United States of America et al v. United States Steel Corporation

Filing 54

OPINION AND ORDER granting in part and denying in part 14 Motion to Dismiss for Failure to State a Claim; granting 34 Motion to Correct and/or Clarify Transcript of Oral Argument Held on 1/24/2013. Signed by Chief Judge Philip P Simon on 8/21/2013. (rmn)

UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF INDIANA
HAMMOND DIVISION
UNITED STATES OF AMERICA, et al.,

)
)
)
)
)
)
)
)
)

Plaintiffs,
v.
UNITED STATES STEEL CORPORATION,
Defendants.

2:12-cv-304

OPINION and ORDER
The Environmental Protection Agency and three similar state entities from Indiana,
Michigan, and Illinois initiated this action against U.S. Steel in an 18-count complaint. (For
simplicity’s sake, I will refer to the Plaintiffs in this case collectively as the EPA). The EPA
alleges that three of U.S. Steel’s plants – one in each of the three states – are and have been
violating the Clean Air Act. U.S. Steel says its done no such thing, but the merits of the case
will be decided later. The only thing presently before me is whether Count 1 and parts of Count
3 are barred by the applicable statute of limitations and must therefore be immediately dismissed.
As explained in detail below, I only buy half of that argument. U.S. Steel’s Motion to Dismiss
will therefore be granted in part and denied in part.
BACKGROUND
In 1990, U.S. Steel performed a “reline” of its No. 4 furnace at its Gary Works facility in
Gary, Indiana. A reline is the process whereby the furnace is completely shut down and the
interior lining is updated. The legal issues raised in U.S. Steel’s motion to dismiss all really
emanate from this 1990 relining. One might reasonably wonder why a court in the year 2013 is

1

dealing with actions taken by a company some 23 years earlier; U.S. Steel is wondering the same
thing. In all events, U.S. Steel believes it could do the relining of the furnace without a permit,
but the EPA disagrees. To understand the parameters of this disagreement, we have to dive into
obscure world of the Clean Air Act (“CAA”), its regulations, and the myriad of acronyms that
flow from them.
Here’s a broad outline of the parts of the CAA applicable to this case. The CAA
establishes two types of air permits: construction permits and operating permits. Both types of
permits are at issue in this case. The construction permitting program emerged from the 1977
amendments to the CAA which required all new “sources” – factories, power plants, refineries,
etc. – to undergo a “New Source Review” (NSR). Under the NSR process, all post-1977 sources
are required to meet various emission limits for pollutants. NSR consists of two programs –
prevention of significant deterioration in air quality (PSD) and non-attainment with air quality
standards (NA/NSR). Both of those programs are at issue in this case, but the differences are
fairly immaterial to U.S. Steel’s motion to dismiss, so to simplify matters I’ll just refer to the
whole construction permitting process as “NSR.”
Under the NSR program, the new source must obtain a pre-construction permit from the
EPA, demonstrate that the construction will not increase emissions above a certain threshold,
and show that facility operations are in continuous compliance with the best available control
technology (BACT). The NSR permit then specifies what construction is allowed, what
emission limits must be met, and how often the emissions source must be operated.
In addition to these construction permits, most sources also have to obtain operating
permits, which are known as Title V permits. A Title V permit grants a source permission to

2

operate, and it includes all air pollution requirements that apply to the source, including
emissions limits and monitoring, record keeping, and reporting requirements.
Most of this CAA permitting – both NSR and Title V – is actually done at the state level.
The EPA establishes the basic requirements for the CAA, and states can then take those
minimum thresholds and develop their own unique requirements and procedures in what is called
a State Implementation Plan (“SIP”). Indiana has codified its SIP at Titles 326 and 327 of the
Indiana Administrative Code. Indiana’s SIP generally seems to mirror the federal requirements,
including requirements that new sources must obtain a pre-construction NSR permit and install
BACT.
All of this applies to new sources, those built after 1977. The Gary Works plant,
however, has a lot of dust on it; steel production began there in the early part of the last century.
That doesn’t take the plant out of the regulatory woods, however, because the CAA says that
even pre-1977 plants must comply with the NSR permitting program any time a “major
modification” is made to them. Enter the 1990 relining project. U.S. Steel concluded that the
reline wasn’t a major modification, so it didn’t bother seeking a preconstruction NSR permit.
Then, in 1996, U.S. Steel applied for a Title V operating permit, which was eventually issued by
the state of Indiana. The EPA now argues that the 1990 reline was a major modification, that
U.S. Steel thus should have sought a preconstruction NSR permit, and that consequently the
Title V permit was improperly granted and should be invalidated. More specifically, the EPA
alleges in Count 1 that U.S. Steel performed a major modification, which violated the CAA and
Indiana’s SIP by operating Gary Works without an NSR permit and without BACT. In Count 3,

3

the EPA alleges that U.S. Steel is operating Gary Works in violation of the Title V permitting
program under the CAA and Indiana’s SIP.
Whether or not the 1990 reline was a major modification is likely to be a difficult
question to answer. See, e.g., U.S. v. Midwest Generation, LLC, __ F.3d __, 2013 WL 3379319
at *1 (7th Cir. July 8, 2013) (“The question ‘how much repair or change requires a permit?’ has
been contentious and difficult.”). But that substantive question is not yet at issue here. Instead,
the key question in U.S. Steel’s motion to dismiss is procedural: Did the EPA blow the statute of
limitations by waiting until 2012 to file a lawsuit about the 1990 reline?
ANALYSIS
A motion to dismiss pursuant to Rule 12(b)(6) “challenges the sufficiency of the
complaint to state a claim upon which relief may be granted.” Hallinan v. Fraternal Order of
Police of Chi. Lodge No. 7, 570 F.3d 811, 820 (7th Cir. 2009). A plaintiff’s failure to adhere to a
statute of limitations is an affirmative defense and “is rarely a good reason to dismiss under Rule
12(b)(6).” Reiser v. Residential Funding Corp., 380 F.3d 1027, 1030 (7th Cir. 2004). At the
motion to dismiss stage, “the question is only whether there is any set of facts that if proven
would establish a defense to the statute of limitations[.]” Clark v. City of Braidwood, 318 F.3d
764, 768 (7th Cir. 2003). A plaintiff need not anticipate affirmative defenses in his complaint in
order to survive a motion to dismiss. United States v. Lewis, 411 F.3d 838, 842 (7th Cir. 2005)
(citing Gomez v. Toledo, 446 U.S. 635, 640 (1980)); see also Indep. Trust Corp. v. Stewart Info.
Serv. Corp., 665 F.3d 930, 935 (7th Cir. 2012). “The exception occurs where ... the allegations of
the complaint itself set forth everything necessary to satisfy the affirmative defense, such as

