the crime rate in their precinct started to drift upward.165 The visibility of the police was multiplied by a mandate to go after nuisance crimes like graffiti, littering, aggressive panhandling, drinking liquor or urinating in public, and extorting cash from drivers at stoplights after a cursory wipe of their windshield with a filthy squeegee. The rationale, originally articulated by James Q. Wilson and George Kelling in their famous Broken Windows theory, was that an orderly environment serves as a reminder that police and residents are dedicated to keeping the peace, whereas a vandalized and unruly one is a signal that no one is in charge.166
Did these bigger and smarter police forces actually drive down crime? Research on this question is the usual social science rat’s nest of confounded variables, but the big picture suggests that the answer is “yes, in part,” even if we can’t pinpoint which of the innovations did the trick. Not only do several analyses suggest that something in the new policing reduced crime, but the jurisdiction that spent the most effort in perfecting its police, New York City, showed the greatest reduction of all. Once the epitome of urban rot, New York is now one of America’s safest cities, having enjoyed a slide in the crime rate that was twice the national average and that continued in the 2000s after the decline in the rest of the country had run out of steam.167 As the criminologist Franklin Zimring put it in The Great American Crime Decline, “If the combination of more cops, more aggressive policing, and management reforms did account for as much as a 35% crime decrease (half the [U.S.] total), it would be by far the biggest crime prevention achievement in the recorded history of metropolitan policing.”168
What about Broken Windows policing in particular? Most academics hate the Broken Windows theory because it seems to vindicate the view of social conservatives (including former New York mayor Rudy Giuliani) that violence rates are driven by law and order rather than by “root causes” such as poverty and racism. And it has been almost impossible to prove that Broken Windows works with the usual correlational methods because the cities that implemented the policy also hired a lot of police at the same time.169 But an ingenious set of studies, recently reported in Science, has supported the theory using the gold standard of science: an experimental manipulation and a matched control group.
Three Dutch researchers picked an alley in Groningen where Netherlanders park their bicycles and attached an advertising flyer to the handlebars of each one. The commuters had to detach the flyer before they could ride their bikes, but the researchers had removed all the wastebaskets, so they either had to carry the flyer home or toss it on the ground. Above the bicycles was a prominent sign prohibiting graffiti and a wall that the experimenters had either covered in graffiti (the experimental condition) or left clean (the control condition). When the commuters were in the presence of the illegal graffiti, twice as many of them threw the flyer on the ground—exactly what the Broken Windows theory predicted. In other studies, people littered more when they saw unreturned shopping carts strewn about, and when they heard illegal firecrackers being set off in the distance. It wasn’t just harmless infractions like littering that were affected. In another experiment, passersby were tempted by an addressed envelope protruding from a mailbox with a five-euro bill visible inside it. When the mailbox was covered in graffiti or surrounded by litter, a quarter of the passersby stole it; when the mailbox was clean, half that many did. The researchers argued that an orderly environment fosters a sense of responsibility not so much by deterrence (since Groningen police rarely penalize litterers) as by the signaling of a social norm: This is the kind of place where people obey the rules.170
Ultimately, we must look to a change in norms to understand the 1990s crime bust, just as it was a change in norms that helped explain the boom three decades earlier. Though policing reforms almost certainly contributed to the headlong decline in American violence, particularly in New York, remember that Canada and Western Europe saw declines as well (albeit not by the same amount), and they did not bulk up their prisons or police to nearly the same degree. Even some of the hardest-headed crime statisticians have thrown up their hands and concluded that much of the explanation must lie in difficult-to-quantify cultural and psychological