Texas v. United States Environmental Protection Agency

Case: 16-60118 Document: 00513595283 Page: 1 Date Filed: 07/15/2016

IN THE UNITED STATES COURT OF APPEALS
FOR THE FIFTH CIRCUIT United States Court of Appeals
Fifth Circuit

FILED
July 15, 2016
No. 16-60118
Lyle W. Cayce
Clerk
STATE OF TEXAS; TEXAS COMMISSION ON ENVIRONMENTAL
QUALITY; PUBLIC UTILITY COMMISSION OF TEXAS; LUMINANT
GENERATION COMPANY, L.L.C.; BIG BROWN POWER COMPANY,
L.L.C.; LUMINANT MINING COMPANY, L.L.C.; BIG BROWN LIGNITE
COMPANY, L.L.C.; LUMINANT BIG BROWN MINING COMPANY, L.L.C.;
SOUTHWESTERN PUBLIC SERVICE COMPANY; UTILITY AIR
REGULATORY GROUP; COLETO CREEK POWER, L.P.; NRG TEXAS
POWER, L.L.C.; NUCOR CORPORATION,

Petitioners,

v.

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY; GINA
MCCARTHY, in her official capacity as Administrator of the United States
Environmental Protection Agency,

Respondents.

Petitions for Review from the
United States Environmental Protection Agency

Before CLEMENT, ELROD, and SOUTHWICK, Circuit Judges.
JENNIFER WALKER ELROD, Circuit Judge:
The State of Texas, numerous energy companies, power plants, steel
mills, consumer organizations, state regulators, and a labor union in Texas
(collectively, “Petitioners”) challenge the Environmental Protection Agency’s
 Case: 16-60118 Document: 00513595283 Page: 2 Date Filed: 07/15/2016

No. 16-60118
action disapproving Oklahoma’s and Texas’s plans for controlling regional
haze and imposing EPA’s own plans instead. Petitioners contend that EPA
has acted outside its statutory authority and seek a stay pending review of
the rule on the merits. EPA moves to dismiss or transfer the petition because
it asserts this court lacks jurisdiction over the petition. Because the Clean Air
Act gives jurisdiction over petitions for review to the courts of appeal
generally and because the Act’s forum selection clause designates the
regional circuit as the appropriate venue for this challenge, we DENY EPA’s
motion to dismiss or transfer. Because Petitioners have demonstrated a
strong likelihood of success on the merits, because they are likely to suffer
irreparable injury in the absence of a stay while EPA has not shown similar
injury from the issuance of a stay, and because the public interest weighs in
favor of a stay, we GRANT the motion for a stay pending resolution of the
petitions for review on the merits.
I.
A. The Clean Air Act’s Regulatory Process

The Clean Air Act is “an experiment in cooperative federalism.”
Michigan v. EPA, 268 F.3d 1075, 1083 (D.C. Cir. 2001). It “establishes a
comprehensive program for controlling and improving the nation’s air quality
through state and federal regulation.” BCCA Appeal Grp. v. EPA, 355 F.3d
817, 821–22 (5th Cir. 2003). The Act requires the states and the federal
government to set and seek to achieve targets for visibility in protected
national parks and wildlife areas by modifying regulations that control air
pollutants in ambient air. 42 U.S.C. §§ 7410, 7491, 7492(e)(2). While the
federal government has the primary responsibility for identifying air
pollutants and setting standards, the states “bear ‘the primary responsibility’
for implementing those standards” by promulgating state implementation

2
 Case: 16-60118 Document: 00513595283 Page: 3 Date Filed: 07/15/2016

No. 16-60118
plans (“SIPs”). Luminant Generation Co. v. EPA, 675 F.3d 917, 921 (5th Cir.
2012) (quoting BCCA Appeal Grp., 355 F.3d at 822).
The Clean Air Act gives each state “wide discretion in formulating its
plan” for achieving the air quality standards set by EPA. Union Elec. Co. v.
EPA, 427 U.S. 246, 250 (1976). “[S]o long as the ultimate effect of a State’s
choice of emission limitations is compliance with the national standards for
ambient air, the State is at liberty to adopt whatever mix of emission
limitations it deems best suited to its particular situation.” Train v. Nat. Res.
Def. Council, Inc., 421 U.S. 60, 79 (1975).
The Clean Air Act confines EPA’s role in implementing air quality
standards “to the ministerial function of reviewing SIPs for consistency with
the Act’s requirements.” Luminant, 675 F.3d at 921. The statute mandates
that “the administrator shall approve such [a state implementation plan] as a
whole if it meets all of the applicable requirements of this chapter.” 42 U.S.C.
§ 7410(k)(3) (emphasis added); see also Fla. Power & Light Co. v. Costle, 650
F.2d 579, 587 (5th Cir. 1981) (“The great flexibility accorded the states under
the Clean Air Act is . . . illustrated by the sharply contrasting, narrow role to
be played by EPA.”); Michigan, 268 F.3d at 1083 (EPA’s “overarching role is
in setting standards, not in implementation.”). “This division of responsibility
between the states and the federal government ‘reflects the balance of state
and federal rights and responsibilities characteristic of our federal system of
government.’” Luminant, 675 F.3d at 921 (quoting Fla. Power & Light, 650
F.2d at 581). The structure of the Clean Air Act indicates a congressional
preference that states, not EPA, drive the regulatory process. As our sister
circuit recently observed, “[d]isagreeing with Congress’s expressly codified
policy choices isn’t a luxury administrative agencies enjoy.” Central United
Life Ins. Co. v. Burwell, ___ F.3d ___, No. 15-5310, 2016 WL 3568084, at *2
(D.C. Cir. July 1, 2016).
3
 Case: 16-60118 Document: 00513595283 Page: 4 Date Filed: 07/15/2016

No. 16-60118
Only if the state has not complied with the requirements of the Clean
Air Act does EPA assume the role of primary regulator by drafting a state-
specific plan. At that point, after disapproving a state implementation plan,
EPA has two years to promulgate a federal implementation plan (“FIP”). 42
U.S.C. § 7410(c)(1). EPA promulgates the federal implementation plan “to fill
all or a portion of a gap . . . in a State implementation plan.” 42 U.S.C.
§ 7602. As a result, EPA’s obligations and authority to promulgate the federal
implementation plan are the same the state had when promulgating its
implementation plan. See, e.g., 77 Fed. Reg. 40,150, 40,164 (July 6, 2012) (“At
the point EPA becomes obligated to promulgate a FIP, EPA steps into the
State’s shoes, and must meet the same requirements. . . .”).
Within this framework, one provision of the Clean Air Act requires
EPA and the states to jointly act to improve visibility at certain protected
federal lands. 42 U.S.C. § 7491. EPA’s obligations under this provision begin
with identifying the federal lands that need improved visibility. 42 U.S.C.
§ 7491(a)(2); 40 C.F.R. §§ 81.400–81.437. After EPA has identified areas for
targeted haze reduction, the Act requires EPA to write regulations providing
the guidelines that states will use to design state implementation plans to
reduce haze in the affected areas. 42 U.S.C. § 7491(b)(1), (2). In 1999, EPA
promulgated the Regional Haze Rule. 40 C.F.R. § 51.308; 64 Fed. Reg. 35,714
(July 1, 1999). The Regional Haze Rule established the guidelines for state
compliance with the air visibility requirements of § 7491. 1
The Regional Haze Rule requires five elements in a state
implementation plan. For each affected wilderness and national park, the
plan must: (1) set “reasonable progress goals” toward achieving natural

1 The guidelines were updated in 2005 after the 1999 version was partially vacated.
Am. Corn Growers Ass’n v. EPA, 291 F.3d 1 (D.C. Cir. 2002) (vacating 1999 Haze Rule
because it imposed technology requirements stricter than those permitted by the Clean Air
Act); 70 Fed. Reg. 39,104 (July 6, 2005) (replacing vacated 1999 Haze Rule).
4
 Case: 16-60118 Document: 00513595283 Page: 5 Date Filed: 07/15/2016

No. 16-60118
visibility conditions that ensure improvements in visibility on the most
impaired days over the period of the implementation plan; (2) calculate
baseline visibility and natural visibility conditions; (3) devise a long-term
strategy with enforceable emissions limitations, compliance schedules, and
other measures necessary to achieve the reasonable progress goals;
(4) develop a monitoring strategy for measuring and reporting visibility; and
(5) list the best available retrofit technology (“BART”) that emission sources
in the state will have to adopt to achieve the visibility goals, along with a
schedule for implementing BART. 40 C.F.R. § 51.308(d), (e).
The Regional Haze Rule also prescribes how states may calculate their
reasonable progress goals. A state begins by calculating the steady linear rate
of decreasing emissions that would achieve natural visibility in the covered
wildernesses and national parks by the year 2064. 40 C.F.R. § 51.308. If a
state determines that the linear rate would result in unreasonable
regulations, it must propose an alternative set of reasonable progress goals
and demonstrate both that the linear rate is unreasonable and that the
alternative goals are reasonable. Id. § 51.308(d)(1)(ii). The Clean Air Act and
the Regional Haze Rule require a state to consider four factors when setting
reasonable progress goals: “the costs of compliance, the time necessary for
compliance, and the energy and nonair quality environmental impacts of
compliance, and the remaining useful life of any existing source subject to
such requirements.” 42 U.S.C. § 7491(g)(1); see also 40 C.F.R.
§ 51.308(d)(1)(i)(A) (repeating the factors listed in § 7491(g)(1)).
BART is the only portion of the implementation plan that is enforced
against emission sources in a state. See 64 Fed. Reg. at 35,733 (“Once a State
has adopted a reasonable progress goal and determined what progress will be
made toward that goal over a 10-year period, the goal itself is not enforceable.

5
 Case: 16-60118 Document: 00513595283 Page: 6 Date Filed: 07/15/2016

No. 16-60118
All that is ‘enforceable’ is the set of control measures which the State has
adopted to meet that goal.”).
If emissions in one state are anticipated to impact the visibility
conditions in protected areas in another state, the Regional Haze Rule
requires the states to consult with one another and develop a coordinated
emission strategy. 40 C.F.R. § 51.308(d)(3). States may participate in regional
planning organizations that jointly approve the technical analyses assessing
the cross-border impact of emissions. Id.

The Regulatory Process under the Clean Air Act
1 2 3 4 5
Clean Air Act Interior and States write EPA approves States
Establishes EPA designate SIPs with SIP or enforce
Regulatory wilderness areas reasonable disapproves BART
Framework and set SIP progress goals and imposes
guidelines and BART FIP

B. The Texas and Oklahoma SIPs
The rulemaking under challenge here concerns visibility in two
national parks and one federal wildlife refuge. 2 The Wichita Mountains
Wildlife Refuge in southwestern Oklahoma was established in 1901 and
protected by statute in 1905. 16 U.S.C. § 684. The refuge is home to scores of
bird, mammal, and fish species, including reintroduced populations of turkey,
bison, and elk. 3 Big Bend National Park was established in 1935 with lands
in southwest Texas donated by the state to the federal government. 16 U.S.C.

