The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Affordable Care Act

ACA: The Lay of the Land

Where things stand after oral argument.


The oral argument transcript in California v. Texas (the ACA case) makes for interesting reading. Here's my sense of where things stand—not a prediction of how the Justices will vote, but how I'd evaluate the arguments at this point.

TL;DR: The plaintiffs have standing, but only under their 'bank-shot' theory of inseverability. Because they lose on severability, the Court shouldn't even get to the merits.


Direct standing. The plaintiffs' 'direct' standing theories—the ones citing injuries following directly from the mandate—are pretty thin gruel. The individuals say they're injured by the mandate itself: they don't want to comply and buy health insurance, but they also don't want to shirk their legal duties. The states claim a different kind of direct injury: the more people who feel compelled to buy insurance under the ACA, the more money they'll have to spend for various other reasons.

Neither claim seems right. Imagine that Congress passed the following statute:

§ 1. The Church of Jesus Christ of Latter-day Saints is the established church of the United States, and all citizens shall comply with its tenets.

If anything violates the Establishment Clause, that does. But can anyone go to court over it? Like the ACA, this statute has no enforcement mechanism, so there's no one to sue: the plaintiff's quarrel is with the statute book, not with any particular defendant. As Justice Barrett noted, you can't sue Congress to make them repeal a law; and in any case, courts don't issue "writs of erasure," stripping pages out of the Statutes at Large.

This may seem kind of crazy, that a literal establishment of religion could go without challenge. But a core lesson of standing doctrine is that a law can be unconstitutional—even wildly unconstitutional—and yet no single plaintiff might have standing to challenge it. That's why standing is separate from the merits: because courts have to wait for a certain kind of case to weigh in, even if what they're waiting to say is true. (I have other reservations about modern standing doctrine, but that part seems right to me.)

In this hypo, even if an individual plaintiff objects to the new legal duty, there's no defendant to whom that injury is traceable; there's no one for a court to enjoin, much less to award damages against. And, as Justice Gorsuch noted, you can't get declaratory relief just to answer an abstract question; you need the possibility of some other kind of relief in some other kind of suit. (Maybe the establishment could be challenged in private lawsuits—say, if LDS tenets rendered alcohol contracts among private persons void as against public policy. But you'd still need an actual defendant.)

The Texas SG noted that the United States is a defendant in the ACA case (as has Josh Blackman in a previous post). Maybe our hypothetical objector could sue the U.S. too, assuming an appropriate waiver of sovereign immunity. But this may not help. If the U.S. can sue private citizens in equity, obtaining injunctions to force them to adhere to the new religion (as Justice Gorsuch suggested with respect to 26 U.S.C. § 7402(a)), then sure, an individual defendant could invoke the Establishment Clause in defense. Maybe that individual could also, as plaintiff, race to the courthouse for an anti-suit injunction, or else for some kind of declaratory judgment establishing their rights and liabilities relative to the United States. But the former requires some real threat of enforcement, and the latter a "case of actual controversy." If the United States agrees that the law is unconstitutional, and if no one is going after you, then what are the plaintiffs worried about?

The ACA plaintiffs seem to be assuming that unconstitutional statutes really do impose legal obligations, until a court officially announces that they don't. But that's not how it works. "[A]n act of the legislature, repugnant to the constitution, is void," even before any court says so. A void statute imposes no obligations whatsoever: that's why a court must refuse to apply it when the statute is invoked. If a government officer were to come and throw you in jail for not buying health insurance, then yes, you could challenge that detention in court; and if they were going to throw you in jail, then you might get pre-enforcement relief to stop them. But nothing is actually going to happen to the individual plaintiffs, other than the fact that they think Congress passed a void mandate, which falsely purports to impose on them an unenforceable legal obligation. The injury is just in their heads. The correct response to a law that's both void and incapable of enforcement is to ignore it, not to seek Supreme Court review because you lack the courage of your convictions.