4

when a complaint plainly reveals that an action is untimely under the governing statute of
limitations.” Lewis, 411 F.3d at 842 (citing Leavell v. Kieffer, 189 F.3d 492, 495 (7th Cir.1999)).
The applicable statute of limitations here is 28 U.S.C. § 2462, which states:
Except as otherwise provided by Act of Congress, an action, suit or proceeding
for the enforcement of any civil fine, penalty, or forfeiture, pecuniary or
otherwise, shall not be entertained unless commenced within five years from the
date when the claim first accrued if, within the same period, the offender or the
property is found within the United States in order that proper service may be
made thereon.
Count 1 and Count 3 of the Complaint seek both injunctive relief and civil penalties. Section
2462 explicitly applies to the latter remedy, but it is less clear whether it applies to the former.
As explained in detail below, I ultimately conclude that Section 2462 does not apply to claims
for injunctive relief, but does apply to claims for civil penalties. I’ll start by analyzing the
question of civil penalties, and then turn to the more difficult issue of injunctive relief.
Before moving to this analysis, however, I need to clear up one issue with respect to how
Counts 1 and 3 are interrelated. The EPA argues that the Counts should be independently
analyzed: “Regardless of how the Court decides the issue of whether NSR violations are
ongoing, as Plaintiffs urge, or one-time past violations, as Defendant argues, the Court should
deny Defendant’s Motion with regard to Plaintiffs’ Third Claim.” [DE 20 at 38.]
I disagree, and will instead analyze both Counts in the same way here. Here’s why. The
main focus of the motion to dismiss really has to do with Count 1 of the Complaint, which is
entirely centered on the NSR permitting process related to the 1990 reline. Count 3 deals with
the Title V permitting process. The 1990 reline is only one part of Count 3, and U.S. Steel has
only moved to dismiss the paragraphs in Count 3 that deal specifically with the 1990 reline (not
all of Count 3). The central conceit of Count 3 is that U.S. Steel’s Title V permit has numerous
5

deficiencies, one of which is that U.S. Steel’s application for the permit lacked information about
the 1990 reline. More specifically, Count 3 asserts that since U.S. Steel did not have an NSR
preconstruction permit for its 1990 reline, its 1996 application for its Title V permit was
deficient.
The problem with this position is that the Title V permit is only deficient in this way if
U.S. Steel was required to apply for an NSR preconstruction permit in 1990. U.S. Steel argues
adamantly that the statute of limitations bars the EPA from even challenging the 1990 decision
not to apply for the permit. I agree with U.S. Steel that if it is correct that under the statute of
limitations “the Government cannot challenge the lack of a [NSR preconstruction] permit at this
late date, [then] it cannot assert a claim that U. S. Steel failed to disclose that such a permit
should have been sought” in the context of its Title V application. [DE 14-1 at 10.] That’s an
important point, because it means that the forthcoming statute-of-limitations analysis applies not
just to Count 1, but also to the paragraphs of Count 3 that have to do with the 1990 relining.
I.

Civil Penalties
The issue of whether the EPA is barred from bringing claims for civil penalties based on

the 1990 reline is fairly straightforward. Section 2462 clearly applies to claims for civil
penalties, which means all such claims must be brought within 5 years of their accrual. U.S.
Steel argue that since the permitting violations are alleged to have occurred in 1990, the
government had to bring any civil penalty claims by 1995. Any claims brought thereafter are
barred by the statute of limitations.
The logic of this is simple and undoubtedly correct. The EPA, nevertheless, has grasped
for various statute-of-limitations straws that would extend their time to file suit beyond five

6

years. First, they argued that the discovery rule extends the limitations period, but they have
now rightly abandoned that argument in the face of the Supreme Court’s recent decision in
Gabelli et al. v. Securities and Exchange Commission, __ U.S. __, 133 S.Ct. 1216 (2013).
The EPA also argues that the statute of limitations was extended beyond 1995 because
there was not just one violation in 1990, but rather a series of ongoing and continuous violations.
The theory is that there is a new violation (and thus a new 5-year limitations period) every day
since 1990 that U.S. Steel was out of compliance with the federal NSR permitting requirements
and the Indiana SIP permitting requirements. This “continuing violation” argument is a
somewhat more complicated question as various circuit and district courts have split on the
issue. But the question got a whole lot easier to answer (for me, at least) a few weeks ago, as the
Seventh Circuit addressed this very issue in U.S. v. Midwest Generation, LLC, __ F.3d __, 2013
WL 3379319 (7th Cir. July 8, 2013).1 In Midwest Generation the Court explained that “[t]he
continuing-violation argument is that every day a plant operates without a § 7475 permit is a
fresh violation of the Clean Air Act.” Id. at *2. The Court definitively rejected that argument:
“Nothing in the text of § 7475 even hints at the possibility that a fresh violation occurs every day
until the end of the universe if an owner that lacks a construction permit operates a completed
facility. . . . [Thus,] operating a new or modified plant, despite failure to obtain a construction
permit, is [not] a new violation of § 7475(a).” Id.