2 Changes in visibility are measured in deciviews. A higher deciview measurement
indicates more haze and less visibility. 40 C.F.R. § 51.301. A single deciview is around the
increment that the average person can perceive with the naked eye. Nat’l Parks
Conservation Ass’n v. EPA, 788 F.3d 1134, 1139 n.2 (9th Cir. 2015).
3 Wichita Mountains Wildlife Refuge: About the Refuge, U.S. Fish & Wildlife Serv.,

http://www.fws.gov/refuge/Wichita_Mountains/about.html (last visited May 3, 2016).
6
 Case: 16-60118 Document: 00513595283 Page: 7 Date Filed: 07/15/2016

No. 16-60118
§ 156. Today the park spans over 801,000 acres of desert, mountain, and river
valley and is home to cultural, geological, and biological treasures of the
United States. 4 Guadalupe Mountains National Park in west Texas is the
world’s premier example of a fossilized reef from the Permian Era, more than
250 million years ago. 5 Now it is home to 60 species of mammals, 289 species
of birds and 55 species of reptiles including black bear, grey fox, porcupine,
mountain short-horned lizard, and mountain lion. 6
The Regional Haze Rule requires states to develop an implementation
plan for the period from 2009–2018 and to submit revised plans for each ten-
year period thereafter. 40 C.F.R. § 51.308(b), (f). 7 Initial state
implementation plans were due December 17, 2007. 40 C.F.R. § 51.308(b). In
January 2009, EPA found that thirty-seven states, including Texas and
Oklahoma, had missed the deadline. 74 Fed. Reg. 2392–01 (Jan. 15, 2009).
This finding triggered a two-year deadline for EPA to promulgate a federal
implementation plan. 42 U.S.C. § 7410(c)(1). Before EPA drafted its federal
implementation plan, Texas submitted a state implementation plan on March
31, 2009. More than three years later—and three years into the ten-year
window for this round of implementation plans—EPA issued a limited

4 Michael Welsh, Landscape of Ghosts, River of Dreams: A History of Big Bend
National Park 2–6 (2002).
5 Guadalupe Mountains National Park: Geologic Formations, Nat’l Park Serv.,

https://www.nps.gov/gumo/learn/nature/geologicformations.htm (last visited May 4, 2016).
6 Guadalupe Mountains: Animals, Nat’l Park Serv., https://www.nps.gov/gumo/

learn/nature/animals.htm (last visited May 4, 2016).
7 The Clean Air Act gives EPA flexibility to determine the length of time between

revisions to implementation plans and the length of time each implementation plan will
cover. See 42 U.S.C. § 7491(b)(2)(B) (requiring EPA to draft regulations requiring states to
develop “a long-term (ten to fifteen years) strategy for making reasonable progress toward
meeting the national goal. . . .”). When it promulgated the Regional Haze Rule, EPA elected
to bind states to a ten-year revision period. 40 C.F.R. § 51.308(f). When EPA steps into the
shoes of a state to develop a federal implementation plan, that period is binding on EPA as
it was on the state. See 77 Fed. Reg. at 40,164 (“At the point EPA becomes obligated to
promulgate a FIP, EPA steps into the State’s shoes, and must meet the same
requirements. . . .”).
7
 Case: 16-60118 Document: 00513595283 Page: 8 Date Filed: 07/15/2016

No. 16-60118
disapproval of Texas’s plan. 77 Fed. Reg. 33,642, 33,653 (June 7, 2012). 8
Oklahoma first submitted its state implementation plan in February 2010.
EPA partially disapproved its plan in 2011. 76 Fed. Reg. 81,728 (Dec. 28,
2011).
In 2014, five years after receiving Texas’s SIP and four years after
receiving Oklahoma’s SIP, EPA proposed a federal implementation plan to
replace the parts of the Texas and Oklahoma state implementation plans
that EPA found deficient. 79 Fed. Reg. 74,818 (Dec. 16, 2014). Finally, in
2016—nearly seven years after Texas submitted its implementation plan and
nearly six years after Oklahoma submitted its implementation plan—EPA
promulgated a final rule (“the Final Rule”) partially approving and partially
disapproving the Texas and Oklahoma plans and replacing portions of them
with a federal implementation plan. 81 Fed. Reg. 296 (Jan. 5, 2016). The
Final Rule imposes federal reasonable progress goals for wildlife refuges and
national parks in Texas and Oklahoma but only requires emission controls in
Texas. No Oklahoma power plants or emission sources are affected.
Texas’s state implementation plan included each of the five elements
required by the Regional Haze Rule. Texas concluded that the linear rate
required to achieve natural visibility by 2064 was unreasonable and set an
alternative series of reasonable progress goals. It coordinated with eight
other states—Louisiana, Oklahoma, Arkansas, Kansas, Minnesota, Missouri,
Nebraska, and Iowa—through the Central Regional Air Planning Association
(“CENRAP”) and used CENRAP analysis to assess the impact of Texas

8 State implementation plans for haze reduction need not impose additional
requirements if other emissions controls will achieve the necessary visibility improvements.
Texas relied on the requirements of the Clean Air Interstate Rule (which imposed National
Ambient Air Quality Standards) to meet its haze obligations. After the D.C. Circuit vacated
the Clean Air Interstate Rule, North Carolina v. EPA, 531 F.3d 896, 929 (D.C. Cir. 2008),
modified by 550 F.3d 1176 (remanding for reconsideration without vacatur), EPA
disapproved Texas’s haze plan because it had relied on the vacated Rule’s requirements.
8
 Case: 16-60118 Document: 00513595283 Page: 9 Date Filed: 07/15/2016

No. 16-60118
emissions on protected areas in other states. CENRAP modeled visibility
estimates for 2018 (the close of the current SIP window) and compared the
2018 estimates to the linear rate of progress. Texas determined that current
actual visibility conditions in the covered areas were already better than the
reasonable progress goals it set for 2018. 81 Fed. Reg. at 341; 79 Fed. Reg. at
74,887 (finding that measured visibility already exceeds reasonable progress
goals under both Texas’s state implementation plan and EPA’s alternative
federal implementation plan). Texas considered various emission control
technologies and concluded that any additional technologies would impose
more costs than benefits.
EPA partially approved and partially disapproved Texas’s proposed
plan. 81 Fed. Reg. at 298–99. EPA approved Texas’s measurements of current
visibility conditions in Big Bend, Guadalupe Mountains, and Wichita
Mountains. These measurements show that visibility at all three
wildernesses is already better than the reasonable progress goals in both the
disapproved Texas plan and the replacement federal implementation plan.

Reasonable Progress Goals under Texas’s and EPA’s plans
Texas Goal EPA Goal Texas goal less EPA Actual
(2018) (2018) goal (% of Tex. goal) 9 Visibility 10
Big Bend 16.6 dv 16.57 dv 0.03 dv (0.18%) 16.3 dv
Guadalupe
16.3 dv 16.26 dv 0.04 dv (0.24%) 15.3 dv
Mountains
Wichita
21.47 dv 21.33 dv 0.14 dv (0.65%) 21.2 dv
Mountains

EPA agreed that the linear rate necessary to achieve natural visibility
by 2064 was unreasonable but disapproved Texas’s alternative reasonable

9 A difference of 1.0 deciview is the smallest change perceptible to the average
person.
10This is the recorded visibility on the 20% of days with the worst visibility from
2009–2013. 79 Fed. Reg. at 74,843, 74,870.
9
 Case: 16-60118 Document: 00513595283 Page: 10 Date Filed: 07/15/2016

No. 16-60118
progress goals on the grounds that Texas’s analysis “was not appropriately
refined, targeted, or focused on those sources having the most significant and
potentially cost-effective visibility benefits.” 81 Fed. Reg. at 299. EPA also
disagreed with Texas’s calculation of natural visibility conditions while
approving Texas’s calculation of current baseline visibility. EPA based its
disapproval on a disagreement over the amount of dust that is naturally
occurring in the protected regions. 81 Fed. Reg. at 300. EPA further
disapproved Texas’s long-term strategy, arguing that Texas’s failure to
conduct source-specific analysis also invalidated this portion of the Texas
plan. Id. at 302.

EPA also, in the same rulemaking, disapproved Oklahoma’s plan. EPA
disapproved only the reasonable progress goals in the Oklahoma plan and did
so exclusively because of the effects of Texas emissions on the Wichita
Mountains Wildlife Refuge in Oklahoma. Id. EPA argued that Oklahoma’s
consultation with Texas was “flawed” because Texas “denied [Oklahoma] the
knowledge it needed—the extent to which cost-effective controls were
available for those sources or groups of sources in Texas with the greatest
potential to impact visibility at the Wichita Mountains—in order to properly
construct its reasonable progress goal for the Wichita Mountains.” Id.
Essentially, the only flaw in Oklahoma’s plan springs from EPA’s conclusion
that Texas was required to conduct a source-specific analysis and impose
restrictions on specific sources in Texas. 11
EPA then imposed a federal implementation plan for Texas and
Oklahoma. 81 Fed. Reg. at 303–07. It analyzed emissions individually by

11 The alleged inadequacy of the consultation and the reasonable progress goals that
Oklahoma set as a result of the consultation seem to be Oklahoma’s only involvement in the
Final Rule. All of the additional emission controls in EPA’s federal implementation plan
affect only electrical generating units located in Texas. 81 Fed. Reg. at 298. Oklahoma has
not sought to participate in any of the petitions for review.
10
 Case: 16-60118 Document: 00513595283 Page: 11 Date Filed: 07/15/2016

No. 16-60118
source and selected fifteen electrical generating units with the greatest
impact on the protected areas. The federal implementation plan required
specific emission controls only for these specific sources without consideration
of other controls at other sources. 81 Fed. Reg. at 304–05. EPA demanded
scrubber upgrades at eight facilities and scrubber retrofits at an additional
seven facilities with installation deadlines of 2019 and 2021. 81 Fed. Reg. at
305.
By the time EPA promulgated the Final Rule, only two years remained
in the 2008–18 regulatory window. After promulgating the Final Rule, EPA
issued a notice of proposed rulemaking indicating that it would amend the
Regional Haze Rule and change the governing standards for the second and
all subsequent ten-year planning periods. 81 Fed. Reg. 26,942 (May 4, 2016)
(proposing revisions to 40 C.F.R. § 51.308(f)).
The power companies, labor unions, consumer groups, state regulatory
agencies, steel manufacturers, and state of Texas petitioned for review of the
Final Rule. 12 Petitioners argue that EPA, under the guise of requiring
imperceptible haze reductions, has actually targeted coal-fired power plants.
According to Petitioners, the proposed changes at the targeted power plants
would cost $2 billion, rendering them uneconomical and forcing the plants to
close. The association that manages the Texas power grid, ERCOT, concluded
that EPA’s proposal, which became the Final Rule, would close plants and

12 After filing this challenge here, Petitioners also filed for review of the same rule in
the D.C. Circuit, Texas v. EPA, No. 16-1078 (filed Mar. 4, 2016), consolidated with Nos. 16-
1086, 16-1087, 16-1083, 16-1091, 16-1084, 16-1085, and in the Tenth Circuit, Luminant
Generation Co. v. EPA, No. 16-9508 (filed Mar. 2, 2016), consolidated with Nos. 16-9509, 16-
9511, 16-9512. The D.C. Circuit has suspended the filing deadlines in the consolidated
petitions and has yet to set a briefing schedule. Texas v. EPA, No. 16-1078, (D.C. Cir. Apr.
6, 2016) (order suspending filing deadlines). The Tenth Circuit has ordered the petitions
abated pending resolution of EPA’s motion to dismiss or transfer by this court. Luminant
Generation Co. v. EPA, No. 16-9508 (10th Cir. Jun. 16, 2016) (order granting motions to
intervene, consolidate, and abate).

