[N.B.: the entire post below is written by Adrian Vermeule, both the part in quotes, and the part that comes after it.]
Adrian Vermeule writes in:
Is the following an admissible legal argument? If not, why?
“Under Chevron, let us assume, the government wins so long as the agency offers a reasonable interpretation of statutory meaning, even if it is not clearly correct. The challengers have to show that the agency’s interpretation is clearly wrong, as a matter of the statute’s ordinary meaning.
So far nine federal judges have voted on the merits of the statutory challenge to subsidies on federal exchanges. (The nine comprise the six appellate judges who voted on the merits in King and Halbig; the two district judges in those cases; and one district judge in Oklahoma). To date, six votes have been cast in favor of the government’s position (some on the ground that the agency’s position is reasonable, some on the ground that it is clearly correct). Three votes have been cast against the agency’s view.
In light of these votes, to say that the statute has an ordinary meaning contrary to the agency’s interpretation verges on self-refutation. It implies that the judges in the majority of six can’t read English. It is logically possible that the sample of judges is severely biased in the government’s favor, but it is not actually true. The challengers have had broad latitude to choose their playing fields, and have been unable even to muster a majority of judicial votes, let alone the supermajority that would be necessary to suggest that the statute’s ordinary meaning clearly supports their case.”
(1) An admissible legal argument need not be conclusive, of course. The argument would have to be weighed against other admissible arguments.
(2) The argument, if admissible, would always counsel deference to agencies in Chevron cases where there is a circuit split, other arguments being equal. (Thanks to Abbe Gluck for this observation). I’m fine with that. Is it a problem?
(3) Another implication is that there should be no Chevron cases in which the agency loses by a 5-4 vote at the Supreme Court (and one may make appropriate modifications for other courts). In this setting, the legal rule itself specifies which party should win if reasonable disagreement is present: the agency should win. Accordingly, if a straw vote among the Justices shows a 5-4 split, then all the Justices should update their views; they should realize that there is reasonable disagreement in the case. If so, the agency should win 9-0. Under any other approach, judges in effect throw away valuable information — the information contained in their colleagues’ votes.
An earlier version of this post miscounted the votes, stating them as 5-4 in the government’s favor, instead of 6-3. Jonathan Adler, against interest, graciously corrected the record.