citation omitted)).
And while the EPA implies that its proposed schedule contemplates the use of outside
contractors and other non-SPPD agency resources, it has not demonstrated that it is
impossible for the agency to hire new personnel, including outside contractors, or to
reprogram additional existing agency resources right away, in order to accelerate the
commencement of the instant overdue rulemakings. (See Pls.’ Reply at 27–28.) To be
sure, making such a showing might be difficult, given the agency’s current
representations that it does not need additional funding from Congress to satisfy its
obligations at this time, and in fact can fulfill its “Core Mission” of “Improv[ing] Air
12 Quality” with an almost fifty percent reduction in its air quality-related budget from
Fiscal Year 2017 levels. (See Pls.’ 2d Notice at 4 (internal quotation marks omitted);
Ex. G to Pls.’ 2d Notice, ECF No. 41-9, at 2.) But if the agency’s proposed schedule is
to be countenanced, then it must at least try to make a persuasive claim that it is
currently so cash strapped that it cannot possibly start the mandated reviews given the
resources that it already has, notwithstanding its recent representations.
Second, the EPA’s admission that at least some “of the SPPD’s resources are
devoted to projects that are not subject to statutory deadlines” (Def.’s Reply at 10–11)
undermines its claim that all potentially available resources are already fully utilized on
other court-mandated rulemakings. The agency concedes that at least three “fulltime
equivalents” (“FTEs”) within SPPD are currently dedicated to discretionary, non-
statutorily-mandated projects. (See Decl. of Panagiotis Tsirigotis (“Sept. Tsirigotis
Decl.”) ECF No. 34-1, ¶ 7.) 4 And as Plaintiffs correctly observe, this admission
pertains to “just the number of FTEs within SPPD that [the] EPA has allocated to” some
of the discretionary projects that Plaintiffs have identified in their summary judgment
motion. (Pls.’ Surreply at 4 (emphasis omitted); see also Pls.’ Reply at 28–29 (listing
“discretionary” activities in which the EPA is purportedly engaged and which are not
“statutorily mandated”).) There may well be others; the EPA has made no
representations regarding the total number of otherwise-available FTEs that are
currently assigned to the agency’s other discretionary actions, and while Plaintiffs have
identified other discretionary activities that the agency has taken while this case has
4 The term fulltime equivalent (“FTE”) commonly refers to the total hours that one fulltime employee works in a year, or the equivalent hours worked by several part-time employees.
13 been pending—e.g., efforts to amend previously finalized rules in order to extend
industry compliance deadlines, and steps to repeal certain existing emission guidelines
(see Pls.’ Surreply at 4–6; Pls.’ Notice at 2; Pls.’ 2d Notice at 1–4)—the EPA has
neither denied its participation in such discretionary activities, nor provided any
information about the number of SPPD or other agency staff who are working on such
matters. Without that evidence (which the agency bears the burden of providing under
the circumstances presented here), it certainly seems plausible that the EPA could
reallocate additional FTEs from those kinds of discretionary activities to address the
overdue RTR rulemakings.
Even if no more than three or four FTEs are reallocated to the RTRs in this case,
the record indicates that such an increase would boost the manpower that the EPA
currently has dedicated to completing its mandatory statutory obligations by nearly ten
percent (see Sept. Tsirigotis Decl. ¶ 5 (explaining that thirty-seven FTEs are currently
available to perform the RTR reviews)), which is not a “relatively small” amount (id.
¶ 7), all things considered. Indeed, the EPA admits that relatively few FTEs are needed
to complete each RTR rulemaking. (See id. ¶ 6 (acknowledging that “the least complex
[RTR] projects have historically required about 1 FTE within SPPD[, while t]he most
complex [RTR] projects have required about 3 FTEs within SPPD”).) On a complexity
scale of one to three, the EPA has classified five of the source categories at issue here
as a “one” (the least complex), four source categories as a “two” (medium complexity),
and none as a “three” (the most complex). (See June Tsirigotis Decl. ¶¶ 11–12.) Thus,
it seems that three or more additional FTEs could have a measurable impact on the
EPA’s ability to complete the overdue rulemakings expeditiously.
14 To be clear, this Court does not conclude that other agency activities for which
there is no specific statutory deadline are “frivolous” (Def.’s Reply at 10), and it agrees
with the EPA’s suggestion that attending to such matters as “enforcement actions or
providing guidance to states regarding implementation of environmental regulations”
(id.) is important. Be that as it may, when Congress has ordered the EPA to act by a
certain deadline, “it is inappropriate for [the] agency to divert[,] to purely discretionary
rulemaking[,] resources that conceivably could go towards fulfilling obligations clearly
mandated by Congress.” Sierra Club, 444 F. Supp. 2d at 57. And, indeed, “[t]o accept
such [a practice] in the face of a congressional direction would effectively amount to
condoning a fully discretionary approach to a nondiscretionary duty.” Id. at 54
(internal quotation marks and citation omitted).
