Opposition Brief — Alabama Power Co. v. Thomas

Pate cS f=)

ee tee WV

Nos. 87-2068, 38-60, and 88-61

In the Supreme Court of the Gnit

OCTOBER TERM, 1988

ALABAMA POWER CO., ET AL., PETITIONERS

Vv.

LEE M. THOMAS, ET AL.

OHIO POWER CO. AND ORMET CoRrP., ET AL., PETITIONERS

Vv.

LEE M. THOMAS, ET AL.

NATIONAL COAL ASSOCIATION AND ALABAMA POWER CoO.,

ET AL., PETITIONERS

Vv.

NATURAL RESOURCES DEFENSE COUNCIL, ET AL.

ON PETITIONS FOR A WRIT OF CERTIORARI

TO THE UNITED STATES COURT OF APPEALS

FOR THE DISTRICT OF COLUMBIA CIRCUIT

BRIEF FOR THE FEDERAL RESPONDENTS IN OPPOSITION

DONALD B. AYER

Acting Solicitor General

ROGER J. MARZULLA

Assistant Attorney General

LISA F. RYAN

Altorney

Depariment of Justice

Washington, D.C. 20530

(202) 633-2217

QUESTIONS PRESENTED

1. Whether the Environmental Protection Agency

(EPA) violated the notice requirement of former Section 4

of the Administrative Procedure Act (APA), 5 U.S.C.

553, when it issued its final rule in these cases (No.

87-2068).

2. Whether the EPA could lawfully require presump-

tive compliance with a uniform standard without a finding

that the standard is attainable by most persons subject to

it, where the EPA allows an exception whenever the stand-

ard is not attainable (No. 87-2068).

3. Whether the EPA engaged in unlawful retroactive

rulemaking by issuing a regulation that applies to future

emissions of existing pollution sources (No. 88-60).

4. Whether the court of appeals properly refused to

affirm the EPA’s decision that certain pre-October 1983

stack height increases to levels below the heights specified

by “good engineering practice” formulas need not be

justified by demonstrations that they are needed to avoid

excessive localized pollution (No. 88-61).

5. Whether the court of appeals violated the doctrines

of res judicata and law of the case in requiring the EPA to

reconsider its treatment of certain stack height increases

(No. 88-61).

(I)

eed

TABLE OF CONTENTS

Page

es cc ke iisina scene css taees |

EE a ge ga Wa ts vies ov de sna ee ks oe 2

aya gos oe 6 sa oss soa s ence vnese 2

ne ee 2

a 5

CC se sea swe ete edsaescess 12

EE ES 16

te Vd ca eg ais vie wesc sees seenes 27

TABLE OF AUTHORITIES

Cases:

American Methyl Corp. v. EPA, 749 F.2d 826 (1984),

vacated, 768 F.2d 385 (D.C. Cir. 1985), cert. denied,

Ee 22-23

BASF Wyandotte Corp. v. Costle, 598 F.2d 637 (ist Cir.

1979), cert. denied, 444 U.S. 1096 (1980) .... 2.2002... 16

Chocolate Mfrs. Ass’n v. Block, 755 F.2d 1098 (4th Cir.

ER ee 16, 18

Citizens to Save Spencer County v. EPA, 600 F.2d 844

re a ae 21

Connecticul Light & Power Co. v. NRC, 673 F.2d 525

(D.C. Cir.), cert. denied, 459 U.S. 835 (1982) ......... 16

BE. 1. du Pont de Nemours & Co. v. Train, 541 &.2d

1018 (4th Cir. 1976) atfd in part and rev’d in part, 430

dea we dada ces vcoek es 18, 19, 20

Georgetown University Hospital \. Bowen, 821 .2d 750

(D.C. Cir. 1987), cert. granted, No. 87-1097 (Feb 29,

a a vs a ca ec vas’ 14, 20, 22

International Harvester Co. \. Ruckelshaus, 478 &.2d 615

ep ys bn Gacsuussc veces 16

Sierra Club v. EPA, 719 F.2d 436 (D.C. Cir. 1983), cert.

rn wae ee .. Bae CISE4).................5. 7, 8, 25, 26

Utah Intl, Inc. \. Andrus, 488 F. Supp. 976 (D. Colo.

ea gy cabelas iavicsev cacvers tz 23

IV

Statutes and regulations: Page

Administrative Procedure Act, 5 U.S.C. 551 ef seq.:

TEs i Ee so ee cad ask ckne x etka nn canahe 12, 16

5 U.S.C. 553(b)(3) SEAR. . cede wake nuasseskebe wes 5 16

Clear Air Amendments of 1970 42 U.S.C. (& Supp. IV)

Pe OE a eee a hes eR RR Gh e es Teeaese 2

SO ae a PU ro baka esas cca adeeeye as 6

DSRS Dh, Gee We PIED 3c ww vccseunevass ,

Be St EY ome ee 2

6 10GtaMED, 42 U.S.C. TORREOE) . oo cc ccc sees 3

§ 110(a)(2)(A), 42 U.S.C. 7410(a)(2)(A) ... 2.2... .-. 3

§ 110(a)(2)(H), 42 U.S.C. 7410{a)(2)(H) «2.2... 3, 22

Se Ge ees PIE cin raw cee sens eens 23

BU we gt f ooh”) nr paren ayree 3

SEG tr. MUI sk doses wad ae anda eees 3

Gee ee es ss Sink oes eee cae 2, 4, 5, 6,

13, 14, 17, 22, 25

SR: Gee een, ia aie aaa he eee 4

Bee "We ge thy | 2”. BAe re meee 4

pie hth Se oie Of, Ere ee 6

GS Fes ee och bk ee EAs ee ne a ee 6

ee, et aarcsd teas co poeewnteescinwetenaus 6

Miscellaneous:

Se FG. Te, BNO ok cw osc ks back vale ae 6405 hueek 5

OE FO, eh, BOOTIE ow coc cnn eed ch seedeenceas 5

Se Pe: SAE 6 kc Scan katoi beundcrceyess 6

ae ED spin cn a be ag aa Sone awn ness 6

47 Fea. Ree. SOGB-SOGP CI9GZ) wc ccs cca ncae 6

tek Ge Pee eee 10, 23

47 Fed. Reg. SO6B-SBGP (19GZ) .... we ceases 6

ee. es I ROE 8 sc kw ae ee ean 8

i : : SO ON sds os ee ees ate 8

H.R. Rep. 95-294, 95th Cong., Ist Sess. (1977) ........5, 14, 18

Jn the Supreme Court of the Anited States

OCTOBER TERM, 1988

No. 87-2068

ALABAMA POWER CoO., ET AL., PETITIONERS

V.

