• Email
  • Single Page
  • Print

Is Health Care Reform Unconstitutional?

Nicholas Kamm/AFP/Getty Images
Tea Party supporters protesting against the health care bill, Washington, D.C., March 2010


Can Republicans defeat Obama’s health care bill by persuading the courts that mandatory health insurance is unconstitutional? On December 13, 2010, Henry Hudson, a federal judge in Virginia, declared unconstitutional the central provision of the health care reform law. Judge Hudson reasoned that the law’s command that citizens purchase health care insurance extended beyond Congress’s authority to legislate. It has long been established that Congress may regulate citizens’ economic activities, such as entering into contracts, producing or purchasing goods and services, or shipping goods across state lines. But it is entirely unprecedented, Judge Hudson said, for Congress to regulate “inactivity”—a failure to buy insurance.

Obama dismissed the opinion as just “one ruling by one federal district court.” But to others, it came as a shock. Dahlia Lithwick, Slate’s legal affairs correspondent, said that before Judge Hudson’s decision, most experts thought the legal challenges would fail in the Supreme Court by a large margin, 8–1 or 7–2, but that after the ruling, the betting is that the Court will split 5–4, with Justice Anthony Kennedy likely casting the decisive vote.1

It is easy to see why commentators might expect the case to be closely divided. Health care reform has been nothing if not intensely partisan. The Patient Protection and Affordable Care Act was passed without a single Republican vote, and Republicans in the House have already voted to repeal it. The fight over its enactment helped promote the rise of the Tea Party and the Republican victories in the midterm elections. Most of the state attorneys-general who have challenged the law’s constitutionality in court are Republicans. Several Democratic state attorneys-general have filed a brief supporting the law. So far, three judges have ruled on the merits of the challenges. Two, both appointed by Democratic presidents, upheld the act. Judge Hudson, the first to rule otherwise, is a Republican appointed by George W. Bush. Another Republican-appointed judge, Roger Vinson, has a similar case pending in Florida, and he is likely to side with Judge Hudson.1a

The roots of the ideological divide, moreover, run deep. The principal constitutional issue at stake—the extent of Congress’s authority to pass laws governing Americans’ lives—has separated conservatives and liberals since the beginning of the Republic. “States’ rights” was the South’s rallying cry in its effort to retain slavery before the Civil War, and to defend racial segregation from federal intervention thereafter. From the turn of the century through the early years of the New Deal, conservatives successfully invoked “states’ rights” to interpret Congress’s power over interstate commerce narrowly and thereby invalidate progressive federal laws designed to protect workers and consumers from big business. And the last two times that the Supreme Court struck down laws as reaching beyond Congress’s Commerce Clause power, in 1995 and 2000, the Court split 5–4, with Chief Justice William Rehnquist writing the majority decision, over dissents by Justices Stephen Breyer, David Souter, John Paul Stevens, and Ruth Bader Ginsburg.2

As Judge Hudson sees it, the health care reform law poses an unprecedented question: Can Congress, under its power to regulate “commerce among the states,” regulate “inactivity” by compelling citizens who are not engaged in commerce to purchase insurance? If it is indeed a novel question, there may be plenty of room for political preconceptions to color legal analysis. And given the current makeup of the Supreme Court, that worries the law’s supporters.

But the concerns are overstated. In fact, defenders of the law have both the better argument and the force of history on their side. Judge Hudson’s decision reads as if it were written at the beginning of the twentieth rather than the twenty-first century. It rests on formalistic distinctions—between “activity” and “inactivity,” and between “taxing” and “regulating”—that recall jurisprudence the Supreme Court has long since abandoned, and abandoned for good reason. To uphold Judge Hudson’s decision would require the rewriting of several major and well-established tenets of constitutional law. Even this Supreme Court, as conservative a court as we have had in living memory, is unlikely to do that.

The objections to health care reform are ultimately founded not on a genuine concern about preserving state prerogative, but on a libertarian opposition to compelling individuals to act for the collective good, no matter who imposes the obligation. The Constitution recognizes no such right, however, so the opponents have opportunistically invoked “states’ rights.” But their arguments fail under either heading. With the help of the filibuster, the opponents of health care reform came close to defeating it politically. The legal case should not be a close call.


The provision that Judge Hudson struck down requires all Americans, unless exempted on religious or other grounds, to purchase health care insurance. (Most Americans are already covered through their employment or Medicare or Medicaid, so for them this law would have no impact.) Those who do not obtain insurance must pay a penalty in the form of a special tax.

The individual mandate is aimed at so-called “free riders”—people who fail to get insurance, and then cannot pay the cost of their own health care when they need it. Under our current system, in which hospitals must treat people regardless of ability to pay or insurance coverage, hospitals are able to recover only about 10 percent of the cost of treating uninsured individuals. That cost is ultimately borne by the rest of us. The federal government picks up much of the tab, and hospitals and insurers pass on the rest to their paying customers in higher fees. The Congressional Budget Office estimated that in 2008 the uninsured shifted $43 billion of health care costs to others.

Without the individual mandate, the health care law’s more popular reforms—such as the bar on insurance companies denying coverage because of “preexisting conditions”—would actually make the insurance crisis worse. Knowing that insurers could not deny coverage or charge more for preexisting conditions, people could wait to buy insurance until they were sick. But then more and more of the people insured would be the sickest, defeating the very purpose of insurance—to spread the risk by creating a pool of funds that can be drawn on for payments. Premiums would skyrocket, meaning that even fewer people could afford insurance, and that would in turn induce still more people to opt out. As Wake Forest University Professor Mark Hall testified in Congress, “a health insurance market could never survive or even form if people could buy their insurance on the way to the hospital.”

