As an essential part of its effort to achieve near universal coverage, the Affordable Care Act (ACA) extends sizable tax credits to most people who buy insurance on the newly established health care exchanges. Yet several lawsuits have been filed challenging the availability of those tax credits in the thirty-four states that refused to set up their own exchanges. The lawsuits are premised on a strained interpretation of the ACA that, if accepted, would make a hash of other provisions of the statute and undermine its effort to extend coverage to the uninsured. The courts should reject this latest effort to dismantle a critical feature of the ACA.
Editor's Note: JHPPL has started an ACA Scholar-Practitioner Network (ASPN). The ASPN assembles people of different backgrounds (practitioners, stakeholders, and researchers) involved in state-level health reform implementation across the United States. The newly developed ASPN website documents ACA implementation research projects to assist policy makers, researchers, and journalists in identifying and integrating scholarly work on state-level implementation of the ACA. If you would like your work included on the ASPN website, please contact web coordinator Phillip Singer at firstname.lastname@example.org. You can visit the site at //ssascholars.uchicago.edu/jhppl/.
JHPPL seeks to bring this important and timely work to the fore in Report on Health Reform Implementation, a recurring special section. Thanks to funding from the Robert Wood Johnson Foundation, all essays in the section are published open access.
—Colleen M. Grogan
Put yourself in the shoes of my kids' piano teacher. She's got a part-time job accompanying music students at the University of Michigan and stitches together the rest of her modest income from teaching and performing at church services, weddings, funerals, and the like. She doesn't get health coverage through the university — she only works part-time — nor is she so poor that she can go on Medicaid.
Instead, she has to find coverage on the private market. Before the Affordable Care Act (ACA), however, private coverage was out of reach for her. Even without a preexisting condition, she couldn't afford to pay thousands of dollars for an individual health plan. She was barely scraping by as it was.
To help people like my kids' piano teacher, the ACA extends tax credits to anyone earning between one and four times the poverty level who buys a qualified health plan (26 U.S.C. § 36B(c)(1)(A)), so long as the individual is ineligible for Medicaid and doesn't have access to employer-sponsored coverage. The tax credits are substantial, averaging about $2,890 per person (Levitt, Claxton, and Damico 2014). And they are an essential part of the ACA's effort to achieve what the statute calls “near-universal coverage” (42 U.S.C. § 18091(2)(D)).
As Jonathan H. Adler and Michael F. Cannon read the ACA, however, my kids' piano teacher can't get tax credits at all. Nor should roughly 9.5 million other people scattered throughout the country (Levitt and Claxton 2014). In their view — a view endorsed by a panel of the D.C. Circuit (Halbig v. Burwell, 758 F.3d 390 (D.C. Cir. 2014)) — the ACA instead armed hostile state governments to thwart the ACA's effort to cover the uninsured. Their contention has assumed new urgency now that the Supreme Court has agreed to resolve the statutory dispute in King v. Burwell.
Yet Adler and Cannon fail to offer persuasive reasons to adopt their peculiar interpretation of the ACA. To the contrary, the government's alternative reading makes much better sense of the statute as a whole and avoids assigning a meaning to the ACA that is blatantly at odds with what the statute aims to accomplish.
The Supposed Glitch
To better organize the individual and small-group insurance markets, section 1311 of the ACA instructs the states to establish health care exchanges. From the consumer's perspective, the exchanges are just websites that allow for the easy comparison of health plans sold in a particular area. The hope is that the exchanges will enable price and quality competition in an unruly insurance market.
Congress anticipated that the states would welcome the opportunity to establish their own exchanges (Pear 2012). Better to retain state control over health insurers than cede that authority to the federal government. Nonetheless, Congress recognized that not every state might wish to establish an exchange. In states that declined to do so, section 1321 of the ACA tells the secretary of the US Department of Health and Human Services (HHS) to “establish and operate” the state's exchange.
As it happened, the choice of whether to establish an exchange got caught up in the political furor over Obamacare. For the thirty-four states that refused to establish exchanges, the federal government had to shoulder the unexpected burden of setting up the exchanges.
In Adler and Cannon's (2011) telling, that's where the ACA's supposed “glitch” comes in. The provision of the ACA governing the calculation of tax credits links the amount of the credit to the price of a health plan purchased on “an Exchange established by the State under 1311” (26 U.S.C. § 36B(b)(2)(A)). But what about people living in states whose exchanges were established by the federal government? Literal adherence to the statutory formula would suggest that subsidies aren't available on those exchanges. My kids' piano teacher would be out of luck.
The Obama administration isn't buying this interpretation of the statute. In 2012 the Internal Revenue Service (IRS) issued a rule extending tax credits to eligible individuals, whether or not their state established an exchange. That rule has now been challenged in four separate lawsuits. As Adler and Cannon see it, the rule ought to be struck down because it's inconsistent with the ACA. If they're right, tax credits that millions of people now depend on to buy health insurance would evaporate.