LEE M. THOMAS, ET AL.

No. 88-60

OHiO POWER CO. AND ORMET CorpP., ET AL., PETITIONERS

V.

LEE M. THOMAS, ET AL.

No. 88-61

’ NATIONAL COAL ASSOCIATION AND ALABAMA POWER Co.,

ET AL., PETITIONERS

v.

NATURAL RESOURCES DEFENSE COUNCIL, ET AL.

ON PETITIONS FOR A WRIT OF CERTIORARI

TO THE UNITED STATES COURT OF APPEALS

FOR THE DISTRICT OF COLUMBIA CIRCUIT

BRIEF FOR THE FEDERAL RESPONDENTS IN OPPOSITION

OPINION BELOW

The opinion of the court of appeals (Pet. App. la-64a)!

is reported at 838 F.2d 1224.

'“Pet. App.” refers to the appendix to the petition in No. 87-2068.

(1)

ee

JURISDICTION

The judgment of the court of appeals was entered on

January 22, 1988. Petitions for rehearing and suggestions

of rehearing en banc were denied on April 13, 1988 (Pet.

App. 65a-68a). The petition for a writ of certiorari in No.

87-2068 was filed on June 17, 1988, and those in Nos.

88-60 and 88-61 were filed on July 12, 1988. The jurisdic-

tion of this Court is invoked under 28 U.S.C. 1254(1).

STATEMENT

Since enactment of the Clean Air Amendments of 1970

(the Act) (42 U.S.C. (& Supp. IV) 7401 ef seg.), Congress

and the United States Environmental Protection Agency

(EPA) have sought to prevent persons who are subject to

emission control obligations under the Act from using

dispersion of air pollutants as a means of meeting those

obligations. Section 123 of the Act, 42 U.S.C. 7423, which

was added in 1977, attempts to limit the dispersion of air

pollutants by tall stacks. These petitions challenge the

decision of the United States Court of Appeals for the

District of Columbia Circuit affirming in part and re-

manding in part the EPA’s most recent regulations that

were issued to implement Section 123.

A. Statutory Background. |. The basic tools set forth in

the Act for reducing and controlling air pollution are the

national ambient air quality standards (NAAQS).? After

the EPA sets the standards for various pollutants, the

States must adopt plans that provide for the “implementa-

tion, maintenance, and enforcement” of such standards

? The ambient air quality standards are classified as either primary

or secondary. Primary standards are those necessary to protect public

health. 42 U.S.C. 7409%(b)(1). Secondary standards are those necessary

to protect the public welfare from any adverse effects that are

associated with an air pollutant. 42 U.S.C. 7409(b)(2)

3

within each air quality control region within the State. 42

U.S.C. 7410(a)(1). Those state implementation plans

(SIPs) specify emission limitations and other measures

that are necessary to implement the standards, including

control requirements for existing pollution sources. Those

requirements are supplemented by new source perform-

ance standards (NSPS), which are issued by the EPA Ad-

ministrator under Section 111 of the Act, 42 U.S.C. 7411.

NSPS apply both to new facilities and to major modifica-

tions of existing facilities, and they reflect “the degree of

emission limitation and the percentage reduction

achievable through application of the best technological

system of continuous emission reduction.” 42 U.S.C.

741 1(a).

State implementation plans, which are submitted to the

EPA for approval, must include emission limitations for

sources located within the State, timetables for complying

with such limitations, and provisions for any other

measures that are necessary to ensure that the applicable

air quality standards are attained and maintained. 42

U.S.C. 7410(a)2)(A). State plans must be revised

periodically to take account of revisions of the ambient air

quality standards and the availability of improved or more

expeditious methods of achieving the standard. The EPA

may also require changes if the Administrator finds that

the existing plan is substantially inadequate to achieve the

standards or to ensure compliance with the requirements

of the 1977 Amendments to the Act. 42 U.S.C.

7410(a)(2)(H).

The ambient air quality standards limit local, ground-

level concentrations of pollutants. Many regulated

sources, however, emit air pollutants through stacks or

chimneys. In the absence of any restrictions on their use,

taller stacks can be used to disperse pollutants over a wider

is |

4

area, thus permitting a utility or other source to reduce

local ground-level pollution concentrations without ac-

tually reducing emissions.

2. Perceiving that regulated sources were opting to

raise the height of stacks rather than installing emission

controls, Congress added a new Section 123 in 1977. 42

U.S.C. 7423. That Section, which applies to all stacks con-

structed after December 31, 1970, places restrictions on

the use of tall stacks and other dispersion techniques to at-

tain applicable air quality standards. Specifically, Section

123(a) provides that a source cannot receive credit, in

calculating its emission limitation, for dispersion from a

stack whose height exceeds that prescribed by “good

engineering practice.” 42 U.S.C. 7423(a).

Under the statute, “good engineering practice” means

the height that is necessary to ensure that emissions from

the stack do not result in “excessive” ground-level concen-

trations of pollutants in the immediate vicinity of the

source “as a result of atmospheric downwash, eddies and

wakes which may be created by the source itself, nearby

structures Or nearby terrain obstacles (as determined by

the Administrator).” 42 U.S.C. 7423(c). That height is

presumed not to exceed two and one-half times the height

of the source (2.5H) (42 U.S.C. 7423(c)), which is general-

ly referred to as “formula height” (see, e.g., Pet. App.

10a). Under the statute, however, a source may qualify for

greater than formula height if its owner or operaior

demonstrates to the satisfaction of the EPA that a greater

height is necessary to avoid “excessive concentration” in

the immediate vicinity. 42 U.S.C. 7423(c).

To make the specified showing of need to exceed for-

mula height, an owner or Operator generally conducts a

fluid modeling demonstration, in which the effects of the

source, nearby structures and terrain, and atmospheric

conditions are simulated in a wind tunnel. In that way, it is

possible to predict the effect of a source on air quality

prior to construction of the stack. Where the required

demonstration is made, the statute allows the EPA to ap-

prove credits for stack heights above formula height. As

the House Report accompanying the 1977 legislation

states, however, “the latitude given the Administrator to

allow full credit for such stack height will be exercised with

circumspection and utmost caution in those rare circum-

stances proven to justify its use.” H.R. Rep. 95-294, 95th

Cong., Ist Sess. 93 (1977).

