State of Indiana v. EPA , 796 F.3d 803 ( 2015 )

In the
United States Court of Appeals
For the Seventh Circuit
____________________
No. 14-3214
STATE OF INDIANA,
Petitioner,
v.
ENVIRONMENTAL PROTECTION AGENCY,
Respondent.
____________________
ARGUED JUNE 5, 2015 — DECIDED AUGUST 7, 2015
____________________
Before WOOD, Chief Judge, and FLAUM and EASTERBROOK,
Circuit Judges.
FLAUM, Circuit Judge. Pursuant to the Clean Air Act
(“CAA”), 42 U.S.C. § 7401 et seq., the Environmental Protec-
tion Agency sets standards that specify the maximum per-
missible atmospheric concentrations for certain harmful air
pollutants, including ozone. Geographic areas (which do not
necessarily respect state borders) are classified by EPA as
“attainment” or “nonattainment” depending on whether
they meet the standard for a given pollutant. All states are
required to draft a State Implementation Plan (“SIP”) for
each pollutant, which outlines the state’s plan for how it
seeks to achieve or maintain attainment, and which must be
2 No. 14-3214
approved by EPA. All SIPs are subject to certain statutory
and regulatory requirements; the requirements are more
stringent for states with areas in nonattainment. States can-
not revise their SIPs without EPA approval.
If an area within a state is in nonattainment for ozone, the
state’s SIP must include an automobile emissions testing
program. States have some leeway in designing their pro-
grams, as long as they meet certain performance standards.
Prior to 2005, Illinois used an emissions testing program that
tested the emissions of vehicles from all model years; that
program was included in the state’s SIP. In 2005, though, Illi-
nois passed a law which relaxed its emissions testing pro-
gram by exempting pre-1996 model-year vehicles that met
certain standards. That change went into effect in 2007, but
Illinois did not seek EPA approval at the time. Finally, in late
2012, Illinois submitted a proposed SIP revision to EPA seek-
ing approval of the changes to its emissions testing program.
After the requisite notice-and-comment period, during
which the state of Indiana objected to the proposed change,
EPA approved Illinois’s SIP revision in 2014. Indiana then
filed this petition for review, challenging the EPA approval.
Indiana argues, essentially, that the relaxation of Illinois’s
emissions testing program will decrease the likelihood that
the “Chicago area”—which includes two Indiana counties—
will achieve attainment with regard to ozone in the near fu-
ture. As evidence, Indiana points to its own scientific analy-
sis, which suggests that Illinois’s (unauthorized) use of a re-
laxed testing procedure was a but-for cause of a single
measured Chicago-area violation of the national ozone
standard in 2011. That single violation, in turn, resulted in
the Chicago area being classified as nonattainment. In other
No. 14-3214 3
words, Indiana argues that, if Illinois had not relaxed its test-
ing program, Chicago’s ozone levels would have met the na-
tional standard, and the area would now be in attainment.
According to Indiana, this evidence demonstrates that the
change in Illinois’s testing program will impermissibly “in-
terfere with … attainment,” and therefore that the SIP revi-
sion should have been disallowed by EPA pursuant to Sec-
tion 110(l) of the CAA. 42 U.S.C. § 7410(l).
On the preliminary question of whether a justiciable con-
troversy exists, we conclude that Indiana has standing to
bring this petition for review. However, because EPA did not
act arbitrarily and capriciously in approving the SIP revi-
sion, we deny Indiana’s petition.
I. Background
A. Regulatory Background
We summarized much of the relevant regulatory back-
ground for this case in Sierra Club v. U.S. EPA, 774 F.3d 383
(7th Cir. 2014):
The Clean Air Act (“CAA”), 42 U.S.C. § 7401 et
seq., establishes a comprehensive program for
controlling and improving the nation’s air
quality through both state and federal regula-
tion. Title I of the CAA charges the Environ-
mental Protection Agency Administrator with
identifying air pollutants that endanger public
health and welfare and with formulating Na-
tional Ambient Air Quality Standards
(“NAAQS”) that specify the maximum permis-
sible concentration of those pollutants in the
ambient air. Id. §§ 7408–09. Pursuant to the
4 No. 14-3214
CAA, EPA designates areas of the country as
either “attainment,” “nonattainment,” or “un-
classifiable” for specific NAAQS, based on
whether the area has attained the standard
and/or contributes to a nearby area’s nonat-
tainment. Id. § 7407(d).
Primary responsibility for ensuring that ambi-
ent air quality satisfies the NAAQS falls to the
states. Id. § 7407(a). Each state must draft a
state implementation plan (“SIP”) for each pol-
lutant, the review of which is conducted by
EPA according to the process outlined in sec-
tion 110(k) of the CAA. Id. § 7410(a), (k). Alt-
hough certain SIP requirements apply to an ar-
ea regardless of its designation, nonattainment
areas are subject to more regulations as com-
pared to attainment areas. See id. § 7501–15.1.