(The same goes for the states. Suppose that the more people who adhere to LDS tenets, including their 10% tithing requirement, the more tax revenue the states will lose in charitable deductions. That still doesn't give the state anyone to sue: the problem isn't that the United States government is doing anything to them, but that their residents mistakenly think that they're bound by a void statute. Imagine that another version of the statute, properly construed, actually imposed no statutory obligation to adopt LDS tenets at all, but that lots of ordinary people nonetheless misread it and thought that it did. Could a state run to court to get a declaratory judgment, to save it from the costs of its citizens' interpretive errors? If most people don't check PACER, would a final judgment from a district court be of any help? Can my Civ Pro students run to court for some declaratory judgments, if they'll be injured by their own misreadings when they take my final exam? Etc.)


Bank-shot standing. A second theory of standing works better. The individual and state plaintiffs are unambiguously injured by the defendants' conduct under other parts of the ACA. Federal officials, right now, are preventing the individuals from buying cheaper kinds of health plans, and they're also imposing a variety of obligations on the states. If the mandate is unconstitutional, and if everything else in the ACA is inseverable, then the rest of the law is inoperative (the bank shot), and these officers' actions may properly be enjoined.

This argument is controversial, but it seems plainly correct to me. Consider two more provisions of our imaginary statute:

§ 2. Any person who drives over 55 mph in the District of Columbia is guilty of a felony and shall be imprisoned for not more than two years.

§ 3. Sections 1 and 2 of this Act are inseverable; if either is unconstitutional, the other is wholly inoperative.

Say that someone in D.C. is pulled over for driving 56 mph. As Justice Alito suggested, and as Josh argued in a prior post, surely the defendant could invoke inseverability in his defense at trial: either driving 56 mph in D.C. is either a felony or not, and we have to answer the severability question to find out. The California SG conceded as much, and that concession strikes me as fatal. There's definitely an injury-in-fact here: the defendant is going to jail (!), and he's contesting the legal status of the very provision applied against him. True, his constitutional concern involves some other provision; but that's a necessary premise of his statutory argument for why the criminal prohibition applied to him lacks force. And if you can raise a defense of inseverability in the criminal proceeding, even though no court has ruled on the religious establishment yet (and you have no problem with it yourself), then assuming you've checked the other necessary boxes, you might qualify for pre-enforcement relief.

Some Justices reacted quite negatively to this idea, because it puts severability, a seemingly back-end concern, at the front of the train. The reason may be a disagreement about the nature of inseverability. Consider the following two versions of Section 3:

§ 3. Sections 1 and 2 of this Act are inseverable; if either is unconstitutional, the other is wholly inoperative.

§ 3*. Sections 1 and 2 of this Act are inseverable; if either is ever held unconstitutional by the final judgment of a court of competent jurisdiction, the other is from that date on, and not until then, inoperative.

Strange as it may seem, many inseverability clauses are written explicitly along the lines of § 3*: they depend, not on another provision being unconstitutional, but on its being held unconstitutional by a court. These kinds of clauses represent time-limited legislation, sunsetting one provision once a court holds another invalid. Under a statute like that, a defendant who's prosecuted under Section 2 has no defense regarding Section 1; on the date of his offense, Section 2 hadn't sunset yet, and it was fully in force (and stays in force for him under the general savings statute). If the criminal defendant can't raise the argument, then it makes sense that someone seeking pre-enforcement review can't either: allowing a suit for the sole purpose of triggering the sunset would be letting the plaintiffs pull themselves up by their own bootstraps.

Applying all this to the ACA is pretty confusing, because (as discussed below) the ACA doesn't even have an inseverability clause. But whether there's some invisible-ink inseverability there—and, if so, of what kind—is a question of statutory construction, not of standing.