This is not just an academic prediction. When in 1994 Kentucky enacted similar reforms regarding preexisting conditions, but without an individual mandate, insurance costs rose so steeply that they became untenable, and insurers pulled out of the market altogether. Kentucky was forced to repeal the reform. Initiatives in New York and New Jersey faced similar problems. In Massachusetts, by contrast, where health insurance reform was coupled with an individual mandate, the system has worked; since 2006, insurance premiums there have fallen 40 percent, while the national average has increased 14 percent.

Judge Hudson acknowledged, as do the law’s challengers, that Congress has power to regulate any economic activity that, in the aggregate, affects interstate commerce—no matter how minimal the activity’s effects are standing alone. But the decision not to buy health insurance, Hudson reasoned, is not “activity” at all. It is “inactivity.” Rather than setting rules for those who choose to engage in interstate commerce, the individual mandate compels a citizen who has chosen not to engage in commerce to do so by purchasing a product he does not want. If Congress can regulate such “inactivity,” Hudson warned, there would be no limit to its powers, contravening the bedrock principle that the Constitution granted the federal government only limited powers.

Judge Hudson’s reasoning is not without precedent—but the precedents that his rationale reflects have all been overturned. In the early twentieth century, the Supreme Court ruled that the Commerce Clause authorized Congress to regulate only “interstate” business, not “local” business; only “commerce,” not production, manufacturing, farming, or mining. The Court also ruled that Congress could regulate only conduct that “directly” affects interstate commerce, not conduct that “indirectly” affects interstate commerce. Like Judge Hudson, the Supreme Court warned that unless it enforced these formal categorical constraints, there would be no limit to Congress’s power. Thus, for example, in 1936, the Court struck down a federal law that established minimum wages and maximum hours for coal miners, reasoning that mining was local, not interstate; entailed production, not commerce; and had only “indirect” effects on interstate commerce.3 Using this approach, the Court invalidated many of the laws enacted during the early days of the New Deal.

Around 1937, however, the Court reversed course. It recognized what economists (and the Court’s dissenters) had long argued, and what the Depression had driven home—that in a modern-day, interdependent national economy, local production necessarily affects interstate commerce, and there is no meaningful distinction between “direct” and “indirect” effects. In the local, agrarian economy of the Constitution’s framers, it might have made sense to draw such distinctions, but in an industrialized (and now postindustrialized) America, the local and the national economies are inextricably interlinked.

As a result, Congress’s power to regulate “interstate commerce” became, in effect, the power to regulate “commerce” generally. The Court rejected as empty formalisms the distinctions it had previously drawn, between local and interstate, between production and commerce, and between “direct” and “indirect” effects. Since 1937, the Supreme Court has found only two laws to be beyond Congress’s Commerce Clause power. Both laws governed noneconomic activity—simple possession of a gun in a school zone and assaults against women, respectively—and were unconnected to any broader regulation of commerce.4 But the Court has repeatedly made clear that Congress can regulate any economic activity, and even noneconomic activity where doing so is “an essential part of a larger regulation of economic activity.”

On this theory, the Supreme Court has upheld federal laws that restricted farmers’ ability to grow wheat for their own consumption and that made it a crime to grow marijuana for personal medicinal use, even though in both instances the people concerned sought to stay out of the market altogether.5 The Court reasoned that even such personal consumption affects interstate commerce in the aggregate by altering supply and demand, and that therefore leaving it unregulated would undercut Congress’s broader regulatory scheme.

Under these precedents, a citizen’s decision to forgo insurance, like the farmer’s decision to forgo the wheat market and grow wheat at home, easily falls within Congress’s Commerce Clause power. When aggregated, those decisions will shift billions of dollars of costs each year from the uninsured to taxpayers and the insured. As a practical matter, there is no opting out of the health care market, since everyone eventually needs medical treatment, and very few can afford to pay their way when the time comes. (Those who refuse all medical treatment for religious scruples are an exception, but they are exempt from the mandate.) That one might affix the label “inactivity” to a decision to shift one’s own costs to others does not negate the fact that such economic decisions have substantial effects on the insurance market, and that their regulation is “an essential part of a larger regulation of economic activity.”

  1. 1

    It is not certain that Judge Hudson’s decision will reach the Supreme Court. He found that Virginia had standing to sue because its legislature had enacted a law giving citizens the right not to buy health insurance, as part of a national campaign by health care reform opponents to create obstacles in the states. It is far from clear that a state can manufacture a dispute by enacting such a law, so it is possible that the case could be thrown out on appeal for failing to present a concrete controversy. However, at some point the health care law will plainly be subject to a constitutional challenge, at a minimum by any individual who objects to being required to purchase health insurance or pay the tax. So whether in this case or some other, the courts will eventually have to address the law’s constitutionality. 

  2. 1a

    On January 31, 2011, after this article went to press in the print edition, Judge Vinson issued his decision. Like Judge Hudson, he ruled unconstitutional the provision requiring individuals to purchase health insurance, because it regulates “inactivity.” He went further than Judge Hudson, however, by declaring the entire health care reform law unconstitutional, not just the provision requiring individuals to purchase health insurance, because he concluded Congress would not have enacted the rest of the statute without the “individual mandate” provision. His decision, like Judge Hudson’s will certainly be appealed. 

  3. 2

    United States v. Morrison, 529 U.S. 598 (2000); United States v. Lopez, 514 U.S. 549 (1995). 

  4. 3

    Carter v. Carter Coal Co., 298 U.S. 238 (1936). 

  5. 4

    See cases cited in note 2. 

  6. 5

    Gonzales v. Raich, 545 U.S. 1 (2005); Wickard v. Filburn, 317 U.S. 111 (1942). 

  • Email
  • Single Page
  • Print