Last week the Utah Department of Department (UDOT) released revised traffic projections that purport to prove that the proposed Legacy Highway and other major highway expansions will be needed to meet transportation demands in Davis County within the next 20 years. These numbers come out of an opaque box known as a "travel demand model." Travel demand models are used to try to predict how many trips people will take at some future time, how long the trips will be (from what origins to what destinations), and what modes of transportation (car, bus, train, bike, walk) will be used to get there.

But like all computer models, the results depend on a range of assumptions, the legitimacy of which dictate whether the numbers can be trusted.

The following is a user-friendly guide to some of the ways in which the current model used by UDOT cannot yet be trusted to produce reliable information. Although UDOT and its partners at the Wasatch Front Regional Council are, to their credit, making improvements to the models for future use, critical decisions continue to be made based on older, badly-flawed approaches. While no computer model is perfect, much better procedures are being used in other communities. And this is no mere technicality. UDOT proposes to spend over a half a billion dollars on the Legacy Highway and other highway construction in Davis County. Yet improved models might predict that traffic will be worse after these roads are built than it is now. More important, they might show that better traffic relief could be obtained through sound investments in trains, buses and other transit systems, coupled with improved development patterns that allow people more choice about where they live and work, and how they will travel around their communities.

These improved tools could be in place within a matter of months, for roughly what it costs to install one set of traffic lights.

Why should we spend so much public money on transportation solutions developed with traffic numbers that may not be correct, and for solutions that might not work?

Do we want the Wasatch Front to develop over the next two critical decades in ways that continue to promote more roads, more cars, more air pollution and even more traffic? Or do we want to shift to transportation and land use solutions that allow people more choice to live closer to where they work, learn and play; and to have more choices about how to get there? Sound travel models would provide much better information with which to answer these questions.

The Transportation/Traffic/Land Use Connection
Studies conducted all over the world indicate that highways tend to generate more traffic, a phenomenon known as "induced travel." While some local officials claimed derisively last week that this is "unproven dogma," in fact it is quite well documented. In a comprehensive review of all of the studies of this issue, the U.S. Environmental Protection Agency (EPA) concluded earlier this year that research "has not only built a strong case for the existence of induced travel effects, but in some cases suggests that a large fraction of [traffic] is directly attributed to increases in road capacity." Similar conclusions were reached by the Transportation Research Board and other experts.

The idea that roads breed more traffic is also a matter of simple common sense. In part, it is due to the basic law of supply and demand. When more roads are built, traffic congestion diminishes in the short-term. With less traffic and higher auto speeds, people make different choices about where to live, how many trips they will make, where they will travel, and by what means.

Less traffic encourages people to make more and longer auto trips. With more congested roads people tend to be more efficient, for example, by living closer to work, combining trips, making shorter trips, and using other modes of travel than the auto.

Extensive road construction ultimately breeds different development patterns that, in turn, generate more traffic. Highways allow and encourage more sprawling, auto-dependent development than do investments in transit, which tend to be accompanied by more compact development patterns (and as an important side-benefit, save more open space and wildlife habitat on the edges of the region).

Opening the first proposed phase of the Legacy Highway in Davis County, for example, will likely encourage more auto-dependent development in Davis and Weber Counties, along and north of the Legacy corridor. This new traffic will, in time, clog the new highway lanes that UDOT proposes to build now, which will then be used to justify later phases of Legacy and other highway expansion. And so on. Just like we did in Los Angeles and other cities since the 1950s.

Older travel demand models, including the one used by UDOT to justify Legacy, do not account properly for these cause-and-effect relationships. The model used by UDOT, for example, does not account for highway-induced land use changes. It does not account for the fact that people will change when they travel to avoid congested peak periods.

While some recent improvements to the model do predict some induced travel, they do so using crude and incomplete methods. In sum, the model underestimates likely levels of congestion after the highway is built, and overestimates congestion if the highway is not built.

