Duke-ing It Out at the High Court: The End of New Source Review?


To what extent can the EPA force utilities to update aging fleets with expensive pollution-control technology?

Fortnightly Magazine - February 2007

Utility executives take notice: The U.S. Supreme Court will issue a contentious and potentially far-reaching Clean Air Act (CAA) decision soon in Environmental Defense v. Duke Energy Corp., a case involving the nation’s third-largest energy company. The Supreme Court heard oral arguments in early November and likely will issue an opinion before June 2007. Although the press has been focused on the other environmental case pending before the court, which deals with ability of the Environmental Protection Agency (EPA) to regulate greenhouse gases, the upside of the Duke Energy case for the electric industry could be much more significant—particularly for utilities with aging coal-fired fleets.

The Duke Energy case marks a significant opportunity for the court to clarify one of the most litigious issues in the electric-power industry today: How and to what extent can the EPA force utilities to update their aging fleets with expensive pollution-control technology? Under the EPA rules at issue in the case, utilities that modify their plants often are forced to retrofit them with expensive pollution controls like selective catalytic reduction and flue-gas desulfurization. These controls typically cost hundreds of millions of dollars and require long installation outages. According to Scott Segal, the Electric Reliability Coordinating Council’s director, “Clarity regarding these concepts is essential to improved efficiency, reduced emissions, enhanced workplace safety, and electric reliability.”1

The EPA rules are so divisive because, as plants grow older, they become less reliable and emit air contaminants at a higher rate than newer vintage plants. Yet, the rules at issue in the Duke case would inhibit utilities from upgrading or modifying their older facilities. Indeed, because these older plants are not fitted with many pollution controls, they operate at much lower costs than new or controlled plants, thereby keeping the cost of baseload power low. In other words, the subject provisions have put environmentalists and electric industry officials at loggerheads. To keep the cost of power low, utilities and merchant generators cannot significantly modify their plants. To keep our air clean, however, environmentalists and the EPA must hover over the electric industry, looking for a triggering modification and suing if they spot one.

The pending case at the Supreme Court is not the only noteworthy New Source Review news from the last few months. Two recent economic studies out of Yale and the Center for Economic Development and Research have shed some light on EPA’s New Source Review enforcement tendencies, how utilities respond to increased regulatory scrutiny, and what effect the Supreme Court’s decision might have on rural economies.2 Regardless of whether these studies affect the Supreme Court’s decision-making process, utility managers and decision makers should be keenly aware of some of their findings.

The Modification Rule

According to a 2002 congressional report, power-plant pollution causes approximately 30,000 deaths annually in this nation—more than drunk driving and homicides combined.3 Yet, because of New Source Review and various other congressionally created grandfathering provisions in the CAA and its amendments, today only about one-quarter of the nation’s large-scale generators have updated pollution-control equipment.4 The sole issue in Duke Energy stems from a grandfathering provision, commonly known as the New Source Review modification rule.5 Stated simply, this rule provides that power plants built before 1977 are not required to upgrade their pollution-control equipment as long as the plant is not significantly modified. The idea behind the significant modification standard was that as plants grew older, they either would have to be modified or retired, causing newer, cleaner plants to come online over time.

For more than 20 years, EPA rarely enforced the New Source Review modification rule. Just before leaving office, however, the Clinton administration began a large-scale enforcement initiative—initially suing 36 power plants for billions of dollars—in an attempt to force plant upgrades. While most of these cases have long since been settled (see figure, "Sample New Source Review Settlements"), a few are still being litigated, including the Duke Energy case.

What’s At Issue?

