an unanticipated consequence of
Jack M. Balkin
Jack Balkin: jackbalkin at yahoo.com
Bruce Ackerman bruce.ackerman at yale.edu
Ian Ayres ian.ayres at yale.edu
Mary Dudziak mary.l.dudziak at emory.edu
Joey Fishkin joey.fishkin at gmail.com
Heather Gerken heather.gerken at yale.edu
Abbe Gluck abbe.gluck at yale.edu
Mark Graber mgraber at law.umaryland.edu
Stephen Griffin sgriffin at tulane.edu
Bernard Harcourt harcourt at uchicago.edu
Scott Horton shorto at law.columbia.edu
Andrew Koppelman akoppelman at law.northwestern.edu
Marty Lederman msl46 at law.georgetown.edu
Sanford Levinson slevinson at law.utexas.edu
David Luban david.luban at gmail.com
Gerard Magliocca gmaglioc at iupui.edu
Jason Mazzone mazzonej at illinois.edu
Linda McClain lmcclain at bu.edu
John Mikhail mikhail at law.georgetown.edu
Frank Pasquale pasquale.frank at gmail.com
Nate Persily npersily at gmail.com
Michael Stokes Paulsen michaelstokespaulsen at gmail.com
Deborah Pearlstein dpearlst at princeton.edu
Rick Pildes rick.pildes at nyu.edu
Alice Ristroph alice.ristroph at shu.edu
Neil Siegel siegel at law.duke.edu
Brian Tamanaha btamanaha at wulaw.wustl.edu
Mark Tushnet mtushnet at law.harvard.edu
Adam Winkler winkler at ucla.edu
ObamaCare Subsidies Before the D.C. Circuit: Clearing Up Some ACA Facts and History
The ObamaCare subsidies challenge, Halbig v. Sebelius, will be argued tomorrow in the D.C. Circuit. In this latest round of litigation, the challengers have made a variety of kitchen-sink arguments, many of which are incompatible with the basic principles of statutory construction on which their briefs purport to rely and which evince a misunderstanding of the ACA's procedural history. The Government's briefing and the lower court opinions understandably chose not to respond in detail to all of them; but given the intensity with which some of these arguments have appeared on the blogs in recent days, I will use this post to try to clear up some of these matters.
For readers just getting up to speed, the lawsuit aims to prevent the Administration from subsidizing insurance purchases in those states in which the federal government, rather than the states, is operating the Affordable Care Act’s (ACA) new insurance marketplaces (called “exchanges”). The case is a big deal: more than 12.5 million people are expected to be eligible for this financial assistance on the federal exchanges; the subsidies average more than $5,000 per person; and the subsidies are essential to supporting the insurance-purchase mandate—which in turn supports the insurance-market reforms at the heart of the ACA. The challengers' argument is based on a line in the ACA that provides that the subsidies shall be available to individuals enrolled in insurance “through an Exchange established by the State under section 1311,” which they argue clearly excludes individuals enrolled through federally-operated exchanges. I have already written extensively (here, hereandhere) about why that textual argument—although superficially appealing—should be rejected on its own terms: The ACA’s overall textual structure and context dictate the opposite result. The D.C. district court reached the same conclusion; and it is on those grounds that this case will hopefully be decided on appeal. This post has a different aim: namely, to clear the air on some of those additional, throw-and-see-if-it-sticks assertions that have been made, and to provide the factual and legal background.
#1: This is not a conditional spending program analogous to Medicaid.
The challengers' strategy in this round has been to contend that the subsidies are part of an overarching ACA "carrots and sticks" strategy to lure states into health reform and penalize them if they decline. On that version of the story, it might make sense that subsidies would be unavailable in states that do not run their own exchanges. In their view, the subsidies are therefore exactly like the ACA’s Medicaid provision (from appellants’brief: “The ACA’s subsidy provision offered an analogous ‘deal’ to entice states to establish Exchanges—because Congress (wisely, in hindsight) knew it had to offer huge incentives for the states to assume responsibility for that logistically nightmarish and politically toxic task.”)
