United States of America v. Jupiter Aluminum Corporation

Filing 102

OPINION AND ORDER that Jupiter's Motion to Modify the Consent Decree and to Abate Penalties DE 21 and its Motion for Declaratory Judgment and Preliminary Injunction DE 57 are DENIED. This Opinion and Order SUPERCEDES the Court's September 1, 2009 Sealed Opinion and Order DE 98 . The Court DIRECTS the Clerk to STRIKE that Opinion and Order DE 98 . Signed by Judge Philip P Simon on 9/14/09. (sda)

UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF INDIANA HAMMOND DIVISION UNITED STATES OF AMERICA, et al., Plaintiffs, v. JUPITER ALUMINUM CORPORATION, Defendant. ) ) ) ) ) ) ) ) )

CAUSE NO. 2:07-CV-262 PPS

OPINION AND ORDER On October 10, 2007, Jupiter Aluminum entered into a Consent Decree with federal, state and local environmental agencies to settle claims that its Hammond, Indiana secondary aluminum facility was in violation of federal emissions standards. Among other things, the Consent Decree requires Jupiter to make certain agency-approved modifications to its emissions control equipment at the Hammond facility. The Consent Decree also prohibits Jupiter from processing "other than clean" or "OTC" scrap until those modifications are completed. As the national economy ­ and the demand for aluminum products ­ tanked, Jupiter decided that it couldn't wait for the Agencies to approve its proposed modifications. Therefore, Jupiter made interim upgrades to its emissions control equipment without the Agencies' approval, continued to process OTC scrap, and filed a motion with the Court requesting that the Consent Decree be modified. On February 18, 2009, I entered an Order denying most of Jupiter's requests to modify the Consent Decree. But given Jupiter's ostensible financial crisis, I permitted Jupiter to: (a)

1

prove at an evidentiary hearing that its interim upgrades were compliant with the federal emissions standards; and (b) make its case that, in light of its financial condition, the company should be allowed to melt OTC scrap pending the completion of the agency-approved modifications. Shortly after the evidentiary hearing, I gave the parties an oral ruling on the pending motions; this Order reduces that ruling to writing. Although Jupiter has persuaded me that its interim upgrades exceed the federal emissions standards, I find that Jupiter has not demonstrated a sufficient change in circumstances to satisfy the extraordinarily high standard for modifying a judicially-approved consent decree. So for the reasons stated on the record and for the reasons outlined below, Jupiter's Motion to Modify the Consent Decree [DE 21] is denied. I. FACTUAL BACKGROUND Jupiter operates a secondary aluminum production facility in Hammond, Indiana, where it manufactures aluminum products using recycled aluminum. As a secondary aluminum production facility, Jupiter is subject to the U.S. Environmental Protection Agency's national emissions standards for that industry, known as Subpart RRR. See 40 C.F.R. § 1500 et seq. Subpart RRR imposes limits on emissions from furnace operations and mandates that secondary aluminum production facilities maintain capture and control equipment according to certain guidelines. See id. Jupiter is also subject to the regulatory authority of the Indiana Department of Environmental Management ("IDEM") and the Hammond Department of Environmental Management ("HDEM"). In October 2006, the EPA informed Jupiter that its Hammond facility was not in compliance with Subpart RRR. One month later, Jupiter's Hammond facility suffered a

2

catastrophic fire that forced Jupiter to shut down its operations. The timing of the fire could not have been worse for Jupiter. While already under a regulatory microscope, Jupiter sought immediate approval from the Agencies to quickly rebuild the facility. According to Jupiter, prompt action was necessary for it to remain viable. But, as Jupiter puts it, the Agencies held all the cards. And they played those cards in the form of a Hobson's choice: rebuild the facility and resolve the Subpart RRR under the Agencies' terms, or the permit request will be denied. Facing this ultimatum, Jupiter felt that it had no option but to agree to the Agencies' conditions. One of those conditions, discussed more thoroughly below, involved Jupiter possibly having to implement emissions control systems that substantially exceeded the minimum Subpart RRR standards and that were far more cumbersome than what would have ordinarily been required by the regulations. Acting on the advice of experienced environmental lawyers, Jupiter reluctantly agreed to the terms. As a result, the Agencies filed the instant case against Jupiter in August 2007 and immediately filed a proposed consent decree to settle the matter. On October 10, 2007, I approved the Consent Decree with all of the parties' consent. (DE 20.) The following provisions of the Consent Decree are presently at issue: Jupiter was required to conduct initial stack tests for its two melting furnaces and again if visible emissions were observed. (DE 20 at 9-12.) (Stack tests are procedures used to determine an equipment's pollution emissions rate.) If Jupiter failed the tests, it would not be allowed to melt OTC scrap until it first modified its capture and control system in accordance with negotiated specifications set forth in Appendix A to the Consent Decree. (Id. at 17-18.) Appendix A prescribes certain

