One of the most anticipated cases of the Supreme Court’s 2014-2015 term is King v. Burwell. In it, the Supreme Court is confronted with what should be a straightforward question of statutory interpretation about the scope of subsidies available under the Affordable Care Act (ACA). Section 1311 of the ACA states that “each state shall, not later than January 1, 2014, establish an American Health Benefit Exchange.” Another part of the law, section 1321, then qualifies that apparently absolute duty by providing that if the state does not “elect” to establish that exchange by January 1, 2014, or if it otherwise fails to meet the federal requirements for an exchange, “the Secretary [of HHS] shall . . . establish and operate such exchange within the state.”
The question of whether a state establishes this exchange determines far more than where individuals can buy their health care coverage. It also determines whether any purchaser of health insurance is entitled to a tax credit against his or her cost of coverage, as that subsidy is limited to taxpayers who are enrolled in a qualifying plan “through an Exchange established by the state” under Section 1311. Internal Revenue Service regulations interpreted the ACA requirement so that its tax subsidies were available to all individuals whether they enrolled in an exchange established by the state or by HHS when the state elected or failed to do so. The plaintiffs’ challenge to the regulation was in essence that the plain language of the ACA precluded the IRS from expanding the scope of the subsidy by this sleight of hand. King would have been an open-and-shut victory for the plaintiffs if the disputed interpretation had been some run-of-the-mill tax provision. But 36 states did not establish these exchanges even though they knew that their citizens could not obtain any federal tax subsidies. They made a business judgment that the hassles of compliance were not worth the benefits.
Unfortunately, high stakes litigation invites fevered argument, which was on ample display in the Supreme Court on Wednesday, March 4, 2015. Matters got off on the wrong foot when Justice Ruth Bader Ginsburg challenged the four petitioners’ standing to bring the suit at all. Technicalities aside, it would be a real travesty of justice if the Court postponed its decision until some other party with a sufficient financial stake knocked on its door. Millions of individuals, and thousands of public officials, need to know their duties sooner rather than later. Indeed, one real tragedy of this entire episode is that our twisted law of standing has allowed no one to challenge the IRS regulation the day it was promulgated. The long delay in litigation added no useful information about how to read the various provisions of the ACA, but it did force everyone to make a major business decision when major legal issues were left unresolved.
Once this initial diversion was pushed to one side, the battle switched to three distinct clusters of issues. The first of these addresses the proper reading of sections 1311 and 1321. The second asks whether the ACA should be found constitutional if it were read to allow subsidies only on state exchanges. The third asks about the social consequences of the IRS regulation being struck down.
To put the textual question in perspective, we should ask what would happen if the IRS regulations would limit the said subsidies only to policies purchased through state exchanges. Without question, a regulation that tracks the statute is consistent with it, even if it adds the gloss that the “state” is not the federal government. How then is it possible to turn the provisions upside down so that federal exchanges are treated as state exchanges?
In an effort to avoid that conclusion, Justices Breyer, Kagan, and Sotomayor stressed the use of the word “such” in section 1321, which they claimed meant that when the state chose or failed to create the exchange, the federally created exchange was the state exchange. But that weird interpretation makes hash out of the rest of the statute, where the natural meaning of “such” is that the federal government will establish its own exchange that is the functional equivalent to the state exchange, except for the key fact that it will not be eligible for the subsidy on the state exchange. Elsewhere, the ACA provides that territorial exchanges “shall be treated as a state.” Those words don’t appear in the provision governing the creation of the federal exchanges.
Equally implausible is the argument of Justice Elena Kagan that it would be odd to put such a critical provision in the taxation section of the ACA when it should have been featured more prominently elsewhere. But there is no canon of statutory construction that says the meaning of a general provision is determined by its location in the statute, especially when it makes perfectly good sense to put provisions on tax benefits in a section that is devoted to, well, tax credits. Indeed, King is an easy case of statutory construction because from these two short sentences, the ACA contains no other moving parts that have to be integrated with this coverage provision.
In contrast, other statutes have broad definitional provisions that do need adjustment. To this effect, Solicitor General Donald Verrilli compared King with the case FDA v Brown & Williamson, which was about whether tobacco products count as “drugs” that can be regulated by the federal government. But the two cases are at opposite poles. The Food, Drug and Cosmetic Act defines the term drug to include “articles (other than food) intended to affect the structure or any function of the body.” By this definition, taken literally, it is hard to think of any product that is ingested that is not a drug, including not only tobacco, but also mouthwash. But that broad reading is at war with the ordinary meaning of the term drug, and it doesn’t make sense in the context of the rest of the statute, unless one thinks that it is imperative to conduct clinical trials to see which diseases tobacco is intended to cure.
Long before the tobacco case arose, this definitional section had always been read to apply only to those products for which its manufacturer claimed that it had therapeutic effects. When in the 1950s tobacco companies did make such claims, they were rightly slapped down by the FDA. But the definition of drug to cover tobacco when no such claims are made represents an inexcusable deviation from settled understandings.
