TaxProf Blog op-ed: Preliminary Thoughts on Habig and King, by Bryan Camp (Texas Tech):
Yesterday two different U.S. Courts of Appeal disagreed on the validity of a tax regulation. It is not a rare event for two federal courts of appeals to disagree on the same issue of law. What made yesterday’s rulings a rare event was that the disagreement arose within a matter of hours and involved a key provision of the Affordable Care Act (ACA). At issue was the validity of a Treasury regulation on how certain ACA tax credits are to be calculated. At trial, both the district court for the District of Columbia and the district court for the Eastern District of Virginia had upheld the regulation. On appeal, the D.C. Circuit struck down the regulation in Halbig v. Burwell but, hours later, the 4th Circuit upheld the same regulation in King v. Burwell.
The two panels of three judges produced five written opinions. The D.C. Circuit panel found the Treasury regulation an invalid interpretation of the statute by a 2-1 vote. Each judge wrote an opinion. The majority opinion was penned by Judge Griffith, with Senior Judge Randolph joining and writing a short concurrence to emphasize his view that the government’s arguments really sucked wind. Senior Judge Edwards wrote an impassioned dissent. The 4th Cir. vote was 3-0, with two opinions. Judge Gregory, joined by Judge Thacker and Senior Judge Davis, penned the opinion for the Court. Senior Judge Davis added a short concurrence to emphasize his view that the plaintiff’s arguments really sucked wind.
This post will summarize the arguments and the opinions, then make three brief observations about (1) a non-barking dog, (2) plain language pizza, and (3) what these cases might teach about who---as between courts, Congress, or agencies---ought to be cleaning up statutory messes created by poor drafting.
Given the process that produced the ACA, it should come as little surprise that parts of the statute were not well drafted. These two cases involve one particularly embarrassing example, a provision codified at 26 USC §36B, dealing with the calculation of the federal tax subsidies for certain purchasers of a Qualified Health Plan (QHP).
Section 36B(a) authorizes a tax credit equal to “the premium assistance credit amount” as determined by §36(b). Section 36(b)(1) then explains that the premium assistance credit amount is calculated by adding together all the “premium assistance amounts” for all “coverage months.” It is in the definition of these two additional terms of art where we find the drafting embarrassment.
The first term of art (premium assistance amounts) is defined in §36(b)(2) as being the monthly premiums for a QHP “offered in the individual market within a State...and...enrolled in through an Exchange established by the State under [section] 1311 [of the ACA]...” (emphasis supplied) The second term of art (coverage month) is defined deeper down in §36B(c)(2) as “any month if...as of the first day of such month 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]...” (emphasis supplied)
The embarrassing part is the bolded language. Did Congress really mean to deny the federal tax credit to all those taxpayers enrolled in QHPs through exchanges established by the federal government? Or was the statutory language simply badly drafted? These are nontrivial questions: only about 14 states have established Exchanges under ACA §1311. In the other states, the Exchanges have been established by the federal government under ACA §1321.
Treas. Reg. 1.36B-2, T.D. 9590, 77 FR 30377-01 (May 23, 2012), addresses this embarrassment by treating the statutory language as poorly drafted. Thus, subsection (a) of the regulation provides that applicable taxpayers are “allowed a premium assistance amount...for any month that [the taxpayer] ....(1) Is enrolled in one or more qualified health plans through an Exchange...” The regulation goes on to say that the term “Exchange” includes State Exchanges, regional Exchanges, subsidiary Exchanges, and Federally-facilitated Exchanges. While the Preamble to the regulation acknowledged that commentators had raised concerns that the statutory language prohibited the regulatory approach, the Preamble dismisses those concerns (with no analysis) by asserting that that the regulatory approach was “consistent with the language, purpose, and structure of section 36B and the Affordable Care Act as a whole.”
