There are over 400,000 words in the Affordable Care Act.  The challengers in King v. Burwell rely upon a single one of those words—a simple preposition (“by”) buried in a provision (26 U.S.C. § 36B) setting forth the formula for individuals' monthly tax credits—as the basis for an interpretation of the Act that would unravel Congress’s efforts to guarantee affordable health care for all Americans.

According to the challengers, Congress’s use of the word “by” in the phrase “an Exchange established by the State” (rather than, for example, referring to an Exchange established “within” or “for” the State) has a world-changing impact:  On their reading, when a State chooses to allow the federal government to set up a health-insurance Exchange for its residents—an option the Act plainly allows, and one that almost three dozen states have adopted—that choice would have catastrophic consequences, namely, the denial of tax credits for all of the State’s residents who wish to purchase insurance on that Exchange . . . which would in turn lead to the virtual destruction of the insurance market in that State, thereby making the State’s residents much worse off than if Congress had not enacted the ACA at all.  See, e.g., NFIB v. Sebelius, 132 S. Ct. at 2674 (Scalia, Kennedy, Thomas, and Alito, JJ., dissenting) (“[The Act’s] system of incentives collapses if the federal subsidies are invalidated.… With fewer buyers and even fewer sellers, the exchanges would not operate as Congress intended and may not operate at all.”).

For good reason, the challengers make little effort to demonstrate that any members of Congress, let alone majorities of both houses and the President, actually intended to put the States to such a terrible choice, with such ruinous consequences if a State chooses one of the options Congress has offered.  (Indeed, such an argument would be belied by the fact that no legislators mentioned, or were aware, that they had done so--and that no one else realized it either, until an attorney stumbled upon the idea nine months after the law's enactment.)  

Even so, the challengers say, Congress’s actual intent “is legally irrelevant” (page 14 of their reply brief).  It is enough, they contend, that this sort of bullet-to-the-head of the States would facilitate one of Congress’s subsidiary goals—namely, to induce the States to create Exchanges—even if the cost of not fully realizing that goal would be not only to render unachievable the principal objective of the Act (guaranteeing affordable health care for all Americans), but also to destroy the existing health-insurance markets, such that health care would be much less affordable than it was before Congress enacted the law.

A great deal has already been written—in the briefs, online and elsewhere—about the substantive merits of the challengers’ account of the Act, and, in particular, about the proper interpretation of the subsidy-calculation phrase in section 36B on which the challengers rely: “an Exchange established by the State under section 1311 of the Patient Protection and Affordable Care Act.”  To oversimplify a bit, there are three basic positions regarding the interpretation of that phrase:

-- The government argues, and I agree, that the phrase, read in the context of the Act's text as a whole, plainly does not disqualify individuals who purchase a plan through a federal Exchange from being eligible for the tax credits that make such a purchase possible, because when a State declines to create an Exchange under section 1311 of the Act, and HHS steps in to set up “such Exchange” under section 1321, that federal Exchange qualifies as an “Exchange established by the State under section 1311” for purposes of the Act.

-- Others, such as the majority of the Fourth Circuit panel in King, conclude that the phrase in section 36B is ambiguous on that question—in which case the proper outcome for the Court (under Chevron) would be to defer to the IRS’s reasonable interpretation that such credits are available in States where the federal government has established the Exchange.

-- The challengers stake out the third position—that the “plain meaning” of the contested phrase in section 36B precludes such tax credits in States where the federal government operates the Exchange.

For the purposes of this post, I will assume the challengers are correct, i.e., that the “plain meaning” of the words of section 36B isolation, would preclude tax credits in States where HHS has established the Exchange.  As noted above, I think the opposite is true—that the government’s reading is compelled.  But the purpose of this post is to examine what the Court should do if it were to agree with the challengers on the “plain” meaning of that particular statutory language.  In particular, and as explained more fully below, such a plain meaning would not resolve the case in the challengers' favor, not only because it would establish a fundamental absurdity at the heart of the statute, but also because of a federalism canon of statutory construction that the Court invoked and applied just last Term, but that the challengers entirely ignore--a canon that is the focus of an important amicus brief filed by Jim Feldman on behalf of Professors Tom Merrill, Gillian Metzger, Abbe Gross and Nick Bagley.

The Absurdity Canon

“[T]he only scenario in which a court can disregard plain statutory text,” say the challengers at page 14 of their reply brief, is where the plain reading “is objectively absurd.”