1

After Midwest Generation came out, the United States filed a motion [DE 49] that
sought to defer ruling on U.S. Steel’s motion to dismiss on the theory that it might appeal the
Seventh Circuit’s ruling. Magistrate Judge Rodovich rightly denied that motion. If the United
States decides to appeal and if the ruling in Midwest Generation is altered as a result, then I will
obviously revisit my holding here. In the meantime, U.S. Steel has waited long enough for a
ruling on its motion, which I purposely put on hold pending the Seventh Circuit’s ruling in
Midwest Generation.
7

Given Midwest Generation, it is clear that the EPA’s civil penalty claims based on
federal law (i.e., 42 U.S.C. § 7475) are barred by the statue of limitations. Nevertheless,
Midwest Generation did leave one small door open – whether U.S. Steel violated Indiana’s state
regulations (the aforementioned “SIP”). Midwest Generation explicitly left this question for the
district court to figure out. Here’s how the Seventh Circuit explained the issue:
Plaintiffs stress that § 7475(a)(4) says that newly built or modified sources are
“subject to” the need for the best available control technology. That obligation,
they insist, continues after the construction work is done, which leads them to say
that National Parks and Conservation Association Inc. v. Tennessee Valley
Authority, 480 F.3d 410 (6th Cir. 2007), disagrees with the eighth and eleventh
circuits [and, now, the seventh circuit]. Yet the sixth circuit’s decision rests on
Tennessee statutes and implementation plans that require certain sources to use
the best available control technology, while § 7475 deals only with conditions
precedent to construction or modification. Perhaps an Illinois statute, regulation,
or implementation plan provides that any plant “subject to” BACT by virtue of §
7475(a)(4) must use it in operation, but any claim of that sort remains pending in
the district court. What BACT entails is plant-specific. 40 C.F.R. § 52.21(b)(12).
All we have for decision is a claim directly under § 7475. Plaintiffs maintain that
415 ILCS 5/9.1(d)(2) works the same way as the Tennessee requirements that the
sixth circuit considered. To the extent that this contention is independent of §
7475, we leave it to the district judge in the first instance.
Id. at *3. Take this directive, substitute “Indiana” for “Illinois,” and it is now clearly left to me
in this case to decide, “in the first instance,” whether the Indiana regulations at issue here work
“the same way as the Tennessee requirements that the sixth circuit considered” in National
Parks. Id.
With these marching orders, I have to comb the depths of Tennessee environmental law
and see how it compares with Indiana. But before getting there it’s worth pointing out that the
Sixth Circuit’s decision in National Parks is clearly an outlier.2 Three other circuits – the

2

And not to be confused with the Eleventh Circuit’s decision in a case of the same name.
8

Seventh, Eighth, and Eleventh Circuits – have all held that there is no “continuing violation” of
federal law when an emitting source fails to properly follow the NSR permitting process and
fails to install BACT. See, respectively, Midwest Generation, 2013 WL 3379319 at *2-3; Sierra
Club v. Otter Tail Power Co., 615 F.3d 1008, 1017-18 (8th Cir. 2010); National Parks and
Conservation Association, Inc. v. Tennessee Valley Authority, 502 F.3d 1316, 1324-25 (11th Cir.
2007). The Seventh Circuit dodged the question as to whether it is a continuing violation of
state law to fail to properly follow the NSR process and to install BACT, but both the Eighth and
Eleventh Circuits analyzed the state law at issue in their cases (South Dakota and Alabama,
respectively) and found that the continuing violation theory was inapplicable.
So the Sixth Circuit’s decision in National Parks is now the one case where a circuit
court upheld the continuing violation theory. How did it get there? Under the Sixth Circuit’s
reasoning, two provisions of Tennessee law dictated its result. Here’s the court’s analysis of one
of those provisions, Tenn. Comp. R. & Regs. § 1200-3-9-.01(1)(e):
[U]nder the Tennessee SIP, the obligation to obtain an appropriate permit is
ongoing, and applies even to those sources that did not obtain the appropriate
permits before construction: “In the case where a source or modification was
constructed without first obtaining a construction permit, a construction permit
may be issued to the source or modification to establish as conditions of the
permit, the necessary emissions limits and requirements to assure that these
regulatory requirements are met.” Id. § 1200-3-9-.01(1)(e). In other words, while
§ 1200-3-9-.01(1)(a) contains the Tennessee SIP’s pre-construction permitting
requirement, § 1200-3-9-.01(1)(e) establishes that the duty to obtain a
construction permit containing the proper emissions limits is ongoing, even
post-construction.
National Parks, 480 F.3d at 419. So because this specific provision of Tennessee’s SIP required
ongoing NSR permitting requirements, rather than just preconstruction requirements, the
violation of Tennessee’s SIP was ongoing.

