The reason for these stops is a policing approach called “broken windows,” first articulated by scholars James Q. Wilson and George L. Kelling in a 1982 Atlantic Monthly essay and later adopted by the NYPD in 1993. Broken windows prioritizes cracking down on minor offenses on the theory that doing so can preempt serious crime. Or, to use the metaphor of the idea, actual broken windows create the appearance of disorder, which creates actual disorder as criminals take advantage of the inviting environment. Rather than wait for the serious crimes to begin, police should “repair the windows”—focus on petty crime like loitering, and you’ll stop the worse crime from taking hold.

It’s an elegant concept, but there’s little evidence it works. “Taken together,” notes a 2006 study from the University of Chicago, “the evidence from New York City and from the five-city social experiment provides no support for a simple first-order disorder-crime relationship as hypothesized by Wilson and Kelling nor for the proposition that broken windows policing is the optimal use of scarce law enforcement resources.” Yes, the massive New York crime decline of the 1990s coincided with broken windows policing, but chances are it had more to do with a reversion to the mean (“what goes up, must come down, and what goes up the most, tends to come down the most”) than any new approach.

If broken windows were just a waste of resources, it wouldn’t be a huge concern. But as a policy, broken windows has also had the effect of terrorizing Black and Latino New Yorkers.