United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued September 8, 1998 Decided November 3, 1998
No. 97-1651
George E. Warren Corporation,
Petitioner
v.
U.S. Environmental Protection Agency
and Carol M. Browner, Administrator,
Respondents
Friends of the Earth, et al.,
Intervenors
Consolidated with
97-1656
---------
On Petitions for Review of an Order of the
Environmental Protection Agency
Jeffrey C. Bates argued the cause for petitioner George E.
Warren Corporation. With him on the briefs were Harvey
M. Sheldon and Ellen S. Tenenbaum.
Michael F. McBride argued the cause for petitioner Inde-
pendent Refiners Coalition. With him on the briefs were
Gene E. Godley and Daniel C. Esty.
Seth M. Barsky, Attorney, U.S. Department of Justice,
argued the cause for respondents. With him on the brief
were Lois J. Schiffer, Assistant Attorney General, Jonathan
Z. Cannon, General Counsel, Environmental Protection
Agency, and John Hannon, Attorney.
J. Martin Wagner argued the cause for intervenors
Friends of the Earth, Inc., et al. With him on the brief were
Patti Goldman, Marcus W. Sisk, Jr. and Lori A. Lange.
V. Robert Denham, Jr. argued the cause for intervenors
Petroleo Brasileiro, S.A., and Petrobras America, Inc. With
him on the brief were Rex R. Veal and N. David Palmeter.
Before: Ginsburg, Sentelle, and Rogers, Circuit Judges.
Opinion for the court filed by Circuit Judge Ginsburg.
Ginsburg, Circuit Judge: Before the court are petitions to
review a rule promulgated by the Environmental Protection
Agency in 1997 to implement the anti-dumping provision of
the reformulated gasoline program established by the Clean
Air Act Amendments of 1990. See 42 U.S.C. s 7545(k)(8).
The challenged rule regulates emissions from conventional
gasoline for motor vehicles, and changes the way importers
and foreign refiners of conventional gasoline sold in the
United States had been treated under the prior rule. The
petitioners are the George E. Warren Corporation, an import-
er of gasoline, and the Independent Refiners Coalition, a
trade organization composed of domestic gasoline refiners.
Three environmental pressure groups intervene on behalf of
the petitioners, and two foreign refiners intervene on behalf
of the EPA.
The petitioners and their supporters (hereinafter collective-
ly the petitioners) contend that in promulgating the 1997 rule
the EPA acted beyond its statutory authority, arbitrarily and
capriciously, and in reliance upon comments submitted after
the close of the comment period. For the reasons set out
below, we reject each of these challenges and deny the
petition for review.
I. Background
The Clean Air Act Amendments of 1990 require the refor-
mulation of conventional gasoline to reduce motor vehicle
emissions in certain large urban regions with elevated levels
of ozone. See 42 U.S.C. s 7545(k)(1) (nonattainment areas).
In those areas reformulated gasoline alone may be sold to
consumers; conventional gasoline may be sold in the remain-
der of the country. See id. s 7545(k)(1), (5).
To prevent pollutants being transferred from reformulated
gasoline to conventional gasoline in the refining process, the
Congress included an "anti-dumping" provision, see id.
s 7545(k)(8),* which requires generally that the conventional
__________
* Section 7545(k)(8) provides:
(A) In General
[T]he Administrator shall promulgate regulations applicable
to each refiner, blender, or importer of gasoline ensuring that
gasoline sold or introduced into commerce by such refiner,
blender, or importer (other than reformulated gasoline subject
to the requirements of paragraph (1)) does not result in aver-
age per gallon emissions (measured on a mass basis) of (i)
volatile organic compounds, (ii) oxides of nitrogen, (iii) carbon
monoxide, and (iv) toxic air pollutants in excess of such emis-
sions of such pollutants attributable to gasoline sold or intro-
duced into commerce in calendar year 1990 by that refiner,
blender, or importer.
....
(E) Baseline for determining compliance
If the Administrator determines that no adequate and reli-
able data exists [sic] regarding the composition of gasoline sold
or introduced into commerce by a refiner, blender, or importer
in calendar year 1990, for such refiner, blender, or importer,
baseline gasoline shall be substituted for such 1990 gasoline in
determining compliance with subparagraph (A).
The statutory "baseline gasoline" referred to in s 7545(k)(8)(E) is
set out in s 7545(k)(10)(B).
gasoline of each supplier (including both domestic and foreign
refiners and importers) remain as clean as it was in 1990.
Compliance is measured for each of several specified pollu-
tants by comparing the yearly average emissions per gallon
attributable to each supplier's conventional gasoline with an
individual baseline representing the quality of gasoline that
supplier introduced into the United States in 1990. If, how-
ever, the Administrator of the EPA "determines that no
adequate and reliable data exist[]" to set an individual base-
line for a particular supplier, then that supplier's compliance
is measured against a statutory baseline representing the
average emissions per gallon for all gasoline introduced into
commerce in 1990. And like it or not, thereby hangs a tale.
A.The 1994 Rule
In 1994 the EPA announced standards for setting individu-
al baselines under which domestic refiners, foreign refiners,
and importers were all treated differently. See Final Rule:
Regulation of Fuels and Fuel Additives: Standards for
Reformulated and Conventional Gasoline, 59 Fed. Reg.
7,716. Domestic refiners were each required to establish an
individual baseline because the EPA determined they have
adequate and reliable data with which to do so. Foreign
refiners were not regulated, but their gasoline was subject to
the regulations applicable to the importer thereof. Each
importer was permitted to establish an individual baseline,
but because they generally lacked the data to do so, import-
ers were in practice assigned the statutory baseline. See id.
at 7,785/2-3, 7,786/2.
