In Utility Air Regulatory Group v. EPA, seven members of the Supreme Court upheld the most important feature of the EPA’s Prevention of Significant Deterioration (PSD) program: the ability to require the vast majority of new and modified sources to install the “Best Available Control Technology” for reducing greenhouse gases (GHGs). As a consequence, eighty-three percent of significant new and modified sources will continue to be subject to the BACT requirement for their GHG emissions. Although the Court reversed, by a five-to-four vote, EPA’s contention that greenhouse gas emissions alone could trigger the PSD program, that reversal will have little impact because it will eliminate PSD requirements for only about three percent of significant stationary GHG sources. Justice Scalia’s majority opinion had some choice words for EPA, but it remains to be seen whether those words spell trouble for newly emerging climate regulations.
The Backdrop: EPA Regulation of Stationary Sources under the PSD Program
After Massachusetts v. EPA held that GHGs are “air pollutants” under the CAA, EPA initiated a series of regulatory measures. EPA concluded that GHGs “endanger” public health and welfare and proceeded to regulate GHG emissions from mobile sources under Title II of the Clean Air Act. Regulation under the mobile source provisions of the CAA triggered the applicability of the PSD program for large stationary sources, since, according to EPA’s long-standing interpretation, the PSD program applies to any pollutant once it’s regulated under the Act. (EPA’s regulations are described further below.) EPA then had to deal with a tricky issue: the CAA applies PSD only to ”major emitting facilities,” which the statute defines as sources emitting over 250 or 100 tons per year (depending upon the industry). However, although these thresholds make sense for defining “major” sources of conventional pollutants, they don’t work for GHGs (particularly carbon emissions) because these thresholds would include many small sources, like hospitals, restaurants, etc. Therefore, EPA, in its “Tailoring Rule,” limited the PSD program’s applicability to sources emitting 100,000 and 75,000 ton per year of GHGs, thresholds that would better capture the large sources that Congress had intended to target.
Industry challenged all of these measures in the D.C. Circuit. The court, in Coalition for Responsible Regulation v. EPA, upheld the Endangerment Finding and agreed with EPA that regulation under the mobile source provisions triggered the applicability of the PSD program. The D.C. Circuit did not address the validity of the Tailoring Rule because the rule lightened the challengers’ regulatory burden and therefore failed to generate the injury necessary to prove standing.
Industry petitioned for certiorari. The Supreme Court denied cert on all claims except one: whether regulation under the mobile source provisions triggered the PSD program, in general or in part. The claim raises two questions. The first is whether GHG emissions alone trigger the applicability of the PSD program. The second is whether, even if GHG emissions alone do not trigger the applicability of the PSD program, facilities that are regulated under the PSD program anyway, due to their conventional pollutant emissions, must apply BACT to their GHG emissions.
Whether GHG Emissions Alone Trigger the PSD Program
The PSD program applies only to “major emitting facilities.” The Clean Air Act defines “major emitting facilities” as “stationary sources … which emit, or have the potential to emit … above 250 or 100 tons per year of any air pollutant. 42 U.S.C. § 7479. As noted above, EPA’s Tailoring Rule adjusted the threshold quantity so that the program would apply only to large GHG sources.
A key interpretive puzzle is how to define the term “any air pollutant.” As all parties and the Court concede, the term can’t be taken literally because the definition of “air pollutant” is very broad, including many non-harmful substances. Since the inception of the PSD program, EPA has interpreted the term “any air pollutant” to mean any regulated air pollutant. 40 C.F.R. § 52.21(b). In that way, EPA has regulated only harmful pollutants deemed worthy of regulation. Because EPA limited the PSD program to pollutants regulated under the Act, it did not apply the PSD program until GHGs were regulated under Title II, the mobile source provisions.