3

dimensions, air flow rates and compliance standards for these modifications. (Id. at 79-80.) The Agencies must approve any proposals for Appendix A modifications before they are implemented. (Id. at 17-18.) Within months of signing the Consent Decree, both of Jupiter's melting furnaces failed stack tests, thus triggering the Appendix A requirements. Jupiter then submitted several proposals for Appendix A modifications. All were rejected by the Agencies. Jupiter says the Agencies were playing hardball; the Agencies counter that the proposals were inadequate. In April 2008, in the midst of this standstill, Jupiter took matters into its own hands and implemented interim short-term upgrades to its equipment. Jupiter insists that the current configurations now meet the Subpart RRR standards. In May 2008, Jupiter filed a Motion to Modify the Consent Decree requesting that several key provisions of the Consent Decree be discarded. (DE 21.) Notably, Jupiter asked to be relieved from the ban on OTC scrap. Jupiter argued that, because OTC scrap is much cheaper and more profitable to recycle than clean scrap, it could not compete in the industry if it had to make its products using only clean scrap. Jupiter further contended that the OTC scrap ban no longer served its purpose of preventing excessive emissions because the company's interim upgrades met the Subpart RRR standards. At the parties' request, I withheld my rulings on the motion for several months while the parties tried to resolve their differences through mediation. The mediation attempt was unsuccessful, but the Agencies did approve Jupiter's most recent Appendix A proposal on December 28, 2008. The approved Appendix A modifications are scheduled to be completed sometime in February or March of 2010. (DE 66, Hrg. Tr. at 9.)

4

On February 18, 2009, I entered an Order denying most of Jupiter's modification requests. (DE 59.) However, in light of the recent economic calamity ­ which particularly affected the demand for aluminum products in the auto and housing industries ­ I reserved judgment on the question of whether the Consent Decree should be modified to permit Jupiter to burn OTC scrap while the Appendix A modifications were being implemented. If Jupiter was correct that its interim upgrades met the Subpart RRR standards, and that the company was facing a severe financial crisis, then it made little sense to me why Jupiter should be prohibited from burning OTC scrap. At a March 24, 2009 status hearing, the parties informed me that the most effective way to determine whether Jupiter was in compliance with Subpart RRR was to conduct stack tests for both melting furnaces. Jupiter also requested that the stack tests and evidentiary hearing occur as quickly as possible because the uncertainty of this litigation was affecting Jupiter's banking relationships, which the company relied on for capital and operating loans. This presented a logistical problem because stack tests typically require numerous formalities and months of preparation and negotiation to establish test protocols. However, under these unique circumstances, the parties agreed to cooperate on protocols and waive certain notice periods and formalities in order to get the stack tests done. To the parties' credit, the stack tests were conducted on April 14-15, 2009. The test was administered by Air Compliance Testing, Inc. ("ACT"), a third party contractor hired by the Agencies. Without getting into all the technical details, here's what took place: Jupiter has two melting furnaces that lead to three baghouse exhaust smokestacks. The test protocols provided

5

that all three baghouses would be tested simultaneously during three, three-hour runs. (The second run was aborted because of production issues, so a fourth run was added.) In each threehour run, Jupiter was required to perform several charging cycles at stressed conditions. A complete charging cycle generally consists of two processes: charging and fluxing/drossing. First, the furnaces are charged with aluminum scrap. Jupiter's typical practice is to melt a combination of 40% clean scrap and 60% percent OTC scrap. (DE 91, Hrg. Tr. at 41.) The testing protocols however required that Jupiter charge 100% OTC scrap to simulate worst-case operating conditions. (Def. Ex. N.) (Id.) From Jupiter's point of view this requirement was already a bit onerous because they virtually never burn 100% OTC. Similarly, Jupiter was required to charge a minimum of 18 tons of scrap per hour, which represents Jupiter's maximum operational charging rate. (Def. Ex. N.) Second, flux ­ a salt-based substance ­ is added to the furnace to aid in the formation of dross, which are the impurities in the scrap. (DE 91, Hrg. Tr. at 35-36.) While the flux is added, the dross is simultaneously removed. (Id.) It took Jupiter between 40 to 60 minutes to charge the furnaces with 18 tons of scrap and an additional 15 to 20 minutes to flux and dross. (Def. Ex. S.) In between cycles, the Agencies asked Jupiter to pause for at least another 15 to 20 minutes so that the testing equipment could be repositioned. (DE 91 Hrg. Tr. at 53-60.) During these pauses, Jupiter continued to charge scrap because the caster was running and needed metal. (Id.) However, the Agencies didn't include the scrap added during these pauses in their calculation of the charge or emissions rates. (Id.; see also Def. Ex. S-1.) On April 27, 2009, ACT issued a Compliance Stack Emission Test Report. (Plf. Ex. 38.)