After almost six years of litigation, the Duke Energy case has dwindled down to one central issue: How should the EPA calculate an emissions increase under the New Source Review modification rule? When determining whether a New Source Review triggering modification has occurred at a plant, the EPA uses two criteria: The modification must increase emissions from the plant, and the modification cannot constitute routine maintenance. A recent appellate court decision severely limited the breadth of the routine maintenance exemption; thus, the importance of the emissions increase test is now vital.6

The EPA historically has used two tests for calculating whether a modification will cause emissions to increase significantly—an hourly test and an annual test. The hourly test is used to force small and large companies to meet baseline emissions standards, and the annual test is used to force major sources to install the best currently available technology. With an hourly test, a modification occurs only if the plant’s hourly emissions would increase as a result of the change, but with an annual test, a modification occurs only if the plant’s total annual emissions would increase. Stated more simply, an hourly increase generally will occur only if the modification will increase the plant’s total capacity, while an annual increase will occur for much smaller modifications that might only increase the plant’s capacity factor.

In its decision, the 4th Circuit invalidated EPA’s use of the annual test. It did so by relying on an old Supreme Court tax determination, which says that an agency cannot interpret the same statutory term differently. In other words, EPA could not use an annual test for large sources and an hourly test for small sources, when both tests stemmed from EPA’s interpretation of the same term, “modification.” Although none of the initial briefs cited this argument or this tax case, the 4th Circuit relied only on this case in invalidating EPA’s use of the annual test.

The Hourly/Annual Distinction Is Significant

Why is this apparently small, technical interpretation so significant? Is there really a considerable distinction between the annual and hourly tests? Though it is easy to get caught up in the case’s technical jargon, the distinction between the two tests really is rather simple. The hourly test allows a utility to almost entirely refurbish its generation units without triggering environmental control requirements, whereas the annual test does not. Under the hourly test, EPA’s only consideration is whether plant-wide emissions per hour increase, even if the new, more efficient plant can operate for more hours per year. Conversely, the annual test accounts for all possible capacity factor increases and, as a result, virtually any modification allowing a plant to operate more will require upgraded pollution controls.

Does the Modification Rule Work?

Economists long have argued that the modification rule retards capital investment, thereby inhibiting the turnover of the electric industry’s generation stocks.7 They argue that instead of ensuring that these older plants modernize, the modification rule actually creates the opposite incentive because un-modified plants have lower fixed and variable costs. Of course, a utility’s incentives differ depending on the state regulatory environment. In rate-of-return states, utilities typically can recoup their capital costs for environmental controls plus some reasonable rate-of-return. In restructured states, however, the utilities are not guaranteed a return on capital expenditures; therefore, economists argue that utilities in restructured markets are less likely to make capital investments. In fact, recent empirical evidence out of the University of California, Berkeley suggests that power plants in restructured states tend to install significantly less environmental control technology than those in rate-of-return states.8

On the other side, environmentalists argue that these provisions are necessary to ensure that older plants do not continue running unabated. As Sean Donahue, the lead attorney for the environmental petitioners, put it, “We’re very pleased that the court agreed to review the Duke Energy decision, which rests on a flawed interpretation of the Clean Air Act, and which industry has been citing in numerous other cases in an effort to undermine essential pollution controls applicable to some of the nation’s largest sources of air pollution.”

The Load Split

Because the 4th Circuit overturned the annual test, leaving only the hourly test, facilities in the five states in that circuit (Maryland, West Virginia, Virginia, North Carolina, and South Carolina) can make significant modifications to their plants without triggering expensive environmental-control requirements. This gives facilities in these states a competitive advantage over facilities in neighboring states that serve the same customer load. Even so, because of the Supreme Court’s review,9 facilities in the 4th Circuit probably will wait on the decision before making any major modifications.

Adding to this regulatory drama, 7th Circuit Judge Richard Posner—one of the most well-respected appellate court judges in the country—authored a recent opinion expressly disagreeing with the 4th Circuit’s decision in Duke Energy.10 According to Vickie Patton, a senior attorney for Environmental Defense and a former attorney in the General Counsel’s office at EPA, “This pivotal decision by one of the most influential federal court of appeals in the nation sends a powerful signal that it is time to address the serious human health and environmental impacts of coal-fired power plants. The court’s strong, clear decision also puts its considerable weight behind a protective interpretation of the Clean Air Act’s clean-up requirements for coal plants at the same time that the Supreme Court is considering … these very questions.”11 Judge Posner’s decision, which involved Duke subsidiary Cinergy, expressly refuted the 4th Circuit’s analysis, arguing that “the 4th Circuit stepped out of bounds” in deciding the issue.12

Bush Administration Flip-Flopping?

