Introduction

Federalism scholarship and doctrine have long viewed the states as monoliths.1 It is New York that is commandeered,2 Florida’s sovereign immunity that is violated,3 and Indiana that is coerced4—not officials, agencies, or political parties within the state, but the state qua state. We assume that the federal government does not see the politically contested underbellies of the states, but instead neutrally waits for the conflicts between a state’s governor and legislature, agencies and lawmakers to be resolved before listening for a unified voice. But what happens when the federal government not only sees, but intervenes in, these intrastate political contests? Elsewhere I have argued that federalism theory and doctrine struggle to conceptually accommodate this kind of intervention.5 And yet it remains pervasive. The federal government often disaggregates the states when seeking their consent to join cooperative programs.6 It does this by, among other things, designating a particular state official who can speak for the state.7 The federal government also disaggregates the states in the implementation of cooperative programs, seeking out state officials who are willing to lend a helping hand to the federal effort even as other state actors resist.

This kind of disaggregated collaboration has important theoretical implications because it complicates the paradigm of cooperative federalism as engagement between unified sovereigns. But in the Affordable Care Act (ACA)—one of the largest and most complex cooperative programs ever established—we also see a more immediate and practical significance of the practice of disaggregation. Here I focus in particular on the way that disaggregation shaped the creation of the health insurance exchanges that form the centerpiece of the Act.

The ACA provides that “[e]ach State shall . . . establish” a health insurance exchange.8 The Department of Health and Human Services (HHS) calls those exchanges “state-based exchanges.” But the Act also gives the states an escape hatch. If a state declines to establish a state-based exchange, it is deemed “not an electing State” and the statute directs HHS to “establish and operate” a federal exchange “within the state.”9

After the ACA’s passage, HHS moved quickly to issue regulations and guidance to determine howthe states could establish such exchanges.10 Specifically, HHS required a state’s governor to sign any document indicating the state’s interest in establishing a state-based exchange.11 HHS thus disaggregated the state into its component parts by interpreting the word “State” in the ACA to mean “state governor,” thereby letting the governor determine the ultimate fate of the state’s exchange.12

But as far more governors than expected declined to establish state-based exchanges, HHS sought other ways to keep the states engaged in the exchange project. In particular, it became clear that even in states with governors who staunchly opposed state-based exchanges, state insurance commissioners remained interested in state involvement with the exchanges. To work with those enthusiastic state collaborators, HHS innovated a new model of exchange—the “hybrid” exchange.13 The hybrid exchanges are technically variations on the federal exchanges, but they devolve significant authority back to state insurance commissioners.14 HHS thus disaggregated the states again by identifying officials within the states who favored the exchanges and developing formal pathways for them to wield authority and influence within the federal exchanges. Fourteen states have adopted hybrid exchanges.15 That’s just over forty percent of the thirty-four states that declined to operate state-based exchanges.16

My goal in this short Essay is to document and explore the significance of these acts of disaggregation. By explaining how the various exchange models came to be, Part I shows how HHS’s shift from collaborating with governors to collaborating with insurance commissioners richly refutes the idea that the federal government is blind to intrastate politics. This case study therefore challenges some of the long-standing theoretical assumptions animating cooperative federalism. But HHS’s variety of exchange models may have more pressing implications in light of the ongoing litigation in King v. Burwell,17 which will decide the fate of the exchange scheme later this month. As I will show in Part II, that case reveals the difficulty of mapping the practice of disaggregation onto the text of federal statutes and the principles of federalism doctrine. It shows, in other words, some of the risks and possibilities of disaggregation.

I. the creation of state-based and hybrid exchanges

After the ACA’s passage, HHS immediately began helping the states establish exchanges.18 But, almost from the beginning, it was clear that significantly fewer states than expected would elect to operate their own state-based exchanges. California was the only state to establish an exchange in 2010, the year the ACA was enacted.19 And by July 2011, only nine more states had expressed their intention to set up an exchange.20 These numbers were ominous to many in the administration, and for good reason.21 When the law passed, there was a widespread expectation that most of states would run their own exchanges.22 This expectation was so strong, in fact, that it was embedded in the funding Congress appropriated to support exchange development. While the Act allocated virtually unlimited funds to support the establishment of state exchanges,23 it squeezed the budget for federal exchanges, appropriating only one billion dollars to HHS to get the federal exchange up and running.24 HHS, therefore, had both a political and a financial incentive to encourage as much state participation in the administration of the exchanges as possible.