11
 Case: 16-60118 Document: 00513595283 Page: 12 Date Filed: 07/15/2016

No. 16-60118
remove 3,000 MW to 8,400 MW of generating capacity in Texas. Petitioners
further argue that it would be cost-prohibitive to reopen the affected plants
even if courts ultimately conclude that EPA acted unlawfully. Highlighting
this installment deadlines of 2019 and 2021, Petitioners argue that EPA’s
plan produces no benefits in the period it is intended to cover (2009–2018)
while imposing significant compliance costs.
Petitioners also argue that closure of the plants would substantially
threaten grid reliability in Texas. Electricity rates would increase for Texas
consumers and businesses. Power companies in the state would need to
undertake costly construction projects to create new transmission
infrastructure that would supply power to central Texas. Until infrastructure
projects are completed or new electrical generation facilities are opened in the
state, Petitioners argue that Texas could face power shortages and grid
failures because the federal implementation plan does not include any
exemption for grid reliability.
While the petitions for review are pending, Petitioners request a stay of
the Final Rule to avoid the irreparable damage they assert the federal
implementation plan would impose. EPA has filed a motion to dismiss or
transfer this petition for review to the D.C. Circuit pursuant to the Clean Air
Act’s mandate that an EPA ruling “based on a determination of nationwide
scope or effect” can only be brought in the D.C. Circuit. 42 U.S.C.
§ 7607(b)(1). Petitioners argue, by contrast, that the Final Rule is only
“locally or regionally applicable” and therefore cannot be brought in the D.C.
Circuit. Id.
II.
EPA moved to dismiss or transfer this petition for review arguing that
jurisdiction lies only in the D.C. Circuit and that this court lacks the power to

12
 Case: 16-60118 Document: 00513595283 Page: 13 Date Filed: 07/15/2016

No. 16-60118
consider Petitioners’ challenge. Petitioners respond that jurisdiction and
venue are appropriate in this court.
A.
Section 7607(b)(1) of the Clean Air Act delineates the appropriate
forum for petitions for review. It provides, in relevant part, that:
A petition for review of . . . any . . . nationally applicable
regulations promulgated, or final action taken, by the
Administrator under this chapter may be filed only in the United
States Court of Appeals for the District of Columbia. A petition
for review of the Administrator’s action in approving or
promulgating any implementation plan . . . or any other final
action of the Administrator under this chapter . . . which is
locally or regionally applicable may be filed only in the United
States Court of Appeals for the appropriate circuit.
Notwithstanding the preceding sentence a petition for review of
any action referred to in such sentence may be filed only in the
United States Court of Appeals for the District of Columbia if
such action is based on a determination of nationwide scope or
effect and if in taking such action the Administrator finds and
publishes that such action is based on such a determination.
42 U.S.C. § 7607(b)(1); see also Texas v. EPA, No. 10-60961, 2011 WL 710598,
at *3 (5th Cir. Feb. 24, 2011). Because “the determination of our jurisdiction
is exclusively for the court to decide,” we do not defer to the agency’s
interpretation of this section. Lopez-Elias v. Reno, 209 F.3d 788, 791 (5th Cir.
2000); see also Exelon Wind 1, L.L.C. v. Nelson, 766 F.3d 380, 392 (5th Cir.
2014) (“‘[T]he courts, however, have to make their own determination [about]
jurisdiction, rather than defer to the [federal agency] in the first instance.’”
(first alteration added) (quoting Reeb v. Econ. Opportunity Atlanta, Inc. 516
F.2d 924, 926 (5th Cir. 1975))). 13 Nor do we defer to the agency’s

13Our sister circuits also determine the jurisdiction of the court without deference.
Shweika v. Dep’t of Homeland Sec., 723 F.3d 710, 718 (6th Cir. 2013) (“Chevron deference
does not apply to an agency’s interpretation of a federal court’s jurisdiction.”); Lindstrom v.
United States, 510 F.3d 1191, 1195 n.3 (10th Cir. 2007) (“Determining federal court
13
 Case: 16-60118 Document: 00513595283 Page: 14 Date Filed: 07/15/2016

No. 16-60118
interpretation when determining venue. See Smith v. Aegon Co. Pension
Plan, 769 F.3d 922, 928 (6th Cir. 2014).
Section 7607(b)(1) divides challenges into three general categories.
Petitions for review of nationally applicable actions may only be filed in the
D.C. Circuit. Petitions for review of locally or regionally applicable actions
may only be filed in the regional circuit courts of appeal. 14 Petitions for
review of locally or regionally applicable actions based on a determination
that has nationwide scope or effect may only be filed in the D.C. Circuit.
Texas, 2011 WL 710598, at *3.
EPA argues that the division between the three groups is jurisdictional.
In prior cases, we have treated § 7607(b)(1) as a venue provision without
deciding whether it is jurisdictional. Texas, 2011 WL 710598, at *3 n.28.
Here, the parties continue to dispute whether this provision governs
jurisdiction or venue and, therefore, dispute whether we have power to
consider the motion for a stay. We conclude that § 7607(b)(1) is a two-fold

jurisdiction is exclusively the province of the courts regardless of what an agency may say.”
(internal quotation marks omitted)); Bechtel v. Competitive Techs., Inc., 448 F.3d 469, 478
(2d Cir. 2006) (Leval, J., concurring in the judgment) (“[B]ecause the statutory
interpretation at issue concerns the scope of federal court jurisdiction, it is not a proper
subject of deference under Chevron.”); Verizon Md., Inc. v. Global NAPS, Inc., 377 F.3d 355,
383 (4th Cir. 2004) (Niemeyer, J., concurring in part and dissenting in part) (“Chevron
deference is not required when the ultimate question is about federal jurisdiction.”);
Murphy Exploration & Prod. Co. v. Dep’t of Interior, 252 F.3d 473, 478–80 (D.C. Cir. 2001)
(“Chevron does not apply to statutes that . . . confer jurisdiction on the federal courts.”),
modified on denial of petition for reh’g, 270 F.3d 957 (D.C. Cir. 2001), cited with approval in
NetCoalition v. SEC, 715 F.3d 342, 348 (D.C. Cir. 2013).
14 State implementation plans under the Regional Haze Rule are the subject of

frequent litigation in the regional courts of appeal. See, e.g., Arizona v. EPA, 815 F.3d 519
(9th Cir. 2016) (upholding EPA’s partial disapproval of Arizona’s state implementation plan
and the replacement federal implementation plan); Wildearth Guardians v. EPA, 770 F.3d
919 (10th Cir. 2014) (upholding EPA’s approval of state implementation plans for Wyoming,
New Mexico, and Utah); North Dakota v. EPA, 730 F.3d 750 (8th Cir. 2013) (vacating in
part and upholding in part EPA’s disapproval of North Dakota’s implementation plan and
the replacement federal implementation plan); Oklahoma v. EPA, 723 F.3d 1201 (10th Cir.
2013) (upholding EPA’s partial disapproval of Oklahoma’s state implementation plan and
the replacement federal implementation plan).
14
 Case: 16-60118 Document: 00513595283 Page: 15 Date Filed: 07/15/2016

No. 16-60118
provision. First, it is a “conferral of jurisdiction upon the courts of appeals.”
Harrison v. PPG Indus., Inc., 446 U.S. 578, 593 (1980). Second, its allocation
of petitions between the regional circuits and the D.C. Circuit delineates the
appropriate venue for challenges. 15
This reading accords with the text of the statute, which empowers both
the D.C. Circuit and the regional circuits. The statute is not framed as a
limitation on the power of the courts but as an instruction to petitioners. Our
two-fold reading comports with that of our sister circuit. Dalton Trucking,
Inc. v. EPA, 808 F.3d 875, 878–80 (D.C. Cir. 2015) (holding that § 7607(b)(1)
confers jurisdiction on all courts of appeal and divides venue among them).
B.
Having concluded that § 7607(b)(1)’s three categories delineate venue,
we now conclude that venue is appropriate in the Fifth Circuit because the
present challenge addresses a “locally or regionally applicable” action, which
is not based on a determination that has nationwide scope or effect.
Section 7607(b)(1) is not a model of statutory clarity. Interpreting it
requires close attention to detail. We begin by defining the significant
statutory terms. Section 7607(b)(1) categorizes petitions for review according
to the “action” under challenge. The “action” is the rule or other final action
taken by the agency that the petitioner seeks to prevent or overturn. Here,
the “action” in question is EPA’s final rule disapproving portions of the Texas
and Oklahoma SIPs and imposing a FIP. Section 7607(b)(1) then looks to the
“determination” that the challenged action is “based on.” These
determinations are the justifications the agency gives for the action and they

15 In reading the allocation of petitions between circuit courts as a venue provision,
we are mindful of the Supreme Court’s consistent instruction that courts should not infer
jurisdictional limitations when a statute does not expressly limit jurisdiction. See, e.g., V.L.
v. E.L., 136 S. Ct. 1017, 1021–22 (2016); Arbaugh v. Y&H Corp., 546 U.S. 500, 513–16
(2006).
15
 Case: 16-60118 Document: 00513595283 Page: 16 Date Filed: 07/15/2016

No. 16-60118
can be found in the agency’s explanation of its action. They are the reason the
agency takes the action that it does. For example, because EPA determined
that Texas failed to appropriately calculate reasonable progress goals, EPA
disapproved Texas’s reasonable progress goals and imposed its own. EPA’s
determination about Texas’s goals provided the basis for EPA’s action.
Because the statute speaks of the determinations the action “is based on,” the
relevant determinations are those that lie at the core of the agency action.
Merely peripheral or extraneous determinations are not relevant—the agency
should identify the core determinations in the action. Last, § 7607(b)(1)
speaks of a “find[ing]” and publication by the Administrator. This finding is
an independent, post hoc, conclusion by the agency about the nature of the
determinations; the finding is not, itself, the determination. The finding must
be published as part of the action itself.
Section 7607(b)(1) categorizes petitions for review according to the
nature of the action that the petition challenges. The statute separates
petitions for review of nationally applicable actions from petitions for review
of locally or regionally applicable actions. The question of applicability turns
on the legal impact of the action as a whole. See, e.g., Texas, 2011 WL 710598,
at *3 (action disapproving SIPs from thirteen widely dispersed states and
issuing SIP call requiring states to submit new SIPs is nationally applicable);
Am. Road & Transp. Builders Ass’n v. EPA, 705 F.3d 453, 455–56 (D. C. Cir.
2013) (action disapproving California SIP is locally applicable and must be
filed in Ninth Circuit); ATK Launch Sys., Inc. v. EPA, 651 F.3d 1194, 1199
(10th Cir. 2011) (“[SIPs are] an undisputably regional action [and] the nature

16
 Case: 16-60118 Document: 00513595283 Page: 17 Date Filed: 07/15/2016

No. 16-60118
of the regulation . . . controls.”). The parties agree that the Final Rule under
review is a locally or regionally applicable action. 16
Next, § 7607(b)(1) subdivides challenges to locally or regionally
applicable actions. The default presumption is that petitions for review of
locally or regionally applicable actions “may only be filed in the United States
Court of Appeal for the appropriate circuit.” 42 U.S.C. § 7607(b)(1). The
statute creates an exception, however, for actions “based on a determination
of nationwide scope or effect.” Id. If a challenged action is based on such a
determination, § 7607(b)(1) gives the Administrator the discretion to move
venue to the D.C. Circuit by publishing a finding declaring the
Administrator’s belief that the action is based on a determination of
nationwide scope or effect. EPA argues that the Administrator’s finding alone
is conclusive, but this does not comport with the statutory text. Section
7607(b)(1)’s exception lists two criteria: “(1) if such action is based on a
determination of nationwide scope or effect and (2) if in taking such action
the Administrator finds and publishes that such action is based on such a
determination.” 42 U.S.C. § 7607(b)(1) (emphasis and numerals added).
Because these criteria are listed “if . . . and if . . . ,” both criteria must be
satisfied to transfer venue from the appropriate regional circuit to the D.C.
Circuit.
To determine whether the exception applies, we must answer two
questions: (1) is the action based on a determination that has nationwide

