Those who oppose making affordable health insurance available to lower- and middle-income Americans are not giving up easily. Having lost their two-year long battle to have the Affordable Care Act nullified in its entirety by the federal courts, opponents have come up with a new theory that they believe will at least deprive millions of middle-income Americans of the tax credits that Congress has offered through the ACA to help make health insurance affordable.
While this theory has little chance in succeeding in the courts, and even less chance of being addressed by the courts anytime in the near future because of jurisdictional problems, it may very well convince conservative state legislators and governors to refuse to establish health insurance exchanges in their states. This would in turn lead to the establishment of federally facilitated exchanges in states that decide not to establish their own exchanges. Thus, ironically, we see the spectacle of ACA opponents fighting for increased taxes on the middle class and for a greater federal government “takeover” of health insurance markets.
The issue is this. Section 1401 of the ACA creates section 36B of the Internal Revenue Code, which provides tax credits to middle-income Americans to make health insurance affordable. In defining who is eligible for tax credits, it refers to persons “enrolled in [a qualified health plan] through an Exchange established by the State under section 1311.” ACA opponents argue that this language precludes premium tax credits being issued through the exchanges operated in the states by the federal government, (called federally facilitated exchanges), which section 1321 of the ACA authorizes where states fail to establish exchanges themselves under section 1311. If this is true, it is likely that many—perhaps most–Americans otherwise eligible for premium tax credits would not have access to premium tax credits in 2014, as it now appears that many states will, at least initially, have federally facilitated exchanges.
In a recent article, Michael Cannon of the libertarian Cato Institute and Professor Jonathan Adler of Case Western University claim that this language is not only unambiguous but also intentional, that Congress intended to punish states that refused to establish exchanges by refusing premium tax credits to their residents. Cannon and Adler further claim that final rules promulgated by the IRS making premium tax credits available through federal as well as state exchanges are unauthorized by law, and thus illegal.
These claims are simply false. The legislative history of the ACA establishes that Congress understood that premium tax credits would be available through both federal and state exchanges. The availability of tax credits through federally facilitated exchanges is further demonstrated through the language of the ACA, as amended by the Health Care and Education Reconciliation Act (HCERA), which, although it could have been clearer, supports the interpretation of the statute adopted by the IRS.
The IRS is explicitly authorized by Congress to interpret the statute and its interpretation of the law will be given deference by the courts. The existence of exchanges in every state was assumed both by the Congressional Budget Office and by both proponents and opponents of the ACA as it was being debated. Moreover, the structure and purpose of the ACA requires that state or federal exchanges offer premium tax credits in every state. Finally, as Cannon and Adler concede, employers, the only persons with standing to challenge the IRS’s interpretation of the ACA, would be barred from doing so by the Tax Anti-injunction Act, as recently interpreted by the Supreme Court, probably until 2015.
The Legislative History Shows That Congress Intended Premium Subsidies To Be Available In All Exchanges
First, the legislative history. The ACA is composed of the Senate version of the Patient Protection and Affordable Care Act, Public Law 111-148, and the Health Care and Education Reconciliation Act, Public Law 111-152. PPACA was further derived from the S 1679, the Senate Health, Education, Labor and Pensions Committee bill, and S 1796, which emerged later from the Senate Finance Committee. Each of these bills included state and federal exchanges, which were called Gateways in the HELP bill.
The HELP bill (section 142, adding section 3104 of the Public Health Services Act) created an elaborate structure under which states could either establish exchanges themselves (“establishing states”), request the federal government to establish an exchange in the states (“participating states”), or fail to do either, in which case four years after the enactment of the statue the federal government would create a fallback exchange in the state. Premium tax credits were available in establishing and participating states, but would only be available through the federal fallback exchanges in states that complied with the employer responsibility provisions for state and local employees. In other words, the states were threatened with loss of premium tax credits, not for failing to establish exchanges but for not complying with the employer responsibility provisions for their employees.
The Finance Committee bill did not use this elaborate structure. In fact, the rules it creates are very similar to the final bill. It creates section 2235 of the Social Security Act, which provides that states “shall” establish an exchange, and sets out the duties of the exchange. Section 2225(b) provides, in language very similar to current ACA section 1321, that HHS shall contract with a nongovernmental entity to “establish and operate” an exchange in states that fail to create one within 24 months. In a number of places, including the precursor of the current premium tax credit provision, the bill refers to exchanges “established by the state,” but nowhere does it provide, as did the HELP bill, that premium tax credits would not be available in any of the nongovernmental exchanges created by the federal government.