The EPA had considered giving foreign refiners the option
of either petitioning to establish an individual baseline or
accepting the statutory baseline, but rejected this approach
for three reasons. First, foreign refiners generally would not
have adequate data with which to establish an individual
baseline. Second, the agency was concerned with the poten-
tial for "gaming": a foreign refiner presumably would choose
to apply for an individual baseline only if doing so would allow
the refiner to sell gasoline dirtier than would be allowed
under the statutory baseline, which might result in overall
degradation of air quality. Finally, the EPA determined that
because foreign refiners, unlike domestic refiners, are not
"subject to the full panoply of EPA's regulatory jurisdiction,"
the agency could not impose upon them mechanisms to
ensure compliance monitoring and enforcement, functions
that it considered "integral to the establishment of accurate
and verifiable baselines." Id. at 7,786/2. Because a foreign
refiner's gasoline typically passes through many channels and
may be blended with gasoline from other refineries before it
reaches the United States, and because gasoline cannot be
identified by its refinery of origin, the EPA would have only
the paperwork accompanying the imported gasoline to identi-
fy its source. See id. at 7,787/1-2. The agency also consid-
ered requiring each foreign refiner, like each domestic refin-
er, to establish an individual baseline, but ultimately rejected
this mandatory approach for two of the three reasons it
rejected the optional approach: most foreign refiners lack the
necessary data, and it would be difficult to monitor for and to
enforce compliance. See id. at 7,786/1-2.
B.The 1997 Rule
In 1995 the World Trade Organization held the 1994 rule
violated the anti-discrimination norm of the General Agree-
ment on Tariffs and Trade because domestic refiners were
allowed to set individual baselines while foreign refiners were
not. The United States Trade Representative in June 1996
advised the WTO that the United States intended to comply
with that decision. See 19 U.S.C. s 3533 (specifying proce-
dures agencies must follow before amending a rule held
inconsistent with GATT). The EPA accordingly solicited
public comment to identify its choices, see Invitation For
Public Comment: World Trade Organization (WTO) Deci-
sion on Gasoline Rule (Reformulated and Conventional Gas-
oline), 61 Fed. Reg. 33,703 (1996), and later proposed to
adopt the optional baseline approach it had rejected in the
1994 rule, allowing each foreign refiner either to accept the
statutory baseline or to petition the EPA for permission to
establish an individual baseline. See Notice of Proposed
Rulemaking: Regulation of Fuels and Fuel Additives: Base-
line Requirements for Gasoline Produced By Foreign Refin-
ers, 62 Fed. Reg. 24,776 (1997).
In 1997 the EPA promulgated a final rule to that effect.
See Final Rule: Regulation of Fuels and Fuel Additives:
Baseline Requirements for Gasoline Produced By Foreign
Refiners, 62 Fed. Reg. 45,533 (codified at 40 C.F.R. pt. 80).
The new rule conditions the assignment of an individual
baseline upon the foreign refiner's agreement to certain
conditions necessary to ensure compliance and to facilitate
enforcement. See id. at 45,539-41, 45,550-59. The EPA
concluded that the risk of an adverse environmental impact
from gaming by foreign refiners, although "difficult to quanti-
fy," id. at 45,537/2, was acceptable; indeed, during the first
year in which the 1994 rule was in effect imported gasoline on
average actually had been somewhat cleaner than required by
the statutory baseline. See id. at 45,542/2 & n.13. Neverthe-
less, the EPA again acknowledged, as it had in 1994, that
allowing foreign refiners selectively to petition for individual
baselines could have an adverse environmental effect. Ac-
cordingly, as a remedial measure, the agency announced that
it will monitor the average quality of imported gasoline for
emissions of oxides of nitrogen (NOx)--the pollutants which
in its judgment are most likely to increase over 1990 levels--
and will compare the results with the average quality of
gasoline imported in 1990; if the 1990 levels are ever exceed-
ed, then the permissible levels of NOx for gasoline imported
under the statutory baseline will automatically be decreased
by an offsetting amount, in order thereby to maintain the
"environmental neutrality" of the rule. Id. at 45,537.
II. Analysis
The petitioners bring both substantive and procedural chal-
lenges to the 1997 rule. As to substance, the petitioners
argue the rule is beyond the EPA's statutory authority
because (1) allowing foreign refiners the option to petition the
EPA for an individual baseline may result in a degradation of
air quality; (2) in promulgating the rule the EPA considered
factors other than air quality, namely (a) the WTO's decision
that the 1994 rule was inconsistent with the GATT, and (b)
the likely effect of regulation upon the price and supply of
gasoline in the U.S. market; (3) the EPA may not allow
foreign refiners to choose whether to establish an individual
baseline but must assign them each one unless it determines
with respect to a particular refiner that there are not ade-
quate and reliable data with which to do so; and (4) the
remedial provision impermissibly alters the statutory baseline
set by the Congress. The petitioners also challenge the
agency's reasoning as arbitrary and capricious in several
respects. As to procedure, the petitioners challenge the
EPA's reliance upon comments filed by the Department of
Energy after the close of the period for public comment, and
the agency's failure to consider certain factors when fashion-
ing the remedial provision. We address the latter procedural
point in conjunction with the petitioners' substantive chal-
lenge to the remedial provision, to which it is closely related.
A. Justiciability
Before discussing the merits of this dispute we consider the
EPA's two threshold objections to this proceeding. First, the
EPA contends the IRC lacks prudential standing because it is
not within the zone of interests protected by the Clean Air
Act. The agency then concedes, however, that because War-
ren raises "essentially the same claims as IRC," our holding
that the IRC lacks standing would "not reduce the number of
issues before the Court." See, e.g., Mountain States Legal
Found. v. Glickman, 92 F.3d 1228, 1232 (D.C. Cir. 1996)
("For each claim, if constitutional and prudential standing can
be shown for at least one plaintiff, we need not consider the
standing of the other plaintiffs to raise that claim"). Con-
trary to the EPA's concession, however, we observe that
several of the arbitrary and capricious challenges in this case
are advanced only by the IRC. We must therefore consider
whether the IRC has standing to bring them. See Louisiana
Envtl. Action Network v. Browner, 87 F.3d 1379, 1382 (1996)
("before we reach the merits of any claim, we must first
assure ourselves that the dispute lies within the constitutional
and prudential boundaries of our jurisdiction").