As I explain below, the challengers are mistaken to assume that absurdity is the only ground the Court has recognized as a basis for deviating from a text’s plain meaning.  For what it’s worth, however, the challengers should lose even under that test, because their reading is “objectively absurd.”  In particular, if the challengers were correct that tax credits are unavailable for insurance purchased on a federally established Exchange, Congress’s directive to the Secretary of HHS to establish and operate such federal Exchanges in States that have failed to do so would make no sense at all, as the Solicitor General explains at pages 24 and 38-39 of his brief.

Subsection 1321(c)(1) of the Act provides that if a State elects not to set up an Exchange, or if a State does try to set up an Exchange but misses the deadline, or fails to satisfy all the relevant requirements for an Exchange, the HHS Secretary “shall . . . establish and operate such Exchange within the State and the Secretary shall take such actions as are necessary to implement such other requirements.”  The challengers concede (opening brief at 22) that this provision is designed to require HHS to establish the “same Exchange” that would exist if the State established the Exchange for itself, and that "[t]he HHS Exchange should operate just like the Exchange the state would otherwise have established.”   

If tax credits were unavailable with respect to insurance policies purchased on an HHS Exchange, however, that Exchange would not operate anything like its State-run counterpart.  Without the tax credits, consumers could not afford to buy insurance on the Exchange.  And the HHS-facilitated Exchanges would then collapse as insurers dropped out.  It would be absurd, indeed, for Congress to have insisted that HHS to set up such dysfunctional Exchanges, without the tax credits that are crucial to their operation.  As the SG puts it (p.24), “[a]n Exchange without credits would be a rump Exchange bearing little resemblance to its state-run counterpart—if it could operate at all.”  This wouldn’t merely be Hamlet without the Prince; it would be Hamlet without the Danish monarchy . . . more like Rosencrantz and Guildenstern Are Dead.

Think of it this way:  In their opening brief, the challengers hypothesize what they describe as an analogous statute (pp. 22-23) that instructs States to build airports with particular specifications—airports designed to facilitate affordable air travel for all, let’s say—and then includes a provision explaining that if a State fails to do so, the U.S. Secretary of Transportation must construct “such airports.”  (The purported point of the hypo is to support the challengers’ argument that no one would ever refer to the latter as “state-constructed airports.”)  Under the challengers’ reading of the Act, however, the DOT airports would have empty hangers, and be useless, because no one could afford to fly.

In their reply brief, the challengers respond to this argument by insisting (p.19) that, even on their view, there would be good reason for HHS to establish Exchanges, because those Exchanges would offer an “organized and transparent marketplace” for consumers to “shop and compare health insurance options,” just like Orbitz, or Expedia.  Needless to say, however, Orbitz and Expedia are not airports, nor are they “just like” airports.  And even with respect to their alleged value as “marketplaces,” the HHS Exchanges would be worthless on the challengers’ reading, because there would be no affordable “health insurance options” for residents of the State to “shop and compare.”  That truly would be an absurd result—which is reason enough to reject the challengers’ reading of the Act.  [UPDATE, per Nick Bagley:  "If Congress had meant to punish uncooperative states, it didn’t have to saddle them with dysfunctional exchanges.  It could have left them with no exchanges at all.  Congress created a fallback because it wanted to enable everyone — even people in states that objected to health care reform — to secure affordable insurance."]

The Gregory/Bond Federalism Canon

In addition, however, the challengers are wrong to assume that “absurdity” is the only grounds for the Court to depart from plain meaning.  For example, as then-Justice Rehnquist wrote in Griffin v. Oceanic Contractors, Inc., 458 U.S. at 571, “in rare cases the literal application of a statute will produce a result demonstrably at odds with the intentions of its drafters, and those intentions must be controlling.”  Thus, as Justice Rehnquist explained, the Court has “reserved some scope for adopting a restricted rather than a literal or usual meaning of [a provision's] words where acceptance of that meaning ... would thwart the obvious purpose of the statute” (internal citations omitted).  In this case, the challengers’ reading of section 36B surely would “thwart the obvious purpose of the [ACA],” which would be another sufficient ground to reject it. 

In this post, however, I want to focus on yet another basis for rejecting the challengers’ alleged “plain meaning” interpretation—the federalism canon of statutory construction discussed in the Merrill, et al. amicus brief.

Just nine months ago, the Chief Justice, on behalf of six Justices, invoked “the well-established principle” of Gregory v. Ashcroft that “‘it is incumbent upon the federal courts to be certain of Congress' intent before finding that federal law overrides' the ‘usual constitutional balance of federal and state powers.’”  Bond v. United States, slip op. at 12 (quoting Gregory, 501 U.S. at 460).  Chief Justice Roberts explained that this heavy burden—requiring the court’s certitude of Congress’s intent to override the federal/state balance—“‘assures that the legislature has in fact faced, and intended to bring into issue, the critical matters involved in the judicial decision.’”  Id. (quoting United States v. Bass, 404 U.S. at 349) (emphasis added).

