The Supreme Court seemed inclined to seek a middle way with respect to the EPA’s Clean Air Act (CAA) regulations regarding stationary source emissions of greenhouse gases (GHGs) at yesterday’s oral argument in Utility Air Regulatory Group v. EPA, with at least Chief Justice Roberts showing interest in an interpretation of the Act that would allow the EPA to regulate most large GHG sources under the CAA’s Prevention of Significant Deterioration (PSD) provisions.

The Supreme Court granted certiorari on a single question regarding the suite of EPA rules regulating GHG emissions from motor vehicles and stationary sources that were upheld by the D.C. Circuit in Coalition for Responsible Regulation v. EPA in June 2012: whether the EPA "permissibly determined that its regulation of greenhouse gas emissions from new motor vehicles triggered permitting requirements under the Clean Air Act for stationary sources that emit greenhouse gases" under the CAA's PSD and Title V provisions. The crux of this issue is the validity of the EPA's decision that it was compelled to regulate GHG-emitting stationary sources under the PSD and Title V programs as sources of "any air pollutant" regulated under the CAA, but that it would not abide by the specific numeric emission thresholds in the statutory definitions of "major" facilities subject to PSD and Title V permitting requirements – thresholds that are so low that they would sweep in millions of previously unregulated sources, triggering billions of dollars of costs and years of delays in the permitting process. The EPA instead issued the "Tailoring Rule," which provided that the EPA would presently regulate only sources that emit significantly larger amounts of GHGs. In departing from the plain statutory language, the EPA primarily relied on the rationale that applying the statutory thresholds would lead to "absurd results," although the agency also invoked the "administrative necessity" and "one-step-at-a-time" doctrines as a basis for delaying regulation of smaller sources.

The petitioners advanced two main arguments against the EPA's approach. First, they broadly asserted that GHGs should not be considered an "air pollutant" for purposes of regulation under the CAA's PSD and Title V provisions, given other language in the PSD provisions that would support limiting the term "air pollutant" to those pollutants with local environmental effects and avoid the absurd results cited by the EPA. In making this argument, the petitioners sought to distinguish the 2007 decision Massachusetts v. EPA, in which the Supreme Court held that GHGs do constitute an "air pollutant" under the CAA, as limited to the motor vehicle provisions in Title II that were at issue in that case.

In the alternative, the petitioners advocated for an interpretation of the CAA's PSD provisions that would limit their application to those GHG-emitting stationary sources that emit major quantities of pollutants subject to National Ambient Air Quality Standard (NAAQS) for which a geographic area is in attainment – often generally termed "anyway" sources because they would be subject to the CAA's PSD requirements because of their emission of NAAQS pollutants regardless of their GHG emissions. This definition would still allow for application of Best Available Control Technology (BACT) to GHG emissions from these sources based on the statutory language requiring BACT for “each pollutant subject to regulation under” the CAA.

Peter Keisler, arguing for the private petitioners, focused on the first of these possibilities: that the PSD program was designed to address only air pollutants with “local” effects, and thus GHGs should be excluded in interpreting the ambiguous term “air pollutant” as used in the PSD provisions rather than allowing the EPA to “rewrite” the unambiguous statutory thresholds for major sources. This reading was predictably unpopular with the liberal justices, with Justice Kagan suggesting that the petitioners’ construction would be incompatible with the plain language meaning of “air pollutant,” particularly as construed in Massachusetts v. EPA. Justice Kagan proposed that the EPA therefore acted within its authority to pick the “most sensible” approach to implementing the CAA that would avoid absurd results while abiding by Congress’s intent to apply PSD to large sources of air pollution. Justice Breyer went a step further and speculated that the EPA might have the power to read in an “unwritten exception” in limited situations where it would not make sense to apply a statute, although he, the rest of the court, and all of the parties struggled to identify any precedent that might support such authority.

Justices Scalia, Alito, and Roberts expressed discomfort with allowing the EPA to decide what regulations might make sense without congressional guidance. Texas Solicitor General Jonathan Mitchell, representing the state petitioners, hammered this point in arguing that the EPA’s approach was “atextual” and should not be endorsed given the flexibility of the term “air pollutant.” 