Plaintiffs and their Amici:
a. Text: The plaintiff’s textual argument is very straightforward: the plain language at issue creates three requirements: the applicable taxpayer must be enrolled in a QHP through (1) an Exchange, (2) established by the State, (3) under ACA 1311. An Exchange set up by the Federal government fails the second requirement. If Congress had meant to include Exchanges other than State exchanges in the tax credit calculation, it would have used the language “an established Exchange” or something like that.
b. Statutory Context: Congress elsewhere provided that a U.S. Territory that set up an Exchange “shall be treated as a State.” 42 U.S.C. §18042(a)(1). There is no equivalent language for Exchanges set up by the Federal government.
c. Statutory Purpose: Congress wanted to create incentives for states to create exchanges. Providing a tax subsidy only for individual coverage purchased through state-established Exchanges serves that purpose.
Government and its Amici:
a. Text: The government’s textual argument starts simple, but quickly becomes convoluted. The plain language at issue creates TWO requirements: the applicable taxpayer must be enrolled in a QHP through (1) an Exchange (2) established by the State under ACA §1311. Federally established exchanges meet both requirements because §1321 says so. While §1311 provides that States “shall” establish an exchange, that seemingly mandatory verb is altered by ACA §1321 which allows States to opt out. If a state opts out, §1321 then requires the federal government to “establish and operate such Exchange within the State.” The term “Exchange” is defined in a part of the ACA that was codified in 42 U.S.C. to be “an American Health Benefit Exchange established under section 18031 of this title.” 42 USC §300gg-91(d)(21) (“section §18031 of this title” was the codification of §1311). So the term “such Exchange within the State” in ACA §1321 means, by substitution of definition, “such American Health Benefit Exchange established under section 1311 within the State.” A plain reading of ACA §1311 shows that Exchanges “established under §1311” are, again by definition, established by a State. See e.g. (b)(1) “Each State shall, not later than January 1, 2014, establish an American Health Benefit Exchange (referred to in this title as an “Exchange”) for the State...” Therefore, unless one wants to make the argument that the term “Exchange” itself necessarily excludes federally-created Exchanges, this textual analysis supports the notion that Exchanges set up by the federal government under the command of ACA §1321 are indeed “established by the State under 1311.”
b. Statutory Context: Other statutory provisions appear to presume that the tax credit would be available for taxpayers regardless of whether the Exchange was established directly by a State under ACA §1311 or indirectly by the federal government stepping into the shoes of the State as required by ACA §1321. First, IRC §36B(f)(3) subjects all Exchanges to a variety of reporting requirements among which are several that make no sense if the Exchange did not qualify to support a tax credit. Second, ACA §1312 provides that only an individual who “resides in a State that established the Exchange” can purchase a QHP from an Exchange. If federally-established Exchanges were not fully equivalent to State-created Exchanges, §1312 would mean only individuals living in the 14 or so States that have established Exchanges could purchase a QHP.
c. Statutory Purpose: The government makes an “everybody knows” argument that Congress intended the ACA to increase the availability of health insurance and created a “three-legged stool” to implement that goal. The first leg was to mandate coverage, regardless of pre-existing conditions, and create certain other requirements that must be offered by every insurer. The second leg was the Individual Mandate, a stick designed to counter the adverse selection problem (Jordon Barry and I discuss that in our article “Is The Individual Mandate Really Mandatory?” available here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2091556 ) The third leg was the tax credits, also designed to counter the adverse selection problem with a carrot for those taxpayers who would not be affected by the Individual Mandate.
d. Chevron. The Supreme Court in Chevron told courts that the first step in reviewing a contested agency interpretation of a statute was to see if Congress had directly spoken to the precise question at issue. If so, the statutory text controlled. If not, the second step was to evaluate the agency’s interpretation of the applicable statutory text for reasonableness. Only if the agency was off the wall, out to lunch, in left field, or whatever other colorful metaphor floats your boat, should a court exercise power to overrule an agency interpretation of ambiguous statutory text. Power is allocated in the federal system so that agencies have more power than courts (if that power is properly invoked and exercised) to interpret ambiguous statutory text. Here the government argued for ambiguity. If it could convince a court of that, then there would be no question that the Treasury Department properly invoked and exercised its regulation-writing powers.
Here is how each court dealt with the three sets of arguments present.