9

The Eighth and Eleventh Circuits agreed with this reading of Tennessee’s SIP, but also
contrasted it to the other state SIPs that did not contain similar provisions. As the Eighth Circuit
explained:
Because [Tenn. Comp. R. & Regs. § 1200-3-9-.01(1)(e)] would allow a facility to
obtain a determination of BACT limits through a postconstruction permitting
process, the Sixth Circuit’s conclusion that in Tennessee there were ongoing
duties both to obtain a permit and to employ BACT made sense. Sierra Club has
pointed to no similar rule in South Dakota, and its contention that there is an
independent ongoing duty to apply BACT is therefore not persuasive.
Sierra Club, 615 F.3d at 1017. In its National Parks decision, the Eleventh Circuit reached an
identical conclusion with respect to Alabama law. National Parks, 502 F.3d at 1325. Just as in
those cases, the EPA here has pointed to no similar rule in Indiana’s SIP that requires ongoing
NSR permitting. Thus, the Sixth Circuit’s line of reasoning is not applicable here and there is no
“continuing violation” under Indiana’s SIP.
The Sixth Circuit believed it had another, independent reason for applying the continuing
violation doctrine to Tennessee’s SIP, however. The court pointed to this provision: “A major
modification shall apply best available control technology for any pollutant for which it would
result in a significant net emissions increase at the source.” Tenn. Comp. R. & Regs. §
1200-3-9-.01(4)(j)(3). The Sixth Circuit found that “[t]his provision, by its own terms, creates
an ongoing obligation to apply BACT, regardless of what terms a preconstruction permit may or
may not contain.” National Parks, 480 F.3d at 418. As a result, the defendant had no statute-oflimitations defense because “failing to apply BACT is actionable, and this cause of action
manifests itself anew each day a plant operates without BACT limits on emissions.” Id. at 419.
Indiana’s SIP has a nearly identical “shall apply” provision. Indiana’s provision states:
“A major modification shall apply best available control technology for each regulated NSR
10

pollutant for which the modification would result in a significant net emissions increase at the
source. This requirement applies to each proposed emissions unit at which a net emissions
increase of the pollutant would occur as a result of a physical change or change in the method of
operation in the unit.” 326 IAC 2-2-3. The EPA thus argues that I should follow the Sixth
Circuit’s lead to find that this provision creates an ongoing obligation to apply BACT and that
there is a continuing violation of Indiana’s SIP every day it is not applying BACT.
The Sixth’s Circuit’s reasoning on this point, however, was heavily criticized by both the
Eighth and Eleventh Circuit. The Eighth Circuit, for instance, had this detailed analysis:
Emphasizing the imperative phrase “shall apply,” the Sixth Circuit interpreted . . .
Tennessee’s SIP as creating “an ongoing obligation to apply BACT, regardless of
what terms a preconstruction permit may or may not contain.” Nat’l Parks 6th
Cir., 480 F.3d at 418. We cannot agree with this interpretation. The context of
[40 C.F.R. ] § 52.21(j)(3) shows that the command to apply BACT is not a
freestanding requirement. Rather, it is tied specifically to the construction
process. Subsection (j)(3) “applies to each proposed ... unit” (emphasis added);
see Nat’l Parks 11th Cir., 502 F.3d at 1324–25. . . . In light of these provisions, §
52.21(j)(3) is best understood as requiring that BACT limits be incorporated into
a facility’s construction plans and [NSR] permits, not as establishing an ongoing
duty to apply BACT independent of the permitting process.
This conclusion is bolstered by the practical nature of BACT. BACT limits are
tailored to each facility “on a case-by-case basis” during the [NSR] permitting
process. 42 U.S.C. § 7479(3); see also 40 C.F.R. § 52.21(b)(12). They are thus
inextricably linked to the permitting process. Having determined that there is no
ongoing duty to obtain [NSR] permits, it would make little sense to conclude that
Otter Tail must nevertheless abide by the BACT limits which are a product of
such permits. . . . The duty to obtain a [NSR] permit and the duty to apply BACT
are most sensibly construed as going hand in hand.
Sierra Club, 615 F.3d at 1016-17.
Similarly, the Eleventh Circuit also disagreed with the Sixth Circuit’s reasoning:
To the extent the Sixth Circuit interpreted the “shall apply” language by itself as
creating an ongoing obligation to apply Best Available Control Technology every
time the source operates . . . we do not adopt that interpretation of the similarly
11

worded Alabama regulations. Without a provision imposing an ongoing duty to
meet preconstruction requirements that were not met at the time of construction,
there is no basis for us to conclude that National Parks’ and the Sierra Club’s
New Source Review allegations in this case are timely under the governing
Alabama regulations.
National Parks, 502 F.3d at 1325.
I find the analysis of the Eighth and Eleventh Circuits much more persuasive than the
Sixth Circuit approach. The “shall apply” language in Indiana’s SIP can only sensibly be read to
be tied to the construction process because it applies to “each proposed emissions unit.” 326
IAC 2-2-3. What this means is that if a company is contemplating building a new facility or
making a major modification to an existing facility – i.e. a “proposed” emission unit – it must
apply the best available control techniques. It doesn’t mean that there is an ongoing duty to apply
the BACT limits independent of the permitting process. Sierra Club, 615 F.3d at 17. Thus,
since “the duty to obtain a [NSR] permit and the duty to apply BACT are most sensibly
construed as going hand in hand,” id., the failure to apply BACT is not a new violation each day
the plant operates. Rather, it is a one-time violation, simultaneous with the failure to obtain the
proper NSR permitting.
This conclusion is also consistent with the Seventh Circuit’s general holding that NSR
permitting violations – including the failure to install BACT – all relate to the one-time
construction (in this case, the modification) and not to the continuous operation:
The violation is complete when construction commences without a permit in
hand. . . . [O]perating a new or modified plant, despite failure to obtain a
construction permit, is [not] a new violation of § 7475(a). . . . Although plaintiffs
insist that the construction permit has “an operational component,” they mean
only that under § 7475(a)(4) the operator must install the best available control
technology. Section 7475(a)(4) specifies what must be built, not how the source
operates after construction. If the owners ripped out or deactivated the best
available control technology after finishing construction that would not violate §
12

