Response to application from respondent EPA — Ohio, et al., Applicants v. Environmental Protection Agency, et al.

Supreme Court briefOct 30, 2023

Ask Donna

What actually matters in this document.

Text

Nos. 23A349, 23A350, 23A351

________________________________________________________________

IN THE SUPREME COURT OF THE UNITED STATES

_______________

OHIO, ET AL. (No. 23A349);

KINDER MORGAN, INC., ET AL. (No. 23A350);

AMERICAN FOREST & PAPER ASSOCIATION, ET AL. (No. 23A351),

APPLICANTS

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, ET AL.

_______________

RESPONSE IN OPPOSITION TO THE APPLICATIONS FOR A STAY

_______________

ELIZABETH B. PRELOGAR

Solicitor General

Counsel of Record

Department of Justice

Washington, D.C. 20530-0001

SupremeCtBriefs@usdoj.gov

(202) 514-2217

________________________________________________________________

IN THE SUPREME COURT OF THE UNITED STATES

_______________

Nos. 23A349, 23A350, 23A351

OHIO, ET AL. (No. 23A349);

KINDER MORGAN, INC., ET AL. (No. 23A350);

AMERICAN FOREST & PAPER ASSOCIATION, ET AL. (No. 23A351),

APPLICANTS

UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, ET AL.

_______________

RESPONSE IN OPPOSITION TO THE APPLICATIONS FOR A STAY

_______________

The Solicitor General, on behalf of the United States Environmental Protection Agency (EPA) and Michael S. Regan, Administrator of the EPA, respectfully submits this response in opposition

to the applications for a stay pending the disposition of the

petitions for review.

This case concerns various challenges to the EPA’s final rule

entitled Federal “Good Neighbor Plan” for the 2015 Ozone National

Ambient Air Quality Standards, 88 Fed. Reg. 36,654 (June 5, 2023)

(Rule).

The Rule implements a provision of the Clean Air Act (CAA

or Act), 42 U.S.C. 7401 et seq., that ensures that sources in

upwind States whose pollutant emissions are affecting air quality

in downwind States take the necessary steps to reduce that pollution. See 42 U.S.C. 7410(a)(2)(D)(i)(I). The CAA gives each State

the initial opportunity to submit a plan that will achieve com-

pliance with that requirement.

If a State fails to submit an

adequate plan, however, EPA must promulgate a federal plan to

address the requirements in the State’s place.

42 U.S.C. 7410(c).

The provision is referred to as the Good Neighbor Provision.

accordance with that provision, in February 2023, EPA concluded

that 23 States had failed to submit adequate plans to comply with

revised ozone standards.

EPA then promulgated the Rule to estab-

lish an emissions-control program for large industrial polluters

in those States, based on the same core methodology that this Court

has approved and that EPA has used for decades.

In separate litigation that is not the subject of this suit,

various parties filed petitions for review challenging EPA’s disapproval of 12 state plans, and the relevant regional courts of

appeals stayed the disapproval as to those 12 plans pending the

disposition of those petitions.

EPA recognized that the stays

precluded application of the Rule to sources in those 12 States.

Applicants here -- three States and various industry participants

-- took a different course.

They filed petitions for review in

the D.C. Circuit challenging the federal plan (i.e., the Rule) as

arbitrary and capricious, and they sought a stay of the plan’s

implementation pending the disposition of their petitions for review.

The D.C. Circuit correctly declined to enter a stay.

This

Court should likewise deny applicants’ request for extraordinary

interim relief.

Applicants

contend

that

the

circuit-court

stays

some

state-plan disapprovals in separate litigation undermine the Rule.

But the validity of those disapprovals is not the subject of this

suit and has not been finally determined by any court. The circuit

courts did not stay the disapprovals until after EPA had adopted

the Rule, and those subsequent court actions could not render the

Rule retroactively invalid.

And in any event, EPA’s original

rationales for regulating emissions sources in the 11 States currently subject to the Rule continue to apply with full force and

the Rule continues to function properly in those States, even

though the Rule does not presently apply to sources in the other

12 States.

Applicants also challenge several technical aspects of the

Rule, including its consideration of costs, its applicability criteria, its compliance timeline, and its control requirements for

various industries.

But each of those challenged features of the

Rule is reasonable and adequately explained.

The Rule appropri-

ately regulates industrial sources based on technical and policy

determinations that are supported by a detailed record, and it

provides a variety of compliance flexibilities to ensure that the

necessary emissions reductions can be achieved without overcontrolling or overburdening the industry.

Consistent with the CAA’s

requirements, the Rule thus strikes a proper balance between the

interests of upwind and downwind States.

And applicants further

have not established that their case-specific and record-intensive

objections to the Rule would warrant this Court’s review.

Applicants also have not demonstrated that they will suffer

irreparable harm absent the extraordinary relief they seek.

Many

of the Rule’s challenged aspects do not alter applicants’ obligations until 2026.

EPA’s analysis indicates that near-term capital

expenditures to achieve compliance need not be extensive and will

not endanger natural-gas supply or power-grid operations.

On the

other side of the balance, staying the Rule’s implementation would

significantly harm the public interest.

It would delay efforts to

control pollution that contributes to unhealthy air in downwind

States, which is contrary to Congress’s express directive that

sources in upwind States must assume responsibility for their contributions to emissions levels in downwind States.

By leaving air

pollution caused by upwind States unabated, applicants’ requested

extraordinary relief would impose negative health consequences and

additional regulatory burdens on downwind States and their citizens

-- thus violating the central aim of the Good Neighbor Provision.

The applications should be denied.

STATEMENT

The CAA seeks “to protect and enhance the quality of the

Nation’s air resources so as to promote the public health and

welfare,” 42 U.S.C. 7401(b)(1), and to control air pollution

through a system of shared federal and state responsibility, see

General Motors Corp. v. United States, 496 U.S. 530, 532 (1990).

Title I of the Act requires EPA to establish national ambient air

quality standards (NAAQS or air quality standards) for particular

pollutants at levels that will protect the public health and welfare.

42 U.S.C. 7408, 7409.

The Act also directs States to submit

to EPA state implementation plans to meet those standards.

U.S.C. 7410(a). If EPA determines that a state plan is inadequate,

or if a State fails to submit a plan, EPA must issue a federal

implementation plan within two years after making that determination.

42 U.S.C. 7410(c)(1).

Those provisions reflect Congress’s

effort to “sharply increase[] federal authority and responsibility

in the continuing effort to combat air pollution.”

Train v. Nat-

ural Res. Def. Council, Inc., 421 U.S. 60, 64 (1975).

The Act’s requirements for state plans recognize that “[a]ir

pollution is transient, heedless of state boundaries,” and may be

“transported by air currents” from upwind to downwind States.

EPA

v. EME Homer City Generation, L. P., 572 U.S. 489, 496 (2014).

When air pollution travels beyond the originating State’s boundaries, that State is “relieved of the associated costs,” which are

“borne instead by downwind States, whose ability to achieve and

maintain satisfactory air quality is hampered by the steady stream

of infiltrating pollution.”

Ibid.

To account for that “complex

challenge,” ibid., state plans must include “adequate provisions

* * *

prohibiting

* * *

any source or other type of emissions

activity within the State from emitting any air pollutant in

amounts which will

* * *

contribute significantly to nonattain-

ment in, or interfere with maintenance by, any other State with

respect

any

7410(a)(2)(D)(i)(I).

[air

quality

standard],”

U.S.C.

This statutory requirement, known as the

Good Neighbor Provision, is Congress’s chosen method of balancing

the interests of upwind and downwind States.

EME Homer, 572 U.S.

at 498-499.

EPA has engaged in numerous rulemakings pursuant to the Good

Neighbor Provision. In 1998, EPA limited the emissions of nitrogen

oxide -- a precursor to ozone -- for both power plants and nonelectricity generating units (non-EGUs), including pipeline engines, in 23 upwind States upon finding their existing plans inadequate.

See 63 Fed. Reg. 57,356, 57,358 (Oct. 27, 1998).

The

D.C. Circuit largely upheld that regulation against challenges

brought by power plants, non-EGUs, and States.

See generally

Michigan v. EPA, 213 F.3d 663 (D.C. Cir. 2000) (per curiam), cert.

denied, 532 U.S. 903, and 532 U.S. 904 (2001).

More recently,

this Court upheld a rule that curtailed emissions of 27 upwind

States to assist downwind attainment of three different air quality

standards.

See EME Homer, 572 U.S. at 524.

The Court rejected

contentions that EPA had intruded on state authority or had erred

in using cost as a factor in allocating responsibility among upwind

States.

Ibid.

In many of its rulemakings pursuant to the Good Neighbor

Provision, including those involving ozone, EPA proceeds in four

steps.

First, EPA uses air quality modeling and monitoring data

across the 48 contiguous States to identify areas, known as “receptors,” that are expected to have difficulty attaining or maintaining compliance with the given air quality standard. See, e.g.,

88 Fed. Reg. at 36,659. Second, EPA uses that modeling to quantify

pollutant contributions from upwind States to receptors in downwind States. Ibid. EPA identifies upwind States that are “linked”

to downwind pollution by determining which upwind States contribute more than one percent of the air quality standard to ambient

concentrations of the relevant pollutant at downwind-state receptors.

Ibid.

Third, EPA identifies upwind emissions that “con-

tribute significantly” to nonattainment or interfere with maintenance of air quality standards in downwind States.

7410(a)(2)(D)(i)(I).

42 U.S.C.

In doing so, to ensure that each linked

upwind State does its fair share to reduce the States’ collective

contribution, EPA considers the cost-effectiveness of potential

emissions controls and the total emissions reductions that may be

achieved by requiring such controls, and it evaluates the effect

such reductions would have on air quality in the downwind States.

88 Fed. Reg. at 36,659.

Emissions in excess of the emissions-

control strategies that EPA finds justified under this analysis

are deemed “significant” and therefore prohibited under the CAA.

Id. at 36,659-36,660; EME Homer, 572 U.S. at 519-520.

Fourth, EPA

imposes enforceable control measures to prohibit those “significant” emissions.

88 Fed. Reg. at 36,659-36,664.

The Rule challenged here applies that same regulatory

framework, which has been upheld by both this Court and the D.C.

Circuit.

See EME Homer, 572 U.S. at 524; Michigan, 213 F.3d at

674-679.

In 2015, EPA revised the applicable air quality standard for

ozone, triggering the States’ obligations to submit implementation

plans to comply with that standard.

Upon reviewing those submis-

sions, EPA disapproved 21 state plans for failing to satisfy the

Good Neighbor Provision.

2023).

88 Fed. Reg. 9,336, 9,338 (Feb. 13,

Each of those States had proposed to take no action to

assist downwind neighbors.

Ibid.

EPA then promulgated a federal

plan covering those 21 States, as well as two other States that

had failed to submit plans altogether.

88 Fed. Reg. at 36,654.

EPA first concluded that those 23 States are contributing

significantly to air pollution in other States.

36,659-36,665.

88 Fed. Reg. at

That analysis indicated that within those 23

States, many power plants’ nitrogen-oxide emissions could be more

effectively limited through improved operation of existing controls and by installing control technologies that have been widely

adopted across the industry. Id. at 36,660-36,661. Upon analyzing

non-EGU emissions sources, EPA found that similarly cost-effective

and feasible emissions reductions were available at high-emitting

sources in nine industries, including natural-gas pipelines, cement kilns, steel mills, and paper mills.

Id. at 36,661, 36,664.

To eliminate those emissions that “significantly contribut[e]” to

nonattainment or “interfer[e] with maintenance” of the air quality

standard, EPA adopted an emissions-reduction program covering all

23 States.

Id. at 36,667

For power plants, consistent with previous rules, EPA created

for each covered State a “budget[]” of permissible emissions by

modeling the quantity of pollutants that each source in the upwind

State would emit if all emission reductions EPA identified as

necessary to eliminate significant contributions were implemented.

88 Fed. Reg. at 36,761.

Instead of imposing source-specific emis-

sions limits to reach that budget, the Rule permits sources to

achieve the necessary reductions through an interstate, marketbased trading program that allows covered sources to buy, sell,

and bank emissions allowances, including from sources in other

States.

Id. at 36,904-36,918.

Within this program, power plants

are allocated allowances authorizing emissions at a given level,

with all allowances in the aggregate authorizing emissions only up

to the States’ combined budgets.

commodities.

Allowances are traded like other

Sources that can reduce emissions less expensively

than others therefore may sell their unneeded allowances.

Con-

versely, sources that cannot reduce their emissions as costeffectively may purchase additional allowances on the market.

Although EPA established similar trading programs in previous

rules, EPA’s experience with those programs showed that too much

flexibility could undermine the program’s intended stringency.

The Rule thus announced several enhancements to ensure that emissions deemed “significant” are adequately mitigated.

at 36,657.

88 Fed. Reg.

As relevant here, beginning in the 2026 ozone season

(May 1 through September 30), the Rule implements a dynamic emissions budget-setting procedure.

Id. at 36,765.

EPA explained

that the efficacy of the trading program depends on the stability

of power-plant fleet composition over time.

Id. at 36,764.

For

example, if EPA had required every power plant to implement a given

cost-effective control technology to eliminate its significant

contribution, the retirement of one or more power plants would not

affect the obligations of others.

In the trading program, by

contrast, if multiple power plants retire unexpectedly, EPA’s

budgeted emissions allowances no longer reflect the cost-effective

emissions abatement potential of the remaining plants -- and thus

no longer ensure that significant emissions from those plants are

abated. EPA observed that, under prior rules, the preset emissions

budgets had not kept pace with changes in power-plant fleet composition, so that covered sources had surplus allowances and could

increase emissions in later years, even though decreasing emis-

sions would have been achievable at the cost threshold EPA had

identified as appropriate.

Ibid.

To ensure that the program continues to require the abatement

of significant emissions to the same degree as source-specific

controls, the Rule tailors emissions budgets in later years to the

actual composition of the power-plant fleet.

36,777-36,779.

gradually.

88 Fed. Reg. at

The Rule implements the new budgeting methodology

Through 2029, a preset budget will set the floor, but

a dynamic budget using updated fleet-composition data will be used

to increase the budget if appropriate.

Id. at 36,778.

Starting

in 2030, the budgets will be set exclusively by dynamic budgeting,

allowing the budgets to rise or fall based on fleet composition.

Id. at 36,779.

The Rule also provides for annual recalibration of “bank[s]”

of unused emissions allowances.

88 Fed. Reg. at 36,788.

dynamic budgeting, this change serves to “prevent allowance surpluses from accumulating and adversely impacting the ability of

the trading program in future control periods to maintain” the

“control stringency” that EPA deemed necessary.

Ibid.

Thus, each

year, EPA will set an appropriate percentage of the total budget

that may be banked and will deduct any allowances exceeding that

amount.

Ibid.

For non-EGUs, EPA conducted an initial screening assessment

to identify which industries have the greatest impact on air qual-

ity in downwind States.

88 Fed. Reg. at 36,732-36,733.

See EPA,

Screening Assessment of Potential Emissions Reductions, Air Quality Impacts, and Costs from Non-EGU Emissions Units for 2026 (Feb.

28,

2022)

(Screening

Assessment),

https://perma.cc/AFJ9-7G7Y.

Among the industries it identified, EPA analyzed emissions units

that had emitted more than 100 tons of nitrogen oxide per year.

88 Fed. Reg. at 36,732-36,733.

EPA considered potential air qual-

ity improvements that could be provided to downwind areas by applying various emissions-control strategies to those sources.

making that assessment, EPA identified a marginal cost threshold

-- the point at which further emissions controls generally appear

to become less cost-effective -- at $7500 per ton.

EPA explained

that “this threshold is not intended to represent the maximum cost

any facility may need to expend.”

Id. at 36,733.

Instead, it was

intended as a starting point to begin assessing various industries

and “evaluating technologies” that might be imposed at “different

levels of stringency.”

Ibid.

Based on that threshold, EPA created an initial list of nonEGU emissions units for potential coverage under the Rule. Screening Assessment 3-4.

Using that list as a starting point, EPA

performed a more detailed review of potential emissions controls,

taking into account state and federal emissions standards, technical literature, consent decrees, and permit limits for similar

source types.

88 Fed. Reg. at 36,740.

Based on the updated

analysis, EPA concluded that the $7500-per-ton threshold “does not

reflect the full range of cost-effectiveness values that are likely

present” given the many different types of non-EGU industries and

emissions units.

Id. at 36,746.

Rather, EPA found that a range

of reductions would be cost-effective across the industries, averaging from $939 per ton to $14,595 per ton, with an overall

average of $5339 per ton.

Ibid.

EPA explained that this range

“compares favorably with the values used to evaluate” power plants,

which face representative costs of $11,000 per ton.

Ibid.

EPA

found that the control strategies would meaningfully improve downwind air quality, producing approximately one-third of the total

air quality benefits of the Rule.

Id. at 36,748.

Based on that

assessment, EPA imposed controls on a variety of non-EGU emissions

sources, including natural-gas pipelines, cement kilns, steelindustry reheat furnaces, and paper-industry boilers.