To make their case, however, it's not enough for Adler and Cannon to show that it's possible to read the ACA as withdrawing tax credits from refusal states. If the statute is ambiguous on that point, it's black-letter law that the courts must defer to the IRS's authoritative interpretation (Chevron U.S.A. v. Natural Res. Def. Council, 467 U.S. 837 (1984)). Adler and Cannon instead have to demonstrate that the statute unambiguously withdraws tax credits from people in refusal states — and that the IRS's contrary interpretation is downright unreasonable. They haven't come close to making such a demonstration.
The Statutory Text
When the federal government sets up an exchange on a state's behalf, it does not create some sort of federal exchange. Rather, it creates a state exchange. The text of the ACA is crisp on this point. Recall that section 1311 instructs states to establish their own exchanges. Under section 1321, when a state fails to set up the “required Exchange” — which is to say, a state exchange under section 1311 — the secretary of HHS must “establish and operate such Exchange within the State” (my emphasis).
In other words, the backup exchange isn't established under section 1321. It is established under section 1311, just as the provision authorizing tax credits requires. The secretary merely stands in the shoes of state officials in setting up their states' exchanges. Any exchange that the secretary establishes is the legal and functional equivalent of a state exchange.
Adler and Cannon seem to acknowledge the force of this argument, as did the D.C. Circuit panel that invalidated the IRS rule. They nonetheless resist the implication that tax credits are available in states with federally established exchanges. Their argument hinges on the claim that Congress linked tax credits to exchanges “established by the State under 1311” (my emphasis). Yes, they seem to say, a federally established exchange may be established “under 1311.” But they maintain that Congress used those three words — “by the State” — to signal that tax credits would be unavailable on federally established exchanges.
Reading the statute as a whole, however, it quickly becomes apparent that Congress never meant those three words to bear the immense weight that Adler and Cannon would assign to them. In several places in the ACA, Congress used similar “Exchange established by the State” language to refer generally to exchanges — including federally established exchanges. If that's so elsewhere in the statute, then why not in the provision governing the calculation of tax credits?
Most obviously, the ACA limits who can buy insurance on an exchange to those who “resid[e] in the State that established the Exchange” (42 U.S.C. § 18032(f )(1)(A)). If Adler and Cannon were correct that Congress scrupulously distinguished between state-established exchanges and exchanges in general, then no one in a state with a federally established exchange could go on that exchange to buy a health plan. As the D.C. district court put it, “The federal Exchanges would have no customers, and no purpose” (Halbig v. Sebelius, No. 13-0623 (D.D.C. 2014)). Adler and Cannon claim that the limitation on the exchange's customers is “moot” because it appears in a provision governing state exchanges, not in the provision governing federally established exchanges. But federally established exchanges are state exchanges. And Congress could not possibly have meant for those federally established exchanges to be an empty gesture.
Elsewhere, the ACA says that states have to maintain their Medicaid eligibility standards until “an Exchange established by the State” is up and running (42 U.S.C. § 1396a(gg)(1)). This maintenance-of-effort provision was meant to provide stopgap protection for Medicaid beneficiaries until the exchanges went live. On Adler and Cannon's telling, however, it would forbid a state that declined to establish an exchange from ever relaxing its Medicaid standards. There is zero evidence that Congress meant the ACA to freeze state Medicaid programs into perpetuity.
Brushing past these statutory signals, Adler and Cannon aver that the phrase “established by the State” appears in a number of places in the act. But it doesn't matter how many times Congress used the phrase. The question is what Congress meant by the phrase. Its repetition is perfectly consistent with the view that Congress — anticipating that nearly all the states would establish exchanges — just meant to refer to exchanges in general. At a minimum, Congress's slipshod use of the “by the State” language gives rise to ambiguity as to what Congress meant by it. And when there's ambiguity about federal taxes, it's up to the IRS to resolve that ambiguity.
Adler and Cannon nonetheless insist that Congress meant the “by the State” language in the tax-credit calculation to serve some distinct purpose. What purpose do they have in mind? As they see it, Congress withheld tax credits from states that declined to create their own exchanges in order to goad them into establishing exchanges. Congress was making a threat: set up your own exchanges or you'll lose out on tax credits.
This claim is quite implausible. To begin with, Congress knows how to threaten states with financial consequences when it wants to. Adler and Cannon demonstrate as much by identifying a number of other bills — not the ACA — that contain clear language conditioning federal money on certain forms of state compliance. The absence of any such clear language in the ACA is powerful evidence that Congress never meant the availability of tax credits to depend on whether a state established an exchange. After all, there was no need to bully the states into doing what everyone assumed they would gladly do.
More to the point, threats must be communicated. When Vito Corleone made the proverbial offer that can't be refused, he didn't just say “sign the contract.” He had Luca Brasi hold a gun to the head of a guy and “assured him that either his brains or his signature would be on the contract.” Without the gun, there's no threat (Bagley 2014). In the ACA, however, the threat that Congress supposedly meant to level was so well hidden that the states never noticed it. A careful report from the Georgetown University Health Policy Institute, documenting what states knew and when they knew it, has found no contemporaneous evidence that the availability of tax credits played a role in state decisions over whether to establish exchanges (Dash, Monahan, and Lucia 2013; Monahan 2014). How can Adler and Cannon say that Congress unambiguously threatened the states if the states never got the message?