7475—though it might well violate some other statute, regulation, or
implementation plan prescribing how polluters run their facilities.
Midwest Generation, 2013 WL 3379319 at *2.
In sum, I find that the five-year statute of limitations bars any claims for damages under
either federal law or Indiana state law for U.S. Steel’s alleged failure to obtain the proper NSR
pre-construction permit and to install the best available control techniques. Consequently,
Plaintiffs’ claims for damages in its Count 1, and the paragraphs in Count 3 related to the NSR
permitting (¶¶122-127), are time-barred and will be dismissed.
II. Injunctive Relief
That leaves the issue of whether Section 2462's five-year time limit applies to the EPA’s
claim for injunctive relief. Just like the civil penalties analysis above, the analysis here would
seem to be relatively straightforward: Section 2462 only explicitly applies to fines and penalties,
so its five-year limitations period does not apply to equitable claims.
Of course, since every successful claim for injunctive relief penalizes the defendant in
some way by forcing it to do what it otherwise wouldn’t have done, it can always be argued that
an injunction functions as some sort of “penalty” and should thus be subject to Section 2462's
limitations period. Numerous courts have addressed this issue, and they have converged on a
relatively simple analysis. Section 2462's limitations period does not apply to injunctive relief if
the injunction is actually remedial – i.e., if it “seeks to undo prior damage or protect the public
from future harm.” S.E.C. v. Wyly, 2013 WL 2450545 at *6 (S.D.N.Y. June 6, 2013). On the
other hand, Section 2462 does bar injunctive relief if it is really just a facade for a penalty or a
forfeiture – i.e., “where the injunctive relief sought is punitive in nature.” Id. See also SEC v.
Bartek, 484 Fed. Appx 949, 957 (5th Cir. 2012); Coghlan v. NTSB, 470 F.3d 1300, 1304-06
13

(11th Cir.2006); United States v. Telluride, 146 F.3d 1241, 1245–46 (10th Cir.1998); Johnson v.
SEC, 87 F.3d 484, 486–92 (D.C. Cir. 1996) .
This structure of analysis makes some intuitive sense. If a defendant is simply not
complying with a law, perhaps the government should always have the right to at least bring a
suit demanding that it come into compliance. It would make little sense if a defendant could
strategically comply with some CAA regulations for five years and then immediately stop doing
so on the theory that it would be immune from suit. On the other hand, it also makes sense that
the government can’t get around the limitations period of Section 2462 just by slapping the word
“injunction” on a claim that is functionally really a penalty.
So is the injunctive relief sought by the EPA here really a remedial injunction or a
penalty? The Tenth Circuit’s analysis of the Clean Water Act in United States v. Telluride Co.,
146 F.3d 1241 (10th Cir. 1998) is helpful in analyzing this question. Telluride holds that an
injunction that “remedies damage caused by the defendant is not a penalty for purposes of §
2462 even though it is imposed by the government.” Id. at 1246. The Tenth Circuit held in that
case that the injunctive relief sought by the government – enjoining the defendant from
discharging additional pollutants and requiring it to restore damaged wetlands to their prior
condition or create new wetlands to replace those that could not be restored – was not a penalty:
“Consistent with our definition [of penalty], the restorative injunction in this case is not a penalty
because it seeks to restore only the wetlands damaged by Telco’s acts to the status quo or to
create new wetlands for those that cannot be restored. . . . [W]e do not consider the
Government’s request for injunctive relief an action for ‘civil penalties’ barred by § 2462.”
Telluride, 146 F.3d 1241 at 1246.

14

The same is true here: the government is attempting to remedy the damage that U.S.
Steel is alleged to have caused by being out of compliance with the Clean Air Act. It is thus
clear that the EPA is not barred by Section 2462 from bringing its claim for injunctive relief.
U.S. Steel has two arguments as to why the claims for injunctive relief are barred: 1) it says that
this case is different and the injunction is actually a penalty/forfeiture; and 2) the concurrentremedy doctrine applies. I will consider each of these arguments in turn.
A. Injunction vs. Penalty/Forfeiture
First, U.S. Steel argues that the injunction is, in truth, a penalty or a forfeiture and thus
subject to the five-year limitations period. If an injunction were to be entered it would force U.S.
Steel to apply for an NSR permit, install and operate BACT, and demonstrate compliance with
the CAA. U.S. Steel says this amounts to a penalty:
The injunctive relief sought by the Government is undoubtedly unjustified and
would be unreasonably severe. It asks this Court to undo, and require U.S. Steel
and IDEM to redo, an operating permit that took the parties 10 years to develop in
the first place. In addition, the Government requests an injunction requiring U. S.
Steel to undergo a preconstruction permitting review for a project that was
completed 22 years ago. Most relevant, however . . . the Complaint simply does
not seek to address the “injury caused by the defendant” and, therefore, cannot be
remedial in nature. Since there is nothing remedial in its requested relief, by
asking this Court to allow the Government to void Gary Works’ operating permit
and enjoin U.S. Steel to obtain new permits with new (and, as is undoubtedly the
Government’s intent, more onerous) limits, the Government is seeking a penalty.
[DE 14-1 at 14.]
I disagree that the injunctive relief sought here functions solely as a penalty. If a factual
determination is made in the course of this case that the 1990 reline was a major modification
such that U.S. Steel should have gone through the NSR permitting process and installed BACT,
then it may very well be that the proper remedy would be an order forcing U.S. Steel to install

15

BACT and/or to reapply for a new Title V operating permit. Any such order would be
“remedial” in that its effect would be to bring U.S. Steel into compliance with the CAA.
U.S. Steel similarly argues that the injunctive relief is actually a forfeiture because it
could involve the revocation of its Title V operating permit. U.S. Steel points to Article II Gun
Shop, Inc. v. Gonzales, 441 F.3d 492 (7th Cir. 2006) as applicable on this point. U.S. Steel
argues that the Seventh Circuit held in that case that Section 2462 bars any attempt by the
government to revoke a store owner’s gun license for violations more than five years old, and
that the Plaintiffs here are similarly barred from injunctive relief that would essentially revoke
U.S. Steel’s Title V operating permit.
I disagree that Article II Gun Shop is applicable here. First, the “holding” U.S Steel
teases out of that case is actually just very weak dicta. The true holding of that case was that the
government’s revocation of the gun shop’s license was timely because it was brought within five
years of violations that occurred in 2000. Article II Gun Shop, 441 F.3d at 496-97. It is true
that, in dicta, the court indicated that it agreed with the government that it would not have been
timely to bring a revocation action for violations that occurred in 1981, but this is far from the
holding in the case. Moreover, even if it were the holding of Article II Gun Shop, I don’t see that
it would be particularly germane to this case. That case was entirely about revocation: the
government sought to revoke the gun shop’s federal license to sell firearms. By contrast, in this
case any “revocation” of U.S. Steel’s Title V operating permit would be essentially a collateral
effect of finding that U.S. Steel had violated the CAA and now had to come into compliance
with it.