16Section 7607(b)(1) directs that “[a] petition for review of the Administrator’s action
in approving or promulgating any implementation plan . . . or any other final action . . .
which is locally or regionally applicable” may only be filed in the regional circuit. 42 U.S.C.
§ 7607(b)(1). Regardless of whether “which is locally or regionally applicable” modifies only
“any other final action” or both “any other final action” and “action in approving or
promulgating any implementation plan,” see Barnhart v. Thomas, 540 U.S. 20, 27–28
(2003) (describing the last-antecedent canon), the statutory text places review of SIP
approvals or disapprovals in the regional circuits while providing an exception for review of
a small subset of those actions in the D.C. Circuit.
17
 Case: 16-60118 Document: 00513595283 Page: 18 Date Filed: 07/15/2016

No. 16-60118
scope or effect; and (2) did the Administrator publish an adequate finding? 17
EPA argues that the court has no role in assessing whether the action is
based on a determination of nationwide scope or effect. 18 EPA, however,
cannot point to any part of § 7607(b)(1) giving it such exclusive authority. 19
Rather, the statute provides a clear metric by which a court can assess the
scope or effect of the relevant determinations. The reviewing court merely
asks whether the scope 20 or effect 21 of the determinations is nationwide. 22
Numerous cases demonstrate that courts must assess the
“applicability” of the action. See, e.g., Texas, 2011 WL 710598, at *3
(challenged action is nationally applicable); Am. Road & Transp. Builders
Ass’n v. EPA, 705 F.3d 453, 455–56 (D. C. Cir. 2013) (SIP is locally or
regionally applicable); ATK Launch Sys., 651 F.3d at 1199 (SIP call is
nationally applicable); Madison Gas & Elec. Co. v. EPA, 4 F.3d 529, 530–31

17 EPA concedes that courts of appeals can and should independently consider
whether the Administrator has published a suitable finding. If a circuit court holds that the
Administrator made no publication, or made an inadequate publication, the exception
transferring venue to the D.C. Circuit does not apply. See, e.g., Nat’l Parks Conservation
Ass’n v. McCarthy, 816 F.3d 989, 993 (8th Cir. 2016) (EPA made no publication); Lion Oil
Co. v. EPA, 792 F.3d 978, 981–82 (8th Cir. 2015) (EPA publication was defective because
EPA notified regulated party privately rather than publishing finding in Federal Register).
18 In the alternative, EPA argues that only the D.C. Circuit can assess the scope or

effect of the determinations. This argument is unsupported by any statutory text and is
directly contrary to the familiar maxim that “[w]hen judicial review depends on a particular
fact or legal conclusion, then a court may determine whether that condition exists.” Okoro
v. INS, 125 F.3d 920, 925 n.10 (5th Cir. 1997) (internal quotation marks omitted).
19 Nor does EPA establish that it has unreviewable discretion under the standard

test for reviewability. An agency’s conclusions are unreviewable only in limited
circumstances when “(1) statutes preclude judicial review . . . [or (2)] the statute is drawn so
that a court would have no meaningful standard against which to judge the agency’s
exercise of discretion.” Heckler v. Chaney, 470 U.S. 821, 828, 830 (1985); see also 5 U.S.C.
§ 701(a). Although the Administrator may have discretion over the decision to publish a
finding, the statutory text does not give it exclusive discretion to assess the scope or effect
of the determinations.
20 “The area covered by a given activity or subject.” Scope, American Heritage

Dictionary (2d. College Ed. 1982).
21 “Something brought about by a cause or agent; result.” Effect, Id.
22 “Throughout a whole nation.” Nationwide, Id.

18
 Case: 16-60118 Document: 00513595283 Page: 19 Date Filed: 07/15/2016

No. 16-60118
(7th Cir. 1993) (EPA regulation allocating emissions allowances to listed
power plants is locally or regionally applicable). Courts make this decision by
asking whether the action is “nationally applicable” or “locally or regionally
applicable.” Assessing whether a determination’s scope or effect is
“nationwide” requires a similar judicial inquiry.
We consider whether the first condition is satisfied by assessing the
scope or effect of the determinations underlying the challenged action.
Petitioners suggest that we must make our inquiry as to the scope or effect of
the determinations independent from EPA’s finding because the inquiry
governs the powers of the court rather than those of the agency. EPA, by
contrast, argues that we review whether its finding is arbitrary or capricious.
We agree with Petitioners. The exception for locally or regionally applicable
actions based on a determination of nationwide scope or effect has two
conditions. First the action must be based on such a determination and
second the agency must so find and publish. The statute gives EPA discretion
to transfer venue only if the first condition is also satisfied. Because the
answer to the first condition controls the role of the court, we are persuaded
that we must make an independent assessment of the scope of the
determinations just as we make an independent assessment of the
applicability of the action. See Exelon Wind, 766 F.3d at 392; Texas, 2011 WL
710598, at *3–*4 (conducting an independent inquiry into the applicability of
SIP call); ATK Launch Sys., 651 F.3d at 1196–97 (same).
Addressing the first condition de novo, we conclude that the Final Rule
is not based on any determinations that have nationwide scope or effect. EPA
based its disapproval of the Texas and Oklahoma SIPs and its FIP on a

19
 Case: 16-60118 Document: 00513595283 Page: 20 Date Filed: 07/15/2016

No. 16-60118
number of intensely factual determinations. 23 These determinations all
related to the particularities of the emissions sources in Texas and the
confluence of factors impacting visibility at two locations in Texas and one in
southwest Oklahoma. 24
Even assuming, arguendo, that our task is to review the agency’s
conclusion about the scope or effect of the determinations—rather than to

23 EPA’s determinations that Texas’s and Oklahoma’s SIPs have inadequate
reasonable progress goals and long term strategies are the core determinations that are
relevant for the § 7607(b)(1) inquiry, but even assuming, arguendo, that the action is “based
on” all the determinations EPA made in the course of promulgating the Final Rule, we see
none that has nationwide scope or effect. EPA “determined that Texas’ analysis [setting
reasonable progress goals] is inadequate because it does not provide the information
necessary to determine the reasonableness of controls at those sources. . . .” 81 Fed. Reg. at
301. It “determined that Texas has satisfied [the monitoring] requirement.” Id. It
“determined that nine facilities (with 21 units) merited further modeling to assess what the
visibility benefits might be. . . .” 81 Fed. Reg. at 303. It “determined the amount the
baseline visibility values exceeded the natural visibility conditions to calculate visibility
impairment for each area.” 81 Fed. Reg. at 306. EPA “determined that the CAMx
photochemical model was best suited” to model visibility impacts at the affected sites on the
20% worst days. 81 Fed. Reg. at 304. EPA “determine[d] that under a proper assessment of
reasonable progress factors, additional controls for some sources in Texas are warranted for
the first planning period.” 81 Fed. Reg. at 316. EPA “determined that additional controls on
Parish and Welsh were not required for reasonable progress for the first planning period.”
81 Fed. Reg. at 331. Because EPA “determined that the visibility impacts due to [certain]
facilities was almost entirely due to their sulfate emissions [EPA] determined that to
address the visibility impacts on 20% worst days from these sources, it was only necessary
to evaluate sulfate controls for this planning period.” 81 Fed. Reg. at 332. EPA “determined
that it was reasonable to focus [its] analysis on point sources of SO2 and NOX.” Id.
24 Although the SIP process is generally highly fact-bound and particular to the

individual state, EPA has made determinations in other SIP approvals that may have
nationwide scope or effect. For example, in another rulemaking, it determined that CSAPR,
a nationwide regulation governing NAAQS, “would provide for greater reasonable progress
than BART and established regulations that allow certain states to rely on CSAPR to meet
the SO2 and NOX BART requirements.” 81 Fed. Reg. at 301–02 (discussing a determination
made in Regional Haze: Revisions to Provisions Governing Alternatives to Source-Specific
Best Available Retrofit Technology Determinations, 77 Fed. Reg. 33642-01 (June 7, 2012)). A
determination that a national standard satisfies a particular requirement in each state may
be a determination that has nationwide scope or effect. But EPA explicitly declined to make
that determination in the Final Rule: “Given the uncertainty arising from the remand of
Texas’ CSAPR budgets, we have concluded that it would not be appropriate to finalize our
proposed determination to rely on CSAPR as an alternative to SO2 and NOX BART for
EGUs in Texas at this time.” 81 Fed. Reg. at 302.
20
 Case: 16-60118 Document: 00513595283 Page: 21 Date Filed: 07/15/2016

No. 16-60118
make our own independent inquiry—and ask only whether the agency’s
finding is arbitrary or capricious, 25 we are not persuaded that the Final Rule
is based on a determination of nationwide scope or effect. 26 EPA could not
identify with particularity the determinations with nationwide scope or effect
that formed the basis of the Final Rule. EPA advanced two general
arguments in support of its position that the Final Rule is based on a
determination of nationwide scope or effect. These arguments are legal in
nature and do not rely on EPA’s factual conclusions or expertise. First, EPA
argues that because the Final Rule disapproves SIPs from states in two
different circuits, it has nationwide scope or effect. Second, EPA argues that
its rejection of the Texas and Oklahoma reasonable progress goals and long-
term strategies relied on novel interpretations of the Regional Haze Rule and
that the novel interpretation will guide other states designing future SIPs.
EPA’s first argument is unpersuasive and improperly focuses on the
nature of the rule as a whole and not on the determinations on which the
Final Rule is based. It speaks to applicability of the rule, not to the scope or
effect of the relevant determinations. 27

25 [A]n agency must give adequate reasons for its decisions. The agency must
examine the relevant data and articulate a satisfactory explanation for its
action including a rational connection between the facts and the choice made.
That requirement is satisfied when the agency’s explanation is clear enough
that its path may be reasonably discerned. But where the agency has failed to
provide even that minimal level of analysis, its action is arbitrary and
capricious and so cannot carry the force of law.
Encino Motorcars, LLC v. Navarro, ___ S. Ct. ___, No. 15-415, 2016 WL 3369424, at *7
(June 20, 2016) (internal quotation marks omitted).
26 In the Final Rule, EPA found that “our action on the Texas and Oklahoma

regional haze SIPs, which includes the promulgation of a partial FIP for each state, is based
on a determination of nationwide scope and effect.” 81 Fed. Reg. at 346.
27 EPA supports its argument by reference to a single comment in one House Report

commenting on 1977 amendments to § 7607(b)(1). We do not consider passing commentary
in the legislative history, however, when the statutory text itself yields a single meaning.
United States v. Ron Pair Enters., Inc., 489 U.S. 235, 241 (1989); see also Bank One
Chicago, N.A. v. Midwest Bank & Trust Co., 516 U.S. 264, 280 (Scalia, J., concurring)
21
 Case: 16-60118 Document: 00513595283 Page: 22 Date Filed: 07/15/2016

