The IRS and its defenders in Halbig v. Sebelius, one of several cases challenging IRS’s interpretation of the Affordable Care Act, have raised questions about whether Congress really meant to deny tax subsidies to taxpayers if their states didn’t set up insurance exchanges. As we shall see, this was one of several mechanisms to “encourage” states to set up exchanges by threatening their residents with the loss of benefits if they didn’t join the party.
The central issue in Halbig, you’ll recall, is whether insurance plan tax subsidies established under 26 U.S.C. § 36B apply only to plans purchased on an exchange “established by the State,” as the statute says, or whether the subsidies also go to plans purchased on a federally-established exchange. The IRS issued a regulation taking the latter point of view, roughly doubling the number of states in which plans will receive subsidies. After the plaintiffs brought suit, the trial court refused to strike down an IRS regulation that relies on the interpretation, so the case is now before a D.C. Circuit appeals panel.
At oral argument in March, the majority seemed to be leaning in favor of the challengers. Judge Edwards, a Carter appointee, by contrast, seemed firmly against the challengers’ position. The challengers’ attorney, Michael Carvin, argued that the text, structure, and legislative history of the ACA supported the plain meaning, while Stuart Delery, the DOJ lawyer defending the regulation, at one point argued that the court should construe the statute in accordance with the purpose stated in its title: to make care affordable.
Let’s assume for the moment that finding the general purpose of a vast, convoluted statute like the ACA resolves a significant portion of the interpretive questions. What evidence is there that Congress purposely constructed the tax subsidies so they would not go to residents of states that declined to set up exchanges?
Quite a lot, as it turns out. First, let’s revisit how the tax subsidy works. The ACA applies tax penalties to certain individuals who don’t have health insurance or purchase it on an exchange. For individuals who purchase insurance on an exchange, the ACA applies a variable tax subsidy which is supposed to make insurance plans more affordable. According to Section 36B of the Internal Revenue Code, the subsidy goes to qualified health plans that were “enrolled in through an Exchange established by the State under [Section] 1311” of the ACA. Section 1311 (codified at 42 U.S.C. § 18031) sets out procedures for a state to set up its own exchange, whereas a different section (Section 1321, codified at 42 U.S.C. § 18041) sets out procedures for the federal government to set up an exchange.
Generally speaking, it’s clear that the ACA intended states to play an integral role in carrying out the ACA and was not shy about saying so. The ACA explicitly directs that states “shall” set up exchanges, even though this mandate would not be enforceable in light of anticommandeering doctrines established in Printz v. United States and New York v. United States. The ACA also uses carrots and sticks, such as funding cutoffs and startup funds, as financial incentives for states to set up their own exchanges. In other portions of the ACA, the ACA applied punitive measures to states that refused to get with the program: Congress initially tried to secure expansion of state Medicaid programs by conditioning Medicaid funding on each state’s expansion of its Medicaid program. This, the Supreme Court held in NFIB v. Sebelius, was an unconstitutionally coercive “gun to the head.”
Think for a moment about how the Medicaid expansion incentives would have worked. Medicaid funds do not go to state treasuries for general purposes, but are rather used for the state Medicaid program’s services, training, and administration. If the federal government suddenly turned off the funding spigot (which averages 57% of state program funding), the state would be forced to scale back its programs. This effectively increases pressure on the state officials to do whatever is necessary to restore federal money, so as to avoid accusations that they were selling out their poor citizens, public employees, and so forth. Fundamentally, then, financial pressure from the federal government was supposed to produce a political response by the officials of the targeted states, and if all else fails, by the people of the state. That, of course, is coercion. You can almost hear Tony Soprano saying, “Say, that’s a nice Medicaid program you got there. It’d be a shame if anything happened to it.”
As it turns out, the same sort of maneuver also appears in 26 U.S.C. § 36B(f)(3), which sets out a seemingly innocuous reporting requirement. This provision, added in the reconciliation bill that Congress used to dodge the Origination Clause, requires state and federal exchanges to tell each purchaser how much they are paying in premiums and how much (or how little) they are receiving in subsidies. There are several legitimate functions of this provision, such as allowing HHS to collect data about how the program is working.
But it just so happens that the reporting requirement also gives HHS the opportunity to appeal directly to citizens of non-cooperating states. Although more subtle than the unconstitutional Medicaid expansion, (f)(3)’s reporting requirement is both stick and carrot. In states that declined to set up exchanges (and thus blocked the tax subsidies), the federal exchange would send a letter to residents of the state telling them how much their insurance costs ((f)(3)(B)), that the subsidy is zero ((f)(3)(C)), and that, by the way, state officials can trigger the subsidy by opting into the ACA ((f)(3)(E)). Judge Randolph picked up on this issue during oral argument (at 1:18:55), construing the reporting requirement primarily as a stick:
The report [under Section 36B(f)(3)] goes to the Secretary of the Treasury, but it also goes to each individual citizen, and in the states that have federal exchanges those people are going to get reports from the federal government saying that your subsidy, we’re afraid, is zero. And that puts tremendous political pressure it seems to me on the governors and the state legislators in those states who haven’t set [up] exchanges.
Professor Jonathan Adler and Michael Cannon interpret the incentive as a carrot:
[A]pplying these reporting requirements to federal Exchanges enables those Exchanges and the Treasury Secretary to notify individual taxpayers of the tax credits for which they would become eligible and to publicize to state officials the number of taxpayers who would benefit if the state were to establish its own Exchange.
Ironically, the IRS is relying on the reporting requirement to argue that Congress meant tax credits to be available on both federal and state exchanges. But the reporting requirement was added in the reconciliation bill, which was a separate piece of legislation. And allowing tax subsidies on federal exchanges would actually undermine Congress’s intention to punish noncomplying states.
As a side note, this is yet another reminder that broad purposivism doesn’t really make sense for complicated statutes like the ACA. It is Congress, not the President or the courts, that makes choices about means and ends. Yet when judges use a general purpose (like affordability) as a justification to override a specific purpose that is identified clearly in a statute, they are effectively privileging their own general preference for how the statute should work over how Congress specifically said it must work. That approach is not only highly manipulable, but undermines democratic self-government. It’s not the courts’ job to save Congress from itself, even when its legislation is unnecessarily punitive or fails to accomplish its stated goals.