But the states continued to reject the option of running their own exchanges, largely because of opposition from state governors. This political fact was very much in HHS’s sights. A particularly striking battle among Mississippi’s elected officials played out in letters between those state officials and HHS leadership. The confrontation pitted the elected Mississippi Commissioner of Insurance, Mike Chaney—who favored a state-based exchange—against Mississippi’s Governor, Phil Bryant, who advocated total federal control over the state’s insurance exchange. On November 12, 2012, Chaney sent a letter to Gary Cohen, HHS’s primary contact person for state exchange decisions, declaring Mississippi’s “intent to implement and operate a State-based Exchange.”25 Two weeks later, Governor Bryant wrote to HHS Secretary Kathleen Sebelius, explaining that he felt “compelled to notify her of his ‘complete disagreement with this move.’”26 In a candid acknowledgement that this was a contested issue in the state, he told Secretary Sebelius that he was “exploring my options” to block a state-based Mississippi exchange. That turned out to be unnecessary because HHS rejected Commissioner Chaney’s exchange declaration as inconsistent with its requirement that declarations regarding state-based exchanges be signed by the state’s governor.27

A. Hybrid Model One: State Partnership Exchanges (SPE)

In early 2012, as battles like Mississippi’s bloomed across the country,28 HHS floated the idea of developing more flexible ways for eager state officials to participate in exchange administration. HHS’s initial effort would be its first hybrid model: the “State Partnership Exchange” (SPE). These exchanges delegate several important regulatory powers back to participating states.29 The guidance establishing them notes that SPEs give states the opportunity to “assume primary responsibility for many of the functions of the Federally-facilitated Exchange permanently or as they work towards running a State-based Exchange.”30 One of the biggest selling points of the SPE is that “states can continue to serve as the primary points of contact for issuers and consumers.”31 By allowing state officials to “provide input and guidance” into the otherwise federally facilitated exchange, the SPE model allows these officials to “take ownership over significant components of the [exchange’s] operation.”32 Finally, the guidance commits that even in “areas where the law prohibits HHS from completely delegating responsibility to a state, HHS will work with states to agree upon processes that maximize the probability that HHS will accept state recommendations without the need for duplicative reviews from HHS.”33 By establishing an SPE, in other words, states could retain important responsibilities while avoiding some of the political and financial consequences of running their own separate exchanges.

The SPE guidance document, moreover, was unmistakably written to appeal to state insurance commissioners. It noted that “State Departments of Insurance (DOIs) have a longstanding regulatory role with the health insurance issuers” and noted HHS’s “beli[ef]” in the importance of “preserving the DOI’s traditional roles and responsibilities in the insurance market.”34

To establish an SPE, the state needed to submit a “blueprint” explaining how the state would meet its hybrid exchange obligations.35Seven states agreed to establish SPEs by the deadline.36 But there was a snag: the blueprint still required the signature of the state’s governor. Although the model lowered the political stakes for governors to support HHS’s exchange effort by creating a more limited form of participation, it did not provide the kind of flexibility that state insurance commissioners like Mike Chaney needed to independently partner with HHS.37

B. Hybrid Model Two: Marketplace Plan Management Exchanges (MPM)

To accommodate insurance commissioners in states in which the governor declined to establish a state-based exchange or SPE, HHS developed a second hybrid exchange: the “Marketplace Plan Management” (MPM) exchange. In states that pursue an MPM exchange, state insurance agencies are responsible for providing recommendations regarding which “health plans meet QHP certification requirements.”38 HHS then commits to “rely” on these recommendations, “[a]ssuming a state continues to act in accordance with its attestations.”39 Thus, although the exchange was technically federally managed, state officials retain a meaningful role.

