A recent report announced that the huge financial company UBS will be moving back from a suburb of New York into Manhattan, “because it has come to realize it is more difficult to recruit talented people in their 20s to work in the suburbs.” What a (literal) turnaround!
For about a generation, roughly from the 1970s through the 1990s, the equation, big city equals violent crime, was a taken-for-granted part of how we understood urban life. Movie-makers needed only to zoom in on a block of New York or Chicago to give audiences a dose of stomach-churning anxiety; comedians delivered dark jokes about getting mugged in cities, especially in Manhattan; suburban teens reported fearing the central cities a short drive away. It was a big reason that major corporations moved out of Manhattan.
Yet, today, the image of the big city (and of Manhattan in particular) seems different, infused with romance, allure, excitement, luxury, a place of aspirations – think Sex and the City — rather than fear and avoidance. This is actually a back-to-the-‘50s sensibility about the big city.
In an earlier (and since updated)post, I discussed how violent crime in the United States dropped suddenly and continuously from about 1990 through 2010 and showed how the drop was the last stage of a down-up-down cycle starting after World War II. One aspect of recent drop is the way it has changed the geography of violence. The connection between big city and big crime has weakened greatly; big cities are not as dramatically the places of danger. We may be moving toward an older pattern: that cities are the safer places to live.
City and Suburb (and Country)
In a recent Brookings Institution paper, Elizabeth Kneebone and Stephen Raphael looked at violent and property crime rates in the United States from 1990 to 2008 in the 100 largest metropolitan areas of the country, areas where two out of three Americans live — urban conglomerations that range in size from Ogden, UT, and Modesto, CA, to Chicago and New York City. The authors describe the trends in crime separately for the central cities versus the suburbs of those regions.
On average, the rate of violent crime – the risk of being victimized by murder, robbery, assault, rape and the like –dropped about 30 percent in the center cities of metropolitan areas and by only about 7 percent in their suburbs. Most dramatically, rates of violent crime dropped in the center cities of the New York region by about 75% and of the Chicago region by 80%. The center cities still have higher rates of violent crime than suburbs do, but instead of being about triple the suburban rates as they were in 1990, they are now less than double the suburban ones. The city-suburb gaps narrowed in 62 of the 100 areas.
The Kneebone-Raphael analysis is more complex than this summary suggests. For one, they also looked at property crimes; the city-suburban convergence there is even sharper. And they divided suburban communities into four types – dense inner suburbs, “mature suburbs,” “emerging suburbs,” and the still pretty rural “exurbs.” As the graph below shows, on average, the inner suburbs also experienced major declines in rates of violent crime, although not as great as the center cities. But the more outlying suburbs experienced little change or even increases in rates of violent crime. The inner regions of our metropolitan areas have shown major declines, the outer regions modest ones, at best.
(By the way, the authors’ analysis shows that a growing Hispanic population is associated with a decline in violent and property crime. See also here.)
Pushing back further in history, it turns out that the violent crime wave which surged upwards in the 1960s-‘80s occurred first, fastest, and foremost in the centers of the large metro areas and rose only later and lesser in more outlying areas. I describe that widening of city versus small town experiences in an old paper (Fischer_Spread of Violent Crime_RurSoc_1980). That is, violent crime became a particularly big-city phenomenon about 40 years ago and now has become less so in the last 20 years.
Indeed, if one pushes yet further back in time, the general rule is that rural, particularly frontier, areas were more crime-ridden than the big cities. An early statistician wrote a paper on murder in Massachusetts in the last decades of the 19th century and concluded that “there is less resistance to the homicidal impulse in non-urban communities than in the urban ones.” Homicide is higher in “the sparsely settled rural districts, particularly in the four western counties, [which] have the blackest records.” Waldo Cook described the murders in one of those rural counties:
There was the murder by the man who had a grudge, and who had brooded over it for a long time in the solitude of the woods; there was the murder for the sake of money by depraved country boys; there was the murder and attempted suicide caused by unbridled lust of woman; and, finally, there was the murder and attempted self-murder by a man whose mind had evidently become unbalanced through the hard conditions of a small farmer’s life on the hillsides of Western Massachusetts.
In the cycle of post-World War II criminal violence – a decline to about 1960, a sharp rise to about 1985-90, a sharp drop to today – the large cities led the way down, up, and down (or so I argued in the 1980 paper). Or perhaps it’s just that crime cycles sooner and faster in larger cities.
It’s hard to make sense of the cyclical and the geographic pattern if one thinks about crime and violence as direct reactions to material circumstances. But if we think of crime as having a cultural element, then we might find one way of understanding what’s been going on.
In a 1996 article, science popularizer Malcolm Gladwell asked rhetorically whether crime was like an epidemic, by which he meant that it spread from one person to one another in ways that created huge peaks of “infection” and then dissipated. Epidemics of disease are essentially one kind of “diffusion” – diffusion of germs, not unlike diffusion of fashions, of ideas. Perhaps crime, too, spreads in the same way that new norms, habits, and products spread, and that is, they typically spread from large cities outward.
Perhaps there’s another explanation for the widening and closing of city-suburban-rural differences in violent crime. Kneebone and Raphael tried to explain the recent closing of the gap by looking at the demographic and economic features of communities, but those did not account for the pattern. Something else is going on.
Whatever is going on, the convergence of violent crime rates suggests that the automatic reflex Americans developed in the ‘70s and ‘80s that associated big cities with big danger may be fading. Welcome back, UBS.
Cross-posted from Claude Fischer’s blog, Made in America: Notes on American life from American history.