This brings us to the other critique of bank-shot standing: that it would open the floodgates to unrelated claims. Right now, if some plaintiff wants to argue that, say, fossil fuel subsidies are unconstitutional, standing makes it harder for a court to hear it. In particular, Steel Co. v. Citizens for Better Environment requires that a court address jurisdiction before the merits. But if the plaintiff is headed to prison for an unrelated arson conviction, and if he wants to claim that 18 U.S.C. § 81 is somehow inseverable from the government's subsidies for fossil fuels ("I just burned down some houses, but they're burning down the planet!"), then his injury-in-fact is already present; he's going to jail, and he has a legal theory for why he shouldn't. Steel Co. doesn't prescribe any order of battle here; a friendly Ninth Circuit panel could theoretically reach the merits, hold fossil fuel subsidies unconstitutional, and then conclude that they're nonetheless severable from the arson prohibition, denying the plaintiff any relief while keeping the merits holding on the books (and, indeed, insulating the decision from cert review).

But if this is really an order-of-battle problem, then (per Acting SG Wall) it has an order-of-battle solution. Current doctrine already holds that frivolous claims don't support federal-question jurisdiction. If a plaintiff's pre-enforcement challenge rests on a frivolous severability argument, that would have to be addressed first. Or, if the severability argument is nonfrivolous, the proper use of judicial discretion would still be to address it first. In cases like these, inseverability arguments are ultimately arguments of statutory construction; under Ashwander, statutory arguments get resolved before the constitutional ones. Resolving the severability argument first means that courts couldn't run off to address constitutional challenges unless the merits of those challenges really do affect the plaintiff's entitlement to relief for actual injuries. So it seems doubtful that allowing bank-shot standing would open the floodgates to speculative constitutional claims; and it seems quite certain, by contrast, that at least some bank-shot claims deserve their day in court.


Severability. If the Court does turn to severability first, it should find the ACA severable and send the plaintiffs home. This issue has been discussed pretty exhaustively by others, so I'll add only a few thoughts here. As Kevin Walsh has argued, the original approach to severability was to recognize that both the statute and the Constitution are law, but that the Constitution is superior law; whenever a statute commands what the Constitution forbids, the court must give effect to the Constitution, which means failing to give inconsistent effect to the statute. ("It is not the terms of the law, but its effect, that is inhibited by the constitution.") But if the statute has other effects, not inconsistent with the Constitution, those the Court must respect.

Modern courts approach severability very differently, of course; and legislatures can always provide their own instructions for deeming portions of their laws inoperative. But the 2010 Congress doesn't seem to have done this. The primary provision cited by the plaintiffs, now codified at 42 U.S.C. § 18091(2)(I), is as follows:

* * * [I]f there were no requirement, many individuals would wait to purchase health insurance until they needed care. By significantly increasing health insurance coverage, the requirement, together with the other provisions of this Act, will minimize this adverse selection and broaden the health insurance risk pool to include healthy individuals, which will lower health insurance premiums. The requirement is essential to creating effective health insurance markets in which improved health insurance products that are guaranteed issue and do not exclude coverage of pre-existing conditions can be sold.

The 2017 Congress didn't repeal this finding, but it didn't have to: as an inoperative provision, it reports the view of the 2010 Congress that compliance with the mandate was very important. The 2017 Congress zeroed out the penalty, which suggests a view that compliance with the mandate isn't very important after all. In any case, we don't need to identify "what Congress really thought" to conclude, as the Chief Justice and Justice Kavanaugh suggested, that this particular statutory finding shouldn't be read as a mandatory inseverability clause.

If the rest of the law is severable, it remains in effect regardless of whether or not the mandate is valid. That being so, the plaintiffs have no good argument that their injuries—the real ones, the bank-shot ones—are being unlawfully inflicted. To determine whether these plaintiffs get any relief, it's unnecessary to decide whether 26 U.S.C. § 5000A now imposes a $0 tax or a bare legal duty to buy insurance—and, in the latter case, whether that duty may constitutionally be imposed. The Court can stop there, and it should.