To pursue the MPM model, the state’s insurance commissioner need only attest to HHS that he or she had the authority under state law to perform the relevant oversight functions.40 Thus, a state insurance commissioner can initiate an MPM exchange without submitting a blueprint signed by the governor.

HHS used the MPM option to engage state insurance commissioners who were unable to mobilize support for state exchanges or SPEs. In 2012, for instance, when HHS replied to Mississippi Commissioner Mike Chaney, it emphasized that although it was forced to reject his declaration for a state-based exchange “because of the Mississippi Governor’s stated intent to oppose implementation” of the exchange, it was able to offer him an “additional opportunity for Mississippi’s Insurance Department” to take part in an MPM exchange.41

Other state insurance commissioners took advantage of the MPM model to bypass political disunity within the state. Sandy Praeger, the Kansas Commissioner of Insurance, was the first to sign up. In another surprisingly candid letter to HHS, she gave a full overview of the state’s exchange-related political climate. She first noted that “[w]hen it became clear that neither Governor Sam Brownback nor the majority of the members of the Kansas Legislature supported the development of a state-based exchange, the Kansas Insurance Department (KID) . . . had hoped that Kansas might be able to . . . [participate in] a partnership” exchange.42 But, she wrote, although “KID staff spent countless hours preparing the Blueprint document required to accompany a partnership declaration letter,” the agency has “now determined that there is no political support for a partnership arrangement and we are unable to secure the governor’s letter of support.”43 Accordingly, Commissioner Praeger concluded, HHS should accept “this letter as our formal offer and request” that her department would be able to take advantage of the MPM model.44

To date, seven states have elected to participate in a Marketplace Plan Management exchange.45 Tellingly, all but one of the letters expressing the state’s intention to participate came from the state’s insurance commissioner.46 Moreover, several insurance commissioners who opted for the MPM exchanges clarified that they were emphatically not embarking on formal state-based exchanges or SPE hybrid exchanges.47 Perhaps this is because six of the seven states participating in the MPM model were also led by governors who had publically announced their opposition to the establishment of a state exchange.48 Disaggregation, therefore, served its intended purpose—it allowed HHS to identify potential state collaborators and established pathways for their participation in the federal exchange project even in the face of political opposition within their states.

II. implications for the future of the exchanges

In implementing the exchange provision of the ACA, HHS took exactly the kind of step to which federalism doctrine and the theory remain blind. It disaggregated the states—first by identifying governors as the consent agent for the state exchanges, and then by developing ways to collaborate directly with state insurance commissioners who favored the exchange project even when their governors did not.

The ongoing litigation in King v. Burwell—which is awaiting a decision from the Supreme Court as this goes to press—reveals some of the ways that disaggregation, and the various exchange models it created, could complicate the interpretation of the ACA.

That litigation seizes on a provision in the ACA that makes federal health insurance subsidies—which fuel the exchange scheme by reducing the cost of insurance for exchange users—available only on an “Exchange established by the State under section 1311” of the Act.49 The government argues that “Exchange established by the State” is a term of art used to describe all exchanges, state and federal. The challengers argue that, read literally, the word “State” must be understood to make subsidies available only to consumers who purchase insurance on state-based exchanges.50

But as Part I highlights, whether or not the word “State” appears unambiguous on the face of the statute, what it means for an exchange to have a “State” character in practice is far from clear. Which state officials and which state actions count as the “State,” and who decides? The statute doesn’t say. First, consider the fifteen state-based exchanges. It would be a radical simplification to describe these exchanges as the product of a decision made by the “State” without saying something more about how that decision was made. HHS’s practice of disaggregation suggests that it is most accurate to describe the state exchanges as exchanges “established by the [governor of the] State under section 1311.”