With respect to natural-gas pipelines, EPA’s analysis revealed the potential to eliminate approximately 32,247 tons of

ozone-season nitrogen-oxide emissions from pipeline engines (the

highest level of emissions reductions from any non-EGU industry

covered by the Rule) at an average cost per ton of $4981 -- well

within the representative values that EPA had found justified.

Fed. Reg. at 36,739 (Tbls. V.C.2-1, V.C.2-3). EPA thus established

emissions

limitations

horsepower or greater.

applicable

pipeline

engines

1000

The Rule permits operators to implement an

averaging plan, allowing them to prioritize the most cost-effective

emissions reductions across multiple engines in a facility so long

as the total emissions reductions are at least equivalent to those

that would be individually required.

40 C.F.R. 52.41(d); 88 Fed.

Reg. at 36,823-36,824. Operators may also seek approval for higher

emissions limits if they cannot comply with the applicable limit

“due to technical impossibility or extreme economic hardship.”

C.F.R. 52.40(e).

Pipelines must comply with the emissions limits

by May 1, 2026, with the possibility of compliance extensions

premised upon a showing that the operator cannot meet the compliance date “due to circumstances entirely beyond [its] control.”

40 C.F.R. 52.40(d)(1).

on March 15, 2023.

EPA finalized the Rule and made it public

See Press Release, EPA, EPA Announces Final

“Good Neighbor” Plan to Cut Harmful Smog, Protecting Health of

Millions from Power Plant, Industrial Air Pollution (Mar. 15, 2023)

(EPA Press Release), https://perma.cc/8EUA-7YFG.

In separate litigation, various States and industry

groups challenged EPA’s disapproval of 12 state plans by filing

petitions for review in various federal regional courts of appeals.

Months after EPA had promulgated the Rule implementing the federal

plan, those courts stayed the challenged state-plan disapprovals

pending further review.1

Because EPA’s authority to promulgate a

See Texas v. EPA, No. 23-60069 (5th Cir. May 1, 2023 and

June 8, 2023); Arkansas v. EPA, No. 23-1320 (8th Cir. May 25,

federal plan in those States depended on the agency’s antecedent

determinations that the covered States had not submitted adequate

state plans, EPA recognized that those stays currently preclude

application of the Rule to the 12 States for which stays of the

state-plan disapprovals have been entered.

EPA has issued interim

final rules to address applicable standards in those States while

the stays remain in effect.

See 88 Fed. Reg. 49,295 (July 31,

2023); 88 Fed. Reg. 67,102 (Sept. 29, 2023).

In this case, the applicants here (the States of Ohio,

Indiana, and West Virginia, along with members of industries subject to the Rule as power plants and non-EGU sources) petitioned

for review of the Rule in the D.C. Circuit.2

Shortly thereafter,

applicants moved to stay the Rule pending the disposition of their

petitions for review.

On September 25, 2023, the D.C. Circuit

denied the stay applications.

State Appl. App. A1.

Judge Walker

dissented. Ibid.

2023); Missouri v. EPA, No. 23-1719 (8th Cir. May 26, 2023); Nevada

Cement Co. v. EPA, No. 23-682 (9th Cir. July 3, 2023); ALLETE,

Inc. v. EPA, No. 23-1776 (8th Cir. July 5, 2023); Kentucky v. EPA,

No. 23-3216 (6th Cir. July 25, 2023); Utah v. EPA, No. 23-9509

(10th Cir. July 27, 2023); West Virginia v. EPA, No. 23-1418 (4th

Cir. Aug. 10, 2023) (administrative stay pending disposition of

motions to stay or transfer); Alabama v. EPA, No. 23-11173 (11th

Cir. Aug. 17, 2023).

This brief refers to applicants in No. 23A349 as State

Applicants; applicants in No. 23A350 as Pipeline Applicants; and

applicants in No. 23A351 as AFPA Applicants.

ARGUMENT

The applications should be denied.

Applicants seek what in

practical effect is an injunction against enforcement of the Rule

pending review. To obtain such an injunction, applicants generally

must show that their “claims are likely to prevail, that denying

them relief would lead to irreparable injury, and that granting

relief would not harm the public interest.” Roman Catholic Diocese

v. Cuomo, 141 S. Ct. 63, 66 (2020) (per curiam). A similar standard

applies to a request for a stay.

418, 434 (2009).

See Nken v. Holder, 556 U.S.

But because a request for an injunction seeks

judicial relief that a lower court has withheld, it “‘demands a

significantly higher justification’ than a request for a stay.”

Respect Maine PAC v. McKee, 562 U.S. 996 (2010) (citation omitted).

Such an injunction should be granted “sparingly and only in the

most critical and exigent circumstances,” Wisconsin Right to Life,

Inc. v. FEC, 542 U.S. 1305, 1306 (2004) (Rehnquist, C.J., in chambers) (citation omitted), as when “the legal rights at issue are

‘indisputably clear,’” ibid. (citation omitted); see Roman Catholic Diocese, 141 S. Ct. at 66 (granting injunction where “applicants ha[d] clearly established their entitlement to relief”).

In considering whether a party seeking extraordinary relief

from this Court has made the requisite showing, moreover, the Court

not only considers “the underlying merits” but also makes “a discretionary judgment about whether the Court should grant review in

the case.”

Does 1-3 v. Mills, 142 S. Ct. 17, 18 (2021) (Barrett,

J., concurring in the denial of application for injunctive relief)

(citing Hollingsworth v. Perry, 558 U.S. 183, 190 (2010) (per

curiam)).

“Were the standard otherwise, applicants could use the

emergency docket to force the Court to give a merits preview in

cases that it would be unlikely to take -- and to do so on a short

fuse without benefit of full briefing and oral argument.”

Ibid.

Applicants have not satisfied the standard for a stay, much

less the more demanding standard for an injunction pending review.

Their various challenges to the Rule are not likely to succeed on

the merits because the Rule is a reasonable exercise of EPA’s

authority under the CAA and is not arbitrary, capricious, or otherwise contrary to law.

Applicants also have not shown that any

of those factbound challenges would warrant this Court’s review.

And the balance of equities and the public interest tip decisively

in favor of allowing the Rule to remain in effect, since the Rule

provides important public benefits in reducing harmful ozone levels across the United States.

A stay of the Rule could result in years of delays for the

phase-in of significant reductions in emissions. Such delays would

seriously harm the downwind States that suffer from their upwind

neighbors’

emissions,

placing

the

entire

burden

achieving

healthy air quality on those States and exposing their residents

to public-health risks. On the other side of the scale, applicants

cannot show that they will be irreparably harmed if the Rule re-

mains in effect during the pendency of the D.C. Circuit proceedings.

The Rule sets reasonable compliance deadlines for covered

industry participants, and many of the Rule’s programs do not go

into effect until 2026 or later.

The Rule imposes no requirements

on States at all, and it appropriately balances the State Applicants’ interests against those of the downwind States that the

Good Neighbor Provision protects.

APPLICANTS HAVE NOT ESTABLISHED A LIKELIHOOD OF SUCCESS ON

THE MERITS, MUCH LESS A CLEAR ENTITLEMENT TO RELIEF

Applicants assert that a variety of purported flaws render

the Rule arbitrary and capricious. But the arbitrary-and-capricious

standard is “narrow,” and a reviewing court “is not to substitute

its judgment for that of the agency.”

Motor Vehicle Mfrs. Ass’n

of the U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29,

43 (1983).

The court must assess “whether the decision was based

on a consideration of the relevant factors and whether there has

been a clear error of judgment.”

Ibid. (citation omitted).

The

Rule readily survives review under that deferential standard.

The Stays Of Various State-Plan Disapprovals Entered By

Regional Circuits In Other Litigation Do Not Retroactively Render the Rule Invalid.

Applicants rely substantially on the fact that, months after

EPA finalized the Rule, various regional circuits in other litigation entered orders temporarily staying EPA’s disapprovals of 12

state plans pending judicial review.

Applicants argue that those

stays undermine the Rule and its continued application to the 11

remaining upwind States within its original coverage.

That chal-

lenge is barred by the CAA’s judicial-review provision, and it

lacks merit in any event.

Applicants provide no basis for con-

sidering the Rule’s reasonableness based on events that postdated

its promulgation.

Those arguments do not cast doubt on the Rule’s

validity at the time it was originally promulgated.

Rather, they

are better viewed as claims that the agency should have reconsidered its decision or taken a subsequent action in light of laterarising events.

In any event, EPA’s original rationales for the

Rule continue to apply with full force and the Rule can continue

to function properly in the remaining 11 States, even though the

Rule currently applies to a smaller set of upwind States than EPA

had originally envisioned.

Under the CAA, “[o]nly an objection to a rule

* * *

which was raised with reasonable specificity during the period for

public comment

* * *

U.S.C. 7607(d)(7)(B).

may be raised during judicial review.”

If “it was impracticable to raise such

objection within such time or if the grounds for such objection

arose after the period for public comment,” a party who seeks to

pursue the objection must move for “reconsideration of the rule.”

Ibid.

Only if EPA “refuses to convene such a proceeding” may a

party “seek [judicial] review of such refusal.”

Ibid.

In commenting on the Rule, applicants did not raise before

EPA any objection to the Rule’s continued applicability in cir-

cumstances like those presented here.

Indeed, contrary to State

Applicants’ contention (Appl. 19-20), none of EPA’s state-plan

disapprovals had been stayed before the Rule was finalized.

The

Rule was signed and publicized on March 15, 2023, see EPA Press

Release, and the first stay of a state-plan disapproval was entered

on May 1, 2023, see Texas v. EPA, No. 23-60069 (5th Cir. May 1,

2023).

Applicants thus “did not and could not have raised” a

specific challenge based on those stays “during the period for

public comment”; rather, “the only appropriate path for [applicants] to raise this issue is through an initial petition for

reconsideration.”

EME Homer City Generation, L.P. v. EPA, 795

F.3d 118, 137 (D.C. Cir. 2015).

that requirement.

But applicants have not exhausted

The Court therefore is “without authority” to

reach the question they press here.

Ibid.

In basing their claims on the stays of the state-plan disapprovals, applicants attempt to circumvent the statutorily-mandated

process.

Because those stay orders were entered months after the

Rule was signed and publicized, they have no bearing on whether

the Rule was lawful when it was promulgated.

Applicants cite no

authority for their implicit premise that an agency action may be

rendered retroactively arbitrary and capricious based on events

that occur after it is finalized. Indeed, given the Court’s demand

that agency action be supported based only on the justifications

given “at the time of the agency action,” see, e.g., DHS v. Regents

of the Univ. of Cal., 140 S. Ct. 1891, 1907 (2020) (citation and

emphasis omitted), it would make little sense to evaluate the

reasonableness of those justifications by reference to circumstances that did not yet exist.

Rather, to the extent a change in circumstances merits reconsideration of an agency action, regulated parties can petition

for such reconsideration or other agency action.

But such a claim

is not appropriately brought as a challenge to the original agency

action; it is a challenge to a subsequent failure to reconsider or

take other action in response to later-arising events.

Addressing

applicants’ arguments here would allow them to evade that process

and the “important values of administrative law” that it serves.

Regents, 140 S. Ct. at 1909.

Even if this Court concludes that the applicants’ argu-

ments based on post-promulgation events can properly be asserted

within their current challenge to the Rule, those arguments lack

merit.

The agency action under review in the D.C. Circuit is the

federal plan (i.e., the Rule), not the separate agency action

disapproving state plans that has been stayed in 12 States in other

proceedings that are currently pending before various regional

circuits.

The orders entered by those courts stay the effective-

ness of disapproval as to the 12 States during the pendency of the

review proceedings, but they do not reflect any final judicial

determination as to the legality of EPA’s state-plan disapprovals.

And neither this Court nor the D.C. Circuit is in a position to

determine in this case whether the state-plan disapprovals were

lawful.

Moreover, even if the Court views the post-promulgation

regional-circuit proceedings as relevant to the proper disposition

of applicants’ challenge to the Rule itself, the only question at

this juncture is whether, during the pendency of those proceedings,

it is rational for EPA to continue to apply the Rule to the 11

States whose plan disapprovals have not been challenged.

EPA’s

record and the reasoning underlying the Rule make plain that it

is.

And in any event, the question whether the Rule can continue

to apply in 11 States during the pendency of the regional-circuit

proceedings does not warrant this Court’s review.

As an initial matter, while various regional circuits have

stayed particular state-plan disapprovals, none of those courts

has finally determined whether the challenged disapprovals were

unlawful.

In those proceedings, the government is defending EPA’s

state-plan disapprovals on the merits and has argued that the D.C.

Circuit, rather than the regional circuits, is the proper venue

for those challenges.

It therefore is unclear whether any of the

challenged state-plan disapprovals will ultimately be declared unlawful.

At the conclusion of the various proceedings, there may

be anywhere between 11 and 23 States to which the Rule might

validly apply.

Applicants do not specify what minimum number of

States they believe the Rule must cover in order to constitute a

rational exercise of agency authority.

For States that challenged their state-plan disapprovals, the

various stays put the effectiveness of the disapproval on hold,

thereby limiting EPA’s duty to implement a federal plan as to those

States.

But for States that chose not to challenge their disap-

provals, there is no reason to question EPA’s judgment that the

state plans are invalid.

EPA’s unchallenged state-plan disapprov-

als triggered a statutory obligation to establish, for each of

those 11 States, a federal plan that “achieves something measurable

toward the goal of prohibiting sources ‘within the State’ from

contributing to nonattainment or interfering with maintenance ‘in

any other State.’”

North Carolina v. EPA, 531 F.3d 896, 907 (D.C.

Cir. 2008) (per curiam) (quoting 42 U.S.C. 7410(a)(2)(D)(i)(I)).

And barring a showing of impossibility, EPA must do so in time to

“bring th[ose] State[s] into compliance before upcoming attainment

deadlines.”

Wisconsin v. EPA, 938 F.3d 303, 318 (D.C. Cir. 2019)

(per curiam); see id. at 318-319.

Although applicants argue it is arbitrary and capricious for

the Rule to continue to apply to those 11 States, they do not

identify any alternative interim rule that might apply.

applicants

appear

contemplate

that

emissions

Rather,

within

those

States will not be controlled at all during the pendency of these

proceedings.

That approach ignores the obligations that EPA as-

sumes when it disapproves a state implementation plan.

And given

the uncertain landscape and the present possibility that some or

all of the challenged state-plan disapprovals will ultimately be

upheld, it is particularly reasonable for EPA to fulfill its statutory obligations by continuing to apply the Rule pending resolution of the disapproval litigation.

Nothing about the Rule’s operation undermines that conclusion.

EPA explicitly provided that the Rule is “severable along

* * *

[S]tate and/or tribal jurisdictional lines, such that the

[R]ule can continue to be implemented as to any remaining jurisdictions” even if it is invalidated elsewhere.

88 Fed. Reg. at

36,693. That statement reflected EPA’s recognition that the Rule’s

viability and validity do not depend on the number of jurisdictions

it covers.

See App., infra, 3a-4a.

Consistent with the CAA, EPA could have promulgated 23 separate rules, one for each of the States that lacked an approved

state plan.

Although EPA instead found it efficient to promulgate

a single Rule covering emissions sources in all such States, the

Rule need not apply to any minimum number of States in order to

operate coherently.

App., infra, 3a-7a.

Neither the statutory

Good Neighbor Provision nor the Rule is premised on accomplishing

some minimum total of emissions reductions. Id. at 6a-7a. Rather,

each State must eliminate its own “significant contribution” to

air pollution in downwind States.

See 42 U.S.C. 7410(a)(2)(D)(i).

And it would contradict both the Act and the Rule to allow one

State’s significant contributions to continue unabated merely because EPA’s efforts to abate pollution from other States have been

stayed.

The Rule identified a “uniform level of emissions reduction”

to equitably allocate responsibility among the States, 88 Fed.

Reg. at 36,676, and it applied “emissions control strategies on a

uniform basis,” id. at 36,741; but covered sources in each State

are responsible for eliminating their own significant contribution

to downwind pollution regardless of whether other contributors do

so, see Wisconsin, 938 F.3d at 324-325.

Applicants note (Pipeline

Appl. 11-12; AFPA Appl. 18; State Appl. 17) that the Rule now

regulates fewer emissions than it would if it applied to all 23 of

the States it originally covered, because the stays lessen the

restrictions on emissions in the 12 States for which EPA’s stateplan disapprovals have been stayed, without triggering increased

control obligations on emissions sources in other covered States.

But that simply indicates that (a) the Rule operates State by State

and (b) the judicial stays that currently preclude the Rule’s

application to 12 States do not alter the obligations that the

Rule imposes on emissions sources in the remaining 11, since those

sources remain responsible for their own significant contribution

to downwind pollution.

Nor does the allowance-trading program require participation

of any particular number of States.

EPA’s determinations concern-

ing the amounts of emissions reductions required from covered

sources does not depend on use of the trading program at all.

App., infra, 40a-41a.

The trading program simply smooths the cost

curve by incentivizing sources with cheap reductions to overperform, thereby generating credits that can be purchased by sources

for which reductions would be more expensive.

36,754.

88 Fed. Reg. at

In any event, applicants have not demonstrated that the

trading program will no longer serve its purpose if fewer States

participate.