The DC majority agreed with key parts of the plaintiff’s textual argument. That is, the majority interpreted IRC §36B as created three requirements of (1) an Exchange, (2) established by the State, (3) under ACA 1311. The majority agreed with the government on its substitution of definition analysis but believed that analysis met only two of the three requirements: “Thus, §1321 creates equivalence between state and federal Exchanges in two respects: in terms of what they are and the statutory authority under which they are established.” But this was not good enough because of that second requirement “established by the State.” “Of the three elements of that provision....federal Exchanges satisfy only two: they are Exchanges established under section 1311. Nothing in section 1321 deems federally-established Exchanges to be “Exchanges established by the State.”
The DC dissent appeared to concede the textual analysis, relying instead on the statutory context and purpose to repeatedly warn that the straight textual reading would “crumble” the ACA by sawing off one of the three legs of the stool.
The 4th Cir. Panel accepted the government’s definitional substitution argument, but cautioned that the government’s position was “only slightly” more convincing than the “common-sense appeal” of the taxpayer position. While the opinion did not articulate exactly why it was convinced, my reading of the way in which the Court set up the parties’ arguments convinces me it was because the Court read §26B as containing only two requirements, not three. That is, the phrase “established by the State under section 1311” was a single requirement and therefore, in the court’s words, “it makes sense to read §132(c)’s directive that HHS establish “such Exchange” to mean that the federal government acts on behalf of the state when it establishes its own Exchange.” I, personally, cannot see any other reason to prefer the government’s textual argument. This idea of two requirements is more clearly expressed in Judge Davis’ concurrence who argues that the phrase “established by the State” is a term of art.
b. Statutory Context:
Once the DC majority concluded that pure textual analysis answers the question presented, it considered any other interpretation to be a “juridical rewriting of an act of Congress.” While the majority was willing to do that, it repeatedly required that the government make an “extraordinary showing” that contrary statutory context or statutory purpose trumped the text. Otherwise, the text controlled and the agency would lose under Chevron step 1.
The DC majority addressed the government’s statutory context arguments and concluded that neither of the two sections were “unworkable” under the majority’s interpretation of §36B. Sure, these provisions might be a bit awkward, but that is a long way from being absurd. “A provision thus may seem odd without being absurd and in such instances it is up to Congress rather than the courts to fix it even if it had been an unintentional drafting gap.” (internal quotes omitted). First, as to the reporting requirements in §36B(f)(3), the majority pointed out that the information that appears to relate to the tax credits also relates to the Individual Mandate, so the information would still serve a purpose, just not as complete a purpose as it would under the government’s interpretation. Second, as to §1312, the majority suggests that the government misreads that section to create the asserted absurdity. Specifially, the majority denies that §1312 acts to limit the population who can purchase QHP (“the obvious flaw...is that the word ‘only’ does not appear in the provision.”)
In contrast, the 4th Circuit declared “we remain unpersuaded by either side.” It recites the arguments from both sides, but discounts them both by noting “it is not especially surprising that in a bill of this size...there would be one or more conflicting provisions.” Curiously, it does not mention the idea, found in the DC majority opinion, that the information reporting requirements would be useful for enforcing the Individual Mandate as well as the credits. Considering the significant overlap of attorneys in the two cases (the plaintiffs in each cases were represented by the same law firm and the same two attorneys wrote the briefs in both Courts, and many of the amici in one case also appeared and submitted briefs in the other), this suggests to me that the idea may not have originated from counsel. Or it could be that since the 4th Cir. punted on statutory context, it did not want to take up space describing all the arguments put out by all sides and amici.
c. Statutory Purpose:
The DC majority found no useful information about the statutory purpose of the ACA. It even-handedly discounted both plaintiff and government versions of the statute’s purpose. wanted to create incentives for states to create exchanges. Providing a tax subsidy only for individual coverage purchased through state-established Exchanges serves that purpose.
The DC dissent played the statutory purpose card, declaring that the purpose of Congress to increase the availability of health insurance, and the execution of that purpose through the “three-legged stool” approach was so clear as to trump the cramped interpretation of the statute. The 4th Cir. concurrence also characterized the plaintiffs’ textual argument as “not literal but cramped.”
The 4th Cir. was likewise unimpressed with either side’s arguments about statutory purpose.