Relevant to this case, ozone is among the pollu-
tants that EPA has identified and, consequent-
ly, for which EPA has promulgated NAAQS.
See 40 C.F.R. pt. 50.
Id. at 386–87. At issue in this case is the 2008 ozone standard,
which is set to 0.075 parts per million (“ppm”) measured
over an eight-hour period. 40 C.F.R. § 50.15.
The purpose of SIPs is to “provide[] for implementation,
maintenance, and enforcement of [the NAAQS] in each air
quality control region (or portion thereof) within” a state. 42
U.S.C. § 7410(a)(1). The SIP must include “enforceable emis-
sion limitations and other control measures, means, or tech-
niques” that will be implemented so that each area in the
No. 14-3214 5
state can maintain or achieve attainment for a given pollu-
tant by the area’s statutory attainment deadline. Id.
§ 7410(a)(2)(A). States have primary responsibility in formu-
lating and revising the rules in their SIPs, but it is the EPA
that must review and approve SIP modifications. Id.
§ 7410(a), (k). If EPA determines that a SIP is complete and
meets all applicable requirements, “the Administrator shall
approve” the SIP. Id. § 7410(k)(2)–(3) (emphasis added).
However, Section 110(l) of the CAA—the “antibacksliding”
provision—states that EPA “shall not approve a revision of a
[SIP] if the revision would interfere with any applicable re-
quirement concerning attainment … or any other applicable
requirement” under the Act. 42 U.S.C. § 7410(I). Once it is
approved by EPA, a state rule embodied in a SIP becomes
enforceable federal law. See Gen. Motors Corp. v. United States,
496 U.S. 530, 540 (1990). EPA can enforce existing SIP provi-
sions through a variety of sanctions, 42 U.S.C. §§ 7413, 7509;
citizen suits to enforce certain SIP requirements, including
vehicle emissions inspection programs, are also authorized.
Id. § 7604(a)(1).
EPA has interpreted Section 110(l) to allow states to
demonstrate that a SIP revision will not “interfere” with at-
tainment in one of two ways: a state may either submit an air
quality analysis, or a state may identify “substitute equiva-
lent emissions reductions to compensate for any change to a
SIP approved program, as long as actual emissions in the air
are not increased.” 78 Fed. Reg. 68,378, 68,382 (Nov. 14,
2013). The second option allows a state seeking approval of a
SIP revision to provide EPA with (1) an estimate of how
much excess pollution will result from the revision and (2) a
list of contemporaneous pollution-control measures—such
as factory closures—along with estimates of the extent to
6 No. 14-3214
which each of these measures is expected to decrease pollu-
tion. If the pollution reductions listed in (2) are equal to or
greater than the pollution increase described in (1), EPA will
conclude that the SIP revision does not “interfere with”
NAAQS attainment. See id. (defining “equivalent emissions
reductions”). The purpose behind this interpretation, accord-
ing to EPA, is to allow states to pick and choose the manner
in which they seek to achieve attainment, as long as they can
show that net emissions are decreasing. In order to be con-
sidered in this calculus, however, substitute emissions re-
ductions must be permanent, enforceable, and quantifiable.
Id. They must also be “contemporaneous” with the proposed
SIP revision. Id. Finally, they must be “surplus,” meaning
that they have not been “otherwise relied on to meet air
quality attainment requirements in air quality programs re-
lated to” the SIP. EPA, Roadmap for Incorporating Energy Effi-
ciency/Renewable Energy Policies and Programs into State and
Tribal Implementation Plans (2012), available at
http://www.epa.gov/airquality/eere/pdfs/appendixC.pdf. In
other words, a given emissions reduction can only be credit-
ed once; that ensures that the same factory closure, for ex-
ample, cannot be used over and over again by the state as a
compensatory offset for multiple emissions increases.
The CAA requires that a state with areas designated as
nonattainment for ozone include motor vehicle inspection
and maintenance programs (“I/M programs”) in its SIP. 42
U.S.C. § 7511a. I/M programs help to reduce emissions from
automobiles, thereby improving air quality. Depending on
the severity of an area’s nonattainment (which can be “mar-
ginal,” “moderate,” “serious,” “severe,” or “extreme,”), the
state must implement either a “Basic” or a more stringent
“Enhanced” I/M program. Id. EPA has established recom-
No. 14-3214 7
mended I/M programs, but states are free to design and im-
plement alternate programs that meet or exceed minimum
performance standards. See 40 C.F.R. § 51.350–.372.
