Clearing the Air

A critical review of Clean Air Act legal definitions

• By Chuck Wehland, Laurie Earl
• Jun 01, 2004

Recent legal developments show that the Clean Air Act is broken. The U.S. Environmental Protection Agency (EPA) and the courts are responsible for the legal malfunction. The U.S. Congress has the authority to repair the statute, but may not have either the institutional will or a clear blueprint for doing it.

A bedrock principle of American law is that courts should reach the same legal outcome every time they are presented with the same set of facts. This principle, called stare decisis, has taken a beating in the court decisions in the Clean Air Act enforcement cases against power plants.

Shifting Winds: Recent Court Decisions
EPA initiated enforcement actions against seven different utilities in 1999. Each of the enforcement actions involves substantially the same facts. Each utility performed life extension projects at coal-fired power plants. The details of the equipment involved vary from generating unit to generating unit. In each case, substantial elements of the unit, such as economizers, steam drums and turbine blades, were refurbished or replaced.

The utilities believed these projects were routine maintenance, repair or replacement exempt from Clean Air Act permit requirements. According to the EPA enforcement actions, the projects were not routine and constituted major modifications for which Prevention of Significant Deterioration (PSD) permits were required.

The ramifications of the PSD characterization are significant. There is no requirement to install emission control equipment when routine maintenance, repair or replacement occurs. By contrast, the best available control technology (BACT) must be installed to obtain a PSD permit for a major modification. The expense of retrofitting a coal-fired electric generating unit with this equipment can be substantial.

Four cases among the EPA actions described above have been decided in the past year, in four different courts. In the spirit of stare decisis, the outcome of each case should be the same. Instead, the four courts reached wildly different conclusions.

The first decision came in U.S. v. Southern Indiana Gas & Electric Co. 245 F.Supp. 2d 994 (S.D. Ind. 2003). The district court determined that the defendant utility had fair warning about EPA's interpretation of the routine maintenance rules, despite evidence provided by the utility that EPA was changing the way it interpreted the rules. The utility also strenuously argued that EPA should look at industry practice rather than practice at a particular generating unit in order to determine if a particular maintenance project is "routine" because life extension projects occur infrequently at individual generating units.

The court decided to look only at the practice at the particular generating unit in question to determine if a specific project was routine. The court found that EPA's interpretation was reasonable, persuasive and consistent with the plain language of the regulation and, consequently, the utility consequently had been fairly warned of the rules. On this reasoning, the court deferred to EPA's interpretation of routine maintenance and reserved a final determination on the new source review issue for a trial.

The court in U.S. v. Ohio Edison agreed with the Southern Indiana analysis of the fair notice and industry practice issues, saying that "the frequency of the activity at a particular unit . . . is most instructive in the analysis of what can be considered 'routine'" 276 F.Supp. 2d 829, 856 (S.D. Ohio 2003). At trial, the court considered many factors in determining that the utility's projects had not been routine, such as: (1) how the projects were budgeted and accounted, (2) the purpose and cost of the projects and (3) the project's resulting increase of net emissions.

In calculating net emissions, the court adopted EPA's approach for calculating projected future actual emissions in evaluating the significance of the increase. EPA's approach counted increased annual emissions resulting from projected reductions in outages and increases in utilization. The utility had argued for a calculation that relies only on changes in hourly emission rates and ignores projected increases in hours of operation. The Ohio Edison court concluded that the plain language of the Clean Air Act gave the utility fair warning of its need to obtain PSD permits for several life extension projects.

The district court in U.S. v. Duke Energy adopted a fundamentally different approach. 278 F.Supp.2d 619 (M.D.N.C. 2003). The Duke court concluded that EPA's past practice in other cases called for an examination of industry practice (rather than work on a particular unit) to determine if a project is routine. Examining this and other factors, the court issued a summary judgment decision that adopted significant portions of the utility's legal argument.

For example, the court gave no deference to EPA's interpretation of how to calculate net emissions increases in a PSD determination. As a threshold matter, the court determined that projected actual emissions are the correct gauge for measuring a net emission increase as the utility argued, instead of future potential emissions as argued by EPA. The court noted that its decision on this issue directly conflicted with Ohio Edison.

After lengthy examination of the legislative history and early EPA rules in the PSD program, the court indicated that "modification" in the PSD program should be defined the same as "modification" in the New Source Performance Standards (NSPS) program. In the NSPS program, a change in the maximum actual hourly emissions rate at an affected facility is necessary for a modification to occur. The court used this definition to determine that projected future actual emissions had to be determined holding constant the hours of operation and other utilization factors as they existed during the baseline period.

Tennessee Valley Authority v. U.S. is the only representative case so far that has been decided by a circuit court of appeals 33 Envtl. L.Rep. 20,231 (11^th Cir. 2003). The ruling was widely expected to establish a precedent for the district court cases. But the circuit court punted, issuing a procedural decision instead of a substantive one on the merits of the case. EPA had used a peculiar administrative process to initiate action against the Tennessee Valley Authority (TVA), while all of the previously discussed cases began as judicial civil enforcement proceedings. The circuit court decided that the administrative order mechanism used against TVA had unconstitutionally deprived it of rights to a hearing. This decision does not directly affect the other new source review actions, but it ultimately could have significant ramifications for EPA's enforcement powers.

Stare decisis does not exist in these four cases. These cases share the same underlying facts, but the results are very different and at times directly opposed to each other. When higher courts address the substance of the cases, some consistency between the courts may be achieved. But even if this occurs, EPA's current efforts to update its rules and enforcement policies may only serve to generate more confusion in the future.

Seeking Consistency
In October 2003, EPA issued a rule interpreting "routine maintenance, repair and replacement." Unlike the prior PSD rule, the new interpretation establishes bright lines and exemptions for projects that will be considered routine. Equipment replacement projects that cost less than 20 percent of the total cost of replacing an emissions unit are considered routine maintenance under the new rule. Several states filed suit to stop this rule change. The federal court stayed the implementation of the rule until the litigation is resolved.

Then, in November 2003, EPA dropped its investigations of PSD violations based on the previous versions of the rules. But in January 2004, EPA announced that investigations were continuing and that new cases would be brought. The uncertainty created by these changing regulatory directives makes development of a sound, cost-effective compliance program very difficult.

The law is broken when it can be interpreted in so many different ways by the courts and the agencies. Neither the courts nor the executive agencies seem capable of fixing the mess. Congress is the only authority with the power to impose order on the Clean Air Act chaos. Despite the Clear Skies Initiative proposed by the Bush administration, Congress does not seem ready to act.

Perhaps Congress does not want to act because the Clear Skies Initiative does not directly address the controversial issues associated with new source review and climate change. Instead, it proposes a cap and trade program for existing criteria pollutants. It leaves the new source review program unchanged, and it ignores climate change.

One possible legislative solution involves a critical examination of new source review program and creative use of existing Clean Air Act tools. The new source review program does not achieve the objectives desired by environmental advocates because its effect on air quality is indirect and diffuse, and businesses dislike its unpredictability. An alternative is to eliminate the new source review and incorporate plant-wide emission limits in operating permits. Sources could be required to model in order to show that emissions at these levels do not cause ambient air problems. Control technology still would be required through the NSPS and maximum achievable control technology (MACT) programs. This solution directly protects air quality while giving predictability to businesses.

This article originally appeared in the 06/01/2004 issue of Environmental Protection.