The DC Cir. believed that the statutory text was clear. Accordingly, the contrary Treasury Regulation was invalid.
The 4th Cir. believed that the statutory text was unclear. Accordingly, it reviewed the regulation for reasonableness, concluding that in light of “the Act’s ambiguity, the IRS crafted a rule ensuring the credits’ broad availability and furthering the goals of the law. In the face of this permissible construction, we must defer to the IRS rule.”
Note here that while the 4th Cir. was clear that the statutory purpose was not sufficient for it to interpret the statutory text itself in Chevron Step 1, the statutory purpose was sufficient to evaluate the reasonableness of the contested regulation in Chevron Step 2.
1. The Dog That Did Not Bark.
While the six judges differed considerably on the merits of the substantive question---the validity of the Treasury Regulation---they uniformly agreed on the procedural questions of standing and remedy. The government argued that various plaintiffs did not have standing and those that did were not authorized to maintain the suit under the Administrative Procedure Act (APA) because the APA gave access to courts only when “there is no other adequate remedy in a court” for the plaintiff’s complaint. Not one judge agreed. As to standing, the plaintiffs’ argued that they lived in states with no State-established Exchange. Accordingly, without the tax credits, the financial hit to them to purchase insurance would be so great that they would qualify for the unaffordability exemption from the Individual Mandate of IRC §5000A. Under the Treasury Regulation’s extension of the tax credits to their states, however, they could afford to buy insurance according to the law, and so would be subject to the Individual Mandate penalty if they did not purchase insurance. This is a good argument for standing purposes because courts have long refused to require citizens to willfully violate a regulation or statute in order to test its validity.
What is interesting to me is the procedural argument the government did NOT make: the Anti-Injunction Act (AIA), located in IRC 7421. That statute says that “no suit for the purpose of restraining the assessment or collection of any tax shall be maintained in any court by any person, whether or not such person is the person against whom such tax was assessed.” This was a big procedural issue in National Federation of Independent Business v. Sebelius, the 2012 Supreme Court decision upholding the constitutionality of the ACA. For more on that see my article “Jesus and the Anti-Injunction Act” available here.
But the government did not even try the AIA argument here. Perhaps that is because it keeps losing this argument in many cases, most notably the recent DC Circuit opinion in Cohen v. United State, 650 F.3d 717 (DC Cir. 2011).
For example, while it used to be that tax-exempt organizations who were protesting their treatment by the IRS were routinely denied judicial relief until after the IRS had ruled on their tax exempt status---see e.g. Alexander v. "Americans United" Inc., 416 U.S. 752 (1974)---it now appears that Cohen has opened the courthouse door to such suits and taxpayers can attack any IRS procedure they believe is unfair before the IRS rules on their application. See Z St., Inc. v. Koskinen, 2014 U.S. Dist. LEXIS 71638 (D.D.C. 2014).
I am not clear on why the government did not argue the AIA. This could be simply my failure to understand that statute but it seems to me that the plaintiffs were here seeking an injunction to prevent the IRS from calculating their taxes under this regulation --- the very kind of lawsuit that the AIA was written to prevent. Did the Supreme Court’s decision on the AIA in National Federation of Independent Business (NFIB) preclude application of the statute here? I do not see why. To be sure, in NFIB the Supremes held that because Congress called the Individual Mandate penalty a “penalty” and not a “tax” the AIA could not apply to suits to restrain the assessment or collection of the Individual Mandate. However, these suits are to nullify a regulation that connects directly to what Congress has called “taxes.” That is, the whole purpose of the tax credits, as §36B(a) makes clear is to reduce “the tax imposed by this subtitle.” If so, then the AIA should apply. The AIA does not simply restrain suits to prevent the assessment or collection of “larger taxes.” It says “any tax.” Again, since I am writing this hastily and on jet-lag from an overseas trip, I am probably missing something obvious and I invite readers who have made it this far to email me their thoughts or put them in the comments.