In the Duke Energy case, EPA continually has flip-flopped its position on the emissions-increase issue. EPA began the case as a plaintiff, suing Duke for modifying its facility and increasing its annual emissions without adding pollution-control technology. After losing in the appellate court, though, EPA abandoned its position and now is proposing to change its nationwide rule to an hourly test.14 Environmentalists are arguing that this is unacceptable and unquestionably will lead to a significant increase in nationwide pollution. Electric industry officials, on the other hand, posit that EPA’s new tradable emissions rules have eliminated the need for the New Source Review modification rule because these trading programs cap total nationwide emissions. In an even stranger move, EPA now is back to its original position on the emissions-increase rule in front of the Supreme Court—arguing for the annual test even though it simultaneously is proposing a nationwide hourly rule change.

The High Court Dukes It Out: The Oral Arguments

Surprisingly, Duke’s attorneys won the oral argument bout. Before the arguments, most legal scholars believed that the court would not side with Duke. After the arguments, however, the consensus changed. There is now a strong possibility that the Supreme Court will uphold the 4th Circuit’s decision.

The Supreme Court arguments were much livelier than many expected, with Chief Justice John Roberts delivering an unusually sharp jab to Donahue, the counsel for environmental petitioners. When Donahue explained in passing that the regulations at issue were “clear on their face,” Roberts couldn’t let it go. “That’s an audacious statement,” he opined. After much courtroom laughter, Justice Antonin Scalia, who seemed to side with Duke, stated that the justices had “wrestled with that for several days,” referring to the meaning of the regulations. Justice Samuel Alito then came in for the finishing below on this issue, pointing out that experts have interpreted it differently: “Doesn’t that show that they’re not clear on their face?” he asked.

Of course, there also was the “whipsaw” argument, which now has snowballed into a highly contentious issue—in fact, the argument still has environmental groups fuming.15 Duke’s attorney, Carter Phillips, who also is managing partner of Sidley Austin’s Washington office, argued that EPA only recently has started using the annual test. It only took a few minutes of Donahue’s argument for Scalia to follow up.

“What I am concerned about,” he mused “is that companies can get whipsawed. They don’t challenge the regulations when they come out. … And then some years later, when it turns out the agency is using a different interpretation, you have the jurisdictional bar.”

Although both Donahue and the EPA attorney pointed out that the annual test has been around since at least 1988, in his opening remarks, Duke attorney Phillips shot back: “It seems to me it is very clear that the understanding of everyone in the industry, outside the industry, from 1980, candidly well beyond 1988 all the way up to 1999, was the regulations didn’t apply under any circumstances in the absence of an increase in the capacity.” Unfortunately for Mr. Phillips, this is simply incorrect. The environmental groups dumped literature proving otherwise on many reporters the day after the case, including one of Duke’s own company training guides from 1991 that references the annual test.16

The Economists Chime In

In October, just before the November arguments, two professors affiliated with the Center for Economic Development and Research released a timely study at a Capitol Hill briefing in Washington, D.C., arguing: “Should Environmental Defense v. Duke Energy be overturned by the U.S. Supreme Court, rising electricity costs from compliance with the EPA’s new interpretation of NSR requirements likely will fall disproportionately on rural businesses and households, especially those with the least financial ability to pay higher utility rates.”17

While this paper garnered considerable press and agency attention,18 its findings are not supported by any empirical evidence.