EPA’s data show the opposite:

prices for emissions

allowances “have dropped significantly in the past several months

and are at the lowest levels since EPA proposed the [Rule],” indicating that allowances are readily available and are likely to

remain so.

App., infra, 40a; see id. at 9a.

Consistent with that

analysis, EPA has previously implemented, with “no issues,” similar trading programs covering 12 or fewer States, including one

that covers sources in a single State.

Id. at 41a; see id. at

18a-19a.

Applicants fare no better in attempting to frame the problem

as a failure to consider, or to provide notice and comment on, a

Rule that currently applies to a smaller number of States than EPA

had anticipated.

See Pipeline Appl. 19; AFPA Appl. 18-19.

already explained, see pp. 20-21, supra, none of EPA’s state-plan

disapprovals had been stayed before the Rule was finalized.

EPA

was not required to foresee that courts of appeals would stay some

subset of the State disapprovals -- temporarily or otherwise.

In any event, EPA’s discussion of severability in its preamble

to the Rule makes clear that the agency did consider whether the

Rule could cogently be applied to a subset of the 23 covered States

-- and concluded that it could.

See App., infra, 4a.

EPA observed

that the Rule established a federal implementation plan for each

covered State and for tribal jurisdictions within those States.

88 Fed. Reg. at 36,693.

The agency then stated that, “[s]hould

any jurisdiction-specific aspect of the final rule be found invalid, the EPA views this rule as severable along those state and/or

tribal jurisdictional lines, such that the rule can continue to be

implemented as to any remaining jurisdictions.”

Ibid.

The agency

explained that this approach to severability “reflects the important public health and environmental benefits of this rulemaking in eliminating significant contribution and to ensure to the

greatest extent possible the ability of both upwind states and

downwind states and other relevant stakeholders to be able to rely

on this final rule in their planning.”

Ibid.

To be sure, EPA did not anticipate the precise combination of

States for which judicial rulings would temporarily preclude the

application of the Rule.

But EPA made clear at the time of prom-

ulgation that it viewed the Rule as capable of coherent application

to a subset of the covered States.

And in the 11 States where

stays of state-plan disapprovals have not been entered, the Rule

continues to subject covered emissions sources to the same requirements that would have applied to them if no stays were in

effect and all 23 States were covered.

Continued application of

the Rule to those sources is both wholly reasonable and demonstrably consistent with EPA’s expressed intent in promulgating the

Rule.3

The Rule’s Regulation Of Both Power Plants And Non-EGU

Emissions Sources Is Reasonable

Pipeline Applicants and AFPA Applicants also challenge numerous technical aspects of the Rule’s regulation of both power

plants and non-EGU emissions sources.

EPA’s determinations as to

the amount of permissible emissions, the applicability criteria,

Pipeline Applicants assert in passing (Appl. 8-9 n.4) that

the Rule implicates the major questions doctrine because it “universally disapprov[es] state plans in favor of a federal plan.”

Even setting aside that EPA’s state-plan disapprovals are not the

agency action at issue here, the major questions doctrine has no

relevance to this case.

The CAA unambiguously requires EPA to

disapprove a state plan and to promulgate a federal plan when a

state plan will not achieve emissions reductions necessary to protect downwind States.

42 U.S.C. 7410(a)(2)(D)(i)(I), (c), and

(k).

This Court has upheld EPA’s use of that authority.

EME

Homer, 572 U.S. at 509-510. And the Rule currently applies only

in States that did not challenge EPA’s disapprovals of their own

implementation plans. The circumstances here thus are far removed

from those in which the Court has previously applied the major

questions doctrine. See, e.g., West Virginia v. EPA, 142 S. Ct.

2587, 2610 (2022) (applying the major questions doctrine when the

agency purported to use an “unheralded power representing a transformative expansion in [its] regulatory authority”) (citation and

quotation marks omitted; brackets in original).

and the implementation timeline are reasonable and well-supported,

as are the enhancements it placed on the power-plant trading program.

EPA lawfully determined the amount of emissions reductions to be required from covered non-EGU

sources

Pipeline Applicants claim (Appl. 13-17) that EPA acted arbitrarily and departed from past practice by failing to adopt a cost

threshold when determining the amount of upwind emissions from

natural-gas pipeline engines that would be deemed significant.

Applicants misunderstand EPA’s methodology and its application

here.

Under EPA’s longstanding approach, the amount of emissions

that will be viewed as significantly contributing to downwind pollution is “that amount of emissions that is in excess of the

emissions control strategies the EPA has deemed cost-effective”

for potentially impactful industries.

88 Fed. Reg. at 36,676; see

id. at 36,678 (noting that the current Rule applies “the same

approach as the prior three” rulemakings); see also EME Homer, 572

U.S. at 519-520 (approving use of this methodology).

To determine

cost-effectiveness, EPA applies a multifactor analysis that assesses cost-per-ton estimates along with comparative emissions reductions and air-quality benefits available from different control

strategies.

88 Fed. Reg. at 36,678-36,679, 36,718-36,719, 36,741.

In performing that analysis here, EPA confirmed that the

available control technologies for pipeline engines were well

within the range of anticipated costs deemed appropriate for other

sources.4 The average cost-per-ton for available pipeline controls

was $4981, see 88 Fed. Reg. at 36,746-36,747, far below the representative control cost for power plants of $11,000 per ton, id.

at 36,746.

And in assessing the efficacy of available controls,

EPA found no drop-off in air quality benefits as the cost-per-ton

rose to the selected stringency level.

Id. at 36,741.

EPA explained that the representative costs it identified do

not establish a cap on what any individual source might spend to

comply. See 88 Fed. Reg. at 36,746.

They are instead intended to

facilitate a comparison of different available controls in different industries, which allows EPA to determine which controls

would optimally yield significant emission reductions with downwind benefits.

Ibid.

Such a use of average representative costs

is consistent with EPA’s approach in prior Good Neighbor rules.

See id. at 36,660, 36,746-36,747 (citing prior Good Neighbor

rules).

And to the extent that certain sources are unable to

implement the presumptive controls due to technical impossibility

In addressing the 2008 ozone standards, EPA performed a

similar comparison of potential reductions at a comparable cost

for power plants and for non-EGU sources. See 88 Fed. Reg. at

36,678. Based on that assessment, EPA determined that emissions

reductions from non-EGU sources were not necessary at that time to

eliminate significant contribution to downwind air quality problems. Ibid.

or extreme economic hardship, EPA allows those sources to comply

with alternative emissions limits.

Id. at 36,818.

Pipeline Applicants contend (Appl. 13-17) that EPA selected

a $7500 threshold in its proposed rule and then improperly abandoned it in the final Rule. That is incorrect. The $7500 threshold

was never intended to serve as a cap for purposes of the “significant contribution” determination.

88 Fed. Reg. at 36,740.

Ra-

ther, EPA used that metric in its initial Screening Assessment to

determine what non-EGU industries and emissions-unit types had

potential for meaningful emissions reductions.

See ibid.; EPA,

Federal “Good Neighbor Plan” for the 2015 Ozone National Ambient

Air Quality Standards, Response to Public Comments on Proposed

Rule

(Mar.

2023)

(RTC),

https://perma.cc/6DY8-Y5G4.

That

screening allowed EPA to focus on nine industries out of the 41

the agency had initially identified, and it served as a starting

point for the analysis of appropriate controls.

RTC 97.

EPA

explained, however, that “the results of the Screening Assessment

should not be confused with regulatory requirements, applicability

determinations, or emissions limits.”

RTC 99.

EPA’s post-Screening

Assessment analysis accounts for additional data, improved understanding, and consideration of comments.

After undertaking that

analysis, EPA reasonably determined an appropriate emissions reduction for natural-gas pipeline engines that is consistent with

past practice and the controls for power-plant emissions.

EPA reasonably applied a 1000-horsepower applicability criterion for natural-gas pipeline engines

Pipeline Applicants object (Appl. 21-23) to the Rule’s 1000horsepower applicability criterion for natural-gas pipeline engines.

Their arguments on that point reflect the same misunder-

standing as their objection to EPA’s determination of the appropriate amount of emissions reductions.

In the initial Screening Assessment, EPA evaluated units with

100 tons per year of actual, historical emissions to identify the

most impactful potential emissions-reductions opportunities.

Fed. Reg. at 36,732-36,733; see Screening Assessment 2-3.

the $7500 threshold, this initial assessment provided a starting

point that allowed EPA to focus its analysis on the largest emitters with the most significant potential reductions.

It did not

reflect an EPA determination that particular pipeline engines make

a significant contribution to downwind air pollution.

After EPA

had performed the Screening Assessment, the agency considered numerous additional factors to determine what reductions to require.

88 Fed. Reg. at 36,740.

When it completed that assessment, EPA determined that most

of the existing standards that the agency had reviewed establish

applicability criteria for pipeline engines based on design capacity rather than historic emissions.

88 Fed. Reg. at 36,821.

For consistency with those requirements, EPA selected a design

capacity of 1000 horsepower, which it determined would capture the

relevant engines.

Ibid.

EPA recognized that use of a design-

capacity threshold “may capture low-use units and some units with

emissions of less than 100 tons per year.”

Ibid.

EPA nonetheless

viewed the horsepower-based measure as appropriate because operators could otherwise shift emissions between controlled and uncontrolled units, thereby evading the limits.

36,746; RTC 123.

Ibid.; see id. at

To further respond to concerns that the 1000-

horsepower threshold would cover some pipeline engines that produce small quantities of emissions, EPA established facility-wide

emissions averaging, “allow[ing] facilities to prioritize emissions reductions from larger, higher-emitting units,” and reducing

the number of engines that must have controls installed.

88 Fed.

Reg. at 36,821.

Pipeline Applicants claim (Appl. 23) that EPA’s concern with

shifting emissions cannot justify its decision because the agency

could instead impose reporting obligations.5 But EPA is authorized

to regulate “any source” and any “emissions activity” that significantly contributes to nonattainment in a downwind State.

U.S.C. 7410(a)(2)(D)(i).

Even where particular pipeline engines

individually emit at lower levels, their aggregate contribution to

The reporting requirements for boilers that the Pipeline

Applicants cite (Appl. 23) are not comparable. The cited provision

applies to a low-use exemption for boilers that operate less than

10% of the year, and it is not equivalent to a 100-tons-per-year

threshold. See 88 Fed. Reg. at 36,819, 36,833.

nonattainment may be significant.

88 Fed. Reg. at 36,680-36,684;

see RTC 109-110. EPA determined that, although the 1000-horsepower

criterion captures more units than the agency had estimated at the

time of the proposal, that threshold still allows for cost-effective

emissions reductions of 32,247 tons of ozone-season nitrogen oxide, at an average cost per ton value of $4921.

RTC 124.

EPA’s

technical determinations are due significant deference, see Marsh

v. Oregon Natural Res. Council, 490 U.S. 360, 377 (1989), and the

Pipeline Applicants have not shown that the agency’s approach was

arbitrary and capricious.

EPA’s compliance timeline adequately accounted for

reliability concerns for natural-gas pipelines

The Rule does not require non-EGU sources to implement reductions until the 2026 ozone season, giving those sources at least

three years before compliance obligations begin.

36,755-,36,757.

88 Fed. Reg. at

Pipeline Applicants nonetheless claim (Appl. 17-

21) that the Rule’s timeline is impossible; that it will threaten

the reliable delivery of natural gas; and that EPA failed to consider those harms.

Contrary to those contentions, EPA thoroughly

considered the compliance schedule, including by commissioning a

report to review the timing needs associated with installing controls for covered non-EGU emissions sources like pipeline engines,

see SC&A, NOx Emission Control Technology Installation Timing for

Non-EGU Sources: Final Report (Mar. 14, 2023) (Timing Report),

https://perma.cc/4HYP-R62J, and by establishing a process for in-

dividual non-EGU sources to seek compliance extensions of up to

three years, 88 Fed. Reg. at 36,759-36,760.

As the Timing Report explains, natural gas compressor stations are located every 50 to 100 miles along a transmission pipeline and use engines to raise the pressure of the gas to help it

flow through the pipeline.

Timing Report 8.

The Timing Report’s

analysis of available data showed that 80% of compressor stations

have more than one unit, about 25% of units operate at less than

40% capacity, and more than 40% of units operate at less than 80%

capacity.

Timing Report ES-8, 8.

For engine controls, “[e]quip-

ment [i]nstallation” is estimated to take a single month, or between three and seven months, depending on the technology used.

Id. at 22, 32.

Those data cast substantial doubt on Pipeline

Applicants’ claim (Appl. 18) that lengthy outages will be required.

To be sure, due to review required by the Federal Energy Regulatory

Commission (FERC), the Timing Report was “not able to complete an

evaluation” of delays that could occur.

Timing Report ES-8.

The

Timing Report noted, however, that the findings regarding relatively low capacity utilization and the “ability to coordinate

outages and work with FERC may not present a substantial basis for

assuming much if any delay in control installation timing.”

Ibid.

In light of the excess capacity and short installation period, EPA

reasonably determined that individual unit outages may be staggered and need not interrupt natural-gas supply.

See 88 Fed. Reg.

at 36,759-36,760; RTC 877.

Pipeline Applicants contend (Appl. 19) that, because “engines

are spread every 40 to 100 miles along the pipeline network,” an

engine immediately ahead of or behind an offline engine may not be

able to substitute for its capacity in periods of “high demand.”

But given the extended compliance timeline and the short installation period, there is no reason that the pipelines should need

to complete the installation during such peak periods.

In cir-

cumstances where unforeseen events would require an operator to

install equipment during peak season in order to comply with the

deadline, and the required installation would threaten the reliability of gas supply, EPA has provided for compliance extensions

of up to three additional years.

88 Fed. Reg. at 36,760.

EPA has

also provided an exemption for emergency engines that could help

to accommodate unforeseen circumstances.

Id. at 36,820-36,821.

Pipeline Applicants thus have identified no sound reason to

believe that the Rule will lead to widespread reliability concerns.

Indeed, EPA modeled the Rule’s requirements on many similar federal

and State requirements applicable to pipeline engines, and applicants have not suggested that those requirements interfered with

gas supply.

See EPA, Final Non-EGU Sectors TSD 5-18 (Mar. 2023),

https://perma.cc/FCR2-F2R6. EPA adequately considered reliability

concerns in adopting the compliance timeline; it simply did “not

agree with all of the

* * *

assertions regarding the time [ap-

plicants] claim is needed for control installation.”

at 36,755.

88 Fed. Reg.

That technical, predictive judgment is reasonable and

should not be disturbed.

See Baltimore Gas & Elec. Co. v. NRDC,

462 U.S. 87, 103 (1983) (noting that a reviewing court is “at its

most deferential” when an agency “is making predictions, within

its area of special expertise”).

The Rule’s provisions that regulate emissions from

cement kilns, the paper industry, and the steel

industry are well-supported

AFPA Applicants make a series of perfunctory allegations

(Appl. 23-24) that EPA relied on flawed assumptions or failed to

provide adequate opportunities to comment when it promulgated the

Rule’s provisions governing emissions from cement kilns, the paper

industry, and the steel industry.

Those arguments lack merit.

EPA recognized that many non-EGU emissions sources, including

cement kilns, already have controls installed or are achieving

reductions at or below the limits the Rule set.

36,827.

88 Fed. Reg. at

The Rule is simply intended to bring all units within

each industry up to a specified level of compliance.

Ibid.

Any

kiln whose existing control technology enables it to meet the

applicable emissions limit need not change anything to comply with

the Rule.

RTC 117.

EPA’s regulation of the paper industry was likewise reasonable.

EPA found that paper-industry boilers will account for an

approximately 1836-ton reduction in nitrogen-oxide emissions --

the highest level of emissions reductions from boilers in any nonEGU industry covered by the Rule.

RTC 121.

As EPA explained in

response to AFPA’s comments, AFPA’s objections rely on commentersubmitted data that could not be verified and reflect misunderstandings of the Screening Assessment and $7500 threshold.

119-121.

RTC

And in response to comments concerning potential imple-

mentation challenges for boilers, EPA exempted low-use boilers and

boilers burning less than 90% fossil-fuel; adopted a formula to

calculate emissions limits for boilers that burn a combination of

fossil-fuel types; and allowed for case-by-case exemptions and

alternative emissions limits.

88 Fed. Reg. at 36,819, 36,833-

36,836, 36,844.

AFPA Applicants’ contentions regarding the steel industry

also miss the mark.6

AFPA Applicants claim (Appl. 24) that EPA

deprived them of an opportunity to comment on emissions limits for

the steel industry.

The proposed rule identified a single emis-

sions limit for steel-industry reheat furnaces.

See 87 Fed. Reg.

20,036, 20,145 & Tbl. VII.C-3 (Apr. 6, 2022).

After considering

industry comments on that aspect of the proposed rule, however,

EPA concluded that the wide variability of performance made a

In a separate order, the D.C. Circuit unanimously rejected

a stay motion specific to the steel industry, which included similar arguments. See United States Steel Corp. v. EPA, No. 23-1207

(D.C. Cir. Oct. 11, 2023), stay application pending, No. 23A___

(filed Oct. 26, 2023).

single limit inappropriate.