Their reading is anomalous for another reason. In the absence of tax credits on federally established exchanges, many healthy people would forgo coverage. Those with serious health problems, however, would likely remain in the market. The risk pools for exchange plans would become skewed toward sicker enrollees, which would in turn increase the cost of exchange plans. As costs went up, relatively healthy people would decline insurance, leading to sicker risk pools and even higher prices, which would drive away still more healthy people. In short, the exchanges would stop working. As a just-released study from the RAND Corporation explains, “If the ACA's subsidies are eliminated entirely, our model predicts a near death spiral — that is, sharp premium increases and drastic enrollment declines in the individual market” (Eibner and Saltzman 2014: 25). Why construe the ACA to create dysfunctional exchanges when an alternative reading would avoid that result?
If Congress did mean to use the three words to threaten the states, it should be easy to find evidence of that threat in the legislative record. Yet there is none. Adler and Cannon themselves prove the point. Over the past few years, they've worked tirelessly to sift through the legislative record, attentive for any hint that might support their preferred interpretation. Yet they've found nothing.
Take, for example, the letter they discuss from the eleven Texas Democrats in the House of Representatives. In the letter, the Democratic legislators implored the president not to give the states the power to create their own exchanges. “In Texas,” they explained, “we know from experience that the dangers to the uninsured from greater state authority are real” (Doggett 2010). Better by far, the legislators argued, to establish a national exchange — one that Texas officials couldn't interfere with.
Adler and Cannon assert that the legislators were voicing concerns that Texas citizens might not get tax credits. Yet the letter says not one word about tax credits. Not one. The absence of any mention of tax credits is the dog that didn't bark in the night — a clue that the legislators had no inkling of any supposed threat. The legislators were voicing concerns about “obstruction” from Texas officials, not about the loss of tax credits from the federal government. And they were right to be worried. Texas not only declined to establish its own exchange. It also announced that it wouldn't enforce the ACA's insurance rules (Luthra 2013), imposed onerous training requirements on the navigators who help people buy insurance (Aaronson 2014), and refused to expand Medicaid (Ramshaw 2012).
What else do Adler and Cannon point to? After the ACA was signed into law, the House of Representatives passed the Reconciliation Act to make a few changes. Among those changes, the House clarified that territories were to be treated as states if they chose to establish exchanges. “It strains credulity,” Adler and Cannon assert in an amicus brief they submitted to the D.C. Circuit, “that Congress … would notice and remedy the bill's failure to authorize [tax credits] in territorial Exchanges, but would not notice its failure to authorize them in federal Exchanges” (Brief of Amici Curiae Jonathan H. Adler and Michael F. Cannon, Halbig v. Burwell, No. 14-5018 (D.C. Cir. 2014)). But it doesn't strain credulity at all. It's only natural that Congress wouldn't notice — much less assign significance to — three innocuous words in a dense, hard-to-follow statutory formula. What is truly inconceivable is that Congress would extend tax credits to territorial exchanges but — without so much as a whisper — deliberately refuse to extend them to federal exchanges.
Finally, Adler and Cannon trace the drafting history of the ACA in an effort to support their view that Congress meant that three-word phrase, “by the State,” to do important work. All they demonstrate, however, is that Congress kept using the phrase during the frantic negotiations over the ACA. At the time, however, everyone thought that the exchanges would be established by the states. Not a shred of evidence supports Adler and Cannon's conjecture that “Senate leaders and White House officials” engineered the insertion of the phrase in the tax-credit calculation in order to strip tax credits from the citizens of refusal states.
I could go on. I've made my point, however. In poring through the legislative history, Adler and Cannon have committed the cardinal sin of looking over a crowd to pick out their friends. Only here, they have no friends.
Adler and Cannon have offered a strained interpretation of the ACA that, if accepted, would make a hash of other provisions of the statute and undermine its stated purpose of extending insurance to nearly all Americans. The more natural reading — one that makes far better sense of the statute as a whole — is that tax credits are available in both the states that established exchanges and those that did not. On this view, the “by the State” language just reflects Congress's assumption, unchallenged at the time, that the states would establish their own exchanges.
But even if you think that Adler and Cannon's claim is plausible, maybe even attractive, the contrary interpretation offered by the government is at least reasonable. That brings me to the aspect of their argument that troubles me the most: their unyielding conviction that they've identified the only possible construction of the ACA. Nowhere do they so much as acknowledge the possibility that maybe, just maybe, they're wrong.
That's because they can't admit to doubt. Because of the deference extended to agency interpretation, doubt means they lose. But their unwillingness even to acknowledge ambiguity reflects an important difference between legal advocacy and neutral interpretation. To be clear, Adler and Cannon deserve immense credit for their lawyerly ingenuity: they've constructed a facially plausible argument in support of an exceedingly strange interpretation of the ACA. But the courts would violate their obligation of fidelity in statutory construction if they mistook that ingenuity for genuine obeisance to congressional will. The latest challenge to the ACA is political activism masquerading as statutory restraint.