B. Regulatory History. 1. The EPA first proposed

regulations under Section 123 on January 12, 1979. Find-

ing that the statutory formula of 2.5H overestimated

“good engineering practice” stack height for some sources,

the Agency proposed a second and more restrictive for-

mula for prospective application based on the height of

the source plus one and one-half the lesser of the height or

width (H + 1.5L). 44 Fed. Reg. 2610, 2614. A source seek-

ing greater stack height credit than accorded under the ap-

plicable formulas could demonstrate, through the use of

an actual field study or a fluid model, that greater height

was necessary “to ensure that emissions from the stack do

not result in excessive concentrations of any air pollutant

in the immediate vicinity of the source” (id. at 2614). The

regulation defined “excessive concentrations” generally to

mean concentrations that (a) are at least 40% more than

the conceritrations that would exist if there were no

downwash, wakes, or eddy effects caused by nearby struc-

tures or terrain (/bid.) and (b) exceed either the ambient air

> The Report also recognizes that the need for extra stack height is

most pronounced for sources that are located in hilly or rugged ter-

rain. The Report states: “[i]t is the expectation of this committee that

persons responsible for siting new facilities will not locate them next to

terrain features which will produce such downwash.” H.R. Rep.

95-294, 95th Cong., Ist Sess. 93 (1977).

iia i

6

quality standards or, in the case of sources subject to the

Prevention of Significant Deterioration (PSD) program

(40 C.F.R. 51.24 (1985); 40 C.F.R. 52.21), the permitted

PSD increment.‘

In response to comments on the 1979 proposed regula-

tions, the EPA issued a revised proposal on October 7,

1981. That proposal, among other things, modified the

definition of “excessive concentrations” to delete the re-

quirement that a source demonstrate that an ambient air

quality standard or PSD increment was exceeded before

the source could qualify for stack height credit above that

provided by the formula. 46 Fed. Reg. 49816. Under the

revised proposal, a source was required only to

demonstrate a 40% increase in pollutant concentrations in

order to qualify for greater than formula credit.

The EPA promulgated final regulations under Section

123 on February 8, 1982. 47 Fed. Reg. 5864. Those regula-

tions adopted the agency’s refined good-engineering-

practice formula of H+1.5L, but they allowed sources

that commenced construction of stacks prior to January

12, 1979, »e date the formula was first proposed, to retain

credit based on the more lenient 2.5H formula. /d. at

5866, 5868. The final regulations also retained the 1981

definition for excessive concentrations. Jd. at 5868-5869.

Neither the 1982 final regulation, the 1979 proposal, nor

the 1981 proposal specified any emission level to be used in

carrying out the Section 123 fluid modeling demonstra-

tions.

+ PSD increments apply in areas that have attained the air quality |

standards established by Section 109 of the Act, 42 U.S.C. 7409. See

42 U.S.C. 7471-7479. They set the amount of increased emissions per-

mitted in such areas and are aimed at limiting the air quality deteriora-

tion in such an area even if the standard itself may not be threatened.

2. The United States Court of Appeals for the District

of Columbia Circuit affirmed in part, reversed in part,

and remanded for the EPA to reconsider certain portions

of the regulations. Sierra Club v. EPA, 719 F.2d 436

(1983), cert. denied, 468 U.S. 1204 (1984). No party to

that litigation challenged the EPA’s adoption of the re-

fined H+1.5L formula. Sierra Club, the Natural

Resources Defense Council (NRDC), and the Common-

wealth of Pennsylvania, however, challenged the agency’s

decision to permit sources that commenced construction

of their stacks prior to the date the new formula was pro-

posed to retain credit based on 2.5H. The court generally

affirmed the agency’s decision to “grandfather” that

category of sources, but it remanded the regulation to the

agency with directions to limit the grandfathering benefit

to sources that actually relied on the 2.5H formula in con-

structing their stacks. 719 F.2d at 468.

Petitioners in Sierra Club also challenged the agency’s

decision not to require any demonstration of need —that

is, of the need for a particular stack height in order to

avoid excessive localized pollution—by owners or

operators of sources that increased the height of existing

stacks up to, but not exceeding, formula height. The court

found that the EPA had not adequately evaluated the

presumption that existing stacks were originally built to

the height prescribed by good engineering practice. It

therefore remanded for the EPA “to reconsider whether,

in light of its new understanding of ‘excessive concentra-

tions,’ demonstrations are necessary before stack heights

may be raised, even if the final height will not exceed for-

mula height.” 719 F.2d at 459-460.

Finally, the court remanded for reconsideration of the

EPA’s definition of “excessive concentrations,” ruling that

the agency had erred in failing to relate the definition to

some level of air pollution that could be deemed to en-

danger health and welfare. 719 F.2d at 450. The court ex-

pressed its approval of the definition originally proposed

by the EPA in 1979, which would have required a source

owner or Operator to show an “exceedance” of an air

quality standard or PSD increment as well as a percentage

increase in pollutant concentrations before credit could be

awarded for stack height that exceeded formula height. /d.

at 468.

3. In response to the court’s remand, the EPA pub-

lished proposed regulations on November 9, 1984. 49 Fed.

Reg. 44878. The agency reformulated its definition of “ex-

cessive concentrations” as it applied to fluid modeling

demonstrations that are conducted for the purpose of

qualifying for above-formula stack height credit. In brief,

the EPA reproposed the definition set forth in its 1979

proposed regulations and required sources to demonstrate

both a 40% increase in local pollutant concentrations and

the exceedance of an ambient air quality standard or PSD

increment. Jd. at 44887.