88 Fed. Reg. at 36,828.

Based on

industry comments, EPA finalized an approach that requires installation of low-nitrogen-oxide burners (or equivalent technology)

and performance testing to determine an appropriate limit for each

unit that is achievable using that technology. Ibid. The agency’s

incorporation of greater flexibility into the final rule in response to industry comments is a desirable feature of notice-andcomment rulemaking, not an indication that the Rule is arbitrary

and capricious.

The Rule’s enhancements to the power-plant trading

program are reasonable

In regulating power plants, EPA relied on its longstanding

framework for determining appropriate reductions, and the agency

chose to implement those reductions by allowing covered sources to

use a market-based trading program.

Based on EPA’s experience

operating such trading programs, however, EPA included various

enhancements designed to “better sustain over time the incentives

created by the trading program to achieve the degree of emissions

control for [power plants] that the EPA has determined is necessary

to address [S]tates’ good neighbor obligations.”

36,762.

88 Fed. Reg. at

EPA explained that prior trading programs had resulted in

lower stringency over time because the dynamic nature of the industry had hindered EPA’s ability to predict future developments

when the agency set allowance budgets.

In prior trading programs, EPA had established fixed budgets

based on current power-plant fleet composition.

36,764.

88 Fed. Reg. at

When fleet composition changed over the subsequent years,

excessive amounts of emissions allowances accumulated and could be

banked for future years, allowing sources to idle controls despite

EPA’s determinations that such controls were necessary.

36,720-36,724, 36,752-36,753.

Id. at

As EPA noted, that was inconsistent

with its intention in prior rules.

See id. at 36,688.

To prevent

that sequence of events from recurring, the Rule provides for

dynamic budgeting to ensure that the number of allowances matches

the actual composition of the State’s power-plant sources, and for

recalibration of allowance banks to ensure that unused allowances

cannot build up over time to a degree that undermines the elimination of significant contributions to downwind pollution.

Id. at

36,657.

AFPA Applicants object (Appl. 21-22) to those enhancements,

claiming that they result in unlawful over-control of emissions.

That argument reflects a misunderstanding of the nature of the

enhancements.

The enhancements are not designed to reduce emis-

sions beyond States’ significant contribution; they are intended

to ensure that EPA’s trading program is not undercut by changed

conditions in later years.

88 Fed. Reg. at 36,764.

EPA’s over-

control analysis confirmed that the Rule does not compel any State

to reduce emissions to a greater degree than is necessary to eliminate its significant contribution.

Id. at 36,748-36,754.

“[W]hile EPA has a statutory duty to avoid over-control, the

Agency also has a statutory obligation to avoid ‘under-control.’”

EME Homer, 572 U.S. at 523.

EPA determined that the enhancements

in the Rule are an appropriate method of ensuring that the trading

program adheres to both requirements. Applicants have not provided

any reason to second-guess EPA’s judgment or to presume that any

over-control will occur.

And even if (contrary to EPA’s expecta-

tion and intent) the trading-program enhancements were to result

in over-control as they are implemented in later years, the appropriate remedy would be for applicants to “bring a particularized, as-applied challenge,” not to seek “judicial condemnation of

the rule in its entirety.”

II.

Id. at 524.

THE COURT IS UNLIKELY TO GRANT CERTIORARI IF THE D.C. CIRCUIT

UPHOLDS THE RULE

Applicants’ request for extraordinary interim relief should

also be denied because they have failed to show that this Court

would likely grant certiorari if the D.C. Circuit upholds the Rule.

Each of the issues applicants raise is a case-specific question

regarding the application of arbitrary-and-capricious review to

highly complex and technical facts.

See pp. 18-41, supra.

Those

issues are not recurring legal questions of broader importance

that would warrant this Court’s review.

Applicants barely even attempt to show otherwise.

State Ap-

plicants fail to address the issue altogether. Pipeline Applicants

include a footnote claiming (Appl. 10-11 n.7) they do not need to

address the likelihood of certiorari.

But see Mills, 142 S. Ct.

at 18 (Barrett, J., concurring in the denial of application for

injunctive relief) (noting that whether to grant extraordinary

relief includes consideration of “whether the Court should grant

review in the case”).

They then tack on a single conclusory

sentence claiming that they satisfy the standard because of “the

importance of the issues and the Rule’s significant legal flaws.”

Appl. 11 n.7.

As already explained, however, applicants have

failed to show any such legal flaws, and the issues applicants

press are highly technical and case-specific.

AFPA Applicants’ arguments (Appl. 13) on this score are similarly sparse.

They focus on the costs of the Rule and note that

the Court has granted petitions for certiorari in other CAA cases.

Ibid. (citing West Virginia v. EPA, 142 S. Ct. 2587 (2022); Michigan v. EPA, 576 U.S. 743 (2015); EME Homer, 572 U.S. at 506).

But in each of those cases -- unlike this one -- the Court was

presented with important questions about the interpretation of

certain provisions of the CAA, not just fact-dependent arbitraryand-capricious challenges like those present here.

See West Vir-

ginia, 142 S. Ct. at 2610 (considering EPA’s interpretation of 42

U.S.C. 7411(d)); Michigan, 576 U.S. at 750 (considering EPA’s interpretation of 42 U.S.C. 7412); EME Homer, 572 U.S. at 509 (considering EPA’s interpretation of 42 U.S.C. 7410).

And although

AFPA Applicants attempt to characterize the Rule as an “unprece-

dented

abrogation

the

congressionally

granted

rights

States,” Appl. 13, the Rule is in fact an exercise of authority

Congress expressly granted to EPA to regulate sources of pollution

in States when those States fail to do so in the first instance.

Accordingly, none of the applicants has shown that the issues in

this case warrant the Court’s discretionary review, and that “counsels against a grant of extraordinary relief in this case.” Mills,

142 S. Ct. at 18 (Barrett, J., concurring in the denial of application for injunctive relief).

III. THE REMAINING EQUITABLE FACTORS WEIGH HEAVILY AGAINST INJUNCTIVE RELIEF

Applicants’ request to enjoin the Rule should be re-

jected for the additional reason that they have not demonstrated

irreparable harm.

To satisfy that requirement, applicants must do

more than “simply show[] some ‘possibility of irreparable injury.’”

Nken, 556 U.S. at 434 (citation omitted); see Winter v.

NRDC, Inc., 555 U.S. 7, 22 (2008).

They have not done so here.

Applicants contend (Pipeline Appl. 27-28; AFPA Appl. 25-27)

that they will be required to spend hundreds of millions of dollars

in compliance costs in the months following the Rule’s effective

date.

But non-EGU sources do not need to meet emissions deadlines

until May 2026 at the earliest, with the potential for compliance

extensions of up to three additional years.

36,755-36,760.

See 88 Fed. Reg. at

In view of the expected timeline for installation,

EPA concluded that “the controls for non-EGU sources needed to

comply with this final rule are generally not expected to be installed significantly before the 2026 ozone season.”

36,759.

Id. at

The Timing Report similarly suggests that applicants

should be able to avoid significant expenditures pending judicial

review. The report estimates that retrofitting of pipeline engines

would generally take between three and six months of design, analysis, and permitting before installation begins.

25.7

Timing Report

And the entire retrofitting process is estimated to take

between six and 19 months (depending on the type of technology

used), inclusive of fabrication and installation.

Id. at 25, 32.

To the extent applicants may undertake some initial design and

planning during judicial review, their potential costs are likely

to be minimal.

A similar analysis applies to power plants’ compliance costs.

For the 2023 through 2025 ozone seasons, the Rule contemplates

familiar control strategies that are not meaningfully different

from those included in EPA’s two most recent Good Neighbor rules.

App., infra, 25a.

2026.

More stringent budgets do not phase in until

And “the preliminary analysis and engineering steps” re-

quired for the relevant control strategies “involve no capital

The Timing Report notes that supply-chain delays could

impact timing, but that those disruptions are easing.

Timing

Report 50-54; 88 Fed. Reg. at 36,759-36,760. In any event, the

availability of compliance extensions adequately addresses concerns with shortages and with the need for specialized labor.

Timing Report 59-60.

costs.”

Id. at 33a.

For power plants that choose to comply with

emissions budgets by installing technology, “much of the first

year of project work typically need not entail substantial capital

outlays.”

Id. at 34a.

Applicants have thus failed to demonstrate

that they will incur substantial compliance costs pending judicial

review, which can proceed expeditiously in the D.C. Circuit.

Pipeline Applicants attempt to bolster (Appl. 24-27) their

claims of financial injury by alleging that the Rule will harm

natural-gas reliability.

Those arguments do not show irreparable

harm for the same reason they fail on the merits:

the Rule’s

compliance timeline does not threaten reliability.

See pp. 34-

37, supra. The record reveals that the vast majority of compressor

stations contain multiple units and that those units have excess

capacity, indicating that pipeline operators can manage unit outages for pollution-control upgrades without endangering service.

Timing Report 8.

Applicants’ contrary claims depend on inflated

estimates of installation time (Pipeline Appl. 24) and unwarranted

assertions that pipelines will be required to complete the installations during peak periods (id. at 25-26).

Applicants also minimize the significance of the Rule’s compliance flexibilities (Pipeline Appl. 26-27), but those provisions

cannot be so easily disregarded.

The availability of emissions

averaging “means that of the approximately 3,000 engines subject

to the [Rule]’s applicability criteria for pipeline engines, less

than one-third (or about 900) are estimated to need to improve

emissions performance to achieve full compliance.”

53a.

App., infra,

Applicants dispute that conclusion and criticize EPA’s data

sample.

Pipeline Appl. 26.

But EPA relied on data from a sta-

tistically significant number of facilities across the affected

States.

EPA,

Final

Non-EGU

Sectors

TSD

(Mar.

2023),

https://perma.cc/FCR2-F2R6.

EPA’s use of that sample was reasonable, particularly because

many facilities are not required to submit annual emissions inventories.

See National Ass’n for Surface Finishing v. EPA, 795

F.3d 1, 12 (D.C. Cir. 2015).

Courts “generally defer to an

agency’s decision to proceed on the basis of imperfect scientific

information, rather than to invest the resources to conduct the

perfect study.”

Ibid. (citation omitted).

And if unforeseen

circumstances arise -- like increased demand due to anomalous

weather conditions during a scheduled outage -- the Rule authorizes

compliance extensions of up to three years.

App., infra, 52a.

The exemption for emergency engines could also serve to avoid

reliability concerns.

Id. at 53a.

Applicants thus have shown no

more than a “possibility” of irreparable injury, which is an insufficient basis for the injunctive relief they seek.

Nken, 556

U.S. at 434 (citation omitted).

State Applicants’ assertions (Appl. 24-25) of irreparable

harm are no more persuasive. Like the other applicants, the States

identify potential compliance costs as a form of irreparable harm.

Appl. 24-25.

But the Rule imposes requirements only on covered

sources, not on the States.

The States cite the burden of pro-

cessing permit applications and ensuring compliance (the latter of

which is at the States’ discretion), but executing traditional

permitting functions within their regular duties is not an irreparable injury.

Treating such routine costs as irreparable injury

would be “inconsistent with [the] characterization of [equitable]

relief as an extraordinary remedy.”

Winter, 555 U.S. at 22.

State Applicants also assert (Appl. 25-26) that the Rule will

cause electricity-grid destabilization, with accompanying economic

impacts.

But State Applicants provide no more than speculative

allegations that some power plants may opt to retire at some future

point.

EPA analyzed that possibility and concluded that the Rule

would not degrade electric-system reliability because neither the

power-plant emissions reductions nor the trading program requires

that any power plant retire.

See 88 Fed. Reg. at 36,770-36,775;

see also App., infra, 27a, 29a-31a, 43a.

And in the event that

some power plants decide to retire, they must comply with the

procedures established by the relevant Regional Transmission Organization, which is charged with maintaining grid reliability.

88 Fed. Reg. at 36,771.

The States likewise cannot show irreparable harm by claiming

(Appl. 26) an intrusion on their “sovereign authority to regulate

air quality within their borders.”

Even assuming that abstract

interest could give rise to a cognizable injury, the federal government has a weighty countervailing sovereign interest in enforcing the Rule -- and “[t]he Federal Government holds a decided

advantage in this delicate balance:

the Supremacy Clause.”

ory v. Ashcroft, 501 U.S. 452, 460 (1991).

Greg-

Under the CAA, each

State has an opportunity to regulate emissions under an appropriate

state implementation plan that accounts for the harm that emissions

from within its borders impose on downwind States.

When a State

fails to submit an adequate plan, the Act requires EPA to step in

to balance the interests of upwind and downwind States.

Because

EPA promulgated the Rule pursuant to that congressional command,

and because the Rule currently applies only in States that did not

challenge EPA’s disapprovals of their plans, the State applicants

cannot rely on abstract conceptions of sovereignty to justify the

extraordinary relief they seek.

Any injury that applicants have demonstrated cannot out-

weigh the injuries to the government and the public interest -which merge in this context, see Nken, 556 U.S. at 435 -- that a

stay of the Rule would entail.

Most fundamentally, emissions

reductions under the Rule will provide significant benefits to the

residents of downwind States.

A delay in the implementation of

the Rule would eliminate the incentive to improve emissions performance in the short-term -- an incentive that has already re-

sulted in a “substantial reduction in emissions,” with some sources

improving their emissions by more than 75%.

44a.

App., infra, 35a,

And a stay would likely delay the phase-in of more signifi-

cant reductions for both power plants and non-EGU emissions sources

that are slated to begin in 2026.

Stays of two prior rules im-

plementing the Good Neighbor Provision led to implementation delays of up to three years, even though the rules were later largely

upheld.

Id. at 21a-22a; see Michigan, 213 F.3d at 695; EME Homer,

795 F.3d at 132.

A stay here could similarly delay elimination of

upwind States’ significant contributions until at least 2029.

App., infra, 21a-22a.

During that delay, downwind States would suffer significant

harms.

The emissions that contribute to cross-state air pollution

represent a public health hazard in downwind communities, associated with worsened asthma and increased mortality.

at 36,671.

88 Fed. Reg.

In addition, those emissions generate economic harm in

downwind States as areas in violation of ozone standards can face

increasingly stringent regulatory burdens mandated by the CAA to

ensure those States attain the standards.

See 42 U.S.C. 7511a;

see also App., infra, 46a-47a.

Applicants suggest that EPA’s timing in disapproving the various state plans and promulgating the Rule demonstrates a lack of

any urgency in implementing the Rule’s requirements.

Appl. 29; State Appl. 27.

See Pipeline

But the delay in promulgating the Rule

arose in part because of litigation on the preceding rule implementing the Good Neighbor provision.

And the D.C. Circuit has

held that the Act requires elimination of upwind emissions in time

for the next downwind attainment deadlines.

at 318-319.

Wisconsin, 938 F.3d

Here, that requires all feasible reductions by May

2026 at the latest, making the Rule’s compliance deadlines consistent with the Act’s requirements.

Fed. Reg. at 36,755-36,756.

See 42 U.S.C. 7410(c)(1), 88

In any event, downwind States and

their residents cannot fairly be punished for any delay in EPA’s

promulgation of the Rule.

In light of the significant public

benefits the Rule provides and the significant public harms a stay

would impose, the balance of equities strongly disfavors the extraordinary relief applicants seek.

AFPA Applicants seek relief that is greatly dispropor-

tionate to the vast majority of the errors they allege.

Those

applicants focus on particular aspects of the Rule that apply to

particular industries, yet they request that the whole Rule be

stayed.

See AFPA Appl. 29.

To the extent those arguments have

any merit, they would not justify staying the Rule in its entirety.

Cf. 88 Fed. Reg. at 36,693 (noting that the Rule “promulgates

discrete emissions control requirements for the power sector and

for each of [nine] other industries,” and that “[s]hould any industry-specific aspect of the final rule be found invalid, the EPA

views this rule as severable as between the different industries

and different types of emissions control requirements”).

Pipeline

Applicants, by contrast, ask only that the Court stay “the Rule’s

provisions for pipeline engines.”

Appl. 29.

If the Court concludes relief is warranted with respect to

any discrete aspects of the Rule, it should tailor the relief

instead of granting a sweeping stay that would more broadly disrupt

the Rule’s protection of downwind States and their residents.

But

because applicants cannot satisfy the standards for extraordinary

relief, the better course is to deny the applications in full.

CONCLUSION

The applications should be denied.

Respectfully submitted.

ELIZABETH B. PRELOGAR

Solicitor General

OCTOBER 2023

APPENDIX

Declaration of Joseph Goffman (Oct. 28, 2023)..................1a

Declaration of Rona Birnbaum (Aug. 17, 2023)..................15a

Declaration of Scott Mathias (Aug. 11, 2023)..................48a

IN THE SUPREME COURT OF THE UNITED STATES

____________________________________

)

STATE OF OHIO, et al.,

)

KINDER MORGAN, INC., et al.,

)

AMERICAN FOREST & PAPER

)

ASSOCIATION, et al.

)

Applicants

)

Nos. 23A349, 23A350, 23A351

)

UNITED STATES ENVIRONMENTAL )

PROTECTION AGENCY, et al.

