Respondents Brief — Nat'l Mining Ass'n v. Envtl. Prot. Agency, 135 S. Ct. 703 (2014) (No. 14-49)

Nos. 14-46, 14-47, 14-49

IN THE

Supreme Coot of the Hnited Stites

MICHIGAN, ET AL., Petitioners,

Vv.

ENVIRONMENTAL PROTECTION AGENCY, ET AL., Respondents.

UTILITY AIR REGULATORY GROUP, Petitioner,

Vv.

ENVIRONMENTAL PROTECTION AGENCY, ET AL., Respondents.

NATIONAL MINING ASSOCIATION, Petitioner,

Vv.

ENVIRONMENTAL PROTECTION AGENCY, ET AL., Respondents.

On Writs Of Certiorari To The

United States Court Of Appeals

For The District of Columbia Circuit

BRIEF OF INDUSTRY RESPONDENTS CALPINE

CORPORATION, EXELON CORPORATION,

NATIONAL GRID GENERATION LLC, AND PUBLIC

SERVICE ENTERPRISE GROUP, INC.

BRENDAN K. COLLINS PAUL M. SMITH

Counsel of Record MATTHEW E. PRICE

ROBERT B. MCKINSTRY, JR. ERICA L. Ross

RONALD M. VARNUM JENNER & BLOCK LLP

LORENE L. BOUDREAU 1099 New York Ave., NW

BALLARD SPAHR LLP Suite 900

1735 Market Street Washington, DC 20001

51st Floor (202) 639-6000

Philadelphia, PA 19103 psmith@jenner.com

(215) 665-8500 Librery of Congress

collins@ballardspahr.com Lisiv o102cty

Counsel for Industry Respondents

RULE 29.6 DISCLOSURE STATEMENT

Respondents Calpine Corporation, Exelon

Corporation and Public Service Enterprise Group, Inc.,

are publicly traded corporations and have no parent

companies. No publicly-held company owns 10% or

more of their stock. Respondent National Grid

Generation LLC is a wholly-owned subsidiary of

KeySpan Corporation. KeySpan Corporation is a

wholly-owned subsidiary of National Grid USA.

National Grid USA is wholly-owned by National Grid

North America Inc., which is wholly-owned by National

Grid (US) Partner 1 Limited. National Grid (US)

Partner 1 Limited is wholly-owned by National Grid

(US) Investments 4 Limited, which is wholly-owned by

National Grid (US) Holdings Limited, which is wholly-

owned by National Grid ple. National Grid plc is a

publicly traded company that has no parent companies,

and no publicly-held company holds 10% or more of its

shares.

i

TABLE OF CONTENTS

RULE 29.6 DISCLOSURE STATEMENT.................... i

pe ey eee Vv

LIN IUTE siicchitnesesiriihcnsianiseniicibeimiingtibdeiatiaditiiniimeiabiaeiasisiiibins l

A. Regulation of Hazardous Air

Pollutants From Power Plants.................. 1

1. Section 112’s Regulatory

BUTI, ‘ccccsttininslecbininaniccnsianesisiniiinninnini 2

2. The Utility Study and

EPA’s Regulatory

ID: scccnieicdissicnnismnacictasnintsitlstieniiivniints 5

B. The Economics of Pollution

Control in Wholesale Electricity

BND cnvedcetnnescaninsenensinniatiimeataqmenmmesenisinanes 7

C. FN Tae iisenieiiiiiniicimnien eimai 11

D. pe er 12

kg gy, 14

TTD ceniticninsintesterbiniietisnniinbinouininiiisiesiiaineidianiiiiiabimaatasinntin 16

I. EPA Permissibly Chose to Regulate

Power Plants in the Same Manner That

Congress Mandated for All Other Sources

of Hazardous Air Pollutants. ..........................000.- 16

A. Section 112(n)(1)(A) Does Not

Unambiguously Mandate

Consideration of Costs. .................c..scss--.- 17

Il.

A.

B.

Surrounding Provisions Confirm

That EPA’s Interpretation Is

Deferring Consideration of Costs

Until the Standard-Setting Phase

Bee Ne I sicisiensncndccnnsinsrsnnnssentionnionncen 24

EPA’s Approach Is Particularly

Suitable Given the Highly

Competitive Nature of Wholesale

Electricity Markets. ............cssssssssssssessres 28

Even If the Court Determines That EPA

Should Have Considered Costs in

Deciding to Regulate, the Court Should

Affirm Because FPA Determined That

Benefits Massively Exceed Costs. .................00000. 33

Petitioners Misrepresent EPA’s

Cost-Benefit Findings. ......................0000+- 34

The Rule Should Be Sustained

Because EPA Has Determined

That Its Benefits Exceed Its

PIII cic ce nes cegieaemusiaekiecmmedninandidaimmumédaeaceienitens 39

& A Remand Would’ Be

Pointless, But Would Inflict

Further Competitive Harm

on Generators That Have

Already Invested in

POMUIOTI COMET OL, ...cccccececccccsccccecees 39

iv

2. NMA’s Argument That

EPA Must Separately

Evaluate the Costs and

Benefits of Regulating Each

Hazardous Air Pollutant Is

Contrary to Statutory Text. ........ 41

epee MIU YS elicited alata esis 44

Vv

TABLE OF AUTHORITIES

CASES

American Textile Manufacturers Institute,

Inc. v. Donovan, 452 U.S. 490 (1981) ................... 19

Baltimore Gas & Electric Co. v. Natural

Resources Defense Council, Inc., 462 U.S.

I RD ccicerseecactceraieahiecearcneei a tan 31

Center for Biological Diversity v. NHTSA,

Fe fly Le en 37

Chevron U.S.A. Inc. v. Natural Resources

Defense Council, Inc., 467 U.S. 843 (1984).... 16, 18

Competitive Enterprise Institute v. NHTSA,

SS FS GRE CIA. Cir, Bi )insccaccnccsesescesnccssscseseses 37

Entergy Corp. v. Riverkeeper, Inc., 556 U.S.

SE IUITTEIITED scrslcisiscnonesionnhiaictiindensslestapisitininibaiaiesi 18, 19, 35

EPA v. National Crushed Stone Ass’n, 449

a ee ieaeiciasinsonscincinsecietcantencindoonaceiesiniiciessiniasiinisiannns 29

Glover v. United States, 531 U.S. 198 (2001)............. 31

Morgan Stanley Capital Group Inc. v. Public

Utility District Number 1, 554 U.S. 527

EE SE ES ane 39-40

National Lime Ass’n v. EPA, 233 F.3d 625

Saas a: BEE csiaihnalacariinissiinseaiastneabiaiidioeiiimeiiibinanianaiinian 44

NLRB v. Wyman-Gordon Co., 394 U.S. 759

TOIT siicniuciinitieihasihnuiaiatibacsinietsiansiubgiiaincaiaiii 40, 41

Natural Resources Defense Council, Inc. v.

EPA, 824 F.2d 1146 (D.C. Cir. 1987)

vi

New Jersey v. EPA, 517 F.3d 574 (D.C. Cir.

