Abstract:A seeming truism, repeated countless times in university transportation courses, holds that "travel is a derived demand." That is, travel occurs because someone wants to do something somewhere else. This basic proposition underlies most policies designed to reduce motorized travel and thereby reduce congestion, increase safety, improve air quality, or reduce consumption of nonrenewable energy resources.

For example, some land use policies seek to place residences at high densities close to work and shopping, with the aim of reducing trips to reach frequently visited destinations. Telecommuting, teleconferencing, and teleshopping present possibilities for people to engage in their usual activities, substituting communication for transportation. The implicit assumption that travelers are cost minimizers also underlies various pricing strategies aimed to induce fewer or shorter trips (congestion pricing, higher fuel taxes, higher parking fees). By increasing the cost of getting there, the theory contends, we can reduce the attractiveness of some destinations and thus reduce the volume of travel.

However, two types of travel appear to be at odds with the derived-demand concept. First is the phenomenon of joyriding—trips themselves being the desired activity. These trips have received little if any attention in trip-generation models, implying that their magnitude is considered too small to be important, or that their complexity and variation defy our modeling capabilities, or simply that they fail to fit that truism underlying the models.

Second is the phenomenon of excess travel—unnecessary mileage attached to routine trips, such as the journey to work. Our research suggests that some excess mileage can be attributed to nothing more than the desire to travel. The benefits enjoyed are independent of and in addition to those associated with getting to the destination.

None of this is news, of course. Joyriding has long been associated with automobiles, but recreational travel has been an outgrowth of virtually every means of transportation ever known (consider horseback riding or sailing). A recent MIT study observes that the transition from slow transit to fast automobiles and airplanes is making for more miles traveled per capita. An Australian study finds that satisfaction with hypothetical commuting times is highest at about fifteen minutes, implying both longer and shorter times are less satisfactory. Some of our own earlier work on the demand for telecommuting found that not everyone who is able to telecommute chooses to do so—that some enjoy the regular commute even when they don't have to take it.

There's been a lot of commentary recently on attitudes toward automobiles, much of it reinforcing these notions that people desire travel for its own sake. Automobile advertisements frequently play to the desire for mobility: "It's an unrestricted round trip ticket to anywhere" (Acura Integra); it "takes me places roads don't even go" (Ford Explorer); "a car so advanced, it might set telecommuting back a few years" (Honda Accord). These themes doubtless resonate with a number of automobile consumers, such as the one recently spotted on a California highway with a vanity license plate reading "BRN4TRVL."