The Separation of Powers Restoration Act (in the Age of Trump), by Adrian Vermeule

by Chris Walker — Thursday, Nov. 10, 2016@chris_j_walker

There are apparently swirling rumors that the Trump people, and Republican senators, may agree to enact the Separation of Powers Restoration Act of 2016. The Act, having passed the House, is now before the Senate Judiciary Committee. It instructs courts, inter alia, to “decide de novo all relevant questions of law, including the interpretation of constitutional and statutory provisions….”

Some thoughts and reactions:

  1. I’m going to go have a nap; wake me when it’s actually signed into law. I believe that rational self-interest with a generous rate of time discounting will quickly take over, in which case, under the new political and institutional configuration: (a) the Trump administration may think twice about whether abolishing deference to its own agencies is a great idea; (b) the Republican senators may realize that the short-run choice is between deference to the Trump administration, over which they will have real influence, and de novo interpretation by a largely Obama-appointed judiciary, over which they have no influence, and which is their ideological opposite.
  2. If the Act is enacted, there is a problem with its text. The Act instructs federal judges to decide legal questions de novo, including the interpretation of statutory provisions, but that is not necessarily inconsistent with the Monahan/Mead justification for Chevron. The judges may determine the law for themselves, de novo, interpreting the relevant statutes, only to decide that those statutes, rightly understood, make deference itself the law. If the law itself is best understood to say that Congress (in a particular organic statute, as interpreted by the judges) delegated to the agency the power to make rules with the force of law, then the judges, deciding the legal question de novo, might decide that deference just is what the statutory law commands. Of course the background purpose and legislative history do show that the point was to abolish Chevron deference — setting us all up for the delicious spectacle of a rapid switch of hats between quondam textualist Justices who want to abolish Chevron, and who will rely on the background purpose and legislative history to do so, and quondam purposivist Justices who do not want to abolish Chevron, and who may well refuse to do so absent a clear textual statement to that effect.
  3. If the Act does pass, and bracketing the interpretive issue above, the Act is of course constitutional. Not even the Baconian defenders of Stuart tyranny, among whom I proudly count myself (are there any others? Eric, you in the office today?) think that Article II requires courts to follow Chevron, in the face of a contrary congressional command.
  4. That said, I will persist in thinking the congressional command a misguided one, for the consequentialist reasons given in the Chevron opinion (expertise and accountability), and in Peter Strauss’ justification (interpretive centralization). But Congress is entitled to enact misguided laws.
  5. One of the main reasons that the Act is misguided is that deference to agencies on questions of law did not begin with Chevron, and will hardly end — at least as a de facto, practical matter — even if Chevron is overturned by statute. I believe, and argue in a book just out, that the law has arrived at deference by a long process of internal legal argumentation; but it is also true, and not inconsistent with that argument, to observe that from an external perspective deference also arises from longstanding institutional and structural pressures on generalist judges. Speaking only of explicit deference in SCOTUS opinions, it appears no later than 1904; speaking of the implicit general practices of Anglo-American courts, deference of one sort or another is assuredly ancient, as witness Coke’s 17th-century complaint that “in a doubtful thing, interpretation goes always for the King.” No generalist Article III judge will refuse to give real weight to an agency’s views under the sort of insanely complex regulatory statutes that roam like giant beasts through the modern legal landscape. After legal realism, it is impossible to disentangle or even distinguish interpretive questions from policy questions. Of course not all judges are realists, but if realism is true, as I think it is, then a sort of institutional pressure arises that is impossible for even non-realist judges to evade. (And, let me say impatiently, semantic distinctions between “persuasive Skidmore deference” and “binding Chevron deference” miss the point.)

The result of the Act, then, will not be no deference, but instead deference sub rosa. One might argue that the sub rosa character of deference will keep it within tolerable bounds; one might also argue that by becoming less visible, deference will become even more routine and expansive. Whatever the truth of that issue, I very much doubt it will disappear altogether, whatever the law on the books may say.

Adrian Vermeule is the Ralph S. Tyler, Jr. Professor of Constitutional Law at Harvard Law School. Follow Professor Vermeule on Twitter here.

Cite As: Author Name, Title, 36 Yale J. on Reg.: Notice & Comment (date), URL.

This entry was tagged .

About Chris Walker

Christopher Walker is a law professor at The Ohio State University Moritz College of Law. Prior to joining the law faculty, Professor Walker clerked for Justice Anthony Kennedy of the U.S. Supreme Court and worked on the Civil Appellate Staff at the U.S. Department of Justice. His publications have appeared in the California Law Review, Michigan Law Review, Stanford Law Review, and University of Pennsylvania Law Review, among others. Outside the law school, he serves as one of forty Public Members of the Administrative Conference of the United States and as Vice-Chair of the American Bar Association’s Section on Administrative Law and Regulatory Practice. He blogs regularly at the Yale Journal on Regulation.

Leave a Reply

Your email address will not be published. Required fields are marked *