The “Alternate Low Enhanced” I/M performance stand-
ard is available to areas that meet certain emissions require-
ments. See id. § 51.351(g)(7). The model program elements
for that standard include testing of vehicles from all model
years 1968 and newer. But, a state may exempt certain model
years from its program, provided that the state demonstrates
that the performance standard will be met. Special model-
year exemptions must be included in a state’s SIP revision
proposals, along with an estimate of the exemptions’ effect
on emissions output.
B. Factual and Procedural Background
Prior to the events at issue in this case, Illinois’s vehicle
emissions testing program covered all cars from model years
1968 and newer.1 That program was part of Illinois’s EPA-
approved SIP. In 2005, however, Illinois changed its I/M pro-
gram to exempt pre-1996 model-year vehicles that met cer-
tain idle exhaust and gas cap pressure testing requirements.
See Vehicle Emissions Inspection Law of 2005, 625 Ill. Comp.
Stat. 5/13C. That law took effect in 2007. EPA approval of the
change was not initially sought or granted.
1 Illinois’s I/M program was required by federal law because the Chicago
area was classified as nonattainment, first, in the 1990s, under the one-
hour federal ozone standard, and then, in 2004, under the 1997 8-hour
ozone standard. In 2008, the 8-hour ozone standard was reduced from
0.08 to 0.075 ppm, and, as we describe above, the Chicago area was clas-
sified as nonattainment for that standard in 2012. The earlier ozone
standards are not at issue in this case.
8 No. 14-3214
In the years following the change to Illinois’s I/M pro-
gram, ozone levels in the Chicago areas were consistently
measured at levels below the 2008 eight-hour ozone
NAAQS; indeed, between 2008 and 2010, no ambient air
monitors in the area detected ozone levels exceeding the
NAAQS. In 2011, though, that pattern was broken when a
monitor located in Zion, Illinois measured an ozone concen-
tration of 0.076 ppm—just slightly (one part per billion)
above the permitted maximum. Based largely on that single
exceedance, EPA in 2012 designated the Chicago area as
“marginal nonattainment” for the 2008 revised eight-hour
ozone NAAQS. EPA included two Indiana counties—Lake
County and Porter County—in this nonattainment designa-
tion based on a finding that those counties “contributed” to
the recorded NAAQS violation.2 According to Indiana, the
Zion exceedance would not have occurred if Illinois had
maintained its prior I/M program rather than making the
2007 switch to a more relaxed program.
2 EPA’s inclusion of Lake and Porter Counties in the Chicago nonattain-
ment area was recently confirmed by the D.C. Circuit. Miss. Comm’n on
Envtl. Quality v. EPA, No. 12-1309, et al., 2015 WL 3461262, at *19–20
(D.C. Cir. June 2, 2015). That court found that it was not arbitrary and
capricious for EPA to conclude that emissions from those two counties
contributed to the violation in Zion. The D.C. Circuit also made clear that
“a ‘contributing’ county need not be a but-for cause of a violation in or-
der to warrant a nonattainment designation.” Id. at *19. As the D.C. Cir-
cuit noted, its holding does not necessarily mean that Indiana’s main
contention in this case—that Illinois’s unauthorized I/M change was a
but-for cause of the Zion violation—is factually incorrect. See id. at *19
n.12. That claim, the D.C. Circuit explained, was irrelevant to EPA’s
“contributing” determination: “the Illinois change … in no way dimin-
ished the contribution of the Indiana counties.” Id. at *20.
No. 14-3214 9
On November 29, 2012, almost six years after switching
to an I/M program that was technically not authorized by
federal law, Illinois finally submitted to EPA a proposed SIP
revision seeking approval of the new program. In its sub-
mission, Illinois included modeling demonstrating that the
new I/M program complied with the Alternate Low En-
hanced I/M performance standard. Illinois also argued that
the proposed revision would not interfere with attainment of
the NAAQS or any other CAA requirement. It did so by list-
ing substitute emissions reductions that would fully com-
pensate for the expected emissions increase caused by the
change to a more lenient I/M program. Specifically, Illinois
included a list of permanent shutdowns of facilities that
were permitted to emit ozone precursors; the list included
over 1,000 facilities that were closed between 2002 and 2012.
The emissions reductions from those closures, Illinois con-
cluded, more than outweighed the increased emissions
caused by the relaxation of the state’s I/M program.
EPA published a proposed approval of the Illinois SIP re-
vision on November 14, 2013. It concluded that the revised
I/M program would meet the Alternate Low Enhanced
standard, and that the listed substitute emissions reductions
were permanent, enforceable, contemporaneous, and sur-
plus, and would more than offset the increased emissions
coming from the change to the I/M program.
During the public comment period, Indiana submitted a
letter accompanied by a technical analysis. The details of this
letter are not overly important to the outcome of this case.