No. 16-60118
EPA’s prior actions also call into question EPA’s assertion here that the
involvement of two circuits gives nationwide scope or effect to the relevant
determinations. While § 7607(b)(1) gives EPA discretion to make and publish
a finding in certain actions while declining to do so in other similar actions,
general principles of administrative law demand that EPA explain its basis
for doing so. See FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515
(2009) (requiring agency to “provide reasoned explanation for its action”).
In another SIP assessment for Michigan and Minnesota—neighboring
states falling in different circuits—EPA did not argue that determinations of
nationwide scope or effect existed even though the plans regulated facilities
in both states. Approval and Promulgation of Regional Haze Implementation
Plan for States of Minnesota and Michigan, 78 Fed. Reg. 8706, 8733 (Feb. 6,
2013) (petitions for judicial review of federal implementation plans “must be
filed in the United States Court of Appeals for the appropriate circuit.”); see
also Disapproval of Interstate Transport Requirements for the 2008 Ozone
NAAQS, 81 Fed. Reg. 38,957-01 (June 15, 2016) (disapproving SIPs for Ohio
(6th Cir.) and Indiana (7th Cir.) without any suggestion of nationwide scope
or effect). EPA may treat different actions differently but EPA must provide
an explanation for its varying treatment. While EPA’s briefing advanced
explanations for treating the Michigan and Minnesota approval differently

(“Legislative history that does not represent the intent of the whole Congress is
nonprobative; and legislative history that does represent the intent of the whole Congress is
fanciful.”). Furthermore, even if we were to consider the House Report, it would weigh in
favor of review in this circuit. The House Report from which EPA extracted its passing
reference explicitly adopted the views expressed by the Administrative Conference of the
United States, which observed that “available transfer provisions” could prevent any
“undue duplication of proceedings.” 41 Fed. Reg. 56,767 (Dec. 30, 1976). The normal
governing transfer provision dictates that venue for a petition for review is proper in the
circuit where a petition for review is first filed. 28 U.S.C. § 2112(a). In the present case,
Petitioners’ petition for review in this circuit was the first filed.
22
 Case: 16-60118 Document: 00513595283 Page: 23 Date Filed: 07/15/2016

No. 16-60118
from the Texas and Oklahoma disapproval, no such explanation appears in
the Final Rule.
Nor are we persuaded by EPA’s second argument that whatever
precedential effect the Final Rule has shows that it is based on
determinations with nationwide scope or effect. Specifically, EPA argues that
its determination that Oklahoma and Texas conducted insufficient
discussions about the impact of Texas emissions on the Wichita Mountains
Wildlife Refuge in Oklahoma will provide guidance to future interstate
consultations. This argument focuses on the appropriate level—the scope or
effect of the determinations that are the basis of the Final Rule—but it too is
unconvincing. To begin with, the argument sweeps too broadly. All SIPs are
likely informed by EPA’s assessment of SIPs from other states. It would
reverse § 7607(b)(1)’s presumption that review of implementation plans
should take place in regional circuits if the guidance one SIP approval
provides another state necessarily gave nationwide scope or effect to EPA’s
determinations. See Am. Road & Transp. Builders, 705 F.3d at 456.
Furthermore, EPA’s argument that a SIP disapproval like the Final
Rule provides interpretive guidance with nationwide scope or effect is
undermined by EPA’s own actions and arguments. In the Final Rule itself,
EPA responded to commenters who alleged inconsistency with other SIP
approvals by arguing that its regulations “do not require uniformity
between . . . actions in all circumstances and instead ‘allow for some
variation’ in actions taken in different regions.” 81 Fed. Reg. at 326 (quoting
Amendments to Regional Consistency Requirements, 80 Fed. Reg. 50,250,
50,258 (Aug. 19, 2015)). Some variation is to be expected because SIP
approvals or disapprovals are highly fact-dependent actions. As EPA itself
insisted during oral argument, the agency’s particular interpretations of the

23
 Case: 16-60118 Document: 00513595283 Page: 24 Date Filed: 07/15/2016

No. 16-60118
Regional Haze Rule in the Final Rule cannot be divorced from the particular
factual context of the Texas and Oklahoma plans.
EPA’s practices confirm that SIP analyses are contextual and variable.
Two years ago, EPA approved a New Mexico SIP with a less stringent
reasonable progress goal for the monitor at the Guadalupe Mountains than
the reasonable progress goal in the disapproved Texas SIP. 28 Had Texas
looked to New Mexico’s SIP for guidance, it would have been misled precisely
because the SIP process is fact-intensive and will vary from state to state.
Finally, as a practical matter, the determinations in the Final Rule will
have no nationwide precedential scope or effect because every other state has
already submitted its SIP for the 2008–18 round. Nor will the Final Rule
impact SIPs considered in the next round because EPA has proposed
revisions of the exact portions of the Regional Haze Rule that EPA claims to
have definitively interpreted in the Final Rule. Protection of Visibility:
Amendments to Requirements for State Plans, 81 Fed. Reg. 26,942, 26,952
(May 4, 2016). If the Final Rule has the impact on other SIPs that EPA
argues it has, no revision of the Regional Haze Rule would be needed.
Because the Final Rule is a locally or regionally applicable action, the
default presumption of § 7607(b)(1) requires review in this circuit. Because
the action is not based on any determinations that have nationwide scope or
effect, the exception to the default presumption does not apply. Therefore,
review is appropriate here, in the regional circuit.

28Compare 77 Fed. Reg. 36,044, 36,071, 36,078 (June 15, 2012) (setting a visibility
goal of 16.92 deciviews for the Carlsbad Caverns National Park measured at “the
IMPROVE monitoring site . . . located in Guadalupe Mountains National Park, Texas”),
with 81 Fed. Reg. at 301, 347 (setting a visibility goal of 16.26 for the Guadalupe Mountains
National Park measured using the IMPROVE monitoring site in Guadalupe Mountains).
24
 Case: 16-60118 Document: 00513595283 Page: 25 Date Filed: 07/15/2016

No. 16-60118
III.
We consider four factors when deciding whether to grant a stay
pending appeal: “(1) whether the stay applicant has made a strong showing
that he is likely to succeed on the merits; (2) whether the applicant will be
irreparably injured absent a stay; (3) whether issuance of the stay will
substantially injure the other parties interested in the proceeding; and (4)
where the public interest lies.” Nken v. Holder, 556 U.S. 418, 426 (2009)
(quoting Hilton v. Braunskill, 481 U.S. 770, 776 (1987)). “A stay is not a
matter of right, even if irreparable injury might otherwise result” to the
appellant. Id. at 433 (quoting Va. Ry. Co. v. United States, 272 U.S. 658, 672
(1926)). Similarly situated petitioners challenging other federal
implementation plans under the Regional Haze Rule have often obtained
stays. See, e.g., Oklahoma v. EPA, 723 F.3d at 1206–07 (staying
implementation of rule imposing $1.2 billion in costs); Wyoming v. EPA, Nos.
14-9529, 14-9530, 14-9533, 14-9534 (10th Cir. Sept. 9, 2014) (staying and
tolling compliance deadlines for rules imposing $700 million in costs); Cliffs
Nat. Res. Inc. v. EPA, Nos. 13-1758, 13-1761 (8th Cir. June 14, 2013) (staying
and tolling compliance deadlines for rules imposing $200 million in costs).
A.
To show a strong likelihood of success on the merits, Petitioners must
demonstrate that EPA acted arbitrarily, capriciously, or unlawfully. 29
Because the BART requirements—the portion of the Final Rule imposing

29 Our determination of Petitioners’ likelihood of success on the merits is for the
purposes of the stay only and does not bind the merits panel. See Mattern v. Eastman
Kodak Co., 104 F.3d 702, 704 (5th Cir. 1997), abrogated on other grounds by Burlington N.
& Santa Fe Ry. Co. v. White, 548 U.S. 53 (2006); see also Planned Parenthood of Greater
Tex. Surgical Health Servs. v. Abbott, 734 F.3d 406, 419 (5th Cir. 2013). By contrast, our
conclusions regarding jurisdiction and venue in evaluating the motion to dismiss or transfer
are not preliminary.

25
 Case: 16-60118 Document: 00513595283 Page: 26 Date Filed: 07/15/2016

No. 16-60118
injury on Petitioners—flow from the federal implementation plan, and
because EPA only has the power to promulgate a federal implementation
plan if it disapproves the state implementation plan, Petitioners can
demonstrate a strong likelihood of success on the merits either by showing
that EPA acted unlawfully when it disapproved the Texas and Oklahoma
SIPs or that EPA acted unlawfully when it drafted the FIP.
The Clean Air Act permits a reviewing court to invalidate any action
taken by EPA that is “arbitrary, capricious, an abuse of discretion, or
otherwise not in accordance with the law . . . in excess of statutory
jurisdiction, authority, or limitations, or short of statutory right; or without
observance of procedure required by law. . . .” 42 U.S.C. § 7607(d)(9); see
Luminant, 675 F.3d at 925 (standard of review of Clean Air Act actions
tracks standards provided by Administrative Procedure Act, 5 U.S.C. § 706).
Agency action:
is arbitrary and capricious “if the agency has relied on factors
which Congress has not intended it to consider, entirely failed to
consider an important aspect of the problem, offered an
explanation for its decision that runs counter to the evidence
before the agency, or is so implausible that it could not be
ascribed to a difference in view or the product of agency
expertise.”
Tex. Oil & Gas Ass’n v. EPA, 161 F.3d 923, 933 (5th Cir. 1998) (quoting Motor
Vehicle Mfrs. Ass’n v. State Farm, 463 U.S. 29, 43 (1983)). “We must
disregard any post hoc rationalizations of EPA’s action and evaluate it solely
on the basis of the agency’s stated rationale at the time of its decision.”
Luminant, 675 F.3d at 925; see Motor Vehicle Mfrs. Ass’n, 463 U.S. at 50 (“It
is well-established that an agency’s action must be upheld, if at all, on the
basis articulated by the agency itself.”); Burlington Truck Lines, Inc. v.
United States, 371 U.S. 156, 168–69 (1962) (“The courts may not accept
appellate counsel’s post hoc rationalizations for agency action; Chenery
26
 Case: 16-60118 Document: 00513595283 Page: 27 Date Filed: 07/15/2016

No. 16-60118
requires that an agency’s discretionary order be upheld, if at all, on the same
basis articulated in the order by the agency itself.” (citing SEC v. Chenery
Corp., 332 U.S. 194, 196 (1947))).
We review factual findings to determine if they are supported by
substantial evidence, but legal conclusions are reviewed de novo. Bd. of Miss.
Levee Comm’rs v. EPA, 674 F.3d 409, 417 (5th Cir. 2012). Because EPA’s
disapproval of Texas’s plan occurred in the form of a rulemaking and has the
force of law, our assessment of EPA’s disapproval is deferential to EPA’s
interpretation of the Clean Air Act if the statute is susceptible to multiple
reasonable interpretations. See United States v. Mead Corp., 533 U.S. 218,
226–27 (2001); Chevron, U.S.A., Inc. v. Nat. Res. Def. Council, 467 U.S. 837,
844 (1984).
If EPA’s action is not permitted by the statute, we must overturn the
action. 5 U.S.C. § 706(2) (“[T]he reviewing court shall . . . hold unlawful and
set aside agency action . . . found to be . . . arbitrary, capricious, an abuse of
discretion, or otherwise not in accordance with law.”); Food and Drug Admin.
v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 125 (“Regardless of how
serious the problem an administrative agency seeks to address, . . . it may not
exercise its authority ‘in a manner that is inconsistent with the
administrative structure that Congress enacted into law.’” (quoting ETSI
Pipeline Project v. Missouri, 484 U.S. 495, 517 (1988))).
Petitioners assert two grounds on which EPA’s disapproval of Texas’s
state implementation plan was unlawful: (1) that EPA exceeded its powers
when it disapproved Texas’s reasonable progress goals and the resulting
long-term strategy despite their compliance with the Clean Air Act; (2) that
EPA acted arbitrarily and capriciously when it disapproved Texas’s
consultation with Oklahoma. Petitioners also assert three independent
grounds on which EPA’s alternative federal implementation plan is unlawful:
27
 Case: 16-60118 Document: 00513595283 Page: 28 Date Filed: 07/15/2016

