Lead Opinion
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 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 appro
I.
A. The Clean Air Act’s Regulatory Process
The Clean Air Act is “an experiment in cooperative federalism.” Michigan v. EPA,
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,
The Clean Air Act confínes EPA’s role in implementing air quality standards “to the ministerial function of reviewing SIPs for consistency with the Act’s requirements.” Luminant,
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.
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 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)(l)(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 re
BART is the only portion of the implementation plan that is enforced against emission sources in a state. See
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.
[[Image here]]
B. The Texas and Oklahoma SIPs
The rulemaking under challenge here concerns visibility in two national parks and one federal wildlife refuge.
The Regional Haze Rule requires states to develop an implementation plan for the
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 (Dee. 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 (“CEN-RAP”) and used CENRAP analysis to assess the impact of Texas 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 cur
EPA partially approved and partially disapproved Texas’s proposed plan.
[[Image here]]
EPA agreed that the linear rate necessary to achieve natural visibility by 2064 was unreasonable but disapproved Texas’s alternative reasonable 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.”
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
EPA then imposed a federal implementation plan for Texas and Oklahoma.
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.
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
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 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,
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.
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,
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,
B.
Having concluded that § 7607(b)(l)’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) 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,
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 determine whether the exception applies, we must answer two questions: (1) is the action based on a determination that has nationwide scope or effect; and (2) did the Administrator publish an adequate finding?
Numerous cases demonstrate that courts must assess the “applicability” of the action. See, e.g., Texas,
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,
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 number of intensely factual determinations.
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.
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 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 ef-feet 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)(l)’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,
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 com-menters 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.”
EPA’s practices confirm that SIP analy-ses 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.
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.
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,
A.
To show a strong likelihood of success on the merits, Petitioners must demonstrate that EPA acted arbitrarily, capri
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,
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,
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,
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; (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, — U.S. -,
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,
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 “estimate] the degree 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
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.
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.
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
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,
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 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)(h), (hi).
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.
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)(l)(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)(l)(i)(B). When the EPA 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.
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
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.”
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.
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 re
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,
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.
[[Image here]]
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.,
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. 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.
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,
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,
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 intervenors.
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.,
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 comment, it has waived any argument about the scope of the stay. See United States v. Green,
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’
Notes
. The guidelines were updated in 2005 after the 1999 version was partially vacated. Am. Corn Growers Ass'n v. EPA,
. Changes in visibility are measured in deci-views. 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,
. 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).
. Michael Welsh, Landscape of Ghosts, River of Dreams: A History of Big Bend National Park 2-6 (2002).
. Guadalupe Mountains National Park: Geologic Formations, Nat’l Park Serv., https:// www.nps.gov/gumoAearn/nature/geologic formations.htm (last visited May 4, 2016).
. Guadalupe Mountains: Animals, Nat’l Park Serv., https://www.nps.gov/gumoAearn/ nature/animals.htm (last visited May 4, 2016).
. 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
. 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,
. 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.
. 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).
. Our sister circuits also determine the jurisdiction of the court without deference. Shweika v. Dep't of Homeland Sec.,
. 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,
. 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., - U.S. -,
. Section 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,
. 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,
. 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,
. Nor does EPA establish that it has unre-viewable 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,
. "The area covered by a given activity or subject.” Scope, American Heritage Dictionary (2d. College Ed. 1982).
. "Something brought about by a cause or agent; result.” Effect, Id.
. "Throughout a whole nation.” Nationwide, Id.
. 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....”
. 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 rule-making, it determined that CSAPR, a nation
.[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, - U.S. -,
. 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.”
. 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.,
. Compare 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
. 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.,
. 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/tl/memoranda/rh_envcurhr_gd.pdf.
. Because EPA’s use of a source-specific analysis provides a sufficient basis for con-eluding 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.
. [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 rea
. 40 C.F.R. § 51.308(f).
. 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,
. Under the proposed amendments, Petitioners’ argument would be much less compelling.
.
. 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 authority to compel power plants to stay online in order to guarantee adequate electrical supply. Id.
. Petitioners argue that it is economically infeasible to reopen a plant even if the regulations are invalidated on the merits.
. 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.
. 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. PERC,
. 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
Concurrence Opinion
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 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)(l)(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
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.