Essentially, the paper argues that “because rural America relies so heavily on coal-fired electricity, the EPA reinterpretation of NSR would be especially harmful to these already distressed communities.”19 It points out that rural businesses consume a higher percentage of electricity than metropolitan ones.20 It then essentially argues that these two facts, coupled with the Supreme Court overturning the 4th Circuit’s decision, would wreak havoc on rural America. Other than the obvious lack of empirical proof, the primary flaw of the paper is similar to that argued by Duke attorney Phillips. Both assume EPA’s 1999 lawsuits triggered a change in EPA’s interpretation of “modification” from the hourly test to the annual test. This is simply incorrect. If the Supreme Court overturns the 4th Circuit’s decision, the rule would revert back to EPA’s annual test, which has been around since at least 1988. Moreover, EPA might change its rule to the hourly test anyway. Fortunately, the other economics study out of Yale was much more thought-provoking and useful to the regulated community.

The Yale study, which also was released in October, uses data from the U.S. electric power industry to explore two questions: (1) What factors increase the likelihood of EPA New Source Review enforcement? and (2) How did utilities respond to the threat of future New Source Review enforcement and actual enforcement?

As to the first question, the study found that “larger firms were more likely to be sued, with a 15 percent increase in the size corresponding to a one percent increase in the probability of a lawsuit.”21 The authors posit that this is likely due to EPA choosing higher-profile cases to maximize deterrence.22 The study also finds that increases in historical emissions and capital investments lead to a corresponding increase in the probability of enforcement: “A 1,000-ton rise in a plant’s maximum year-on-year emissions increase boosted the probability of a lawsuit by one percent,” as did a change of $23 million in the maximum investment in a single year.23

As for how the utilities responded to enforcement, the study finds that as the probability of being sued increased, the plant’s emission rates decreased.24 According to the study, the “findings suggest that firms reduced emissions in 1999 in response to the threat of lawsuits, regardless of whether their plants were sued or not.”25 Perhaps the most surprising finding is that the actual lawsuits had no statistically significant effect. In other words, the lawsuits themselves had no effect on the utilities’ emissions behavior.26

So what will the Supreme Court justices make of the Yale study? The results seem to imply that New Source Review, and more particularly the threat of agency enforcement of New Source Review, can cause emission reductions. If the Supreme Court upholds the 4th Circuit’s decision, utilities will worry less about New Source Review enforcement, as almost all modifications will pass the weaker hourly test. It follows, then, that the study’s findings support the use of the annual test.

The Duke Decision’s Ramifications

Let’s hope that the Supreme Court will put an end to the incessant fighting over the New Source Review modification rule. The court basically has three choices. It could do nothing, it could leave the rule as it is (the annual test), or it could change the nationwide rule, at least temporarily, to an hourly test. Because of a technical jurisdictional argument from the environmental petitioners, the court could overturn the 4th Circuit’s decision, without overturning the 4th Circuit’s analysis on the New Source Review modification rule. In so doing, the court would not need to address the New Source Review issues. While this would eliminate the split in the circuits, it would leave lingering doubts as to the validity of EPA’s use of the annual test. Although EPA has flip-flopped its position in the case, if this occurs, EPA may continue with its proposed rulemaking, thereby changing the national rule to the hourly test. This would allow utilities to modify their equipment without requiring additional environmental controls. Keep in mind, though, that EPA rule changes can take years and often are subject to court challenges.

The court’s other two alternatives involve deciding the primary issue in the case. If the Supreme Court overturns the 4th Circuit, EPA’s annual emissions test stands, and the modification rule continues unabated. If the Supreme Court upholds the 4th Circuit, the national rule, at least for a time, would revert to the hourly emissions test, giving utilities a small window across the country to modify their plants without repercussion. Utilities would have to act quickly, though, primarily because many states would scurry to change their state rules to prevent modifications from occurring. Even so, for a time, the electric industry would be almost entirely free of the New Source Review modification rule and its strenuous requirements.

