After September 11, as New York’s Finest turned some of their attention away from ordinary policing and focused on preventing another terrorist strike, violent crime spiked up in some areas of the city. The spike suggested how activist policing really does cut crime: if you stop doing it, this is what happens. The experience served as a reminder that Gotham’s amazing crime turnaround during the 1990s, with murder falling 70 percent and violent crimes in general down by more than half, rested on the innovations of Giuliani-era policing—and above all on the “broken-windows” strategy of policing such “quality of life” offenses as aggressive panhandling and public urination, on the assumption that tolerating such disorder gives wrongdoers the impression that no one is in charge and encourages more serious offenses.

Most academic criminologists, however, don’t buy the idea that changes in policing can change the crime rate. They believe that “root causes” or vast impersonal forces—not policy choices—explain changes in the crime rate. According to these academics, crime declined in the nineties because the crack epidemic burned itself out; the economy boomed, so poor people weren’t forced into a life of crime; and there were fewer testosterone-charged, crime-prone young males around. And if policing did contribute a little to lowering crime, in the ivory-tower view, the costs were excessive, as activist cops spread a reign of terror in poor minority neighborhoods.

To test these views, in 1999, under the auspices of the Manhattan Institute, we decided to collect and analyze data to see what really did cause crime to fall so dramatically in the city. We collected precinct-level data on violent crime and misdemeanors, and data on some of the “vast forces” criminologists credit with the crime drop: drug-use patterns, school enrollment, unemployment. Then we examined the relationships.

The results strongly indicated that broken-windows policing does matter in cutting crime—a lot. None of the other leading explanations seems plausible. Broken-windows policing significantly and strongly associates with sharp reductions in violent crime; in fact, we estimate that it prevented at least 60,000 violent crimes in New York between 1989 and 1998. By contrast, changes in the number of high school–age males bore no significant association with changes in the amount of violent crime. Nor did shifts in drug use. As for the economy, changes in unemployment levels did relate weakly to reduced violent crime—but in exactly the opposite way than criminologists had theorized. Over the ten-year period of our analysis, violent crime actually dipped as unemployment worsened. Our findings leave all the conventional explanations with little to support them.

In addition, we did what criminologists usually don’t do: we went and looked for ourselves, firsthand, spending time at six precincts. We sat in on strategy meetings, interviewed officers, and observed them as they went about their work. Contrary to the critics’ charge that the NYPD is a mindless, oppressive force, what we discovered is that officers who enforced broken-windows laws routinely evaluated the actual situation at hand and the potential consequences of their actions. While arresting a minor offender was always an option, for example, police used discretion and often warned the offender verbally instead.

Precinct leaders, we also learned from our up-close observations, regularly used the now-famous Compstat system of computerized crime tracking and establishing accountability—another Giuliani-Bratton innovation—to identify potential crime problems and come up with smart ways to deal with them before they materialized. Though we don’t have data that prove Compstat has helped slash New York’s crime rate, everything that we saw suggests that it has. We noted that commanders successfully called their shots: when Compstat focused their attention on fighting a given category of crime, that kind of crime usually decreased.

No matter what happens in the future to the “root causes” and “vast forces,” there’s no reason to fear that the NYPD, if it builds on its current strengths, will be unable to keep crime under control.


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next