As it turns out, however, we find it unnecessary to resolve
the question of the IRC's prudential standing because the
citizen suit provision of the Clean Air Act, 42 U.S.C. s 7604(a)
("any citizen may commence a civil action"), apparently re-
moves that "judicially self-imposed limit[] on the exercise of
federal jurisdiction." Bennett v. Spear, 520 U.S. 154, 162
(1997). In the cited case the Court concluded that the
Congress had removed the zone-of-interests barrier to suits
brought under the Endangered Species Act by including a
citizen suit provision that the Court described as follows:
The first operative portion of the provision says that
"any person may commence a civil suit"--an authoriza-
tion of remarkable breadth when compared with the
language Congress ordinarily uses. Even in some other
environmental statutes, Congress has used more restric-
tive formulations.... Our readiness to take the term
"any person" at face value is greatly augmented by two
interrelated considerations: that the overall subject mat-
ter of this legislation is the environment (a matter in
which it is common to think all persons have an interest)
and that the obvious purpose of the particular provision
in question is to encourage enforcement by so-called
"private attorneys general"--evidenced by its elimination
of the usual amount-in-controversy and diversity-of-
citizenship requirements, its provision for recovery of the
costs of litigation (including even expert witness fees),
and its reservation to the Government of a right of first
refusal to pursue the action initially and a right to
intervene later.
Id. at 164-65. The citizen suit provision in the Clean Air Act
is indistinguishable, for it contains all of the elements the
Court found significant in the corresponding provision of the
ESA. See 42 U.S.C. s 7604(a) ("any person may commence a
civil action"); id. (eliminating amount-in-controversy and di-
versity-of-citizenship requirements); s 7604(d) (providing for
recovery of costs of litigation, including expert witness fees);
s 7604(b)(1)(A) (Administrator must be given right of first
refusal); s 7604(c)(2) (Administrator, if not a party, may
intervene at any time). Accordingly, we hold the IRC has
standing.
Second, the EPA claims that Warren's challenge to the
remedial provision is unripe because the remedial provision
may never be implemented; if ever it is, that will constitute
"final action" from which Warren may then appeal. Warren
responds with the broad nostrum that under the Clean Air
Act "immediate challenge of a regulation is the prudent
course." Despite the generality of its response, we agree
with Warren that its challenge is ripe.
To determine the ripeness of a controversy for judicial
review ordinarily requires the court to evaluate "both the
fitness of the issues for judicial decision and the hardship to
the parties of withholding court consideration." Ohio Forest-
ry Ass'n v. Sierra Club, 118 S.Ct. 1665, 1670 (1998) (quoting
Abbott Lab. v. Gardner, 387 U.S. 136, 149 (1967)). "In
determining the fitness of an issue for judicial review we look
to see whether the issue is purely legal, whether consider-
ation of the issue would benefit from a more concrete setting,
and whether the agency's action is sufficiently final." Clean
Air Implementation Project v. EPA, 150 F.3d 1200, 1204
(D.C. Cir. 1998). All three criteria are met here: Whether
the remedial provision is consistent with the statute is purely
an issue of law. That issue is as concrete now as it will ever
be because the rule operates automatically to lower the
statutory baseline if and when average emissions of NOx
from gasoline imported thereunder exceed corresponding
1990 levels; no future factual developments will add to our
present understanding of the effect the remedial provision
will have upon Warren, when and if it is implemented. Cf. id.
at 1205 (contention that EPA's credible evidence rule altered
emission standards not fit for review because effects of rule
could not be adequately determined prior to its application in
concrete factual setting). And the rule is undoubtedly final.
Under these circumstances, we need not address the EPA's
argument that Warren will suffer no hardship unless and
until the remedial provision is implemented. "[W]here the
first prong of the [Abbott Laboratories] ripeness test is met
and Congress has emphatically declared a preference for
immediate review [as it has under the Clean Air Act, 42
U.S.C. s 7607(b)] ... no purpose is served by proceeding to
the second prong." NRDC v. EPA, 22 F.3d 1125, 1133 (D.C.
Cir. 1994). We conclude, therefore, that Warren's challenge
to the rule's remedial provision is ripe for review.
B.Substantive Challenges
Because the EPA is charged with administering the Clean
Air Act, we evaluate a challenge to its statutory authority
under the familiar two-step analysis of Chevron U.S.A., Inc.
v. NRDC, 467 U.S. 837 (1984). See Motor Vehicle Mfrs.
Assoc. v. EPA, 768 F.2d 385, 389 n.6 (D.C. Cir. 1985) (Chev-
ron analysis applies to questions of statutory construction
under Clean Air Act). Under Chevron step one the court
asks "whether Congress has directly spoken to the precise
question at issue," id. at 842; if so, then we "must give effect
to the unambiguously expressed intent of Congress." Id. at
843. If the Congress has not addressed the issue, however,
then under Chevron step two we will defer to the agency's
interpretation if it is reasonable in light of the structure and
purpose of the statute. See id.
We review a challenge to the agency's actions as arbitrary
and capricious under the same standards that we apply when
reviewing a rule pursuant to the Administrative Procedure
Act. See 42 U.S.C. s 7607(d)(9)(A); Chemical Mfrs. Ass'n v.
EPA, 28 F.3d 1259, 1263 (D.C. Cir. 1994). As we have
pointed out before, this inquiry may overlap with our analysis
under step two of Chevron, see Republican Nat'l Comm. v.