No. 16-60118
(3) that the federal plan impermissibly relied on effects outside the ten-year
regulatory window in requiring emission controls; (4) that the federal plan
failed to adequately consider costs as required by Michigan v. EPA, 135 S. Ct.
2707 (2015); and (5) that the federal plan failed to consider the effects on grid
reliability in Texas.
We now turn to the particular challenges Petitioners raise regarding
EPA’s disapproval of Texas’s implementation plan and EPA’s alternative
federal implementation and consider whether Petitioners have a strong
likelihood of success on the merits.
1. Reasonable Progress Goals
Petitioners have a strong likelihood of success in showing that EPA
exceeded its statutory authority by disapproving the Texas and Oklahoma
reasonable progress goals even though the goals complied with the Clean Air
Act’s standards. EPA cannot base disapproval on any requirements other
than those listed in the Clean Air Act because EPA has “no authority to
question the wisdom of a State’s choices of emission limitations if they are
part of a plan which satisfies the standards of [§ 7410(a)(2)].” Train, 421 U.S.
at 79. “[T]he Agency may devise and promulgate a specific plan of its own
only if a State fails to submit an implementation plan which satisfies those
standards.” Id.
The Clean Air Act imposes a multi-step process for setting visibility
targets. States begin by estimating natural visibility conditions at the
protected federal lands. The state then assesses the changes necessary to
achieve natural visibility by 2064. If those changes impose unreasonable
costs, the state must devise alternative reasonable progress goals for the
close of the current regulatory window.
The Regional Haze Rule grants states considerable flexibility when
they estimate natural conditions, requiring that they “estimat[e] the degree
28
 Case: 16-60118 Document: 00513595283 Page: 29 Date Filed: 07/15/2016

No. 16-60118
of visibility impairment existing under natural conditions for the most
impaired days and least impaired days, based on available monitoring
information and appropriate data analysis techniques . . . .” 40 C.F.R.
§ 51.308(d)(2)(iii). EPA’s natural visibility guidance expressly permits states
to use refined approaches for the calculation and to “identify [other
approaches] that are more appropriate for their own situations.” 30 Yet EPA
disapproved Texas’s refined estimates for Big Bend and Guadalupe
Mountains because Texas assumed that natural factors such as dust storms
and wildfires contributed 100% of coarse mass and soil to the air on the 20%
of days with the most visibility impairment. EPA devised its own estimate
using an 80% contribution from natural factors. Petitioners argue that EPA
provided scant justification for its substitution other than to allege that
Texas’s estimate was inadequately supported. 81 Fed. Reg. at 325. 31
Once a state has estimated natural visibility conditions, the Regional
Haze Rule requires the state to calculate the changes necessary to achieve
natural visibility by 2064. If those changes are unreasonable, a state is
required to set reasonable progress goals that ensure more gradual progress
toward natural visibility conditions. Both EPA and Texas agree that the
regulatory changes necessary to achieve natural visibility by 2064 are
unreasonable. 81 Fed. Reg. 299. As required by the Clean Air Act and the
Regional Haze Rule, Texas’s state implementation plan proposed an
alternative set of reasonable progress goals.

30 Environmental Protection Agency, Guidance for Estimating Natural Visibility
Conditions Under the Regional Haze Program, 3-2 (2003), available at https://www3.epa.gov
/ttn/caaa/t1/memoranda/rh_envcurhr_gd.pdf.
31 Because EPA’s use of a source-specific analysis provides a sufficient basis for

concluding that EPA’s disapproval of the reasonable progress goals was “arbitrary,
capricious, an abuse of discretion, or otherwise not in accordance with the law,” we do not
address whether EPA acted arbitrarily or exceeded its statutory authority by disapproving
the natural visibility estimates.
29
 Case: 16-60118 Document: 00513595283 Page: 30 Date Filed: 07/15/2016

No. 16-60118
Texas considered emissions from a broad range of sources in Texas and
conducted a holistic analysis of emissions controls for this range of sources.
EPA’s disapproval asserted that this approach was unreasonable and instead
substituted its own source-specific analysis examining the particular costs
and benefits of emissions from particular power plants. 81 Fed. Reg. 298–99.
Texas set 2018 reasonable progress goals of 16.6 dv, 16.3 dv, and 21.47 dv for
Big Bend, Guadalupe Mountains, and Wichita Mountains, respectively.
EPA’s replacement reasonable progress goals instead demanded 16.57 dv,
16.26 dv, and 21.33 dv. 81 Fed. Reg. at 306–07; 79 Fed. Reg. at 74,887 tbl. 43.
EPA’s reasonable progress targets are less than 1% lower than the Texas
goals that EPA found inadequate and current visibility conditions are already
better than the targets set in either Texas’s or EPA’s reasonable progress
goals. EPA’s reasonable progress goals would require a number of costly
changes including: installation of sulfur dioxide scrubbers at seven electrical
generating units and upgrades of existing scrubbers at seven other electrical
generating units. 81 Fed. Reg. at 298.
EPA disapproved both Texas’s and Oklahoma’s goals by arguing that
Texas incorrectly weighed the four statutory factors that govern the
development of reasonable progress goals. See 42 U.S.C. § 7491(g)(1) (“costs of
compliance, the time necessary for compliance, and the energy and nonair
quality environmental impacts of compliance, and the remaining useful life of
any existing source subject to such requirements”). EPA argues that it had
several grounds for disapproving the Texas and Oklahoma goals and suggests
each alone provides a sufficient basis for the disapproval. Most of these
“independent” grounds boil down to EPA’s insistence that Texas should have
conducted a source-specific requirement. Other grounds for disapproval were
asserted in the proposed rule but were not finalized in the Final Rule.
Compare 79 Fed. Reg. at 74,842–43 (proposing disapproval because of Texas’s
30
 Case: 16-60118 Document: 00513595283 Page: 31 Date Filed: 07/15/2016

No. 16-60118
cost threshold, weighing of factors for individual sources, reliance on CAIR
reductions, assumptions about efficiency of SO2 scrubbers, evaluation of
potential improvements, order of magnitude estimate, and scrubber upgrade
estimates), with 81 Fed. Reg. at 298–300 (finalizing disapproval because of
lack of source-specific analysis and estimation of natural visibility
conditions).
EPA’s requirement that Texas conduct a source-specific analysis is not
supported by the Clean Air Act or the Regional Haze Rule. As our sister
circuit held, “[n]either the Clean Air Act nor the Regional Haze Rule requires
source-specific analysis in the determination of reasonable progress goals.”
Wildearth Guardians v. EPA, 770 F.3d 919, 944 (10th Cir. 2014). Indeed,
EPA itself has repeatedly argued that states are not required to use a source-
specific analysis. See, e.g., Wildearth, 770 F.3d at 944; Arizona, 815 F.3d at
539–40 (upholding EPA decision not to conduct source-specific analysis). If
the Clean Air Act empowered EPA to draft reasonable progress goals on a
blank slate, EPA’s action may be permissible, but the Clean Air Act limits
EPA to a deferential role. EPA must defer to Texas’s goals so long as the
Texas goals comply with the Act. See 42 U.S.C. § 7410(k)(3) (EPA “shall
approve” a state implementation plan that satisfies the requirements of the
Act); Luminant, 675 F.3d at 921. EPA’s lack of deference to the state inverts
the agency’s “ministerial function” in this system of “cooperative federalism.”
Luminant, 675 F.3d at 921.
Petitioners are likely to establish that EPA improperly failed to defer to
Texas’s application of the statutory factors and improperly required a source-
specific analysis not found in the Act or Regional Haze Rule.
2. Consultation between Texas and Oklahoma
The Regional Haze Rule requires states to “consult with the other
State(s) in order to develop coordinated emission management strategies” if
31
 Case: 16-60118 Document: 00513595283 Page: 32 Date Filed: 07/15/2016

No. 16-60118
emissions in one state affect visibility at a protected area in another state. 40
C.F.R. § 51.308(d)(3)(i). The Regional Haze Rule adds that states may collect
and project visibility data using a regional planning process and that a state
implementation plan must document its compliance with any agreements
that the regional planning process produces. 40 C.F.R. § 51.308(d)(3)(ii), (iii).
Texas and Oklahoma consulted through CENRAP, the regional
planning association. CENRAP assessed the impact each state’s emissions
had on visibility in other member states. Texas and Oklahoma relied on the
CENRAP process to ensure they satisfied the requirement to control
emissions causing visibility impairment in downwind states.
EPA disapproved Oklahoma’s consultation with Texas because EPA
disagreed with Oklahoma’s decision not to demand further emissions controls
at plants located in Texas. 81 Fed. Reg. at 302–03. EPA’s disapproval seems
to stem in large part from its assertion that Texas had to conduct a source-
specific analysis and provide Oklahoma with that source-specific analysis. 32
Neither the Regional Haze Rule nor the Clean Air Act explicitly
requires upwind states to provide downwind states with source-specific
emission control analysis. The Regional Haze Rule only requires that
“[w]here the State has emissions that are reasonably anticipated to
contribute to visibility in [protected areas] in another State or States, the
State must consult with the other State(s) in order to develop coordinated

32[W]e find that this additional analysis [that Texas provided to Oklahoma]
was inadequate because the large control set Texas selected was not
appropriately refined, targeted, or focused on those sources having significant
and potentially cost-effective visibility benefits and did not provide the
information necessary to determine the reasonableness of controls at those
sources in Texas that have the greatest visibility impacts at the Wichita
Mountains.
81 Fed. Reg. at 300–01.
32
 Case: 16-60118 Document: 00513595283 Page: 33 Date Filed: 07/15/2016

No. 16-60118
emission management strategies.” 40 C.F.R. § 51.308(d)(3)(i). The Clean Air
Act makes no mention of interstate consultation at all.
Given the absence of a regulation or statute requiring source-specific
consultations, the extent of negotiations between CENRAP states, the volume
of analysis produced by CENRAP, and the fact that EPA has never before
disapproved the consultation between states under the Regional Haze Rule,
Petitioners have a strong likelihood of success in showing that EPA’s
disapproval of the consultation between Oklahoma and Texas was arbitrary
and capricious.
3. Effective Date of Emissions Controls
Petitioners also have a strong likelihood of success in establishing that
EPA exceeded its statutory authority by imposing emissions controls that go
into effect years after the period of time covered by the current round of
implementation plans.
The Regional Haze Rule requires states to “consider . . . the emission
reduction measures needed to achieve [the reasonable progress goal] for the
period covered by the implementation plan,” and to impose “enforceable
emissions limitations, compliance schedules, and other measures, as
necessary to achieve the reasonable progress goals.” 40 C.F.R.
§ 51.308(d)(1)(i)(B), (d)(3) (emphasis added). The Regional Haze Rule provides
that each implementation plan will cover a ten-year period; before the close of
each ten-year period, the state must submit a comprehensive revision to
cover the next ten-year period. 40 C.F.R. § 51.308(b), (f) (first implementation
plan due December 2007; first “comprehensive periodic revision” due July 31,
2018, and every ten years thereafter).
The emissions controls included in a state implementation plan,
therefore, must be those designed to achieve the reasonable progress goal for
the period covered by the plan. 40 C.F.R. § 51.308(d)(1)(i)(B). When the EPA
33
 Case: 16-60118 Document: 00513595283 Page: 34 Date Filed: 07/15/2016