Having reintroduced into the definition of “excessive

concentrations” the absolute levels of localized pollution,

the agency also stated for the first time its intention to

prescribe specific emission limitations to be used in per-

forming the fluid modeling demonstrations. As the court

of appeals later explained (Pet. App. 14a-15a), ground-

level concentrations are a function of both stack height

and emission levels, and therefore, in order to determine

the stack height that is needed to ensure that ground-level

concentrations do not exceed specified levels, one needs

first to specify an emission level: the higher the emission

level, the greater the stack height required for ground-level

concentrations to be reduced to a given level.’ The EPA

’ A higher assumed emission level means a greater stack height,

which means a greater credit, which in turn means a higher permitted

emission rate. Hence, to maximize the permitted emission rate one

9

proposed the emission levels to be used in the modeling as

follows (Pet. App. 150a):

[I]t will be necessary to specify an emission rate * * *

in order to determine whether a NAAQS or PSD in-

crement is being exceeded. Consequently, the Agency

will require in its technical support document that the

emission limitation be established based on either: (1)

The existing, approved emission limit; (2) any ap-

plicable technology-based emission limit, such as the

new source performance standards (NSPS); or (3) the

» emission limit that would result from the use of GEP

formula stack height, whichever is applicable to the

source being modeled.®

In response to comments, the EPA eliminated two of

the three emission levels set forth in the November 1984

notice of proposed rulemaking. The final regulation thus |

requires that a source seeking to justify extra stack height

perform its fluid modeling demonstration using the emis-

sion rate prescribed by the NSPS applicable to that source

category unless the owner or operator demonstrates that

the NSPS emission limit cannot feasibly be met. Pet. App.

96a. The agency explained (ibid. (footnote omitted)):

The EPA believes that in cases where greater than for-

mula height may be needed to prevent excessive con-

would seek to maximize the assumed emission level in the modeling;

lower assumed levels correspond with lower permitted pollution rates.

® Petitioners observe (87-2068 Pet. 6-7) that fluid modeling

demonstrations that were conducted in the late 1970s and early 1980s

used the sources’ actual or allowable emission levels in calculating ex-

pected ground-level pollution concentrations from particular stack

heights. Nevertheless, the 1982 final regulation, the 1979 proposal,

and the 1981 technical guidelines cited by petitioners all were silent on

whether such use was proper. The issue was not addressed by the EPA

until November 1984. .

10

centrations, sources should first attempt to eliminate

such concentrations by reducing their emissions. For

this reason EPA is requiring that the emission rate to

be met by a source seeking to conduct a demonstra-

tion to justify stack height credit above the formula

be equivalent to the emission rate prescribed by NSPS

applicable to the industrial source category. In doing

this, EPA is making the presumption that this limit

can be met by all sources seeking to justify stack

heights above formula height. Sources may rebut this

presumption, establishing an alternative emission

limitation, on a case-by-case basis, by demonstrating

to the reviewing authority that the NSPS emission

limitation may not feasibly be met, given the

characteristics of the particular source.

The agency also explained that it rejected the option of

allowing a source to use its actual emission limit because

“to the extent that limit relied on greater than formula

height, it would amount to using a tall stack to justify

itself” (id. at 97a).

Several owners and operators of sources had previously

performed fluid modeling demonstrations for the purpose

of qualifying for above-formula stack height credit.’ Ac-

’ For example, Ohio Power Company, a petitioner in No. 88-60,

had completed a fluid modeling demonstration for its Kammer Power

plant in West Virginia. Ohio Power had commenced construction of

its Kammer stack in the mid-1970s. The enactment of Section 123 in

1977 jeopardized Ohio Power’s ability to receive credit for the new

stack. Accordingly, Ohio Power, in consultation with the EPA, con-

ducted a fluid modeling demonstration under the 1979 proposed

regulations. That demonstration was tentatively approved by the

agency on August 17, 1982. 47 Fed. Reg. 35784. A mid-level employee

of the EPA subsequently sent a letter indicating that the demonstra-

tion had been found consistent with the 1982 final regulations. 88-60

Pet. 12; C.A. App. 152. Although the August 17, 1982, notice had an-

Se

11

cordingly, the agency had to decide, based on the court of

appeals’ decision in Sierra Club, whether those sources

should be permitted to retain the extra credit under a

grandfathering provision or should instead be required to

perform fluid modeling demonstrations under the new

regulations. The agency decided not to grandfather that

category of sources. It explained (Pet. App. 99a):

EPA is making this part of the regulations retroactive

to December 31, 1970. In the terms of the court’s

retroactivity analysis, stacks greater than formula

height represent a situation that Congress did af-

firmatively “intend to alter” in section 123. Moreover,

EPA regulatory pronouncements since 1970 have

placed a stricter burden on sources raising stacks

above formula height than on others.

As required by the court of appeals’ remand order, the

EPA also considered whether demonstrations should be

required for owners and operators of sources who in-

creased existing stacks up to formula height. The agency

adopted a new regulation that requires fluid modeling

demonstrations in such circumstanees, but only where the

height of existing stacks was increased up to formula

height after October 11, 1983, the date of the Sierra Club

decision. Pet. App. 100a-1034, 132a. The agency ex-

plained the grandfathering of pre-Sierra Club increases by

observing that the court of appeals’ requirement of

demonstrations for within-formula stack height increases

was a Significant departure from prior agency policy and

regulatory proposals. Hence, there was little reason prior

to the court of appeals’ decision in Sierra Club to an-

ticipated that the results of the fluid modeling demonstration would

be the basis for a state plan revision authorizing a relaxed emission

rate for the Kammer plant, such a revision was never approved by the

agency

12

ticipate that credit for such an increase—up to the 2.5H

formula height before 1979, and up to the H+ 1.5L for-

mula height for increases commenced between January 12,

1979, and October 11, 1983—would not be forthcoming

without a demonstration. Pet. App. 102a-103a. The agen-

cy decided, however, that States would remain free to re-

quire a fluid modeling demonstration (or field studies) for

within-formula height increases if they believed that ap-

plication of the formulas overestimated good engineering

practice stack height. Jd. at 103a-108a.

C. The Opinion Below. Industry petitioners challenged

the agency’s decision to establish a presumption that the

NSPS emission limitations were the proper ones to assume

in conducting fluid modeling demonstrations to justify

credit for above-formula stack height. They asserted that

the EPA failed to provide notice as required by former

Section 4 of the Administrative Procedure Act (5 U.S.C.

553), and that it had not developed a record to support the

presumption that NSPS could be met for all sources. In

addition, certain industry petitioners challenged the EPA’s

decision not to exempt from the new demonstration re-

quirements sources that had qualified for above-formula

stack height credit under prior agency policies. The court

of appeals rejected those arguments.