The provisions of the current ACA addressing this issue are taken largely from the Finance Committee bill, which makes sense because the Finance Committee has jurisdiction over tax matters. The punitive provisions of the HELP bill were abandoned. Section 1311 provides that states “shall” establish an exchange by January 1, 2014. The Constitution, however, prohibits the federal government from literally requiring states to establish exchanges, so section 1321 provides that “the Secretary shall (directly or through agreement with a not-for-profit entity) establish and operate such Exchange within the State.”
By “such Exchange” Congress meant the “required exchange” mandated by section 1311. Thus when several subsequent sections refer to “an Exchange established by the State under section 1311,” including the provisions of Internal Revenue Code section 36B on which Cannon and Adler rest their argument (as well as the provision of section 2001 that prohibits states from reducing Medicaid eligibility until an exchange “established by the State under section 1311” is operational), they are referring both to state exchanges and to “such exchanges” established within states by the Secretary.
Although the PPACA could be clearer on this matter, the issue is clarified by section 1004 of the later-adopted HCERA. This section directs both 1311 and 1321 exchanges to provide the Secretary and individuals eligible for tax credits with information concerning the amount of advance premium tax credits, as well as information necessary to determine eligibility for premium tax credits and whether an individual has received excess tax credits. As a later-adopted statute, HCERA would take precedence over PPACA if there were a contradiction.
Moreover, since the adoption of HCERA was necessary to secure House adoption of the Senate bill, it is doubly important that the provisions of HCERA be taken seriously. The House bill contained only a federal exchange, through which all federal insurance subsidies were offered. But this HCERA provision merely clarifies the most plausible reading of the PPACA — that premium tax credits are available through federally facilitated and non-federally facilitated state exchanges. (Cannon and Adler admit the existence of this provision but simply say it is meaningless, as 1321 exchanges cannot authorize premium tax credits, thus violating another canon of statutory construction, that every provision of a congressional enactment should be given effect.)
The Senate debated the ACA extensively during November and December 2009. The version of the Act they were considering included both state and federally facilitated Exchanges. Throughout the debate, Senators assumed that tax credits would be available in all 50 states. Thus Senator Bingaman stated on December 4, 2009, that the ACA “includes creation of a new health insurance exchange in each State which will provide Americans a centralized source of meaningful private insurance as well as refundable premium tax credits to ensure that coverage is affordable.” 155 Cong. Rec. S12358. Senator Johnson stated on December 17, “The legislation will also form health insurance exchanges in every State,” which will “provide tax credits to significantly reduce the cost of purchasing that [insurance] coverage.” 155 Cong. Rec. S13375.
If Congress had meant to limit premium subsidies to state-established exchanges, as an incentive to States, one would have expected the Finance Committee report on S. 1796 (Senate Report 111-89) to have mentioned this, and for at least one Senator to have pointed this out during the debate in November and December 2009.
CBO’s assumptions. Most importantly, the Congressional Budget Office (together with the Joint Committee on Taxation) provided Congress on November 30, 2009, an analysis of the impact of the legislation on premiums that assumed that premium tax credits would be available in all states, making no distinction between federal and state exchanges. Over the next few days, this analysis was discussed by Republican Senators Grassley (155 Cong. Rec. S12107, 12/2/09), Enzi (155 Cong. Rec. S12378, 12/4/09), and Coburn (155 Cong. Rec. S13687). None raised what Cannon and Adler see as an obvious point, that the CBO analysis was flawed because it failed to recognize that premium tax credits would not be available though federally facilitated (sec. 1321) exchanges.
In fact, the CBO repeatedly provided cost estimates of the ACA and HCERA in late 2009 and early 2010, but never suggested that premium tax credits might be reduced if states failed to establish exchanges. As Professor Abbe Gluck notes in a recent blog post (and forthcoming article), Senators often don’t listen to each other, but they all listen to the CBO, which assumed that premium tax credits would be available to all Americans in all states.
Cannon and Adler claim, however, to have found a smoking gun, a colloquy between Senators Baucus and Ensign during the Finance Committee debate on the bill, in which, they claim, Senator Baucus admits that premium tax credits could not be made available through federal exchanges. In fact, the colloquy, which Canon and Adler edit beyond recognition, had nothing to do with federally facilitated exchanges, but rather with whether the Finance Committee had jurisdiction over malpractice reform legislation that Ensign wanted to attach to the bill. The fact that this is the only fig leaf of legislative debate that Cannon and Adler can find to support their theory demonstrates how implausible it is.
