A few weeks ago, I wrote a short piece (paywalled) on gun regulation that touched on rising suburban crime rates. While the homicide rate in large U.S. cities declined by 16.7 percent from 2001 to 2010, it increased by 16.9 percent in suburbs over the same period. Though the overall number of murders had declined over this period, the gap between large cities and suburbs decreased considerably. I mention this because Joel Kotkin and Wendell Cox have a new article on rising poverty levels that makes the following point:
Despite substantial improvement in crime rates in “core cities” over the past two decades, suburban areas generally have substantially lower crime rates, according to Brookings Institution’s own research. Yet at the same time suburban burgs dominate the list of safest cities over 100,000 led by Irvine and Temecula, Calif., followed by Cary, N.C. Overall suburban crime remains far lower than that in core cities.
A review of 2011 crime data, as reported by the FBI, indicates that the violent-crime rate in the core cities of major metropolitan areas was approximately 3.4 times that of the suburbs. (The data covers 47 of the 51 metropolitan areas with more than 1 million population, with data not being available for Chicago, Las Vegas, Minneapolis-St. Paul, and Providence.)
In the least suburbanized core cities, that is places that have annexed little or no territory since before World War II (New York, Philadelphia, Washington, etc.) the violent crime rate was 4.3 times the suburban rate. Among the 24 metropolitan areas that had strong central cities at the beginning of World War II but which have significant amounts of postwar suburban territory (Portland, Seattle, Milwaukee, Los Angeles, etc.), the violent crime rate is 3.1 times the suburban rate. Among the metropolitan areas that did not have strong pre–World War II core cities (San Jose, Austin, Phoenix, etc.), the violent crime rate was 2.2 times the suburban rate. Basically, the more suburban the metropolis, the lower the crime rate. Rather than castigating suburbs for exaggerated dysfunction, retro-urbanists would be much better served focusing on how to correct and confront the issue of poverty, which continues to concentrate heavily in the urban core and elsewhere in America.
Moreover, Kotkin and Cox’s reference to crime levels across cities raises the question of whether or not there are other variables at work other than density. What happens when we correct for police officers per capita, for example, or the demographic composition of the population? It is certainly true that average density varies across Philadelphia and Portland and San Jose, but of course these cities are different in many other ways as well — mean January temperatures, concentration of college-educated workers, etc. Not all core cities are created equal.
More broadly, I think Kotkin and Cox misunderstand why the rise of suburban poverty is a problem. They note that poverty levels continue to be much higher in core cities than suburbs, which is of course true, and that the rise of suburban poverty in part reflects the fact that there has been a continuing shift from central cities to suburbs:
Many poor suburbs are developing because minorities and working-class populations are moving to suburbs. Yet even accounting for these shifts, cities continue to contain pockets of wealth and gentrification that give way to swathes of poverty. In Brooklyn, it’s a short walk east from designer shoe stores and locavore eateries to vast stretches of slumscape. The sad fact is that in American cities, poor people—not hipsters or yuppies—constitute the fastest-growing population. In the core cities of the 51 metropolitan areas, 81 percent of the population increase over the past decade was under the poverty line, compared to 32 percent of the suburban population increase.
In Chicago, oft cited as an exemplar of “the great inversion” of affluence from suburbs to cities, the city poverty rate stands at 22.5 percent, compared to 10 percent in the suburbs. In New York, roughly 20 percent of the city population lives in poverty, compared to only 9 percent in the suburbs.
As Cox has acknowledged, however, core-city population growth has been small in absolute terms. And the core cities of the 51 metropolitan areas in question vary dramatically in terms of economic health. New York City is an increasingly desirable destination. Cleveland and Detroit are less desirable destinations now than they might have been in decades past. Lumping these cities together tells us something, but I’m not sure it tells us something we don’t already know, e.g., that there is a kind of tournament dynamic at work in U.S. urban regions, in which the most productive and amenity-rich regions are gaining at the expense of others, subject to affordability constraints.
And this leads us to the most important issue: the fact that a fifth of New York city’s population lives in poverty while the same is true of only 9 percent of the population in its suburbs doesn’t represent a failing — rather, it reflects the fact that density and the widespread availability of mass transit are particularly valuable to the poor, who find it more difficult to purchase and maintain automobiles and for whom density facilitates greater access to service jobs. Commuting from Hempstead, a Long Island community with relatively high poverty levels, to a service job in New York city’s urban core is more time-consuming and expensive than commuting from Brooklyn or Queens. Commuting from Hempstead to one of Long Island’s affluent suburban neighboods can also be time-consuming and expensive, particularly when there are no direct transit links, as is often the case. So suburban poverty poses problems that poverty in dense cities well-served by transit does not. The problem we face is that the U.S. has relatively few dense cities that are well-served by transit, as such cities can greatly facilitate upward mobility for the very poor.