The Supreme Court did not question or appear to unsettle EPA’s basic regulatory interpretation that “any air pollutant” means any air pollutant regulated under the Act. Justice Scalia, writing for the five-to-four majority on this issue, recognized that the meaning of terms must often be read in context. Because “any air pollutant” is too broad to be taken literally, the question is how to limit the term. Here, the Court concluded that GHGs could not be considered “air pollutants” for purposes of triggering the PSD program because, when applied to GHGs, the statutory thresholds do not work as intended. The 100/250 ton thresholds would bring in thousands of additional sources, creating extensive regulatory burdens on industry and agencies alike, as well as extending regulatory authority over small sources that Congress had never intended to regulate under the PSD program. (slip op. at 18-20) Although EPA attempted to avoid this situation with the Tailoring Rule, the Court held that EPA did not have the power to rewrite the statute and must, instead, forgo treating GHGs as air pollutants for purposes of defining major emitting facilities subject to the PSD program. In other words, because the existing statutory thresholds make application to GHGs unworkable and contrary to the statute, the statute cannot be applied to GHGs. (slip op. at 23-24)
Justice Breyer, in dissent, recognized that EPA has a problem – that the statutory thresholds for defining major emitting facilities do not work for GHGs. As he says: “What is to be done?” (Breyer dissent, slip op. at 4.) He notes that there are two choices: (1) create an exception to the term “air pollutant” for GHGs, so that PSD can be triggered by any air pollutant regulated under the Act except GHGs (the majority’s approach) (see Breyer dissent, slip op. at 4-5); or (2) create an exception for small sources by adjusting the statutory thresholds for GHGs (Breyer dissent, slip op. at 7), as EPA did with its Tailoring Rule. He argued that, in a context where some exception must be made, the Court picked the wrong exception. According to Justice Breyer, creating an exception for small sources is more consistent with congressional intent than creating an exception from the normal definition of air pollutants for GHGs. (Breyer dissent, slip op. at 7-11) It makes more sense to adjust the statutory threshold to ensure that it captures only large sources (EPA’s approach) than it does to exclude a clearly harmful pollutant from regulation under the statute (the Court’s approach). Justice Breyer argued that his approach is truer to Congressional purpose because it allows EPA to regulate GHGs, which EPA has determined “endanger human health and welfare.” (Breyer dissent, slip op. at 11)
Although the question of whether GHGs alone trigger the PSD program is the primary focus of the majority and of J. Breyer’s dissent, its practical impact on GHGs and on the PSD program more broadly is limited. Eighty-three percent of large GHG sources fall under the PSD program because of their conventional pollutant emissions, emissions which exceed the existing statutory thresholds; only three percent of large GHG sources would trigger the PSD program based upon their GHG emissions alone. (slip op. at 9-10) The fate of EPA’s regulation of GHGs under the PSD program thus turns largely on whether EPA can apply BACT to GHG emissions from sources who come under the PSD program due to their conventional pollutant emissions – known as “anyway sources” because they come under PSD “anyway.”
Before turning to that question, however, it is important to note that the Court’s ruling largely preserved EPA’s traditional PSD regulations, creating a narrow exception that applies only to GHGs. Petitioners had advanced arguments that sliced more deeply at EPA’s PSD regulations, and that could have had farther-reaching impacts on EPA’s PSD program. For example, Petitioners had argued that the PSD program should apply only to criteria pollutants for which NAAQS have been established. Under that approach, regulation of any non-criteria pollutants, not just GHG emissions, would no longer trigger the PSD program. Petitioners had also argued that only pollutants with local effects, not global effects, should trigger the PSD program. Under that approach, regulation of pollutants like ozone-depleting substances that have global rather than local impacts would not trigger PSD. By limiting the exception to GHGs, the Court left EPA’s PSD program for all other air pollutants intact.
Applying the Best Available Control Technology Requirement to GHGs
Even if GHGs are not considered “air pollutants” that could trigger PSD on their own, the story is not over. The Clean Air Act explicitly states that all facilities subject to PSD must install BACT “for each pollutant subject to regulation under this chapter” – with “chapter” referring to the CAA as a whole. 42 U.S.C. § 7475(a)(4). In other words, facilities that are “major emitting facilities” because they emit conventional pollutants above the statutory thresholds must apply BACT to all pollutants regulated under the Clean Air Act, not just the pollutants that triggered the applicability of the PSD program. So, if a power plant emitted above the threshold quantity of sulfur dioxide, a criteria pollutant, then PSD would be triggered. Once PSD is triggered by the criteria pollutant emissions, the facility must apply BACT to GHG emissions above a de minimus level, since GHGs are “regulated” under the CAA’s mobile source provisions.
Seven members of the Court upheld EPA’s application of BACT to GHG emissions from “anyway sources” – sources that would be subject to PSD anyway, regardless of their GHG emissions. The statutory language clearly requires the agency to apply BACT to “any pollutant regulated” under the statute. It is therefore less ambiguous than the language for triggering PSD in the first place, where the phrase “any air pollutant” cannot be interpreted literally and therefore creates uncertainty about its appropriate scope. And because the BACT provision, unlike the “major emitting facility” provision, does not contain emissions thresholds, the provision does not push EPA into the awkward position of having to deviate from the statutory terms to apply the provision.
Somewhat unclear is whether EPA will have to engage in further rulemaking to clarify that its 75,000/100,000 ton thresholds constitute the “de minimis” levels that trigger BACT. The Court noted that the agency had not made this finding (slip op. at 28-29), and it is possible that an additional proceeding will be necessary to clarify that these are the “right” levels for determining what constitutes a greater-than-de-minimis level of emissions that triggers the BACT requirement.
Justices Alito and Thomas dissented, arguing that it makes little sense to argue that GHGs don’t trigger the PSD program and then apply the most significant feature of that program to them anyway. They also embraced Petitioners’ arguments that the PSD program, with its careful case-by-case and context-specific analysis, is simply ill-suited and inappropriate for global pollutants like GHGs.
Because 83 percent of large new sources of GHGs are likely to trigger PSD based upon their conventional pollutants, and only three percent of large new sources would trigger PSD based upon their GHG emissions alone, the ability to apply BACT to “anyway” sources largely preserves EPA’s approach to GHGs. The Court trimmed EPA’s PSD program for GHGs, but only slightly.