RENEE EELS aN Se ee Oe ee 3, 7

Robinson v. Shell Oil Co., 519 U.S. 337 (1997).......... 20

Russello v. United States, 464 U.S. 16 (1983) .......... 20

SEC v. Chenery Corp., 318 U.S. 80 (1943) ................ 40

STATUTES

Sy See ET ciniciitistiniadinisnidscnsetaiiinaniaiimlcdant 2

I See I iinet 2, 20, 42

ee era, IN aici htiacectiieccelieiniacsininccbenatbiasebaiieh 20

A ae a UIE Saniscisnsticiaisetiptinedeitcniciatolennnctieaiiddiicn 3

Oe ts I iieiicichiniciccitanseinssninasietbnnsiiinaieniss 3, 7

Be Ree, Se iaiiiiinticcideiuitieninlinteoicnincenenisibacciainciscaalaidaee 3

Oe Sas eI citric sansciniiiapieiiksptitcseedinsenaieticasia 3, 43

Oe a: SE iiccdicsndiniisnsincisininasiincsiareiasineianiaiin 3, 20, 26

a a iiss siciicnieicieiiiiintitia c 3

ee Ba: I siinieiicibisdisininiscescicniniaiicbcinitcanianetui 25

Ue ie CD icstaiseiinieittiticinieisierininicisiinns 20

EE aac I cesincitinnncusictianciscssiitian 20

I ees SE seitaistctistnci ectiivnsiniscisniiainiaa 24, 33

ke eR LS eee 20, 23

1 Sr as IIE cceuticnsnincishasstnisdainiontsiiateguanieiieamaiiigie 24

Oe a IE TP ciciccntsintinsesissrsenicnonsiosniaieeteidtasieintinanteiiiedi 27

Oe eee TD snticsincnsiseciictepncttiheenitinisiiniiinesnintiainn 27

Sank HIND ninnnicirtiensinsinicdcidinaisonnseionenititin 27

Vii

4B UTES. FIESTA sccccccnccceccstccvesicceness 4, 18, 21, 43

Pg at |: | 4,21

ee I Sra ttietaneeiieniunvtsionsinisniiiniaiencinavaiéa 24

ON cesareintinticennnivinsenscnnnicnninanamnsciniatn 5

Clean Air Act Amendments, Pub. L. No. 101-

BED, 106 Beak. BESS CRISS) nnccccccceccsccscccescscsescsssecsescsess 4

LEGISLATIVE MATERIALS

S. Rep. No. 101-228 (1989), reprinted in 1990

ease tes TANUD cinbeiinciionaiasiceiispcnicnbnieisiieaiinessiiiat 2, 43

OTHER AUTHORITIES

40 C.F.R. Part 68, Subpt. DDODDD .............scccccesssereceee 2

65 Fed. Reg. 79,825 (Dec. 20, 2000) ............:ccescceeeees 6, 7

70 Fed. Reg. 15,994 (Mar. 29, 2005)............cccccseeesseeees 7

76 Fed. Reg. 3821 (Jan. 18, 2011) .............sccceccscsseseecees 40

76 Fed. Reg. 24,976 (May 3, 2011)............ 11, 29, 35, 42

77 Fed. Reg. 9304 (Feb. 16, 2012) ..................0.. passim

EPA, Guidelines for Preparing Economic

Analyses (Dec. 17, 2010) (updated May

SU Uiici asieathiencssecaiabaiebsicicasepeiciniel siciacaianiasierialeicebiingasonisceling 36

EPA, Regulatory Impact Analysis for the

Final Mercury and Air Toxics Standards

(Dec. 2011), http//www.epa.gov/

ttnecas1/regdata/RIAs/matsriafinal.pdf ............. 30

EPA, Study of Hazardous Air Pollutant

Emissions from Electric Utility Steam

Generating Units - Final Report to

Congress, Vol. I (Feb. 1998), available at

http://www.epa.gov/ttn/atw/combust/utilto

ISO/RTO Council, The Value of Independent

Regional Grid Operators (Nov. 2005),

http//www.nyiso.com/public/webdocs/med

ia_room/press_releases/2005/isortowhitepa

per_fimall1112005.pdf...........ccccvssssssssesssssssseesenen

Mich. Admin. Code 336.2502a (2013) ..vcc..s:.ccsssese-e

Mich. Admin. Code 336.2503 (2009)..........csssssseseees

E.J. Mishan & Euston Quah, Cost Benefit

Analysis 4-7 (5th ed. 2007) ...ccccocssssessecsessssneeeeee:

Michael Niven & Neil Powell, Coal unit

retirements, conversions continue to

sweep through power sector, SNL

Financial (Oct. 14, 2014),

http://www.snl.com/InteractiveX/Article.a

Spx ?edid=A-29431641-13357.........cccssecseeeeseeee

OMB, Circular A-4 (Sept. 17, 2003) ..........ccccceeeseees

U.S. Energy Information Administration,

Today in Energy, AEO2014 Projects More

Coal-Fired Power Plant Retirements by

2016 Than Have Been Scheduled (Feb. 14,

2014), available at http://www.eia.gov/

todayinenergy/detail.cfm?7id=15081 ...................

sibs 8

Industry Respondents are engaged in the electric

generation business. Together they represent 80

gigawatts of generation capacity, enough to power 60

million homes, using coal, oil, gas, nuclear, wind, solar,

and other energy sources. Industry Respondents’

experience gives them significant insight into the

practical operation of the Clean Air Act’s regulatory

scheme and its consequences for competitive electricity

markets.

The Mercury and Air Toxics Standards (“the Rule”)

imposed by EPA are economically practicable and have

already been achieved by a large portion of the power

sector. Industry Respondents, along with many other

market participants, have invested billions in installing

emissions controls and developing state-of-the-art,

highly efficient, low- or zero-emissions electric

generation units. Yet, until the Rule takes effect, such

plants will continue to be competitively disadvantaged

relative to old, high-emitting facilities that do not bear

the cost of controlling emissions of hazardous air

pollutants, thereby discouraging further investments to

modernize the Nation’s generation fleet.

STATEMENT

A. Regulation of Hazardous Air Pollutants From

Power Plants.

Large power plants are by far the largest source of

mercury and certain other hazardous air pollutants. 77

Fed. Reg. 9304, 9310 (Feb. 16, 2012) (“Final Rule”).

Yet until the Rule takes effect on April 15, 2015—

nearly 25 years after Section 112 was adopted in its

2

current form—there will be no federal requirement

that power plants limit the hazardous air pollutants

they emit. By contrast, EPA has imposed stringent

regulation on source categories emitting far smaller

amounts of such pollutants, including industrial boilers

that are technologically similar to power plants, but are

smaller or sell less of their output to a utility

distribution system. The Rule changes that, by

requiring all coal- and oil-fired power plants to match

the emissions limitations already achieved in practice

by their best-performing competitors.

3 Section 112’s Regulatory Scheme.

Congress amended Section 112 in 1990 in response

to EPA’s failure to aggressively regulate hazardous air

pollutants. Michigan Pet. App. 8a-9a (“Pet. App.”).

Congress had previously permitted EPA to “list”

pollutants for regulation, but in two decades, EPA had

listed only seven. S. Rep. No. 101-228, at 128 (1989),

reprinted in 1990 U.S.C.C.A.N. 3385, 3513.

In 1990, Congress itself designated 189 pollutants

for which it required EPA to develop emission

standards on an expedited schedule. 42 U.S.C.

$7412(b); see Pet. App. 9a-10a. Congress gave EPA one

year to “list” categories of sources that emit those

' See 42 U.S.C. §7412(aX8) (defining “electric utility steam

generating unit”); 40 C.F.R. Part 63, Subpt. DDDDD (national

emission standards for hazardous air pollutants from industrial,

commercial, and institutional boilers and process heaters).

3

pollutants above certain quantities. §7412(c)(1). EPA

was then required to regulate those sources. §7412(d).

Congress did not permit EPA to consider cost in its

initial listing analysis. §7412(c)(1). Similarly, once a

source category is listed, Congress permitted EPA to

“delist” it only if EPA determines that “no source in

the category” emits hazardous pollutants at levels that

threaten public health; cost’ is __ irrelevant.

§7412(c)(9)(B); New Jersey v. EPA, 517 F.3d 574, 581-82

(D.C. Cir. 2008).

EPA must set emission standards for source

categories it has listed. §7412(d), (e)(1). At that stage,

Congress assured that cost would be given due weight.

It directed EPA to establish emission standards that

assure the “maximum degree of reduction” that EPA

“determines is achievable,” “taking into consideration

the cost” of such regulation, as well as other factors.

§7412(d)(2). Congress further mandated that the

minimum. standards for each source category “shall not

be less stringent than ... the average emission

limitation achieved by the best performing 12 percent

of the existing sources.” Jd. §7412(d)(3); see also Pet.

App. 9a-10a. These minimum “Floor Standards”

implicitly reflect cost considerations. Because they are

based on what the “best performing” sources in the

same category are already achieving, they necessarily

have proven to be economically practicable for those

operators. See Pet. App. 29a.

Congress required an additional threshold step

before the largest power plants would be regulated

4

under Section 112. When it amended the statute,

Congress was unsure whether other programs

applicable to power plants would have the ancillary

benefit of substantially reducing their emissions of

hazardous air pollutants. Specifically, to comply with

the Title IV Acid Rain Program (“Title IV”), which

targeted sulfur dioxide (“SO2”) and nitrogen oxides

(“NO,”) emissions from the largest coal-fired plants,

those plants could have adopted controls that would

also reduce emissions of hazardous air pollutants. See

Clean Air Act Amendments, Pub. L. No. 101-549, 104

Stat. 2399 (1990).

Thus, Congress required EPA to conduct, within

three years, a study (“the Utility Study”) “of the

hazards to public health reasonably anticipated to occur

as a result of emissions by electric utility steam

generating units of [hazardous air pollutants) after

imposition of the requirements of this chapter.”

§7412(n)(1)(A). EPA also was to consider “alternative

control strategies for emissions which may warrant

regulation under this section.” Id.” Congress directed

EPA to regulate power plants under Section 112 if it

determined that “such regulation is appropriate and

necessary after considering the results of the study.”

Id.

* Congress also directed EPA to perform a study of mercury

emissions from power plants and other sources (“the Mercury

Study”) on a longer, four-year timeframe. 42 U.S.C.

§7412(nX1XB). In contrast to the Utility Study, the Mercury

Study was to consider “the costs” of available control technologies.

Id.

5

2. The Utility Study and EPA’s

Regulatory Response.

EPA submitted the Utility Study to Congress in

1998. The Study determined that strategies

implemented by power plants to comply with Title IV

would not significantly reduce emissions of hazardous

air pollutants.

Unlike Section 112, Title IV does not impose

mandatory emission limits on individual power plants.

Instead, it uses tradable pollution allowances to give

generators an economic incentive to reduce pollution

contributing to acid rain. See 42 U.S.C. $§7651-7651o.

