Simulating sprawl reduction: We're not England
A recent post at Atlantic Cities, reporting on an article in the Journal of the American Planning Association, suggests that reining in urban sprawl may produce more costs than benefits. “Smart growth” advocates have vigorously attacked the article for not clarifying the assumptions behind its simulation model of growth and its impacts in 30 English cities.
Less has been said, however, about whether and how the simulations should matter to policymakers on this side of the pond. The same methods applied to American metropolitan areas would likely produce greater benefits and lower costs, for at least two reasons.
First, England’s baseline conditions are much more compact than those in the United States. For example, the median lot size of new single-family houses in Atlanta in the early 2000s was six-tenths of an acre, according to the Census’s American Housing Survey. In Dallas, the median was about a quarter acre. If every lot needs an equal amount of land for streets and community facilities, 100,000 new houses built at the median lot size would need 120,000 acres of land in Atlanta but 50,000 in Dallas. Atlanta’s large lots are at least partly a product of exclusionary zoning by suburban jurisdictions whose minimum lot requirements exceed what buyers appear to prefer in less-regulated metro areas such as Dallas and Houston. Deregulation in Atlanta, and in many other American metropolitan areas, would therefore increase density at the urban fringe. England’s land-use regulations work in just the opposite way, producing densities higher than many households prefer. Even a small increase in density there might reduce satisfaction, while a doubling of density in suburban and exurban Atlanta might even increase overall satisfaction.
At the same time, boosting density at the far-flung margins of America’s fastest-sprawling metropolitan areas will reduce the total number of miles driven by vehicles in the area (measured as “vehicle-miles traveled” or VMT) much more than a similar proportionate increase in England. Boosting density here would sometimes mean shrinking single-family lot sizes from, for example, half acres to quarter acres, while in England it would mean shifting from houses on small lots to townhouses. And the number of destinations one can reach in a given period of time increases much more—and potentially, vehicle-miles traveled will decrease more—when single-family lot sizes shrink.
Second, England expects much less population growth in the coming decades than does the United States. Consequently, new land development naturally will play a smaller part in its path to greenhouse gas emissions than it might in the United States.
The benefit-cost ratio from compact development also varies among U.S. metro areas. The gross impact of allowing suburban lots to reach market-equilibrium sizes in metros stretching from Richmond to Birmingham would appreciably reduce vehicle-miles traveled compared with what will occur if lots continue to be so large. In metro areas such as Los Angeles and Las Vegas, by contrast, new development is already occurring at relatively high densities, so their path to VMT reduction will depend more on changing how people travel than will be true in Atlanta or Nashville. New development is occurring at low densities in Pennsylvania and Ohio too, but neither of these states is projected to account for much of the nation’s growth.
Ultimately, urban simulation models hold significant promise for decision-making. They offer ways to avoid one-size-fits-all policy prescriptions, showing how and where various policies threaten vulnerable people and important environmental assets. It’s ironic that the main message emerging into the popular consciousness from this article is a one-size-fits-all conclusion that may not apply to a single U.S. metropolitan area—much less to all of them.
[A note on sources: A few people have asked me since the first posting for suggested sources, especially to follow up my contention about density and VMT. Some excellent recent ones include:
Bento, A.M., Cropper, M.L., Mobarak, A.M., Vinha, K., 2005. The Effects of Urban Spatial Structure on Travel Demand in the United States. The Review of Economics and Statistics 87, 466–478.
Boarnet, M.G., Houston, D., Ferguson, G., Spears, S., 2011a. Land use and vehicle miles of travel in the climate change debate: getting smarter than your average bear. In: Hong, Y.H., Ingram, G. (Eds.), Climate Change and Land Policies. Lincoln Institute of Land Policy, Cambridge, MA.
Boarnet, M.G., Joh, K., Siembab, W., Fulton, W., Nguyen, M., 2011b. Retrofitting the suburbs to increase walking: evidence from a land use – travel study. Urban Studies 48, 129–159.
Salon, D., 2009. Neighborhoods, cars, and commuting in New York City: a discrete choice approach. Transportation Research Part A 43, 180–196.
Salon, Deborah, Marlon G. Boarnet, Susan Handy, Steven Spears, and Gil Tal. 2012. How do local actions affect VMT? A critical review of the empirical evidence. Transportation Research Part D: Transport and Environment, Volume 17, Issue 7, Pages 495–508.]