As President Obama’s signature health care law comes before the Supreme Court, most of the constitutional commentary has focused on whether Congress has the power to require individuals to purchase health insurance. Relatively little attention has been given to a more complicated and potentially more far-reaching issue: If the Supreme Court strikes down the individual mandate, what should it do with the rest of the law?
The Court has scheduled ninety minutes of oral argument for the “severability” issue, and the possible outcomes are legion. The act could be tossed out in its entirety, as held by the trial court; or the remainder of the act could be allowed to stand as is, without the individual mandate, as held in the Eleventh Circuit Court of Appeals. But the Court could also take a middle road, stripping the act of some combination of core provisions—probably those focused on health insurance reforms, Medicaid expansion, and the exchange related subsidies—and allowing the rest of the law to stand.
The stakes could hardly be greater: If the Court takes the wrong approach on severability, it could leave the nation with a stripped-down law that did not command, and never could have commanded, congressional support at the time of passage. That judicially constructed law, moreover, could well leave us with an underfunded health care system that could unravel in spectacular fashion, pushing the private health insurance industry to the brink of collapse.
Well-established Supreme Court precedent holds that when one part of a law is held unconstitutional, the remainder is normally upheld if (1) it will be “fully operative” in the way Congress intended1, unless (2) it is evident that Congress would not have enacted the remainder without the invalid part. This two-part test should not be watered down to allow the rest of the law to survive merely because it has some legal effect after severance; courts should not create entirely new laws that Congress never passed and were never presented to the President.
The inquiry should therefore turn on which parts of the ACA operate independently from the mandate and which are interwoven with it. The two lower courts directed woefully little analysis to this decisive question. The trial court held the mandate unconstitutional and struck down the entirety of the ACA along with it. The 11th Circuit Court of Appeals agreed that the mandate was unconstitutional, but reversed on severability, holding that the rest of the Act could function without the mandate. Neither court conducted the thorough analysis of statutory interactions that severability doctrine seems to require for each separate portion of the act.
In our Supreme Court amicus curiae brief, we focused exclusively on the heart of the law—namely the insurance “reforms”, Medicaid expansion, and premium subsidies (Titles I and II of the ACA). The better the Court understands the vital interrelation of those provisions with the mandate in the original legislative design, the clearer it will be that these core provisions are wholly interwoven with the mandate and must be struck down with it. Without the mandate, these other provisions will not function as Congress intended and would never have passed.
In fact, implementing those other core provisions without the mandate is likely to result in a financial meltdown. The reason lies in the unyielding economics of health insurance, which the policy reforms at the heart of ACA sought to reshape.
The ACA’s central goal was to provide universal insurance coverage for a full range of medical benefits. The crux of the law establishes the “guaranteed issue” of health insurance for all applicants, which requires all private plans to take all comers up to their capacity.
By itself, guaranteed issue flips the economics of insurance on its head. The standard view of insurance recognizes that misconduct by either the insurer or the insured could imperil the long-term success of the contract. On the insurer’s side, there is the risk of misappropriation of premium dollars before payout, for which financial oversight is required. On the insured’s side, the key risks are moral hazard and adverse selection. The former refers to the possibility that an insured will engage in riskier conduct precisely because it is insured. The latter refers to the possibility that an insured will use private information about his or her own health care status in choosing what kind of insurance to acquire and when to acquire it.
Well-drafted private policies must guard against these risks by the insured. To combat moral hazard, most policies deny coverage for various types of losses, including participation in dangerous activities. To combat adverse selection, most companies require individuals to take examinations that determine their health care status, and to purchase coverage in advance and hold it for a definite period of time.
The ACA’s guaranteed issue provision does nothing to encourage healthy people to pay for “insurance”; on the contrary, it is a powerful incentive for people to wait until they are sick to buy it. That behavior leads to the “adverse selection spiral” under which people buy health care insurance only when they know it is worth more to them than what they pay for it. As healthy people leave the rolls, the per-unit cost of insuring the remaining (riskier) insured rises, which pushes premiums up, which in turn drives more healthy people off the rolls. In the end, the only people who enroll are those with known medical conditions, such that premiums approach the actual cost of health care, and the insurance industry collapses.
The concerns here are not just theoretical. When the Clinton-era health reform efforts failed, eight states passed “guaranteed issue” reforms in the individual health insurance market (that 10 percent of the insurance market outside the federally regulated market for employer-provided group health plans).2 These state plans also included “community rating” compression, by lowering the rates for the elderly below cost, and raising the rates of the young above cost, thereby consciously creating a cross-subsidy system. The state plans allowed insurers to exclude applicants with preexisting conditions. But even that concession to market pressures did not prevent the individual health care market in these states from imploding because of excessive adverse selection.