6

The report contained Jupiter's emissions rates for three categories of pollutants: First, for particulate matter emissions, the furnaces averaged less than 25% of the applicable limits. In other words, the furnaces were producing only one-fourth of the maximum allowable emissions for this type of pollutant. Second, for the pollutant dioxin/furan, the test results showed emissions that were minuscule ­ less than 5% of the applicable limits. Finally, for the pollutant hydrogen chloride, the test results showed emissions of less than 25% of the applicable limits. (Id.) So from Jupiter's point of view, since the results were well under the Subpart RRR standards, they pass the test with flying colors. The agencies believe otherwise. Between April 30, 2009 and May 4, 2009, I held an evidentiary hearing to determine whether or not Jupiter's furnaces are currently in compliance with Subpart RRR and whether the consent decree should be modified. Jupiter presented two witnesses on this issue: Mark Volkmann, Jupiter's Health and Safety Director; and Marcus Cooke, an expert in environmental compliance. The Agencies presented several witnesses of their own including: Sara Breneman, an EPA enforcement officer; Charles Simon, a Senior Scientist at ACT; environmental engineering expert Patrick Atkins; and Steven Friend, an IDEM compliance officer. In addition, Paul Chevalier, Jupiter's Executive Vice President, testified about the company's financial condition. II. DISCUSSION To put it bluntly, Jupiter signed a bad deal. But Jupiter knew what it was getting into because during the public comment process, the parties received comments from a trade association and one of Jupiter's competitors complaining that the proposed consent decree was

7

too inflexible and imposed obligations that exceeded regulatory standards. Jupiter ignored the warning signs and signed the decree nonetheless and I approved it in October 2007. It is stating the obvious that much has changed since October 2007. The national and world economies have suffered an unexpected and dramatic recession ­ far worse than any recession since the Great Depression. The bottom line for the purposes of this case is that the two primary consumers of Jupiter's products ­ the auto and housing industries ­ have been hit the hardest. According to Jupiter, bids for supplier contracts in the secondary aluminum industry are won and lost based on a tenth of a cent per pound. If one believes Jupiter, it is impossible in the secondary aluminum industry for a company to stay financially viable if it's forced to use only clean scrap, which is more costly. Jupiter claims it has retrofitted its plant to meet the Subpart RRR. If that's true, it seems to me that the Agencies are overreaching by insisting on the temporary ban on OTC scrap. What legitimate purpose is served by prohibiting Jupiter from running its business ­ within the confines of the federal emissions standards ­ until the Appendix A modifications are completed? These concerns prompted me to permit Jupiter to make its case that this provision of the Consent Decree should be modified. But the bar that Jupiter must clear is very high. It is extremely difficult to get out of a consent decree and for good reason. This is because a consent decree serves both as a judgment of the court and as the parties' negotiated settlement of litigation. See United States v. Alshabkhoun, 277 F.3d 930, 934 (7th Cir. 2002). Federal Rule of Civil Procedure 60(b)(5) gives me some discretion to grant a party relief from a judgment "if applying it prospectively is no longer equitable." FED R. CIV. P. 60(b)(5). But because of the importance of finality that

