Thanks to star libertarian lawprof and Cato senior fellow Randy Barnett for pointing out something that has needed saying for a while: most proposals in the U.S. Congress to address medical malpractice law run into serious federalism problems.

Most medical malpractice suits go forward in state courts under state law. If the U.S. Congress wishes to impose a nationwide rule on these suits, such as by limiting damages for pain and suffering, it first needs to answer the question: under which of the federal government’s constitutionally prescribed powers is it acting? Even if it can identify such authority, it should also ask: is it a wise idea—consistent with what one might call a prudential federalism—to gather yet more power in Washington at the expense of the states?

Unfortunately, the backers of the current federal med-mal bill have chosen to rely on the Supreme Court’s very expansive “substantial effects” doctrine, which as Barnett explains:

allows Congress to regulate any economic activity in the country that can be said, in the aggregate, to have a “substantial effect” on interstate commerce. This doctrine was unknown before the 1940s, and goes far beyond the original power to regulate trade between states. This is how most of Congress’ regulatory power has been justified since then.

Although it is followed even by conservative justices, Justice Clarence Thomas has long criticized the Substantial Effects Doctrine on the ground that it exceeds the original meaning of the Constitution.

Let’s step back for a moment to review what’s not at issue here. First, this is not an argument over whether liability reform of some sort is a good idea: in fact Prof. Barnett “strongly support[s] reforming our malpractice laws to protect honest doctors from false claims and out-of-control state juries.” (So do I.)

Nor is this an argument over whether the federal government should simply leave the state courts alone as a general proposition, as some late-blooming friends of federalism on the left side of the aisle seem to suggest. Our constitutional scheme of government is entirely consistent with federal-level supervision of state courts when those courts behave in certain ways, as by violating litigants’ due process, impairing the obligation of contract, or abridging the privileges and immunities of citizens of other states, to name but a few. Article IV, Section 1 confers on Congress a broad charter to prescribe to states “by general Laws” how they are to accord full faith and credit to other states’ enactments. That’s not even counting Congress’s genuine interstate commerce power (as opposed to the on-steroids New Deal version) or various other powers.

But observe the pattern. Again and again, the Constitution contemplates federal supervision of state courts when they reach out to assert power over transactions and litigants outside their own boundaries. It has far less to say about intruding upon the authority of those courts over disputes that arose between their own residents and are unmistakably under their own law. That general game plan—oversee the interstate but mostly not the intrastate doings of state courts—comports well with the insight of public choice scholars who point out that states face an ongoing temptation to stack liability proceedings so as to enrich their own citizens at the expense of out-of-state litigants obliged to appear in their courts.

Where does this leave federal-level liability reform? It suggests a very real difference between areas like product liability and nationwide class actions—in which suits ordinarily cross state lines, and the majority of runaway verdicts are against out-of-state defendants—and more conventional kinds of tort litigation arising from car crashes, slip-and-falls, and medical misadventure, where cases are mostly filed against locally present defendants. As a rough rule of thumb, it’s worth presuming that most of the local suits do not externalize heavy costs across state lines and should accordingly be left alone by Congress unless it is itself vindicating some constitutional right or coordinating the functioning of some constitutionally authorized federal government activity.

That doesn’t mean federal policymakers are to be left with no role at all. For example, if Washington is paying for a large share of hospital stays, it may make sense as a cost containment measure for it to steer beneficiaries into lower-cost ways of resolving disputes over care quality, or even to ask beneficiaries as a condition of treatment to agree not to file certain suits at all. But that would require stepping back toward a more careful—and more Constitutionally appropriate—view of the federal role.