Photo by Ross Helen/Thinkstock

In March, the Supreme Court heard oral arguments in King v. Burwell, the case that will decide whether the language of the Affordable Care Act (ACA) allows only those who purchase health insurance through state-established exchanges—not federal health exchanges—to qualify for federal subsidies. Justices Sonia Sotomayor and Anthony Kennedy suggested that they may be forced to allow the subsidies for federal exchanges because limiting subsidies to state exchanges might unconstitutionally intrude on the federal-state relationship by coercing states into forming their own health-insurance exchanges. This federalism argument, however, is based on speculation leading to flawed legal reasoning. It shouldn’t determine the outcome the case.

The ACA’s statutory language seems to limit federal subsidies to people enrolled “through an Exchange established by the State under section 1311,” the ACA section directing states to establish health exchanges. The statute makes no mention of subsidies for enrollees on the federal exchanges authorized under a different ACA section, 1321. The plaintiffs in the case argue that this is no accident: Congress, they maintain, intended to encourage states to create their own exchanges by offering subsidies only to state-created exchanges.

Justice Sotomayor asked if interpreting the ACA to disallow federal-exchange subsidies would “intrude on the federal-state relationship, because then the states are going to be coerced into establishing their own exchanges.” Justice Kennedy, widely viewed as the Court’s swing vote, amplified the argument, suggesting that under “the standard of constitutional avoidance,” the plaintiffs’ interpretation might be prohibited because it could lead to unconstitutional coercion. “States are being told either create your own exchange, or we’ll send your insurance market into a death spiral,” he said.

The justices suggest that a phenomenon known as “adverse selection” would occur. Disallowing federal-exchange subsidies will make insurance much less affordable for the 87 percent of exchange enrollees currently receiving subsidies. These people would no longer be required to buy insurance, since the ACA’s individual mandate only applies to individuals with affordable insurance options. Further, the ACA’s imposition of “community rating,” requiring the same premium for all individuals in a given plan with only minimal adjustments for their risk characteristics, and “guaranteed issue” of insurance regardless of the enrollee’s health, means that the old and unhealthy will continue to buy coverage but the young and healthy will forego it. The predominance of older, sicker enrollees will then increase the average cost for individuals remaining in the non-group insurance market, both on and off the exchanges, resulting in increased premiums. If this adverse selection leads to a self-reinforcing cycle of rising premiums, additional exodus of healthy patients from the market, additional premium increases, and eventual market collapse, it becomes the “death spiral” Justice Kennedy referred to.

But would a “death spiral” and unconstitutional coercion actually occur? The Urban Institute predicts that discontinuing federal-exchange subsidies would result in premium increases of 35 percent and enrollment declines of 69 percent in the individual health-insurance market. The Rand Corporation has made similar predictions. But these forecasts are inconsistent with studies of earlier state insurance-market “reforms” that created market conditions similar to those that would result if the plaintiffs in King prevail.

In the 1990s, a few states imposed community rating and guaranteed issue in their health-insurance markets, as the ACA has done, but without providing any subsidies—the same arrangement that will result on the federal exchanges from a plaintiffs’ victory in King. Though many analysts predicted that these states would experience an adverse-selection death spiral, a National Bureau of Economic Research (NBER) paper found otherwise. Comparing New York, which has both community rating and guaranteed issue, with neighboring states, the NBER researchers “found no evidence for the conventional wisdom that the imposition of pure community rating tends to an adverse selection death spiral.” Similarly, a later NBER paper compared states with community rating and guaranteed issue with states that have no such regulations. They found a small increase in the number of uninsured but did “not observe a strong positive relationship between risk status and the likelihood of being covered that would be consistent with so-called ‘death spirals.’”

If no death spiral happened in New York, which had pure community rating with no adjustment for any type of individual risk factors, then such a spiral would be unlikely to occur under the ACA, which permits risk adjustment by age bands and by habits such as tobacco use. Since some riskier patients already pay higher premiums under the ACA, adverse selection will be mitigated.