8

judgments bring, courts may exercise that discretion only in exceptional situations. See Harrington v. City of Chicago, 433 F.3d 542, 546 (7th Cir. 2006). This requires Jupiter to demonstrate a significant change in circumstances. Rufo v. Inmates of Suffolk County Jail, 502 U.S. 367, 383-84 (1992). Boiled to its essence, Jupiter has attempted to meet this burden by demonstrating two changed circumstances: First, that Jupiter's furnaces, with the interim upgrades, are currently in compliance with Subpart RRR. Second, that Jupiter's financial condition, along with the unanticipated national economic crisis, render it impossible for Jupiter to survive until next spring by burning only clean scrap. A. Compliance with Subpart RRR Depending on who you ask, the results of the stack tests were either a complete success for Jupiter, or tell us nothing at all. Jupiter insists that the number's don't lie: ACT's Test Report unequivocally demonstrates that Jupiter is emitting less than 25% ­ and for some substances, less than 5% ­ of the permissible limits. At the outset, it's worth pointing out that these results were achieved under extreme conditions which required 100% OTC charge . The more OTC that is used, the more pollutants that are produced. But this 100% OTC requirement was a bit unrealistic because, as mentioned above, Jupiter almost never charges 100% OTC; they usually charge at a 60/40 ratio of OTC to clean scrap. And the regulations that govern testing protocols only require testing at the "highest production level with charge materials representative of the range of materials processed by the unit." 40 C.F.R. § 63.1511(b)(1) (emphasis added). So to be clear, the test was conducted under extreme conditions ­ indeed,

9

conditions that rarely exist. What's more, Jupiter's emissions compliance expert, Dr. William Cooke, testified that the reported test results are artificially higher than Jupiter's actual emissions rates for other reasons. Dr. Cooke is extremely well qualified to testify on Subpart RRR compliance and testing standards. He holds a doctorate in chemistry from Virginia Tech, serves on the board at Cal Tech and has taught chemistry at universities around the world. He has provided consulting services to numerous federal and state agencies, including the EPA, Department of Energy and NASA, relating to environmental pollution. He also presently serves an EPA ombudsman for EPA Region Four in Atlanta, Georgia, where he is responsible for investigating non-compliance with consent agreements. Dr. Cooke also knows a thing or two about stack tests ­ he has been involved in hundreds of them. Dr. Cooke explained that Subpart RRR does not place limits on the emissions from dross pans (protective pans holding the skimmed-off dross). (DE 91, Tr. at 142.) However, the test protocols required that the Jupiter's dross pans be left at the hood. (Id. at 143.) This means that the dross pan emissions were captured in the flue and control system and showed up in the reported emissions levels. (Id.) Dr. Cooke also credibly testified that the test results were distorted when the Agencies paused the test to reposition their testing equipment. These pauses were neither part of the test protocols nor are they consistent with standard testing procedures. (Id. at 140-42.) Jupiter loaded significant amounts of scrap during these pauses to keep the caster running, yet none of these loads were included in the emissions calculations. (Id. at 60.) But because OTC scrap emissions are released over time, much of the emissions from these

10

scrap amounts were counted later on when the testing started up again. (Id. at 145.) The Agencies have a much different take on the stack test results. Their main beef is that Jupiter never charged scrap at a rate of 18 tons per hour as they required to do by the test protocols. There are two ways to measure charge rates under Subpart RRR: by batch process or continuous process. (DE 92, Hrg. Tr. at 21.) The former measures charge loaded per full cycle. (Id.) (Recall that a full cycle includes charging plus fluxing and drossing.) The latter measures charge loaded per hour. (Id.) The Agencies say that Jupiter has always represented to them that it operates under a continuous process. (Id.) According to the Agencies, this means that Jupiter was required to sustain a continuous charge rate of 18 tons per hour for the entire three-hour test. During the test, it took Jupiter only 40-60 minutes to charge 18 tons of scrap. (Def. Ex. S.) However, Jupiter spent another 15-20 minutes fluxing and drossing, during which time the clock was running. (Id.) Each full cycle took more than an hour to complete, which means Jupiter was never able to sustain a continuous charge rate of 18 tons per hour. (Id.) The Agencies therefore contend that Jupiter failed to meet test protocols and that the test is wholly unreliable. Jupiter counters that the Agencies are holding Jupiter to an unrealistic testing protocol. From Jupiter's perspective, the 18 tons per hour requirement simply means that Jupiter was required in each cycle to charge 18 tons of scrap in less than an hour. (DE 91, Hrg. Tr. at 32-35.) Because Jupiter needed time to flux and dross, Jupiter says it would've been impossible to meet a continuous charge rate of 18 tons per hour unless Jupiter charged an additional 5 to 6 tons per cycle (or 23 to 24 tons in an hour), which it contends it has never done. It appears there was a fundamental misunderstanding as to how much scrap Jupiter was