As the Merrill amicus brief explains, the Court has applied the Gregory/Bond “certainty of congressional intent to before unbalancing” canon in various different contexts, implicating various different relationships between the federal and state governments—including, most importantly for present purposes, in construing “cooperative federalism” statutes in which Congress assigns the States a particular role in addressing a problem, with a federal fallback if the States do not participate.  See New York v. United States, 505 U.S. at 169-170. 

Here, I’ll focus primarily on the most recent case in which the Court invoked, and relied upon, the Gregory federalism canon—Bond itself.

A federal statute makes it unlawful to possess and use “chemical weapons.”  The statute specifically defines “chemical weapon” in relevant part as “[a] toxic chemical and its precursors,” except where the chemical is intended for “[a]ny peaceful purpose related to an industrial, agricultural, research, medical, or pharmaceutical activity or other activity.”  The statute then specifically defines “toxic chemical,” in turn, as “any chemical which through its chemical action on life processes can cause death, temporary incapacitation or permanent harm to humans or animals . . . regardless of their origin or of their method of production, and regardless of whether they are produced in facilities, in munitions or elsewhere.”

The defendant in Bond deliberately exposed a former friend to two chemicals that are toxic to humans and, in high enough doses, potentially lethal.   Thus, as the three dissenting Justices explained, applying the statutory definition to the defendant’s conduct was “hardly complicated”:

Bond possessed and used “chemical[s] which through [their] chemical action on life processes can cause death, temporary incapacitation or permanent harm.”  Thus, she possessed “toxic chemicals.”  And, because they were not possessed or used only for a “purpose not prohibited,” they were “chemical weapons.”  Ergo, Bond violated the Act.  End of statutory analysis.

According to the dissent, it was “utterly clear” that the statutory definition encompassed the defendant’s acts.  More importantly, the majority did not disagree. 

Nevertheless, citing the Gregory canon, Chief Justice Roberts and five other Justices refused to apply the statute to the defendant’s conduct because it would significantly upset the federal/state balance for Congress to make virtually all local poisonings unlawful; and, although the statutory definition certainly covered the case, the Court could not discern any certainty that Congress intended such a dramatic recalibration of the state/federal relationship.

Although the Court did not suggest, let alone conclude, that the statutory definition of “chemical weapon” was ambiguous (because it was not), it did find that there was “ambiguity” as to whether the statute covered such quintessentially local conduct—ambiguity that “derive[d] from the improbably broad reach of the key statutory definition given the term . . . being defined,” as well as the “deeply serious consequences of adopting such a boundless reading; and the lack of any apparent need to do so in light of the context from which the statute arose.”  Slip op. at 14.  The Court then invoked the Gregory canon, explaining that it was appropriate to “insist on a clear indication that Congress meant to reach purely local crimes, before interpreting the statute's expansive language in a way that intrudes on the police power of the States,” ibid.  The Court concluded that although the key term was “defined extremely broadly,” that “general definition does not constitute a clear statement that Congress meant the statute to reach local criminal conduct.”  Ibid.

Bond is thus the latest example of a case in which the Court has invoked the Gregory federalism canon, or its equivalent, in order to avoid the implications of a single provision’s alleged “plain meaning” where there was no certainty that Congress intended the statute, read as a whole, to have the significant impact on the States that such a reading would compel.  See also, e.g., New York, 505 U.S. at 169-170 (invoking Gregory canon as a ground to read a statute not to require the States to regulate a solution to radioactive waste in their domain, even though one provision of the law stated that "[e]ach State shall be responsible for providing . . . for the disposal of . . . low-level radioactive waste generated within the State"); Pennhurst State School & Hosp. v. Halderman, 451 U.S. at 22-27 (invoking federalism principles as a reason not to construe a federal law to require States, as a condition of receiving federal funds, to remedy unsanitary, inhumane, and dangerous conditions in a state-operated, federally funded facility for the care and treatment of the mentally retarded, even though the law expressly stated that “[p]ersons with developmental disabilities have a right to appropriate treatment, services, and habilitation for such disabilities” and that state governments “have an obligation to assure that public funds are not provided to any institutio[n] ... that . . . does not provide treatment, services, and habilitation which is appropriate to the needs of such person”).