And if no death spiral occurred in state markets with enrollees of all income types, little reason exists to believe that a death spiral would follow a plaintiffs’ victory in King. In a post-King setting, the trigger for adverse selection would presumably be the removal of subsidies for people who would otherwise be unable to afford insurance. Those few who manage to come up with premium money will quickly reach a point where they cannot afford higher premiums. The market will settle at an equilibrium-risk profile and premium level where no additional patients, high- or low-risk, are willing or able to buy insurance. The number of uninsured will increase, but adverse selection, to the extent that it occurs, will not spiral into market failure.

Predictions of a death spiral are also likely inaccurate because the market is not static; it is already adjusting to the prospect of a ruling that favors the plaintiffs. Insurers’ main interest is retaining enrollees, and they are making contingency plans to keep insurance affordable and companies solvent without subsidies. The NBER reports found that community rating sparked a structural shift in the small-group and individual markets, away from indemnity insurance toward less costly, managed-care plans. The ACA has already produced a similar shift toward lower-cost plans with narrow provider networks and no out-of-network benefits. A plaintiffs’ victory would accelerate this trend.

Questionable forecasts of complex and rapidly changing market dynamics should not influence the Supreme Court’s deliberations. But even if adverse selection occurred, it would not represent unconstitutional coercion. Congress can induce states to accept conditional grants, as it has in other programs, but it may not compel or coerce them into participating. NFIB v. Sebelius, the 2012 challenge to the ACA, is the only Supreme Court case to invalidate a congressional spending program as unconstitutionally coercive. The ACA threatened to withhold funding for states’ pre-existing Medicaid programs—amounts that exceeded 20 percent of most states’ budgets—unless they expanded those programs. The Court concluded that “the financial ‘inducement’ Congress has chosen is much more than ‘relatively mild encouragement’—it is a gun to the head . . . [that] leaves the States with no real option but to acquiesce in the Medicaid expansion.” States had to have a real choice whether to expand their Medicaid programs without being coerced by a loss of preexisting Medicaid funds.

But King is different than NFIB. States that decline to establish a state-based exchange don’t risk losing a large amount of funding for an established program and crippling their budgets. At most, they risk destabilizing coverage for a small percentage of their citizens. Most Americans get health coverage through employer, large-group, or government insurance. The individual and small-group markets cover 10 percent or less of the insured in most states. Moreover, the supposed burden would not, as in NFIB, fall directly on the states as sovereign entities, but rather on their citizens, who lose their subsidies, and on insurers, who might financially suffer as a result. Federal statutes that only burden states indirectly by imposing costs on their citizens and businesses are permitted.

Unlike the situation in NFIB, any burdens on states could be mitigated without invoking constitutional avoidance. Congress could moderate adverse selection by liberalizing the ACA’s community-rating standards. Prior to the ACA, most states let insurers charge older, sicker patients up to five times as much as younger patients. The ACA limited this adjustment to three times. Reverting to the previous limit would make it more affordable for younger, healthier people to stay in the insurance market without subsidies. Congress could also relax the ACA’s generous minimum-essential-benefits requirements to make insurance more affordable. It would be more consistent with traditional notions of federalism for Congress to scrap the ACA benefits requirements and restore states’ historical role in setting affordable insurance standards for their citizens.

The justices should also consider that disallowing federal-exchange subsidies may be more respectful of federalism than upholding the government’s position. The ACA’s “employer mandate” is only triggered when an employee receives a subsidy on an exchange. The ACA’s “individual mandate” only applies to people for whom insurance costs less than 8 percent of household income, a group that was markedly expanded by the provision of subsidies on federal exchanges. States should be allowed to decide if they want their citizens to benefit from a federal insurance subsidy or to opt for a federal exchange, thereby protecting their businesses from the employer mandate and their citizens from the individual mandate. Seven states filed an amicus brief in King stating that they were aware that subsidies were limited to state exchanges but chose to avoid “the enormous burden associated with creating and operating a program of such complexity and the hardships establishing an Exchange would impose on employers.”

Disallowing federal subsidies will not, as the conventional wisdom claims, lead to unavoidable insurance-market collapse or unconstitutional coercion of the states. The justices should decide King v. Burwell by interpreting the statute as written and not be influenced by bogeymen. Ultimately, it is Congress, not the Supreme Court, that should determine the best way to provide health insurance.


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next