11

required to charge per cycle or per hour. Unfortunately, the written test protocols themselves do not clarify the matter. (Def. Ex. N.) I am mindful of the fact that the parties did not conduct the stack tests under ideal circumstances given the time constraints. If not for the urgency to get the tests done as soon as possible, the parties would likely have had sorted out these issues well beforehand. Under these circumstances, I can't really blame either side for the confusion. However, I do find Jupiter's position to be much more sensible. The whole purpose of the 18 tons per hour requirement was to replicate worst-case conditions. Charging records submitted to the Court show that Jupiter's maximum charge rate under normal operating conditions was just over 18 tons in an hour. (Plfs. Exs. 20-25.) But there is no evidence in the record that Jupiter ever charged scrap at a continuous rate of 18 tons per hour over a period of several consecutive hours. Therefore, I am convinced that the tests were conducted while the furnaces were operating at maximum, or close to maximum, stressed conditions. The Agencies also contend that the test results shouldn't be trusted because Jupiter failed to effectively capture all the emissions it was producing. To make their case, the Agencies played a video recording of the stack tests showing visible smoke ­ referred to as fugitive emissions ­ escaping from the hearth doors. To be sure, the EPA recognizes that there will always be a certain level of fugitive emission. The goal is to minimize fugitive emissions, not eliminate them. See 65 Fed. Reg. Vol. 15,698 (Mar. 23, 2000). The EPA's enforcement officer, Sara Breneman, concluded that the fugitive emissions seen in this case rendered the test results invalid. I found Ms. Breneman's testimony to be completely unpersuasive. Breneman admitted

12

that Subpart RRR doesn't specify how much fugitive emissions is too much. (DE 92, Hrg. Tr. at 63.) Rather, she said that she is the one who makes the decision based on "enforcement discretion." (Id.) Yet she could not provide any objective criteria for determining how much was too much. (Id.) She also confessed on cross examination that she had never actually seen a stack test before this one. (Id. at 67.) Yet she insisted that the fugitive emissions in these stack tests were excessive. But how would she know if she had no objective standards to rely on or any basis for comparison? The Agencies stack test compliance expert, Dr. Charles Simon, testified that the smoke produced during the first stack test run was so dark and thick that he had to leave the building. (DE 92, Hrg. Tr. at 127-28.) While Dr. Simon went into great technical detail about what caused the smoke and what the smoke consisted of, he gave little insight into whether the smoke violated Subpart RRR, or whether the fugitive emissions ­ if captured ­ would have put Jupiter over the emissions limits. Steven Friend of IDEM also testified about the visible emissions. Mr. Friend only observed the first stack test run and testified, based on his experience observing other stack tests, that the fugitive emissions in that test were more than what he has observed in the past. (DE 92, Hrg. Tr. at 137.) But similar to Ms. Breneman, Friend didn't account for variable factors, such as test protocols, nor did he provide any objective basis for his conclusions other than his own subjective impressions. Dr. Cooke testified on rebuttal that none of the testimony offered by the Agencies indicated that the fugitive emissions produced during the stack tests put Jupiter above the regulatory limits. Dr. Cooke reported that he is certified by the EPA as an opacity reader, which

13

means he is qualified to make conclusions about visible emissions by following certain observational criteria. (DE 92, Hrg. Tr. at 195.) He explained that the stack test protocols didn't call for conditions that would have made it possible to accurately judge the level of fugitive emissions just by looking at them. (Id.) According to Dr. Cooke, the visible emissions shown in the video recordings are deceptive. (Id. at 192-200.) In his opinion, the amount of fugitive emissions during the stack test would have an insignificant or de minimis effect on the overall emissions rate. (Id. at 203-07.) This makes sense since the fugitive emissions were coming out of the hearth door, which in comparison to the cubic feet of the three baghouses is minuscule. So it's difficult to see how the fugitive emissions would have significantly altered the test results. But even if they would have, to say that they would have led to four times more emissions at the stack ­ or twenty times for the one pollutant! ­ seems preposterous. It's clear that some emissions were sneaking out of the furnaces that shouldn't have. But the regulations allow for this. So I am not persuaded that the presence of visible fugitive emissions invalidates the test results. Jupiter's reported emissions levels were so low, that it is incredibly unlikely that the emissions that escaped through the hearth door were high enough to account for the additional 75% to 95% of emissions that would have put Jupiter over the Subpart RRR limits. Ultimately, I find Dr. Cooke's conclusion that Jupiter is operating well below the federal emissions standards to be persuasive. Both Dr. Cooke and Dr. Simon have impressive credentials, but on balance I found Dr. Cooke's testimony and experience to be more relevant to the issues at hand.

