The great policy challenge to cannabis legalization is discouraging problem use. Most consumers have no trouble keeping their consumption within reasonable bounds, but 10 to 15 percent lose control for months at a time, and some of them develop chronic problems. At any one time, there are a couple of million Americans who self-report that they’re trying and failing to cut down on their cannabis use. That hasn’t kept zealous advocates from convincing themselves and others that “pot isn’t addictive.”

No one intends to become habituated to cannabis, any more than anyone intends to develop any other bad habit. Abuse creeps up on some people because their long-term desire to avoid habituation, and to function well in school, at work, and at home, has a hard time competing with the short-term allure of getting high.

One way to limit the number of problem users is to prevent the creation of an industry of for-profit firms eager to sell to them (see Jonathan P. Caulkins, “Nonprofit Motive,” page 39). Another possibly complementary approach would be to keep consumers mindful of how much they’re actually using, compared to how much they intend to use, with a system of user-chosen monthly purchase quotas. Since using more, or more often, than intended is among the defining characteristics of substance abuse, helping users enforce on themselves their own chosen consumption patterns would address the problem at its root. This is one version of the sort of “choice architecture” or “libertarian paternalist” strategy described by Richard Thaler and Cass Sunstein in their book Nudge.

Under a quota system, each cannabis user would have to sign up and receive an account number. (There would be no need to match the number with a name or any other identifier.) On signing up, each user would be asked to set his or her own monthly purchase limit. A user who wanted to get high only once a week, and hadn’t yet built up a tolerance, might set a limit of 200 milligrams of THC per month. How to provide appropriate guidance at that stage—and how to convince a new user that habituation might happen to him, not just to someone else—constitutes the hardest operational challenge. One approach would be to set some modest arbitrary number as the “default option” for users who don’t choose a higher or lower limit for themselves.

A user could set any limit he wanted to, but once set, that limit would be binding until changed. If a user asked to buy more than was left out of his or her self-selected monthly allowance, the clerk would have to refuse. Those limits would not be carved in stone; users could change them, but only—and here’s the key—after some delay. (Two days? Two weeks? There would have to be a similar waiting period between applying for a new card and its effective date.) Since testing-and-labeling rules mean that each package will contain a measured quantity of THC, the system would only have to track the monthly total for each account number: not a daunting data-management task.

Of course, no such system is foolproof. Someone will always invent a bigger fool. Some users who run out of quota would simply set a higher limit, and do so again when that higher limit didn’t keep pace with their growing appetite for intoxication. For them, the system might not do much good. However, others—and there’s no way to guess in advance how many—would not reset their limits, choosing instead to keep them in place as external props for occasionally weak wills. Those, and the other users for whom a quota system is an encouragement to think about how much they’re using, would be the beneficiaries of this “nudge” toward temperance.

Like most nudge strategies, there’s a “chicken soup” aspect to a personal user quota: it might not do much good, but it couldn’t hurt.

Return to “Saving Marijuana Legalization.”

Mark Kleiman

Mark Kleiman is a professor of public policy at the New York University Marron Institute.