Symposium: Six Humpty Dumptys playing Calvinball

Michael F. Cannon (@mfcannon) is director of health policy studies at the libertarian Cato Institute. He has been described as “the intellectual father” of King v. Burwell and is co-author (with Jonathan H. Adler) of Taxation Without Representation: The Illegal IRS Rule to Expand Tax Credits Under the PPACA, Health Matrix: Journal of Law-Medicine 23, No. 1 (2013): 119-195.

In King v. Burwell, all nine Supreme Court Justices agreed on one thing. The King challengers claimed the Patient Protection and Affordable Care Act (ACA) authorizes the Internal Revenue Service to issue tax credits and impose the related penalties only “through an Exchange established by the State,” and not through exchanges established by the federal government. “Petitioners’ arguments about the plain meaning of Section 36B are strong,” Chief Justice John Roberts wrote, and their interpretation is “the most natural reading of the pertinent statutory phrase.” Justice Antonin Scalia agreed, finding the meaning of that phrase “so obvious there would hardly be a need for the Supreme Court to hear a case about it.”

There was no dissent about the plain meaning of the phrase “through an Exchange established by the State.” All seven of the other Justices joined one of those two opinions. Nor was there dissent about the fact that that phrase, used repeatedly in the statute, is the only provision of the Act that speaks directly to the question presented. Not a single Justice lent credence to the government’s assertions that this was a meritless case, or one that the Court should never have accepted. Nor was there dissent about the consequences of that provision’s plain meaning in the face of broad state resistance to the ACA. All agreed that withholding tax credits in the thirty-four states with federal exchanges could lead to adverse selection in those states, with premiums climbing higher and higher in a “death spiral.”

Where disagreement emerged was over the question of whether the latter should alter the former – whether the potential for adverse consequences “compels” the Court to disregard the universally acknowledged meaning of the operative text. The Court split six to three in favor of rewriting plain text, and rendering the requirement “established by the State” a nullity. The Chief Justice wrote for the majority, Scalia for the dissent. The effect of the ruling is that this and future administrations must do what everyone agrees the plain meaning of the operative text does not permit: spend hundreds of billions of dollars and tax 70 million employers and individuals in those thirty-four states.

Others can speak with greater authority about what this ruling means for textualism, contextualism, purposivism, and other legal doctrines. I can speak as one who relied on the plain meaning of that provision; who encouraged states not to establish exchanges because of the power that provision gave them to affect the course of the ACA; who helped lay the groundwork for King and three related cases; who can reasonably claim to have done more research on this aspect of the statute and its legislative history than any living human being; and who followed King all the way to the Supreme Court.

A number of things stand out about the opinion of the Court.

First, the Chief Justice would have benefited from spending more time with the statute. One of the factors that leads him to conclude “the Act may not always use the phrase ‘established by the State’ in its most natural sense [and] the meaning of that phrase may not be as clear as it appears” has to do with the Act’s definition of “qualified individuals.” He writes:

[Section 1312] defines the term “qualified individual” in part as an individual who “resides in the State that established the Exchange.”…And that’s a problem: If we give the phrase “the State that established the Exchange” its most natural meaning, there would be no “qualified individuals” on Federal Exchanges. But the Act clearly contemplates that there will be qualified individuals on every Exchange… It would be odd indeed for Congress to write such detailed instructions about customers on a State Exchange, while having nothing to say about those on a Federal Exchange.

As Jonathan Adler and I explain elsewhere, Roberts discovers a tension that simply isn’t there. Congress naturally referred to “the State that established the Exchange” because in Section 1312, Congress was speaking to the states and under the presumption that they would comply with the directive in the previous section (Section 1311) that each state “shall” establish an Exchange. Roberts is also incorrect that the Act “ha[s] nothing to say” about qualified individuals on federal exchanges. Just two sections later, indeed the instant Congress finished laying out the rules for state-established exchanges, Section 1321 drops the presumption of state cooperation and provides that if a state doesn’t establish its own exchange, “the Secretary shall take such actions as are necessary to implement such [a] requirement[].”

