Supreme Court Holds that EPA Is Required to Consider Costs When Determining Whether Regulating Certain Power Plants Is “Appropriate And Necessary”


Yesterday in Michigan v. EPA, No. 14-46, the U.S. Supreme Court held that the EPA unreasonably determined that it should not consider costs in determining whether regulation of power plant emissions is “appropriate and necessary.”1 The Court emphasized that “appropriate and necessary” is a broad standard that instructs the agency to consider all relevant factors, and that it was unreasonable for the EPA to conclude that costs should not be one of those considerations.   This decision reaffirms the importance of cost considerations in agency decision making and makes clear that agencies must ordinarily take costs into account when regulating pursuant to broadly worded statutory authority. The decision may also reflect a continuing trend of the Supreme Court weakening Chevron deference to agency interpretations of statutes, and it emphasizes that administrative decisions must be evaluated solely on the grounds put forth by the agency during the administrative process. The decision does not vacate the  mercury regulation at issue,  so litigation regarding  that regulation and  other emissions regulations should only be expected to continue.


In the Clean Air Act Amendments of 1990, Congress required the Environmental Protection Agency (EPA) to study whether to regulate power plants.2  Based on that study, if the EPA determines that regulating power plants is “appropriate and necessary,” then it must set specific emissions standards to reduce pollution.3 The EPA eventually completed its study and decided that regulating coal- and oil-fired plants, but not natural gas-fired plants, is appropriate and necessary based primarily on the emission of mercury from such plants. The EPA then went on, after various administrative proceedings, to establish emissions standards for power plants.4 Notably, the EPA did not consider the costs of regulation at the first step (i.e., determining whether to regulate coal and oil plants at all), but did consider costs at the second step (i.e., determining where to set the emissions standards for coal and oil plants).5

Various regulated parties brought the current challenge, arguing that the EPA was required to consider costs in determining whether the regulation of power plants is “appropriate and necessary.” A divided panel of the D.C. Circuit disagreed, concluding—over a partial dissent by Judge Kavanaugh—that the EPA reasonably decided that it should not consider the costs associated with regulation to determine whether such regulation is “appropriate” under the circumstances.6    The Supreme Court then granted review.


In yesterday’s decision, the Supreme Court held that the EPA unreasonably interpreted the statutory term “appropriate” to “not allow for the consideration of costs” when deciding whether it would regulate power plants. By refusing to consider costs in determining whether regulation of power plants was appropriate, the Court held, the EPA’s decision did not rest “on a consideration of the relevant factors” and “strayed far beyond” the “bounds of reasonable interpretation” of the applicable statute.7

The Court noted that the word “appropriate” is “the classic broad and all-encompassing term that naturally and traditionally includes consideration of all the relevant factors.”8 In the Court’s view, by requiring the EPA to determine whether “regulation is appropriate and necessary,” the statute implicitly required the agency to consider both the “advantages and the disadvantages” of its decision to regulate.9 As “[a]gencies have long treated cost as a centrally relevant factor when deciding whether to regulate,” the Court found it “unreasonable to read an instruction to an administrative agency to determine whether regulation is appropriate and necessary as an invitation to ignore cost.”10 The Court found further support for its conclusion in the surrounding statutory context, noting that one of two other statutorily required studies—which the EPA interpreted as jointly providing the framework for its decision making process regarding power plant emissions regulations—expressly instructed the EPA to consider costs of technologies that could control emissions.11

The Court found the EPA’s arguments in favor of interpreting “appropriate” to exclude consideration of costs unpersuasive. The EPA pointed to other provisions of the Clean Air Act that expressly mention cost, but the Court held that the “broad reference to appropriateness” in the provision at issue in this case encompasses multiple relevant factors,” including cost.12   The Court also distinguished this case from its prior decision in Whitman v. American Trucking Assns., Inc., 531 U.S. 457 (2001), which held that the EPA could not consider costs under a different statutory provision when setting ambient air quality standards at levels “requisite to protect the public health” with an “adequate margin of safety,” concluding that “‘[a]ppropriate and necessary’ is a far more comprehensive criterion than ‘requisite to protect the public health.’”13 Finally, the Court rejected the EPA’s argument that because costs will necessarily factor into the second step of determining how much to regulate power plant emissions, the agency did not need to consider costs at the first step of determining whether to regulate power plant emissions. The Court explained that the only question before it was the meaning of “appropriate and necessary,” and that the fact that costs would be relevant at step two did not suggest that costs were irrelevant at step one.14

