Cars and the 'Burbs: A Policy Question

by: Ryan Avent

Liberal bloggers appear to have gotten results, and Tyler Cowen has recently been blogging on metropolitan development. On the whole, his posts are pretty interesting reads, though urbanist readers won’t find much new there. Tyler has also generated some nice responses, including a couple from Mike Konczal.

At the heart of many of Tyler’s posts are questions about what drove the growth of the suburbs. This is a question that really must be framed appropriately to make sense.

If one refers to suburbs as development outside the traditional or center city, then suburbs have basically been around as long as cities. Some share of population growth has always been accommodated by development at the urban fringe. Development of new transportation technologies changed the pace and the form of suburban developments over time, as did the increase in population growth associated with industrial development. But outward growth is a natural part of urban development.

When we talk about the phenomenon of rapid suburban growth in America, we’re largely asking questions about why that growth took a particular form and why that growth coincided with decline in center cities. Obviously, new suburban growth forms were largely a function of the automobile, but that’s not the end of the story. Specific choices were made to accommodate the automobile in various ways, and those choices affected decisions at the margins, and given the importance of feedback loops in urban settings these choices were potentially quite powerful in certain circumstances.

Was the automobile so disruptive a technology that center cities would have experienced some decline no matter how governments chose to accommodate it? Probably, but it’s difficult to know given the many, many factors influencing metropolitan growth over the last half century.

Economic and income growth is one of those factors, and I think that suburbia would have arisen no matter what based in part on such growth. But other issues were key. Desegregation cannot be underestimated as a factor driving suburbanization. The phenomenon of tipping points is quite real. Discriminatory lending reinforced this process. If you wanted your kids to play with other middle-class white kids and go to school with other middle-class white kids, you moved to the suburbs.

Highway construction, particularly within center cities, had powerful impacts. The destructive effect of construction in cities on neighborhoods reduced the attractiveness of living in such neighborhoods (cut off from other parts of the city, overshadowed by noisy, polluting highways) while also reducing the cost of commuting in from the suburbs. As Tyler mentions, policy competition, including tax rate competition, was another factor driving sprawl. Given the minimal infrastructure load in suburbs and low levels of poverty and crime, residents could move just a few miles out and substantially lower their tax burdens.

Throughout these processes there were feedback loops. The government would have had to work hard to prevent urban decline given the social and economic changes taking place, and as it happened the government worked pretty hard to encourage growth in suburbs.

Suburbanization was partly about preferences, partly about technological change, partly about government policy choices (at local, state, and federal levels), and partly about complicated and unexpected interactions of the three. It was not the perfect expression of consumer demand, or the result of the untrammeled working of the free market.

Stepping back a bit, choices about where and how to live (and build) come down to a tension between a household desire for space and privacy and a desire for access to economic opportunities in production and consumption. Here again we see a world in which preferences vary across the population (and change within members of the population over time), where policy choices influence decisions at the margins, where household decisions are influenced by what other households are choosing, and where the net result is a complex and dynamic system which can behave in unpredictable ways.

In such a system, questions like “How much road should we build,” are really difficult to answer.