No. 16-60118
disapproves a SIP and proposes a FIP, it stands in the position of the state
with all the same requirements and powers the state had in initially drafting
its SIP. Here, the state implementation plans under review only cover the
period up to 2018. 33 Yet EPA’s federal implementation plan requires power
plants in Texas to meet reasonable progress goals by installing scrubbers in
2019 and 2021. 81 Fed. Reg. at 347. Petitioners persuasively argue that this
exceeds the power granted by the Regional Haze Rule.
EPA responds that it has the statutory authority to impose emission
control requirements outside the ten-year window because the Clean Air Act
gave EPA the flexibility to require revised implementation plans at ten- to
fifteen-year intervals. 42 U.S.C. § 7491(b)(2)(B). The minimal deference owed
to an agency interpretation first raised during the course of litigation is
insufficient to persuade us that EPA’s interpretation of the Clean Air Act and
Regional Haze Rule is reasonable. 34 The Clean Air Act may grant EPA the
authority to require new state implementation plans with fifteen-year
coverage periods, but EPA bound states (and accordingly bound itself) to a
ten-year window when it promulgated the Regional Haze Rule.
As Petitioners observe, if EPA wishes to extend the ten-year regulatory
window, it may do so by amending the Regional Haze Rule. EPA, apparently
recognizing this, has already proposed amendments to the Regional Haze
Rule to remove the language tying emissions controls to the reasonable
progress controls “for the period covered by the implementation plan.” 81

33 40 C.F.R. § 51.308(f).
34 EPA raised this argument for the first time in its response to Petitioners’ motion
for a stay and asks the court to defer to this novel interpretation of the Clean Air Act. As we
have stated in prior Clean Air Act cases, Chevron deference does not apply to statutory
interpretations an agency advances in documents, like litigation documents, that do not
bind with the force of law. Luminant, 675 F.3d at 928; see also Calix v. Lynch, 784 F.3d
1000, 1007 (5th Cir. 2015) (Agency interpretations “warrant Chevron deference so long as
they were established prior to the case under consideration.”).
34
 Case: 16-60118 Document: 00513595283 Page: 35 Date Filed: 07/15/2016

No. 16-60118
Fed. Reg. at 26,972. 35 The regulations in effect when Texas and Oklahoma
submitted their plans, however, require states to set the emissions controls
necessary to achieve the reasonable progress goal for 2018. Agency actions
must be assessed according to the statutes and regulations in effect at the
time of the relevant activity. See Caring Hearts Personal Home Servs., Inc. v.
Burwell, No. 14-3243, 2016 WL 3064870 (10th Cir. May 31, 2016) (vacating
CMS sanctions imposed under regulations that came into effect years after
the relevant claims were filed). “[I]t is elementary that an agency must
adhere to its own rules and regulations. Ad hoc departures from those rules,
even to achieve laudable aims, cannot be sanctioned.” Reuters Ltd. v. FCC,
781 F.2d 946, 950 (D.C. Cir. 1986).
EPA also defends the 2019 and 2021 deadlines by arguing that Texas
ought not to benefit from Texas’s delay in promulgating a state
implementation plan. Were there some evidence that Texas’s alleged
intransigence caused the delay in the promulgation of the Final Rule, we
might be inclined to consider this argument. But Texas submitted its
implementation plan for approval in 2009. EPA waited seven years before
finalizing its disapproval in 2016. It does not seem that Texas created
inordinate delay in order to obstruct EPA. EPA may not use its own delay as
an excuse for imposing burdens on Texas that the Regional Haze Rule does
not permit.
Petitioners have a strong likelihood of showing that EPA acted in
excess of its statutory power when it disapproved the Texas state
implementation plan for failing to require scrubbers that will not be installed
until the state implementation plan is no longer in effect.

35 Under the proposed amendments, Petitioners’ argument would be much less
compelling.
35
 Case: 16-60118 Document: 00513595283 Page: 36 Date Filed: 07/15/2016

No. 16-60118
4. Consideration of Costs
Petitioners further challenge EPA’s federal implementation plan,
arguing that EPA did not adequately consider the costs of the power plant
changes—mostly installation of sulfur dioxide scrubbers—when it imposed
the plan.
The Clean Air Act requires EPA to consider “costs of compliance” when
it develops its reasonable progress goals and sets the emission controls
necessary to obtain them. 42 U.S.C. § 7491(g)(1). EPA set a cost threshold for
emissions controls in terms of cost per ton of emissions reduction.
Commenters suggested that because the purpose of the Regional Haze Rule is
to improve visibility, EPA should instead assess costs in terms of cost per
deciview of visibility improvement. EPA declined to do so. 81 Fed. Reg. at
319. Notably, even though the designated areas have all achieved better
visibility than their targets for 2018, 36 the federal implementation plan
imposes emissions controls that will cost $2 billion without achieving any
visibility changes in the time period covered by the plan. Because, as
discussed infra, Petitioners have a strong likelihood of establishing other
flaws in EPA’s federal implementation plan, we need not decide whether
EPA’s use of $/ton metrics instead of $/dv metrics fell short of its obligation to
consider the costs of its regulations, or whether the costs imposed are
unreasonable as a whole in light of the minimal visibility benefits the FIP
would achieve in the relevant time period. See Michigan v. EPA, 135 S. Ct. at
2712; Entergy Corp. v. Riverkeeper, Inc., 556 U.S. 208, 225–26 (2009)
(“[W]hether it is ‘reasonable’ to bear a particular cost may well depend on the
resulting benefits; if the only relevant factor was the feasibility of the costs,
their reasonableness would be irrelevant.”).

3681 Fed. Reg. at 341 (acknowledging that visibility on the 20% worst days is better
than the agency’s reasonable progress goals).
36
 Case: 16-60118 Document: 00513595283 Page: 37 Date Filed: 07/15/2016

No. 16-60118
5. Grid Reliability
The Clean Air Act requires EPA to consider “the energy . . . impacts of
compliance” with the emission controls in a SIP or FIP. 42 U.S.C.
§ 7491(g)(1). Petitioners have a strong likelihood of success in showing that
EPA failed to do so when it devised its FIP because the Final Rule would
render several of the affected electrical generating units uneconomical and
cause the closure of 3,000 to 8,400 MW of generating capacity in Texas.
In its electrical grid, as in so many things, Texas stands alone. While
all the other states in the Union have extensive interconnections with
neighboring states, nearly 90% of Texas is covered by a single isolated grid
with limited connections to external power supplies. This grid shares the
name of its governing board, the Electric Reliability Council of Texas
(ERCOT). Pub. Utility Comm’n of Tex. v. City Pub. Serv. Bd. of San Antonio,
53 S.W. 3d 310, 312 (Tex. 2001). ERCOT’s independence makes the Texas
electrical grid uniquely vulnerable to sudden power shortages when power
plants in the state unexpectedly close because each power plant provides a
larger fraction of the grid’s total power than individual power plants in either
the Western or Eastern Interconnections.
EPA, responding to concerns ERCOT submitted in the notice-and-
comment period, attempted to address all questions about the impact of the
Final Rule on power capacity in Texas in half of one page of the Federal
Register. 81 Fed. Reg. at 345. EPA criticized ERCOT’s rules for failing to
require “meaningful notice” from producers planning to close plants and
suggested that any reliability concerns arose not from the Final Rule but
from ERCOT’s regulatory system. 37 Id. EPA also criticized ERCOT’s

37 ERCOT rules require a plant operator to provide 90 days’ notice before closing an
electrical generating unit in the system. 16 Tex. Admin. Code § 25.502. ERCOT also has
37
 Case: 16-60118 Document: 00513595283 Page: 38 Date Filed: 07/15/2016

No. 16-60118
comment for understating new gas turbine capacity in the state and for
overstating the likelihood of plant closures (although the power company
petitioners in the present case agree with ERCOT’s projections for plant
closures). Id. EPA summarily dismissed concerns about grid reliability in
Texas and relied on a report prepared by a private expert outside the agency
to bolster its conclusion. 81 Fed. Reg. at 345.

Figure 1. Power Grids in the United States. 38

authority to compel power plants to stay online in order to guarantee adequate electrical
supply. Id.
38 Brad Plumer, What Happens If You Add Lots of Wind and Solar Power to the

Grid?, Wash. Post, Sept. 25, 2013, https://www.washingtonpost.com/news/wonk/wp/2013/09/
25/what-happens-if-you-add-lots-of-wind-and-solar-power-to-the-grid/.
38
 Case: 16-60118 Document: 00513595283 Page: 39 Date Filed: 07/15/2016

No. 16-60118
The Final Rule, other than pointing to the report of EPA’s outside
expert, does not detail why the emissions controls in question would not
endanger reliability or cause the closure of up to 8,400 MW of generating
capacity, as ERCOT’s studies suggest. While the agency is free to rely on
outside experts to support its conclusions, the level of deference owed to an
agency’s conclusions is substantially diminished when the subject matter in
question lies beyond the agency’s expertise. Cf. Brown & Williamson Tobacco
Corp., 529 U.S. at 132 (deference is justified by “the agency’s greater
familiarity with the ever-changing facts and circumstances surrounding the
subjects regulated.”). As EPA’s reliance on an outside expert demonstrates,
EPA has no expertise on grid reliability—its sister agency FERC, uninvolved
in this regulatory scheme or this rulemaking, is the federal expert in that
area. Therefore the deference owed to EPA’s assertions about grid reliability
are diminished and the agency must support its arguments more thoroughly
than in those areas in which it has considerable expertise and knowledge.
Particularly when contrasted with the expert report of ERCOT, the group
with the greatest knowledge regarding questions of grid reliability in Texas,
EPA’s truncated discussion of grid reliability indicates that the agency may
not have fulfilled its statutory obligation to consider the energy impacts of
the FIP.
Given the exceptional complexity of grid reliability concerns in Texas,
EPA’s limited authority to dictate how ERCOT should run the Texas grid,
and the explicit directive in the Clean Air Act that implementation plans
“take[] into consideration . . . the energy . . . impacts of compliance,” 42 U.S.C.
§ 7491(g)(1), it is noteworthy that the Final Rule provides neither an
exemption from compliance when necessary to preserve the power supply nor
a more rigorous exploration of the impact of the Final Rule on grid reliability.

39
 Case: 16-60118 Document: 00513595283 Page: 40 Date Filed: 07/15/2016

No. 16-60118
Petitioners have a strong likelihood of success in showing that the Final
Rule’s failure to include either may render it arbitrary and capricious.
B.
Petitioners have demonstrated several irreparable injuries if the Final
Rule is not stayed. They argue that compliance with the Final Rule would
impose $2 billion in costs on power companies, businesses, and consumers.
Because plant emission controls take several years to install, the regulated
companies will have to begin installation almost immediately. The costs of
compliance would not only increase rates for consumers but would also
endanger the reliability of power in ERCOT if plant operators close facilities
rather than install or upgrade uneconomical emissions controls. These
closures would permanently shut down plants with up to 8,400 MW of
generating capacity. 39 The petitioner steel mills and business associations
allege they would also suffer injury as their input costs rise substantially. 40
The petitioner unions argue that their members would lose their employment
at the various power and industrial plants that are threatened by the rule.
Petitioners also state that the absence of a stay would require the Public
Utility Commission of Texas to spend significant resources enforcing
compliance with a voided federal implementation plan rather than enforcing
a valid state implementation plan. Finally, Petitioners assert that allowing
the Final Rule to stand pending the appeal would disrupt the system of
cooperative federalism enshrined in the Clean Air Act. See Michigan v. EPA,
268 F.3d at 1083.