The casualty list is impressive. New Hampshire and Kentucky both adopted guaranteed-issue reforms in 1994. Kentucky repealed guaranteed-issue in 1998, and backed away from minimum-coverage in 2004; New Hampshire repealed its reforms in 2002. Vermont, Washington, and New Jersey passed similar reforms with similar results: adverse selection set in, dramatically increasing premiums and the number of uninsured. By 1999, Washington allowed insurers to exclude the sickest people, while Vermont and New Jersey retreated to allowing very high deductibles. Maine and New York struggled for years to fix the disaster created by their health reforms, without success, until they were overtaken by Obamacare.
In 2006 in Massachusetts, then-Governor Mitt Romney attempted to fix the disaster created by his state’s 1990s-era reforms with a return to market principles. An overwhelmingly Democratic state legislature imposed a far more state-centered set of reforms, overturning each of his eight vetoes. Romney secured a market-based insurance exchange that provided a way to make insurance portable, but was forced to accept employer and individual mandates, both of which he had originally opposed. The purpose of the mandate was to roll back the crushing adverse-selection spiral that had resulted from the other reforms.
At first, it seemed to work. Massachusetts insurance premiums dropped some 40 percent. But that did not last. Premiums in Massachusetts eventually rose again—but only after Congress was well down the road to developing a sweeping health care law backed by the individual mandate, which Congress then thought was the silver bullet for the adverse-selection spiral. The mandate concept figured prominently in nearly every committee hearing related to health reform in 2009.
Congress placed such faith in the mandate that its resulting legislation ambitiously prohibited any exclusions for preexisting conditions, on top of the other health insurance reforms. This choice is particularly crucial to understanding the vital importance of the individual mandate in the overall congressional design, for it supplied virtually the only corrective to a scheme of guaranteed issue without exclusions for preexisting conditions.
The Eleventh Circuit utterly failed to understand this. It thus concluded incorrectly that the remainder of the health care coverage could function independently of the mandate, notwithstanding its critical role in preventing an adverse-selection spiral. The use of the mandate was needed, or so Congress thought, to make the broad guaranteed issue provision at the heart of the ACA work without undue risk of adverse selection. Congress knew that without the individual mandate, the ACA’s other reforms were a perfect recipe for adverse selection on an epic scale.
Any fair-minded look at the original congressional design shows that all of the health insurance market reforms depended on the individual mandate—not just “guaranteed issue” and the (related) prohibition on exclusions for preexisting conditions, but also “community rating”, the prohibition on annual benefit limits, comprehensive coverage requirements, limitations on co-pays and deductibles, preventive care coverage requirements, and even the reduction in subsidies to hospitals that care for the indigent. All of these provisions raise the cost of insurance; and in combination with guaranteed issue, they create relentless adverse selection pressure.
Transcripts of committee hearings, floor statements, and record votes as various proposals wound their way through the legislative process show, in abundant detail, that that proponents of health reform fully understood the mandate’s foundational importance to the whole scheme. It was for that reason that the law’s proponents rejected every effort to remove the mandate from the law in committee vote after committee vote.
The law seeks to sustain elaborate cross-subsidies that could never survive in a voluntary market. Without the mandate to temper individual abuses, the needed subsidies would force the Federal government to raise new revenues to cope with adverse selection by all individuals earning less than 400 percent of the federal poverty level (or about two-thirds of all Americans). Among the top third of income earners, health insurance will virtually disappear without the mandate, because healthy people will, quite rationally, wait until they are sick to get health insurance. The likely result is that, without the mandate, overall insurance coverage will shrink as the top third of income-earners flee the system, while the subsidies to lower-income people widen the federal deficit beyond any estimate. Even the Congressional Budget Office estimates that without the individual mandate the rest of the law will generate at most half the hoped-for increases in the ranks of insured.
Proponents of the law have argued that the close connection between the mandate and the rest of the statute makes the mandate constitutional under the Necessary and Proper Clause of the Constitution. But that conclusion misunderstands the severability argument, which is that once this law has passed, the remainder cannot not operate without the mandate and is “necessary” in that sense. Before passage, Congress could have replaced the mandate with other devices to combat adverse selection—such as allowing exclusions for preexisting conditions, or even establishing a fully subsidized single-payer system.
If the Court strikes down the individual mandate but upholds the rest of the law, it will ensure a repeat of the states’ disastrous experience with “guaranteed issue” reforms—on a far grander scale. The Supreme Court should not craft out of whole cloth a new law that Congress never enacted and would never have enacted. And future attempts at health reform will hopefully be far more wary of distorting market forces than recent reforms have been.
1The modern Supreme Court precedents truncate this first inquiry by asking only whether the remainder would be “fully operative as a law”, apparently regardless of whether those provisions would operate in the way Congress intended. But that loses the insight of the older cases, which was that the remaining provisions could be left in place if they were independent of the invalid provisions and would be “fully operative” in the way Congress intended. The parties have unfortunately embraced the modern reading, which in effect focuses all attention on the second prong.
2The latter were regulated by the Health Insurance Portability and Accountability Act of 1996 (HIPAA) until the ACA was adopted.