14

For their part, the Agencies spent most of their time at the hearing trying to convince me that the tests should be invalidated for failure to follow protocols. Just prior to the hearing, the Agencies represented that the stack tests were the best way to determine compliance with Subpart RRR. (DE 66, Hrg. Tr. at 16-17.) They hedged that statement by saying: (1) if Jupiter fails, then we know for sure that it is not in compliance; but (2) if Jupiter passes, there are a litany of reasons why the test results would be inconclusive. (Id.) That's a "heads I win, tails you lose" proposition if I've ever seen one. How in the world would a company ever know if its in compliance? Now that the stack tests show that Jupiter is operating substantially below the federal emissions limits, the Agencies want me to throw out the test results. This I will not do. I appreciate the Agencies' point that the test results shouldn't be oversimplified. For sure, Subpart RRR is a complex regulatory scheme and that there are many rules and protocols that companies must follow to ensure that the relevant agencies have the most reliable data to effectuate compliance. But if one adopts the Agencies' inflexible position in this case, there is no way a company can ever prove that it is in compliance ­ they'd be trying to hit a moving target. One of the purposes of the evidentiary hearing was to do determine whether Jupiter was operating below the federal emissions limits. Despite the imperfections in the stack test, Jupiter has demonstrated to the Court's satisfaction ­ through the stack test results and Dr. Cooke's testimony ­ that its furnaces are currently operating well below these limits. While Jupiter must do other things to ensure that its operational practices also satisfy the regulatory mandates, those issues don't change the fact that the interim upgrades made by Jupiter are effectively controlling emissions within the federal guidelines.

15

B.

Economic Hardship As noted earlier, the threshold for modifying a Consent Decree based on changed

circumstances is extremely high, almost impossibly so. The fact that Jupiter is currently operating within the federal emissions limits is not enough to meet this standard. For one thing, this change wasn't caused by external circumstances, such as a change in the law or regulations. Jupiter unilaterally caused the change without the Agencies' approval. Moreover, the standard under Rule 60(b)(5) requires a showing that unforeseen or unanticipated circumstances have made the prospective enforcement of the consent decree unworkable or substantially more onerous. Rufo, 502 U.S. at 383-84. Jupiter's upgrades alone have no bearing on the feasibility of the Consent Decree. The current economic meltdown, along with Jupiter's particular financial circumstances, prompted the post-judgment motions; so the question before me is whether such economic hardship is sufficient to merit a modification of the Consent Decree. First, it must be noted that the Consent Decree has a force majeure provision which explicitly exempts Jupiter's financial hardship as a legitimate force majeure event. (DE 20 at 62.) The Seventh Circuit has routinely enforced these force majeure provisions. See, e.g., United States v. Rueth Dev. Co., 335 F.3d 598, 606 (7th Cir. 2008); United States v. Ashabkhoun, 277 F.3d 930, 936 (7th Cir. 2002). So from the outset, by relying on its financial hardship, Jupiter is asking me to brush aside what it specifically agreed to. That is not to say that a company can never rely on economic hardship as a basis for relief from a consent decree. Jupiter relies heavily on a D.C. Circuit case, Evans v. Williams, 206 F.3d 1292 (D.C. Cir. 2000), to make its case. In Evans, the District of Columbia's public

16

mental health institution entered into a consent decree to resolve a class action lawsuit challenging poor conditions at the facility. Id. at 1293. The decree required the District to place the residents in community institutions and insure that all vendors were paid within thirty days of the submission of proper vouchers. Id. Several years later, the District suffered "financial problems of horrendous proportions and faced its worst crisis in over a century." Id. Since the District didn't have enough cash on hand to pay its bills, it failed to pay some of its vendors according to the thirty-day timetable established in the consent decree. Id. The Court held that the District was entitled to relief from this timetable in light of its unanticipated financial crisis, which the parties had not contemplated in their negotiations. Id. at 1297-99. In doing so, the Court noted that "flexibility is especially important in institutional reform litigation[.] Because consent decrees often remain in place for extended periods of time, the likelihood of significant changes occurring during the life of the decree is increased." Id. at 1298 (citing Rufo, 502 U.S. at 380). The Supreme Court recently reiterated the need for flexibility in institutional reform litigation. See Horne v. Flores, 129 S.Ct. 2579 (2009). In Horne, a class of English LanguageLearner students and their parents sued the State of Arizona under the Equal Education Opportunities Act for failing to appropriately fund programs to overcome the students' language barriers. Id. at 2589-90. The district court concluded that the defendants violated the EEOA and entered a declaratory judgment along with some injunctions to ensure that funding was allocated appropriately. Id. Several years later, the state legislature passed a bill designed to implement a permanent funding solution to the problems originally identified by the district court, but which