)

____________________________________)

DECLARATION OF JOSEPH GOFFMAN

I, Joseph Goffman, under penalty of perjury, affirm and declare that the following

statements are true and correct to the best of my knowledge and belief, and are based on my own

personal knowledge or on information contained in the records of the United States

Environmental Protection Agency (EPA) or supplied to me by EPA employees under my

supervision.

I am Principal Deputy Assistant Administrator performing delegated duties of

Assistant Administrator for the United States Environmental Protection Agency Office of Air

and Radiation (OAR), which is located at 1200 Pennsylvania Avenue, NW, Washington, D.C.

20460.

OAR is the EPA headquarters-based unit with primary responsibility for

administration of the Clean Air Act (CAA or the Act). As the Principal Deputy Assistant

Administrator performing delegated duties of Assistant Administrator for OAR, I serve as the

principal advisor to the Administrator of EPA on matters pertaining to air and radiation

programs, and I am responsible for managing these programs, including program policy

(1a)

development and evaluation; development of emissions standards; program policy guidance and

overview; and technical support and evaluation of regional air and radiation program activities.

As part of my duties as Principal Deputy Assistant Administrator performing

delegated duties of Assistant Administrator of OAR, I oversee the development and

implementation of actions, regulations, policy, and guidance associated with the review and

establishment of National Ambient Air Quality Standards (NAAQS) under sections 108 and 109

of the CAA, 42 U.S.C. §§ 7408 and 7409, and implementation of the NAAQS under section 110

of the CAA, 42 U.S.C. § 7410, including the Good Neighbor Provision, 42 U.S.C. §

7410(a)(2)(D)(i)(I), and relevant other parts of Title I of the Act, including Part D (plan

requirements for nonattainment areas).

This declaration is filed in support of EPA’s opposition to applications for a stay

of the “Federal ‘Good Neighbor Plan’ for the 2015 Ozone National Ambient Air Quality

Standards,” 88 Fed. Reg. 36654 (June 5, 2023) (the Good Neighbor Plan or Plan), filed before

the U.S. Supreme Court in applications 23A349, 23A350, and 23A351.

The purpose of this declaration is to provide the Court information pertaining to

the regulatory design of the Good Neighbor Plan, to explain why it is EPA’s view that the Plan

can and should remain in effect for each of the 11 states that it currently covers, notwithstanding

temporary administrative stays of the Plan for 12 other states. As I will explain, the Plan is

designed to eliminate each covered state’s “significant contribution,” using the same analytical

framework the Supreme Court upheld in EME Homer City Generation, L.P. v. EPA, 572 U.S.

489 (2014). The Plan can be implemented in each state to accomplish this statutory objective as

to each state, irrespective of the total number of states the Plan covers.

EPA issued stays of the Plan for 12 states through two interim final rules. 88 Fed.

Reg. 49295 (July 31, 2023) (First IFR); 88 Fed. Reg. 67102 (Sept. 29, 2023) (Second IFR). The

IFRs were issued to ensure EPA’s compliance with preliminary stay orders several regional

circuit courts issued pending judicial review of a separate EPA action disapproving 21 states’

state implementation plan (SIP) submissions addressing Good Neighbor obligations for the 2015

ozone NAAQS, 88 Fed. Reg. 9336 (Feb. 12, 2023) (the Disapproval). Because the Disapproval is

a predicate to EPA’s authority to issue the Good Neighbor Plan for most states, EPA had no

choice but to stay the effectiveness of the Plan as to those states once the preliminary stay orders

issued.1

The stay of the Plan as to some states has no bearing on the lawfulness and

appropriateness of its application in other states. The Plan is comprised of a series of 23 federal

implementation plans (FIPs), promulgated for 23 states. Depending on the analysis of each

state’s contribution to downwind ozone problems and cost-effective emissions control

opportunities (using EPA’s nationwide “4-step interstate transport” analytical framework), the

Plan subjects the states to certain emissions control programs for electric generating units

(EGUs) and for certain affected units in other industries (often referred to in this context as nonEGUs).

Neither the legal basis for the Plan, nor its design and implementation, preclude it

from being carried out in only a subset of the states that were originally covered. For the 11

In the case of Utah, EPA had separate authority to issue the FIP through a predicate “finding of

failure to submit” issued in 2019. See 88 Fed. Reg. at 36689. The Tenth Circuit in staying the

Disapproval concluded that its stay order as to the Disapproval blocked the Good Neighbor Plan

from taking effect for Utah. State of Utah v. EPA, 23-9509 (10th Cir.), ECF No. 11016742. EPA

has complied with the Tenth Circuit order by staying the Plan as to Utah. However, it has raised

this issue to the Tenth Circuit in its merits brief. Id. ECF No. 010110917156, at 82-83.

states that remain covered, EPA's determination that the emissions control requirements in the

Good Neighbor Plan are necessary to eliminate their “significant contribution” remains

unchanged.2 The implementation and rationale of the Good Neighbor Plan do not depend on the

specific number of states that it covers.

For this reason, EPA found in the Good Neighbor Plan that the Plan is severable

by state. 88 Fed. Reg. at 36693. While the analytical methods, policy judgments, and technical

analyses that informed the Plan are conducted at a national scale, the actual definition of

significant contribution is determined at the state level, and the implementation of the measures

necessary to eliminate significant contribution is fully achievable by the sources within each

state, irrespective of other states’ participation.

10.

The Plan determines on a state-by-state basis which of the EGU (i.e., power plant)

and non-EGU emissions-control programs should apply. See 40 C.F.R. § 52.38(b)(2) (as

amended by 88 Fed. Reg. at 36862-63) (identifying states subject to the Good Neighbor Plan’s

“Group 3” EGU emissions trading program promulgated at 40 C.F.R. Pt. 97, subpart GGGGG);

40 C.F.R. § 52.40(c)(2) (as promulgated at 88 Fed. Reg. at 36869) (identifying states subject to

non-EGU emissions control requirements promulgated at id. §§ 52.41-46). The regulations at 40

C.F.R. Pt. 97, subpart GGGGG, and 40 C.F.R. §§ 52.41-46 are uniform in nature. But states are

“enrolled” into these requirements based on state-specific findings regarding the level of their

“Significant contribution” is often used as a shorthand to refer to the identification of those

amounts of emissions that significantly contribute to nonattainment or interfere with

maintenance of the NAAQS in other states and therefore must be prohibited under the Good

Neighbor Provision. See 42 U.S.C. § 7410(a)(2)(D)(i)(I).

contribution to other states’ ozone problems and how long that contribution is projected to

continue into the future.3

11.

In other words, it is through the application of those uniform programs, as

appropriate, in each state, that the Good Neighbor Plan eliminates each covered state’s

significant contribution, as required by 42 U.S.C. § 7410(a)(2)(D)(i)(I).

12.

The state-specific coverage of the Plan (at the time it was promulgated on March

15, 2023), by regulatory program, is as follows:

EGUs in all covered states except California (22 states total) are required

to participate in the Group 3 EGU emissions trading program at the level of stringency

associated with near term emissions-control strategies that EPA found can be

implemented in 2023 and 2024.

EGUs in Alabama, Minnesota, and Wisconsin are only subject to this

“near-term” stringency level within the Group 3 Trading Program, and no more, because

EPA found these states are no longer linked to downwind ozone problems in the year

2026.

EGUs in 19 states (excluding Alabama, Minnesota, and Wisconsin) that

are covered by the Group 3 trading program, are subject to the enhanced stringency in the

budgets that takes effect over 2026 and 2027 because these states are linked through the

2026 analytic year.

This is identical in structure to how EPA has promulgated Good Neighbor federal requirements

through multiple prior rulemakings. See 40 CFR § 52.38-39 (identifying the enrollment of states

into emissions trading programs for ozone season NOX, annual NOX, and annual sulfur dioxide

promulgated as subparts to 40 C.F.R. Pt. 97, as necessary to address Good Neighbor obligations

for other ozone and particulate matter NAAQS).

EPA found California has no cost-effective fossil-fuel fired EGU

emissions reductions available at the stringency levels determined in the Good Neighbor

Plan and so is not subject to the Group 3 Trading Program at all.

Non-EGUs in 20 states are subject to the uniform emissions control

regulations for non-EGUs. Because EPA found these requirements may take up to three

years to be implemented (i.e., until 2026), this number excludes Alabama, Minnesota,

and Wisconsin, for the same reason as above: these states are not “linked” in 2026.

13.

These state groupings illustrate how the application of each set of regulatory

requirements promulgated in the Plan depend on the circumstances of each state, as determined

through the application of the nationwide 4-step analytical framework. In no case are all of the

Plan’s EGU and non-EGU control programs applicable in all 23 states.

14.

The Good Neighbor Plan was never premised on an assumption that it must be

applicable in specifically 23 states.4 As further illustration of this fact, the Good Neighbor Plan,

like all prior Good Neighbor federal rulemakings before it, recognizes that states may choose to

replace their FIP with a SIP. See, e.g., 88 Fed. Reg. at 36838-42. In doing so, states may opt to

leave the interstate trading program for EGUs in favor of an adequate, alternative approach to

addressing their Good Neighbor obligations. Id. at 36841-42; see also, e.g., Cross-State Air

Pollution Rule, 76 Fed. Reg. 48208, 48328 (Aug. 8, 2011) (CSAPR).

15.

The emissions control requirements are not in any way dependent on a minimum

number of states’ enrollment in the Plan. This is not just a function of the regulatory structure of

As EPA acknowledged in the Plan, there are several additional states that may have Good

Neighbor obligations, which EPA is still in the process of addressing. See 88 Fed. Reg. at 36658.

However, there is no evident reason why the Good Neighbor Plan should be considered improper

or un-implementable because those states are not currently included in it.

the Plan as described above. It is because, as a factual matter, EPA’s determinations as to the

level of emissions reductions that are appropriate to eliminate significant contribution (even

though evaluated on a national scale) do not depend on the application of the Plan’s requirements

in multiple states. The Plan is not premised on accomplishing a minimum total of emissions

reductions but rather in holding the sources in each linked upwind state to minimum levels of

emissions performance deemed to be cost-effective. 88 Fed. Reg. at 36741. In establishing what

that level of stringency should be, the Plan identified conventional, at-the-source, NOX emissions

control technologies that have been available in the covered industries for many years. See, e.g.,

88 Fed. Reg. 36738 (identifying control technologies for EGUs); id. at 36739 (identifying control

technologies for non-EGUs). The feasibility of these control technologies is not in any way

dependent on the inclusion of a minimum number of states in the Rule.

16.

This is true even in the case of the interstate trading program for EGUs. Interstate

trading for EGUs is a feature of this rule as with prior Good Neighbor rules like CSAPR.

Interstate trading can help make compliance more efficient, but even before the Good Neighbor

Plan, EPA took measures to ensure that interstate trading does not undermine the obligation to

eliminate each state’s significant contribution. See North Carolina, 531 F.3d 896, 921 (D.C. Cir.

2008), modified on reh’g, 550 F.3d 1176. See, e.g., 76 Fed. Reg. at 48268-71; 88 Fed. Reg. at

36752-53.

17.

While interstate trading would generally increase the size of the allowance trading

market and thus may increase market liquidity in ways that can improve market efficiency, there

is no reason that the program cannot be implemented on a state-by-state level.5 Indeed, each

In fact, the size of the trading region is not the only determinant of liquidity; the relative

demand for allowances is an important factor. For example, sources that are not well-controlled

state’s budget is set in the Plan at levels that provide sufficient allowances for each state

assuming EGUs achieve a level of reduction equivalent to what can be achieved by the at-thesource technologies identified to eliminate significant contribution.

18.

Trading regions have always varied in size over the history of implementation of

the Good Neighbor provision. This has never posed a challenge to compliance feasibility, nor

does EPA have any evidence of allowance shortages occurring in any of these programs. For

example:

Currently, Georgia is the only state whose EGUs remain in the original

CSAPR “Group 1” ozone season NOX trading program.

In 2021, the Revised CSAPR Update created a 12-state trading region to

complete the remedy to significant contribution for the 2008 ozone NAAQS (ie, the

original “Group 3” program).

With the Revised CSAPR Update in place, the 2016 CSAPR Update

“Group 2” program trading region was reduced from 22 states to 10 states.

Currently, with the stay of the Good Neighbor Plan as to 12 states, EGUs

in three states (Kentucky, Louisiana, and West Virginia) are in a temporary trading

program (“Expanded Group 2”) to maintain status quo regulatory requirements for these

for NOX would tend to put upward pressure on allowance prices (and potentially reduce

liquidity). If such sources are removed from the Group 3 trading program, for example due to

judicial stays as to the states in which they are located, this may put downward pressure on

allowance prices (and potentially increase liquidity). See paragraph 22 below on current Group 3

allowance prices.

EGUs under the Revised CSAPR Update during the pendency of litigation over the

Disapproval.6

19.

Illustrating the long history of successful implementation of emissions trading

programs, allowance prices have tended to decline substantially over time as emissions

reductions are implemented and a bank of unused allowances builds up. See 88 Fed. Reg. at

36687 (discussing experience in prior programs).7

20.

So far, the experience with the Good Neighbor Plan has been no different—even

with the stays in place for 12 states. Allowance prices have been declining substantially since

EPA promulgated the Plan in March 2023. The current price of a Group 3 allowance now stands

at less than $2000/ton (reflecting a drop of 90% from where Group 3 allowance prices were a

year ago and reflecting a continuing decline in allowance prices over 2023, despite reductions in

the number of states covered by the program resulting from the stays). This continuing price drop

illustrates that the Plan remains achievable within the current 10-state trading region, and there is

no shortage of allowances available for compliance.

21.

Finally, the non-EGU emissions control strategies do not entail interstate trading.

These control programs are based on a regulatory structure where emissions limitations and

associated compliance assurance requirements apply directly to individual covered emissions

The emissions and allowance-availability data indicate that there will not be compliance

challenges for this group. Their combined EGU ozone season NOX emissions were 40,648 tons

in 2021, and 35,403 tons in 2022. Their combined budget in 2023 and each subsequent year (so

long as they remain in the “Expanded Group 2” program) is 41,753 tons. Taking into account

already-banked allowances, they will have a total of 61,011 allowances available for compliance

for the 2023 ozone season. Assuming their 2023 emissions are similar to 2022 emissions, they

will therefore carry over a substantial bank of allowances for use in 2024 and later years.

For example, according to S&P Global Market Intelligence, a subscription-based reporting

service, allowances in each of the original CSAPR annual SO2 and annual NOX trading programs

are currently trading between $2.00 and $3.00 per ton.

10a

units. These requirements are unaffected by the number of states that are covered by the Good

Neighbor Plan. And as with the power plant control strategies, these limits are based on at-thesource emissions-control technologies that are well-demonstrated and in fact mandated in many

downwind states with ozone nonattainment areas.

22.

The stays of the SIP Disapproval that have been entered constitute preliminary

rulings that do not bind the merits panels, and have been issued on a temporary basis, only for

the purpose of preserving the status quo pending judicial review. EPA has argued that once these

cases are briefed on the merits, the regional circuit courts should transfer those cases to the D.C.

Circuit or dismiss them on the basis that venue is improper in those courts, uphold EPA on the

merits of the Disapproval, or remand without vacatur even if some error is identified. In any of

these scenarios, the preliminary stays will be lifted, allowing EPA to bring the Good Neighbor

Plan into effect for these states.

23.

Should any court vacate the Disapproval as to any state upon adjudication of the

merits, EPA anticipates that its course of action in that circumstance would be dependent on the

grounds of that court’s ruling, including evaluation in consultation with the Department of

Justice whether there are grounds for rehearing or appeal.

24.

At this time, EPA has not identified any policy or legal justification that would

warrant an agency action staying the Good Neighbor Plan as to the 11 states where it is presently

in effect. For these states, the Disapproval or findings of failure to submit remain unchallenged

and effective and thus EPA has the authority and obligation to promulgate FIPs for these states.8

In the case of Pennsylvania and Virginia, EPA’s FIP authority stems from a 2019 finding of

failure to submit. See 88 Fed. Reg. at 36689.

11a

25.

By contrast, the First and Second IFRs staying the Good Neighbor Plan were done

to comply with the regional circuits’ preliminary stay orders. EPA cannot implement a FIP for

any state for which it lacks predicate authority through disapproval of a SIP or a finding of

failure to submit a complete SIP. See 42 U.S.C. § 7410(c)(1). Thus, so long as the preliminary

court orders are in place, EPA must abide by such orders and did so by staying the effectiveness

of the Good Neighbor Plan FIPs for those states until such time as its authority to implement

those FIPs is restored. 88 Fed. Reg. 49295, 49297 (July 31, 2023); 88 Fed. Reg. 67102, 67103

(Sept. 29, 2023).

26.

Because the judicial stays were of the underlying Disapproval, not the FIP, EPA

determined that to comply with the stay orders and to provide regulatory certainty to relevant

sources, the proper course was to promulgate the interim final rules (IFRs) that administratively

stay the GNP for the 12 states covered by judicial stays.

27.

