Randal O'Toole: Highway Funding and Urban Form

Many opponents of low-density suburbs — areas they derisively call “sprawl” — argue that Americans would not have chosen to live in such areas unless they were subsidized or forced to do so. One of the most important such subsidies, they claim, is the Interstate Highway System.

“For more than a generation,” argues former Milwaukee Mayor and current head of the Congress for the New Urbanism John Norquest, “urban sprawl sprung up with federal assistance [such as] excessive road building . . . that interfered with the free market.” He adds that, “urban superhighways should be relegated to the scrap heap of history.”

Would our cities look a lot different if the federal government had not built the urban interstates (which were the first major urban highways built with federal assistance)? I argue that the differences would be minor.

First, unlike many local roads, 100 percent of the cost of the interstate system was paid for out of gas taxes and other user fees (tolls plus taxes on autos, trucks, and tires that were created to pay for such roads). The urban interstates make up 42 percent of the lane miles of the interstate system but carry 66 percent of the vehicle miles of travel, which makes up for much of the differences in the costs of urban vs. rural roads. While gas taxes are a poor user fee, this shows that people were willing to pay build the urban interstates, so most would have been built even if they were left to the states or private highway companies.

More important, the interstates were not the primary force shaping our urban areas. This is illustrated by a 1927 New York Times review of what is now considered a classic film, Metropolis. This movie depicted a futuristic city consisting almost entirely of skyscrapers, but the reviewer — none other than H.G. Wells — argued that this was “silly.” Such a “vertical city of the future” is “highly improbable,” he said, because cities had already begun decentralizing (or, as the English put it, “centrifugal”) years before the automobile became popular. “The British census returns of 1901 proved clearly that city populations were becoming centrifugal,” says Wells, “and that every increase in horizontal traffic facilities produced a further distribution.”

Those “horizontal traffic facilities” include the horsecar in 1832, the electric streetcar in 1888, and Henry Ford’s mass-produced automobile in 1913. Each of these gave a new class of people the mobility they needed to escape the dense cities — and escape they did. By 1922, Ford himself predicted that most people would soon live outside of the cities. “Cities are doomed,” he said, adding, “We shall solve the city problem by leaving the city.”

In the same year, Frank Lloyd Wright pointed out that the automobile, electricity, and telephones together eliminated any advantages to living or working in dense areas. The “skyscraperites,” as Wright dismissively called those who wanted to build up, not out, were following a “blind alley.”

Census data show that the population of Manhattan, the closest American city to the Metropolis vision, peaked in 1910. By 1950, before any interstate had been built, its population had fallen by 16 percent. By 1960, when most interstates were still only on the drawing boards, Manhattan’s population had fallen a total of 28 percent. Today it is around 30 percent less than in 1910, meaning only a small portion of the loss took place after most interstates were built.

The same story can be told of other American core cities. Yes, some of them lost population after the interstates were built, but they were already declining before.

The automobile, not federal highways, enabled people to move out of the cities. If anything, as Harvard planning Professor Alan Altshuler once pointed out, the interstates actually slowed the decline of downtowns by relieving the traffic congestion that many people and businesses were trying to escape.

The automobile would not have led people to move to low-density suburbs if they didn’t want to live in such suburbs in the first place. As previously noted, before 1900 wealthy and middle-class people were already moving to suburban areas thanks to electric streetcars. The automobile democratized mobility and the suburbs, offering working-class people the opportunity to own their own homes. (Sadly, it too often appears that what many anti-sprawl types want is a return to a more stratified society in which only the well-to-do enjoy low-density suburbs and the working class are confined to higher-density areas.)

Nor are suburbs a purely American phenomenon. When Peter Newman and Jeffrey Kenworthy compiled their massive databook on Cities and Auto Dependence, they thought they were proving that European cities were less auto-dependent than American and Australian ones. Instead, their 1960-1990 time series data showed that all cities were rapidly decentralizing thanks to increasing auto driving. Since Asian and European cities were denser to start with (and had lower average incomes), they just appeared to be less affected by decentralization than their younger counterparts if you only looked at one year, not the entire time series. What Newman and Kenworthy call “auto dependence” is really “auto liberation.”

I am not saying our urban areas would look exactly the way they do today without the federal interstate highway program. A system of toll roads driven by profits, or at least the need for each road to cover its costs, would have avoided many of the costlier parts of the Interstate Highway System. It is impossible to predict just exactly what differences this would make.

But with or without interstates, most Americans would still be living in single-family homes on moderately large lots in neighborhoods that separate residential uses from businesses. Those who advocate a return to nineteenth- or early-twentieth-century patterns of development mast face the fact that they are going against the desires of most Americans. The regulatory regimes required for a return to more compact development are not only undemocratic, they will have huge unintended and undesirable consequences.


This post, composed by Randal O'Toole, is jointly posted at the Antiplanner blog.