39 Petitioners argue that it is economically infeasible to reopen a plant even if the
regulations are invalidated on the merits.
40 For petitioner Nucor, for example, electricity costs are the second largest

production cost. Substantial rate increases would threaten Nucor’s plants in the state and
the livelihoods of the employees who work there.
40
 Case: 16-60118 Document: 00513595283 Page: 41 Date Filed: 07/15/2016

No. 16-60118
The losses Petitioners allege are sufficient to satisfy the irreparable
injury prong of the stay test. The tremendous costs of the emissions controls
impose a substantial financial injury on the petitioner power companies
which, in this circuit, “may also be sufficient to show irreparable injury.”
Enter. Int’l Inc. v. Corp. Estatal Petrolera Ecuatoriana, 762 F.2d 464, 472–73
(5th Cir. 1985). Indeed “complying with a regulation later held invalid almost
always produces the irreparable harm of nonrecoverable compliance costs.”
Thunder Basin Coal Co. v. Reich, 510 U.S. 200, 220–21 (1994) (Scalia, J.,
concurring in part and in the judgment). When determining whether injury is
irreparable, “it is not so much the magnitude but the irreparability that
counts. . . .” Enter. Int’l, 762 F.2d at 472. No mechanism here exists for the
power companies to recover the compliance costs they will incur if the Final
Rule is invalidated on the merits. 41
EPA relies on precedent from other circuits to argue that all the alleged
injuries are purely financial, and cannot satisfy the irreparable injury prong.
See Am. Hosp. Ass’n v. Harris, 625 F.2d 1328, 1331 (7th Cir. 1980); Freedom
Holdings, Inc. v. Spitzer, 408 F.3d 112, 115 (2d Cir. 2005); Mexichem
Specialty Resins, Inc. v. EPA, 787 F.3d 544 (D.C. Cir. 2015). These same
cases, however, recognize Petitioners’ injury because our sister circuits
categorize financial losses as irreparable injury “where no ‘adequate
compensatory or other corrective relief will be available at a later date, in the
ordinary course of litigation,’” Mexichem, 787 F.3d at 555 (quoting Wis. Gas,

41 EPA suggests that the power companies can request cost recovery from their
customers through the state rate recovery process but this does not eliminate the threat of
injury to Petitioners. The power company petitioners are not guaranteed any rate recovery
would be approved. Approval, if granted, would merely spread the injury more broadly and
increase further the damage to the manufacturing petitioners and consumer groups.
Furthermore, the costs imposed on parties are irreparable where they cannot be recovered
“in the ordinary course of litigation.” Wis. Gas Co. v. FERC, 758 F.2d 669, 674 (D.C. Cir.
1985). Even recoverable costs may constitute irreparable harm “where the loss threatens
the very existence of the movant’s business.” Id.
41
 Case: 16-60118 Document: 00513595283 Page: 42 Date Filed: 07/15/2016

No. 16-60118
758 F.2d at 674), or “where the loss threatens the very existence of the
movant’s business,” Wis. Gas Co., 758 F.2d at 674. The plant closures here
threaten the very existence of some of Petitioners’ businesses and, even
assuming, arguendo, that the plant operators could recover their costs from
ERCOT or their consumers, this would not be a recovery made in the course
of the litigation.
Here Petitioners have raised threatened harms—including
unemployment and the permanent closure of plants—that would arise during
the litigation if a stay is not granted, that are irreparable, and that are great
in magnitude. Even setting aside the costs of compliance for the power
company petitioners, if the Final Rule causes plant closures, the threat of
grid instability and potential brownouts alone constitute irreparable injury to
Texans. Similarly, the institutional injury to Texas from the inversion of the
federalism principles enshrined in the Clean Air Act may constitute
irreparable injury. In sum, Petitioners have shown irreparable injury.
C.
The third and fourth factors ask whether the absence of a stay will
injure other parties and whether the public interest favors or disfavors a
stay. EPA asserts that a stay would injure the public by delaying the
achievement of natural visibility at covered areas under the Clean Air Act.
But EPA’s asserted injury from a stay is unconvincing as it acknowledges
that its proposed implementation plan would not reduce emissions for at
least three years, after the next revision window has opened. Given the
miniscule difference between the 2018 visibility goals in the federal and state
implementation plans, given that current visibility already exceeds even the
federal goals for 2018, and given that the major emissions controls will not
take effect until 2019 and 2021 (although producers would incur significant
costs imminently), we are not persuaded that a stay would injure EPA or the
42
 Case: 16-60118 Document: 00513595283 Page: 43 Date Filed: 07/15/2016

No. 16-60118
intervenors. 42 Petitioners, who themselves represent a broad spectrum of the
public including labor unions, manufacturers, power providers, and consumer
groups, argue that because of the increase in electricity costs and the threat
to grid reliability, the public interest favors a stay.
We agree with Petitioners that the public’s interest in ready access to
affordable electricity outweighs the inconsequential visibility differences that
the federal implementation plan would achieve in the near future. See, e.g.,
Sierra Club v. Ga. Power Co., 180 F.3d 1309, 1311 (11th Cir. 1999) (“[A]
steady supply of electricity during the summer months, especially in the form
of air conditioning to the elderly, hospitals and day care centers, is critical.”);
Tri-State Generation & Transmission Ass’n v. Shoshone River Power, Inc.,
805 F.2d 351, 357 (10th Cir. 1986) (holding that public interest favored an
injunction pending appeal when necessary to preserve power supply to the
public).
D.
We have the power to stay the agency’s action “to the extent necessary
to prevent irreparable injury[.]” 5 U.S.C. § 705. Petitioners request a stay of
the Final Rule in its entirety. EPA, in passing, requests that any stay be
“narrowly tailored.” Because EPA offers nothing beyond this cursory

42 Intervenors Sierra Club and National Parks Conservation Association introduce
expert reports to argue that the FIP would produce health benefits for the public. We are
not persuaded this argument is relevant. Section 7491 is concerned not with health but
with the beauty of our nation’s wild spaces. Other provisions of the Clean Air Act which are
not relevant here protect public health. See, e.g., 42 U.S.C. § 7410. As EPA itself observed,
“for purposes of this action, we are not authorized to specifically consider [health, welfare,
and economic benefits, including ecosystem and tourism benefits] under the regional haze
program.” 81 Fed. Reg. at 325. Even assuming, arguendo, that the technological controls
required by the Final Rule would improve health in a manner relevant to the public
interest component of the stay test, because the emissions controls need not be in place
until 2019 at the earliest, these health benefits (unlike the costs of initiating the scrubber
installation process) would only be realized well after this petition for review concludes
regardless of whether a stay is granted.
43
 Case: 16-60118 Document: 00513595283 Page: 44 Date Filed: 07/15/2016

No. 16-60118
comment, it has waived any argument about the scope of the stay. See United
States v. Green, 964 F.3d 365, 371 (5th Cir. 1992) (“Failure to prosecute an
issue on appeal constitutes waiver of the issue.”). Although it is our
understanding that the emissions controls in the Final Rule are the only
portion of the Final Rule with practical effect and therefore a stay of those
requirements effectively operates as a stay of the Final Rule, neither party
has briefed how we might craft a limited stay. Therefore, we stay the Final
Rule in its entirety, including the emissions control requirements, pending
the outcome of this petition for review.
IV.
Section 7607(b)(1) directs that challenges to EPA’s assessment of a
state implementation plan may only be filed in the appropriate regional
circuit. Because the Final Rule is not based on a determination that has
nationwide scope or effect, the narrow exception in § 7607(b)(1) does not
apply. Venue for this challenge is appropriate in this court.
Petitioners have demonstrated a strong likelihood of success in
establishing that EPA acted arbitrarily, capriciously, and in excess of its
statutory authority when it disapproved the Texas and Oklahoma
implementation plans and imposed a federal implementation plan.
Petitioners have also shown a threat of irreparable injury if a stay is not
granted and have demonstrated that EPA will not suffer injury if a stay is
granted. Finally, Petitioners have shown that the balance of public interests
weigh in favor of a stay. Therefore, Petitioners’ motion for a stay of the Final
Rule is GRANTED.

44
 Case: 16-60118 Document: 00513595283 Page: 45 Date Filed: 07/15/2016

No. 16-60118
LESLIE H. SOUTHWICK, Circuit Judge, concurring.
I concur in the opinion but write separately as to Section III.A.3. The
majority concludes it is strongly likely the EPA improperly required scrubber
installations “after the period of time covered by the current round of
implementation plans,” meaning after the end of the current plan in 2018.
The problem as the petitioners put it is that the EPA has in effect
required that parts of the federal plan be implemented after the end of the
ten-year period. That argument focuses on the back-end of the requirement.
It seems equally proper when determining the EPA’s authority to focus on
the front-end, which is that the construction must start during the ten-year
plan period. The EPA found “that five years is an adequate amount of time
to allow for the installation of scrubber retrofits, and three years is an
adequate amount of time to allow for the installation of scrubber upgrades.”
81 Fed. Reg. 296, 305 (Jan. 5, 2016).
Indeed, the fact that construction must begin now is central to the
argument on irreparable injury. As the Texas Energy petitioners say in their
briefing: “Construction of the new scrubbers that EPA’s rule mandates
involves massive expenditures, extensive coordination, and long lead-times
for planning, design, engineering, procurement, permitting, and actual
construction. EPA’s rule provides the bare minimum amount of time for
completing this work, and the clock has already started ticking.” That
argument clearly accepts that the work ordered on the scrubbers must begin
now, well before the end of the current plan.
I do not believe we need to decide whether the EPA has such authority.
It is unclear to me whether the Regional Haze Rule dictates a time frame
within which projects must be completed. One view is that the recurring
ten-year deadlines for submitting revised state plans are administrative
45
 Case: 16-60118 Document: 00513595283 Page: 46 Date Filed: 07/15/2016

No. 16-60118
benchmarks, checkpoints at which the State and the EPA must review
progress and revise reasonable progress goals and long-term strategies. See
40 C.F.R. § 51.308(f). Stated another way, the Rule does not clearly impose a
requirement that everything begun during a plan term must be completed
during that same term. As just discussed, the enormity of some projects that
might limit pollution from coal plants could require years of construction.
Limiting the State’s or the EPA’s authority to establishing plans that require
every project begun during the plan term also be completed seems
impractical and not clearly required by statute or regulation. Indeed, the
goal of the recurring plans is “to attain natural visibility conditions by 2064.”
Id. § 51.308(d)(1)(i)(B). To reach that goal, plans presumably must be shaped
in a way that considers how far along the road to that goal the current plan’s
steps will take the State.
I close with an observation about the problems of requiring each ten-
year plan to be, as it were, self-contained temporally, problems well
exemplified by the record in this case. Texas’s initial implementation plan
remains un-finalized, and how much of it is to be implemented may well not
be resolved before a revised plan is due on July 31, 2018. Similar delays may
be expected in the submission of revised plans for the next ten-year period.
If, as the majority concludes, a plan may only include projects designed to be
completed before the next revision is due, but the approval of that next plan
may be as delayed as the approval of this one, significant construction
projects may never be feasible. Because I do not see clarity in the statute or
regulations, I am reluctant to state categorically that EPA could not require a
construction project to begin within the plan period even though it would not
end until later.

46
 Case: 16-60118 Document: 00513595283 Page: 47 Date Filed: 07/15/2016

No. 16-60118
Notwithstanding these concerns, I concur in the majority’s conclusion
because the petitioners have a strong likelihood of showing the EPA exceeded
its statutory authority by disapproving Texas’s and Oklahoma’s
implementation plans and imposing a federal implementation plan. I would
grant the petitioner’s motion to stay without addressing the import of the
Rule’s recurring ten-year periods.

47