EPA was clear in issuing these IFRs that they entail no exercise of agency

discretion, but rather are necessary to comply with the preliminary stay orders and to preserve

status quo regulatory requirements pending judicial review of the Disapproval. See 88 Fed. Reg.

at 49299.

28.

Because these IFRs are necessary to comply with court orders, were not an

exercise of agency discretion, and were important to implement quickly to provide regulatory

certainty, EPA found good cause to issue the IFRs without prior opportunity for comment on

grounds that comment is both unnecessary and impractical in this circumstance. Id. at 49299300. Nonetheless, recognizing that there was some potential that in executing these stays, some

flaw or unintended consequence might result, EPA provided for 30-day comment periods on

both IFRs in its discretion. See id. at 49300.

12a

29.

So far, EPA has reviewed comments on the First IFR and has identified no

grounds on which its IFRs were issued in error, should be modified, extended to include other

states, or limited to exclude any states. No comments identified any technical flaws or mistakes

in the regulatory changes it made to preserve the status quo. The discretionary comment period

on the Second IFR closes on October 30, 2023. 88 Fed. Reg. at 67102.

30.

There is no basis to extend an administrative stay of the Good Neighbor Plan to

any state not currently subject to a judicial stay of the underlying SIP disapproval. As to each of

the 11 states where the Plan is currently in effect, the requirements of the Act as interpreted

through relevant case law make clear that there are no legal grounds not to continue the Plan in

effect for these states.9

31.

EPA is obligated to address each state’s significant contribution. North Carolina,

531 F.3d at 921. EPA is obligated to do so consistent with the attainment schedule faced by

downwind areas, i.e., for ozone, pursuant to 42 U.S.C. § 7511(a), as expeditiously as practicable

and no later than the next attainment date. Id. at 911; Wisconsin, 938 F.3d 303, 313-20. EPA may

deviate from this mandate only upon a sufficient showing of necessity, taking into consideration

the ultimate objective of timely attainment of the NAAQS in downwind areas. Id. at 320.

32.

An agency action staying the Good Neighbor Plan as to any of the 11 states where

it is currently in effect would not be consistent with this legal framework or sound air quality

planning. First, this would unnecessarily cause and/or extend ongoing harm to air quality and

public health and welfare resulting from these 11 states’ emissions. In the near term,

EPA also lacks authority to issue a stay of this action under 42 U.S.C. § 7607(d)(7)(B), which

authorizes no more than a 3-month stay in any case, and such an action must be predicated on

commencement of a mandatory reconsideration process, which EPA has not done.

13a

improvements in EGUs’ emissions performance that were achieved in the 2023 ozone season

and can continue to be achieved in 2024 and beyond would be stalled. Further, the additional

emissions reductions required to eliminate these states’ significant contribution beginning in

2026 likely would be delayed, possibly by years. The deferral of emissions reductions from these

states would leave downwind states that face increasing regulatory burdens associated with

continuing nonattainment with no relief from these upwind states’ sources. Maryland v. EPA,

958 F.3d 1185, 1200-04 (D.C. Cir. 2020).

33.

The Plan follows the same approach to defining significant contribution as EPA

applied in CSAPR, which the Supreme Court upheld in EME Homer City Generation, L.P. v.

EPA, 572 U.S. 489 (2014). Within that framework, which the Court found to be an “equitable”

and “efficient” solution to the “thorny” causation problem of interstate ozone pollution, the Plan

“requires the most impactful sources in each state . . . to come up to minimum standards of

environmental performance based on demonstrated NOX pollution-control technology.” 88 Fed.

Reg. at 36741 (citing 572 U.S. at 519).

34.

Covered sources in each state that remains subject to the Plan can still meet that

requirement. While it is true that the stays of the Plan for 12 states mean that the Plan’s air

quality benefits for downwind areas will only be partially realized for the time being, this does

not imply grounds for staying the Plan as to 11 more states with valid disapprovals or findings of

failure to submit. The Act requires timely elimination of each state’s significant contribution.

35.

Consistent with its understanding of its statutory authority and duty, EPA will

continue to implement the Good Neighbor Plan in all states where it has the authority to do so.

To the extent that its authority is currently unaffected by court decisions, EPA will proceed with

14a

implementation of the Good Neighbor Plan and take other actions as needed to eliminate

significant contribution for purposes of the 2015 ozone NAAQS.

I declare under penalty of perjury that the foregoing is true and correct.

Executed this 28th day of October, 2023.

_________________________________

Joseph Goffman

Principal Deputy Assistant Administrator

performing the delegated duties of Assistant

Administrator

Office of Air and Radiation

United States Environmental Protection Agency

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 2 of 42

15a

UNITED STATES COURT OF APPEALS

FOR THE DISTRICT OF COLUMBIA CIRCUIT

____________________________________

)

STATE OF UTAH, et al.,

)

Petitioners,

)

No. 23-1157 (and consolidated

)

cases)

UNITED STATES ENVIRONMENTAL )

PROTECTION AGENCY, et al.,

)

Respondents.

)

____________________________________)

DECLARATION OF RONA BIRNBAUM

I, Rona Birnbaum, affirm and declare that the following statements

are true and correct to the best of my knowledge and belief and that they are based

upon my personal knowledge, or on information contained in the records of the

United States Environmental Protection Agency (“EPA” or the “Agency”), or on

information supplied to me by EPA employees.

I am the Director of the Clean Air Markets Division in the Office of

Atmospheric Protection within the Office of Air and Radiation at EPA. The Clean

Air Markets Division, which was initially created to implement the acid rain

provisions of the Clean Air Act Amendments of 1990, designs and operates

market-based programs to reduce emissions of sulfur dioxide (“SO 2”) and nitrogen

oxides (“NOX”), generates and provides public access to power plant emissions

data, facilitates and oversees emissions monitoring and reporting, assesses

emissions control technology options, conducts atmospheric deposition monitoring

and analysis, develops information systems for market-based programs, assesses

environmental and human health effects, assesses benefits and costs of programs,

and educates the public regarding regional air pollution problems and marketbased programs. The currently operated market-based programs were established

under the Acid Rain Program, the Cross-State Air Pollution Rule (“CSAPR”), the

CSAPR Update, and the Revised CSAPR Update.

(Page 75 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 3 of 42

16a

In my current capacity as Director of the Clean Air Markets Division,

I oversee EPA’s implementation of components of the Clean Air Act including

Title IV (acid deposition control) and parts of Title I (air quality standards and

associated emission limitations). In coordination with other EPA offices, I manage

the promulgation and implementation of regulations pursuant to the Clean Air Act

including the suite of CSAPR programs. I manage all of the Clean Air Markets

Division’s activities as listed in paragraph 2, including overseeing EPA’s

collection of emissions data from the power sector (and some other stationary

emissions sources) under the Acid Rain Program and the suite of CSAPR

programs.

Prior to becoming Director of the Clean Air Markets Division in 2022,

I held several management positions in the Office of Atmospheric Protection

including in the early years of the Acid Rain Program. I joined EPA in 1988 and

the Office of Atmospheric Protection in 1991. I hold a bachelor’s and master’s

degree in environmental and natural resource policy from The George Washington

University.

The purpose of this declaration is to provide information responsive to

certain allegations made in several parties’ (“Movants”) Motions for Stay filed by

August 4, 2023 respecting the Good Neighbor Plan’s requirements for power

plants (“electric generating units” or “EGUs”). In addition, Section V concerns the

regulatory, public health, and economic consequences for downwind

nonattainment areas if the Good Neighbor Plan is stayed. Unless otherwise noted,

information and data presented in this declaration regarding the Good Neighbor

Plan reflect the rule as it was signed on March 15, 2023. Section IV addresses the

continuing feasibility of the Good Neighbor Plan’s requirements for EGUs despite

judicial stay orders that have required EPA to stay the rule’s requirements in

eleven states. 1

The Emission Allowance Trading Program Established by the Federal

“Good Neighbor Plan” for the 2015 Ozone National Ambient Air Quality

Standards.

Overview of the Good Neighbor Plan

Once EPA sets new or revised national ambient air quality standards

(“NAAQS,” or “air quality standard”), states must submit state implementation

See EPA Response to Judicial Stay Orders, https://www.epa.gov/csapr/epa-response-judicialstay-orders (last visited Aug. 17, 2023).

(Page 76 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 4 of 42

17a

plans (“SIPs”) to satisfy certain Clean Air Act requirements, including the good

neighbor provision, 42 U.S.C. § 7410(a)(2)(D)(i)(I). With respect to the 2015

NAAQS for ozone, EPA reviewed states’ good neighbor SIPs, and it approved 24

plans, disapproved 19 plans, and partially approved / partially disapproved 2 plans.

See 88 FR 9336 (Feb. 13, 2023). EPA separately found several states failed to

submit complete plans, including Pennsylvania, Utah, and Virginia. See 84 Fed.

Reg. 66612, 66613 (Dec. 5, 2019). A finding of failure to submit or disapproval of

a Good Neighbor SIP imposes no legal obligation on the state or sources within the

state, but rather imposes a legal obligation on EPA to promulgate a federal

implementation plan (“FIP”), at any time, within two years of the disapproval. 42

U.S.C. § 7410(c)(1).

EPA Administrator Michael S. Regan signed a FIP action related to

these requirements, referred to as the “Good Neighbor Plan” 2 (or the “Plan”), on

March 15, 2023, to achieve emissions reductions required by the good neighbor

provision with respect to the 2015 NAAQS for ozone. The Plan establishes federal

requirements for qualifying power-plant and industrial sources in 23 covered

states, to reduce ozone pollution during the May 1-to-September 30 “ozone

season” by reducing emissions of NOX, which is an ozone precursor pollutant.

The objective of the Plan is to eliminate the covered states’ significant

contribution to nonattainment and interference with maintenance of the 2015 ozone

NAAQS in other states as expeditiously as practicable and in alignment with the

statutory attainment schedule.

With respect to fossil fuel-fired power plants in 22 states, this action

will prohibit those emissions by implementing an allowance-based trading program

beginning in the 2023 ozone season, although the majority of the emission

reductions captured in the trading program will not begin until the phase-in of

reductions associated with new post-combustion control technology retrofits over

the 2026 and 2027 ozone seasons. The Plan also prohibits emissions through

emissions limitations and associated requirements for certain other industrial

stationary sources in 19 of those 22 states, and one other state (California),

beginning in the 2026 ozone season.

Federal “Good Neighbor Plan” for the 2015 Ozone National Ambient Air Quality Standards, 88

FR 36654 (June 5, 2023). The rulemaking docket is EPA-HQ-OAR-2021-0668 and can be

accessed through www.regulations.gov. A number of key supporting materials and additional

information are available at EPA’s website, Good Neighbor Plan for 2015 Ozone NAAQS,

https://www.epa.gov/csapr/good-neighbor-plan-2015-ozone-naaqs (last visited June 5, 2023).

(Page 77 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 5 of 42

18a

10. In assisting downwind states with attaining and maintaining the 2015

ozone NAAQS, the Plan will deliver substantial public health and environmental

benefits across wide swaths of the United States. The benefits of the Plan far

exceed its anticipated costs. Like its predecessor programs, the NO X SIP Call, 3

Clean Air Interstate Rule (“CAIR”), 4 and CSAPR, 5 the Plan can be implemented

without disruption to the reliability or affordability of the electrical power supply.

Estimated Monetized Health and Climate Benefits, Compliance Costs, and

Net Benefits of the Good Neighbor Plan, 2023 Through 2042 (Millions 2016$,

Discounted to 2023) 6

3% Discount Rate 7% Discount Rate

Present Value

Equivalent

Annualized

Value

Health Benefits

$200,000

$130,000

Climate

Benefits

$15,000

Compliance

Costs

$14,000

$9,400

Net Benefits

$200,000

$140,000

Health Benefits

$13,000

$12,000

Climate

Benefits

$970

“Finding of Significant Contribution and Rulemaking for Certain States in the Ozone Transport

Assessment Group Region for Purposes of Reducing Regional Transport of Ozone,” 63 FR

57356 (Oct. 27, 1998).

“Rule to Reduce Interstate Transport of Fine Particulate Matter and Ozone (Clean Air Interstate

Rule),” 70 FR 25162 (May 12, 2005).

“Federal Implementation Plans: Interstate Transport of Fine Particulate Matter and Ozone and

Correction of SIP Approvals,” 76 FR 48208 (Aug. 8, 2011) (generally referred to as the CrossState Air Pollution Rule, or “CSAPR”).

Adapted from Good Neighbor Plan Executive Summary. For explanations, caveats, and table

notes associated with these figures, see 88 FR 36654, 36666.

(Page 78 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 6 of 42

19a

Compliance

Costs

$910

$770

Net Benefits

$13,000

$12,000

The estimated annualized compliance costs for the Plan of $910 million (3%

discount rate, 2016$) or $770 million (7% discount rate, 2016$) are comparable to

or less than those prior interstate transport rulemakings. For example, EPA

estimated that the NOX SIP Call would cost $1.7 billion (1990$) annually to

implement. 63 FR at 57478. Similarly, CAIR was estimated to cost the power

sector $2.4 billion in 2010 and $3.4 billion in 2015 (1999$). 70 FR at 25305.

CSAPR was estimated to cost the power sector $810 million in 2014 (2007$). 76

FR at 48215.

11. The Plan will deliver substantial public health and environmental

benefits. On average, the ozone levels at the identified “receptor” locations around

the country are projected to decrease by 0.66 parts per billion (ppb). Good

Neighbor Plan, Table V.D.3-1 (88 FR at 36748). The Plan will help many

downwind areas make substantial progress toward coming into compliance with

the 2015 ozone NAAQS. In some cases, such as for receptors in Colorado, coastal

Connecticut, and Texas, the Plan is projected to make substantial progress toward

achieving full attainment of the standard.

12. According to the air quality analysis for the SIP disapproval Final

Rule, there are 43 air quality monitoring sites throughout the United States that are

identified as “receptors”—i.e., locations that are projected to struggle to attain or

maintain the 2015 ozone NAAQS. See 88 FR at 36706-08. The combined

population of the designated ozone nonattainment areas associated with these

receptors in 2021 is 82.3 million people, representing roughly 25 percent of the

total U.S. population.

13. The air quality benefits of the Plan will also reach many other people

beyond the specific areas where receptor sites are located. The map below

graphically illustrates the reduction in ozone levels that is projected to occur across

the United States with full implementation of the Plan.

(Page 79 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 7 of 42

20a

14. The emissions control strategies on which the Plan is premised are all

conventional, widely-used, at-the-source technologies that have been available to

power plants and industrial sources for decades. This level of control is widely

mandated for these types of sources in downwind areas with ozone air quality

problems. For example, selective catalytic reduction (“SCR”) control technology is

already installed at roughly two-thirds of the coal-fired power plant capacity in the

U.S. fleet. Good Neighbor Plan, 88 FR at 36768.

15. As can be seen in the figures below, many fossil fuel-fired power

plants in the states that are included in the Good Neighbor Plan have relatively

high, poorly controlled NOX emissions contributing to ozone pollution. These

sources along with other anthropogenic emissions sources in the States are

impacting air quality hundreds of miles away. (Emissions from power plants in

states subject to the Good Neighbor Plan are highlighted in red.)

(Page 80 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 8 of 42

21a

16. A delay in the implementation of the Plan would result in the

continuation of significant contribution to harmful levels of air pollution across the

United States. Delays of as long as three years in the implementation of two prior

good neighbor rulemakings (NOX SIP Call and CSAPR) have been experienced as

a result of stay litigation. In both cases, the regulations were largely upheld once

courts were able to adjudicate the merits. EPA is applying this same, now-Supreme

Court-upheld analytical framework in this Plan. A delay of three years or more

here would delay the full elimination of significant contribution under this Plan

(Page 81 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 9 of 42

22a

until 2029 or later. This would be eight years after the 2021 Marginal area

attainment deadline, five years after the 2024 Moderate area attainment deadline,

and two years after the 2027 Serious area attainment deadline. 7 In the meantime,

many Americans could suffer illness and premature death from the harmful

pollution that would be allowed to continue, while downwind areas that fail to

attain the health-based NAAQS will be subject to ever more stringent regulatory

requirements under the Act without relief from the contributing effects of upwindstate pollution. For example, the forgone emissions reductions in 2026 could result

in forgone reductions in avoided premature mortalities and illnesses equal to as

much as $14 billion (2016$, 3% discount rate).

Establishment, Applicability, and Relationship to Other Trading Programs

17. Among other things, the Plan implements a revised and expanded

allowance trading program for electricity generating units – the CSAPR NO X

Ozone Season Group 3 Trading Program (the “Trading Program”). This program

generally applies to fossil fuel-fired boilers and combustion turbines that are

located in covered states and serve generators larger than 25 megawatts producing

electricity for sale. 40 CFR 97.1004.