Applying the Gregory/Bond Federalism Canon in King

Of course, many statutes have some sort of impact on the States, and the Court does not automatically apply the Gregory federalism canon in every such case.  The Court has not had occasion to precisely identify the types of effects on the federal/state balance that would trigger the requirement of finding certainty of congressional intent.  Wherever that line might be drawn, however, King is not a close case at the margins.  If the Court were to adopt the challengers’ reading, every State would be put to a terrible and momentous choice:  a decision not to establish its own insurance Exchange would result in a denial of tax credits for all of the State’s residents, which would lead, in turn, to a degradation and probable destruction of the health insurance market in the State.  Surely, then, this is a heartland case for application of the Gregory canon.

And here, far from any certitude that Congress intended such a draconian effect on the States, the challengers cite virtually no evidence, apart from the contested “established by the State” language itself, that Congress so intended.

Moreover, as both the Merrill amicus brief and the government’s brief demonstrate, there are numerous powerful indications in the Act itself that Congress did not have any such intent to put the States to such a momentous and potentially devastating choice.  As the challengers describe the Act, Congress was so intent on inducing the States to set up Exchanges that it was willing to sacrifice the ACA's signature reform—increased access to affordable health care—in those States that elect not to do so.  But a consideration of the Act as a whole—especially of those provisions that directly address the States and their decision whether to establish Exchanges—demonstrates that Congress’s intent was otherwise. 

Most importantly, the Act actually includes a specific provision explaining what will happen when a State elects not to set up an exchange, or when a State wishes to do so but falls short in implementation—subsection 1321(c), which is entitled “Failure to establish Exchange or implement requirements.”  If Congress had truly intended to use the denial of tax credits—and destruction of the State’s insurance market—as a dramatic means of inducing the States to establish Exchanges, surely there would have been some mention of it there.  But there is not.  As the Merrill brief emphasizes (p.19), “[i]nstead, Congress kept all of Section [1311] scrupulously free of any mention of this crucial consequence, while emphasizing in that provision the States' flexibility to decide, one way or the other, whether to set up Exchanges.”

As the challengers describe it, Congress instead buried the momentous no-subsidy condition in a provision of the tax code directed to individuals, not States.  And it did so implicitly, and only in a roundabout manner.  Section 36B is expressly designed to inform individual taxpayers how much “shall be allowed as a credit against” their income tax.  The operative provision of section 36B, subsection 36B(a), does not so much as mention that many individuals might not be eligible for a credit at all—namely, those who live in States that do not establish Exchanges.  More tellingly still, there is no any mention of such possible ineligibility in the definition of the “applicable taxpayer[s]” who are eligible for the tax credit, subsection 36B(c)(1)(A), which turns on an individual’s income, with no reference to the State in which she lives. 

According to the challengers, Congress instead inserted the most momentous condition in the whole Act in the provision defining a “coverage month” for an applicable taxpayer's tax credits.  (That provision defines a “coverage month” as “any month if … as of the first day of such month the taxpayer, the taxpayer's spouse, or any dependent of the taxpayer is covered by a qualified health plan described in subsection (b)(2)(A) that was enrolled in through an Exchange established by the State under section [1311] of the [ACA] and … the premium for coverage under such plan for such month is paid by the taxpayer.”)  On the challengers’ view of the Act, the number of “coverage months” in nonparticipating States will be zero—and that is the way Congress very stealthily informed the States that their residents will be ineligible for any tax credits if the State chooses not to establish an Exchange.

It is implausible that Congress would have imposed such a momentous and drastic condition on a State choice in such an indirect, counterintuitive—and, not surprisingly, unnoticed—manner.  As Part III of the Merrill amicus brief elaborates in detail, Congress has never used such a “hide the ball” maneuver in the many statutes—both conditional spending statutes and joint federal-state “cooperative federalism” enactments such as the ACA—in which it has imposed costly or significant conditions on choices it has offered the States.  (The crystal clear conditions on the States’ acceptance of Medicaid funding in the ACA itself—a major subject of the Court’s decision in NFIB v. Sebelius—is a prominent case in point.)  And for good reason:  because Congress, like the Court, appreciates that if the States are to be put to such a choice, it is crucial, and consistent with our federal structure, to permit the States and their residents a meaningful opportunity to examine and debate that proposal during the legislative process itself.

In sum, there is nothing in the ACA that could offer the Court any “certainty” of Congress' intent to impose such a radical and momentous choice on the State—and plenty of indications to the contrary.  Therefore, even if the challengers were correct that the contested phrase in section 36B has a “plain meaning” that could only be read in the fashion they propose, the Gregory canon, applied by six Justices of the Court just last Term as grounds for disregarding a very clear statutory definition, would be reason enough to affirm the IRS’s interpretation of the Act.