16

Thus, I find that the injunctive relief sought here is neither a penalty nor a forfeiture, and
it is therefore not barred by Section 2462's five-year limitations period.
B. Concurrent Remedy Doctrine
U.S. Steel also argues that the injunctive relief claims must be dismissed because the
“concurrent remedy doctrine” applies. The concurrent remedy rule provides that when legal and
equitable relief are available “concurrently” (i.e., when an action at law or equity could be
brought on the same facts), “equity will withhold its relief in such a case where the applicable
statute of limitations would bar the concurrent legal remedy.” Cope v. Anderson, 331 U.S. 461,
464 (1947). So, according to U.S. Steel, since the statute of limitations bars the “legal remedy”
here, I must “withhold” the equitable relief as well.
The problem with this argument, however, is that courts have uniformly held that the
“concurrent remedy rule” does not apply to “a suit by the United States in its governmental
capacity.” Telluride, 146 F.3d at 1248. See also United States v. Banks, 115 F.3d 916, 919 (11th
Cir. 1997) (citing the E.I. du Pont de Nemours & Co. v. Davis, 264 U.S. 456, 462 (1924)); Sierra
Club v. Duke Energy Indiana, Inc., 2010 WL 3667002 at *10, n.18 (S.D. Ind. 2010) (“It is
well-established that the concurrent remedy doctrine does not apply to suits brought by the
United States in its official enforcement capacity.”); New York v. Am. Elec. Power Serv. Corp.,
2006 WL 1331543 at *8-9 (S.D. Ohio Mar. 21, 2006); United States v. Cinergy Corp., 397
F.Supp. 2d 1025, 1031 (S. D. Ind. 2005); United States v. Am. Elec. Power Serv. Corp., 136
F.Supp. 2d 808, 811 (S.D. Ohio 2001); United States v. Murphy Oil U.S.A, Inc., 143 F.Supp. 2d
1054, 1086-87 (W.D. Wis. 2001); U.S. ex rel. Zissler v. Regents of the Univ. of Minnesota, 992
F.Supp. 1097, 1110 (D. Minn. 1998).

17

U.S. Steel concedes that the caselaw is uniform on this point. It argues, however, that the
governmental exception to the concurrent remedy doctrine emerges only out of an unfortunate
misreading of the Supreme Court’s 1924 decision in E.I. du Pont , and that the misreading is
now trapped in what can best be described as a doctrinal echo chamber. I actually agree with
U.S. Steel here: after going through something of a byzantine analysis retraced in the pages that
follow – I have found almost no coherent legal analysis that supports the governmental exception
to the concurrent remedy doctrine. But even after taking this journey, I still ultimately conclude
that the concurrent remedy doctrine does not bar the EPA’s claims for injunctive relief here.
Let me explain. First, the idea that the concurrent remedy rule doesn’t apply to the
government comes initially from the Eleventh Circuit’s decision in Banks and its analysis of E.I.
du Pont. The district court cases cited above generally just cite to Banks and/or E.I. du Pont, as
does the Tenth Circuit’s decision in Telluride. Most of these cases more or less just cite Banks
without actually engaging in any independent analysis of the issue, which is understandable,
given how decisive Banks is on the topic.
Nevertheless, on a close read, I think the analysis in Banks is deeply flawed. I’ll get to
those flaws in a moment, but first let’s get a better handle on what the concurrent remedy
doctrine actually is. Here’s the basic principle: “when the jurisdiction of the federal court is
concurrent with that at law, or the suit is brought in aid of a legal right, equity will withhold its
remedy if the legal right is barred by the local statute of limitations.” Russell v. Todd, 309 U.S.
280, 289 (1940). See also Cope, 331 U.S. at 464 (“And equity will withhold its relief in such a
case where the applicable statute of limitations would bar the concurrent legal remedy.”). The
rule emerged out of the old distinction between courts of law and courts of equity, and it is

18

entirely about the limited availability of equitable remedies. Equitable remedies are available
only in two instances: (1) when equity is the exclusive mode for seeking a particular remedy or
(2) when legal and equitable remedies are both available and the statute of limitations on the
legal remedy has not yet expired. When there is both a legal and equitable remedy available and
the statute of limitations on the legal remedy has expired, however, then the equitable remedy is
no longer available. That’s the concurrent remedy rule in a nutshell.
Thus, an important distinction to keep at the forefront here is that the concurrent remedy
doctrine is not a statute of limitations rule; rather, it’s a general rule about the scope of equity’s
reach. The doctrine has nothing to do with whether the applicable statute of limitations (here,
Section 2642) applies to equitable remedies, which is a question I’ve already addressed above.
Rather, the question here is whether a separate principle of equity – the concurrent remedy
doctrine – independently bars the EPA’s equitable claims for relief here.
But what does it actually mean for legal and equitable remedies to be “concurrent”? In
the context of litigation like this, where the EPA seeks to enforce alleged NSR violations under
the CAA, one might argue that the remedies here are not really “concurrent” at all: the civil
penalties sought here would likely serve a completely different purpose than the injunctive relief
being sought. Indeed, that was the conclusion of at least one district court in this circuit:
[R]egardless of a government exception to the concurrent remedy doctrine, the
civil fines and the equitable remedies the Plaintiffs seek are not concurrent,
because they have different goals and effects. For example, the citizen suit
provisions of the Act create an equitable remedy to stop threats to the
environment. Civil fines collected through a citizen suit must be paid to the
United States, not to the citizens; thus, “the primary purpose of civil penalties is
deterrence, not reward to the Plaintiffs.” As a result of this distinction, the
concurrent remedy doctrine does not operate to bar the Citizens' suit for injunctive
relief.