EPA and the Supreme Court
Although the Supreme Court’s ruling allows EPA to apply the central components of the PSD program to GHGs, Justice Scalia had harsh words for EPA’s interpretation of “major emitting facilities” subject to the PSD program – words that appear perhaps unduly harsh in light of the agency’s objectives. Because the statutory thresholds would extend regulation to many small sources, the Court sometimes implies that EPA is attempting to assert that jurisdiction. The court states that EPA’s interpretation would not only be burdensome, but “would bring about an enormous and transformative expansion in EPA’s regulatory authority without clear congressional authorization ….” (slip op. at 19) The Court does acknowledge EPA’s recognition that the CAA was not intended to extend PSD regulation so broadly, but suggests it is “patently unreasonable – not to say outrageous – for EPA to insist on seizing expansive power that it admits the statute is not designed to grant.” (slip op. at 20) But EPA’s Tailoring Rule was designed precisely to avoid the power grab that Justice Scalia accuses the agency of attempting, and so Justice Scalia’s depictions of a power-hungry agency appear over-wrought.
That said, it is possible to see how the Tailoring Rule could prompt some concern. EPA tread on thin ice by attempting to reconcile two irreconcilable positions. On the one hand, EPA recognized that the PSD program was designed to address only large sources and that application to small sources was inconsistent with the statute; hence the Tailoring Rule’s high thresholds. On the other hand, the agency was concerned about ignoring the low thresholds plainly stated in the statute, and so suggested that it might, in the future, meaningfully regulate smaller sources. In so stating, however, EPA raised the possibility (however remote) that it might later extend its authority in the very way it just stated was inconsistent with Congress’ intent. This clearly made Justice Scalia nervous; as he states: “We are not willing to stand on the dock and wave goodbye as EPA embarks on this multiyear voyage of discovery.” (slip op. at 23)
EPA might have avoided the “power grab” accusation, and the appearance of inconsistency, if it had simply created a permanent exception for all small sources and not floated the possibility of future controls. But the opinion suggests that Justice Scalia would not have accepted that either. He states that “an agency may not rewrite clear statutory terms to suit its own sense of how the statute should operate,” and that EPA thus lacked the authority to tailor the Clean Air Act. (slip op. at 23-24) What Justice Scalia does not explain, however, is how his solution is any less a re-writing of the statute. He readily accepts that PSD can be triggered by air pollutants regulated by the Act, and then does his own re-writing of the statute by adding “except if they are GHGs.” As Justice Breyer observed, adjusting the statutory thresholds does less violence to the Clean Air Act’s purpose than excepting GHGs from the class of air pollutants regulated under the Act. If the statutory thresholds don’t work, why would it be more reasonable for EPA to relinquish control over a grave threat to public health and welfare rather than adjust the thresholds to match Congress’ intent to address large sources?
What, if Anything, Does UARG Mean for EPA’s Clean Power Plan?
The question de jure is, of course, what the opinion (and its criticism of EPA) means about the eventual fate of EPA’s newly unfolding climate change regulations, including its controversial proposals to regulate new and existing power plants under Clean Air Act § 111. The proposed Clean Power Plan for controlling existing power plants is particularly noteworthy because EPA set guidelines for states based on an innovative and expansive reading of the “best system of emission reduction.” Rather than focusing solely on source-specific technology-based controls, the agency took a system-wide approach predicated on a wide range of emission-reducing activities, some of which, like renewable power and consumer energy efficiency, do not occur at power plants themselves. Does the Court’s tone in UARG suggest how the Court will respond to EPA’s approach?
The answer is unclear. The Court’s concern about an EPA “power-grab” suggest the Court’s vigilance over efforts to expand authority. In addition, although the Court upheld EPA’s application of BACT to GHGs, the Court expressed some sympathy for the Petitioners’ concern that BACT could lead to potentially far-reaching controls that go beyond site-specific adjustments. (slip op. at 25-27) The Court found that the BACT provision did not justify these fears, but EPA’s proposed Clean Power Plan, with its system-wide approach to reducing power sector emissions, might be more at risk of being viewed as a “power grab” over energy policies that impermissibly extends beyond individual facility controls.
On the other hand, in UARG, the Court was uncomfortable with the blatant re-writing of the statutory provisions that was necessary to effectuate the PSD program’s application to GHGs. Statutory interpretations that do not require such an obvious administrative adjustment to the statute might face less skepticism. EPA’s Clean Power Plan provides a robust defense of the agency’s system-wide approach, and EPA’s interpretation does not require obvious exceptions or adjustments to the statutory text.
And UARG was not the only EPA case decided this term. Tea leaves from two cups of tea must be read; in EME Homer, the Supreme Court upheld EPA’s innovative approach to interstate pollution, even though EPA’s innovative and sophisticated interstate trading program for criteria pollutants was created by the agency with little direction from the statute. Thus, innovation and breadth alone will not doom agency programs.