The owner of any power plant may choose to buy the

allowances necessary to cover its emissions, or it may

choose to reduce its emissions through various means

and sell surplus allowances to other plants. JA117;

EPA, Study of Hazardous Air Pollutant Emissions

from Electric Utility Steam Generating Units - Final

Report to Congress, Vol. I (Feb. 1998) (“Utility Study”)

at 1-3 to 1-4. Generators’ choices among these

compliance options influence their emissions of

hazardous air pollutants.

For example, complying by installing scrubbers

would also reduce emissions of hazardous air pollutants;

however, the Utility Study found that few power plants

had or would install scrubbers to comply with Title [V.

* Portions of the Utility Study are reproduced in the Joint

Appendix. The entire Utility Study is available at

http//www.epa.gov/ttn/atw/combust/utiltox/eurtcl pdf.

6

See JA117-19. Instead, the vast majority of plants

switched to low-sulfur coal or purchased emission

allowances, JA117-18, which had little or no effect on

hazardous air pollutant emissions. Utility Study at 13-

1, 13-3 to 13-7, 1446. The Study also reported that

existing NO, and particulate matter controls were not

expected to reliably reduce mercury emissions. JA94-

95; Utility Study at 13-13 to 13-15, 14-7.

The Utility Study thus concluded that “[ulJtilization

of add-on controls to comply with the acid rain program

{is} not expected to significantly impact [hazardous air

pollutant] emissions due to their limited numbers and

limited [hazardous air pollutant] contro) efficiency

improvement.” JA106-07.

In 2000, after considering the Utility Study, EPA

concluded that it was “appropriate and necessary” to

regulate emissions from power plants. 65 Fed. Reg.

79,825, 79,827-28 (Dec. 20, 2000) (“2000 Finding”). As

EPA explained, such units “are the largest source of

mercury emissions in the U.S.” Jd. at 79,827. Because

“[mjercury is highly’ toxic, persistent, and

bioaccumulates in food chains,” it poses health risks to

humans, and, in particular, to developing fetuses. Jd. at

79,827-29. EPA also identified other metal and acid gas

emissions from power plants—including arsenic,

chromium, nickel, cadmium, dioxins, hydrogen chloride,

and hydrogen fluoride—that were “of potential

concern” because of their health effects. Jd. at 79,827.

Given the health and environmental risks posed by

power plant emissions, and the existence of several

7

options to control them, EPA determined that it was

“appropriate” to regulate coal- and oil-fired plants. 65

Fed. Reg. at 79,830. Such regulation was “necessary”

because implementation of the Act’s other provisions

would “not adequately address” the hazards EPA had

identified. Id.*

In 2005, the agency reversed itself, removing coal-

and oil-fired power plants from the list of source

categories. 70 Fed. Reg. 15,994, 15,994 (Mar. 29, 2005)

(“2005 Revision”). The D.C. Circuit held that decision

unlawful because EPA had failed to find that “no

source in the category” emitted pollutants at levels

that threatened public health. New Jersey, 517 F.3d at

582; see §7412(c)(9)(B). Nonetheless, EPA waited until

2011 to regulate emissions of hazardous air pollutants

by power plants. Meanwhile, large power plants—the

largest source of such emissions—faced no federal

requirement to reduce them.

B. The Economics of Pollution Control in

Wholesale Electricity Markets.

The negative impact of EPA’s regulatory vacuum

has been exacerbated by the economics of wholesale

electricity markets. All power plants are connected to

the nationwide network of electric transmission lines

commonly referred to as the “grid,” where electricity is

transmitted across multi-staie regions to satisfy

. By contrast, EPA declined to regulate natural gas-fired power

plants because emissions of hazardous air pollutants from such

plants were “negligible.” 65 Fed. Reg. at 79,831.

8

demand. Grid operators must balance the amount of

electricity generated with demand in real time, as the

grid has no storage capacity and a supply-demand

imbalance can overload transmission lines or yield

voltage drops that can cause massive blackouts.

In areas served by competitive wholesale electricity

markets—which include many of the Nation’s most

populated areas—grid operators decide which power

plants should be allowed to feed electricity into the grid

at any given time. Power generators offer electricity

for sale at a price that, typically, reflects their marginal

production cost. The grid operator dispatches plants in

order of their bids, selecting progressively more

expensive generation units until supply meets demand.”

The price received by each dispatched generator is

equal to the bid of the most expensive unit needed to

meet demand. This pricing scheme—in which al]

dispatched generators receive the same price—creates

a powerful incentive to reduce operating costs, as

generators with low costs can bid less, be dispatched

more often, and make greater profits when they run.

Operating pollution controls tends to increase a

generator’s marginal costs—for example, costs related

° More than 60% of the electricity supplied to the grid is delivered

through competitive wholesale electricity markets. See ISO/RTO

Council, The Value of Independent Regional Grid Operators (Nov.

2005), at 9410, http//www.nyiso.com/public/webdocs/media.

room/press_releases/20065/Asortowhitepaper_finall11112005.pdf. In

other areas, cost also dictates dispatch, though generators do not

compete to supply power at least cost.

9

to treatment chemicals, waste disposal, and power and

water consumption, in addition to capital costs—and

thus increase the amount it must bid to break even.

Consequently, “clean” generators may be underbid by

similar generators that have not installed or do not

operate controls. Moreover, plants without pollution

controls receive a windfall from participating in the

same market as those with controls: because the

market pays all plants the price needed for supply to

meet demand, plants without pollution controls receive

as profit the higher market price needed to compensate

plants operating with them. These dynamics create

powerful incentives to avoid investing in or operating

pollution controls.

Nonetheless, for several reasons, many coal-fired

plants have invested in control technology that reduces

their emissions of hazardous air pollutants.

First, fourteen states have already adopted limits

on mercury emissions, while others have required

power plants to install mercury monitoring equipment.

See JA306-20; State Resps. Br. 35-36." Some states

have also regulated additional hazardous air pollutants.

JA306-20. Many states that have imposed regulation—

including Illinois, New Jersey, and Maryland—are part

of the same regional grid as states that have not. Even

though plants in the former states suffer a competitive

disadvantage relative to “dirtier” plants in other states,

* Indeed, Petitioner Michigan adopted such a requirement, but

suspended it after EPA adopted the Rule. See Mich. Admin. Code

336.2503 (2009); Mich. Admin. Code 336.2502a (2013); JA314-15.

10

the “clean” plants have remained in business and

continue to compete in the wholesale market.

Second, the applicable federal emission standards

for conventional pollutants (such as SOQz, NO,, and

particulate matter) vary based upon a power plant’s

age. New and modified units must be equipped with

state-of-the-art controls, which may also reduce

emissions of some hazardous air pollutants. Older units

must meet the less stringent emission standards that

were in place when they were built. As a result, plants

constructed before the Clean Air Act was enacted have

few if any modern pollution controls.. Nevertheless,

despite the additional operating costs new plants incur

to meet conventional emission standards, newer units

have remained in the market in part because they are

more efficient than older ones.

Third, many generation owners have anticipated

the Rule and other standards and invested in pollution

control technologies to satisfy them. Since 2008,

Exelon (including Constellation Energy) spent

approximately $1.4 billion to install advanced emission

control technologies on coal-fired generation that it

owned outright or jointly with others. Other companies

likewise have expended significant sums. For example,

between 2001 and 2011, Duke Energy spent $5 billion

retrofitting its existing units with emissions controls,

with the goal of installing scrubbers on 90 percent of its

coal generation fleet. PPL Generation reported in 2011

" In 2007-2008, 59% of the Nation’s coal-fired units, and 34% of the

coal-generation capacity, was over 40 years old. JA579-80.

11

that 96 percent of its competitive coal generation is

scrubbed. JA649-50.

Nevertheless, the economic logic of the wholesale

electricity markets discourages the oldest and least

efficient plants from investing in and operating

pollution controls unless they are required to do so,

and, all else equal, makes pollution more profitable than

pollution control—despite the negative health and

environmental effects that pollution imposes on society.

C. The Rule.

In 2011, EPA proposed to reaffirm its 2000 Finding

that regulation of power plants was “appropriate and

necessary.” EPA reevaluated the record and

considered new data collected after complete

implementation of Title IV. EPA again found that

“(utilities are by far the largest remaining source of

[mercury] in the U.S.,” and are the “largest source” of

several additional hazardous air pollutants. 76 Fed.

Reg. 24,976, 24,999 (May 3, 2011) (“NPRM”). Because

those emissions continue “to pose a hazard to public

health and to the environment,” and because

technologies are available to control them, EPA

proposed to find that it is “appropriate and necessary”

to regulate coal- and oil-fired power plants. Id. In the

same NPRM, EPA proposed emission standards for

these plants. Jd. at 25,026-28.

Following public comment, EPA adopted the Final

Rule. EPA confirmed that it is “appropriate” to

regulate power plants under Section 112 because their

emissions of hazardous air pollutants cause hazards to

12

public health and the environment, and pollution

controls are available. 77 Fed. Reg. at 9310-11. EPA

further concluded that it is “necessary” to regulate

power plants because other requirements of the Act

had not adequately addressed the problem. Jd.

EPA did not consider the costs of regulation in

making the “appropriate and necessary” determination.