19

Cinergy, 397 F. Supp. 2d at 1032 (citation omitted). See also New Jersey v. Reliant Energy
Mid–Atl. Power Holdings, LLC, 2009 WL 3234438 at *12. (E.D. Pa. Sept. 30, 2009) (concluding
that legal and equitable remedies are not concurrent in the context of NSR violations).
As superficially intuitive as that reasoning might seem, however, it’s hard to square it
with the Seventh Circuit’s decision in Nemkov v. O'Hare Chicago Corp., 592 F.2d 351 (7th Cir.
1979). The court held in Nemkov that where “the sole remedy is not in equity and an action at
law can be brought on the same facts, the remedies are concurrent for purposes of the
[concurrent remedy doctrine] even though more effective relief would be available in equity.”
Id. at 355. Thus, under this precedent, when a legal remedy is based “on the same facts” as the
equitable remedy – clearly true in this case – the concurrent remedy doctrine applies.
In fact, when concerned citizen groups (rather than the government) have alleged NSR
violations just like in this case, courts have not hesitated to find that legal and equitable claims
are concurrent such that the concurrent remedy doctrine bars the plaintiffs’ claims for injunctive
relief. See Otter Tail, 615 F.3d at 1019 (“Following [the concurrent remedy] principle, we
conclude that because Sierra Club's [NSR] civil penalty claims are barred by the statute of
limitations, the equitable remedies it seeks under those causes of action are barred as well.”);
National Parks, 502 F.3d at 1327; Duke Energy, 2010 WL 3667002 at *10, n.18. In fact, each of
these decisions explicitly rejected the aforementioned conclusion from Cinergy. For example:
[Cinergy] suggested that, even if the government exception to the concurrent
remedy doctrine had not applied, the civil remedies and equitable relief at issue in
that case were nonetheless not concurrent because “they have different goals and
effects.” Id. We respectfully disagree with this part of his analysis. As discussed
above, our understanding of the doctrine is that it is not the nature of the remedy
that is determinative, but rather whether the equitable relief is available on the
same set of facts as a time barred legal remedy that determines whether the
concurrent remedy doctrine applies. See Nemkov, 592 F.2d at 354-55. Thus, we
20

find that the doctrine acts to bar equitable relief here because legal remedies albeit
time-barred were available to vindicate the same rights for which Plaintiff was
seeking relief in equity.
Id.
It’s thus clear that claims for civil penalties and injunctive relief based on NSR violations
are indeed “concurrent.” Therefore, for the concurrent remedy doctrine not to apply here, it has
to be because there is something special about the government being the plaintiff. This is where
we return to Banks, which concluded that there is in fact something special about that fact.
Banks definitively states that “the concurrent remedy rule cannot properly be invoked
against the government when it seeks equitable relief in its official enforcement capacity.”
Banks, 115 F.3d at 919. But it only reaches this conclusion via a confused and ultimately
untenable logic. Banks starts by analyzing the question of whether Section 2462 applies to
claims for equitable relief and concludes – just as I have – that “the plain language of section
2462 does not apply to equitable relief.” Id. The court then goes on to analyze the concurrent
remedy rule, but in doing so it improperly collapses two concepts: 1) core principles of statuteof-limitations law and related cannons of statutory construction that favor the government, and
2) the concurrent remedy rule. As I noted above, these are separate and distinct principles. The
Eleventh Circuit’s analysis, however, does not treat them as such. The only way to see that is to
quote most of Banks’ four-paragraph analysis here:
Banks, however, urges us to adopt the “concurrent remedy rule,” which provides
that “equity will withhold its relief ... where the applicable statute of limitations
would bar the concurrent legal remedy.” Cope v. Anderson, 331 U.S. 461, 464
(1947). Banks relies chiefly on United States v. Windward Properties, Inc., 821
F.Supp. 690 (N.D.Ga.1993) to support his position. In Windward, the government
sought equitable relief and civil penalties under section 309 of the CWA against
the defendant for unpermitted discharge of dredged or fill materials into streams
and adjacent wetlands. There, the court applied the concurrent remedy rule to bar
21

the government's claims for equitable relief under similar facts to this case. Id. at
693.
The Windward court, however, did not address the well-established rule that “an
action on behalf of the United States in its governmental capacity ... is subject to
no time limitation, in the absence of congressional enactment clearly imposing it,”
E.I. du Pont de Nemours & Co. v. Davis, 264 U.S. 456, 462 (1924); United States
v. Alvarado, 5 F.3d 1425, 1427 (11th Cir.1993), or the canon of statutory
construction that “any statute of limitations sought to be applied against the
United States ‘must receive a strict construction in favor of the Government.’ ”
Alvarado, 5 F.3d at 1428.
Incorporating these principles into the analysis, the properly constructed rule is
that-absent a clear expression of Congress to the contrary-a statute of limitation
does not apply to claims brought by the federal government in its sovereign
capacity. The statute is enforced against the government only when the
government is acting to vindicate private interests, not a sovereign or public
interest. See United States v. Beebe, 127 U.S. 338, 347 (1888).
We conclude, therefore, that the concurrent remedy rule cannot properly be
invoked against the government when it seeks equitable relief in its official
enforcement capacity. Because Congress did not expressly indicate otherwise in
the statutory language of section 2462, its provisions apply only to civil penalties;
the government's equitable claims against Banks are not barred.
Id. (footnotes omitted).
On a close read of this analysis, its flaws become manifest: what starts out as ostensibly
an analysis of the concurrent remedy doctrine quickly changes into an assessment of how the
statute of limitations applies to equitable claims – an entirely separate (and already answered)
question. Indeed, Banks’ citation to E.I. du Pont belies its flaws and underscores my point. E.I.
du Pont has nothing to do with the concurrent remedy rule. It does hold, to be sure, that an
official government action “is subject to no time limitation, in the absence of congressional
enactment clearly imposing it.” E.I. du Pont, 264 U.S. at 462. But that only expresses a general
principle of statute-of-limitations law; it has no impact on the separate equitable doctrine related
to concurrent remedies. The reasoning in Banks, thus, leaves me completely unconvinced that
22