17

differed from the obligations imposed by the court. Id. at 2590-91. The state then moved for relief under Rule 60(b)(5), arguing that the new legislation amounted to changed circumstances. Id. The district court denied the motion and the Ninth Circuit affirmed the denial. Id. at 2591. The Supreme Court reversed and held that the lower courts misapplied the Rule 60(b)(5) standard. Id. at 2598. Specifically, the Court reasoned that injunctions issued in institutional reform litigation typically last for several years, and over time many things are likely to change including governing law, policy insights, and even the nature of the underlying problem itself. Id. at 2593. Given the nature of these types of cases, they often raise sensitive federalism concerns involving areas of core state responsibility and impositions on a state or local government's budget priorities. Id. The Court held that Rule 60(b)(5) serves an important function in these types of cases of giving the Court flexibility to reexamine the appropriateness of the judgment over time. Id. Contrary to Jupiter's argument, I don't find Evans to be particularly relevant to the facts of this case. First, there is no indication that the Evans consent decree had a force majeure clause, like this one does. And although Rule 60(b)(5) relief isn't by any means limited to institutional reform cases, it can't be said that the concerns raised in Evans and Horne apply to this case. For one thing, Jupiter's Consent Decree doesn't impose highly supervised injunctive relief over the course of several years. Instead, Jupiter will be relieved from the ban on OTC scrap when it completes the Appendix A modifications in February or March of 2010, about six months from now. Moreover, there are less public policy concerns than there were in Evans, Rufo, and Horne, where state and local budgets were impacted. Therefore, the same need for

18

flexibility isn't present in a case like this. More fundamentally, Jupiter has not demonstrated to me that it is in fact experiencing a severe financial crisis brought on by changed circumstances. Jupiter's refrain throughout the proceedings has been that a secondary aluminum facility cannot compete in the marketplace if it is forced to make its products using clean scrap alone. The basis for this argument is that OTC scrap costs more per pound than clean scrap, and in an industry where supplier bids are won and lost based on a tenth of a cent per pound, such a differential in raw materials can literally shut a company out of the competitive market. I have no doubt that the temporary OTC scrap ban will impair Jupiter's profitability in the short term. However, several facts suggest to me that the restriction isn't quite the death knell that Jupiter makes it out to be. First, Jupiter's Executive Vice President Paul Chevalier testified that, in December 2007, clean scrap cost seventeen cents more per pound than OTC scrap. (DE 80, Hrg. Tr. at 50.) Therefore, Jupiter was aware of the financial impact of the OTC ban provision when it signed the Consent Decree. So it can't be said that the financial burden of this provision was unanticipated or a changed circumstance. Second, by March 2009, the price differential between clean and OTC scrap had actually narrowed to two cents. (DE 80, Hrg. Tr. at 50.) Chevalier testified that this was due to the impact of China entering the aluminum market on a massive scale, and he predicted that the price differential would likely increase if and when China pulled out. (Id. at 19.) Though I was impressed by Chevalier's knowledge of the macroeconomics of the secondary aluminum industry, trying to make predictions in the commodities market with accuracy is a speculative