18. The Plan amends the existing CSAPR NOX Ozone Season Group 3

Trading Program established for twelve states in 2021 under the Revised CSAPR

Update.8 The CSAPR NOX Ozone Season Group 3 Trading Program was first

established to achieve emissions reductions required by the good neighbor

provision with respect to the 2008 ozone NAAQS. For several other states, the

Trading Program will replace the CSAPR NOX Ozone Season Group 2 Trading

Program established in 2016 under the CSAPR Update and currently still being

implemented for ten states. 9 The CSAPR NOX Ozone Season Group 1 Trading

Program was first established in the original CSAPR rulemaking in 2011 to

address good neighbor obligations associated with the 1997 ozone NAAQS, and

currently applies only in the State of Georgia. 10

19. Under the Plan, power plants in seven states will transition from the

CSAPR Group 2 Trading Program to the CSAPR Group 3 Trading Program, and

power plants in three states not currently covered by a CSAPR trading program for

seasonal NOX emissions will be added to the CSAPR Group 3 Trading Program.

Further discussion of the disruptive consequences of a stay of the Final Rule is in section V

below.

“Revised CSAPR Update for the 2008 Ozone NAAQS,” 86 FR 23054 (April 30, 2021).

“CSAPR Update for the 2008 Ozone NAAQS,” 81 FR 74504 (October 26, 2016).

CSAPR, 76 FR 48208 (Aug. 8, 2011).

(Page 82 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 10 of 42

23a

20. Virtually all of the electric generating units covered by the Plan,

including those in the three states not currently covered by a seasonal NO X

emissions program, nonetheless participate in the Acid Rain emissions trading

program under Title IV of the Clean Air Act and already meet rigorous monitoring

and reporting requirements in compliance with 40 CFR Part 75. 11

21. Power plants have deep familiarity with Clean Air Act compliance

assurance and permitting obligations and face minimal administrative burdens

associated with entry into the Trading Program. All of the units that will participate

in the Trading Program already participate in multiple CSAPR trading programs

for 20 of the 22 states. For the remaining two, Utah and Nevada, nearly all EGUs

already participate in the Acid Rain Program.

22. As finalized (and not accounting for judicial stays, which are

discussed in Section IV), the emissions reduction requirements associated with the

new Trading Program emissions budgets established by the Plan apply as of the

effective date of the Plan, which was August 4, 2023, 60 days after publication of

the Plan in the Federal Register on June 5 (88 FR 36654). 12

23. As finalized (and not accounting for judicial stays, which are

discussed in Section IV), for units in the states already covered by either the

CSAPR NOX Ozone Season Group 2 or Group 3 trading programs, the Plan has

transitional provisions so that the new budgets apply only after the Plan’s effective

date.13 For units in the remaining states that will be newly covered by the Trading

Program, no requirements, either in terms of emissions reductions or in terms of

other administrative requirements, applied until the effective date.

How Emissions Trading Programs Work and the Enhancements to the

Trading Program

24. The Clean Air Markets Division operates or has operated a number of

allowance trading programs, the earliest of which started more than 25 years ago.

These include the NOX Budget Trading Program for ozone-season NOX emissions

Approximately 97 percent of ozone season NOX emissions reported under Part 75 are

determined using continuous emissions monitoring systems (“CEMS”). Gas- or oil-fired units

that qualify as peaking units or low mass emissions units under the regulations have options to

determine reported emissions using other methodologies.

See Good Neighbor Plan Preamble Section VI.B.12.a (88 FR at 36811-13).

See id.

(Page 83 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 11 of 42

24a

under the NOX SIP Call, 14 programs for ozone-season and annual NOX emissions

under CSAPR and CAIR, 15 and programs for annual SO2 emissions under CSAPR,

CAIR, and the Acid Rain Program. 16 Most of the units that are covered by the

Plan’s Trading Program also participate or participated in some of these other

programs. There has not been, nor have Declarants identified, a single instance

where implementation of these EPA trading programs has caused an adverse

reliability impact.

25. EPA provides robust technical analysis for identifying its emission

reduction requirements. For power plants, this includes starting with reported data

for recent historical operations. It further tests these requirements against future

expectations for the sector by using a state-of-the-art, peer-reviewed programming

model of the contiguous U.S. electric power sector (the Integrated Planning Model,

or IPM). IPM provides forecasts of least-cost capacity expansion, electricity

dispatch, and emissions control strategies while meeting energy demand and

environmental, transmission, dispatch, and reliability constraints.

26. The Trading Program, like the other current and former allowance

trading programs operated by the Clean Air Markets Division, does not impose any

fixed limits on the operations or emissions of individual affected units. Instead,

each affected unit is required to monitor and report its emissions, and each source

with affected units is required to hold quantities of emission “allowances” based on

the reported emissions from all its affected units for each “control period” for the

program. (For the Trading Program, the control period is the May-September

ozone season.) Allowances can be traded with other sources covered by the

program in the same or other states or with third parties (e.g., brokers). The

aggregated emissions from all the affected units under such a program are limited

by the total number of allowances issued for use in the program, each of which

authorizes the emission of up to one ton of NOX.

27. The Trading Program budgets are set based on an evaluation of

available NOX mitigation control technologies. As in the prior CSAPR

rulemakings, as well as the earlier CAIR and NOX SIP Call rulemakings, EPA

assessed several well-understood, widely-available, at-the-source emissions control

“Finding of Significant Contribution and Rulemaking for Certain States in the Ozone

Transport Assessment Group Region for Purposes of Reducing Regional Transport of Ozone,”

63 FR 57356 (Oct. 27, 1998).

“Rule to Reduce Interstate Transport of Fine Particulate Matter and Ozone (Clean Air

Interstate Rule),” 70 FR 25162 (May 12, 2005).

CAA subchapter IV-A, 42 U.S.C. 7651-7651o; 40 CFR parts 72-78.

(Page 84 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 12 of 42

25a

strategies. Following a multifactor assessment at “Step 3” of the interstate transport

framework, EPA arrived at a suite of control strategies that obtain cost-effective

emissions reductions delivering meaningful downwind air quality benefits. EPA’s

assessment of these technologies is set forth in Section V.B-C of the Plan

preamble, 88 FR at 36720-40.

28. The primary strategies for power plants that emerged from this

analysis are: starting in 2023, optimizing existing post-combustion controls;

starting in 2024, upgrading to state-of-the-art combustion controls at the few

remaining coal facilities without them; and, over the 2026-2027 ozone seasons,

retrofitting post-combustion controls on large emitting units currently lacking

them. See Good Neighbor Plan Preamble Section VI.A, 88 FR at 36754-58.

29. It bears noting that the 2023 and 2024 strategies of optimizing

existing post-combustion controls and upgrading combustion controls are

essentially identical to the emissions control strategies that were identified in

CSAPR, the CSAPR Update, and the Revised CSAPR Update. EPA’s analysis in

the Good Neighbor Plan is that these strategies remain widely available on a

relatively near-term basis and additional, cost-effective emissions reductions can

be obtained from these strategies across the fleet of existing power plants in the

covered upwind states.

30. For each control period, the total quantity of allowances is initially

allocated among the affected units. 17 Allocations in Good Neighbor trading

programs have followed a similar methodology for many years, relying on

historical heat input and emissions data to determine how many allowances to

allocate to each unit, as well as to new units. The Plan generally follows this

approach with some minor changes from prior programs. See generally Good

Neighbor Plan Preamble Section VI.B.9, 88 FR at 36801-08. Among the features

of the Plan’s allocation methodology, similar to prior programs, is a “new unit set

aside,” which is available for any power plants that would not otherwise receive

allocations, mainly (but not exclusively) new power plants that come online after

the Plan is issued. Power plants that have gone offline but are then returned to

operation also qualify to receive allocations from the new unit set aside if the

plants are no longer eligible to receive allocations as existing units. Often, there are

CSAPR trading programs are designed to allow states to easily replace EPA’s allocation

methodology with their own. States may also leave the FIP through adopting the trading program

in full (in addition to replacing EPA’s allocation methodology) into their state program or

developing their own approaches for approvable SIPs that can replace the FIP. See Good

Neighbor Plan Preamble Section VI.D, 88 FR at 36838-43.

(Page 85 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 13 of 42

26a

allowances remaining in each state’s new unit set aside, and if so, these are

recycled back to existing units in proportion to their original allocation. Thus, the

entire budget for each compliance period is fully allocated, and units typically will

receive more allowances than their initial allocation figures suggest.

31. The allowance allocation methodology is distinct from the method of

determining the emissions budget for each state. The number of initial allocations

each affected unit receives is not an emission limit, nor is it an express or implied

prohibition on how much that source may emit; rather, sources may buy or sell

allowances with any other party and use them for compliance. This incentivizes

units that can reduce their emissions easily or cheaply to make those reductions

and reap the benefits from selling their unneeded allowances, while units with

relatively more expensive reduction opportunities can comply by purchasing those

allowances. For a more comprehensive overview of emissions trading programs,

see the Division’s website at https://www.epa.gov/emissions-trading-resources.

32. Sources with affected units under the Trading Program are not

required to hold allowances to cover their emissions before or at the actual time of

the emissions (e.g., in or during the 2023 ozone season). Instead, each source is

obligated to surrender allowances to cover its affected units’ emissions for a

control period by the program’s “allowance transfer deadline,” which is June 1 of

the year after the year of the control period. 40 CFR 97.1002, 97.1006(c)(1). For

the 2023 ozone season, the allowance transfer deadline will be June 1, 2024. Thus,

a source in the Trading Program has an extended period of time—eight months

after the end of the ozone-season control period on September 30—in which to

acquire any additional allowances that may be needed for compliance.

33. Each state’s budget determines the total number of allowances to be

allocated among the state’s affected units for each control period. However, a state

budget is not a limit on how much NOX a state’s affected units may emit, in the

aggregate, during the control period. Under the Trading Program (like the trading

programs under the original CSAPR), the aggregated emissions from a state’s

affected units can exceed the state’s budget up to a certain level, called the

“assurance level,” without triggering any further obligations beyond each source’s

basic compliance obligation to hold allowances equal to the sum of its affected

units’ emissions. If the aggregated emissions from the affected units in a single

state exceed the state’s assurance level during a control period, the sources that

contributed to the state’s exceedance must surrender two additional allowances for

each ton of their respective shares of the exceedance. 40 CFR 97.1025. The

assurance levels include “variability limits” beyond the respective state emissions

budgets that allow for potential inter-annual variability in operating needs for each

(Page 86 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 14 of 42

27a

state. At the same time, the assurance levels function within the structure of an

interstate trading program to meet the Act’s requirement that each state’s sources

are held to the elimination of the state’s significant contribution.

34. The Trading Program is fully achievable without any need for sources

to reduce their operations or retire, because the emission budgets are premised on

widely available pollution control technologies described above whose use would

achieve the required emission reductions without need to reduce operations or

retire any affected EGU. However, under the Trading Program, no power plant is

required to follow these strategies. In general, a power plant owner has options to

operate the emissions controls identified by the EPA for that particular type of unit

(including installation or upgrade of controls), operate other types of emissions

controls, or adapt the unit’s levels of operation to produce less emissions. The Plan

generally preserves the compliance flexibility of prior transport trading programs

in reserving these decisions to sources’ owners and operators.

35. While preserving the intrinsic emissions trading compliance

flexibilities noted above, the Trading Program contains several enhancements

relative to prior trading programs. These enhancements operate together to ensure

that, within the structure of an interstate trading program, sources continue to

achieve a degree of emissions reduction consistent with the Act’s requirement to

eliminate “significant contribution.” As EPA discusses in the Plan, experience with

prior trading programs has produced evidence that over time sources may not be

properly incentivized to operate emissions controls to the degree needed to

eliminate significant contribution on an ongoing basis. The enhancements included

in the Trading Program continue to provide flexibility while providing greater

assurance that significant contribution will be eliminated on the most critical days

of the ozone season and will remain eliminated on a permanent basis. EPA made

several adjustments to these enhancements from the proposal in light of comments

regarding grid reliability, as discussed in the following section beginning at

paragraph 37.

36. There are four enhancements to the Trading Program in the final Plan,

compared to prior CSAPR programs: dynamic budgeting; annual bank

recalibration; unit-specific backstop daily emissions rates; and a secondary

emissions limitation:

a. Dynamic Budgets: Prior trading rules used a single, fixed emissions

budget, set based on power sector data as of the date of the action. In the

Revised CSAPR Update, EPA established preset budgets for several

years into the future, to better reflect known changes in the power sector

(Page 87 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 15 of 42

28a

over time. In the Good Neighbor Plan, EPA is again establishing preset

budgets as floors for the 2023 to 2029 control periods. Beginning in

2026, dynamic budgets (i.e., budgets set by applying the emission

control strategies selected in the Plan to more recent operating data) will

be calculated for each control period. From 2026 through 2029, a state’s

dynamic budget will be used only if it is higher than the state’s preset

budget for that control period. Beginning in 2030, dynamic budgeting

will be the sole method of budget calculation. See Good Neighbor Plan

Preamble Section VI.B.4, 88 FR at 36777-79.

b. Bank Recalibration: If a source does not use all of its allowances to

demonstrate compliance in a given control period, the Trading Program,

like all the other allowance trading programs operated by the Clean Air

Markets Division, allows the unused allowances to be banked for use in

the program in future control periods. In the CSAPR Update and the

Revised CSAPR Update, EPA executed one-time conversions of

available banked allowances from prior trading programs into initial

allowance banks appropriately scaled to the budgets under the new

trading programs. The Plan carries that process forward by limiting the

collective allowable number of banked allowances for the Trading

Program that can be carried over each year to 21% of the sum of the

states’ emissions budgets, starting with the 2024 ozone season. This will

prevent the buildup of an excessively large bank of allowances that

would undermine program stringency in the latter years of a program.

See Good Neighbor Plan Preamble Section VI.B.6, 88 FR at 36788-91.

c. Unit-specific Backstop Daily Emissions Rates: To ensure more

consistent operation of installed controls on sources with the highest

level of emissions potential throughout each day of ozone seasons going

forward, the Plan includes backstop daily emission rates applied to each

of a subset of the covered sources. This rate applies beginning in 2024

for large, coal-fired sources that have SCR post-combustion emissions

controls already installed. The rate is set at a level that reflects seasonal

optimization of the control (not daily maximal performance), and

sources must surrender additional allowances for the emissions

associated with exceedances of this rate (after a 50-ton threshold, which

accommodates the potential unavoidable emissions sources might have

above the daily rate associated with activities like start-up). The same

rate is applied for large coal-fired units with SCR-retrofit potential in

the second control period after such control is installed or in 2030,

(Page 88 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 16 of 42

29a

whichever occurs first. See Good Neighbor Plan Preamble Section

VI.B.7, 88 FR 36791-97.

d. Secondary Emissions Limitations: To avoid foreseeable exceedances of

the state-by-state assurance levels, the Plan establishes the conditions

for an enforceable Clean Air Act violation in defined circumstances of

egregious failure to operate existing pollution controls, starting with the

2024 ozone season. See Good Neighbor Plan Preamble Section VI.B.8,

88 FR at 36797-801.

Key Changes in the EGU provisions of the Good Neighbor Plan from

Proposal

37. The proposal underwent a comment period of 76 days, and EPA held

many stakeholder meetings, including with electricity reliability coordinators, to

receive feedback as well. This public engagement provided useful information to

the Agency and produced a number of important changes in the Good Neighbor

Plan.

38. The information regarding the contents of the Good Neighbor Plan,

reflective of these changes, became available to the general public on or about

March 15, 2023, with the release of the unofficial, pre-publication copy of the Plan

on EPA’s website.

39. Several changes to the EGU-related provisions in the Good Neighbor

Plan bear directly on the claims of harm put forward by Movants. These changes

respond to concerns raised by commenters that the Plan, as proposed, could have

unintended effects on power sector grid-reliability.

40. Commenters observed that the fleet of fossil-fuel fired power plants is

undergoing a period of transition to cleaner fuels and technologies. Many power

plant owners and operators highlighted their interest in seeing flexibility in this

program that would facilitate their business decisions, while, in their view, the Plan

as proposed could force uneconomical decisions either to retire power plants

earlier than intended or to force expensive pollution-control retrofits for sources

that in their judgment would otherwise not continue in operation for much longer.

See Good Neighbor Plan Preamble Section VI.B.1.d, 88 FR at 36770-75.

41.

During rule development, EPA also actively engaged with key

stakeholders in the electricity sector, including system operators, regional

(Page 89 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 17 of 42

30a

transmission operators (“RTOs”), the U.S. Department of Energy (“DOE”), the

Federal Energy Regulatory Commission (“FERC”), and other parties that have the

responsibility for ensuring reliability. EPA hosted a series of meetings with

reliability organizations who had commented on the proposal to ensure we had a

solid understanding of their concerns and perspectives. See Good Neighbor Plan

Preamble Section III.B.1.c, 88 FR 36678-80.

42. In light of these viewpoints, EPA adopted multiple changes from the

proposal to address the reliability-related concerns identified in comments and

brought into greater focus through consultations with RTOs and other agencies.