Putting aside the fact that no one thought the states wouldn’t want to run the exchanges themselves (indeed, Senators were demanding that option for their states), the exchange provisions simply do not work in the same way as Medicaid. Unlike the ACA’s Medicaid provisions, the exchange provisions have a federal fallback: Medicaid is use it or lose it; the exchanges are do it, or the feds step in and do it for you. In other words, this isn’t Medicaid; it’s the Clean Air Act (CAA). If a state decides not to create its own implementation plan under the CAA, its citizens do not lose the benefit of the federal program—the feds run it. The same goes for the ACA’s exchanges and so it would be nonsensical to deprive citizens in federal-exchange states of the subsidies. More importantly, if we are going to compare apples to oranges, the ACA’s Medicaid provisions have an explicit provision stating that if the state declines to participate, it loses the program funds (this was the provision at issue in NFIB v. Sebelius in 2012). The ACA’s subsidy provisions, in contrast, have no such provision, strong evidence that the subsidies were was not intended to be forfeited if the states did not participate. If the challengers are going to insist on strict textual arguments, this is exclusio unius 101: the rule of interpretation that provides that where Congress includes a specific provision in one part of the statute but does not include an analogous provision elsewhere, that omission is assumed intentional.
#2: The HELP Committee Bill Did Not Deprive All Federal-Exchange States of the Subsidies
As part of their effort to depict the subsidies as "carrots" for state participation, the challengers have relied on an early draft bill from the HELP Committee, much of which was later incorporated into the ACA.Setting aside the fact that the challengers should not be able to simultaneously embrace and reject the ACA’s legislative history, as the briefs do, they have focused on the Senate HELP Committee’s original draft, which was subsequently merged with a Finance Committee draft and further morphed by Majority Leader Reid—all of which should give us pause about relying on these early drafts in the first place. Regardless, challengers have asserted that the HELP Committee bill denied premium credits in states that did not establish exchanges (from the brief: “Under the HELP bill, if a state established an Exchange (“Gateway”), residents could receive “credits” almost immediately. Affordable Health Choices Act, supra, § 3104(b)(1) …. If a state neither established an Exchange nor requested a federal Exchange, “the residents of such State shall not be eligible for credits” until four years after the date of enactment. Id. at § 3104(d) (emphasis added). The HELP bill permanently withheld credits in states that failed to implement the bill’s employer mandate. Id.”)
Pull up theHELP bill for yourself and keep reading. The section of the HELP bill that challengers’ rely upon—§3104--did give the subsidies to the federally-run exchanges. That section gave the states a choice to adopt their own exchanges (which it called “Gateways”) and adopt the supporting insurance provisions (§3104(a)(1)), or “request that the Secretary operate … a Gateway in such State” and adopt the supporting relevant insurance provisions (§3104(a)(2)), or do neither (§3104(a)(3)). The next section required the Secretary to establish the Gateway herself if the state does not adopt its own Gateway under any of the three provisions of section (a). (§3104(b)(c)). Critically, section (b)(3) provides “the State shall be deemed to be an ‘establishing State’ on the date on which the Gateway established by the Secretary is in effect in Such state.” In other words, as soon as the federal exchange was set up, it was to be considered a state exchange for all purposes—the exact argument that the government is now making before the D.C. Circuit. Finally, the section expressly provided that—exactly as in the case of state-run gateways—the subsidies would become available on the federally-run gateways within 60 days. (“Any resident of a State described in paragraph (3) … shall be eligible for credits under section 3111 beginning on the date that is 60 days after the date which such Gateway is established in such State [by the Secretary].” §3104(c)(4). )
It is true that the HELP bill (§3104(d)) also provided that subsidies would be denied for four years in states that both refused to adopt the reforms themselves and also refused to request the federal government to do it for them. But that conceptualization of state-choice did not make it into the ACA. Instead, the ACA replicates only the first two state-flexibility options from the HELP bill--the options for which the subsidies were always available: the states can establish the exchanges themselves or the feds will do it for them. The subsidies were to be given to both. The ACA, unlike the HELP draft, does not give the states the choice to decline the implementation altogether. (Again, it’s the Clean Air Act. It’s not Medicaid.)
