An Index of Sprawl (Pittsburgh Is Worst)

Peter Coy

You could argue that sprawl is a phony issue.

After all, people have to live somewhere. And why should those ensconced in suburban and rural splendor be allowed to close the gates and prevent more people from coming to live with them?
On the other hand, it's clear that sprawl does matter to a lot of people. A 2000 survey by the Pew Center for Civic Journalism found that 18% of Americans said urban sprawl and land development were the most important issue facing their local community. That was the top response, tied with crime and violence. (Does that mean having a development go up in the cornfield down the road street is as bad as getting mugged?)
Now, there's some fascinating research into the extent and causes of sprawl. First, here are some stats from the research, which appears in the May issue of the Quarterly Journal of Economics:

(For metro areas in 1992, the percentage of undeveloped land in the square kilometer surrounding an average residential development.)

Pittsburgh 57.7
Atlanta 55.6
Charlotte 52.7
San Diego 45.6

Chicago 31.8
New York 28.9
Phoenix 27.6
Memphis 27.4
Miami 20.7

The biggest surprise to me is Phoenix ranking so low in sprawl. But it makes sense when you understand the researchers' methodology. They looked at high-altitude photos of the U.S. in 1976 and 1992 to see how sprawl occurred. They chopped the country into nearly 9 billion quarter-acre parcels. They figured that if residential development occurred in areas that were previously empty, that was sprawl. If new houses were squeezed into gaps in already developed areas, that was not sprawl. Phoenix has grown like mad, but its housing consists mostly of big subdivisions, not dinky houses set off by themselves.
One thing that came across is that the United States is one heck of a large country. Even in 1992, they found, only 1.9% of the country was built-up or paved, up from about 1.3% in 1976. Another finding is that sprawl didn't get worse. Yes, more of the country was developed. But builders weren't any more likely to leapfrog into virgin areas in 1992 than they had been in the earlier period.
Here are some of the things that seem to make sprawl more likely:
*Available groundwater
*A temperate climate
*Rugged terrain
*Decentralized employment
*Early public transportation infrastructure
*Uncertainty about metropolitan growth
*Unincorporated land in the urban fringe

The authors of the paper, "Causes of Sprawl: A Portrait from Space," are Marcy Burchfield of the Toronto-based Neptis Foundation, Henry G. Overman of the London School of Economics, and Diego Puga and Matthew A. Turner of the University of Toronto.

Before it's here, it's on the Bloomberg Terminal.