77 Fed. Reg. at 9326-27. EPA observed that the Act

does not require consideration of costs at the listing

stage. Id. In addition, EPA explained that Congress

did not require it to consider costs in other listing

decisions, and that nothing in Section 112(n)(1)(A)

suggests a departure from that approach. /d.

Although EPA was not required to consider costs

under Section 112(n)(1)(A), Executive Orders 12866

and 13563 required it to conduct a cost-benefit analysis

when it established emission standards. EPA did so in

both its proposed and final Regulatory Impact

Analyses (“RIA”). Those analyses were performed

using “methods and assumptions consistent with the

state-of-the-science for human health impact

assessment, economics and air quality analysis.” 77

Fed. Reg. at 9432. The Final RIA determined that the

benefits of the Rule would outweigh its costs by a ratio

of at least 3:1, and potentially by 9:1. Jd. at 9306.

D. The Decision Below.

Several groups challenged the Rule in the D.C.

Circuit, which upheld it in its entirety.

13

As relevant here, State, Industry, ard Labor

Petitioners challenged EPA’s interpretation of the

“appropriate and necessary” standard, contending that

the word “appropriate” required EPA to consider the

costs of regulation. Pet. App. 23a-24a. The D.C.

Circuit majority disagreed. The court explained that

“Tojn its face, § 112(n){1)(A) neither requires EPA to

consider costs nor prohibits EPA from doing so.” Jd. at

25a. The term “appropriate” is “open-ended,”

“ambiguous,” and “inherently context-dependent.” Jd.

at 26a (quotation marks omitted).

Moreover, EPA’s construction of that ambiguous

term was reasonable: in making the “appropriate and

necessary” determination, “Congress directed EPA’s

attention to the conclusions of the study regarding

public health hazards from [power plant] emissions.”

Id. In the remainder of Section 112, “Congress

mentioned costs explicitly where it intended EPA to

consider them.” Jd. Congress’s failure to do so in

Section 112(n)(1)(A) demonstrated that EPA was not

required to consider costs in deciding whether to

regulate power plant emissions—only in deciding how

stringently to do so. Indeed, the majority emphasized,

Petitioners could not “point to a single case” requiring

“EPA to consider costs where the [Clean Air Act] does

not expressly so instruct.” Jd. at 27a-28a.

The majority also rejected the dissent’s view that

Congress would not have “authorized EPA to regulate

without any consideration of regulatory cost.” Jd. at

28a. That argument rested “on a false premise”

because EPA did consider costs, both directly and

14

indirectly, in setting emission standards under Section

112(d). Jd. at 28a-29a. In setting Floor Standards,

EPA required all power plants to meet the emissions

reductions already achieved by their best-performing

peers, thus “leveling the playing field” by requiring

“uncontrolled plants to install and operate technology”

that their competitors were already using in an

economically viable manner. /d. at 32a (quoting

Industry Respondents’ brief). Finally, the majority

pointed out that, contrary to the dissent’s doomsday

predictions, EPA found that “the benefits of th[e] rule

outweigh its*costs by between 3 to 1 or 9 to 1.” Id. at

$2a-33a (quoting 77 Fed. Reg. at 9306).

SUMMARY OF ARGUMENT

The D.C. Circuit did not err in upholding the Rule.

In Section 112(n)(1)(A), Congress asked EPA to decide

whether to proceed with regulation of emissions of

hazardous air pollutants by power plants, applying an

open-ended “appropriate and necessary” standard.

That standard on its face gives EPA discretion to focus

on the health and environmental harms of such

emissions at this initial stage, and to defer

consideration of cost to the later stage when EPA sets

specific emission standards supported by a full cost-

benefit analysis.

Unable to identify any express statutory language

foreclosing EPA’s position, Petitioners argue that

making an initial decision to regulate without

considering cost is so irrational that it falls outside the

range of discretion granted by Congress. That

15

argument fails because EPA simply applied to power

plants the identical regulatory approach that Congress

mandated for every other source of hazardous air

pollutants. Impatient with EPA’s slow pace of

regulation, Congress required in Section 112(c) and

112(d) that EPA establish emission standards for every

other significant source of hazardous air pollutants.

Congress directed EPA to “list” these sources for

regulation without regard to cost. Congress then

expressly directed EPA to consider cost later, in

setting emission standards.

EPA did not act irrationally in applying Congress’s

own chosen regulatory structure to power plants. EPA

conducted the Utility Study as Congress directed. It

concluded that other Clean Air Act programs would not

sufficiently reduce power plants’ emissions of

hazardous air pollutants. And it found that control

technologies were available. Thus, EPA decided that

regulation was “appropriate and necessary.”

That determination was particularly apt given the

nature of competitive electricity markets in which

generators submit bids to sell their electricity based

upon their operating costs. Absent federal regulation,

these markets perversely subsidize uncontrolled plants

and place cleaner plants at a competitive disadvantage.

EPA’s interpretation of the “appropriate and

necessary” standard thus deserves Chevron deference.

Petitioners have not come close to satisfying the heavy

burden of showing that EPA acted unreasonably.

16

The argument for reversal is especially weak

because EPA in fact considered costs and benefits as

part of the process of promulgating the final emission

standards. Using the cost-benefit methodology

mandated by law, it concluded that the Rule’s benefits

would greatly outweigh its costs. Although EPA did

not believe these analyses were legally relevant to its

initial decision to regulate the emission of hazardous air

pollutants by power plants, there is no doubt what the

agency would conclude if it had. For these reasons,

even if EPA misinterpreted the Act (and it did not), the

record provides all the facts and analysis required for

affirmance on grounds of harmless error. A pointless

remand would only further delay regulation that

Congress authorized 25 years ago, and would

perpetuate the competitive advantage enjoyed by

polluting plants in the wholesale electricity markets.

ARGUMENT

I. EPA Permissibly Chose to Regulate Power

Plants in the Same Manner That Congress

Mandated for All Other Sources of Hazardous

Air Pollutants.

In challenging EPA’s interpretation of the statutory

phrase “appropriate and necessary,” Petitioners face a

heavy burden: under Chevron U.S.A., Inc. v. Natural

Resources Defense Council, Inc., 467 U.S. 843 (1984),

they must demonstrate that the statute unambiguously

forecloses EPA’s interpretation. Thus, they strain to

read the word “appropriate” as clearly requiring EPA

to consider costs when deciding whether to regulate,

17

arguing that Congress could not have intended EPA to

make that choice based on public health and other

hazards alone.

That argument founders because with respect to all

other sources—including industrial boilers

technologically similar to large power plants—

Congress made the express legislative choice that

Petitioners condemn as irrational: it decided to

regulate based upon public health and other hazards

alone. It directed EPA to consider costs only in setting

the level of regulation, structuring that second stage so

that minimum standards would be based on the

emission levels that the best-performing comparable

sources already had achieved and requiring EPA to

consider cost expressly before regulating more

stringently than these Floor Standards. Nothing in

Section 112(n)(1(A) prohibits EPA from regulating

power plants in the same manner.

A. Section 112(n)(1)(A) Does Not Unambiguously

Mandate Consideration of Costs.

As Petitioners acknowledge, Michigan Br. 23;

National Mining Association (“NMA”) Br. 21; Utility

Air Regulatory Group (“UARG”) Br. 25, in directing

EPA to regulate power plant emissions of hazardous

air pollutants if “appropriate and necessary,” Congress

set an open-ended standard. Certainly there is no

* Petitioners contend that the breadth of the word “appropriate”

requires EPA to consider all “relevant factors.” E.g., Michigan Br.

93. But even assuming that is true, EPA reasonably determined

that costs were not “relevant” in making the threshold decision

18

“express statutory requirement that the Agency

consider costs in making the appropriate

determination.” 77 Fed. Reg. at 9326-27. Thus,

because “the statute is silent or ambiguous” as to

whether costs must be considered in deciding whether

to regulate, this Court will uphold EPA’s interpretation

so long as it constitutes “a permissible construction of

the statute.” Chevron, 467 U.S. at 843; see also, e.g.,

Entergy Corp. v. Riverkeeper, Inc., 556 U.S. 208, 218

(2009) (to receive Chevron deference, agency

interpretation need not be “the only possible

interpretation, nor even the interpretation deemed

most reasonable by the courts”).

EPA’s construction is reasonable. Congress

directed EPA to decide whether to regulate power

plants “after considering the results of’ the Utility

Study. §7412(n)(1)A). That Study, in turn, focused on

hazards to public health remaining after the

implementation of other Clean Air Act requirements,

as well as availability of alternative control

technologies. Congress did not mandate consideration

of costs in the Study.

After considering the Study’s results, EPA

concluded that regulation of coal- and oil-fired plants is

“appropriate” because their emissions of hazardous air

pollutants “pose hazards to public health” and the

environment, and because “effective controls are

available to reduce” both mercury and non-mercury

whether to regulate, given the statutory context— including that

Congress itself did not consider costs in listing other sources.

19

emissions. 77 Fed. Reg. at 9310; see id. at 9311. (EPA

found that regulation of natural gas-fired units is not

“appropriate,” because those units’ emissions did not

pose health or environmental hazards. See supra n. 4.)

