Density no solution for traffic, obesity problems

Published February 1, 2002

The January issue of Environment & Climate News described a recent report published by Sprawlwatch as a “fake CDC” study. (See “Fake CDC study full of holes,” Environment & Climate News, January 2002.) Although the report was written by Centers for Disease Control employees, it was not published by the CDC and not apparently endorsed by that agency. Yet many news reports described it as a CDC study, not a Sprawlwatch study.

While not the publisher of the study, the CDC does address issues similar to those covered in the report. The agency runs an “Active Community Environments” program that combines land use, transportation, and public health issues. According to its Web site (http://www.cdc.gov/nccdphp/dnpa/aces.htm), the program aims to promote walking and cycling.

The CDC program has published two reports: “How Land Use and Transportation Systems Impact Public Health: A Literature Review,” and “How Land Use and Transportation Systems Impact Public Health: An Annotated Bibliography.”

Although the reports can be downloaded from the CDC Web site, they were published by the Georgia Institute of Technology. Their title page contains the statement, “The content does not necessarily reflect the official views or policies of the Centers for Disease Control and Prevention or the Georgia Institute of Technology.”

Unlike the authors of the Sprawlwatch study, the authors of the GIT reports, Lawrence Frank and Peter Engelke, are not doctors but urban planners affiliated with the Georgia Institute of Technology city planning school.

Biased literature review

Though the first report is described as a “literature review,” it is clear the authors reviewed only the literature that supports their view. For example,

  • They cite Newman and Kenworthy’s claim that miles driven is related to density, based on the fact that Americans drive more than Europeans. The authors fail to cite work by UC Irving economist Charles Lave, who explains that differences between American and European driving are mostly income-related. (“Cars and Demographics,” Access, Fall 1992)
  • Frank and Engelke cite Sierra Club member John Holtzclaw’s study claiming Americans in high-density neighborhoods drive less than Americans in low-density neighborhoods. They fail to cite MIT planner Paul Schimik’s finding that differences in neighborhood driving are almost entirely explained by income and family size. (“Household Motor Vehicle Ownership and Use: How Much Does Residential Density Matter?” Transportation Research Board, 1996)

Frank and Engelke do cite USC planners/real estate experts Genevieve Giuliano and Harry Richardson, who explain that densities would have to be hugely increased to have any effect on driving at all. But Frank and Engelke dismiss the Giuliano/Richardson work by citing, again, Holtzclaw.

Density doesn’t determine driving

The problem with Holtzclaw’s study is that people are likely to choose the neighborhood they live in based on whether or not they want to drive. If they want to drive, they choose a low-density, auto-friendly neighborhood. If not, they choose a denser, transit-intense neighborhood.

If Holtzclaw’s claim that density determines driving were true, then differences in driving among urban areas would be strongly correlated with urban area densities. People may express their preference to drive or not drive when they choose their neighborhoods, but they are less likely to choose the urban area they reside in based on their driving preferences. Thus, the preference question is less important and density becomes more important.

U.S. urban areas range in density from little more than 600 people per square mile to nearly 6,000 people per square mile. Yet there is almost no correlation between density and miles of driving or commute mode. Thus, it is reasonable to conclude that the differences in driving levels found by Holtzclaw are based largely on preferences, not urban form.

For example, the 1990 Census found that 94.6 percent of commuters drove or rode autos to work in the nation’s least-dense urban area, while 90.3 percent drove in the densest urban area. This is not good news for planners who want to increase their urban area’s density.

Fearmongering and propaganda

The Executive Summary of the literature review quotes Benjamin Franklin as saying:

“To get the bad customs of a country changed and the new ones, thought better, introduced, it is necessary first to remove the prejudices of the people, enlighten their ignorance, and convince them that their interests will be promoted by the proposed changes; and this is not the work of a day.”

This is a call for public health officials to join in propaganda campaigns to convince the public to support government action on environmental and social issues. Scientists in the past have stayed out of political debates. Today, more and more scientists appear willing to stretch the truth to raise public attention over problems they believe exist, even when they can’t prove those problems are real.

For example, one of the assertions in the Sprawlwatch report is that American children are getting more obese, and that this is related to suburban driving. While it is true that an “epidemic” of child obesity has emerged in the past two decades, the cause is not the suburbs. No information in the Sprawlwatch report or any other study has indicated suburban children are more obese than urban ones.

Instead, the cause appears to be the hysteria raised over missing and abducted children in the 1980s. Before this campaign, the vast majority of school children walked to school. After it, two out of three are driven to school by their parents. Densifying the suburbs is not going to fix this.

Moreover, child obesity is a problem in Canada (http://www.greenestcity.org/walk/ReasonsToWalk.pdf), Britain (http://www.bupa.co.uk/healthy_living/children/c_obesity/), and even India (http://www.clinic-2000.com/childhood.htm), which can hardly be accused of low-density sprawl.

In all three places, the problem is blamed in part on kids not walking to school. “Either they travel by bus, rickshaw, auto or they get dropped at the school over a motorcycle or by a car by one of the family members,” says the Indian Web page. Since Canada and Britain each have more aggressive smart-growth programs than the U.S., we are not likely to cure child obesity through smart growth.

This is not to say the CDC’s concerns for walking and obesity are not valid. It is also true that more pedestrian-friendly and bicycle-friendly urban designs could help make exercise more common. But such designs do not need to be auto-hostile, which is the real goal of many smart-growth advocates.

It appears the CDC has funded and is distributing an unscientific report that rehashes old claims that urban form influences travel habits. While that report states it is not the official view of the CDC, there can be little doubt that sprawl opponents aim for it to become the official view of the CDC and other public health agencies.


Randal O’Toole ([email protected]) is a senior economist with the Thoreau Institute (http://www.ti.org) and author of the recent book, The Vanishing Automobile and other Urban Myths.


For more information . . .

The two CDC reports–“How Land Use and Transportation Systems Impact Public Health: A Literature Review,” and “How Land Use and Transportation Systems Impact Public Health: An Annotated Bibliography”–can be downloaded from the Internet at http://www.cdc.gov/nccdphp/dnpa/pdf/aces-workingpaper1.pdf and http://www.cdc.gov/nccdphp/dnpa/pdf/aces-workingpaper2.pdf, respectively.