Basically, though, Indiana argued that the increased emis-
sions resulting from Illinois’s 2007 I/M program changes
caused the 2011 Zion ozone exceedance; without those
10 No. 14-3214
changes, the Zion violation would not have occurred. Be-
cause Illinois’s I/M change had, in its opinion, directly
caused nonattainment in the recent past, Indiana argued that
the change would likely also interfere with attainment in the
future, and therefore that EPA should reject the proposed
revision. Finally, Indiana argued that, because Illinois’s I/M
change had been implemented several years before the state
sought EPA approval, there was “extensive actual emissions
data” from those years, and therefore “any analysis [of the
proposed SIP revision’s impact on attainment] … should not
be based solely on emissions modelling or speculative re-
sults, but supported by actual monitoring data.”
Despite Indiana’s concerns, EPA approved Illinois’s SIP
revisions in a Final Rule effective September 12, 2014. In the
Final Rule, EPA specifically addressed Indiana’s comment.
Though EPA now argues that it rejected Indiana’s analysis in
part on scientific grounds, it did not indicate that in the Final
Rule. Indeed, it seemed to accept Indiana’s scientific conclu-
sion that, absent Illinois’s I/M program changes in 2007, the
Zion violation would not have occurred. Rather, EPA reject-
ed Indiana’s claim as irrelevant because its analysis was in-
complete. EPA explained that the fact that Illinois’s change to
the I/M program increased ozone levels—and that this in-
crease caused a NAAQS violation—was not determinative
under EPA’s interpretation of Section 110(l), which considers
only whether emissions increases caused by a SIP revision
are offset by substitute emissions reductions. Because Indi-
ana’s modeling did not take into account Illinois’s substitute
emissions reductions, EPA concluded that its conclusions
were irrelevant to the Section 110(l) analysis.
No. 14-3214 11
II. Discussion
A. Standing
We have jurisdiction under 42 U.S.C. § 7607(b)(1) to hear
challenges to SIP revisions for geographic areas within the
Seventh Circuit. See e.g., Sierra Club, 774 F.3d at 388. Howev-
er, after oral argument, we asked the parties to file supple-
mental briefs on the question of whether there is a justiciable
controversy in this case. We begin, therefore, by addressing
the issue of Indiana’s standing to bring this petition.
Article III of the Constitution limits the jurisdiction of
federal courts to “Cases” and “Controversies.” U.S. Const.
art. III, § 2. No “Case” or “Controversy” exists if the plaintiff
lacks standing to challenge the defendant’s alleged miscon-
duct. Lujan v. Defenders of Wildlife, 504 U.S. 555, 560 (1992).
The plaintiff bears the burden of establishing the required
elements of standing. Kathrein v. City of Evanston, Ill., 752 F.3d
680, 690 (7th Cir. 2014). The “irreducible constitutional min-
imum of standing” requires the plaintiff to show that he has
suffered or is imminently threatened with (1) a concrete and
particularized “injury in fact” (2) that is fairly traceable to
the challenged action of the defendant, and that is (3) likely
to be redressed by a favorable judicial decision. Lujan, 504
U.S. at 560–61. The traceability element is met only if the
plaintiff’s injury was sustained “as a consequence of” the
challenged conduct. Valley Forge Christian Coll. v. Am. United
for Separation of Church & State, Inc., 454 U.S. 464, 485 (1982).
Indiana’s principal argument is that it has standing be-
cause Illinois’s I/M program change allegedly caused the Zi-
on violation, which in turn caused two Indiana counties to
be currently classified as nonattainment. That is a judicially
12 No. 14-3214
cognizable injury—areas in nonattainment must meet more
stringent regulatory requirements—but it’s not traceable to
the action challenged in Indiana’s petition (EPA’s approval of
the Illinois SIP revision). That’s because the nonattainment
classification occurred in 2012, but EPA did not approve Illi-
nois’s SIP revision until 2014.3 The classification of Lake and
Porter Counties as nonattainment, therefore, could not have
occurred “as a consequence of” the challenged EPA action.
Moreover, Indiana has not alleged that the EPA approval has
in any way resulted in its counties being denied reclassifica-
tion into attainment in the time following approval. For ex-
ample, Indiana has not alleged that EPA has reassessed the
attainment status of the Chicago area since September 12,
2014 (the date EPA finally approved Illinois’s SIP revision),
and that the area was again classified as nonattainment be-
cause of emissions resulting from EPA approval of Illinois’s
new I/M program. The fact that Lake and Porter Counties
are currently classified as nonattainment therefore does not
give Indiana standing to bring this petition.
Indiana next argues that, because EPA approved Illinois’s
relaxed I/M program, more ozone will form in the Chicago
area, making it less likely that the area—including Lake and
Porter counties—will be classified as attainment in the near
future. That decreased likelihood of future attainment status
has two possible effects, the first of which does not give In-
diana standing, but the second of which does. First, Indiana
argues that continued nonattainment status will harm busi-
nesses within Lake and Porter Counties because they will
have to comply with more stringent regulatory require-
3 The SIP revision proposal was not even submitted to EPA until after
the 2012 nonattainment classification.