The Court concluded by rejecting arguments in support of the EPA’s rule that the agency itself had not made. Relying on the “foundational principle of administrative law that a court may uphold agency action only on the grounds that the agency invoked when it took the action,” the Court rejected the argument in Justice Kagan’s dissent that the EPA did not need to analyze costs at the first step because other features of the regulatory regime would ensure that the regulation would prove to be cost-effective.15   For the same reason, the Court rejected the contention of other parties supporting the EPA that the regulation should be upheld on the ground that “ancillary benefits”—i.e., benefits that resulted from the reduction of air pollutants other than mercury—would outweigh the imposed costs of the regulation.16 Ultimately, the Court did not require the EPA to “conduct a formal cost-benefit analysis,” but instead stated that it would be “up to the Agency to decide . . . how to account for cost.”17

Justice Thomas concurred, writing separately to question the Chevron doctrine on separation-of-powers grounds insofar as it requires courts to defer to agencies’ interpretations of broadly worded statutes.18 Justice Kagan dissented, joined by Justice Ginsburg, Justice Breyer, and Justice Sotomayor. She argued that, because the EPA knew that it would take costs into consideration “at multiple stages and through multiple means as it set emissions limits for power plants,” it was reasonable for the agency to “decline[] to consider costs at the opening bell of the regulatory process given that it would do so in every round thereafter.”19


Yesterday’s decision reaffirms the centrality of cost considerations to agency decision making pursuant to broadly worded statutory authority. The Court made clear that, when an agency is given authority to determine whether regulating in a certain area is “appropriate,” that determination ordinarily must include consideration of the costs of complying with the contemplated regulation. (In this respect, the decision is consistent with last Term’s decision in EPA v. EME Homer City Generation, L.P., which affirmed the EPA’s interpretation of a different provision of the Clean Air Act to permit consideration of costs.20)

Agencies are not necessarily required to conduct a “formal cost-benefit analysis in which each advantage and disadvantage is assigned a monetary value”; they will instead have flexibility in determining how to account for costs.21 Nevertheless, the obligation to consider costs will likely slow down the regulatory process by imposing an additional burden on agencies to justify their regulations as “appropriate.”

The decision may also reflect a continuing (though inconsistent) trend of weakening Chevron deference to agency interpretations of statutes. Broad statutory language such as “appropriate and necessary” may previously have been thought to afford agencies the widest possible latitude in determining how to act, but the Court in this case found the EPA’s interpretation of that phrase to be beyond the bounds of reasonable interpretation when considered in its proper context. Indeed, Justice Thomas expressly calls for reconsideration of the applicability of Chevron deference in cases involving such capacious statutory language. Another recent case decided this Term refused altogether to apply Chevron when there was reason to doubt that Congress would have delegated the particular question at issue to the administering agency.22    Chevron deference remains the norm (and a high hurdle), but the Court has occasionally shown more willingness to question its applicability in recent years. The Court also reaffirmed the fundamental principle of administrative law that courts should consider an agency’s action solely based on the justification provided by the agency at the conclusion of the administrative process, rather than on any other consideration raised during litigation.

For the energy sector, this case is the latest chapter in the longstanding and politically charged fight over air emissions regulation that began with Massachusetts v. EPA, 549 U.S. 497 (2007), which affirmed the EPA’s authority to regulate carbon dioxide emissions under the Clean Air Act. Because yesterday’s decision did not vacate the regulation at issue, that conflict will continue on remand in the D.C. Circuit, which will have to decide whether to allow the regulation at issue to remain in effect while the EPA conducts the necessary cost-benefit analysis. Those ongoing proceedings, together with the fact that many power plants have already expended substantial sums to install technologies designed to comply with the mercury regulation, may limit the immediate practical effect of yesterday’s decision. But the Court’s holding may fortify industry actors’ resolve in their continued efforts to resist increased EPA regulation. Given the complexity of emissions regulations affecting power plants and the significant costs and regulatory burdens involved, this case demonstrates that the energy sector will continue to be subject to ongoing regulatory uncertainty affecting the timing, scope, and cost of future regulatory compliance obligations.