As to the notice-and-comment argument, the court

stated that the EPA “can obviously promulgate a final

regulation that differs in some respects from its proposed

regulation” and that the agency’s “final rule must only be a

‘logical outgrowth’ of its proposed rule” (Pet. App. 32a

(citation omitted)). The court found that that standard

was met (/d. at 32a-34a). First, it noted that the agency’s

proposed rule in November 1984 included the NSPS limit

as one of the alternatives that would be applicable to some

sources. Second, it pointed out that the idea that even-

13

tually led the EPA to choose that option, and to reject the

use of existing emission limits or the limits that would

result from use of the GEP formula, was one that at least

one of the rulemaking participants, the NRDC, vigorously

pressed on the agency from the outset—that polluters

should “control first” and only afterwards be permitted to

resort to dispersion by increasing stack height.*® Third, the

court found that public comments on the proposed rule

expressly urged the agency to adopt, and thereby put

others on notice that it might adopt, a single technology-

based limit. Finally, the court observed that at least some

of the industry petitioners were given actual prior notice of

the substance of the final rule, and they therefore had a

limited opportunity to focus a direct attack on the NSPS

presumption, as they did.

As to the objection that the “NSPS presumption” lacked

record support, the court of appeals concluded (Pet. App.

30a-31a) that the EPA need not establish a record that

NSPS are attainable by most affected sources. The court

explained: “as EPA allows any source to use a higher emis-

sions rate when NSPS is infeasible, there is no need for

any sort of generic demonstration that it is normally

[feasible]” (/d. at 30a). It further explained that the EPA’s

decision to set a relatively uniform and low presumptive

baseline emission limit for use in the Section 123

demonstrations, and to place the burden of showing in-

feasibility on the source owner, was consistent with Con-

* The court explained that the NSPS option chosen by the EPA “‘is,

of course, a variant of control-first” (Pet. App. 17a). It further ex-

plained that, in selecting that option, the EPA rejected the existing-

limit Option becaaise use of that limit could lead to “ ‘using a tall stack

to justify itself’ ” (id. at 23a (quoting id. at 97a)). The court also

observed (/d. at 33a) that no party before it advocated the formula-

level option originally articulated in the November 1984 proposed

rule.

14

gress’s intent that credit for extra stack height be approved

only in “ ‘rare circumstances’” (Pet. App. 30a-3la

(quoting H.R. Rep. 95-294, supra, at 93)).

The court of appeals also concluded (Pet. App. 35a,

45a-50a) that the agency did not abuse its discretion in

deciding not to grandfather sources that performed fluid

modeling demonstrations to justify above-formula stack

height under prior agency policy and regulations. The

court first found that the decision in Georgetown Univer-

sity Hospital v. Bowen, 821 F.2d 750 (D.C. Cir. 1987),

cert. granted, No. 87-1097 (Feb. 29, 1988), did not apply

in this case. That decision, the court explained (Pet. App.

35a), invalidated as retroactive a rule that would have

limited reimbursements for past transactions. The rule at

issue in this case applies only to future conduct, namely, a

source’s future emissions. |

The court then applied the retroactivity analysis that it

had articulated in its decision in Sierra Club, concluding

(Pet. App. 36a) that “the issue entails a balancing of the

interest in prompt and complete fulfillment of statutory

goals against the inequity of enforcing a new rule against

persons that justifiably made investment decisions in

reliance on a past rule or practice.” The court found that

the agency had relied on the importance of the statutory

goal and on its view that Congress in Section 123 intended

affirmatively to alter industry reliance on above-formula

stack heights (Pet. App. 49a). The court recognized that

source owners that had conducted fluid modeling

demonstrations in the past had valid reliance in-

terests — though those were tempered somewhat by the fact

that “regulatory pronouncements since 1970 had con-

sistently placed a higher burden on credits for above-

formula stacks” (/bid.)—and might have an equitable

argument for grandfathering if they could demonstrate

thai the new requirements would impose a heavy economic

15

burden (ibid.). Because the NSPS presumption could be

rebutted by a showing of infeasibility, however, those

source Owners would “have an opportunity to identify

the[ir] costs and secure such relief as their size may justify”

(id. at 50a).? The court accordingly found that the agency

did not abuse its discretion in deciding not to grandfather

sources that had performed modeling demonstrations and

increased their stacks to above-formula heights.

Finally, the court rejected various other challenges, in-

cluding a challenge to the sufficiency of the agency’s re-

quirement of demonstrations for justifying within-

formula stack height increases (Pet. App. 24a-28a). In

particular, several envireamental groups and States

challenged the EPA’s decisioii grandfather within-

formula stack height increases comnvenced prior to the

ing (Pet. App. 36a-40a) that the EPA had not adequately

considered the degree of reliance on fluctuating agency

policy and therefore had not adequately supported its con-

clusion that grandfathering of this category of stacks was

appropriate. The court stated (id. at 40a): “We do not say

there is no room for grandfathering on these facts, but the

case for it seems unusually weak. Any grandfathering

chosen should fit, to a reasonable degree, the variations in

regulatory history and degrees of reliance.”

° The court observed (Pet. App. 50a) that “the only sunk cost that ts

directly wasted by the new regulations is the cost of the [modeling]

demonstrations themselves” and that, while those costs might not be

negligible in some cases, it would be “a rare case where the costs of

securing data could alone entitle a party to grandfathering.”

en

16

ARGUMENT

1. Petitioners contend (87-2068 Pet. 16-23) that the

EPA violated former Section 4 of the Administrative Pro-

cedure Act (5 U.S.C. 553) by giving inadequate notice

prior to promulgation of its final rule that source owners

would presumptively have to use NSPS as the assumed

emission level in conducting fluid model demonstrations

to qualify for stack height credit above that prescribed by

the formula. The court of appeals correctly rejected that

claim (Pet. App. 3la-34a). The ruling raises no issue that

warrants this Court’s review.

Petitioners received the notice to which APA Section 4

entitled them. See 5 U.S.C. 553(b)(3) (a notice of proposed

rulemaking must convey “either the terms or substance of

the proposed rule or a description of the subject and issues

involved”). A final rule may, of course, differ in some

respects from the initial proposal: the very point of com-

ments is to promote consideration of changes. See

Chocolate Mfrs. Ass’n v. Block, 755 F.2d 1098, 1103-1104

(4th Cir. 1985); International Harvester Co. vy.

