Smoking Joints and Broken Windows

In a forthcoming Criminology and Public Policy article, University of Chicago law professor Bernard Harcourt and Jens Ludwig, a professor of public policy at Georgetown University, debunk the idea that New York City's crackdown on public pot smoking has helped reduce violent crime. As part of New York's "broken windows" law enforcement strategy, misdemeanor arrests for smoking marijuana in public view (MPV) rose from 1,851 in 1994 to a peak of 51,267 in 2000, an increase of 2,670 percent. According to "broken windows" theory, these arrests should have had a broader impact on crime by reducing conspicuous signs of lawbreaking and disorder. Another possible justification is that MPV arrests lead to more serious arrests because public pot smokers are especially likely to be predatory criminals. (A similar rationale has been offered for crackdowns on petty offenses such as loitering and subway turnstile jumping.) A third rationale is that arresting drug users disrupts black market activity and reduces the violence associated with it—not very plausible in this case, since the marijuana trade is substantially less violent than, say, the crack trade.

Whatever the theory for viewing MPV arrests as a crime control tactic, they do not seem to have had the predicted effect. Looking at data across police precincts, Harcourt and Ludwig initially find an association between MPV arrests and reductions in violent crime. But this apparent effect disappears once they take into account regression to the mean—the tendency of the precincts that saw the biggest increases in crime rates during the 1980s and early '90s to see the biggest subsequent drops. It turns out those precincts are also the ones that put the most effort into arresting pot smokers. After adjusting the data for regression to the mean, Harcourt and Ludwig find that, if anything, higher MPV rates are associated with higher violent crime rates. "We find no good evidence that the MPV arrests are associated with reductions in serious violent or property crimes in the city," they conclude.

The apparent lack of results is especially troubling because the MPV arrests have disproportionately involved blacks (who accounted for about 25 percent of New York's population but 52 percent of MPV arrests) and Hispanics (also about 25 percent of the population but 32 percent of MPV arrestees). It may be that blacks and Hispanics are more apt than whites to smoke pot in public, or (more likely) it could be that they disproportionately live in the high-crime neighborhoods where the police department focused its "broken windows" efforts. But Harcourt and Ludwig note that blacks and Hispanics were not only especially likely to be arrested during the study period (1989 to 2000); they also "were more likely than their white counterparts to be detained before arraignment (2.66 and 1.85 times more likely, respectively), convicted (both twice as likely) and sentenced to additional jail time (4 and 3 times more likely, respectively)." One need not accuse the police and courts of deliberate discrimination to be disturbed by figures like these, especially since advocates of the crackdown on pot smokers apparently have nothing to show for it.

Leaving aside the racial angle and the injustice of arresting people for a "crime" that harms no one, these arrests seem to be a big waste of time and money, even when you consider possible indirect effects. As Harcourt and Ludwig put it, "New York City's marijuana policing strategy seems likely to simply divert scarce police resources away from more effective approaches that research suggests is capable of reducing real crime."