Regulation of coal- and oil-fired units is “necessary,”

EPA continued, because those “hazards to public

health” and the environment “are _ reasonably

anticipated to remain after imposition of the

requirements of the [Clean Air Act].” 77 Fed. Reg. at

9311; see id. at 9363.

Because nothing in Section 112(n)(1)(A) requires

EPA to consider costs, its decision to focus on health

and environmental factors, as well as the availability of

control technology, is reasonable. See, e.g., Entergy,

556 U.S. at 222 (“It is eminently reasonable to conclude

that [statutory] silence is meant to convey nothing

more than a refusal to tie the agency’s hands as to

whether cost-benefit analysis should be used, and if so

to what degree.”); Am. Textile Mfrs. Institute, Inc. v.

Donovan, 452 U.S. 490, 510 (1981) (“When Congress

has intended that an agency engage in cost-benefit

analysis, it has clearly indicated such intent on the face

of the statute.”).

B. Surrounding Provisions Confirm That EPA’s

Interpretation Is Reasonable.

The provisions neighboring Section 112(n)(1)(A), as

well as the structure of Section 112 generally, confirm

that the word “appropriate” can reasonably be

construed to focus on public health and environmental

harms, and not to encompass consideration of costs.

20

See, e.g., Robinson v. Shell Oil Co., 519 U.S. 337, 341

(1997) (“The plainness or ambiguity of statutory

language” accounts for “the specific context in which

that language is used, and the broader context of the

statute as a whole.”).

As the D.C. Circuit explained, “(tJhroughout § 112,

Congress mentioned costs explicitly where it intended

EPA to consider them.” Pet. App. 26a. Congress’s

failure to address costs in Section 112(n)(1)(A) thus

indicates that it did not require EPA to consider them.

E.g., Russello v. United States, 464 U.S. 16, 23 (1983).

Specifically, EPA may add pollutants or sources to

be regulated without regard to cost. See §7412(b), (c).

Yet the subsections directing EPA to set emission

standards for listed source categories do expressly

require consideration of cost. Those provisions

address, inter alia, emission standards for new or

existing sources (§7412(d)(2)), emission standards and

work practice regulations for coke ovens and coke oven

batteries (§7412(d)(8)(A)Gi), (B)i)), and emission

standards for addressing the remaining environmental

risks after implementation of the statutory provisions

(§7412(f)(1)(A)). EPA permissibly construed

Subsection 112(n)(1(A}—a iisting provision—

consistent with that general structure.

Moreover, in stark contrast to Congress’s omission

of any reference to costs when describing the Utility

Study in Section 112(n)(1)(A), the very next subsection

requires EPA to conduct the Mercury Study

“consider{ing],” among other things, “the costs of

21

[control] technologies.” §7412(n)(1)(B). Yet while

Congress required EPA to complete the Utility Study

“within 3 years after November 15, 1990,” the Mercury

Study was to be completed “not later than 4 years”

from the same date. §7412(n)(1)(A)}(B). Thus,

Congress plainly understood that EPA might make the

“appropriate and necessary” determination before the

results of the Mercury Study—including its

consideration of costs—were available.”

Petitioners argue that unless the phrase

“appropriate and necessary” includes consideration of

cost, it lacks independent meaning. Not so. Congress

required the Utility Study because it did not know

whether power plant emissions of hazardous air

pollutants would continue to cause significant health

harms, notwithstanding implementation of the other

programs included in the 1990 Amendments. It then

told EPA to consider the results of the study and to act

if those results made regulation “appropriate and

* NMA contends that because the Mercury Study directed

consideration of both “environmental effects” and costs, EPA

erred in considering the former, but not the latter, in making the

appropriate and necessary” determination. NMA Br. 32-33. But

the D.C. Circuit upheld EPA’s conclusion that it could consider the

environmental hazards posed by power plant emissions when

deciding whether to regulate, Pet. App. 35a-37a, and that holding

is outside the Question Presented. In any event, by requiring

EPA to consider the results of the Utility Study, Congress did not

preclude EPA from also taking into account other factors, such as

environmental hazards. See 77 Fed. Reg. at 9825. Thus, while

EPA could have chosen not to consider environmental effects—or

to consider costs—the statute does not require either of those

interpretations.

22

necessary.” There is nothing mysterious (or

superfluous) about that statutory standard when read

in context.

Nor are Petitioners correct in contending that on

EPA’s view, “appropriate” and “necessary” are

redundant. E.g., Michigan Br. 34. Indeed, Petitioners

can make that argument only by misrepresenting

EPA’s analysis. The Petitioner States argue that EPA

“accounted for the existence of health hazards through

its finding that regulation was ‘necessary.” Id. But as

previously noted, EPA determined that regulation was

“necessary” because “the hazards to public health” and

the environment from power plant emissions “are

reasonably anticipated to remain after imposition of the

requirements of the CAA”—a factor that EPA’s

“appropriateness” determination did not consider. 77

Fed. Reg. at 9311 (emphasis added); see id. at 9363.

Similarly, EPA’s “appropriateness” analysis rested in

part on the availability of control technology—a

consideration not relevant to EPA’s determination that

regulation of power plants was “necessary.” Jd.

UARG further contends that EPA must consider

costs in making the “appropriate and necessary”

determination because “[cJonsideration of cost-benefit

relationships is especially relevant for emission

standards that address residual emissions and risks.”

UARG Br. 28. On UARG’s view, because regulation

under Section 112(n)(1)(A) addresses “smaller

increments of emissions” that remain after the

implementation of other provisions, consideration of

costs is particularly “appropriate.” /d.

23

UARG’s argument ffails because Section

112(n)(1)(A) is not a residual-risk provision. Emissions

of hazardous air pollutants by power plants are not

regulated by federal law outside of the Rule enacted

pursuant to Section 112(n)(1)(A). Thus,

Section 112(n1)(A) provides the primary method of

regulating those emissions. Principles regarding

“residual” risk are irrelevant.

In fact, Section 112(n)(1)(A) contrasts sharply with

the statute’s actual residual-risk provisions, which

address sources and emissions already regulated by

Section 112. Section 112(f)(1) requires that EPA study

the “risk to public health remaining, or likely to remain,

from sources subject to regulation under this section

after the application of standards under subsection

(d).” §7412(f)(1)(A) (emphasis added). Thus, unlike

Section 112(n)(1)(A), Section 112(f) focuses on sources

that are already subject to regulation, and asks

whether more must be done to reduce the risks from

those sources.

Further, even in setting residual risk standards

under Section 112(f)(2), Congress provided only that

EPA promulgate standards if necessary either to

provide an “ample margin of safety to protect public

° Because Title IV’s Acid Rain program does not target mercury

and other hazardous air pollutants, NMA is wrong to contend that

EPA’s interpretation of Section 112(n)(1A) makes that provision

the “tail that .. wag{s] the dog{]” of Title IV. NMA Br. 27-31.

Regulation of hazardous air pollutant emissions by power plants

under Section 112(n\(1XA) is not the “tail” to Title [V’s “dog”—.it

is a different animal altogether.

24

health,” or “to prevent, taking into consideration costs

.. and other relevant factors, an adverse

environmental effect.” §7412(f)(2(A). EPA may not

consider costs in making “an initial determination of

what is ‘safe,”” because cost has “no relevance” to that

question. Natural Resources Defense Council v. EPA,

824 F.2d 1146, 116465 (D.C. Cir. 1987) (en banc)

(addressing prior version of section 112(f)). Nothing in

Congress’s treatment of this residual risk scenario

suggests that Congress intended to require EPA to

consider costs under Section 112(n)(1)(A) before it

imposed any emission standards. See also §7412(n)(2)

(requiring a study and _ recommendations for

“economically viable control technologies” to “reduce

residual risks” from coke ovens “after implementation

of the standard under subsection (d) of this section”).

C. Deferring Consideration of Costs Until the

Standard-Setting Phase Is Not Illogical.

Deciding whether to regulate based upon health and

environmental hazards, and then considering cost when

setting emission standards, makes practical sense.

Merely deciding to regulate does not impose any costs

in the abstract; costs result from the actual imposition

of emission standards. Those costs will vary

" Section 112’s judicial review mechanism underscores that EPA’s

decision to regulate power plants is merely a preliminary step that

does not by itself affect substantive rights. The statute provides

that listing decisions are not final agency action subject to review.

§7412(eX4). Instead, review may be had “when the Administrator

issues emission standards” for a category of sources, id.—

standards that will reflect cost considerations.

25

depending on what standards EPA promulgates. That

decision, in turn, requires a detailed understanding of

what emission reductions the best performers in the

source category have already achieved, what

equipment will be required to meet those limits, and

what impacts on health and the environment will result.

Yet Petitioners would require EPA to undertake those

analyses before making the threshold determination

whether power plant emissions of hazardous air

pollutants should be regulated at all. Petitioners put

the cart before the horse.

