On Balance: Major Questions about West Virginia v. EPA and Benefit-Cost Analysis
On June 30, 2022, the Supreme Court of the United States decided West Virginia v. Environmental Protection Agency. The case was about whether EPA in the Clean Power Plan could set the carbon emissions standard for existing power plants at a level that would require the power plants to reduce coal use and shift to or subsidize natural gas or renewable-energy electricity generation (referred to as “generation shifting”). But the focus was not on whether such a level would be benefit-cost justified—but rather on whether the agency was allowed to set a standard requires generation shifting for compliance under this provision of the Clean Air Act. Writing for the majority, Chief Justice Roberts applied the “major questions” doctrine and concluded that EPA could not do this despite reasonable textual support for it; on issues like this one, which have “vast economic and political significance,” Congress must clearly authorize an agency to act in this way.
It is tempting to conclude that this decision about statutory interpretation has no relevance for benefit-cost analysis (BCA). But unfortunately, the decision contains language that suggests the Court misunderstands and disregards the value of using BCA in federal regulatory policy.
The trouble starts when the Court appears to ignore how BCA can help agencies set stringency levels for various standards. The Clean Power Plan, for example, was supported by a BCA that concluded that the rule’s benefits to society would dwarf its costs, with $22.6 billion worth of net health and safety benefits each year in likely scenarios (EPA 2015). But the majority opinion refers to this analysis to highlight only the expected costs of EPA’s approach—without mentioning the expected benefits at all (p. 10). And more concerning, when the EPA argued that its discretion to set an emissions standard is bounded by the fact that it must consider costs, the Court suggested that this admission is itself somehow evidence that EPA’s authority was too vast (p. 25). In fact, the Court seemed to not understand how BCA could reasonably inform and constrain the agency’s choice of the stringency level—or, as the Court called it, the “emissions cap” (p. 29). The Court appeared to think EPA was unconstrained in setting the cap, calling it a cap set to “wherever the Agency sees fit” as opposed to a cap “based on some scientific, objective criterion” (p. 29-30).
It is possible that the Court’s confusion was partly a result of a strategic EPA decision to avoid justifying the cap with BCA, even though it was benefit-cost justified. EPA has previously avoided relying on its BCA in litigation in an attempt to preserve for itself the discretion and authority to set a standard that deviates from benefit-cost principles in the future (see Michigan v. EPA 2015). If so, this strategic decision once again backfired, suggesting unconstrained authority and producing skepticism among the justices that could threaten even BCA considerations.
In particular, the Court deemed it “highly unlikely that Congress would leave to agency discretion the decision of how much coal-based generation there should be over the coming decades” (p. 25). It then went further, speculating that “[t]he basic and consequential tradeoffs involved in such a choice are ones that Congress would likely have intended for itself” (p. 26). The Court tries to limit the reach of this observation by tying it to the availability of the “tool” of generation shifting, but its language could suggest broader criticism of agencies balancing the benefits and costs of regulatory options without explicit congressional authorization.
As an initial matter, a presumption against the availability of a tool for accomplishing statutory goals unless explicitly authorized by Congress reduces the usefulness of BCA. EPA required some generation shifting because it found it to be a benefit-cost justified approach to reducing carbon emissions in many cases. For BCA to be useful in informing agency decision-making, the analysis must consider all relevant benefits and costs associated with different regulatory approaches or stringency levels for meeting statutory goals. By implying statutory limits on the scope of such analysis, West Virginia v. EPA could make it difficult for agencies to consider and adopt creative approaches to accomplish statutory goals in efficient or, at least, lower-cost ways.
But the suggestion that Congress itself should balance tradeoffs—tradeoffs that are inherent in any regulatory decision—is even more concerning for BCA and regulatory policy in general. The Court tries to frame these particular tradeoffs as somehow different, requiring balancing benefits associated with reducing carbon emissions with costs that include energy price effects and grid reliability (p. 25). But is this context really that different than any other context where the agency must balance benefits of its preferred action with costs that often include price effects and otherwise unintended consequences?