It may be reasonable to attribute the governor’s decision to the state as a whole—as, for instance, where the governor is authorized under state law to make such a decision. But it is hardly clear that the governor is so authorized in every state. In many states, there are indicia that the insurance commissioner has significant autonomy over the state’s insurance policy. Take, for instance, states in which the insurance commissioner is elected in her own right, or is appointed for a fixed term and cannot be removed by the governor.51 In those states, HHS’s governor-centric consent procedure may place the wrong state official—at least according to state law—in the driver’s seat. If that’s so, referring to all exchanges established by the state’s governor as “State” exchanges—and, as a corollary, calling any state in which the governor has declined to set up an exchange “not an electing State”52—presumes a very particular definition of “State.”

For similar reasons, it is somewhat simplistic to describe the SPE and MPM exchanges as “federal” exchanges. It’s true that HHS established the hybrid models using its statutory authority to create exchanges in states whose governors declined to establish a state-based exchange.53 But the political character, authority structures, and functional operations of these hybrid exchanges are not entirely federal; indeed, they are shaped in meaningful ways by the judgments of state decision-makers. In the hybrid exchanges, many of the functions enumerated in section 1311 of the Act are performed by or in collaboration with state insurance commissioners. The hybrid exchanges are, in at least some ways, “established by the [insurance commissioner of the] State under section 1311.”

In both cases, HHS’s disaggregation illustrates that there are deeper challenges in defining a state than simply reading the words on the page. Understanding the dynamics of intrastate politics—and acknowledging the disaggregation that made the state and federal exchanges what they are—should complicate any reasoning in King that relies on a strict division between the “State” and the “federal.” For that decision must eventually reckon with the lived federalism born of the statute—the ways that state officials exercise diverse forms of power within the exchanges.

Indeed, this lived federalism could gain strategic importance in the aftermath of King. If the Supreme Court interprets the Act to make subsidies available only on the fifteen state-based exchanges, HHS could take an additional disaggregative step to help insurance commissioners play a more significant role in guiding their states toward such exchanges. It could do this, for instance, by deeming the state insurance commissioner able, under some circumstances, to speak for the state in electing to create a state-based exchange. Of course, HHS would not want to (and likely could not) shift the decision about whether to establish an exchange wholesale from skeptical governors to supportive insurance commissioners. Doing so would risk destabilizing the state-based exchanges that governors have previously established. But HHS could replace its governor-centric approach with an approach that is more deferential to how the states allocate decision-making power over insurance issues. The agency, for instance, could let each state decide which state official is authorized to make the exchange decision for itself.54 And in states that give significant power to elected insurance commissioners, such a deferential regime might smooth the way for a transition from “federal” and “hybrid exchanges” to “state-based exchanges.”

Still, this move is not without risks, in part because we have no doctrine to help us evaluate the legitimacy of disaggregation. In many respects, HHS is flying blind. One intuitive starting point would be to use the standard tools of administrative law to discern whether Congress authorized HHS to collaborate with individual state officials—whether they be governors in the first instance or insurance commissioners later on—in the way the agency has. We might ask whether Congress envisioned HHS working with the state as a unified entity or with a range of individual state actors. This approach is very possible: Abbe Gluck has shown that Congress often embeds novel forms of federal-state collaboration “inside of federal statutes.”55 And there is a growing trend of challenging agency action as inconsistent with a statute’s internal federalism logic. Perhaps, then, the question is whether the ACA itself endorses disaggregation and under what conditions.

But there might also be broader constitutional principles governing disaggregation. Some might see disaggregation running contrary to one of the fundamental premises of federalism—that federalism is a system of layered governments, not of layered officials.

Whether there are administrative or constitutional principles that could affect an agency’s ability to disaggregate the states is a question beyond the scope of this Essay, but it is a question that may soon call for an answer.

Bridget A. Fahey is a member of the Yale Law School J.D. Class of 2014. She thanks Abbe Gluck, Alex Hemmer, Matt Shahabian, and Reynold Strossen for helpful comments on previous drafts.

Preferred Citation: Bridget A. Fahey, Health Care Exchanges and the Disaggregation of States in the Implementation of the Affordable Care Act, 125 Yale L.J. F. 56 (2015), http://www.yalelawjournal.org/forum/health-care-exchanges-and-the-disaggregation-of-states-in-the-implementation-of-the-affordable-care-act.