Petitioners nonetheless contend that EPA’s choice

not to consider costs at the first step of the analysis was

unreasonable because “ignoring costs is an irrational

way to regulate.” Michigan Br. 30. That argument

aims at a straw man. EPA did not ignore costs in

promulgating the Rule; it accounted for them in

numerous ways in deciding what emission standards to

impose.

Most notably, the statutory method for setting

emission standards considers costs both directly and

indirectly. As noted above, Floor Standards generally

are determined by taking the “average emissiori

limitation achieved by the best performing 12 percent

of the existing sources.” §7412(d)(8)(A). Thus,

minimum standards are based on what other, similar

sources already have achieved in practice—a test that

26

necessarily ensures that standards will not impose

industry-wrecking costs.~

To set standards more stringent than the Floor

Standards, EPA must expressly “consider{] the cost of

achieving such emission reduction.” §7412(d)(2). Here,

EPA generally decided not to impose standards more

stringent than the Floor Standards. See 77 Fed. Reg.

at 9367, 98369 (EPA adopted a standard more stringent

than the Floor Standard for mercury from one

subcategory of existing coal-fired plants, but this

standard is less stringent than the mercury emission

standard for all other existing coal-fired plants).

EPA also has numerous other ways to ensure that

compliance is practicable, and it used many of them

here.

For example, EPA permitted existing contiguous,

commonly-controlled power plants in the same

subcategory to demonstrate compliance with emission

standards by averaging their emissions, rather than

meeting the requirements on an individual basis. See

77 Fed. Reg. at 9384-85, 9473-76; Pet. App. 59a-64a

(upholding EPA’s use of averaging). As EPA

* Petitioners at times argue that Section 112(dX3) is entirely

insensitive to costs. See, e.g., NMA Br. 34-35. But if that were

true, it would only highlight the irrationality of Petitioners’

position. The D.C. Circuit found it “improbable” that Congress

intended to force an “all-or-nothing” choice, requiring EPA to

consider costs when making the on-or-off determination whether

to regulate, but prohibiting EPA from considering costs as a basis

for relaxing the Floor Standards. Pet. App. 29a.

27

explained, averaging “can provide sources. the

flexibility to comply in the least costly manner while

still maintaining a regulation that is workable and

enforceable.” 77 Fed. Reg. at 9385.

EPA also controlled costs by establishing work

practice standards, which are qualitative standards

typically less costly to achieve and to monitor than

numerical standards. See §7412(h)(1). In the Rule,

EPA adopted several work practice standards “in lieu

of numeric emission standards” for certain units and

pollutants. 77 Fed. Reg. 9401; see id. at 9369, 9438.

In addition, to ease the costs of compliance, EPA

adopted a three-year compliance period, the longest

initial timeframe permitted by the statute. 77 Fed.

Reg. at 9407-11; see §7412(i)(3)(A). EPA further

suggested that “a fourth year for compliance” should be

permitted “in a broad range of situations.” 77 Fed.

Reg. at 9407; see id. at 9410; JA876-77; §7412(iX3)(B).

Given these other methods for accommodating costs,

Petitioners are simply wrong to contend that EPA was

“cost blind,” e.g., UARG Br. 24, or that EPA regulated

with a “deliberate indifference to the regulation’s cost,”

Michigan Br. 20. EPA’s decision to regulate emissions

of hazardous air pollutants by power plants using the

same two-stage approach that Congress mandated for

all other sources was entirely reasonable.”

* Petitioners claim EPA’s statutory construction could justify

“regulations costing $1 trillion even if the benefit was a mere $1.”

NMA Br. 2. But the concern expressed in that hypothetical is

28

D. EPA’s Approach Is Particularly Suitable

Given the Highly Competitive Nature of

Wholesale Electricity Markets.

EPA’s regulatory approach—setting a regulatory

floor based upon the emissions reductions that have

already been demonstrated to be achievable by the

best-performing plants in the source category—is

particularly appropriate given the nature of

competitive wholesale electricity markets. See supra

at 7-11.

As previously explained, the wholesale electricity

markets’ use of an auction mechanism to determine

which units are dispatched and what price they receive

strongly disincentivizes investment and use of pollution

control technology in the absence of regulation. Yet

despite these market dynamics, EPA determined that,

as of 2010, 69 existing coal-fired power plants—or 27

percent of those that reported data to EPA—already

met all of the final existing source emission limits. 77

Fed. Reg. at 9387; JA875. Other data in the record

indicate that substantially more units had already

installed the necessary controls.“

entirely fanciful, because EPA must find sufficient health harms

for regulation to be “appropriate and necessary,” and the features

of Subsection (d), discussed supra, protect against over-regulation.

™ Nearly 60 percent of all coal-fired boilers that submitted stack

test data to EPA regarding mercury emissions were already

meeting the mercury emissions standard. About 70 percent of all

coal-fired boilers that submitted such data regarding particulate

matter and acid gas emissions were already meeting the standards

29

That so many plants have installed the necessary

controls—and have remained in business despite the

cost advantage that polluting plants enjoy—completely

undercuts the notion that the Rule will financially ruin

the industry. To the contrary, the Rule simply requires

that the remaining coal-fired plants meet the

performance already achieved by many of their peers.”

Indeed, until now, these “dirty” units have

benefited from a regulatory framework that has

effectively subsidized power generation by high-

polluting facilities. All else being equal, these plants

have been able to underbid cleaner plants and thereby

be called for dispatch more frequently than they

otherwise would have been. Yet they still receive the

governing one or both of those pollutants. JA627-29. EPA

additionally noted that it “agree{d] with the findings of ...

independent studies” that “over 50 percent of the fleet is equipped

with scrubbers and the number will increase to nearly 2/3 by

2015.” 77 Fed. Reg. at 9417. As noted above, operating a scrubber

will typically allow a coal-fired plant to satisfy the emissions

standards for acid gases. 76 Fed. Reg. at 25,023, 25,038-40.

* NMA argues that “to the extent” Floor Standards “implicitly

consider{]’ cost, “it cannot be assumed that all units ... can bear

the cost of new controls simply because” the best-performing units

have done so. NMA Br. 34 n.16. But nothing in the Clean Air

Act—or in administrative law more generally—requires EPA to

set emission standards that are economically practicable for each

and every source in the category. Cf. EPA v. Nat'l Crushed Stone

Ass’n, 449 U.S. 64, 73-78 (1980) (rejecting argument that EPA

must provide “variances from otherwise valid regulations where

dischargers cannot afford normal] costs of compliance” because

such variances “would undermine” Clean Water Act’s purpose of

“reducfing] the total pollution produced by an industry”).

30

market-clearing price set by the highest-priced power

generator needed to meet demand. By requiring all

plants to meet minimum emission standards

demonstrated to be achievable by peers that have

remained competitive, the Rule levels the playing field

and eliminates the perverse incentives that previously

allowed coal-fired plants to profit by refusing to install

the same emissions controls used by their peers.

To be sure, some coal-fired plants may choose to

retire rather than make ‘the capital expenditures

required to comply with the Rule. But the plants most

likely to retire are very old—between 40 and 60

years—and nearing the end of their useful lives in any

event. JA546, JA579-80; see also EPA, Regulatory

Impact Analysis for the Final Mercury and Air Toxics

Standards 3-17 (Dec. 2011),

http://www.epa.gov/ttnecas1/regdata/R1As/matsriafina

l.pdf (average retired plant expected to be 52 years

old). Moreover, these plants face other economic

pressures much more significant than this rulemaking

in determining whether they retire. Most important is

the declining price of natural gas, which has allowed

natural gas plants to submit lower bids for dispatch.

See 77 Fed. Reg. at 9407. This has resulted in lower

electricity prices and less frequent dispatch for coal

plants, with the result that coal plants on balance have

greater difficulty covering their costs.

EPA carefully considered whether the Rule would

adversely affect electric reliability, and, based on

abundant record evidence, concluded that it would not.

Using modeling that “has been extensively reviewed

ol

and has been utilized in several rulemakings affecting

the power generation sector over the last 15 years,” 77

Fed. Reg. at 9413, EPA predicted that only 4.7

gigawatts of coal-fired plants would retire as a result of

the Rule, as opposed to other factors such as lower

natural gas prices and greater energy efficiency. /d.;

id. at 9407. The 4.7 gigawatts of expected retirements

amount to less than one-half of one percent of total

generating capacity in the United States, and less than

1.5% of total U.S. coal capacity. Jd. at 9407-08.

Petitioners nonetheless suggest that the Rule “is

forcing numerous plants into retirement” and will

“contribute” to the retirement of 54 gigawatts of coal-

fired generation capacity. NMA Br. 27; UARG Br. 20-

21. Yet in the Court of Appeals, “Petitioners [did] not

challenge [EPA’s] conclusion” that the Rule would

cause the early retirement of less than 2 percent of U.S.

coal-fired capacity. Pet. App. 32a; 77 Fed. Reg. at 9416;

see id. at 9408. They therefore cannot do so here. £.g.,

Glover v. United States, 531 U.S. 198, 205 (2001). And

even if they could, in considering EPA’s predictive

judgment about the effects of the Rule—a technical

judgment within its area of expertise—this Court’s

review would be “at its most deferential.” Baltimore

Gas & Elec. Co. v. Natural Resources Defense Council,

Inc., 462 U.S. 87, 103 (1983). That deference would be

well-warranted: the report on which Petitioners

principally rely, see UARG Br. 21 n.8; NMA Br. 15,

confirms that greater numbers of coal plant

retirements are expected because those plants have

been “under significant economic pressure in recent

32

years because of low natural gas prices and slow

electricity demand growth.”