The Court’s suggestion that Congress should balance tradeoffs underscores the close connection between the major questions doctrine embraced by the Court in West Virginia v. EPA and the nondelegation doctrine. The major questions doctrine is a canon of statutory interpretation that requires Congress to speak clearly when it wants an agency to make decisions of vast “economic and political significance.” The nondelegation doctrine, meanwhile, is an interpretation of the Constitution that prevents Congress from delegating some decisions to agencies, whether implicitly or explicitly. The doctrine has not been aggressively used to police congressional delegations since 1935, but several justices on the current Court have signaled a desire to revisit the contours of permissible delegations under the doctrine.
These questions—whether statutory or constitutional—about an agency’s authority to make some decisions naturally implicate BCA (see Cecot 2021). In the 1980 Benzene Case, for example, Justice Rehnquist wrote a concurrence arguing that the Occupational Safety and Health Administration’s regulation was invalid because Congress impermissibly delegated to the agency the decision of how to balance benefits of regulating benzene with costs, calling that decision “one of the most difficult issues that could confront a decisionmaker” (p. 672). In particular, “Congress was faced with a clear, if difficult, choice between balancing statistical lives and industrial resources or authorizing the Secretary to elevate human life above all concerns save massive dislocation in an affected industry” (p. 685). To Justice Rehnquist, because the statute did not clearly specify how stringently the agency should regulate, the statute violated the nondelegation doctrine by passing that major decision on to the agency.
The current practice of BCA occurs largely without explicit congressional requirements for relying on BCA—or even preparing or considering BCA. Instead, executive orders going back to President Reagan require agencies to conduct BCA and use it to inform regulatory decisionmaking to the extent permitted by relevant agency statutes. And many agencies have used BCA to inform their decisionmaking as long as their statutes do not explicitly prohibit the agency from going so. In fact, just a few years ago, in Michigan v. EPA, the Court appeared to support such a presumption. While the relevant provision did not explicitly mention costs, the Court unanimously agreed that “appropriate” regulation requires at least some comparison of benefits and costs. In particular, Justice Scalia, writing for the majority, declared that “[n]o regulation is ‘appropriate’ if it does significantly more harm than good” (p. 2707), and Justice Kagan, writing for the dissent, agreed that “EPA’s power plant regulation would be unreasonable if ‘[t]he Agency gave cost no thought at all’” (p. 2714). An expanded major questions doctrine—or worse, an expanded nondelegation doctrine—could upend this longstanding practice if the Court is not careful in defining the contours of the doctrines. Will the agency’s ability to consider costs when a statute is silent be once again likened to an “elephant[]” hiding in a “mousehole[],” as Justice Scalia famously proclaimed in Whitman v. American Trucking?
And finally, there is a fine line between courts appropriately constraining a rouge agency’s action outside of its likely authorization and courts inappropriately preventing an agency from acting flexibly and applying its expertise in its authorized domain. Congress often chooses to delegate implementation of some objective to an agency to take advantage of an agency’s expertise and its ability to respond quickly given new scientific and technical developments in a fast-moving area. Requiring the agency to go back to Congress before some (but not all) applications of its expertise, whether new approaches to old problems or old approaches to new problems, could extinguish the main advantage of giving the agency flexible authority in the first place. And the results could be very costly, ranging from insufficient congressional responses to wildly inefficient and inappropriately rigid and prescriptive congressional responses (Cecot 2021).
But maybe that’s the point. Chief Justice Roberts concludes the majority opinion by stating that even if “forc[ing] a nationwide transition away from the use of coal to generate electricity [is] a sensible solution to the crisis of the day,” the “decision . . . rests with Congress itself, or an agency acting pursuant to a clear delegation.” BCA is ultimately a tool that helps agencies pursue regulatory goals in “sensible” ways. And, simply put, the tool becomes less useful if sensibility in government decisionmaking is disregarded in favor of raw politics.
REFERENCES
Caroline Cecot, Congress and Cost-Benefit Analysis, 73 Administrative Law Review 787 (2021).
Environmental Protection Agency, Regulatory Impact Analysis for the Clean Power Plan Final Rule, EPA-452/R-15-003 (Aug. 2015).
The Benzene Case, 448 U.S. 607 (1980).
West Virginia v. EPA, 597 U. S. ____ (2022), slip opinion, https://www.supremecourt.gov/opinions/21pdf/20-1530_n758.pdf.
Whitman v. American Trucking Assns., Inc., 531 U. S. 457 (2001).