Michigan v. EPA: Competing conceptions of deference due to administrative agencies

Today the Supreme Court decided Michigan v. Environmental Protection Agency, with a 5-4 conservative majority finding that the EPA improperly excluded cost considerations from its decision to regulate hazardous air pollutants emitted from power plants.  Though it may be understandably overlooked in the aftermath of King v. Burwell (the Affordable Care Act case) and Obergefell v. Hodges (the gay marriage case), Michigan usefully exposes a divide on the Court not over environmental policy, but over the future of the administrative state. 

It’s impossible to understand the case without knowing some background about Chevron deference, a doctrine dating back to another case involving EPA Clean Air Act rules: Chevron v. NRDC (1984).  In that case, the Court laid down a now-talismanic two-part test for reviewing agency interpretations of statutes: first, judges must determine if the agency’s interpretation of the law is consistent with the clear meaning of the statute’s text; and if there is no clear conflict, then judges should uphold the agency’s interpretation of any ambiguous statutory text as long as it is reasonable. 

For outsiders looking in, Chevron’s two steps have often seemed to say: an agency can take just about any action it wants as long as it doesn’t run smack into statutory language explicitly precluding that action.  That’s surely an oversimplification, but how much? The answer, very much in the weeds of administrative law, is of great importance in determining who decides what the law means in practice.

The two sides in Michigan v. EPA staked out sharply conflicting positions on how demanding Chevron’s first-step requirement of textual obedience should be.  Justice Scalia, writing for the conservative majority, wrote that “Chevron allows agencies to choose among competing reasonable interpretations of a statute; it does not license interpretive gerrymanders under which an agency keeps parts of statutory context it likes while throwing away parts it does not.”  In Scalia’s view, the agency selectively deployed statutory text to justify the action it wanted to take such that the text became just a means to an end, and that is impermissible. 

Under Scalia’s view, the first thing an agency must do when approaching a statute is faithfully interpret its language, which in this case required the EPA to include considerations of cost in its original decision about whether regulation of power plants under § 112 of the Clean Air Act was “appropriate.”  (If that sounds like a fairly narrow and technical question, it is: headlines declaring that this decision represents a decisive smackdown of Obama’s environmental program are engaging in hit-seeking hyperbole, plain and simple.) 

Intriguingly, Justice Thomas, in a concurrence, went even further than the majority opinion.  He signaled that he would prefer not just to make Chevron’s statutory fidelity test more demanding, but to abandon the presumption of deference to agencies altogether.  He worries that decades of experience under Chevron have disrupted the constitutional order by giving agencies the impression that it is their job to legislate (under a thin pretense of “interpretation”), with judges improperly acquiescing to this transfer of power.  There’s nothing to indicate that other justices fully share Thomas’s misgivings (as no other justice joined his concurrence), but it will be interesting to see whether his concurrence foreshadows an open revolt against Chevron that some people (in Congress, for instance) have occasionally pined for.

Justice Kagan’s dissent takes an entirely different approach to the role of statutory interpretation in the totality of the regulatory process—which she accuses the majority of missing because of their “blinkered way” of considering interpretation as occurring in a single moment of a law’s application.  The dissenters see the EPA as making entirely reasonable decisions about where in their process cost considerations should factor in.  Because judges (under Chevron) should “interfere only if the Agency’s way of ordering its regulatory process is unreasonable—i.e., something Congress would never have allowed,” their role in this case should simply end with a recognition of the agency’s reasonableness.

Kagan paints a portrait of regulatory practice that is incremental, sensitive to subtle facts on the ground (e.g., “plants designed to burn low-rank virgin coal” need to be treated differently than “plants designed to burn high-rank coal”), and generally responsive to legitimate commercial interests in minimizing regulatory costs.  Like a manager in a private firm, the EPA’s planning is not a static and front-loaded exercise, but a dynamic process, and the EPA is entitled to view that process as a unified whole—which in turn deemphasizes what the majority sees as the singular moment of interpretation at the beginning.  Given the complexity and duration of its regulatory tasks, the EPA is better suited than courts to figuring out how to handle its statutory responsibilities, and, per Chevron, “Congress has entrusted such matters to [agencies], not to [judges].”

“Entrusted” is a nice word to highlight the differences between the two sides. The Court’s conservatives have become more and more uneasy about trusting regulatory agencies to do the best they can (by their own lights) to make their statutes function. The Court’s liberals look increasingly comfortable giving a leading role to agency-led policymaking, which they think of as highly professional and ultimately responsive to Congress when the legislature stirs itself to speak clearly.  As the liberals see it, agencies have become the front-line legal actors of the twenty-first century—the common law courts of our time, as Cass Sunstein put it—and it couldn’t really be any other way, given the huge expertise needed to thoughtfully administer complex contemporary regulation.

For those whose instincts and experience lead them to think that government ends up muddling through fairly well, Kagan’s opinion in this case (and Justice Breyer’s in many other similar ones) will be convincing; for those who suspect that technocratic bureaucrats are oblivious to important business concerns, their professions of sensitivity notwithstanding, the majority’s insistence on a more up-front and transparent declaration of the agency’s interpretation will seem sensible.

The stakes of this emerging debate about the future of Chevron deference are huge.  It is hard to overstate just how central debates over Chevron’s proper application have been to recent legal developments governing America’s regulatory state—although at the same time very hard to articulate exactly what the practical effect of Chevron is once one moves past official doctrine and to the question of how it is actually applied.  As Connor Raso and William Eskridge have recently argued, Chevron is probably better thought of as akin to a canon of statutory construction—one heuristic among many available to judges attempting to determine what good interpretive practice looks like—rather than as a straightforward rule.  But how the case and the body of law around it guide administrative law is nonetheless of great consequence.

In spite of some of today’s headlines, figuring out the practical import of this case for air pollution regulation is much harder.  With the majority’s opinion in place, there is no guarantee that the rule at issue will ultimately be vacated; as Brad Plumer notes, it’s entirely possible that the lower courts will ask the EPA to engage in some retrospective cost considerations and then leave the rule in place without any changes at all.  Although the connection is too complex to explain here, if the rule were ultimately thrown out, that could be very good news for the Obama administration’s signature regulation of power plants’ greenhouse gas emissions, the Clean Power Plan—but that’s getting quite speculative.

There is one line in the majority’s opinion that could be big news in the world of regulation: in the course of discussing EPA’s interpretation of the word “appropriate,” Justice Scalia says: “No regulation is ‘appropriate’ if it does significantly more harm than good.”  There are some big fights about the what kind of cost-benefit analyses agencies are required to undertake in their rulemakings coming down the pike, and it sounds like the Court’s conservative wing is gearing up to say that an agency may not proceed with rulemaking if it has not clearly established the net social benefits of a rule.  By involving courts more deeply in the practice of cost-benefit analysis, that would be a big change.  But for now, Justice Scalia’s pronouncement is just a dictum, and the case’s holding itself has very little to say about the importance of formal cost-benefit analyses or their results.  If we are to get any serious judicial battles over cost-benefit analysis, they will have to wait for future terms.  Who knows—stranger things have happened—we could even get legislative battles over the same before then.