July 1, 2015

Michigan v. EPA: The Latest in a Series of Anti-Agency Decisions


Clean Air Act, EPA, Justin Pidot, King v. Burwell, Michigan v. EPA, Utility Air Regulatory Group v. EPA

supcourt4.JPG

by Justin Pidot, Associate Professor of Law, University of Denver Sturm College of Law

In its last decision of the 2014 term, the Supreme Court decided Michigan v. EPA, ruling that EPA must consider costs before deciding to regulate toxic air pollutants from power plants.  Lisa Heinzerling has identified the many questions that remain open in the wake of the Court’s decision.  And Dan Farber and Ann Carlson also provide insightful commentary on the meaning of the decision. As all three suggest, the lasting practical effect of the Court’s decision on mercury and other toxic emissions from power plants remains to be seen.

Without retreading ground that has been well-covered already, I want to offer two observations.  First, I want to offer some (very cautious) optimism that the legal rule provided by the Michigan v. EPA decision has little effect.  Read broadly, the decision could require agencies in many contexts to consider costs before regulating.  I’m not convinced, however, that the decision necessarily tells us anything about when agencies must consider costs. 

The Court offers several reasons that EPA unreasonably interpreted its authority to regulate power plants without accounting for the billions of dollars of costs such regulation might impose: 

First, the Court explains that the toxic air pollution provisions of § 112 of the Clean Air Act differentiate between power plants and other stationary sources.  For sources other than power plants, the Act essentially allows EPA to consider, at most, health and environmental effects.  In contrast, the Act requires EPA to regulate power plants only if “necessary and appropriate.”  This contrast, the Court offers, must mean something.

Second, the Court opines that appropriate regulation generally requires an agency to think about both the benefits of regulation and its cots.  This suggests, that could be read to presumptively require agencies to consider costs in making regulatory decisions. 

Third, the Court seizes on an inconsistency in EPA’s rationale for its decision.  That inconsistency involves three facts. Fact one, the Act required EPA to conduct three studies about regulating power plants, and one of those studies had to examine the cost of pollution abatement technology. (Call this the “cost study,” although the study didn’t solely focus on costs).  Fact two, while EPA potentially could have argued that the cost study was irrelevant to its “necessary and appropriate” decision, the agency explicitly stated that it considered all three studies.  Fact three, EPA simultaneously interpreted § 112 as allowing the agency to decide to regulate without considering costs.  The Court viewed fact two and fact three as irreconcilable.  Because EPA both embraced and disclaimed the cost study, the Court found that the agency engaged in “interpretive gerrymanders under which [EPA] keeps parts of statutory context it likes while throwing away parts it does not.” 

Because the Court relied on three considerations, perhaps the decision only applies where all three considerations are present. In other words, if EPA had not both considered and refused to consider the cost study, the decision might have survived. If so, the decision doesn’t even tell us that the statute required EPA to consider costs in this situation.  Under that reading, the decision would tell us essentially zero about any other statute. 

I don’t claim such arguments will necessarily prevail in this Supreme Court, particularly because Justice Kagan’s dissent forthrightly proclaims that agencies must always account for costs when imposing significant burdens. But the composition of the Court changes.  And the lower courts are filled with judges of all stripes. Because this decision can be cabined, it may one day be so cabined, particularly because (as Lisa suggests) a presumption that costs must be considered rests on an infirm legal foundation.

My second observation is decidedly more pessimistic.

To my eye, the threat of Michigan v. EPA has less to do with doctrine and more to do with a recent alarming tendency of the justices to override agencies when they engage in regulatory efforts that involve large dollar figures. Three recent cases typify this tendency.  In last term’s decision in Utility Air Regulatory Group v. EPA, which I discussed here, the Court found that EPA had unreasonably interpreted a different provision of the Clean Air Act because, in part, the Agency’s interpretation required pollution control that was too expensive.  Had Congress desired such a result, the Court reasoned, Congress would have clearly said so.  In ruling in favor of the Obama administration in King v. Burwell, the Court invoked the “economic and political significance” of health insurance subsidies as a reason to ignore an IRS regulation (citing Utility Air Regulatory Group).  Had Congress wanted the IRS to interpret the Affordable Care Act, Congress would have said so.  And now in Michigan v. EPA, the Court overrules EPA acting in the heartland of its expertise with the Court’s eye fixed on the potential that EPA’s rule will cost power companies billions of dollars.   

These three cases suggest new, not yet precisely defined, rules of administrative law that distinguish between agency decisions that involve significant economic costs and those that do not.  Whatever we think of the deference courts give to agency decisions in general, we should fear distinctions of the sort suggested by Michigan v. EPA, King, and Utility Air Regulatory Group.  Singling out agency rules of economic significance for more exacting judicial scrutiny favors deregulation to regulation.  It benefits regulated entities at the expense of regulatory beneficiaries.  And it preferences economic interests over the environment, human health, worker safety and other aspects of the public interest. 

The lasting legacy of Michigan v. EPA may consist of the mood that the Court continues to telegraph, rather than the rule of law it articulates.  I fear that mood may persist, resulting in more decisions in which the Court intervenes to delay or altogether halt federal agency efforts to address pressing social problems, problems that can be addressed only by imposing costs on powerful economic interests that cause those problems in the first place.