Lisa Heinzerling has posted this draft on SSRN (forthcoming, William and Mary Law Review). Here is the abstract:
With three recent decisions – UARG v. EPA, King v. Burwell, and Michigan v. EPA – the Supreme Court has embraced a new trio of canons of statutory interpretation. When an agency charged with administering a long-existing statute asserts regulatory authority it has not previously used, in a matter having large economic and political significance, its interpretation will be met with skepticism. When an agency charged with administering an ambiguous statutory provision answers a question of large economic and political significance, one central to the statutory regime, and the Court believes the agency is not an expert in the matter, the Court may ignore the agency’s interpretation altogether. And when an agency charged with administering a statute interprets an ambiguous provision to permit the agency not to consider costs before deciding to regulate, the agency will likely lose as having acted unreasonably. In each of these cases, the Court put Congress on notice that it would need to speak clearly if it wanted to give administrative agencies interpretive authority over certain kinds of decisions. In each case, the Court took interpretive power from an administrative agency, power that would normally have been the agency’s due under Chevron, and kept it for itself. And in each case, the Court’s seizure of power aligned with its basic distrust of an active administrative state. I call the new canons the “power canons.”
The power canons are both normative and new. They are normative canons, not descriptive or linguistic canons, insofar as they are not based on a meaningful assessment of what Congress likely intended to accomplish by using particular statutory language. The canons are also new. Although two of them draw upon previous decisions alluding to the significance (in two different senses) of an interpretive question as a factor in statutory interpretation, the recent cases both resuscitate that factor after intervening cases had signaled its demise and add new, distinctive parameters. The third canon, on regulatory costs, is utterly new.
As normative canons, the power canons must be normatively justified, and the justification must spring from something other than, and more than, judges’ own political preferences. Yet the Supreme Court made essentially no effort to lay normative foundations for its new trio of canons. Borrowing from Professor Eskridge’s normative framework for evaluating interpretive canons, I argue here that the power canons undermine rather than uphold important normative values. Their unpredictability and subjectivity upset rule-of-law values. Indeed, the asymmetrical thrust of the power canons – pushing statutory regimes away from responsiveness and dynamism, and toward regulatory passivity – is, more than their dilution of Chevron deference, their core problem. The blunt approach of the power canons also ignores details of statutory history and design and thus their application drives a wedge between legislative objectives and judicial outcomes. They undermine the public values of separation of powers and deliberation by enlarging the judicial power at the expense of the legislative and executive branches and by leaning hard against one side of the ongoing debate over the appropriate scope of regulatory power. The Supreme Court should renounce the power canons.