Ruckelshaus, 478 F.2d 615, 632 n.51 (D.C. Cir. 1973). For

example, an agency may eliminate several alternatives

discussed in a notice of proposed rulemaking even if the

result is to make the final rule more stringent than the pro-

posal. See Connecticut Light & Power Co. v. NRC, 673

F.2d 525, 533 (D.C. Cir.) cert. denied, 459 U.S. 835

(1982). More generally, notice is adequate if the changes in

the original proposal are in character with the original

scheme, and the final rule is a logical outgrowth of the

notice and comment previously given. See, e.g., BASF

Wyandotte Corp. v. Costle, 598 F.2d 637, 642 (Ist Cir.

1979), cert. denied, 444 U.S. 1096 (1980); Chocolate Mfrs.

Ass’n v. Block, 755 F.2d at 1105. Those standards were

met here.

17

The changes adopted in the final rule were in character

with the original scheme. The November 1984 notice of

proposed rulemaking identified the NSPS limit as one of

the three emission rates that the EPA proposed would be

applied to various affected sources. All of the rulemaking

participants were aware that one of the issues to be ad-

dressed in the rulemaking was the emission limit to be used

in fluid modeling demonstrations; indeed, the agency

could not do otherwise, for, as we explained above, such a

limit is an essential element of a meaningful demonstra-

tion. The parties to the rulemaking were also aware that at

least two of the principal participants in the process, Sierra

Club and NRDC, were vigorously promoting the idea that

pollution sources should have to exhaust their ability to

control emissions before being . permitted higher

stacks — an idea that, in the context of the modeling issue,

pointed clearly toward adoption of the NSPS limit, which

was the most stringent of the three standards originally

proposed.'® Even more specifically, various public com-

ments expressly promoted adoption of a single tech-

nology-based limit (such as NSPS) for all sources, in-

cluding those whose owners were performing Section 123

fluid modeling demonstrations (Pet. App. 33a). The agen-

cy responded to all of the rulemaking comments by

'® As the court of appeals explained (Pet. App. 33a), no party ad-

vocated the formula-level option, and the existing-level option was re-

jected essentially on control-first grounds. The objection to using ex-

isting limits was that doing so could lead to a ratcheting up of per-

missible stack height. An existing emission level could justify an in

crease in stack height as necessary to avoid excessive ground-level

pollutant concentrations. That would in turn justify a credit on emis-

sion levels that would then justify an increased emission level. The

new higher emission level could then be used to justify a higher stack

height, and so on. A uniform technology-based limit was needed to

forestall that result, which would frustrate the purpose behind Section

123

TE

18

eliminating the two alternatives to the NSPS limit (/d. at

34a). That response, which was firmly based on the con-

gressional policy that above-formula stacks should be per-

mitted only in “rare circumstances” (H.R. Rep. 95-294,

supra, at 93), introduced no novel or unexpected prin-

ciples into the proceeding; rather, it was a_ logical

outgrowth of the original proposal.''

Contrary to petitioners’ suggestion (87-2068 Pet. 17-18),

the decision of the court of appeals does not conflict with

the principles articulated by the Fourth Circuit in

Chocolate Mfrs. Ass’n v. Block, 755 F.2d at 1102-1105.

The Fourth Circuit there recognized that an agency may

promulgate a final rule that is different from its proposal

(id. at 1103-1104), that notice is adequate if the changes in

the original proposal are in character with the original

scheme and a logical outgrowth of the comments received

(id. at 1105), and that application of those principles in a

particular case is an inherently fact-intensive undertaking

(id. at 1104). See also E. 1. du Pont de Nemours & Co. v.

Train, 541 F.2d 1018, 1026 (4th Cir. 1976), aff'd in part

and rev’d in part, 430 U.S. 112 (1977) (“Notice is sufficient

if it provides a description of the subjects and issues in-

volved.”). The court in Chocolate Mfrs. Ass’n found thai,

considering all the circumstances, the proposed rule was

not “sufficiently descriptive to provide interested parties

with a fair opportunity to comment and to participate in

the rulemaking” (541 F.2d at 1104): first, neither the pro-

posed rule itself nor the history of the issue in the agency

'! In addition, as the court of appeals pointed out (Pet. App. 3ia,

33a), certain of petitioners received actual notice of the EPA’s inten-

tion to adopt the NSPS presumption two weeks before promulgation

of the final rule, and they had an opportunity directly to attack the

NSPS rule at least one week before the final rule was issued.

bance ein

19

nor the preamble discussion furnished fair warning that

the position eventually adopted in the final rule would

even be considered (id. at 1106-1107); and, indeed, the

final rule, which banned flavored milk, was exactly the op-

posite of the original proposal, which banned other pro-

ducts but listed flavored milk as part of a permissible diet

(id. at 1103). The court of appeals in the present cases ap-

plied the same principles and correctly feund, on entirely

different facts, that the notice of proposed rulemaking,

considering all the circumstances, accorded petitioners a

fair Opportunity to urge the agency not to adopt a single

technology-based emission limit, such as the NSPS, for

use in modeling demonstrations.

2. Petitioners contend (87-2068 Pet. 23-27) that the

court of appeals erred in sustaining the EPA’s adoption of

the NSPS presumption in the absence of a technical record

establishing that NSPS can be met by most sources. To the

extent that petitioners suggest that the EPA made a factual

finding that most sources can meet the NSPS, that sugges-

tion is incorrect: the EPA merely “ma[de] the presumption

that this limit can be met by all sources seeking to justify

stack heights above formula height,” while allowing any

source Owner to show the infeasibility of meeting the

NSPS (Pet. App. 96a). Hence, there was no factual find-

ing that required record support.

All that the EPA did was to allocate a burden of proof

to those seeking credit for stack heights above the good

engineering practice formula heights. It merely required

such source owners to prove that they cannot feasibly meet

the NSPS emission-control limit, and should therefore be

able to use a higher limit in their modeling demonstra-

tions, in order to justify a taller stack for greater disper-

sion of pollutants. As the court of appeals held, the EPA’s

ruling in this regard is entirely reasonable and is consistent

with the obvious congressional policy that sources seeking

20

above-formula stack height must carry a heavy burden of

justification. Pet. App. 30a-3la.'?