#3: The Question of Subsidies on the Exchanges Had Nothing to Do with the Jurisdiction of the Senate Finance Committee
e challengers also have persisted in citing a stray 2009 remark by Senator Baucus as further proof of their "carrots" argument, even after the factual premise of that remark has been corrected. (From the appellant’s brief: “Senator Max Baucus, used the conditional nature of the subsidies to justify his jurisdiction over the Exchanges and related regulations of health coverage in the draft ACA; that is, the Finance Committee had jurisdiction over health issues only because the bill conditioned “tax credit” subsidies, within its bailiwick, on states creating Exchanges subject to regulation.” One amicus, on the blogs, has made the same argument here.). As an initial matter, the Baucus comment had nothing to do with differentiating between the state and federal exchanges. It was an explanation of why his Committee shared jurisdiction over health reform with the HELP Committee. (Specifically he was explaining why the Finance Committee had jurisdiction over amendments relating to health insurance coverage even though it would not have jurisdiction over medical malpractice amendments.) But even if it were relevant, it tells us nothing about whether the subsidies might be offered on one, the other, or both. (Read the transcript for yourself.) More importantly, it is simply not true, as the challengers claim, that also including subsidies for the federal exchanges would somehow have deprived the Finance Committee of jurisdiction over the ACA. There is zero evidence for any such argument in the record or in the rules of the Senate. Regardless as I have already detailed (here) a stray comment early in the drafting process of a statute with a legislative process as unorthodox as the ACA’s has no place in the judicial decision-making process.
#4: The IRS Has Authority to Issue this Rule
The challengers also argue that the IRS did not have the authority to adopt the rule in question in the case. As DOJ has argued, the provision in question appears in the Internal Revenue Code and expressly states that the Secretary “shall prescribe such regulations as may be necessary to carry out the provisions of this section.” (ACA §36B(g)). That should be enough to defeat this argument. The challengers nevertheless argue that, because the section in question cross-references ACA §1311, and because §1311 falls under HHS authority, the IRS should not have been able to interpret the section. At the same time, they claim: “[c]onversely, the IRS Rule would not be entitled to deference had it been promulgated by HHS rather than the IRS” because “HHS … does not administer the subsidy provision, 26 U.S.C. § 36B.” Right. That’s because the IRS does. In other words, challengers are claiming that no one has the authority to interpret the section! This argument is entirely at odds with a statute that is replete with hundreds of provisions for administrative implementation and that clearly intends the agencies to do the lion’s share of the gap filling--including in this very section. The broader argument here is that Chevron deference for administrative rulemaking is not appropriate at all when multiple agencies are involved. In my view, this is not a multiple-agency question: the section here sits squarely in the Internal Revenue Code and, unlike classic multiple-implementation cases, the section in question does not also mention HHS’s authority. Instead the ACA simply gives the IRS full authority here. In any event, I have previously offered evidence (here)) for why a no-deference rule for multiple implementers is inconsistent with current doctrine.
#5: State Officials Did not Base Their Decisions Not to Participate On The Assumption the Subsidies Would Not Be Offered In the Federal Exchange
A group of state officials has filed amicus briefs claiming that they decided not to run an exchange because they assumed the subsidies would not be offered on a federally-run exchange (the argument here is that these state officials were purportedly trying to relieve their citizens from the insurance mandate: because the subsidies make insurance more affordable and trigger some employer requirements to provide health insurance, without the subsidies more people would fall into the ACA’s hardship exemption and not be subject to the mandate). These assertions are almost certainly false, according to evidence provided by the Center on Health Insurance Reforms at Georgetown (CHIR). CHIR has been tracking the states’ relevant public statements and letters throughout implementation. They have demonstrated that the states’ public justifications for their decisions—their official declaration of intent not to pursue a state exchange—made no mention of this argument whatsoever. Virginia filed an amicus brief in the companion case illustrating the same thing. Instead, the states' official declarations mentioned considerations like lack of state flexibility and not enough guidance from the Administration CHIR also has provided statements from some of these states’ leaders evincing their understanding that some subsidies would indeed be offered to their citizens through the federally run exchanges. In any event, state officials' post hoc justifications should have no import for an inquiry into congressional intent.
