fb-pixel Skip to main content

Numbers matter in public safety reform

IN 2011, 9 out of 10 people targeted in the controversial stop-and-frisk program utilized by the New York Police Department were African-American or Hispanic. The city’s population is 23 percent African-American, 29 percent Hispanic, and 34 percent non-Hispanic white. Indeed, whites were the only demographic that was frisked at a proportion far below its percentage of the population: Only 9 percent of those stopped were white.

The NYPD has held strong to the belief that the program prevents crime and enhances public confidence. Whose confidence is a separate question, as the NYPD faces angry outcries and lawsuits. But the traditional civil-liberties-versus-public-safety argument is actually the wrong prism through which to evaluate this program.

Indeed, claims that lower crime rates can be correlated to any single police tactic should be met with suspicion, especially by people whose inclination is to be tough on crime. Without a more detailed study, police departments can’t begin to assess whether stop-and-frisk tactics are actually working. And that skepticism is not fueled by the racial breakdown of those who are stopped and frisked, however disconcerting it may be, but by a newly released study of the NYPD’s most famous previous contribution to law enforcement: “Broken windows.”


Hundreds of books and academic studies have sought to analyze the major reduction in crime in New York City during mayor Rudy Giuliani’s reign in the 1990s. Giuliani’s basic proposition was that by focusing police efforts on minor crimes — like breaking windows — there would be a corresponding reduction in major crimes. Broken windows were just a symptom, the theory goes, of unstable environments that lead to more serious crimes.

The notion took hold like wildfire. It was easy to understand and seemed intuitively right. As the story goes, “broken windows” made the city safer, Giuliani a national figure, and helped relaunch New York as a financial and cultural hub. Overall, the mayor oversaw a 40 percent reduction in crime.

Of course, every major city achieved a significant drop in crime during the same period. What is now seriously in doubt is whether New York’s reduction can be tied to a specific police tactic. And the answer appears, at least for “broken windows,” to be no.


New York University sociologist David Greenberg systematically analyzed data from 1988 to 2001, which was not easily accessible in the past. This covers the period when “broken windows” was official police policy. His study, released in the journal Justice Quarterly, is regarded as a groundbreaking contribution to the public safety debate.

Previous studies supporting “broken windows” were flawed because they did not include data from individual precincts. Based on crime and jail statistics, police presence, and the demographics of each precinct, Greenberg’s analysis shows one undeniable fact: that the start of New York’s rapid reduction in crime predated the implementation of “broken windows,” and that the pace of decline appears to have been completely unaffected by its implementation. Even the premise of “broken windows” — that arrests for misdemeanors decreased felony crimes in the same areas — is contradicted by the data.

More startling is that it isn’t even clear, based on police reports, that the NYPD was actually enforcing “broken windows” in the communities it promised to target. Thus, if “broken windows” was not responsible for the reduction in crime, then maybe individual police tactics are less important than the social, cultural, and economic environment of the neighborhoods in which they’re deployed.

It’s a coincidence that Greenberg’s new study on “broken windows” was issued just as the NYPD was seeking to justify its most recent, similarly touted, tactic — the stop-and-frisk program.

There is no way now to know whether stop-and-frisk is working because the NYPD has not released the only relevant number: the number of stop-and-frisks that resulted in arrests. Until then, all those who are concerned about public safety should be skeptical about the tactic’s effectiveness. All we know for sure is that, in the highly integrated, socially diverse, but economically divided New York City, there were 685,724 stop-and-frisks.


And that nearly 620,000 of those were of blacks or Hispanics.

Juliette Kayyem can be reached at jkayyem@globe.com and Twitter @juliettekayyem.