U.S. Solicitor General Donald Verilli argued on behalf of the EPA, facing tough questioning from several members of the court. He did manage to articulate a view of the CAA PSD provisions very different from the “local pollution” version imagined by the petitioners, arguing that the PSD program was enacted as a backstop to national source category emissions standards that were being promulgated at a relatively slow pace, and thus was intended to reach beyond NAAQS pollutants alone to “fill the breach” pending the issuance of source category standards. He was less successful at explaining why the EPA’s choice of alternative PSD triggering thresholds does not represent the “rewriting” of the statute charged by the petitioners, struggling to describe the EPA’s approach as one aimed at the eventual reconsideration of how to determine what sources emit at the statutory major threshold levels rather than as a simple substitution of new thresholds. In particular, Chief Justice Roberts questioned what “intelligible principle” would guide the EPA once it departed from the language of the statute. Justice Sotomayor may have rung the death knell for the EPA’s chances at a victory when she asked Verilli how the EPA would prefer to lose, forcing Verilli to endorse the interpretation under which PSD applicability would be triggered only by emissions of NAAQS pollutants, while regulated sources would still be required to implement BACT for GHGs.

The petitioners also appeared to gain an upper hand with the court in arguing that the result of this case is not dictated by the holding of Massachusetts v. EPA that GHGs meet the definition of “air pollutant” for purposes of Title II or the holding of American Electric Power v. Connecticut that the EPA has authority to regulate GHGs under CAA section 111, 42 U.S.C. § 7411, which authorizes the EPA to issue standards of performance for emissions of pollutants from particular source categories. Keisler was careful to distinguish those separate provisions of the CAA as ones that could be applied to GHGs without requiring “extreme measures” or “counterintuitive results,” a question that the court looked to in Massachusetts v. EPA in endorsing a broad interpretation of “air pollutant” for purposes of regulation of GHG emissions from motor vehicles. As Keisler explained, neither of those cases requires that the same definition be applied “every time the word pollutant appears in the” CAA, and the EPA does in fact read the term differently in different situations (for example, confining it to visibility-impairing pollutants in the CAA provision concerning that issue). Thus, the petitioners seemed to successfully map out a winning position – that “air pollutant” is in fact an ambiguous term and therefore, as Justice Scalia suggested, the EPA was obligated to adopt an interpretation of that term that would not create absurdity.

While Justice Kagan appeared to be a full-throated defender of the idea that the court should defer to the EPA’s approach in the situation where there is no way to implement the plain language of a complex statute like the CAA without “violating a statutory term,” there was no indication that any of the five conservative justices might join her in this view. Justice Kennedy, who joined the majority in Massachusetts v. EPA, spoke little during the argument, and his few questions did not reveal much of his thinking regarding the case. However, Chief Justice Roberts pursued a line of questioning that suggests he might be a candidate to support a result that would allow the EPA to continue some regulation of GHGs under the PSD program. Chief Justice Roberts asked both the petitioners and the EPA about the alternative argument that PSD provisions, although not triggered by GHG emissions alone, do require otherwise regulated major sources to implement BACT for their GHG emissions. As the Chief Justice noted, this construction would still allow the EPA to regulate 83% of GHG stationary source emissions, versus the 86% of emissions that would subject to PSD under the EPA’s more expansive reading of the statute. In particular, he asked the EPA why the court should allow a “fight over 3%” to create the myriad problems raised by the EPA’s approach. Although Verilli insisted that this narrowed view of which pollutants could trigger PSD requirements would also require the broader rejection of the EPA’s 34-year interpretation of the triggering definition of major sources as encompassing non-NAAQS as well as NAAQS pollutants, his repeated invocation of sulfuric acid mist as one of the only examples of a non-NAAQS pollutant that might trigger the application of PSD requirements did not seem to concern the justices unduly.

Justice Breyer seemed to crystallize the dilemma facing the court when he asked Mitchell which of the interpretations proposed by the parties “does less violence” to the statute. As a practical matter, there may be five votes agreeing that the best candidate is a reading of the statute that applies PSD to “anyway” sources but goes no further, accommodating both the CAA’s purpose of addressing air pollutants including GHGs and the need to avoid imposing an absurdly onerous permitting burden on the sources regulated by the EPA.

Whatever the result in this case, most agree that the real main event is yet to come, in the form of the EPA's regulation of GHG emissions from power plants under CAA section 111. The EPA has already proposed New Source Performance Standards for natural gas- and coal-fired power plants, and President Obama has directed the agency to finalize a rule for existing power plants under section 111(d) by June 1, 2015. In the wake of this argument, it appears that no one will challenge the EPA’s overall authority to regulate GHG emissions from new power plants. However, the Supreme Court's decision here could have major ramifications for the next regulatory task of addressing existing power plants. The EPA faces significant questions as to what control measures it can require for existing power plants as the "best system of emission reduction" that has been "adequately demonstrated," particularly whether such measures can include requirement for emission reductions "beyond the fenceline" of individual power plants given language in section 111(d) indicating that it applies to individual sources. The court's verdict on the EPA's treatment of the statutory language at issue here may well inform the EPA's consideration of how much leeway it has to depart from that language in service of its regulatory aims as the agency implements section 111(d).