FEC, 76 F.3d 400, 407 (1996); so it does to some extent in
this case.
1. Air Quality Improvement
The petitioners base all their challenges to the EPA's
statutory authority upon the premise that any rule that does
not guarantee the maintenance or improvement of air quality
violates the anti-dumping provision of 42 U.S.C. s 7545(k)(8).
For this they rely upon our remark in American Petroleum
Institute (API) v. EPA, 52 F.3d 1113, 1119 (1995), that the
"sole purpose of the [reformulated gasoline] program is to
reduce air pollution." In doing so, however, the petitioners
misjudge the applicability of that precedent.
The overall goal of the reformulated gasoline program is, of
course, to improve air quality by reducing air pollution; the
means chosen to achieve that end are, first, requiring that
only reformulated gasoline be sold in nonattainment areas,
see s 7545(k)(1), and second, prohibiting the transfer of pollu-
tants in the refining process from reformulated to convention-
al gasoline, see s 7545(k)(8). In API the court was interpret-
ing s 7545(k)(1), which sounds the main theme by requiring
"the greatest reduction in emissions of [specified pollutants]
achievable through the reformulation of conventional gaso-
line." That interpretation has no application to s 7545(k)(8),
which is intended (though for reasons discussed in the next
paragraph is not guaranteed) to maintain the emissions per
gallon attributable to conventional gasoline at levels no higher
than they were in 1990, but does not require that they be
reduced. The petitioners thus conflate the overriding statu-
tory purpose of the reformulated gasoline program, which
was at issue in API, with an express statutory command
applicable to conventional gasoline, which is at issue in this
case.
Indeed, far from requiring that emissions per gallon of
conventional gasoline be reduced--or even capped at 1990
levels, as a bill passed by the House would have done, see S.
1630, 101st Cong. s 212(k)(6) (May 23, 1990), reprinted in
Sen. Comm. on Env't & Pub. Works, 103d Cong., 2 Legislative
History of the Clean Air Act Amendments of 1990, at 2065
(1993)--the anti-dumping scheme as enacted allows such
emissions to increase from 1990 levels under certain circum-
stances. If individual baselines could be established for every
supplier, and if the market share of each were to remain at its
1990 level, then of course emissions per gallon also would
remain at 1990 levels. Because, however, the data needed to
set an individual baseline do not exist for every supplier--a
contingency for which the Congress expressly provided, see
s 7545(k)(8)(E)--the statute inherently tolerates an increase
over 1990 emissions per gallon. While suppliers with an
individual baseline must maintain the quality of their gasoline
at its 1990 level or better, suppliers operating under the
statutory baseline may be providing gasoline that is dirtier
than was the corresponding gasoline on the market in 1990.
In addition, the market share of the latter group may in-
crease, which would also contribute to a lessening of air
quality. In sum, although the general purpose of the anti-
dumping provision is to maintain average emissions per gallon
from conventional gasoline at no more than 1990 levels, the
specific approach adopted by the Congress makes full
achievement of that goal less than certain. This result appar-
ently reflects a legislative compromise between two potential-
ly conflicting goals--avoiding degradation of air quality and
not disrupting the market for conventional gasoline. See 136
Cong. Rec. 35,759 (1990) ("There are several dilemmas ...
that must be equitably addressed by EPA in implementation
of [s 7545(k)(8)], taking into account costs, fuels, availability
and energy penalties to refiners") (statement of Sen.
Simpson); id. at 35,002 ("Control over supply means control
over price. EPA must be sensitive to this danger: New
[s 7545(k) does] not intend to resurrect a 1970's DOE type
scheme of detailed government intervention in U.S. gasoline
markets. An approach of this sort would be especially unac-
ceptable, in view of the fact that DOE's restrictions artificially
lowered prices, whereas EPA restrictions on supply would
artificially raise them") (statement of Rep. Sharp), reprinted
in 1 Legislative History of the Clean Air Act Amendments of
1990 at 1164, 1219.
2. Factors Other Than Air Quality
Again proceeding from the mistaken premise that the
maintenance or improvement of air quality is the sole focus of
the anti-dumping provision, the petitioners argue that the
EPA may not consider factors other than air quality in
promulgating rules under s 7545(k)(8). Thus do they chal-
lenge the EPA's consideration both of the WTO's decision
interpreting the GATT and of the comments of the Depart-
ment of Energy concerning the economic effects of the alter-
natives before the agency. The DOE predicted that mandat-
ing individual baselines for foreign refiners could adversely
affect the price and supply of gasoline by making it more
difficult for the foreign refiners quickly to divert gasoline to
the U.S. market in times of increased demand. See 1997
Rule, 62 Fed. Reg. at 45,536/2. The EPA responds that
nothing in the statute precludes consideration of such factors,
and that its approach is congruent with that employed by the
Congress when it enacted the anti-dumping provision.
The petitioners do not direct our attention to anything in
the text or structure of the statute to indicate that the
Congress intended to preclude the EPA from considering the
effects a proposed rule might have upon the price and supply
of gasoline and the treaty obligations of the United States.
Under step two of Chevron, therefore, we must defer to the
agency's construction if it is reasonable. See NRDC v. EPA,
824 F.2d 1146, 1157 (D.C. Cir. 1987) (en banc) (interpreting 42
U.S.C. s 7412 of the Clean Air Act and rejecting the view
that "as a matter of statutory interpretation, cost and techno-
logical feasibility may never be considered under the Clean
Air Act unless Congress expressly so provides"); Interna-
tional Bhd. of Teamsters v. United States, 735 F.2d 1525,
1529 (D.C. Cir. 1984) ("In the absence of clear congressional
direction to the contrary, we will not deprive the agency of
the power to fine-tune its regulations to accommodate worthy
nonsafety interests" under a statute focused upon safety); see
also Grand Canyon Air Tour Coalition v. FAA, 154 F.3d 455,
475, (D.C. Cir. 1998) (FAA properly considered effects of rule
on air tourism industry where the statute did not forbid such
consideration and required not total but only "substantial
restoration of the natural quiet").
