Policing Broken Windows

In economist Steven D. Levitt's book Freakonomics, he argues against the claim that various innovative policing strategies, such as the Broken Windows strategy, helped to reduce crime (pp. 127-130). When certain other factors, such as demographic change and police force growth, are also accounted for, the adoption of those strategies appears not to have had much of the role in the precipitous crime drop seen in New York City.

But here's a hypothesis that seems consistent with the data. Certain policing strategies might abate risks of mugging, robbery, etc. Before these strategies are used, however, there still wouldn't be all that many muggings, etc., because people use common sense. They know what side streets to avoid. They know to take a bus or taxi, rather than the subway, late at night. They know that certain neighborhoods are off-limits altogether. New policing strategies, such as deploying more plainclothes officers on subways and cracking down on turnstyle jumpers, intoxicated riders, loitering, etc. made the subway an acceptable option for late night travel. In the late 1990s, I rode the subway after midnight any number of times. That would have been unthinkable for me a few years earlier. In many respects, crime (risk) is just a quality of life factor. I was never at much risk of being a victim, because I restricted my movements. I gained a lot more freedom as time went on, thanks, I think, to the changes Levitt has written off.

This really doesn't refute Levitt's claim at all. But it does call into question whether crime rates tell the whole story, if what really matters is public safety. Crime rates matter to public safety, of course, but so do crime risks. Broken Windows might play a role in the latter, if not the former.