Petitioners incorrectly assert (87- 2078 Pet. 26) that the

court of appeals’ decision on this point conflicts with the

ruling of the Fourth Circuit in FE. J. du Pont de Nemours &

Co. v. Train, 541 F.2d at 1028. That ruling is irrelevant to

this case: it does not address what record support is re-

quired for a regulation that is expressly promulgated to

establish a rebuttable presumption of the feasability of a

technological level of pollution control. Rather, the ruling

in the case was simply that EPA effluent-limitation regula-

tions under the Clean Water Act establish neither

uniformly applicable standards that must be included in

every water-pollution permit nor mere guidelines for the

permit issuers but presumptively applicable permit condi-

tions. 541 F.2d at 1027-1028.

3. Petitioners contend (88-60 Pet. 9-15) that the EPA

violated an Administrative Procedure Act proscription on

retroactive rulemaking in determining that all post-1970

sources must comply with the demonstration requirements

announced in this rulemaking proceeding in order to ob-

tain credit for above-formula stack heights. They also con-

tend (ibid.) that the court of appeals, in affirming the

agency’s requirement, departed from its earlier holding in

Georgetown University Hospital v. Bowen, supra, which

generally bars retroactive rulemaking. Petitioners are

wrong on both counts. '

'2 Contrary to petitioners’ suggestion (87-2068 Pet. 23-24), the fact

that the EPA must make a record to support the factual findings

necessary to establish NSPS under Section 1}1 of the Act, as it has

done in that context, does not suggest that a similar record is required

where, as here, the agency has merely allocated a burden of proof and

not made any factual findings.

21

The court of appeals correctly pointed out (Pet. App.

35a) the obvious distinction between Georgetown Univer-

sity Hospital and these cases. The rule at issue in that case -

would have limited monetary reimbursements for past

transactions. The rule here governs only future conduct.

Source owners who conducted fluid modeling demonstra-

tions in the past may have to reduce their future emissions

to comply with the requirements of the new regulation,

but the EPA regulation does not recoup past benefits, and

sources that qualified for extra stack height credit under

earlier regulations suffer no penalty stemming from their

past emissions. Indeed, petitioner Ohio Power has ob-

tained considerable economic benefits from its extra stack

height since its Kammer Plant stack became operational in

1979. The EPA regulation means only that Ohio Power

may not continue to enjoy that credit, and thus emit the

additional pollutants the credit would permit, in the

future. In short, nothing in the EPA’s regulation declares

that any of petitioner’s past conduct was unlawful or seeks

to impose liability based on that conduct: the regulation

applies equally to all who engage in the covered conduct in

the future, regardless of their past behavior. !3

'? Contrary to petitioners’ contention (88-60 Pet. 13 n.29), the court

of appeals’ decision in this case is no more contrary to Citizens to Save

Spencer County v. EPA, 600 F.2d 844, 879-881 (D.C. Cir. 1979),

Which could not in any event create an intercircuit conflict, than it is to

Georgetown University Hospital. The Citizens case involved the ap-

plication of an EPA regulation to construction projects commenced

prior to issuance of the regulation, where the regulation required

preconstruction review: it prohibited construction itself unless certain

conditions were met. Accordingly, that regulation, unlike the regula-

tion at issue here, by its terms applied to past conduct and declared

that the conduct had been unlawful at the time it took place.

22

For those reasons, the court of appeals correctly con-

cluded that the retroactive-rulemaking bar articulated in

Georgetown University Hospital does not apply in this

case.'4 The court of appeals recognized, however, that,

when a regulation may diminish the value of past in-

vestments or upset reliance interests, its validity under

arbitrary-and-capricious or abuse-of-discretion review

depends on a balance of “the interest in prompt and com-

plete fulfillment of statutory goals against the inequity of

enforcing a new rule against persons that justifiably made

investment decisions on a past rule or practice” (Pet. App.

36a). The court correctly concluded that the regulation

was valid under that standard.

On one side of the balance, the main statutory goal in

these cases, of course, is that expressed by Congress when

it declared in 1977 that the Section 123 demonstration re-

quirements would apply to all post-1970 sources whose

owners wish to receive credit for above-formula stacks

(Pet. App. 49a). In addition, the Clean Air Act plainly

contemplates and requires that emission control re-

quirements may be altered and made more restrictive —

e.g., by rulemaking or alteration of a state plan—when

statutory or regulatory requirements change or when ex-

isting requirements prove inadequate to attain the ambient

air quality standards. See 42 U.S.C. 7410(a)(2)(H). In

those ways, the Act substantially undermines any claim

that a source owner may seek reliance-based protection

against agency changes of valid _ pollution-control

measures'S’ and demonstrates congressional recognition

‘+ Because the cases do not present the same issues, the petition for

a writ of certiorari in No. 88-60 should not be held pending the Court's

disposition of Georgetown University Hospital v. Bowen.

'S The decisions cited by petitioners (88-60 Pet. 12 n.27) do not sup-

port the contrary premise. In American Methyl Corp. v. EPA, 749

a

23

of a broad need to subject already-existing pollution

sources tO current requirements. Those considerations

weigh heavily in any balance involving the proper treat-

ment of claims of reliance on pre-1984 EPA regulations,

which, moreover, were known all along not to be final.'®

On the other side of the balance, the court of appeals

observed (Pet. App. 49a) that source owners who per-

formed fluid modeling demonstrations in the past may

have relied on agency guidance in doing so and that, under

some circumstances, those source owners might have an

equitable argument for grandfathering.'’? The court of

F.2d 826 (1984), vacated on other grounds, 768 F.2d 385 (D.C. Cir.

1985), cert. denied, 474 U.S. 1082 (1986), for example, the court did

not find, as petitioners suggest, that the EPA had no authority to

revoke a previous decision. Rather, the court of appeals held that the

EPA had failed to fcllow the revocation procedures prescribed by

statute. In Utah Int'l, Inc. v. Andrus, 488 F. Supp. 976 (D. Colo.

1980), the district court refused to permit the government to reopen a

case based on a change in administrative policy. In the present case,

the EPA revised its regulation at the direction of the court of appeals,

which found that the agency’s prior approach did not implement con-

gressional intent.

'© As the EPA argued to the court of appeals, its approval of the

fluid modeling demonstration for Ohio Power’s Kammer Plant was

not an adjudication but part of its rulemaking function. The fluid

mcdeling demonstration for Kammer has no significance that is in-

dependent of the emission limitation and control strategy applicable

to that plant. See C.A. App. 334. Those limitations and strategies are

imposed through rulemaking under Section 110 of the Act. 42 U.S.C.