Under step two of Chevron, we think the agency's interpre-
tation is permissible. Section 7545(k)(8) specifically allows
foreign refiners that produced dirtier than average gasoline
in 1990 to continue importing gasoline of that quality, pre-
sumably in order to prevent the disruption that might ensue
were those refiners forced to choose between producing
cleaner gasoline than they did in 1990 or quitting the U.S.
market. The agency, following the lead of the Congress,
similarly sought to prevent its rule from disrupting the
market.
In the particular circumstances of this case our usual
reluctance to infer from congressional silence an intention to
preclude the agency from considering factors other than
those listed in a statute is bolstered by the decision of the
WTO lurking in the background. "Since the days of Chief
Justice Marshall, the Supreme Court has consistently held
that congressional statutes must be construed wherever pos-
sible in a manner that will not require the United States 'to
violate the law of nations.' " South African Airways v. Dole,
817 F.2d 119, 125 (D.C. Cir. 1987) (quoting The Schooner
Charming Betsy, 6 U.S. (2 Cranch.) 64, 118 (1804)); see also
Vimar Seguros y Reaseguros, S.A. v. M/V Sky Reefer, 515
U.S. 528, 539 (1995) ("If the United States is to be able to
gain the benefits of international accords and have a role as a
trusted partner in multilateral endeavors, its courts should be
most cautious before interpreting its domestic legislation in
such manner as to violate international agreements").
In sum, we conclude the EPA's consideration of factors
other than air quality is not precluded by anything in
s 7545(k)(8); in this case, moreover, that consideration ap-
pears to be congruent with both the congressional purpose
not to disrupt the market for imported gasoline and the
Supreme Court's instruction to avoid an interpretation that
would put a law of the United States into conflict with a
treaty obligation of the United States. For these reasons we
deem the EPA's interpretation of the anti-dumping provision
a reasonable one; pursuant to Chevron step two, therefore,
we must uphold it.
3. Adequate and Reliable Data
The petitioners argue that in s 7545(k)(8)(E) the "Congress
provided that the Statutory Baseline was not an option but
could be used only if 'the Administrator determines that no
adequate and reliable data exist[] regarding the composition
of gasoline sold or introduced into commerce by a refiner,
blender, or importer in calendar year 1990.' " The EPA does,
however, link assignment of the statutory baseline to the
unavailability of data with which to set an individual baseline.
As we understand the petitioners' argument, therefore, they
object to the manner in which the agency makes this determi-
nation.
Under step one of Chevron, we conclude the Congress did
not speak to the precise question at hand: s 7545(k)(8)(E)
requires the EPA to determine whether the necessary data
are available, but it does not tell the agency how it must go
about the task. Nor could counsel for the petitioners say,
when asked, what the statute requires the agency to do when
a foreign refiner claims it does not have data relevant to
establishing an individual baseline. In view of the lack of
guidance in the statute concerning how the EPA is to deter-
mine whether "adequate and reliable data exist[]" within the
meaning of s 7545(k)(8), we proceed to step two of Chevron
and examine whether the agency's method of making that
determination represents a reasonable construction of the
statute.
The EPA determined implicitly--but nonetheless clearly--
that for the purpose of setting an individual baseline no
adequate and reliable data can exist with respect to a foreign
refiner unless that refiner agrees to certain conditions, sever-
al of which relate directly to the production of information,
necessary to ensure compliance with and enforcement of the
statute. "[W]e defer to an interpretation which was a neces-
sary presupposition of the [agency's] decision," so long as that
interpretation is reasonable. National R.R. Passenger Corp.
v. Boston & Maine Corp., 503 U.S. 407, 420 (1992). (As in
the cited case, it is "noteworthy ... that [no petitioner]
contends the [EPA's] decision was not informed and governed
by this statutory interpretation." Id.)
In the 1994 rulemaking the EPA stated its belief that most
foreign refiners lack the data necessary to establish the
quality of the (often minor) portion of their 1990 output sold
in the United States. See 1994 Rule, 59 Fed. Reg. at 7,786/1-
2. As to those refiners that might have the data, the agency
explained that such data cannot be deemed "adequate and
reliable" unless backed up by appropriate mechanisms for
monitoring and enforcing the refiner's compliance with EPA
regulations:
There is a fundamental distinction between EPA's
ability to monitor and enforce regulatory requirements
that would apply against domestic as opposed to foreign
refiners. Simply put, domestic refiners are subject to
the full panoply of EPA's regulatory jurisdiction and
compliance monitoring, while not all foreign refiners
desiring to produce reformulated and/or conventional
gasoline may be subject to EPA's regulatory jurisdiction
with equivalent certainty. Compliance monitoring and
enforcement are integral to the establishment of accurate
and verifiable baselines, as well as subsequent compli-
ance with standards based on these baselines.
The reformulated gasoline program compliance moni-
toring and enforcement scheme consists of several ele-
ments designed in the aggregate to ensure that the
environmental goals of the Clean Air Act are met, includ-
ing, inter alia: baseline-setting audits; mandatory re-
porting and record keeping; independent laboratory
sampling and testing; tracking of product from point of
production to point of distribution; unannounced EPA
compliance inspections; annual attest engagements by
certified professionals; and an enforcement scheme com-
prised of civil penalties, injunctive relief, and criminal
sanctions. Domestic refiners and importers are subject
to EPA jurisdiction in each of these activities; all foreign
refiners may not be equally amenable to EPA jurisdic-
tion.
Id. at 7,786/2-3.