Of course, this Court cannot ignore the fact that the EPA is subject to other
recent court orders regarding mandatory review obligations that have constrained the
agency’s ability to devote all of its time and energy to addressing the overdue RTR-
associated rulemakings at issue in this case. (See Def.’s Reply at 13 (urging this Court
not to “ignore the real world”).) See also Sierra Club, 2016 WL 1055120, at *3 (“There
is nothing . . . that suggests the Court must consider how long the EPA could complete
the requested rulemakings in isolation.”). But, again, the applicable standard is
impossibility, and in this Court’s view, the EPA has failed to demonstrate that it would
be impossible for the agency to follow a more expeditious schedule than the one that it
proposes in light of its existing responsibilities, and this is especially so with respect to
the proposed commencement date. Because the EPA has not shown that it cannot hire
outside contractors, or reallocate existing EPA staff from other agency subcomponents,
15 or cease discretionary activities, and thereby start working on the rulemakings at issue
here before March of 2020, this Court will order the EPA to begin the instant RTRs and
associated rulemakings sooner, i.e., after its completion of the first tranche of the
existing court-ordered RTR reviews.
B. The EPA Has Not Demonstrated That It Is Impossible To Complete The Nine-Phase Rulemakings More Expeditiously
As noted above, the EPA has identified nine distinct phases of the RTR
rulemaking process, and it has also predicted how long it believes the agency needs to
complete each phase, and thus the entire rulemaking, for each of the nine instant RTR-
associated rulemakings. (See Appendix A.) But in this Court’s judgment, the EPA has
not shown that it would be impossible to complete the overdue rulemakings on a more
expedited schedule, and in certain respects, it is crystal clear that a faster schedule is
possible, because the agency’s proposal includes significant periods of time to complete
work that may never actually occur.
For example, all of the time that the EPA has allocated for Phase III to collect
supplemental information—eighteen months for three of the source categories, and
seven months for another source category—is, at least today, hypothetical; the EPA has
not made “a final determination” that supplemental information is even required, and
will not make that determination until after “the preliminary information collection
phase for each [rulemaking].” (June Tsirigotis Decl. ¶ 15(a).) The EPA has also built
an additional six months into its proposed schedule for all nine source categories, to
facilitate a review of each rule by the Office of Management and Budget (“OMB”)
pursuant to Executive Order 12866 (see id. ¶¶ 18(f), 21(e)); yet such OMB reviews are
completely discretionary. See, e.g., In re United Mine Workers of Am. Int’l Union, 190
16 F.3d 545, 551 (D.C. Cir. 1999) (rejecting argument that the OMB review provision of
the executive order excused party’s failure to meet a statutory deadline); Nat. Res. Def.
Council, Inc. v. Ruckelshaus, No. 84-cv-758, 1984 WL 6092, at *4 (D.D.C. Sept. 14,
1984) (“OMB review is not only unnecessary, but in contravention to applicable law.”).
(See also June Tsirigotis Decl. ¶¶ 18(f), 21(e) (acknowledging that OMB review is only
conducted “for projects that are considered significant regulatory actions”).)
The EPA’s proposed schedule also includes additional hypothetical, contingent
time that may never be needed. Thus, the agency adds two weeks to the review time for
each source category, in order to conduct “a risk-based demographic assessment[,]” but
it admits that no such review will be needed if it turns out that the “risks are well below
a level of concern[.]” (June Tsirigotis Decl. ¶ 17(d).) This Court does not know
whether the risk-based demographic assessment, or any of the other contingent time
included in the EPA’s schedule, will ultimately be necessary for the agency to fashion
well-reasoned final rules, but neither does the agency, and that is precisely the point.
Under the “impossibility” standard that the D.C. Circuit has adopted, an agency must be
ordered to complete the mandated tasks within the most efficient timeframe possible,
and a schedule that affords time for the EPA to complete tasks that may or may not be
necessary does not comport with the requirement that the minimum possible timeframe
be adopted.
But while the EPA’s proposed schedule in the instant case is too lax, it appears
that Plaintiffs’ proposed schedule is much too draconian. Plaintiffs have provided no
evidence to support their contention that the EPA can possibly complete its RTRs and
associated rulemakings within the next two years. And Plaintiffs are mistaken to rely
17 on the “statutory scheme” and “legislative design” of the Clean Air Act as evidence
with respect to this timing issue. (See Pls.’ Mot. at 46–47 (“[T]he Act directs that,
starting from scratch, [the] EPA must, within two years, promulgate emission standards
for over 40 major source categories, as well as other rulemakings.” (emphasis omitted));
Pls.’ Reply at 42 (“The Clean Air Act directs that [the] EPA will conduct many air
toxics rulemakings in no more than two years . . . and complete a combination of as
many as 165 rules and review rulemakings in two to three years.”).) To begin with, it is
clear on the face of the statute that Congress adopted an eight-year—not a two-year—
window within which the EPA must complete the RTRs and rulemakings at issue here,
and the statute does not speak to how much of that eight-year timeframe should be
dedicated to the mandated process. See 42 U.S.C. § 7412(d)(6), (f)(2); see also Cal.