Then again, Roberts managed to conclude that “by the State” could be read to mean “by the federal government,” even though he acknowledged Congress explicitly defined “State” in a way “that does not include the Federal Government.” So perhaps spending more time with the statute would not have helped.

Second, Roberts’s opinion might benefit from basic logic. “The conclusion that Section 36B is ambiguous,” he writes, “is further supported by several provisions that assume tax credits will be available on both State and Federal Exchanges.” But do they? He cites two provisions in the section of the Act where Congress directs states to establish exchanges and requires them to furnish information about “the tax credits [available] under section 36B.” He cites a third that directs state and federal exchanges to report the information on people who do and don’t receive tax credits, and “aggregate amount of any advance payment of such credit,” if there was one. Petitio principii.

Third, for all of Roberts’s appeals to context, his opinion might benefit from spending more time with the broader context of the Act. Roberts concludes that Congress could not have meant “established by the State” literally because “Congress passed the Affordable Care Act to improve health insurance markets, not to destroy them.” Oh? As Roberts acknowledged the last time he adopted a saving construction of the ACA, Congress included in the Act a Medicaid expansion that would have destroyed Medicaid in any state that did not participate. The consequences of that part of the Act would have been even more jarring. Health coverage for the state’s most vulnerable residents would have disappeared.

Congress enacted both a Medicaid expansion and a system of Exchanges that allowed states to destroy what Congress sought to create. Congress did so for the same reason both times: they didn’t have the votes for anything else. Until today, I guess.

Fourth, his opinion might benefit from spending more time with the ACA’s legislative history. Roberts maintains that while Congress might have been willing to tolerate adverse selection in “comparatively minor programs,” it is “implausible” that Congress would have been willing to do so in “the very heart of the Act.” Yet all sides acknowledge that another leading bill advanced by Senate Democrats did exactly that.

In 2009, Democrats on the Senate’s Health, Education, Labor, and Pensions (HELP) Committee reported to the full Senate a bill that would have withheld exchange subsidies in any state that failed to implement that bill’s employer mandate. The HELP bill was ultimately merged with a Finance Committee-reported bill to produce the ACA. All sides agree that condition sat like a dagger pointed at the heart of the Act.

Indeed, the HELP bill’s community-rating price controls were even tighter then the ACA’s. In other words, many ACA supporters were willing to tolerate an even greater risk of adverse selection than would exist under a plain-meaning interpretation of the ACA, which Roberts was “compel[led] to reject” because Congress supposedly would never tolerate such.

Embarrassingly, Roberts even cites testimony from a 2009 HELP Committee hearing to show that Congress understood (and did all it could to avoid) the danger of adverse selection, while ignoring legislation the HELP Committee produced that shows ACA supporters rejected the very lesson Roberts thinks it internalized.

Fifth, the Court likely forestalled any challenges to other ways the IRS is illegally implementing the ACA’s premium tax credits and employer mandate.

University of Iowa professor Andy Grewal has identified several additional instances where the IRS expanded the availability of tax credits beyond the clear and unambiguous eligibility limits Congress imposed. Somewhat ironically, the agency “effectively provides the largest tax credits to persons who don’t satisfy the statutory criteria.” Some of these eligibility expansions would trigger penalties against employers under the ACA’s employer mandate, and therefore “could face judicial challenge.”

If six Justices were willing to rewrite a congressionally defined term like “State,” however, there may not be five votes on the Court to enforce a numerical eligibility limit like the requirement that tax-credit recipients have household income above one hundred percent of poverty. I mean, issuing tax credits to people below the poverty line advances the goal of “ensur[ing] that anyone who wanted to buy health insurance could do so,” right? Even if five votes do somehow exist, this opinion hid them so well I doubt any plaintiff would be willing to go to the trouble. The IRS may understandably see itself as having free rein to expand and contract both the ACA’s premium subsidies—and its mandate penalties. Because democracy.

Sixth, speaking of democracy, the Chief Justice’s majority opinion in King v. Burwell, like his controlling opinion in NFIB v. Sebelius, has further protected the Act from democratic accountability.