The 1997 rule addresses these concerns by allowing a
foreign refiner that is able to establish the quality of the
gasoline it sold in the United States in 1990 to petition the
EPA to establish an individual baseline if the refiner agrees
to specific conditions aimed at ensuring its compliance.
These are: (1) in the case of a state-owned or operated
refinery, that it waive any defense of sovereign immunity in
civil, criminal, or administrative enforcement proceedings;
and (2) in all cases, that the refiner (a) appoint an agent for
the service of process in Washington, D.C.; (b) post a sub-
stantial bond to ensure payment of penalties in the event of
its noncompliance; (c) commit to allowing EPA inspections
and audits of all gasoline produced, regardless whether it is
intended for the U.S. market; (d) submit to the jurisdiction of
United States courts or administrative tribunals in any en-
forcement action; (e) implement detailed tracking and certifi-
cation procedures to ensure its compliance with EPA regula-
tions; and (f) procure independent third-party sampling and
laboratory tests. See 1997 Rule, 62 Fed. Reg. at 45,539-41,
45,550-59.
In effect, then, the agency presumes that foreign refiners
will not have (or will not produce) the 1990 data necessary to
establish an individual baseline; it then provides foreign
refiners that do have those data with the option, subject to
the conditions necessary to ensure their compliance with the
agency's program, of petitioning for an individual baseline.
In view of the evident likelihood that foreign refiners--even if
willing--would be unable to produce the requisite data, and
the significant jurisdictional obstacles to obtaining data from
and enforcing compliance against the unwilling, we think the
EPA's construction of s 7545(k)(8) is a reasonable one. That
data with which to set an individual baseline merely "exist" in
the abstract is meaningless, as the agency concluded, unless
backed by adequate provisions for enforcing the refiner's
compliance with that baseline.
4. The Remedial Provision
Recall that the EPA provided for an automatic tightening
of the statutory baseline for imported conventional gasoline in
the event emissions per gallon of NOx ever exceed a bench-
mark representing the average quality of gasoline imported in
1990. See 1997 Rule, 62 Fed. Reg. at 45,537/2-3. Warren
asserts the EPA has no authority to alter the statutory
baseline set out in s 7545(k)(10)(B), to which the agency
responds that s 7545(c)(1)(A) empowers it to do precisely
that.
Under the latter provision, the EPA has broad authority to
"control" the introduction into commerce of any motor vehicle
fuel "if in the judgment of the Administrator any emission
product of such fuel ... causes, or contributes, to air pollu-
tion which may reasonably be anticipated to endanger the
public health or welfare." It is well-established that under
this provision the EPA may control "fuels or fuel additive[s]
already in commerce," API, 52 F.2d at 1121, which the
agency has long done with respect to leaded gasoline. See
Union Oil Co. v. EPA, 821 F.2d 678, 680 (D.C. Cir. 1987);
Amoco Oil Co. v. EPA, 501 F.2d 722, 744-46 (D.C. Cir. 1974).
The agency may also use the same authority to alter decisions
made under other subsections of s 7545. See Ethyl Corp. v.
EPA, 51 F.3d 1053, 1063-64, 1065 (D.C. Cir. 1995) (explaining
that under s 7545(c) EPA may control or prohibit a fuel
additive previously waived into commerce under s 7545(f)).
Warren does not respond directly to the EPA's claim of
authority under s 7545(c)(1)(A), but instead asserts that
s 7545(k)(10)(B) specifies "exactly" what the statutory base-
line for imported gasoline should be. That is neither formally
nor functionally correct, however. Section 7545(k)(10)(B) re-
fers not to gasoline that contains specific levels of pollutants
but to gasoline that "meets the [listed] specifications"; as a
matter of ordinary language, gasoline that exceeds those
specifications necessarily also "meets" them. See American
Heritage Dictionary 1122 (3d ed. 1992) (def. 11: "[t]o satisfy
(a need, for example); fulfill: meet all the conditions in the
contract"). Nor does anything in s 7545(k) suggest the
Congress intended to limit the Administrator's broad authori-
ty under s 7545(c)(1)(A).
Warren next argues that the agency failed to satisfy a
procedural prerequisite to regulation under s 7545(c)(1)(A):
the Administrator may regulate a fuel under s 7545(c)(1)(A)
only "after consideration of all relevant medical and scientific
evidence available to [her], including consideration of other
technologically or economically feasible means of achieving
emission standards under section 7521 of this title [governing
emissions controls on motor vehicles]." s 7545(c)(2)(A).
This claim founders, however, upon Warren's own failure to
show (indeed, even to attempt to show) that the alleged
oversight was material to the Administrator's decision, as
required by the Act. See s 7607(d)(8) ("In reviewing alleged
procedural errors, the court may invalidate the rule only if
the errors were so serious and related to matters of such
central relevance to the rule that there is a substantial
likelihood that the rule would have been significantly changed
if such errors had not been made"); s 7607(d)(1)(E) (making
s 7607(d) applicable to rules regulating fuels under s 7545);
Texas Mun. Power Agency v. EPA, 89 F.3d 858, 875 (D.C.
Cir. 1996) (under s 7607(d)(8) court may reverse only proce-
dural errors that are material).
Accordingly, we hold that in enacting the remedial provi-
sion the EPA acted within its authority under s 7545(c)(1)(A),
that such authority is not limited by s 7545(k)(10)(B), and
that Warren's procedural challenge to the remedial provision
is barred.