2. Plain Language Pizza.
O.k. You KNOW this issue is going to the Supreme Court. It is a vital part of the ACA and we have a circuit split on the question of whether the statutory text is clear enough to invalidate a Treasury regulation. And who is the greatest textualist of them all? Why, Justice Scalia. And so is it only coincidence that Senior Judge Davis invokes pizza, a metaphor Scalia used to great effect in his dissenting opinion in National Cable & Telecommunications Association et al. v. Brand X Internet Services et al., 545 U.S. 967 (2005)?
In Brand X the government agency was interpreting statutory language about whether Internet Service Providers who offered broadband cable modem service were "information service" providers but were not a "telecommunications service" within the meaning of the statute and so were not subject to mandatory Title II common-carrier regulation. The statute provided that telecommunications service was given by those providers who "offered” telecommunications for a fee directly to the public. 47 U.S.C. § 153. The FCC decided that it could not separate out the information content from the transmission capability and so cable companies offering broadband Internet access did not offer the end user simply a telecommunications service, but instead used telecommunications to provide end users with cable modem service.
Scalia was incredulous and went to a pizza analogy to show the fallacy of the FCC’s logic:
If, for example, I call up a pizzeria and ask whether they offer delivery, both common sense and common usage would prevent them from answering: "No, we do not offer delivery--but if you order a pizza from us, we'll bake it for you and then bring it to your house." The logical response to this would be something on the order of, "so, you do offer delivery." But our pizza-man may continue to deny the obvious and explain, paraphrasing the FCC and the Court: "No, even though we bring the pizza to your house, we are not actually 'offering' you delivery, because the delivery that we provide to our end users is 'part and parcel' of our pizzeria-pizza-at-home service and is 'integral to its other capabilities.'" Any reasonable customer would conclude at that point that his interlocutor was either crazy or following some too-clever-by-half legal advice.
Similarly, Senior Judge Davis reaches for the pizza to reject the plaintiff’s textual argument.
Appellants’ reading is not literal; it’s cramped. *** If I ask for pizza from Pizza Hut for lunch but clarify that I would be fine with a pizza from Domino’s, and I then specify that I want ham and pepperoni on my pizza from Pizza Hut, my friend who returns from Domino’s with a ham and pepperoni pizza has still complied with a literal construction of my lunch order. That is this case: Congress specified that exchanges should be established and run by the states, but the contingency provision permits federal officials to act in place of the state when it fails to establish an exchange.
3. Whose Job Is it Anyway.
To side with the plaintiffs here is to place the burden of fixing this problem on Congress. To side with the government places the burden of fixing the problem on the IRS, as supervised by Treasury and, perhaps more loosely (under Chevron) by the courts themselves. Notably, it is the courts that will exercise the power to decide who gets to fix the problem. It’s kind of like an appointment power. This is, of course, familiar to those who study Chevron.
The DC majority frankly acknowledges the problems its ruling would create but insists that, “high as those stake are, the principle of legislative supremacy that guides us is higher still.” It gives no reasoned explanation for why that is so. Similarly, Senior Judge Edwards appears to believe that the problems to be encountered by the majority’s interpretation are so significant as to themselves be a canon of statutory construction: he repeatedly explains the necessity of the government’s interpretation for the ACA to function and declares “it is inconceivable that Congress intended to give states the power to cause the ACA to crumble.” But Judge Edwards also advances no explanation for why, in this case, the agency should get to fix the statutory drafting problem. I mean that as no criticism for I, too, am clueless on just where and why one would draw the line allocating the fix-it power to Congress or to the agency. Both institutions commit error.
One possible line of analysis here is to allocate the fix-it power to the institution that has the best capability of...well...fixing problems. Normally, that would be agency personnel because of their expertise and experience with how the words on the page actually play out in practice. The 4th Cir. opinions (both the majority and Judge Davis’ concurrence) appear to me to nudge in that direction in that they emphasize that in such a large and complex statute as the ACA, one can readily expect drafting errors and so the agency (or, here, agencies since implementation of the ACA is split) should get first go at fixing problems. That is, they do not explicitly link the fix-it allocation to the size of the problem, but instead seem to link it to the size of the statute.
Those are my afternoon thoughts on these opinions. As I said above, I welcome any comments, either to me directly at firstname.lastname@example.org, or on Paul’s comment section.