The limited retirements anticipated to result from

the Rule will leave grid operators with substantial

capacity reserve margins, particularly after accounting

for anticipated new investments in_ electrical

generation. The industry has anticipated the Rule for

many years, and participants have made numerous

investments that will result in a more modern and

environmentally friendly fleet. Some retiring coal

plants will be re-powered as natural gas-fired plants.”

For example, after acquiring in 2010 two primarily coal-

fired plants in New Jersey and Delaware representing

approximately 780 megawatts of capacity, Calpine

redeployed them to operate primarily on gas. Exelon

has invested billions of dollars in installing emissions

controls on coal plants and in operating low- and zero-

emission generation such as nuclear plants. Similarly,

PSEG has spent more than a billion dollars installing

pollution control technologies on its coal plants. There

* US. Energy Information Administration, Today in Energy,

AEO2014 Projects More Coal-Fired Power Plant Retirements by

2016 Than Have Been Scheduled (Feb. 14, 2014), available at

http://www.eia.gov/todayinenergy/detail.cfm?id=15031.

17

As of October 2014, nearly 11 gigawatts of coal-to-gas-

conversions had been announced or were under construction, more

than twice the 4.7 gigawatts EPA projected to retire due to the

Rule. Michael Niven & Neil Powell, Coal unit retirements,

conversions continue to sweep through power sector, SNL

Financial (Oct. 14, 2014), http:/Awww.snl.com/InteractiveX/

Article.aspx?cdid=A-29431641-13357.

33

also has been phenomenal growth in the number of new

natural gas-fired plants. Between 1999 and 2008, the

electric sector added almost 270 gigawatts of natural

gas-fired generating capacity. JA558. Moreover, gas-

fired plants still have relatively low utilization rates

compared to coal plants because for many years, coal

has been cheaper than gas, see, e.g., JA564-66, meaning

that significant quantities of generation could be

switched from coal-fired plants to gas-fired plants

without any adverse consequences for electric

reliability.

In sum, the record amply supports EPA’s

conclusion that the Rule is economically practicable,

and will have no adverse impact on electric reliability.

II. Even If the Court Determines That EPA

Should Have Considered Costs in Deciding to

Regulate, the Court Should Affirm Because

EPA Determined That Benefits Massively

Exceed Costs.

Even assuming EPA was required to consider costs

when determining whether to regulate hazardous air

pollutant emissions by power plants, the Rule should

still be affirmed. Under Subsection 112(e)(4), a listing

decision is not itself final agency action that should be

reviewed in isolation; instead, the decision to regulate

may be reviewed only “when the Administrator issues

emission standards for such pollutant or category,”

§7412(e)(4), at which time the costs and benefits of the

action will be known.

34

Here, EPA determined that the Rule’s benefits

massively exceed its cost: the Rule’s “annual quantified

net benefits (the difference between benefits and costs)

are $27 to $80 billion using a 3 percent discount rate or

$24 to $71 billion using a 7 percent discount rate.” 77

Fed. Reg. at 9306. In addition, EPA identified many

non-monetized benefits, including various health and

ecosystem effects. Jd. Because EPA has already

determined that the benefits of regulating power plants

exceed the costs, a remand would serve only to further

delay regulation that is already many years overdue

and to harm industry participants that have already

invested in pollution reduction.

A. Petitioners Misrepresent EPA’s Cost-Benefit

Findings.

Petitioners repeatedly claim that “the quantified

costs of hazardous air pollutant regulation are more

than one-thousand times greater than the quantified

benefits: $9.6 billion versus $4 million to $6 million.”

UARG Br. 3; see also id. at 19, 23; Michigan Br. 13, 32,

47; NMA Br. 1, 2, 14, 19, 25. This constant refrain flatly

misrepresents the record. As a matter of fact—made

clear in black and white on the Rule’s opening pages—

EPA found that the benefits outweigh the costs by a

ratio of at least 3:1, and as much as 9:1. 77 Fed. Reg. at

9306. Moreover, EPA explained that many of the

benefits of reducing hazardous air pollutants cannot

easily be quantified, and thus are not accounted for by

these already lopsided ratios. See id. (EPA “could not

monetize some costs and important benefits, such as

some [mercury] benefits” and those for other hazardous

35

air pollutants, but “[u}pon considering these limitations

and uncertainties, it remains clear that the benefits of

this rule ... are substantial and far outweigh the

costs.”). Thus, contrary to Petitioners’ efforts to

obfuscate, this is simply not a case where EPA has

required plants to “spend billions to save one more fish

or plankton.” Entergy, 556 U.S. at 232-33 (Breyer, J.,

concurring) (quotation marks omitted).

To be sure, most of the quantified benefits identified

by EPA are “co-benefits”—that is, benefits resulting

from the Rule that do not arise exclusively from

reducing emissions of hazardous air pollutants.

Petitioners simply pretend that these co-benefits do not

exist. But they are real benefits from the Rule and

cannot be ignored when weighing its benefits against

its costs—just as ancillary or indirect costs cannot be

ignored, either.

Indeed, the many benefits of regulating hazardous

air pollutants are inextricably connected with the

benefits of regulating certain “conventional” pollutants.

For example, some fine particulate matter consists in

part of non-mercury metals. Thus, controlling

emissions of non-mercury metals can also reduce

emissions of particulate matter, and vice-versa. 77 Fed.

Reg. at 9420; 76 Fed. Reg. at 25,038-40; JA105-06;

JA110; Utility Study at 2-10, 2-13, 13-22. Similarly,

limiting hazardous acid gas emissions may also limit

SO, emissions. 76 Fed. Reg. at 25,038-40, 25,050-51.

Petitioners provide no rationale or authority for

ignoring such co-benefits when conducting a cost-

36

benefit analysis. In fact, EPA’s consideration of co-

benefits, as well as ancillary costs, was entirely proper

and consistent with established Office of Management

and Budget (“OMB”) directives and peer-reviewed

EPA guidance. The relevant OMB directive,

promulgated in 2003, explains that an agency’s

analysis should look beyond the direct benefits

and direct costs of [its] rulemaking and consider

any important ancillary benefits and

countervailing risks. An ancillary benefit is a

favorable impact of the rule that is typically

unrelated or secondary to the statutory purpose

of the rulemaking (e.g., reduced refinery

emissions due to more stringent fuel economy

standards for light trucks) while a countervailing

risk is an adverse economic, health, safety, or

environmental consequence that occurs due to a

rule and that is not already accounted for in the

direct cost of the rule (e.g., adverse safety

impacts from more stringent fuel-economy

standards for light trucks).

OMB, Circular A-4 at 26 (Sept. 17, 2003). EPA’s

internal guidelines likewise make clear that “[a]n

economic analysis of regulatory or policy options should

present all identifiable costs and benefits that are

incremental to the regulation or policy under

consideration. These should include directly intended

effects and associated costs, as well as ancillary (or co-)

benefits and costs.” EPA, Guidelines for Preparing

Economic Analyses at 11-2 (Dec. 17, 2010) (updated

May 2014) (emphasis added). Thus, contrary to

37

Petitioners’ insinuations, there was nothing

aberrational or irregular about EPA’s consideration of

co-benefits when weighing the benefits of the Rule

against its costs.

Petitioners do not challenge as arbitrary and

capricious the basic principle that an agency, when

weighing costs and benefits, should consider all the

costs and benefits of regulatory action. The notion that

the agency should count only some benefits, and

presumably only some costs, would have far-reaching

consequences and would be contrary to well-

established economic principles” and case law. See,

e.g., Competitive Enter. Inst. v. NHTSA, 956 F 2d 321,

327 (D.C. Cir. 1992) (reversing agency for failing to

consider whether benefits of fuel economy standards

outweigh ancillary costs in terms of lives lost due to

smaller vehicles); Ctr. for Biological Diversity v.

NHTSA, 538 F.3d 1172, 1198-1203 (9th Cir. 2008)

(reversing agency for quantifying ancillary costs of fuel

economy standards (the impact on vehicle sales and

employment) but not quantifying ancillary

environmental benefits).

a

Petitioners also do not dispute that EPA faithfully

implemented the applicable OMB and internal agency

guidance regarding how to conduct cost-benefit

analysis. Indeed, as EPA acknowledged, studies

submitted by commenters, including Exelon, suggested

that EPA’s estimate of benefits was, if anything,

" See, e.g., EJ. Mishan & Euston Quah, Cost Benefit Analysis 4-7,

104 (5th ed. 2007).