7410(c). The Federal Register notice that tentatively approved the

Kammer demonstration explicitly stated that a request for a state plan

revision was expected. 47 Fed. Reg. 35784 (1982). But the agency

never approved any revision for the Kammer plant. See Gov’t C.A.

Br. at 23-25, Ohio Power v. Thomas, No. 86-1331.

'? Petitioners incorrectly state (88-60 Pet. 12 & n.26 (quoting Pet.

App. 35a)) that the court of appeals found that the regulation “would

‘impinge unfairly On source OWners that made investments or other

a

24

appeals concluded, however, that, whatever the precise

reliance interests and however they may weigh in the

balance in particular cases, petitioners have not shown

that the regulation will impose a substantial burden on

them (/d. at 50a). The court found that each source owner

with a reliance claim can make the claim in the course of

rebutting the NSPS presumption by showing that the

NSPS are infeasible: the opportunity to rebut that

presumption provides “each source owner * * * an oppor-

tunity to identify the[ir} costs and secure such relief as

their size may justify” (/bid.).'® Hence, the rule at issue in-

corporates a mechanism for individualized consideration

of reliance interests and cannot be found invalid on the

ground that it overrides or impairs such interests.

4. Petitioners challenge (88-61 Pet. 16-22) the court of

appeals’ application of a demonstration requirement to

sources whose stacks are raised to levels up to the good

commitments in reasonable reliance on prior understandings.’ ” To

the contrary, the court of appeals recognized only that the regulation

might have such an effect; it then found that the source owner’s ability

to demonstrate that the new requirements could not feasibly be met

would protect the Owners from any undue burden (Pet. App.

49a-50a).

'§ In the case of Ohio Power’s Kammer Plant, the reliance interest

would seem to be slight. Ohio Power commenced construction of its

Kammer Plant stack in 1976, which was prior to enactment of the

1977 Clean Air Act Amendments. Its subsequent fluid modeling

demonstration was an attempt to qualify that stack for good engineer-

ing practice credit after the fact. Thus, Ohio Power did not rely on

agency guidance in constructing its stack. It may, however, have relied

On its ability to retain credit for the stack in negotiating its current fuet

contract. That issue was not addressed in the rulemaking, and there is

accordingly no record on any such claim. Nor did the court of appeals

address that question. In any event, like all claims of reliance interests,

the issue is appropriately addressed in individual cases if and when

Ohio Power claims that it cannot feasibly meet an NSPS emission

limitation.

25

engineering practice formula levels. To the extent that

petitioners contend that the court of appeals erred in

Sierra Club in reversing the EPA’s determination that no

Such requirement applies to within-formula increases (as

opposed to above-formula increases), the issue does not

merit this Court’s review. When Sierra Club was decided,

the EPA, which is the agency whose interpretation of the

Clean Air Act was invalidated, decided not to seek review

in this Court. The issue was raised in an industry petition

for a writ of certiorari (No. 83-1429); the EPA stated in

response that the petition should not be granted, as the

agency would seek to comply with the remand order in

such a way as tO minimize regulatory burdens; and this

Court denied certiorari. 468 U.S. 1204 (1984). The court

of appeals in the present cases, in holding that the EPA’s

grandfathering of pre-Sierra Club within-formula stack

increases was inadequately supported, implicitly relied on

the Sierra Club principle that demonstration requirements

may apply to within-formula stack increases; the EPA has

again determined that review by this Court is not war-

ranted and has therefore not filed a petition for a writ of

certiorari. The EPA intends to proceed in accordance with

the remand order. The correctness of Sierra Club’s rejec-

tion of the EPA’s interpretation of Section 123 does not

warrant this Court’s review now any more than it did in

1984.

To the extent that petitioners complain separately about

the court of appeals’ refusal to sustain the EPA’s decision

to grandfather all within-formula stack height increases

commenced prior to October 11, 1983, petitioners have

again not raised an issue that merits this Court’s review.

To be sure, we believe that the EPA’s decision to grand-

father within-formula stack height increases was a

reasonable one, both under Section 123 and under the

court of appeals’ prior decision in Sierra Club, and it

26

should have been sustanied by the court of appeals. The

court’s refusal to do so, however, does not present any

legal issue that calls for resolution by this Court. Without

repudiating the proper standards of review, the court

determined that the EPA had not adequately assessed the

degree of reliance by source owners on the EPA’s varying

stack height policies over the years (Pet. App. 36a-40a).

That fact-specific ruling is not of general importance and

does not conflict with any decision of this Court or of any

other court of appeals.

5. Petitioners finally contend (88-61 Pet. 22-24) that

the court of appeals ignored the doctrines of law of the

case and res judicata by remanding for the agency to

reconsider its treatment of within-formula stack height in-

creases after that issue was allegedly settled in Sierra Club.

Of course, the EPA agrees that agencies should not be re-

quired to relitigate the merits of previously adopted

regulatory choices every time it engages in supplemental

rulemaking after a remand. The present cases, however,

do not call for this Court’s elaboration or application of

that doctrine, because the issue raised here was not ex-

plicitly raised or addressed in Sierra Club.

The pertinent issue in Sierra Club was whether the EPA

could grandfather sources whose owners had relied on the

2.5H formula (prior to January 12, 1979), or whether the

more restrictive regulatory formula of H+ 1.5L had to be

applied (719 F.2d at 467-468). Here, the issue is more com-

prehensively whether the agency may permissibly grand-

father stack height increases commenced prior to October

11, 1983. Although the agency originally interpreted Sierra

Cluh’s affirmance of the grandfathering provision as ef-

fectively encompassing the latter question (at least as to

pre-1979 stack height increases), the decision is in fact

somewhat ambiguous on this point. See id. at 459-460.

27

Thus, we cannot say that the court of appeals clearly erred

in failing to find that litigation of the issue was foreclosed

under the doctrines of law of the case and res judicata.

CONCLUSION

The petitions for a writ of certiorari should be denied.

Respectfully submitted.

DONALD B. AYER

Acting Solicitor General*

ROGER J. MARZULLA

Assistant Attorney General

LISA F. RYAN

Altorney

SEPTEMBER 1988

* The Solicitor General is disqualified in this case.

US GOVERNMENT PRINTING OFFICE 1988—2C2.037/60653