One possible reason fear of crime remains high is that powerful people have an incentive to ring the alarms anyway. Politicians score points by promising to get "tough on crime," even after those efforts pay off and crime levels hit historic lows. Media play up only the most horrifying deeds. The result is a skewed perception of how dangerous the world is. It's telling, though, that most people believe the danger is not on their doorstep, but beyond it. Fewer people say crime is up when asked about their area than when asked about the whole country.
Another reason for the perception gap is the constant sense that things are getting worse. Only during one short period in the last 20 years have Americans thought there was less crime than in the past. That was in the two years after Sept. 11, 2001. "People were so positive about America that I think the enemy was perceived to be much more outside the country than inside the country," a Gallup pollster told the Boston Globe.
Keep in mind, too, that perceptions of crime lag behind crime itself. Crime went down for several years starting in 1991, but fear remained high. "That was the nastiest and most fearful period in history," says Zimring. Congress passed a federal "truth in sentencing" law in 1994. About half of the states passed "three strikes" laws to punish habitual offenders. All this, even though crime was already on the decline.
Some criminologists think the causal relationship between fear of crime and actual crime is actually the reverse—that fear of crime itself leads to more actual crime. The "broken windows" theory of criminology, made famous by Rudy Giuliani in the 1990s, posits that police can deter serious crimes by enforcing less serious ones. If someone jumps the subway turnstile and the police chase him down, that contributes to the sense that crime is punished and the social order is preserved—which might make a would-be murderer think twice. This notion that when society feels more safe, it is more safe, is a win-win for politicians: They can take credit for reducing crime and reap the benefit of a safer-feeling city. But the fact that crime has dropped while fear of crime remains high shows once again that the relationship is more complicated than the theory suggests.
Fear of crime isn't entirely bad, from a risk-management perspective. Someone who's scared of being assaulted is less likely to leave the house, thus lessening the likelihood of assault. Or maybe they just avoid the street corners that make them most nervous. No matter how far the crime rate falls, these perceptions could still reduce personal risk on the margins. The problem is, shrinking from crime doesn't make it go away. It merely deflects the danger onto someone else. Fear might work as a personal strategy; it doesn't work as a societal one.