Online Poker and Steven Levitt's Daughter

University of Chicago economist Steven Levitt, who says his wife "is kind of a semi-professional poker player," is upset about the Justice Department's crackdown on the online version of the game. "I think it makes no sense at all," he says in a recent Freakonomics radio segment. Unfortunately, he continues after that:

Most things that are made illegal, everyone agrees on—homicide, theft. There's just general agreement. And then there's this other set of activities which fall into a gray area, things like prostitution and gambling—not everyone agrees. But I think poker is so obviously on one side of the gray area relative to legality that it just doesn't make any sense to make it illegal.

Why is poker so obviously on the other side? Because it passes Levitt's "daughter test":

If the prohibited activity is something that I actually think would be good for my daughter to do, then I'm in favor of it being legal. But if the activity is something which I would feel terrible if my daughter did, then I would want it to be illegal.

Since Levitt is fine with the idea of his daughter as a professional poker player, poker should be legal. Since he is not fine with the idea of his daughter as a prostitute, he says, prostitution should be illegal, even though intellectually he is inclined to think, "Why shouldn't prostitution be legal? It's a transaction between two individuals." In case you find the daughter test unsatisfying, Levitt, prodded by his interviewer, throws in some mumbo-jumbo about externalities:

When we think about prohibiting activities, we do [that] either because they directly harm someone else, like homicide or theft, or because indirectly there are spillovers that hurt other people. So we think that drug addiction, for instance, imposes costs on other people, or people argue that secondhand smoke imposes costs on others. If you think about poker…it's hard to see how [the externalities] could be very large, especially when there are sanctioned kinds of gambling that people are allowed to do, like lotteries, which pay out at much worse rates than do poker sites. The easiest message to draw from this is that the kinds of gambling which the government outlaws are those kinds of gambling which the government does not directly financially benefit from.

Levitt is onto something with that last point, although I would add that the financial interests of politically influential businesses (such as Las Vegas casinos and horsetrack operators) are important too. But his daughter test is utterly arbitrary, since other people might well be horrified at the idea of their offspring becoming professional poker players. Indeed, many Americans do not accept the distinction that Levitt draws between "gambling" (which he says falls into the "gray area") and poker (which he says clearly does not).

It's hard to know how seriously to take Levitt's daughter test, since at one point he says, "Would I be in favor of outlawing premarital sex? I probably would. Now that would be a hard law to enforce." But even if Levitt is trying to describe the reasoning of other people who support paternalistic laws, I would guess that the "me test" is more common: If the prohibited activity is something I enjoy, it should be legal. I actually prefer that rule, which is just as arbitrary but without the pretense of fatherly concern.

Levitt (obviously) is not a moral philosopher, but he is an economist. Yet his discussion of the externality rationale for paternalistic laws is even less coherent than his explication of the daughter test. People do claim that drug use imposes costs on third parties, but they claim the same thing about gambling, including poker. Indeed, a similar claim could be made about any activity that can be carried to excess: It disrupts relationships, degrades work performance, fosters crime, etc. If that argument, which I call "the addict's veto," is enough to justify prohibition, pretty much everything falls into Levitt's "gray area."

Levitt argues that poker's social costs can't be very big, especially since lotteries are legal. But that's a non sequitur: Maybe the social costs of both are substantial, and the right policy is to ban both, which is the position  of many anti-gambling activists. After all, Levitt starts from the (correct) premise that existing laws are not necessarily rational or consistent. Speaking of which, his suggestion that drug prohibition is (or might be) justified by externalities raises the obvious question of why alcohol should be legal. Perhaps because Levitt does not mind if his daughter drinks when she grows up but would be terribly upset if she smoked pot or snorted cocaine.

Addendum: A recent study that Levitt co-authored provided evidence that poker is a game of skill. I think there's no serious question that skill matters a lot in poker over the long run, but that does not necessarily resolve the issue of whether poker counts as gambling under state laws. The most common legal test is whether chance is "the dominating element," which leaves room for argument. More important, this distinction is morally irrelevant as a justification for prohibiting consensual interactions between adults.