5. Arbitrary and Capricious Challenges
The petitioners challenge the 1997 rule as arbitrary and
capricious in seven respects. First, they argue the EPA has
not provided a reasoned explanation for departing from its
prediction in the 1994 rulemaking that the optional baseline
approach would adversely affect the average quality of im-
ported gasoline due to "gaming" by foreign refiners. See
Motor Vehicles Mfrs. Ass'n v. State Farm Mut. Auto Ins. Co.,
463 U.S. 29, 57 (1983) ("An agency's view of what is in the
public interest may change, either with or without a change in
circumstances. But an agency changing its course must
supply a reasoned analysis"). In response, the EPA points
out that it continued to recognize the possibility that gaming
by foreign refiners might lessen air quality but concluded that
this risk was not as great as the agency previously had
thought; indeed, one of its basic assumptions in the 1994
rulemaking was contradicted by new data.
In 1994 the agency had simply assumed that refiners would
supply the dirtiest gasoline allowed by law. See 1994 Rule,
59 Fed. Reg. at 7,785/1-2. In the 1997 rulemaking, however,
the EPA noted that data collected under the 1994 rule
showed that for imported gasoline in 1995 the volume-
weighted average of exhaust toxics was significantly lower,
and in the case of NOx was slightly lower, than both the
statutory baseline and the volume weighted average for do-
mestic individual baselines. See 1997 Rule, 62 Fed. Reg. at
45,542/2 & n.13. Those data corroborated data submitted by
a major foreign supplier to the U.S. market showing its
expected individual baseline for exhaust toxics would be
cleaner than the statutory baseline. See id. at 45,542/2. The
evidence therefore showed that although under the 1994 rule
foreign suppliers were required only to meet the statutory
baseline, they were in fact on average producing cleaner
gasoline, apparently for other reasons. The facts having
changed, the agency reasonably changed its position, as it
explained at the time.
Second, and relatedly, the petitioners argue that the after-
the-fact nature of the remedy is arbitrary and capricious
because it allows air quality to degrade in the first place.
The EPA says it was responding, however, to the uncertain-
ties inherent in attempting to predict whether a future degra-
dation of gasoline quality, should it occur, will be due to
gaming by foreign refiners or to shifts in the composition of
the market. (Obviously those foreign refiners that do not
have the data to construct an individual baseline simply can
not game the system; they do not have the option of petition-
ing for an individual baseline.) In monitoring the quality of
imported gasoline the EPA will be using a three-year average
in order to distinguish between these two possible causes; as
the agency explained, a "change in average gasoline quality
during any particular year ... might indicate the effects of
allowing the option for individual baselines,* or it might
__________
* In quoting the EPA decisions, acronyms have been replaced with
the terms they represent.
reflect the unique circumstances [due to shifts in the market]
for that year." Id. at 45,542/1.
The EPA further explained that any adverse impact upon
the quality of conventional gasoline due to gaming by foreign
refiners depends upon a host of factors, to wit:
the number of foreign refiners that receive an individual
baseline, the actual individual baselines assigned to them,
the volume of gasoline included in the individual baseline,
the source and amount of conventional and reformulated
gasoline imported each year, and the extent, if any, to
which foreign refiners whose 1990 exports to the U.S.
were cleaner on average than the statutory baseline
would now ship gasoline that is dirtier than what they
exported to the United States in 1990.
Id. at 45,560/3; see also id. at 45,537/2 (noting similar uncer-
tainties); DOE Comments, 1997 Rulemaking, Pub. Dkt. No.
A-97-26, IV-G-03, at 3 (filed July 8, 1997). It seems to us
entirely reasonable for the EPA to plan now how it will deal
with uncertain future developments but to delay remedial
action until the need for any remedy is known. See North
Carolina v. FERC, 112 F.3d 1175, 1190 (D.C. Cir. 1997) ("In
the face of 'serious uncertainties,' an agency need only 'ex-
plain the evidence which is available, and ... offer a rational
connection between the facts found and the choice made' ")
(quoting State Farm, 463 U.S. at 52); see also NRDC v.
Thomas, 805 F.2d 410, 425 (D.C. Cir. 1986) (emissions averag-
ing method reasonable even though it could lead to worse air
quality under certain scenarios, because tradeoffs in terms of
clean air objectives are difficult to predict).
Third, Warren argues it would be unfair for the EPA, by
tightening the statutory baseline for NOx, to place upon
importers such as itself the burden of compensating for the
dirtier than average gasoline supplied by foreign refiners
operating under individual baselines. As the agency explains,
however, because gasoline supplied under an individual base-
line necessarily reflects the quality of the supplying party's
1990 gasoline, any increase in emissions per gallon from
foreign gasoline must be due to gasoline imported under the
statutory baseline. Restating the argument in other terms,
Warren claims importers and domestic refiners are similarly
situated and must therefore be treated alike. Although War-
ren claims the EPA did not respond to this formulation of the
point in its rulemaking, it is clear that the agency's response
to the first formulation is equally applicable: because all
domestic refiners have individual baselines, all gasoline sup-
plied at the statutory baseline must be imported. Targeting
this pool of gasoline, therefore, is not arbitrary and capri-
cious; on the contrary, it is congruent with the statutory
scheme permitting a refiner of dirtier than average gasoline
in 1990 to continue, by establishing an individual baseline,
selling gasoline at that standard.
Relatedly, Warren observes that if the statutory baseline is
ever tightened, then more foreign refiners will have an incen-
tive to apply for an individual baseline. This is likely true but
it is no more arbitrary and capricious than the underlying
decision, which we uphold, to place the remedial burden upon
gasoline imported under the statutory baseline. Additionally
we note the EPA has limited the volume of gasoline a foreign
refiner may import under its individual baseline to the volume
of gasoline it imported in 1990; gasoline in excess of that
amount must comply with the presumably more stringent
statutory baseline and therefore will not add to Warren's
potential burden under a tightened statutory baseline. See
1997 Rule, 62 Fed. Reg. at 45,550/2 n.23.
Fourth, the IRC challenges the agency's reasoning that the
mandatory baseline approach could cause adverse environ-
mental effects because foreign refiners of cleaner gasoline
might leave the U.S. market in response to the added compli-
ance burdens entailed in having an individual baseline. The
EPA explained that a mandatory baseline approach
might produce incentives that would tend to reduce the
average quality of imported conventional gasoline. For
example, gasoline from refiners with cleaner individual
baselines would be measured against a more stringent
baseline than under the [1994] rules, while gasoline from
refiners with dirtier individual baselines would be mea-
sured against a less stringent baseline than under the
[1994] rules. Additional costs would be associated with
segregation, tracking, and other requirements.... To
the extent these changes put refiners with clean individu-
al baselines at an economic disadvantage compared to
refiners with either the statutory baseline or an individu-
al baseline dirtier than the statutory baseline, it could
potentially push the supply of gasoline away from refin-
ers with clean individual baseline.
