Back to top

Why Transportation Planners Need to Be a Little More Like Mad Men

WINTER 2013 -- Persuading people to abandon their cars for buses, subways, or bikes, may be more difficult than transportation planners have assumed—even as traffic gets worse and travel times increase.

One reason, according to ITS Ph.D. student Ashkay Vij, is that travel demand models fail to account for human frailties—our irrational behavior, our hard-to-break habits, our need to define ourselves by what we consume.

Traditional travel demand models, which are used by planners to forecast transportation infrastructure and service needs 20 or 30 years into the future, assume that we are rational; that we are aware of the full range of transportation alternatives, that we carefully weigh the pros and cons of different travel modes at our disposal and make choices based on the costs and benefits associated with level-of-service attributes as well as the needs of our households.
Madison Avenue knows us better—and has for a long time.
At a recent ITS seminar, Vij projected a series of slides of automobile advertisements from the past.
  • A gleaming white 1948 Studebaker is parked in the moonlight on a beach, where a young man, the driver, lounges on cowhides with three beautiful young women. “All over America the word for style is Studebaker,” the ad trumpets.
  • Another slide shows a jazzy 1973 orange Camaro beside the sea; its attractive, athletic passengers are equipped with scuba tanks. Clearly they’re adventurers. 
  • In a third slide, a self-confident mom stands beside her 1989 Dodge Caravan, extolling the virtues of her van which allows her to transport kids to school and to volleyball, to run to the grocery store, to pick up a load of furniture, then drive to Colorado for Christmas, loaded with kids, luggage and ski equipment.
Madison Avenue has always wooed customers by insinuating our lives should and could be like those of the good-looking, confident and happy people in glossy advertisements and commercials—whether they are young men on the make or soccer moms. What you wear, eat, and drive—what you consume, essentially, defines your lifestyle, defines you—and in transportation parlance defines your modality—how you travel.
But travel demand models aren’t equipped to capture our psyches, explained Vij.
“Existing travel demand models place an overriding emphasis on travel times and travel costs. What I was trying to show with the advertisements was that car-makers have never sold cars as the fastest or cheapest mode—but as part of a lifestyle that comes with each of those cars.”

Limitations of demand models

Vij, who writes short stories in his spare time, came to transportation engineering with a greater interest in psychology, cognitive behavioral science, and sociology than in mathematics.
“I’m good at numbers, but that’s not what keeps me happy,” he explained recently.
Two-and-a-half years ago he began working closely with ITS professor Joan Walker, whose research focus is behavioral modeling.
Typically, planners generate forecasts for future decades by asking thousands of residents how they traveled over a two-day period: where they went and how they got there. To combat increasing congestion and other transportation woes they develop policies aimed at moving people from one mode—usually their cars—to another such as transit.
But Vij argues that failure to understand why people choose to travel the way they do will result in skewed forecasts and, in turn, make efforts to change their transportation modes less effective.
Vij says his father, who lives in Delhi, is an example.
“You could build a bus stop in front of his house and he’s not going to change. He will continue to drive and never take the bus. But once the Delhi metro was built, which is a very different system from the bus system—it’s cleaner and faster—it’s a whole new paradigm, and that actually got him considering transit.
“He’s still never going to take the bus, but he has become conscious of the metro system. He knows the travel time and the waiting time—which he never knew about the bus system. He doesn’t know how to get to work by bus, but he knows how to do that by metro.”

Surprising results

Vij set out to devise a better modeling framework—one that recognizes the more subtle aspects of ourselves—to capture the influence of lifestyle and different modality styles on travel mode choice.
"Car-makers have never sold cars as the fastest or cheapest mode—but as part of a lifestyle that comes with each of those cars."
In his paper “You Can Lead Travelers to the Bus Stop, But You Can’t Make them Ride,”  which he presented earlier this month at TRB, Vij puts it this way:
“…The myriad choices that an individual is daily confronted with result in decisions not only about how to act but who to be. For example, an individual’s proclivity to recycle, her desire to reside in high-density mixed-use neighborhoods, and her inscrutable ability to endure a slow and uncomfortable bus ride as part of her morning commute to work every day, are different manifestations of the same system of beliefs. How that system of beliefs, or the lifestyle that the individual subscribes to, influences her travel behavior is the question motivating this study.”
Because belief systems are not captured in traditional travel models, Vij turned to a different modeling framework, a latent class choice model, first developed in the field of marketing sciences. But such models fail to capture the future influence of, say, service reductions on transit lines or increased freeway congestion, on belief systems. Vij added a feedback mechanism to account for these changes and make long-term forecasting more accurate.
Using data from the Bay Area Travel Survey 2000, a two-day travel survey of 8,737 Bay Area households, he defined a number of modal alternatives: drive alone, shared ride, walk, bike, walk to transit and drive to transit.
Groups distinguished by their modal preferences were revealed and described in a soon-to-be-published paper “Allowing for the Acquisition of Taste: Latent Class Choice Models with Feedback through Consumer Surplus” by Vij and Joan Walker.
The groups, which Vij labeled as "inveterate drivers," "car commuters," "moms in cars," "transit takers," "multimodals," and "empty nesters," were found to differ substantially from one another in terms of both the kinds of individuals that belong to each group and the relative importance that they attach to different level-of-service attributes of the transportation system.
For example, the modality style with the strongest positive predisposition towards public transit--the "transit takers"--is made up primarily of unemployed young people from large, low-income households with low rates of car ownership. Perhaps unsurprisingly it is also the smallest of the six groups at 7.4 percent of the sample population
Similarly, value of time--the amount of money that an individual is willing to pay in dollars to reduce his travel time by an hour--varies across the six modality styles from as little as 0.5 $/hr (indicating near insensitivity to travel times) to as high as 37 $/hr.
Among other discoveries, Vij found that although the Bay Area has an extensive public transit network compared to other metropolitan regions in the U.S., 30 percent of the sampled population belonging to two of the six modality styles, "inveterate drivers" and "car commuters," revealed within the BATS 2000 data set never considers public transit when deciding how to travel.
Even worse, only nine percent of the sampled population belonging to one of the six modality styles even thinks about bicycling as a mode of transportation when deciding how to travel.
Moreover, when he compared various scenarios that would make travel by transit more amenable—such as reducing travel times by half, and ensuring zero waiting and transfer times—the traditional travel demand model was unreasonably optimistic. In contrast, Vij’s latent class model showed that three-quarters of travelers, like Vij’s father, still would not get on the bus.
Clearly, the job of moving people out of their cars and onto transit may be harder than planners understand. Vij suggests it would take a shock to the system, one on the order of London’s congestion pricing system, to break up our love affair with the car. But so far, there is little political will to move to congestion pricing or even increase the gas tax.
For decades advertising executives on Madison Avenue have sold automobiles as part of the American dream. As congestion and air pollution have become nightmarish, perhaps it’s time for transit agencies to engage Mad Men to shine their lights on alternatives--and help extinguish the torch we carry for our cars.
--Christine Cosgrove