If utilities are allowed to extend their fleet’s lives significantly, the environment will suffer, but reliability, safety, and efficiency will prosper. If they are forced to add pollution controls, many of these older plants likely will be retired. These retirements will put a strain on the grid and the ISOs, increasing the need for new baseload generation. For the next few months, then, utility investment and environmental protection hang in the balance, awaiting the high court’s pronouncement.



1. Charles Lane, “Justices to Hear Environmental Appeal on EPA Emissions Rule,” Wash. Post. A04 (May 16, 2006).

2. Nathaniel O. Keohane, Erin T. Mansur, & Andrey Voynov, Averting Enforcement: Strategic Response to the Threat of Environmental Regulation, Yale School of Management and School of Forestry and Environmental Studies Research Paper, available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=935083; Bernard L. Weinstein & Terry L. Clower, Environmental Defense v. Duke Energy: How a Supreme Court Reversal on the Interpretation of New Source Review Could Imperil Rural America, Center for Economic Development and Research, available at http://www.nreca.org/Documents/PublicPolicy/Newsourcereviewupdate10-06.pdf.

3. Hearing Before the Committee on Environment and Public Works, Health Impacts of PM2.5 Associated With Power Plant Emissions, 107th Congress, at 3 (Oct. 2, 2002).

4. See U.S. General Accounting Office (GAO), Air Pollution Emmissions from Older Electric-Generating Units 3 (2002) (GAO 02-709) available at http://www.gao.gov/new.items/d02709.pdf.

5. See 42 U.S.C. §§ 7501-7515, 7470-7492

6. See New York v. EPA, 443 F.3d 880 (D.C. Cir. 2006).

7. See, e.g., Robert N. Stavins, The Effects of Vintage-Differentiated Environmental Regulation, Resources for the Future Discussion Paper (05-12), 1 (March 23, 2005), available at http://www.rff.org/Documents/RFF-DP-05-12.pdf.

8. Meredith Fowlie, Emissions Trading, Electricity Industry Restructuring, and Investment in Pollution Abatement, Unive of Cal. Berkely Job Market Paper, at 22 (“An increase in the capital cost of a compliance option decreases the probability that the option will be chosen by a plant in a restructured electricity market.”).

9. See U.S. Supreme Court, Docket No. 05-848, http://www.supremecourtus.gov/docket/05-848.htm.

10. State of New York, et al. v. Cinergy Corp., No. 06-1224 (Aug. 17, 2006).

11. Environmental Defense Hails 7th Circuit Court Decision Supporting Its Suit to Stop Weakening of Clean Air Act, U.S. Newswire (Aug. 17, 2006) available at http://releases.usnewswire.com/GetRelease.asp?id=70931.

12. Id. at 7.

13. Data taken from U.S. General Accounting Office (GAO), Air Pollution Emissions from Older Electric-Generating Units 3 (2002) (GAO 02-709) available at http://www.gao.gov/new.items/d02709.pdf and Energy Information Administration Web site. The price data is from 2005 and the pre-1971 unit data is from 2000.

14. Prevention of Significant Deterioration (PSD), Nonattainment New Source Review (NSR), and New Source Performance Standards (NSPS): Emissions Test for Electric Generating Units, 70 Fed. Reg. 61,081 (proposed Oct. 20, 2005) (proposed changes to 40 C.F.R. pts. 51 & 52).

15. See Pete Yost, Groups Upset by Claim in Clean Air Suit, Forbes (Nov. 24, 2006).

16. Id.

17. Weinstein & Clower, supra note 2, at 19.

18. See, e.g., On Eve of High Court Suit, Utility Study Lays Ground For New NSR Debate, Inside the EPA (Oct. 27, 2006).

19. Id. at 11 (emphasis added).

20. Id. at 13.

21. Keohane, et al., supra note 2, at 13.

22. Id. at 13-14.

23. Id. at 13.

24. Id. at 15.

25. Id.

26. Id. at 17.