One of the side-debates in the ACA decisions yesterday was between Chief Justice Roberts and Justice Ginsburg over the meaning of the term “regulate.” The Commerce Clause of the Constitution, Art. I, sec. 8, cl. 3, empowers Congress “[t]o regulate commerce . . . among the several states . . . .” Much of the pre-decision debate over the ACA mandate involved whether mandating the purchase of a service — health insurance — fell within the definition of “commerce.” This is where the famous “activity/inactivity” distinction arose: choosing not to buy something is not “commerce,” the argument went, and therefore not within Congress’s Commerce Clause powers.

Chief Justice Roberts didn’t exactly adopt that argument, however, in his opinion denying that Congress had Commerce Clause authority to mandate the purchase of health insurance. (I’m not an expert on Supreme Court voting rules, but there’s considerable debate about whether, even though five justices said the mandate was beyond the Commerce Clause, that’s actually a binding holding of the court.) Instead, what Roberts held was that mandating the purchase of health insurance isn’t regulation:

The language of the Constitution reflects the natural understanding that the power to regulate assumes there is already something to be regulated. . . . The individual mandate, however, does not regulate existing commercial activity. It instead compels individuals to become active in commerce by purchasing a product,on the ground that their failure to do so affects interstate commerce. Construing the Commerce Clause to permit Congress to regulate individuals precisely because they are doing nothing would open a new and potentially vast domain to congressional authority.

Nat’l Fed. of Indep. Bus., slip op. at 19, 20. This is a novel twist on the argument. Chief Justice Roberts is clear that he is not rejecting the idea that choosing not to buy health insurance affects commerce, at least in the same way that Filburn’s growing the wheat his family consumed affected commerce. It’s that a law forbidding individuals from making a choice not to do something doesn’t regulate commerce.

To an economist, perhaps, there is no difference between activity and inactivity; both have measurable economic effects on commerce. But the distinction between doing something and doing nothing would not have been lost on the Framers, who were “practical statesmen,” not metaphysical philosophers. . . . The Framers gave Congress the power to regulate commerce, not to compel it, and for over 200 years both our decisions and Congress’s actions have reflected this understanding. There is no reason to depart from that understanding now.

Perhaps I am missing something (a good friend of mine evaluated my argument below as follows: “Meh”), but I don’t see how this can be right. Much of what Congress does under the Commerce Clause requires individuals or companies to do or to not do something that they haven’t done yet — that is, Congress regulates activity that has not yet occurred. All federal criminal laws passed pursuant to Congress’s Commerce Clause authority — i.e., all federal crimes that don’t involve things such as stealing mail or killing bears in national parks — issue commands to people who have not engaged in the activity in question, and Congress wants to keep it that way. That’s not just a policy preference, it’s actually required by the Constitution. Congress cannot declare actions that happened in the past to be criminal and then punish them; it can only command people to not do things in the future — things that they haven’t done yet — and punish them if they disobey. In other words, Congress “regulates” inactivity, and mandates that the inactivity continue. And no one doubts that Congress can do that under the Commerce Clause.

“Well,” you might say, “that’s different. Punishments for violating criminal laws only kick in after you violate the law by acting. I.e., the law punishes past activity.” That’s both inaccurate and irrelevant. It’s inaccurate because the law imposes lots of penalties for failing to act — failing to make a required disclosure for example. But it’s also irrelevant. The activity/inactivity distinction isn’t about limits on what the government can impose punishment for. It allegedly marks a boundary about who the federal government, under the Commerce Clause, can issue commands to. And the idea is that while the government can command health insurance companies to do lots of things when they are already acting in interstate commerce, it can’t issue commands to individuals who are not engaged in the sort of activity at issue. But federal criminal laws issue commands all the time to people who are not engaged in the activity in question, who indeed are just sitting around doing nothing. And that command has an effect on them — that’s the whole point — even though they have taken no affirmative steps to subject themselves to Congress’s Commerce Clause power. Just the threat of the penalty for acting in violation of federal criminal law weighs on their minds and restricts what they think they can do. Congress in effect says: “Good. Keep it that way, or else.” The inactive are regulated in both cases; the only difference is the goal of the regulation.

“But in the criminal case,” you might argue, “a person is potentially going to act, and the point of the criminal law is to regulate that potential, future activity. Whereas with the health insurance mandate, we have a present failure to act that is being regulated.” But the government in the ACA cases made the argument that the mandate was justified because it attempts to regulate potential future activities, namely the very high likelihood that almost everyone will need to purchase health care sooner or later. Roberts shot this down: Congress cannot “regulate” potential, future activity:

The proposition that Congress may dictate the conduct of an individual today because of prophesied future activity finds no support in our precedent. We have said that Congress can anticipate the effects on commerce of an economic activity. . . . But we have never permitted Congress to anticipate that activity itself in order to regulate individuals not currently engaged in commerce. Each one of our cases, including those cited by JUSTICE GINSBURG, post, at 20–21, involved preexisting economic activity. . . .

Everyone will likely participate in the markets for food, clothing, transportation, shelter, or energy; that does not authorize Congress to direct them to purchase particular products in those or other markets today. The Commerce Clause is not a general license to regulate an individual from cradle to grave, simply because he will predictably engage in particular transactions.

If Congress cannot issue commands to someone until they have engaged in the activity in question, then that means it cannot issue commands in advance at all under its Commerce Clause authority. And criminal commands must be announced in advance under the Ex Post Facto Clause. So under Chief Justice Roberts’s theory, Congress can only pass criminal laws under its Commerce Clause authority that regulate recidivists. And it’s no defense that other people may be widely engaged in the activity in question. That does not mean Congress can regulate person X, where X has not yet trafficked in narcotics. Plenty of people purchase and use health care services on a daily basis. The question is whether Congress can regulate those who don’t. If it can’t, then it can’t regulate people who don’t commit crimes either, by mandating that they not commit crimes in the future.

But plainly Congress can do that. Chief Justice Roberts’s attempt to draw the line of regulation at “preexisting commercial activity of the sort in question by the individual to be regulated” does not work.