#6: Any “Legitimate Method of Statutory Construction” Undermines the Challengers Case
Finally, in an apparent attempt to scare the textualist D.C. Circuit panel, the challengers have littered their briefs with sentences like the following: “No legitimate method of statutory construction would interpret the phrase “Exchange established by the State under section 1311”’ to include the federal exchanges. The challengers have two primary textual arguments: 1) the text just quoted above, standing in isolation and 2) the textual canon known as the rule against superfluities, which counsels courts to give effect to every statutory phrase (and not render any redundant). Any principled textualist will see that there are many more textual arguments on the other side. I already have mentioned the exclusio unius canon above. Here are a few more.
The superfluities argument made by the challengers actually cuts in favor of the Government: ACA §36B(f) provides:
‘‘(3) INFORMATION REQUIREMENT.—As revised by section 1004(c) of HCERA.Each Exchange (or any person carrying out 1 or more responsibilities of an Exchange under section 1311(f)(3) or 1321(c) of the Patient Protection and Affordable Care Act) shall provide the following information to the Secretary and to the taxpayer with respect to any health plan provided through the Exchange:… ‘‘(C) The aggregate amount of any advance payment of such credit or reductions”….
Section 1311 refers to the state exchanges; section 1321 refers to the federal exchanges. Half of this section—which requires reporting to the IRS of the amount of the subsidies offered on both exchanges—would be superfluous the subsidies were not available on the federal exchange. It also would be absurd—another favorite textualist canon.
Furthermore, as noted by the words in the statute in italics, this section, §36B(f), was added to the ACA in the reconciliation legislation, which followed a week after the ACA’s enactment-. That legislation harmonized House/Senate preferences and took the place of Conference, which is typically the last stage of legislation and brings together the versions from each chamber. Every legislation expert knows that this last stage—where differences across the chambers are resolved—is the most important. It’s why the Supreme Court pays special attention to Conference Reports, for example. As a result, §36B(f) is the latest piece to be added to the official original ACA statute and is part of the only document that was negotiated and written by both chambers together. As I have written previously (here), unforeseeable political circumstances made the Senate text of the ACA the text that had to be adopted virtually unchanged by the House. The reconciliation bull was the one and only later-coming piece of the ACA that was negotiated and written by both House and Senate. It is no small matter, then, that §36B(f) clearly contemplates the subsidies on both types of exchanges.
Finally, the challengers have ignored two of the most important textual canons of all: 1) that text must be interpreted in context, a point always espoused by leading textualists, including Justice Scalia and John Manning; and 2) the whole act rule—also known as the non-derogation canon—the rule that statutes must be read as a whole, giving effect to all provisions in a way that makes them work as a coherent whole. The Government’s brief is replete with references to other parts of the ACA that make no sense—like §36(B)(f) above—if the court reads the federal subsidies out of the statute.
In a recent blog post, one amicus’sresponse to this argument that a single phrase should not be able to undermine the entire act is the following: “the Obama administration aborted another PPACA entitlement – the “CLASS Act” – because a single statutory phrase forbade its implementation unless the program could be actuarially solvent.” The CLASS Act was a title of the ACA concerning long term care—and once again, the comparison actually cuts the other way. Concerned that a long-term care insurance program would be economic infeasible, Congress included an economic feasibility requirement in the text of the statute.A “single statutory phrase” can indeed halt an entire statute if Congress clearly so provides its intention for the entire program to be halted. The absence of a similar clear statement in the case of the subsidies provisions is telling (exclusio unius). Moreover, virtually every canon of statutory interpretation counsels courts that, where the statutory meaning is less than clear, they should render the statute coherent—or defer to the agency..
Would it have been preferable had Congress done a better job drafting this behemoth law? Of course. It likewise would be preferable if our politics were not such that no one dare touch the statute to make little fixes that would clean it up, or improve it. But bad politics should not make bad law. This case raises important matters of statutory interpretation doctrine that have significance far beyond the confines of this politically charged case. The District Court applied ordinary rules of statutory construction to sustain the statute. Those of us who study legislation hope that the D.C. Circuit will give the statute the same careful look tomorrow.