38

conservative. 77 Fed. Reg. at 9415; JA689."" EPA also

considered comments that the Rule would result in

indirect economic costs due to reduced employment and

higher electricity prices, but it found that any job losses

were likely to be more than offset by increased

employment in “manufacturing steel, cement and other

materials needed to build pollution control equipment,”

as well as “jobs creating and assembling pollution

control equipment, and jobs installing the equipment at

power plants.” 77 Fed. Reg. at 9414. EPA further

noted “{pljotential job increases from increased output

by lower-emitting facilities.” Jd. As for electricity

prices, EPA found that prices are likely to increase

only around 3 percent, and “the downstream economic

effects” of any such increase is likely “to be small

because electricity is only a small factor in the

production of most goods and services.” Jd.

Petitioners do not challenge these findings as

unsupported by substantial evidence. Instead, they

simply ignore the findings that EPA actually made,

claiming that they are irrelevant because EPA stated

that it did not rely on these co-benefits in deciding that

regulation was “appropriate and necessary.” NMA Br.

17, 41-42; Michigan Br. 47-48. Of course EPA did not

consider these co-benefits when deciding whether to

"Ina peer review of EPA’s analysis for the proposed rule, Dr.

Charles Cicchetti, an economist, concluded that if one were to

consider benefits that EPA had not attempted to monetize as well

as positive impacts on the Natior.’s economy, the Rule would

create $52.5 to $139.5 billion in annual net benefits. JA691; 77 Fed.

Reg. at 9415.

39

regulate. That is because, under EPA’s view of the

statute, the decision whether to regulate should turn on

whether hazardous air pollutant emissions from power

plants were causing public health and environmental

harms. EPA concluded that they were, and had no

cause to inquire further at that stage. But if Section

112(n)(1)(A) had required a cost-benefit analysis as a

predicate to the decision to regulate power plants, then

it would have been irrationa] for EPA to consider only

some of the benefits of regulation—just as it would

have been irrational for EPA to consider only some of

the costs. Thus, there is no basis for Petitioners to

treat tens of billions of dollars of co-benefits resulting

from the Rule as though they do not exist.

B. The Rule Should Be Sustained Because EPA

Has Determined That Its Benefits Exceed Its

Costs.

1. A Remand Would Be Pointless, But

Would Inflict Further Competitive

Harm on Generators That Have

Already Invested in Pollution Control.

In light of EPA’s actual finding that the benefits of

the Rule significantly exceed its costs, this Court

should affirm the Rule even if it determines that EPA

should have considered costs when deciding whether

regulation was “appropriate.”

A remand for EPA to make the very cost-benefit

determination that it already has made in the Rule

would be “an idle and useless formality.” Morgan

Stanley Capital Grp. Inc. v. Public Util. Dist. No. 1,

40

554 U.S. 527, 545 (2008) (quoting NLRB v. Wyman-

Gordon Co., 394 U.S. 759, 766-67 n.6 (1969) (plurality

op.)). EPA received voluminous comments regarding

its proposed cost-benefit analysis, and it considered and

responded to them. 77 Fed. Reg. at 9313-19, 9413-19,

9424-39; e.g., JA885-903. EPA not only found that the

benefits justified the costs, see supra, but it needed to

so find in order to promulgate the Final Rule. Under

Executive Order 13563, an agency may “propose or

adopt a regulation only upon a reasoned determination

that its benefits justify its costs.” 76 Fed. Reg. 3821,

3821 (Jan. 18, 2011) (emphasis added).

Affirming the Rule on the basis of EPA’s cost-

benefit analysis, rather than remanding, would be

consistent with SEC v. Chenery Corp., 318 U.S. 80

(1943). There, the Court explained that the principles

governing judicial review of administrative agency

decisions are akin to those governing judicial review of

jury verdicts. Although a reviewing court is not

permitted to substitute its own policy judgments or

factual findings for the agency’s, it need not remand a

case to an agency merely to reinstate a factual finding

that the agency has already made. Instead, the

agency's decision—like a decision of a lower court or a

jury—“must be affirmed if the result is correct

‘although the lower court relied upon a wrong ground

or gave a wrong reason.” Id. at 88 (quoting Helvering

v. Gowran, 302 U.S. 238, 245 (1937)). In short,

“Chenery does not require that we convert judicial

review of agency action into a ping-pong game.” NLRB

v. Wyman-Gordon Co., 394 U.S. 759, 766-67 n.6 (1969)

(plurality op.).

4}

Given the administrative record before the Court,

“{tJhere is not the slightest uncertainty as to the

outcome” of any further cost consideration this Court

might direct EPA to undertake. Jd. It would therefore

“be meaningless to remand.” Jd. In fact, a remand

would accomplish nothing but further delay, frustrating

Congress’s purpose to expedite regulation of hazardous

air pollutant emissions. That delay also would

perpetuate the distortion of wholesale electric markets

to favor dirtier plants at the expense of cleaner ones.

And it would chill the industry from making further

investments that will modernize America’s generation

fleet and expand the natural gas _ pipeline

infrastructure, as companies are reluctant to build

replacement, cleaner generation or to modernize

existing plants when their competitors are not required

to comply with modern emission standards. See JA425-

27.

2. NMA’s Argument That EPA Must

Separately Evaluate the Costs and

Benefits of Regulating Each Hazardous

Air Pollutant Is Contrary to Statutory

Text.

Finally, NMA contends not only that EPA was

required to consider costs and benefits, but that it

needed to separately evaluate those costs and benefits

for each pollutant, and failed to do so for acid gases.

NMA Br. 38-39 (arguing that EPA failed to identify

health hazards or environmental impacts from acid gas

emissions). Even were this Court to hold that EPA

should have considered cost in making the “appropriate

42

and necessary” determination, NMA’s argument would

provide no basis for a remand.

NMA’s contention, which was unanimously rejected

by the D.C. Circuit, falls outside this Court’s grant of

certiorari. It also is contradicted by the record. EPA

noted the significant health benefits produced by

control of acid gases. 76 Fed Reg. at 25,050-51.

Moreover, as NMA grudgingly acknowledges, EPA

relied upon a study of hydrochloric acid deposition in

the United Kingdom, which showed that hydrochloric

acid can be a significant driver of acidification. 77 Fed.

Reg. at 9362; NMA Br. 39. NMA asserts that this

study is somehow irrelevant because it was conducted

in the United Kingdom, and that EPA could not

identify an example in which “domestic electric

generator hydrochloric acid emissions have affected

acid deposition.” NMA Br. 39. NMA provides no

reason to believe that the laws of chemistry and

atmospheric science are any different in the United

Kingdom than in the United States, and EPA was

plainly within its discretion to credit that study.

In any event, the statute squarely forecloses NMA’s

notion that EPA must separately make an “appropriate

and necessary” finding for each individual pollutant.

Cf. NMA Br. 42-44. Section 112 regulates sources, and,

for covered sources, requires that emission standards

be established for the pollutants listed in Subsection

(b). §7412(b). Thus, Section 112(n)(1)(A) states that

EPA “shall regulate electric utility steam generating

units under this section” if it finds such regulation

43

“appropriate and necessary.” §7412(n)(1)(A) (emphasis

added).

“(Tjhis section,” id., in turn, requires that EPA

“promulgate regulations’ establishing emission

standards for each category or subcategory of major

sources and area sources” of the pollutants that

Congress has listed. §7412(d)(1) (emphasis added).

While the statute goes on to give EPA discretion to

“distinguish among classes, types, and sizes of sources

within a category or subcategory in establishing such

standards,” id., it never suggests that EPA must, or

even may, distinguish among pollutants, and set

emission standards for some pollutants but not for

others.

The legislative history confirms that Congress

intended for EPA to promulgate, for covered sources,

emission standards for all pollutants listed in

Subsection (b). Congress amended Section 112 in 1990

in part because EPA had failed to regulate enough

pollutants under the prior provisions. The Senate

Report, for example, explained that “(ijn 18 years, EPA

has regulated cnly some sources of only seven

chemicals ... The legislation reported by the

Committee would entirely restructure the existing law,

so that toxics might be adequately regulated by the

Federal Government.” S. Rep. No. 101-228, at 128,

reprinted in 1990 U.S.C.C.A.N. at 3513. Congress

therefore added the specific list of pollutants for which,

for each covered source, emission standards must be

set. In sum, the D.C. Circuit correctly and unanimously

rejected “[t]he notion that EPA must ‘pick and choose’

44

among [pollutants] in order to regulate only those

substances it deems most harmful.” Pet. App. 42a; see

Natl Lime Ass’n v. EPA, 233 F.3d 625, 633-34 (D.C.

Cir. 2000).

CONCLUSION

For the foregoing reasons, the decision below

should be affirmed.

February 25, 2015 Respectfully submitted,

Brendan K. Collins Paul M. Smith

Counsel of Record Matthew E. Price

Robert B. McKinstry, Jr. Erica L. Ross

Ronald M. Varnum JENNER & BLOCK LLP

Lorene L. Boudreau 1099 New York Ave. NW

BALLARD SPAHR LLP Suite 900

1735 Market Street Washington, DC 20001

51st Floor (202) 639-6000

Philadelphia, PA 19103 psmith@jenner.com

(215) 665-8500

collins@ballardspahr.com

Counsel for Industry Respondents