These changes have been carefully crafted to ensure the statutory mandate to

eliminate significant contribution to interstate pollution problems under the Clean

Air Act is met without disrupting the reliable operation of the bulk power grid. See

Good Neighbor Plan Preamble Section VI.B.1.d, 88 FR at 36770-75.

a. EPA had proposed to apply “preset” state emissions budgets only for

the control periods in 2023 and 2024, with dynamic budgeting allowing

for changes in the budget both upward and downward beginning in

2025. EPA had proposed to use only one year of data in the dynamic

budget-setting process. In the Final Good Neighbor Plan, preset

budgets will operate as floors from 2023 through 2029. This will

establish predictable minimum quantities of allowances available

during the period when commenters have expressed concern that the

reliability-related need for such predictability is greatest. In addition,

the dynamic budgets will be set using multiple years of operating data

to prevent an anomalous year of data from skewing the budgets. See

Good Neighbor Plan Preamble Section VI.B.1.b.i, 88 FR at 36764-66.

b. The target percentage of the state emission budgets used to annually

recalibrate the allowance bank will not be set at the proposed 10.5

percent level until the 2030 control period. For the control periods from

2024 through 2029, a target percentage of 21 percent will be used

instead. The adoption of the higher target percentage for use through the

2029 control period is intended to enhance the availability of allowances

during this period by allowing power plant owners and operators to

“bank” allowances at a higher level through 2030. See Good Neighbor

Plan Preamble Section VI.B.1.b.ii, 88 FR at 36766-67.

(Page 90 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 18 of 42

31a

c. The application of the backstop daily emissions rate for units without

existing SCR controls is deferred until the 2030 control period from the

2027 control period as EPA had proposed. This change extends by

several years the period during which the highest emitting sources in the

fleet may continue surrendering only one allowance per ton emitted, as

opposed to three allowances per ton emitted, while operating without

widely available pollution control technology within the Trading

Program. See Good Neighbor Plan Preamble Section VI.B.1.c.i, 88 FR

36767-69.

43. Additionally, EPA made several other key changes in the Good

Neighbor Plan from the proposal that will also help ensure it can be implemented

on a feasible and cost-effective basis in light of comments and other record-based

considerations that in EPA’s judgment warranted attention:

a. The Good Neighbor Plan does not require any emission reductions

associated with projected generation shifting using EPA’s Integrated

Planning Model. See Good Neighbor Plan Preamble Section V.B.1.f, 88

FR at 36731-32.

b. EPA finalized a phase-in approach for emission reductions associated

with the SCR-retrofit strategy. These reductions are phased in over

2026-2027 in the final Good Neighbor Plan, as opposed to just 2026 at

proposal. This change provides an additional year for the full

implementation of reductions associated with this strategy relative to the

proposal. See Good Neighbor Plan Preamble Section VI.A, 88 FR at

36757-58.

c. Emissions control stringency associated with combustion control

upgrades does not go into effect for any state until the start of the 2024

ozone season. See Good Neighbor Plan Preamble Section V.A, 88 FR

36754-55.

II.

NOX Mitigation Strategies and Timing: Further Detail

44. The Plan assumes two mitigation strategies in setting emission

budgets for the 2023 ozone season. This is the optimization of two types of

existing post-combustion controls—SCR and selective non-catalytic reduction

(“SNCR”). Therefore, no new pollution control equipment is assumed in 2023,

only the operation of existing equipment. EPA uses its database of reported

(Page 91 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 19 of 42

32a

historical power sector operations and emissions performance to derive state

emissions budgets based on these (and other) strategies. According to EPA data,

power plants have demonstrated through their historical operation (for more than

90% of such units) that they have already achieved this level in the past, in many

cases significantly out-performing the representative performance rates used by

EPA to establish budgets based on these strategies.

45. The vast majority of SCR-controlled units (nationwide and in the 22

states covered by the Trading Program) at least partially operated these controls

during the 2021 and 2022 ozone seasons, based on reported emissions rates.

Existing SCRs operating at partial capacity still provide functioning, maintained

systems that may only require increased frequency or quantity of delivered

chemical reagents (i.e., ammonia or urea), which can be accomplished within a few

weeks. In many cases, units with SCR have historically achieved more efficient

NOX removal rates than their current performance and therefore are capable of

reverting to earlier operation and maintenance plans that achieved demonstrably

better SCR performance.

46. There is ample evidence of units restoring optimal performance of

post-combustion controls within a timeframe of two months or less. See Good

Neighbor Plan Preamble Section V.B.1.a, 88 FR at 36720-25. Not only have units

reactivated SCR performance levels at the start of an ozone season or when

requirements took effect, but unit-level data also shows instances where sources

demonstrated the ability to quickly alter their emissions rate within an ozoneseason and even within the same day in some cases. Moreover, this emissions

control technique is familiar to sources and was analyzed and included in the

Revised CSAPR Update emissions budgets finalized in 2021 and the CSAPR

Update emissions budgets finalized in 2016.

47. The recently implemented Revised CSAPR Update was finalized on

March 15, 2021, with emissions reductions premised on the same technology and

nearly identical implementation schedules as this Plan regarding existing control

optimization. See paragraph 51. Sources were able to comply with a 100% success

rate in meeting their allowance-holding requirements, and units optimized their

controls, showing significant improvement in emissions performance relative to

prior years. Neither sources nor state agencies and reliability authorities reported

any difficulty maintaining compliance with electric reliability standards as a

function of achieving compliance with the Revised CSAPR Update.

48. The recent experiences with both the Revised CSAPR Update and

CSAPR Update underscore the eminently achievable nature of the control

(Page 92 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 20 of 42

33a

strategies informing the establishment of the Trading Program budgets for 2023

and 2024.

49. In the Plan, EPA finds that new SCR retrofit installation is costeffective and is included as part of the overall strategy to eliminate significant

contribution. Corresponding emission reductions are reflected in state emissions

budgets, phasing in over the 2026 and 2027 ozone seasons. EPA extended the

timeframe for installation of SCR controls from 36 months at proposal to 36-48

months in the final Plan. There are many instances of individual SCR-retrofit

projects being completed well within a three-year timeframe; however, a 36-48

month period corresponds with EPA’s expectations regarding timing needs for

fleetwide implementation of this strategy. There is significant engineering

literature and third-party testimonials as to the feasibility of this timing for sources

pursing this compliance option. This technology is widely available. SCR controls

already exist on over 60 percent of the coal fleet in the states covered by the

Trading Program. Nearly every pulverized coal unit larger than 100 MW built in

the last 30 years has installed this control.

50. The timeframes by which the requirements of the Plan go into effect

are all keyed to the finalization of the Plan. Thus, the phasing in of the SCR-retrofit

stringency over the 2026-2027 ozone seasons corresponds to a 36-48 month period

from the date of issuance of the Plan. See Good Neighbor Plan Preamble Sections

V.B.1.e, 88 FR at 36726-31, and VI.A, 88 FR at 36757-58.

51. The implementation of the Plan’s budgets reflecting the 2023 control

strategy as of the effective date 60 days after publication in the Federal Register

accommodates the two-month period EPA found to be the maximum amount of

time needed to implement these strategies. These exact technology and timing

assumptions were just successfully implemented on an identical schedule in the

Agency’s Revised CSAPR Update rule, which was finalized on March 15, 2021

and included emission reduction requirements premised on optimization of existing

controls going into effect upon the effective date of the rule during the 2021 ozone

season. See Good Neighbor Plan Preamble Section V.B.1.a., 88 FR at 36720-21;

see also Revised CSAPR Update, 86 FR 23054.

52. With respect to the control strategies reflected beginning in the 2026

and 2027 ozone seasons, the preliminary analysis and engineering steps involve no

capital costs; these include pre-construction activities, such as engineering studies,

conceptual design, schedule, specifications, and cost estimates. For power plants

that choose to pursue a strategy of retrofitting post-combustion controls such as

(Page 93 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 21 of 42

34a

SCR, much of the first year of project work typically need not entail substantial

capital outlays. 18

III.

Achievability of the Good Neighbor Plan

53. The emissions reductions implemented through the Plan’s Trading

Program are readily achievable for the covered power plants, and the Program is

designed so as not to threaten resource adequacy or otherwise degrade electric

system reliability in any state or region. This section will discuss the achievability

of the Trading Program as finalized in the Good Neighbor Plan on March 15, 2023.

The following section will discuss why the Trading Program remains achievable

for covered states not under judicial stays.

54. Under the Trading Program, for each control period EPA allocates an

amount of allowances equal to each state budget among the affected units in the

respective state. For control periods after 2023, a state may submit a state

implementation plan revision replacing EPA’s unit-level allocations with unit-level

allocations of its choosing, provided that the total number of allocations does not

exceed the state budget. 40 CFR 52.38.

55. The sum of the preset state budgets under the Trading Program for

2023 as designed was 208,119 tons. (For the set of states that would have been

subject to the trading program for the entire 2023 ozone season, prorating of the

budgets to account for the effective date of the plan, as discussed in paragraphs 2223, would have increased this amount by 20,123 tons.) Adding the amount of

allowances in the anticipated starting bank (see paragraph below), EPA estimated

that the total number of allowances that would have been available for compliance

in 2023 was approximately 269,479 allowances prior to any 2023 prorating due to

the August 4 effective date. Under EPA’s prorating approach, the quantity of

allowances available per day of compliance increased in proportion to each day of

delay in the FIP’s effective date beyond May 1, 2023.

56. In addition to allowances allocated for each control period and already

banked under the Group 3 Trading Program, the EPA will convert for use in the

Trading Program an amount of allowances banked under the existing CSAPR NO X

Ozone Season Group 2 trading program. 40 CFR 97.826. Any affected unit (or

other entity) that holds banked allowances issued under the CSAPR Group 2

program will be issued a proportional number of converted allowances that can be

See document titled “Typical SCR and SNCR Schedules 2023” in the docket for the Good

Neighbor Plan (EPA-HQ-OAR-2021-0668-0975).

(Page 94 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 22 of 42

35a

used under the Trading Program just like allowances allocated from the Trading

Program state budgets. Based on emissions data for 2022, in total, the alreadybanked and converted allowances collectively would have constituted a “starting

bank” of approximately 61,360 allowances available for 2023 compliance prior to

any 2023 prorating due to the August 4 effective date.

57. Total emissions from the sources that would be covered by the

Trading Program in 2021 were 239,507 tons, and in 2022 were 207,524 tons. As

EPA has observed in prior CSAPR trading programs, EPA fully anticipates that

sources will in fact optimize existing controls during the 2023 ozone season and/or

pursue other emissions reduction opportunities, in response to the allowance price

signal and in order to maintain or increase the respective amounts of banked

allowances they hold for their own use or for sale to others. Nonetheless, these

numbers indicate that even if no sources had chosen to reduce emissions in 2023

below where they already were in 2022, there would have been adequate

allowances available for compliance. As explained in Section IV, even with

judicial stays limiting the scope of the program, there are more than enough

allowances available for compliance in 2023 for the eleven states currently in the

program.

58. The most recently reported emissions data, which are for the first and

second quarters of 2023, show a substantial reduction in emissions was achieved

among power plants within the Good Neighbor Plan trading region. NO X emissions

decreased by 19 percent for the months of May and June, compared with 2022

levels, from approximately 75,000 to 60,000 tons. (Data available at

https://campd.epa.gov/.)

59. In addition, the data for May and June of 2023 compared with the

same time period in 2021 and 2022 indicate a marked improvement in the

emissions performance of coal-fired EGUs equipped with SCR in the 22 states

covered by the Good Neighbor Plan Trading Program. The data also indicate that

performance remained flat or deteriorated in states not covered by the Good

Neighbor Plan. SCR-controlled coal-fired units in the Good Neighbor Plan (GNP)

footprint improved emission-rate performance by approximately 15% while nonGNP states saw a deterioration in performance among similar units.

(Page 95 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 23 of 42

36a

Average May/June Emission Rate for SCRcontrolled Coal (lb/mmBtu)

0.085

0.082

0.080

0.082

0.081

0.080

0.075

0.073

0.070

0.065

0.060

2021

2022

GNP states

2023

Non-GNP States

SCR performance among the worst performing SCR-controlled units (those

emitting > 0.1 lb/mmBtu in 2021 prior to GNP proposal) has improved

significantly in GNP states but has been flat in non-GNP states.

Average May/June Emission Rate for Poor

Performing SCR-controlled Coal (lb/mmBtu)

0.18

0.16

0.15

0.14

0.13

0.12

0.11

0.09

0.10

0.08

0.06

2021

2022

GNP

2023

Non-GNP States

60. The Good Neighbor Plan will not adversely affect the reliability of

electricity supply. EPA conducted a “resource adequacy” assessment for the Good

Neighbor Plan. This assessment shows that accredited capacity projections, and

therefore reserve margins, are expected to be virtually identical for the power

sector between the baseline and the Good Neighbor Plan “policy case.” In

particular, in 2023, 2025, and 2030, reserve margin projections under the Plan

(Page 96 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 24 of 42

37a

remain consistent with baseline projections and are at or above target reserve

margins.19

61. For all North American Electric Reliability Corporation (NERC)

reliability assessment regions and for all years, adequate reserve margins are

projected to be maintained under the Good Neighbor Plan. Projected changes in

reserve margins under the Plan through 2030 are exceedingly small relative to the

baseline without the rule. 20

62. The Plan’s projected effect on retail electricity prices relative to

baseline projections is also projected to be exceedingly small. In 2023, there is a

0% change projected. In 2025, the changes are on the order of -1% to 1%. In 2030,

the changes are of a similar magnitude, with only one area of the country projected

to see a greater than 2% change in electricity prices. 21

63. Compliance with the Good Neighbor Plan is anticipated to be even

less costly than EPA’s primary analysis of compliance costs in the Plan’s

regulatory impact analysis (RIA) suggests (see paragraph 10). EPA conducted a

supplementary analysis to assess the effects of the Inflation Reduction Act of 2022,

Pub. L. 117-169 (“IRA”). That analysis indicates that the annualized cost of the

Plan for the power sector over the 2023-2045 period declines under the IRA from

$449 million/year to $196 million/year (2016$). See RIA Appendix 4A, Table 4A2. For comparison, the annualized costs of the NOX SIP Call were estimated at

$1.7 billion (1990$), which would be $2.8 billion in 2016$.

64. There has never been a shortage of allowances in any allowance

trading program operated by the Clean Air Markets Division from 1995 – the first

year of the Acid Rain Program’s trading program for SO2 emissions – to the

present. After the allowance transfer deadline for every control period for every

such program, a bank of unused allowances has always been available for

See Resource Adequacy and Reliability Analysis Final Rule TSD 2, Tbl. 1, available at

https://www.epa.gov/system/files/documents/2023-03/Resource%20Adequacy%20and%20

Reliability%20Analysis%20TSD.pdf.

Id. Tbl. A3, B3, C3.

See Regulatory Impact Analysis for the Final Federal Good Neighbor Plan Addressing

Regional Ozone Transport for the 2015 Ozone National Ambient Air Quality Standard 166-68,

Tbl. 4-15, 4-16, 4-17, available at https://www.epa.gov/system/files/documents/202303/SAN%208670%20Federal%20Good%20Neighbor%20Plan%2020230315%20RIA_Final.pdf.

(Page 97 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 25 of 42

38a

carryover to future control periods. See the Division’s progress reports at

https://www3.epa.gov/airmarkets/progress/reports/index.html.

65. Under the Trading Program, like EPA’s other allowance trading

programs, affected units are required to report their hourly emissions data to the

Clean Air Markets Division on a quarterly basis, and all allowance allocations and

transfers are also recorded by the Division. 40 CFR 97.1020–97.1035. The

Division maintains publicly accessible databases of the reported emissions data

and the recorded allocation and transfer data at https://campd.epa.gov/. Sources

and other participants in the market for emissions allowances, such as brokers, can

use these data to assess the potential supply of and demand for allowances and to

identify potential buyers and sellers.

66. Buyers and sellers of allowances are generally not required to report

transaction prices to the Clean Air Markets Division. However, subscription data

services regularly survey and report market prices for allowances in EPA’s

allowance trading programs. As of August 8, 2023, one such service reported a

market price of $763 per Group 2 allowance and $3,625 per Group 3 allowance. 22

The recently reported Group 3 allowance price of $3,625/ton represents a decline

of about 75 percent from reported prices immediately prior to the mid-March prepublication release of the Plan. In particular, prices declined steeply in mid-July,

around the same time that 2023 second quarter emissions data became publicly

available, indicating a substantial decrease in emissions compared to 2022.

$/allowance as

of: 08/08/2023

CSAPR SO2 Group 1

2.31

CSAPR SO2 Group 2

2.81

CSAPR NOx Annual

2.00

CSAPR NOx Seasonal (Group 3)

3,625

CSAPR NOx Seasonal (Group 2)

763

Price data are reported by S&P Global Market Intelligence and are available by subscription at

https://www.SNL.com.

(Page 98 of Total)

USCA Case #23-1157

Document #2013255

Filed: 08/18/2023

Page 26 of 42

39a

67. While prices reported for the first part of the 2022 ozone season were

higher than prices in the later part of and after the ozone season, relatively few

allowance transfers among unrelated parties took place during the period of the

highest reported prices. Moreover, EPA’s data indicates that there were more than

enough allowances available for compliance with the Revised CSAPR Update

Group 3 program in 2022. Our data indicate total emissions in the Revised CSAPR

Update Group 3 program (covering 12 states) of around 90,458 tons in the 2022

ozone season, while ava

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.