Cannon And Adler’s Flawed Argument On Congressional Purpose
Cannon and Adler also argue that Congress prohibited the federal exchanges from offering premium tax credits as a way of encouraging the states to adopt exchanges. It is clear that the federal government favored state exchanges and offered generous grants to the states, which to date have totaled nearly $850 million dollars with more on the way. States that fail to establish exchanges will also lose some control of their insurance markets. But Congress did not try to “coerce” states to create state exchanges by threatening their citizens with loss of billions of dollars of premium tax credits. Indeed, under the Supreme Court’s recent Medicaid decision, such coercion might have been suspect.
Further, the entire structure of the ACA depends on premium tax credits being available in all states. The ACA’s guaranteed issue and community rating requirements apply to insurers in all states, regardless of whether they have federal or state exchanges. So do the ACA’s risk mitigation programs. So does the ACA’s individual mandate. The premium tax credits are intended to bring millions of new participants into insurance markets, and if they are not available in many states, the nature of insurance markets will change dramatically, increasing the risk for insurers and decreasing availability to middle-income Americans. If this was the intent of Congress, it surely would have made it far more evident.
The IRS’s Interpretation Of The ACA Is Consistent With The Statute And Should Be Given Deference
The ACA is not a model of clear drafting, as I have noted earlier. In addition to obvious drafting errors conceded by Cannon and Adler, the ACA contains three sections with the same number (1563) and amends an existing provision of the Public Health Services Act inconsistently twice within the scope of a few pages. The Senate bill was not supposed to be the final law. Only the Senate election in Massachusetts in early 2010 made a conference committee bill that would have reconciled the House and Senate versions and cleaned up the current bill impossible.
I agree with Cannon and Adler that the courts are unlikely to find the “established by the state” language a “scrivener’s error.” But the courts will interpret the language in the context of the ACA’s structure and purpose, in light of the statute’s legislative history, and they will put great weight on the HCERA amendment; they will find that federally facilitated exchanges can in fact issue premium tax credits.
The courts are also likely to put great weight on the interpretation of the provision by the IRS, which is explicitly given authority under the statute to adopt regulations interpreting the provision. Cannon and Adler are correct that under the governing “Chevron doctrine,” courts will not defer to the interpretation of an agency if “Congress has spoken to the precise question at issue.” However, where a statute is ambiguous, “the question for the court is whether the agency’s answer is based on a permissible construction of the statute.” In situations like this, where there is ambiguity, “legislative regulations are given controlling weight unless they are arbitrary, capricious, or manifestly contrary to the statute.” As noted above, the interpretation of the ACA by the IRS is completely consistent with, rather than “manifestly contrary” to, the statute and thus deserves judicial deference.
Premium Tax Credits In Federally Facilitated Exchanges Could Not Be Challenged Until 2015
Finally, there is the matter of jurisdiction. Under our Constitution, the federal courts only have jurisdiction over “cases or controversies,” that is disputes in which someone has actually been injured. As Cannon and Adler admit, the courts will not hear cases brought by someone who simply objects to a law. Indeed, as the Fourth Circuit held in the Virginia challenge to the ACA (relying on long line of cases) even states cannot sue to challenge a law to which they object unless the state is concretely injured by the law.
The only viable challengers to the law are employers who may in the future have to pay an exaction because they fail to offer their employees insurance (or affordable or adequate insurance) and their employees consequently end up receiving tax credits in the federal exchanges. But the employer responsibility provision of the ACA expressly refers to the exaction as a tax, and under the Supreme Court’s recent ACA decision, challenges to it would be barred by the Tax Anti-Injunction Act until it was actually assessed, which will probably not be until sometime in 2015. Of course, by then possibly millions of Americans will have received premium tax credits from federally facilitated exchanges, and the political risk of a lawsuit depriving them of this federal tax advantage will be high.
Which brings us back to the original question: Why are ACA opponents trying to increase the taxes of middle-income American taxpayers? Or more importantly, what is it about extending the benefits of our health care system to millions of uninsured Americans that so troubles opponents of the ACA?
I thank Mark Regan of the Disability Law Center of Alaska for his insight into the legislative history of this provision and Professor Abbe Gluck for reading an earlier draft of this post.