No. 14-3214 13
ments. That theory of standing, however, is blocked by the
doctrine of Massachusetts v. Mellon, 262 U.S. 447 (1923), which
held that “a State, as parens patriae, may [not] institute judi-
cial proceedings to protect citizens of the United States from
the operation of the statutes thereof.” Id. at 485–86. In Michi-
gan v. EPA, 581 F.3d 524, 529 (7th Cir. 2009), we held that
Michigan did not have standing to challenge an EPA rule-
making because the new rules did not “affect Michigan di-
rectly; rather, they affect emitting sources within Michigan
that want to construct new facilities or modify existing
ones.” Michigan, we held, did have an interest in protecting
its own economy; that interest could serve as the basis for
the state’s standing against some defendants, but not against
the United States due to the Mellon doctrine. Id.
The decreased likelihood that the Chicago area will
achieve attainment in the near future has a second effect,
however, that does give Indiana standing to bring this peti-
tion: if the Chicago area stays in nonattainment, the state of
Indiana itself will have to undertake actions to achieve at-
tainment. In other words, because Illinois’s SIP revision will
make it more difficult for the Chicago area to achieve at-
tainment, it is more likely that Indiana will be forced by fed-
eral law to take increased steps to aid the area in achieving
attainment. (Each geographic area has a statutory attainment
deadline, and SIPs must be formulated with the goal of
meeting that deadline. 42 U.S.C. § 7502(c)(1)–(6). So, if it ap-
pears that an area will not achieve attainment by that date,
federal law requires states to take more drastic measures to
meet the deadline.) For instance, Indiana might have to test
the emissions of more of its citizens’ cars, or engage in more
rigorous testing of those cars. That is a burden on the state
itself, and so the state has standing to sue, not to protect the
14 No. 14-3214
rights of its citizens as parens patriae, but rather to assert its
own rights. See Massachusetts v. EPA, 549 U.S. 497, 520 n.17
(2007).4
The burden of establishing standing, though, is on the
plaintiff, and EPA protests that Indiana has not adequately
pled this theory of standing. However, in its supplemental
brief, Indiana states: “[T]he failure to test older vehicles in
Illinois interferes with the attainment of ambient air quality
standards in the Chicago area. The resulting designation of
nonattainment increases the regulatory burden that now
must be borne in Northwest Indiana.” Given Indiana’s use of
the words “in Northwest Indiana” rather than “by North-
west Indiana,” and its subsequent citation to a case noting
that businesses in a nonattainment area must meet increased
regulatory requirements, it seems possible that Indiana was
making a parens patriae argument here, not an argument that
Indiana itself would have to do more to fight pollution. But
the statement can also be fairly read to argue that continued
nonattainment will increase the regulatory burden of the gov-
ernment in Northwest Indiana. We therefore conclude that
Indiana’s statement is enough to carry its burden of estab-
lishing standing.5
4 Because Indiana’s standing in this case does not depend on the 2012
nonattainment classification of Lake and Porter counties, the D.C. Cir-
cuit’s opinion in Mississippi Commission on Environmental Quality, 2015
WL 3461262, does not affect our standing analysis. Even if Indiana’s own
emissions may make future attainment more difficult, what matters to
the standing analysis is that achieving future attainment would be at
least marginally easier without the change to Illinois’s I/M program.
5Indiana also argues that it has standing because EPA’s approval of Illi-
nois’s relaxed I/M program will cause diminished air quality in Indiana.
No. 14-3214 15
B. EPA’s Approval of the Illinois SIP Revision
On the merits, Indiana challenges EPA’s approval of the
Illinois SIP revision. Pursuant to the Administrative Proce-
dure Act, EPA’s approval of a SIP revision must be upheld
unless it was “arbitrary, capricious, an abuse of discretion, or
otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A);
see also Citizens Against Ruining the Env’t v. EPA, 535 F.3d 670,
674 (7th Cir. 2008) (holding that, when the CAA does not
provide a standard for reviewing specific EPA decisions, the
standard of review is supplied by the Administrative Proce-
dure Act). Two other principles of agency deference are po-
tentially at play in this case. First, when an agency’s decision
is based on an interpretation of statutory language subject to
notice and comment, the agency’s interpretation is afforded
Chevron deference. See Chevron, U.S.A., Inc. v. NRDC, 467 U.S.
