A study compared how well old-city street layouts handled traffic versus modern approaches. The results set off a firestorm.
When I drive my neighborhood streets of Brooklyn, which were laid out more than a century ago in a grid style, it's obvious: These city streets do a better job of handling local traffic than the more modern set up of cul-de-sacs, collector streets and arterials. That's because, when I'm heading somewhere, I can choose from five or six local streets as opposed to one or two suburban style "arterials."
Many of us didn't need a study to conclude the merits of the grid, but the U.S. Environmental Protection Agency's Smart Growth office thought otherwise and commissioned one. The 2004 report, "Characteristics and Performance of Regional Transportation Systems," examines transportation data from a handful of older cities, such as Philadelphia, New Orleans and Pittsburgh, and compared them with newer cities such as Atlanta, Houston and Tampa. The idea was to see which approach -- the old or the new -- performed better.
The answer? In the older cities, people drive less and use transit more; they experience fewer traffic delays, die less often in traffic accidents and emit less pollution.
No surprises there, but the study was a big deal in the transportation world, suggesting as it does that maybe the whole modern-suburban approach to street design needs to be reexamined. But as significant as its conclusions were, this report is no longer available. It is, the EPA's Web site says, "being updated and revised to address certain issues regarding the data and methods used in the analysis."
It seems that after the report's release, the Federal Highway Administration, which is charged with building and maintaining the nation's roads, objected to some of the study's methodologies -- objections that in 2004 were quickly followed by a public chorus of criticism from those who favored the conventional "modern" building patterns.
To clear things up, I got Geoff Anderson, director of the EPA Smart Growth program that commissioned the study, on the phone. Anderson says the study contained valid problems that popped up under closer review, such as the "multicollinearity" of some variables.
So the EPA is hard at work on revising the study, right? No. The study has been permanently tabled, Anderson admits, adding that "the idea of getting some sense of what kind of development patterns and what types of transportation systems perform well is a good idea. But it will take a second round of assessment to do that, and probably more resources than we have to make that possible."
Some in the urban planning world have alleged that the study's revision was killed for political reasons, that the road-building community didn't like its probable conclusions. Not so, Anderson insists. "The things that had me the most concerned were the technical issues," he says. "Those were the ones I responded to."
One incendiary aspect of the study was that it not only examined how older street grids performed; it examined whether some regions that built more roads per capita have less traffic congestion. The study found they did not, thus calling into question the utility of most road building. The study found, for example, that congestion delay in Detroit rose from 14 hours per peak-period traveler per year in 1982 to over 50 hours in 2000, but the metro area population grew by an average of only 0.3 percent per year. Moreover, during the same period, urbanized land area expanded 21 percent, and total lane miles increased by 13 percent.
While the study may have had flaws, it's not the only word on the subject. With a copy of the study in hand, I asked several academics to review the report. Norman Garrick's response was typical. Garrick, associate professor of civil and environmental engineering at the University of Connecticut in Hartford, tells me that "the trends are not surprising because they are supported by other studies conducted at various community scales." The differences, he says, are even starker when you compare American cities with Western European ones.
The international comparison brings up a distinction the study failed to look at: walking and cycling in a city. Anderson agrees that the travel surveys don't do a great job of measuring non-auto trips. "A lot of people think that the key difference between European and American cities is in transit use," he says, "but the big difference is the amount of walking."
The point of all this -- controversies aside -- is that we can learn from the transportation systems of older-style cities. In the meantime, Anderson says he welcomes follow-ups to the killed smart growth study, even if his EPA office doesn't have the money to fund them.
I do, too.