MIT health economist Jonathan Gruber infamously explained that architects of the ACA deliberately designed it to hide its taxes and transfers and thereby skirt democratic accountability. If those taxes were transparent, Gruber explained, public opposition would have prevented it from passing. Put differently, the ACA’s authors knew that voters would have rejected the ACA if they understood how it works, so they hid what they were doing. The ACA was an undemocratic enterprise from the start.

In NFIB v. Sebelius, Roberts positively gutted the most important constraint the Constitution imposes on Congress – democratic accountability – to save the ACA. He reasoned that if the individual mandate could fairly be viewed as a tax, it was the Court’s duty to uphold it as an exercise of Congress’s taxing power. How Congress described the mandate (i.e., whether it invoked its taxing power) was not relevant. What mattered was whether any of Congress’s enumerated powers could produce such a measure.

The so-called “magic words” doctrine has merit, but not when it allows Congress to defeat democratic accountability. In this case, Congress deliberately chose not to create the mandate using Congress’s taxing power because doing so would have prevented the bill from passing.Just as the enumerated powers doctrine prevented Congress from enacting an individual mandate under the Commerce Clause, the Constitution’s democratic constraints prevented Congress from imposing a tax on Americans who don’t buy insurance. It is difficult to imagine a clearer instance where democratic accountability prevented Congress from using an enumerated power.

When Roberts applied the “magic words” doctrine to the ACA’s individual mandate, he allowed Congress to blow right through the constitutional constraint of democratic accountability. Enabling Congress to evade this constraint is no more defensible than allowing it to evade the other. Roberts helped Congress enact a tax that Congress never could have imposed by itself.

Three years later, the Court is presented with the fact, and agrees, that the plain terms of the ACA give states the ability to veto the Act’s premium-assistance tax credits, cost-sharing subsidies, employer mandate, and (largely) its individual mandate. If enough states exercise those vetoes, the cost of ACA coverage would become transparent to such an extent that Congress would be forced to open it. Thirty-four states exercised those vetoes. A critical mass if ever there was one.

Indeed, ACA opponents swept into state office in 2010, 2011, and 2012 on waves of public opposition to that law. When the Obama administration chose to implement the disputed taxes and subsidies in those states anyway, it disenfranchised its political opponents in the states. It made the ACA appear more workable than, as all nine Supreme Court Justices acknowledge, the plain language of the statute would suggest. It insulated itself, and its allies in Congress, from democratic accountability for their support for the ACA. By promising to spend, and then spending tens of billions of dollars, it created a constituency with funds that, as nine Supreme Court Justices agree, the plain text of the ACA does not authorize it to spend. The Obama administration and its allies gained tremendous political advantage by departing from the plain meaning of the ACA. By baptizing the administration’s departure from the text of the ACA to its own political advantage – all in service of the ACA’s underlying goals, of course – the Court has once again protected the ACA from democratic accountability.

What it all means is that the ACA’s opponents haven’t failed. They succeeded. They read the bill. They stopped implementation in thirty-four states. Under the plain terms of the Act, that means they exposed the true cost of “ObamaCare” coverage. They forced Congress to reopen the statute. They won, and without abrogating any laws. They won using good, clean democracy.

Seventh, unfortunately, winning only gets you so far when you’re up against six Humpty Dumptys playing Calvinball. When he blew past Congress’ express definition of the term “State,” Roberts inadvertently omitted his cite to Louis Carroll:

“When I use a word,” Humpty Dumpty said, in rather a scornful tone, “it means just what I choose it to mean — neither more nor less.”

“The question is,” said Alice, “whether you can make words mean so many different things.”

“The question is,” said Humpty Dumpty, “which is to be master — that’s all.”

When he rewrote “established by the State” to include federal Exchanges and then glossed over the absurdities that interpretation creates elsewhere in the Act by claiming the phrase “mean[s] different things in different places,” he likewise omitted his citation to Calvin and Hobbes:

The only permanent rule in Calvinball is that you can’t play it the same way twice.

 

As one court-watcher put it, this type of ruling illustrates perfectly why bringing cases like King is so important. Selecting judges, too.