That would normally give the state standing as parens patriae, but not
against the United States. See Mellon, 262 U.S. at 485–86. States have
“special solicitude” to sue the United States, however, if a quasi-
sovereign interest of the state is at stake. See Massachusetts, 549 U.S. at
520–23 (holding that Massachusetts had standing to sue EPA based on an
injury to state-owned coastal lands). EPA, pointing to Michigan v. EPA,
581 F.3d at 529, argues that the “special solicitude” analysis does not ap-
ply to a suit in which a state’s alleged injury is diminished air quality. In
that case, though, we noted that “[i]n contrast to [Massachusetts v. EPA],
in which Massachusetts’s coastal lands were threatened by rising sea
levels, Michigan’s air can only benefit from” the challenged agency action.
Michigan, 581 F.3d at 529 (emphasis added). If anything, that statement
suggests that Michigan might have had standing had it alleged that the
challenged action would harm its air quality. On the other hand, while
Massachusetts actually owns some of its coastal property, Indiana does
not own its air; the state sovereign interest, therefore, was much clearer
in Massachusetts than it is here. Regardless, we need not resolve this issue
because we find that Indiana has standing on other grounds.
16 No. 14-3214
837 (1984). Under Chevron, if “Congress has directly spoken
to the precise question at issue … the court … must give ef-
fect to the unambiguously expressed intent of Congress.” Id.
at 842–43. If, however, “the statute is silent or ambiguous
with respect to the specific issue,” a reviewing court must
defer to the agency’s interpretation if it is reasonable. Id. at
843–44. Finally, when reviewing an agency’s scientific and
technical determinations, “a reviewing court must generally
be at its most deferential.” Balt. Gas & Elec. Co. v. NRDC, 462
U.S. 87, 103 (1983).
Essentially, Indiana argues that it was arbitrary and ca-
pricious for EPA to conclude that Illinois’s SIP revision
would not “interfere with” attainment even though Indiana
put forth scientific data demonstrating that the conduct em-
bodied in that revision had already caused ozone levels to
exceed the NAAQS. Once again though, as with its central
standing argument, Indiana overlooks the importance of
timing. When deciding whether to approve Illinois’s SIP re-
vision, EPA was required to determine whether the revision
would, going forward, interfere with attainment. The fact that
a policy may have prevented attainment in the past does not
necessarily mean that it will do the same in the future. In
fact, the circumstances of this case suggest that the Chicago
area will soon achieve attainment regardless of the EPA ap-
proval. The area is currently classified as nonattainment
based solely on a single, vanishingly small violation of the
NAAQS. And, because EPA confirmed that Illinois’s I/M
change is part of an overall net decrease in ozone precursor
emissions, it seems probable that ozone levels will decrease,
and the area will reach attainment.
No. 14-3214 17
Indiana speaks of the Illinois I/M change as if it is forever
and irretrievably cursed due to its illegitimate beginnings
and its role in causing Chicago’s current nonattainment sta-
tus. But Illinois’s long delay in seeking approval of the
change is irrelevant to this case; Indiana has pointed to no
authority suggesting that EPA should (or even can) reject a
SIP revision due to a state’s premature enactment of the
change embodied in that revision. Remember: the CAA re-
quires that “the [EPA] shall approve” a SIP revision that
meets all applicable requirements, and none of those re-
quirements relate to the state’s compliance with its current
SIP guidelines. 42 U.S.C. § 7410(k) (emphasis added). And by
focusing so heavily on the I/M program’s past as the cause of
current nonattainment, Indiana completely loses sight of the
fact that EPA’s job was to assess the future. If EPA had been
charged with determining whether the proposed Illinois SIP
revision had ever “interfered with” NAAQS attainment, In-
diana would have a compelling argument. But that was not
the relevant issue, and past interference does not necessarily
tell us much about what will happen in the future.
Moreover, Indiana’s argument overlooks EPA’s interpre-
tation of Section 110(l), which allows states to make emis-
sions-increasing SIP revisions if they identify substitute
emissions reductions such that net emissions are not increas-
ing. Indiana concedes that it has “no general objection to the
use of substitute control measures to demonstrate noninter-
ference with the NAAQS when seeking approval for a pro-
spective SIP change in a nonattainment area.” Indeed, all
three courts of appeals that have addressed the issue agree
that EPA’s interpretation of Section 110(l) is reasonable under
Chevron. See Ala. Envtl. Council v. EPA, 711 F.3d 1277, 1292–93
(11th Cir. 2013); Galveston-Hous. Ass’n for Smog Prevention v.
18 No. 14-3214
EPA, 289 Fed. App’x 745, 754 (5th Cir. 2008); Ky. Res. Council,
Inc. v. EPA, 467 F.3d 986, 995 (6th Cir. 2006). Indiana argues,
however, that this interpretation of Section 110(l) should not
apply in a case, such as this one, where measurable emis-
sions data has been generated due to a state’s premature im-
plementation of a SIP change.
Indiana misunderstands the nature of Chevron deference.