19

endeavor. Even assuming Chevalier is correct that the price of OTC scrap is likely to go up, the fact that the price differential has narrowed so dramatically waters down the argument that it is prohibitively costly to only use clean scrap. If the price differential had gone up from two cents to seventeen cents, Jupiter would have had a better argument. Third, Dr. Patrick Atkins, the Agencies' environmental engineering expert, explained that a ton of clean scrap, though more expensive, yields a higher volume of usable material than a ton of OTC scrap because the latter contains impurities which must be discarded in the melting process. (DE 92, Hrg. Tr. at 149-53.) Dr. Atkins reviewed Jupiter's charge records for the stack tests and calculated that Jupiter had yielded a melt loss of 6%. (Id. at 153.) That melt loss may compensate for some, if not all, of the price differential. He also explained that clean charge melts at a faster, more efficient rate, which further improves profitability. (Id. at 152.) Jupiter also contends that the cloud of uncertainty hanging over this litigation is threatening its ability to obtain capital and operating loans. However, Jupiter did not provide the Court with any evidence that its ability to maintain its current banking relationships or pursue new banking relationships will be affected by the outcome of this case. Without more evidence, I cannot conclude with any degree of confidence that Jupiter will lose the ability to obtain credit unless the Consent Decree is modified. At its core, Jupiter's plea of impending financial doom is long on generalizations and short on specifics. Chevalier gave very interesting testimony about the complexities of the global aluminum commodities market, but he offered no details about Jupiter's current financial viability. Jupiter didn't provide the Court with historical financial records, projections, or any

20

other data to demonstrate that its fear of insolvency was justified. I appreciate the fact that Jupiter is a private company and keeps its financial information close to its vest. However, Jupiter is asking the Court to grant an extraordinary remedy. Without more, I am reluctant to modify a consent decree based on such thin evidence. Jupiter is not the first company to cite the current global financial crisis as a basis for modifying a consent decree. Consider the recent case of United States v. Signature Flight Support Corp., 607 F. Supp.2d 56 (D.D.C. 2009). The Signature case was an antitrust matter, where the defendant Signature was required to divest one of its assets (a fixed base operation at Indianapolis Airport) within a month and a half of the entry of the consent decree. Id. at 58. Signature petitioned the court for relief from this deadline, pointing to the difficulty of selling these assets in light of the current global credit crisis. Id. Specifically, Signature pleaded that it would be inequitable to force it to sell its assets at 25% of the original purchase price when neither party anticipated the extent of the global economic meltdown. Id. at 59. The court denied Signature's request because the consent decree was negotiated in the midst of troubling economic news and the parties had specifically countenanced the potential difficulty in selling the assets. Id. at 59-60. Moreover, Signature could not show that it couldn't sell the assets, only that it would be less profitable to do so. Id. This case is very similar. Nobody anticipated the extent of the current global financial crisis and the devastating impact it has had on American businesses like Jupiter. But like Signature, Jupiter knew at the time it signed the Consent Decree that there would be significant financial consequences if things didn't go its way. Moreover, Jupiter was aware that, in October

21

2007, the housing market had already been in decline for eight months and that the company would be in trouble if the OTC scrap ban was triggered. (DE 80, Hrg. Tr. at 18.) Yet Jupiter still agreed to those terms. Perhaps Jupiter was overconfident that it would pass the initial stack tests and that the provision wouldn't be triggered. If that's the case, Jupiter is not entitled to Rule 60(b)(5) just because it is disappointed with the consequences of the settlement. See McCormick v. City of Chicago, 230 F.3d 319, 327 (7th Cir. 2000). C. Motion for Declaratory Judgment and Preliminary Injunction This Order will likely have little effect on Jupiter's current operations, since Jupiter has never stopped burning OTC scrap after it signed the Consent Decree. In February 2008, the Agencies demanded over $4 million in stipulated penalties for these infractions. (DE 21-3 at 3840.) It appears that no further penalty demands have been made since. Jupiter is nonetheless concerned by the big hammer that the Agencies are wielding over its head. It has therefore filed a motion requesting a declaratory judgment and/or preliminary injunction prohibiting further accrual of penalties for charging OTC scrap while the Appendix A modifications are being completed. (DE 57.) This request puts the cart before the horse. The Consent Decree prescribes specific informal and formal dispute resolution procedures that must be followed before such an issue can be presented to the Court. (DE 20 at 59-61.) An order from this Court deeming any future penalties invalid would undermine these negotiated procedures. Until a penalty demand has been made, and Jupiter has completed the dispute resolution process, Jupiter's request for judicial intervention is premature.

22

III. CONCLUSION For the foregoing reasons, Jupiter's Motion to Modify the Consent Decree and to Abate Penalties [DE 21] and its Motion for Declaratory Judgment and Preliminary Injunction [DE 57] are DENIED. This Opinion and Order SUPERCEDES the Court's September 1, 2009 Sealed Opinion and Order [DE 98]. The Court DIRECTS the Clerk to STRIKE that Opinion and Order [DE 98]. SO ORDERED. ENTERED: September 14, 2009 s/ Philip P. Simon PHILIP P. SIMON, JUDGE UNITED STATES DISTRICT COURT