there is something special about the way the concurrent remedy rule is applied when the
government is the plaintiff.
Nevertheless, while Banks and the other courts that have parroted its analysis leave me
wanting for logic, I am still reluctant to find that the concurrent remedy doctrine bars
governmental action. This hesitancy derives from two sources. First, I have not been able to
uncover a single instance where the concurrent remedy doctrine has been used to bar
governmental action. U.S. Steel points me to Federal Election Commission v. Williams, 104
F.3d 237 (9th Cir. 1996) and Reichelt v. United States Army Corps of Eng’rs, 969 F. Supp. 519,
521 (N.D. Ind. 1996), but neither of these cases is persuasive. The analysis of the issue in
Williams could not be more cursory (literally just three sentences), and it manages to be just as
confusing as Banks in the way it conflates the concurrent remedy doctrine with the analysis of
whether Section 2462 applies to claims for equitable relief. Reichel’s discussion of the issue is
similarly sparse and is, moreover, complete dicta.3
My second source of reluctance is this: Can some creaky, esoteric doctrine of equity
really prevent the government from enforcing basic public policy? At a minimum, shouldn’t the
government always be able to bring an equitable action to bring a defendant into compliance

3

I did turn up Bank of U.S. v. Daniel, 37 U.S. 32, 56 (1838), which might use the
concurrent remedy doctrine to bar an equitable governmental action. But drawing that
conclusion would depend, I think, on a finding that the Second Bank of the United States
(chartered from 1817 to 1836) was synonymous with the “government.” That’s a historical
judgment I feel unequipped to make, though some cursory research seems to indicate that
Second Bank of the United States probably should not be considered the “government,” at least
for our purposes here. See Lebron v. National R.R. Passenger Corp., 513 U.S. 374, 386-87
(1995) (noting that the United States subscribed only 20 percent of the Bank's capital stock and
that the President would appointed only 5 of the Bank's 25 directors, with the rest to be elected
annually by shareholders other than the United States).
23

with the law? Counseling in favor of this intuition is the fact that the most prominent equitable
doctrine related to limitations periods – laches – does not apply to government action for this
very reason. As the Supreme Court has explained:
The rule quod nullum tempus occurrit regi – that the sovereign is exempt from the
consequences of its laches, and from the operation of statutes of limitations –
appears to be a vestigial survival of the prerogative of the Crown. But whether or
not that alone accounts for its origin, the source of its continuing vitality where
the royal privilege no longer exists is to be found in the public policy now
underlying the rule even though it may in the beginning have had a different
policy basis. The true reason is to be found in the great public policy of
preserving the public rights, revenues, and property from injury and loss, by the
negligence of public officers. And though this is sometimes called a prerogative
right, it is in fact nothing more than a reservation, or exception, introduced for the
public benefit, and equally applicable to all governments. Regardless of the form
of government and independently of the royal prerogative once thought sufficient
to justify it, the rule is supportable now because its benefit and advantage extend
to every citizen, including the defendant, whose plea of laches or limitation it
precludes; and its uniform survival in the United States has been generally
accounted for and justified on grounds of policy rather than upon any inherited
notions of the personal privilege of the king. So complete has been its acceptance
that the implied immunity of the domestic ‘sovereign,’ state or national, has been
universally deemed to be an exception to local statutes of limitations where the
government, state or national, is not expressly included . . . .
Guaranty Trust Co. of N.Y. v. United States, 304 U.S. 126, 132 (1938). Other courts have also
come to the same conclusion – that laches does not apply to equitable claims brought by the
government. Bostwick Irrigation Dist. v. United States, 900 F.2d 1285, 1291 (8th Cir. 1990);
Goodman v. McDonnell Douglas, 606 F.2d 800, 804 (8th Cir. 1979); U.S. v. American Electric
Power Service Corp., 218 F.Supp.2d 931, 937 (S.D. Ohio 2002); S.E.C. v. McCaskey, 56
F.Supp.2d 323, 327 (S.D. N.Y. 1999).
What is animating all of these cases in the end is the fundamental public policy principle
expressed by the Supreme Court – that “the sovereign is exempt from the consequences of its
laches”– because the government has to have enforcement power to enjoin violations of the law.
24

Guaranty Trust, 304 U.S. at 132. And for that reason, I have arrived at the same substantive
conclusion as Banks, but I have gotten there by a slightly different path: as a matter of public
policy, the government must be allowed to bring violators into compliance with its laws, and
therefore “the concurrent remedy rule cannot properly be invoked against the government when
it seeks equitable relief in its official enforcement capacity.” Banks, 115 F.3d at 919.
CONCLUSION
For the reasons explained above, U.S. Steel’s Motion to Dismiss [DE 14] is GRANTED
in part and DENIED in part, consistent with this opinion. Furthermore, Plaintiffs’ Motion to

25

Correct and/or Clarify Transcript of Oral Argument Held on 1/24/2013 [DE 34] is unopposed
and is thus GRANTED.
SO ORDERED.
ENTERED: August 21, 2013
s/ Philip P. Simon
PHILIP P. SIMON, CHIEF JUDGE
UNITED STATES DISTRICT COURT

26