When a court defers to an agency’s interpretation of an am-
biguous statutory word or phrase, it essentially concludes
that the agency’s interpretation of that word or phrase is rea-
sonable as a general matter. It is not simply saying that a cer-
tain interpretation is reasonable with respect to the case be-
fore the court. Put differently, the court determines that the
agency’s regulation is “facially” valid, rather than valid “as
applied.” Viewed this way, it makes little sense to say that an
interpretation is, in general, reasonable, but should not be
followed in certain circumstances. EPA need not then re-
determine for each new situation that arises how best to in-
terpret those words. If anything, it would be arbitrary for
EPA to now refrain from applying that definition to Illinois’s
SIP revision.
Even if Indiana were correct about how Chevron defer-
ence operates, it has not convinced us that it was unreasona-
ble for EPA to use its substitute emissions reductions analy-
sis in this case. Indiana argues that, because there is data
from the Illinois I/M program’s pre-approval years, EPA
should have required Illinois to base its application on that
actual real-world data. (Recall that a state may support its
SIP revision application either by identifying substitute
emissions reductions or by providing EPA with an air quali-
ty analysis reflecting real-world data.) The main piece of da-
No. 14-3214 19
ta Indiana points to is its estimate of the increased emissions
resulting from the relaxed I/M program. But Indiana’s analy-
sis is missing data regarding the other side of the ledger—
Illinois’s substitute emissions reductions. Indiana concedes
that it is reasonable for EPA to consider the net change in
emissions when it is estimating the future effects of a SIP re-
vision; it does not explain why net emissions are irrelevant
when EPA’s analysis relies on real-world analysis. The only
other data that Indiana points to is the Zion ozone exceed-
ance. EPA addressed that data in its Final Rule when it re-
sponded to Indiana’s comment by stating, “EPA believes
that, had the commenter modeled the ozone impact of the
combined increased emissions from the I/M revision and the
decrease in emissions from the offsetting emissions reduc-
tions, the commenter would have modeled a net decrease in
peak downwind ozone concentrations.” 79 Fed. Reg. 47,377,
47,379 (Aug. 13, 2014). So, even if Indiana were right that
EPA should have looked to real-world data, Indiana has not
shown that the agency’s conclusion would have been any
different.
We conclude, therefore, that EPA did not act arbitrarily
and capriciously when it approved Illinois’s SIP revision.
EPA complied with its concededly reasonable interpretation
of Section 110(l) in determining that Illinois’s changed I/M
program would not interfere with NAAQS attainment. That
the program existed for multiple years pre-approval and ar-
guably led to the classification of the Chicago area as nonat-
tainment in the past does not disturb the validity of EPA’s
determination that, going forward, the increased emissions
from the program will be outweighed by other emissions re-
ductions.
20 No. 14-3214
Two other points merit brief discussion. First is EPA’s de-
termination that Illinois’s substitute emissions reductions
were “contemporaneous” with the 2007 change to Illinois’s
I/M program.6 Illinois’s list of substitute emissions reduc-
tions includes factory closures from the ten-year span (2002–
2012) surrounding 2007. Indiana contends that this was
problematic for two reasons. First, Indiana argues that it is
unreasonable to consider that long of a time span as “con-
temporaneous.” Indeed, though EPA has not set a definitive
rule regarding how near in time an emissions reduction
must be to qualify as “contemporaneous,” it has stated that
it is reasonable to interpret reductions that occur within one
year in either direction of the increase (i.e., a two-year span)
as contemporaneous. We agree with Indiana that the use of
this ten-year period is troubling: it stretches the meaning of
“contemporaneous” to its breaking point. We share Indiana’s
concern that, “Apparently, … ‘contemporaneous’ can be
whatever length of time (short or long) it takes to justify
EPA’s decision (good or bad).”
As it turns out, however, EPA’s reliance on the ten-year
list of factory shutdowns was harmless in this case. EPA ar-
6 Indiana’s arguments regarding EPA’s contemporaneousness conclu-
sion are likely waived, as it did not raise the issue in its comment letter.
See NRDC v. Jackson, 650 F.3d 662, 666 (7th Cir. 2011) (rejecting a chal-
lenge to SIP revisions raised for the first time in a petition for review be-
cause it “was not pointed out to the EPA during the rulemaking and so
has not been preserved for judicial review”). We address them briefly,
however, to provide necessary clarification.
Indiana’s challenge to EPA’s conclusion that the substitute emissions
reductions were surplus is also likely waived for the same reason. How-
ever, because this argument is hardly developed in Indiana’s briefs and,
as best we can tell, lacks merit, it does not warrant further consideration.
No. 14-3214 21
gues, without contradiction by Indiana, that the substitute
emissions reductions from just the years 2005 and 2006 were
more than sufficient to offset the increases from the I/M pro-
gram change. Indiana does not attempt to argue that EPA’s
use of that two-year span for determining contemporane-
ousness is unreasonable.