There is no similar difficulty with the ACA. The government may protest that its subsidies are only available through the state exchanges, but the result is not unintelligible. Indeed the provision makes perfectly good sense if the plan, as often stated by MIT economist Jonathan Gruber, was intended to give states a strong incentive to sign up with the program, even if only 14 states rose to take the bait.
The inescapable truth is that the word “state” is defined in opposition to federal, and no amount of linguistic high-stepping can change its meaning. The four liberal judges were determined to inject ambiguity into this provision in order to take advantage of the so-called Chevron doctrine, dating from 1984, which holds that in those cases where the text is not clear, the courts must give deference to the agency’s interpretation of the statute. In oral argument, Justice Scalia, who is one of Chevron’s champions, held that it did not apply because the statute was clear on its face no matter how much the government regretted the outcome. But unfortunately, the question of what counts as a clear statute is itself subject to serious ambiguities, which the liberal justices hoped to exploit in this case.
The strongest answer to that approach is to protest against the application of the Chevron doctrine altogether. The obvious technical difficulty with the doctrine is that its author, Justice John Paul Stevens, never bothered to cite Section 706 of the 1946 Administrative Procedure Act that treats all questions of statutory interpretation as questions of law to be decided by the courts. That result is surely desirable on matters of this importance. During oral argument, Chief Justice Roberts asked whether a different administration could switch interpretations on the statute to meet its own view. Solicitor General Verrilli answered, incorrectly, that such a switch is possible only if the new government could make out “a very strong case” in view of the “disruptive consequences” that such a shift would have. But in fact, Chevron itself involved just that kind of switch between the Carter and Reagan administrations, and no one required the substantial showing of cause that Verrilli mentioned.
Unfortunately, such disruptive flip-flops are all too possible under Chevron. That is the strongest reason why the question should be treated as a question of law, which makes it impossible for one administration to reverse the decision of its predecessor. In any event, it is hard to think that an agency interpretation that returns to the plain meaning of the statute could ever be attacked because of its inconsistency with the strained reading that the Obama Administration has put on the clause. Verrilli’s response to the Chief Justice is a nonstarter.
In one sense the most novel argument of the day was an ill-thought-out suggestion by Justice Anthony Kennedy that the ACA might be unconstitutional if it were read to deny subsidies to health care policies purchased on the federal exchanges. Justice Kennedy never bothered to state whether his suggestion would require invalidation of the entire statute, or the creation of a massive subsidy that Congress itself had never authorized. There is, fortunately, no need to choose between these two unappetizing alternatives. Kennedy tossed off an argument that no one ever raised throughout the litigation: the denial of the federal subsidies would coerce individual states to set up exchanges in order to benefit their citizens.
The supposed precedent for this argument was the earlier decision in NFIB v. Sebelius that struck down part of Title II of the ACA that required states to forfeit all their current Medicaid benefits if they failed to sign up for the expanded set of Medicaid benefits under the ACA. Chief Justice Roberts held that this provision was coercive. No state could possibly withstand that financial hit. But in this instance, many states refused to establish the exchanges precisely because they were not threatened with the loss of any collateral benefits.
Shortly after the oral argument, Oklahoma Attorney General Scott Pruitt protested Justice Kennedy’s argument, noting that Oklahoma knew what the law required and was prepared to live with its consequences, given the heavy costs that it would have to incur to establish a state exchange. Indeed, the Kennedy provision leads to the bizarre conclusion that the statute would be unconstitutional if it omitted the possibility of purchasing any coverage under the federal exchanges, and his argument would invalidate dozens of statutes under which states are entitled to get benefits if they first undertake some particular action. The section is constitutional as written.
The hard question about King is not found in these far-fetched efforts to defeat its plain meaning. It rests instead on the massive disruption that this interpretation could impose on those individuals who, in good faith, purchased insurance on the federal exchange confident that they could gain the subsidy. The government’s defenders stress this point constantly even though they said little about the massive disruptions when millions of people lost their coverages in the original ACA rollout in the fall of 2013. The sad point here is that this latest transitional fiasco was largely avoidable. The Internal Revenue Service knew full well of the risk that it took and yet decided to plunge forward in its willful course of action. It would be distressing if the government’s conscious administrative overreach were rewarded, and it would be equally distressing if millions of innocent people were left in the lurch.
The hard question is whether it is the job of a court to ratify the sins of a government agency, given the dreadful precedent it would create for all future cases. The better choice, on balance, seems for the Court to strike down the IRS regulation and for Congress to work out some fix. That fix should not include expanding the coverage to federal exchanges, which would allow the Obama administration to work an illicit extension of the initial program. A far better suggestion is to make block grants to the states, which could fund subsidies to pick up the slack when the IRS regulation is struck down. The Republicans might well pass such legislation quickly and dare President Obama to veto it—which he might do to legitimate his own misconceived legislation.