Id. at 45,536/3. The IRC asserts that this reasoning is
illogical because foreign refiners with a dirtier individual
baseline bear the same compliance and tracking costs as
those with a cleaner individual baseline; therefore, we are
told, the two groups of refiners would be equally likely to
abandon the U.S. market for one with lower transaction costs.
It is common ground among all parties, however, that cleaner
gasoline is more expensive to produce. Therefore, it is not
irrational to believe that if all foreign refiners incur an
increased cost, then the suppliers of cleaner gasoline among
them, with their higher costs and therefore smaller profit
margins, will be the first to leave the U.S. market. See Paul
A. Samuelson & William D. Nordhaus, Economics 138-44
(16th ed. 1998) (because profit-maximizing firms will cease
additional production when marginal cost exceeds price, firms
with higher marginal costs will supply relatively fewer goods
at a given price than competitors with lower marginal costs).
Fifth, the IRC contests the EPA's conclusion that because
foreign refiners might abandon the U.S. market when faced
with the significant additional transaction costs associated
with individual baselines, mandating individual baselines for
all foreign refiners could affect the price and supply of
gasoline in the United States. The IRC claims this reasoning
is equally applicable to domestic refiners. As the EPA
explained in its rulemaking decision, however, and no peti-
tioner contests, the 1994 rule mandating individual baselines
for all domestic refiners "has been successfully implemented
without significant disruptions to the supply or price of
conventional gasoline." 1997 Rule, 62 Fed. Reg. at 45,545/3.
Sixth, the IRC claims that to allow foreign refiners the
option to petition for an individual baseline discriminates
arbitrarily against domestic refiners, which are not afforded
that option. [Brief at 28] As the EPA explains, however, and
as we have discussed (in part II.B.3, above), foreign refiners
are situated differently from domestic refiners; the EPA can
impose upon domestic refiners by force the conditions neces-
sary to ensure compliance with individual baselines. Addi-
tionally, because it has determined domestic refiners have
adequate and reliable data to establish individual baselines, to
allow domestic refiners such an option would exceed the
agency's authority under s 7545(k)(8)(A). See id. at 45,545/3
n.18.
Lastly, the IRC claims that because the agency in 1994
allowed domestic refiners to establish individual baselines
using post-1990 data to model 1990 production, see id. at
45,534/2, it "cannot be contended, as EPA did, that a foreign
refiner could not construct its own Individual Baseline using
post-1990 data and submit it to EPA for approval." The
EPA does not respond to this argument in its brief, perhaps
because the agency did not take the position attributed to it;
rather, it concluded in 1994 that modeling the quality of
overall gasoline refinery production in order to establish an
individual baseline "will not work properly for some or most
foreign refineries [because they] ship only a portion of their
production to the U.S. market." 1994 Rule, 59 Fed. Reg. at
7,786/2. Accordingly, in the 1994 rule the agency declined to
permit any foreign refiner to use post-1990 data to establish
an individual baseline. Under the 1997 rule, the agency still
presumes that foreign refiners will be unable to use such
data, see 1997 Rule, 62 Fed. Reg. at 45,536/2, but it gives
them the opportunity to try.
C.Procedural Challenge: Late-Filed Comments
The petitioners challenge the EPA's reliance upon com-
ments filed by the DOE after the close of the comment
period. The EPA argues that the court lacks jurisdiction
over this claim because the IRC failed to exhaust its adminis-
trative remedies. Under the Clean Air Act the court may not
consider a procedural claim unless it has been "raised with
reasonable specificity" before the EPA, either "during the
period for public comment" or in a petition for reconsidera-
tion. 42 U.S.C. s 7607(d)(7)(B); see Texas Mun. Power, 89
F.3d at 871.
In its notice of proposed rulemaking the EPA noted the
DOE's concern, which the DOE had expressed informally,
that the supply and therefore the price of gasoline could be
adversely affected if the EPA were to mandate individual
baselines for all foreign refiners. See 62 Fed. Reg. at 24,-
779/2. At the public hearing the IRC protested the lack of
any data or analysis in the record to justify this concern.
After the period for public comment the DOE submitted its
analysis for the record. The IRC now claims its earlier
objection to the lack of analysis somehow morphed into its
current objection to the late filing of that analysis. The two
points are clearly different, however. Consequently, the
IRC's current challenge to the agency's reliance upon the
DOE's late-filed comments is raised here for the first time,
and for that reason barred.
III. Conclusion
In summary, we hold that the EPA had authority to enact
the optional baseline rule and, in view of the uncertainties
inherent both in the anti-dumping provision of the Act and by
extension in the agency's implementation of that provision, to
provide a prospective remedy against possible degradation of
air quality. We also hold the agency did not act arbitrarily
and capriciously in enacting the rule, for it based its different
treatment of importers, foreign refiners, and domestic refin-
ers upon rational distinctions. Finally, we do not pass upon
the petitioners' two procedural challenges to the rule; one
has not been shown to be material and the other is raised
here for the first time. Accordingly, the petitions for review
are
Denied.