I often write negatively about suburbs, and as a result I’m frequently faced with comments that generally fall into two categories of criticism: 1) people really like living in suburbs, and 2) cities just don’t seem to work as well as suburbs. Today, via Brad DeLong, I found this paper (PDF), which collects and expands upon many of my responses to the above criticisms. A snippet:

In the decades following World War II, American cities underwent a period of rapid suburbanization. The rich were the first to leave central cities, in part because they had a comparative advantage in commuting by car (Leroy and Sonstelie, 1983; Margo, 1992; Glaeser, Kahn and Rappaport, 2000). At the same time, cities were receiving an in-flow of poor migrants from the South. Taken together, these two trends led to an income inversion: while, in 1950, the average city resident was wealthier than the average resident in the surrounding suburban ring, by 1970, the opposite was the case.

*snip*

If part of the attraction of the suburbs is their wealthy, homogenous population, then an initial event that prompts the rich to leave the city – say, the construction of a new highway - could generate follow-on mobility, leading to a downward spiral of population loss and urban decay. This process might explain why some cities went through dramatic and rather sudden declines in the post-War period.

There are a couple of key points in the piece. The first is that there were a number of initial factors that fed into suburbanization. One was the outward movement of the richest residents of cities, but one imagines that pales next to the effect of highway construction (which eased commuting into the city and destroyed or isolated many urban neighborhoods) or of residential mortgage guarantees in the suburbs. Anther factor is the migration of poor blacks out of the south and into cities. One key point about this migration is that unlike European movements of the early part of the century which were due in large part to the pull of economic growth in cities, black migration came largely from the push of poor economic and social conditions in the south. Often these populations entered northern cities just as urban industry was moving out of expensive urban cores, leaving migrants with fewer good economic options and helping to entrench urban black poverty.

The second key point is that initial movements contributed to a feedback loop that continued to tip the balance in favor of suburbs. Outer jurisdictions were wealthy with few infrastructure demands, meaning that excellent services could be provided at a low tax burden. Central cities, by contrast, were growing poorer as infrastructure demands increased. This balance created pressure on upper middle class families in cities to leave for suburbs, which further impoverished cities and led to higher urban taxes and poorer services and still richer suburbs. For decades this feedback loop helped to empty out and bankrupt cities while entrenching suburbs as places of wealth and responsive government.

The final key point is that this loop had, at its core, a desire among people to live near wealth and homogeneity. Whites wanted to live around wealthy whites, and diversity itself was felt to be a cost helping to push city residents outward. The paper linked above interprets this, in part, as an increase in the cost of negotiating appropriate city policies. In a diverse society, policy formation is more costly.

All of which is very interesting. What’s really fun to think about are the ways in which this has all changed recently. On the one hand, suburbs have become more diverse in economic and demographic terms. On the other hand, suburbs have become sufficiently crowded that their relative infrastructure deficiencies are now a serious problem. It’s no longer easy to commute in suburbs or easy to provide a high level of service without a large tax burden (see this). Meanwhile, the density of older cities (and in many cases the diversity) has become an advantage, enabling a higher quality of life.

The result is not hard to predict; once again, the wealthy are leading the way, but this time back into cities. In New York, Washington, San Francisco, Chicago, London, and other metropolises, urban living is now a luxury good. The challenge now is to try and duplicate in suburbs the aspects of urban life that allow cities to succeed amid large and growing populations.

What I would love to study, if I had the time, would be which process (suburbanization or re-urbanization) has contributed more to overall economic growth. Which method of societal organization, in other words, is better for the economy (and for the distribution of economic growth)? I suspect that the latter is preferable, but I’ll be on the lookout for empirical sudies proving or disproving that speculation.