Cmtys., 241 F. Supp. 3d at 205 (“A standard of how often something should take place
does not describe how long the thing should take.” (emphasis in original)). It is likely
that Congress did not intend for the EPA’s rulemaking to take the entire eight years, but
there is simply no basis for importing the Act’s two-year deadline for promulgating
initial emission standards, see 42 U.S.C. § 7412(e)(1)(A), into the entirely separate
eight-year cycle for completing the RTRs and associated rulemakings. Additionally,
Congress did not speak to a specific remedy for a violation of the eight-year review
period, or mandate a time by which overdue rulemakings must be completed. Coupled
with the lack of specifics to support Plaintiffs’ proposed schedule, this Court simply has
no principled basis for adopting the undifferentiated two-year schedule that Plaintiffs
propose.
18 Taking into account all of the above, while “Plaintiffs’ timeline may be ‘simply
too compressed at this stage to afford any reasonable possibility of compliance[,]’” Cal.
Cmtys., 241 F. Supp. 3d at 207 (quoting Sierra Club, 444 F. Supp. 2d at 58); see also
Blue Ridge, 261 F. Supp. 3d at 61 (same), this Court concludes that the EPA has failed
to demonstrate that its proposed schedule is the fastest possible plan for completing the
instant rulemakings. Luckily, there is a middle ground between Plaintiffs’ schedule and
the EPA’s schedule, and given the various representations by the agency, it is
reasonable to expect the EPA to complete all nine of the overdue RTRs and rulemakings
within thirty-three months of the commencement of its work. (Cf. Def.’s Mem. at 28
(touting two and a half years as “a ‘useful benchmark’ in evaluating what deadlines are
possible for the Agency to meet” (quoting Cal. Cmtys., 241 F. Supp. 3d at 205)).) This
means that once the instant reviews get underway, which will be on or before January 1,
2019, the EPA can be reasonably expected to act diligently to complete all of these
rulemakings by October 1, 2021. And if more time is eventually needed, the agency
can “move for an extension of [the] deadline[s] where appropriate,” as it has done in the
past. (Def.’s Mem. at 27–28; see also Hr’g Tr. at 10:2–5 (indicating that Plaintiffs
would “be very happy” to allow for an extension if the EPA does, for example, “need to
gather additional information”).) See also Train, 510 F.2d at 712 (“[P]alliative
measures may be taken with regard to specific categories if indicated at a later date.”).
IV. CONCLUSION
The EPA has not said that it would be impossible to reallocate existing personnel
from other parts of the agency or to hire new personnel, including outside contractors,
to begin performing at least part of the overdue rulemakings now, nor has it shown that
19 qualified agency staff resources are being spent only on mandatory activities. To top it
off, the EPA has indicated publicly that it has no need for additional resources with
respect to its Clean Air Act responsibilities, and the extended schedule it proposes for
the overdue RTRs at issue here includes time for work that may be unnecessary and
thus may never occur. Therefore, this Court will not accede to the agency’s proposed
timeline, but it will also reject the impossibly compressed deadlines that Plaintiffs
suggest. Instead, as set forth in the accompanying Order, the EPA will be required to
begin the instant rulemakings no later than January 1, 2019, after completing its first
tranche of the existing court-ordered RTR reviews, and it will be further ordered to
complete all nine overdue rulemakings no later than October 1, 2021. Consequently,
Plaintiffs’ motion for summary judgment is GRANTED to the extent that it seeks an
order setting a schedule for the mandatory rulemakings, and DENIED with respect to
the schedule proposed, and Defendant’s request for the schedule set forth in its cross-
motion for summary judgment is DENIED.
DATE: March 31, 2018 Ketanji Brown Jackson KETANJI BROWN JACKSON United States District Judge
20 APPENDIX A Appendix A
EPA ESTIMATE FOR FINAL RULEMAKING BY SOURCE CATEGORY
Phase Phase Phase Phase Phase Phase Phase Phase Phase Final Source Category I II III IV V VI VII VIII IX Rule Days Days Days Days Days Days Days Days Days Date 5 Mercury Emissions 30 45 0 45 104 270 90 90 184 July 22, from Mercury Cell 2022 Chlor-Alkali Plants Semiconductor 60 90 0 90 44 321 90 90 184 Nov. 10, Manufacturing 2022 Generic MACT II – 60 90 0 90 44 349 90 90 184 Dec. 8, Cyanide Chemicals 2022 Manufacturing Generic MACT II – 60 90 0 90 44 349 90 90 184 Dec. 8, Spandex Production 2022 Generic MACT II – 60 90 0 90 104 366 90 90 184 Feb. 23, Carbon Black 2023 Production Primary Copper 60 90 210 90 104 422 90 120 238 Feb. 8, Smelting 2024 Flexible Polyurethane 60 90 547 90 44 390 90 120 238 Oct. 10, Foam Fabrication 2024 Operations Refractory Products 60 90 547 90 104 365 90 120 238 Nov. 14, Manufacturing 2024 Primary Magnesium 60 90 547 90 104 421 90 120 245 Jan. 16, Refining 2025
5 Each of the EPA’s proposed final rule dates is calculated using the agency’s requested start date of March 16, 2020.
A1