Indiana also argues that it was unreasonable for EPA to
consider substitute emissions reductions occurring after 2007
as “contemporaneous” because those reductions were likely
not contemplated by Illinois when it implemented the I/M
change, and so likely could not have been relied upon by Il-
linois had it submitted its SIP revision when it was supposed
to, in 2007.
This argument fails for two reasons. First, as we noted
previously, Illinois did not need to rely on post-2007 reduc-
tions because the reductions from 2005 and 2006 were suffi-
cient to offset any emissions increase caused by the I/M pro-
gram change. Second, the inclusion of post-2007 reductions
was harmless for another reason, though some explanation
is necessary to see why. In determining whether Illinois’s SIP
revision was offset by emissions decreases, EPA calculated
year-by-year comparisons of the increased emissions caused
by the I/M change and the decreased emissions caused by
listed factory closures. When analyzing a specific year, how-
ever, EPA did not look only at the factory closures from that
year. Rather, it aggregated the emissions reductions from all
of the previous “contemporaneous” factory closures. For the
year 2008, for example, EPA compared the increased emis-
sions caused by the I/M program relaxation that year to the
emissions reductions resulting from factory closures in the
years 2002–2008. That’s because a factory that closed in 2002
22 No. 14-3214
remained closed in 2008; a factory closure eliminates emis-
sions from that factory not just in the year that it closes, but
for each following year as well.7 Also important is that the
extra emissions caused Illinois’s I/M change are decreasing
over time—the increased pollution caused by the relaxed
program is expected to be less each year, because there are
expected to be fewer exempted vehicles (those from model
years 1968–1995) on the road as time goes on. So, because
emissions reductions outweighed the emissions increase in
2007—the year in which the increase was the largest—those
reductions necessarily also outweigh the increases in each
subsequent year, even if the post-2007 factory closures are ig-
nored.8
Finally, we briefly address EPA’s argument that it is owed
special deference for certain “scientific determinations” it
made in this case. See Balt. Gas & Elec. Co., 462 U.S. at 103. We
agree that some of its determinations are due this added
deference, but EPA’s argument goes too far. For example,
EPA argues that its conclusion that Illinois’s substitute emis-
sions reductions were contemporaneous was a “technical de-
termination.” But, as Indiana points out, EPA has not
demonstrated that there is anything scientific or technical
about its determination of the length of the contemporane-
ousness window. EPA says that determining the contempo-
raneousness time frame is “a case-specific exercise that
7 There may be a point at which it becomes unreasonable for EPA to con-
tinue considering emissions reductions from the distant past. That possi-
bility is not implicated in this case, however, as EPA’s analysis runs only
through the year 2025.
8 This is true even if the pre-2005 factory closures are omitted as not con-
temporaneous.
No. 14-3214 23
should be tailored to the facts of the particular rulemaking,”
but there is no suggestion that this tailoring has anything to
do with science—for example, information regarding how
long certain pollutants remain in the air or how emissions
from different years may interact—rather than policy. In the
absence of any sign that the contemporaneousness inquiry is
based on science, it is not owed any extra deference.
Relatedly, EPA claims that it rejected Indiana’s modeling
of the Zion violation on scientific grounds. But EPA has not
pointed to anything that is scientifically wrong with that
analysis. Rather, the analysis was rejected because Indiana
failed to take into account the effects of Illinois’s substitute
emissions reductions, thereby rendering Indiana’s analysis
irrelevant to the Section 110(l) inquiry that EPA has—as a
matter of policy—deemed appropriate. With minor excep-
tions that are noted in EPA’s briefs (but not in its response to
Indiana’s comment in the Final Rule), EPA does not chal-
lenge the science behind Indiana’s conclusion that, absent
the change to Illinois’s I/M program, the Zion violation
would not have occurred. So, EPA’s rejection of Indiana’s
conclusions for non-scientific reasons does not merit special
deference.9 Even without giving added deference to these
9 Indiana argues that EPA’s conclusion that Illinois’s substitute emissions
reductions outweigh the increased emissions from the I/M program
change does not merit special deference either, because EPA only en-
gaged in the “simple arithmetic” of adding up Illinois’s various emis-
sions reductions. However, EPA did not simply accept Illinois’s claims
without confirming that Illinois’s emissions estimates were scientifically
valid. Rather, it determined that Illinois “used reasonable methods and
appropriate models in estimating the emissions effects of the program
changes.” 78 Fed. Reg. at 68,382. That is a scientific determination, and
therefore EPA’s conclusion that Illinois’s substitute emissions reductions
24 No. 14-3214
determinations, however, we hold that EPA did not act arbi-
trarily and capriciously by approving Illinois’s SIP revision.
III. Conclusion
For the foregoing reasons, Indiana’s petition for review is
DENIED.
outweighed the increased emissions from the I/M program is entitled to
receive added deference.