The way out of this circular logic is to use better travel models or other methods that predict far more accurately how much new traffic will be generated as a result of new highway capacity. Only by measuring that trend will we know whether we will end up increasing traffic demand by more than we have built added highway capacity, and thus whether the new highway has made traffic better or worse.

Highways versus Transit – Leveling the Playing Field
Using a model that can measure induced traffic is also critical if we are to compare fairly the options of building highways or making transit improvements, and the resulting changes in growth patterns. All parties, from UDOT to the Sierra Club, agree that improved transit is an important part of the overall transportation solution in Davis County and other parts of the Wasatch Front. The proposed ¼ cent increase in sales taxes, which will be on the Davis County ballot this fall, is evidence of some increased commitment to transit in the region.

The parties disagree, however, on how much difference transit can make, and how much added highway capacity will be needed as well. Two critical flaws in UDOT's modeling process, however, once again stack the deck in favor of more roads at the expense of more investments in transit.

The first problem with UDOT's process flows directly from the issue discussed above. In comparing highway versus transit options, a model that fails to measure the tendency of highways to induce traffic will produce skewed results. New highway lanes will provide some new capacity, but that capacity will be offset in part by induced traffic. If transit can provide the same new transportation capacity without inducing more auto traffic, the net result will tend to be more favorable under the transit option. By not comparing the two options fairly, UDOT's current model is inherently biased in favor of highways.

Second, even with its existing model, UDOT steadfastly refuses to measure how much of the region's travel demand could be met with more aggressive investments in transit than are currently planned. In the model results released last week, UDOT compared several different highway build options. By contrast, in each of the options tested UDOT used the same, relatively modest increases in transit. For example, each of the options compared assume a single rail option for Davis County in which only a limited number of trains are run in each direction each day. The result – by definition – is that the proposed major highway expansions always were able to meet far more travel demand than the limited transit options tested. An evenhanded comparison of all viable options is required in the Environmental Impact Statement prepared under the National Environmental Policy Act, and under federal air quality and transportation laws.

Transportation Choices – Are Utahns Willing to Change?
An additional flaw in the travel demand model currently used by UDOT would bias even this comparison of highway and transit options if not corrected. In predicting the relative numbers of auto trips as opposed to other modes of travel, modelers must make assumptions about what choices people will make under different circumstances. The model used by UDOT to justify Legacy is based in significant part on a survey of travel patterns by Davis County residents in 1992. Residents were asked to report what trips they made in a 24-hour period, and by what mode of transportation. Based on these data, UDOT assumes that only a relatively small percentage of the public will use transit.

The flaw in this logic is that the data were collected when transit options in the region were limited, and survey respondents could only report actual transportation usage. The correct inquiry is how many people would be likely to use transit (or walking, bicycling and other non-auto modes) if better investments were made in these options. That question cannot be answered using purely historical data. There are two ways to solve this modeling problem. The first is to use "stated preference" surveys in which people are asked what mode(s) of travel they would choose if presented with a wider range of options than are currently available. The second is to use a mode choice model from another community that currently has a wider range of transportation choices.

Ridership on TRAX, Utah's first experiment with a modern transit system, is considerably higher than projected. Perhaps more important, TRAX data thus far indicate that a high percentage of riders are new to transit. TRAX has succeeded in moving people out of their cars and onto the trains. This certainly suggests that UDOT is wrong to assume that transit potential in Utah is so limited.

The quality of travel demand models may sound like the stuff of policy geeks. But the key problems with the models being used by UDOT are easily understood. They ignore the fact that more highways breed more traffic and promote sprawl development, and underestimate other effects of highways on travel behavior. They compare several levels of highway construction, but do not consider increased investments in transit. And they assume that Utahns will make irrational, inefficient decisions to continue to sit in traffic even if better transit options were provided. Hundreds of millions of public dollars are at stake. More important, the quality of our lives and the future quality of our environment are at stake. Yet decisions are being made today based on yesterday's tools.

The Utah public deserves better.

Robert W. Adler is a Professor of Law specializing in environmental law at the Wallace Stegner Center for Land, Resources and the Environment at the University of Utah College of Law.