Chevron or Dole in King v Burwell?

by Michael Dorf

In the aftermath of the King v. Burwell oral argument, what I have previously called "SCOTUS kremlinology" now commences: Lawyers and law professors will dissect every question asked by the various Justices and even matters analogous to the seating arrangements--such as the fact that CJ Roberts was mostly quiet during the argument, with one critical exception, to which I'll return below. I like to play this game too but I'm going to resist in order to suggest that there is a critical argument towards which Justice Kennedy's questions gestured, but that, for institutional reasons, SG Verrilli was not well positioned to make quite as forcefully as it might have been made.

The plaintiffs and the government each argued that they respectively should win the case under both the plain meaning of the statutory text and its meaning understood in context. I think it's nonetheless fair to say that the government's argument relied more on context, whereas the plaintiffs' argument relied more on (somewhat isolated) text. Indeed, even Justice Scalia, who was quite obviously sympathetic to the plaintiffs' position, didn't seem to buy attorney Michael Carvin's argument that the ACA works perfectly well--indeed better!--without subsidies on federally run exchanges than with them; Justice Scalia instead expressed the view that Congress messes stuff up all the time but that's no reason to disregard plan statutory language.

Meanwhile, SG Verrilli expressed the view that, read in light of overall context, the statute is clear that it authorizes subsidies on federal exchanges. That's an argument for affirming based on a different rationale from the one offered by the Fourth Circuit in the proceedings below. That court found that as used in 26 § 36B(c)(2)(A)(i), the crucial term "an Exchange established by the State under section 1311" was unclear, but that pursuant to step 2 of Chevron, the IRS construction of the term to include federal exchanges was reasonable. By contrast, SG Verrilli argued that the government should win at Chevron step 1.

And it's not surprising that he did so. In just about his only intervention on issues other than standing, CJ Roberts asked the SG whether a victory under Chevron step 2 would mean that a future administration could reach a contrary conclusion, finding that subsidies are not available on federal exchanges. SG Verrilli put up some resistance, saying that--in light of the disastrous consequences--such a construction might be unreasonable, but given that avoidance of those same disastrous consequences is part of what informs his conclusion that the government should win at Chevron step 1, if the Court were to reach Chevron step 2 (say, in reviewing an IRS change of heart under a future Republican president), that might not be enough to deem this approach unreasonable.

It thus might appear that in order to win the case on terms that bode well for the long-term survival of the ACA, the SG had to persuade the Court that § 36B(c)(2)(A)(i) is not merely ambiguous but that, when read in overall context, it clearly means what the government says it means. I think he has a decent argument for that conclusion, but it's hardly a slam-dunk.

Nonetheless, the government can win big even if a majority of Justices find that the language is ambiguous. The key would be to win not on the basis of Chevron deference but on the basis of the South Dakota v. Dole plain statement rule. For if the Court holds that, in light of Dole, subsidies are available on federal exchanges, then no future Administration could undo that result (absent repeal of the ACA by a sympatico Congress).

South Dakota v. Dole sets forth conditions on conditional spending by Congress, one of which
states that "if Congress desires to condition the States' receipt of federal funds, it must do so unambiguously." (Internal quotations and citations omitted.) By its terms, that test doesn't apply to  § 36B(c)(2)(A)(i), because it is a provision governing the tax liability of individuals rather than governing the states' receipt of federal funds. However, the logic of Dole and the anti-commandeering case, New York v. United States, very strongly point in the direction of applying it in the present context.

Indeed, at one point Justice Alito asked SG Verrilli directly whether reading § 36B(c)(2)(A)(i) as the plaintiffs proposed would render the law unconstitutionally coercive. Justice Kennedy's earlier line of questions to Mr. Carvin indicated that he thought the answer was yes. However, because SG Verrilli was aware of his obligation to defend the statute if attacked on that ground, what he said was it would be a "novel" question. He then started to say something like if the Court thinks that's a difficult constitutional question, then the Court should invoke the canon of constitutional avoidance and reject the plaintiffs' reading, but Justice Kennedy interrupted him and made the point for him.

And that seems right to me. NY v. US says that Congress can't simply direct a state to pass a law (such as the one needed to set up a state exchange) but that Congress can do so if it conditions the state's receipt of funds on passing a law, so long as it satisfies the Dole rules regarding conditional spending. The "novel" question is whether Congress can evade NY/Dole's requirements by threatening to withhold money (here in the form of tax subsidies) from the citizens of a state rather than from the state itself. The whole logic of NY strongly suggests a negative answer, and that's the way that Justice Kennedy appeared to be leaning.

Two final points are worth noting. First, in the past I (and many others) have criticized SG Verrilli's performance in high-stakes cases, including the last ACA case. Here he did a great job. His inability to seize the lifeline offered by Justice Kennedy with both hands was occasioned by the institutional constraint I've noted, not by any lack of competence--and even then, he did embrace the suggested constitutional avoidance argument.

Second, Justice Alito tried to head off the conditional spending point by noting that there wouldn't be a severe notice problem: Congress would not be pulling the rug out from the states because they could simply set up their own exchanges after a ruling by the Supreme Court. But this argument is doctrinally non-responsive. Under Dole, if a condition is not expressed clearly, it doesn't exist; it doesn't get converted to a phase-in. Moreover, the point of requiring a clear statement in a conditional spending measure is partly to give fair notice to the states, but that's not its only point. The clear statement rule for conditional spending also seeks to leverage the political safeguards of federalism: If Congress really wants to put financial pressure on the states to do its bidding, then it must do so clearly and in the open, so that the states can attempt to use their political power to block it in Congress before it goes into effect; not just by refusing to take the money.