Let’s Pretend

Faulty smart growth survey is misleading

Given the choice between heaven and purgatory, which would you choose? Most rational people would choose heaven. Thus, it wasn’t surprising that when given a choice between an idealized urban village and a suburban nowheresville, a survey released by Smart Growth America (SGA) and the National Association of Realtors (NAR) found that most people chose the former.

According to the press release for SGA-NAR’s 2004 American Community Survey, “Asked to choose between two communities, six in ten prospective homebuyers chose a neighborhood that offered a shorter commute, sidewalks and amenities like shops, restaurants, libraries, schools and public transportation within walking distance over a sprawling community with larger lots, limited options for walking and a longer commute.”

SGA and NAR were able to elicit this outcome only by presenting survey respondents with carefully structured information and choices. Respondents were given a choice between a community with one-acre lots and a 45-minute-plus commute or a mixed-use community where everything is within a few blocks of home (see Figure 1). But these aren’t the choices real people face. SGA-NAR forced respondents to choose between a suburb that sounds like it’s in the middle of nowhere, and an urban nirvana that doesn’t exist anywhere. What’s extraordinary is that they got only 60 percent of the respondents to choose the latter. Even with the deck stacked against suburbs, an amazing 40 percent of Americans apparently find suburban purgatory superior to urban heaven.

Figure 1. Survey Respondents Had to Choose One of these Two Hypothetical Communities

Source: SGA-NAR, 2004 American Community Survey, p. 4.

What’s misleading about these choices? First, few homes are on the large lots imagined by SGA. According to the Census Bureau, 87 percent of new single-family detached homes are built on lots of less than half an acre – the highest lot size for which the data were broken out.[1] Sixty-six percent were on less than a quarter acre. So a realistic suburb would have several times the population density implicit in the SGA-NAR survey.

Second, few Americans commute more than 45 minutes, and the longest commute times occur in the dense urban areas favored by smart growthers. The Census Bureau estimates that the average work commute takes just over 25 minutes, and only 15 percent of workers have a commute longer than 45 minutes.

Of course, a national average could be misleading. Perhaps suburbanites are the ones on the high side of that average. Though it will come as a surprise to those on an information diet high in smart growth orthodoxy, dense cities with extensive public transit systems are where commutes are longest. New York City tips the scales at an average of 38 minutes — the highest in the U.S. Chicago comes in a distant second at 33 minutes. In contrast, sprawling areas like Atlanta, Houston, and Phoenix are right around the national average.

And it isn’t just New York City as a whole that has long commutes. Even in Manhattan, a paragon of density and mixed-use urbanity, the average commute takes about 30 minutes — 30th out of 233 urban counties ranked by the Census. Thirteen out of the worst 30 counties are in the New York metropolitan area, which consequently accounts for a substantial fraction of the roughly 19 million Americans who suffer those 45-minute-plus commutes. Commuting takes longer in denser cities, because denser cities have more traffic congestion and because traveling by transit, which is more common in denser cities, is slower than traveling by car.[2]

Perhaps more misleading than the fake choices is the vague language used to describe them. Note in Figure 1 that in the smart-growth community respondents are told what kind of housing exists in the area, but not the specific kind of housing they would actually live in. Respondents are also led to believe that everything they would want or need to do (other than work) is within walking distance.

Did people who chose the smart-growth community imagine themselves living in one of the single-family homes or in a townhouse; on a large lot or small one? Would their responses have changed if they were told they had only a 25 percent chance of living in a detached home on more than 1/10th of an acre?

When the respondents imagined all those stores and restaurants and leisure activities within walking distance, were they thinking of a random collection of such amenities, or did they assume they would have their particular favorites nearby? Would their response have changed if they knew what fraction of nearby businesses they would actually patronize, or how often they would need to drive or take transit to get to the specific places where they wanted to shop? If they had time to reflect, would anyone seriously believe there’s any community where everyone can have their pasta, sushi, tapas, burritos, bagels, groceries, clothing, books, and half-caf-no-fat lattes all within a few blocks of their house? For good measure, let’s add soccer practice, piano lessons, the pediatrician, and the health club.

What people say in surveys and what they actually do are two different things. The reason for the difference between “say” and “do” is that in surveys you don’t have to make tradeoffs among a range of competing goals and aspirations, but in the real world you do. When it comes to housing and community, I suspect that what most people really want is something like this:

  • To live in a single-family home with a substantial yard, within walking distance of their favorite restaurants and stores, on a street with no traffic, and in a neighborhood with little traffic congestion.
  • A short commute on clear highways, but no highways near their house.
  • Convenient public transit that other people will use.

Since you can’t have all of these things at the same time, people have to make choices and tradeoffs, and they naturally make them based on their particular values, aspirations, and financial means. In the end, most people who can afford it opt for a single-family home with a yard, travel mainly by automobile, and a commute of maybe 20 to 40 minutes on roads that are likely to be relatively congested.

One might argue that these choices are to some extent a function of public policy, rather than innate human aspirations. To be sure, policies such as minimum parking requirements encourage driving, while minimum-lot-size and other zoning requirements can discourage or prevent high-density or mixed-use housing, even when demand for such developments exists. Ideally, we could say good riddance to such meddling. But Europe’s experience suggests that space, privacy, and mobility are deep-seated human desires that go well beyond these more marginal policy effects. Even with extensive public transit and five-dollar-per-gallon gasoline, per-capita driving has skyrocketed in European cities during the last few decades, and most residents of European cities live in suburbs that look very much like those in America.[3]

Rather than a program to help Americans achieve their dreams, smart growth seems to be more about helping anti-suburb activists impose their own vision on the rest of society. Smart-growth advocates wish it were otherwise, but people tend to choose a suburban, auto-based lifestyle once they and their society achieve the necessary wealth to afford it.

Joel Schwartz is an adjunct fellow at Reason Foundation and visiting scholar at American Enterprise Institute.


[1] There is great variation by region, however. In the Northeast, 68% of houses are built on less than half an acre. In the West, the figure is 95%.

[2] Based on data from the 2001 National Household Travel Survey, in metropolitan areas, the average transit commute takes 56 minutes, while the average automobile commute takes 23 minutes. Although per-capita driving decreases with increasing density, it doesn’t decrease nearly enough to offset the concentration of driving into a smaller amount of land area. Peter Gordon, Bumsoo Lee, Harry W. Richardson, “Travel Trends in U.S. Cities: Explaining the 2000 Census Commuting Results” (Los Angeles: Lusk Center for Real Estate, University of Southern California, April 2004), Brian Taylor, “Rethinking Traffic Congestion,” Access, Fall 2002, pp. 8-16, Wendell Cox, “How Higher Densities Make Traffic Worse,” The Public Purpose, 57, 2003,

[3] Randall G. Holcombe and Samuel R. Staley eds., Smarter Growth: Market-Based Strategies for Land-Use Planning in the 21st Century.