A Contribution to the Urban Transportation Debate: An Econometric Model of Urban Residential and Travel Behavior

Abstract

T HIS paper presents a nine-equation econometric model dealing with the decisions about residential density, auto ownership, journey-to-work transportation media, and length of worktrip of workers employed at 254 Detroit workplace locations. Understanding these aspects of household behavior is crucial to a number of enormously important investment decisions being hotly debated in urban communities. The economic stake in urban transportation investments alone is immense, as witness the nearly billion-dollar plans under consideration in the San Francisco-Oakland Bay Area, in Los Angeles, and in Washington, D.C.' The revenue estimates used in all proposed transit plans assume that the post-war decline in transit use not only will halt, but will be reversed as automobile commuters switch to the new rail facilities. The Los Angeles Backbone Plan, for example, expects a ridership of 38 million passengers, of whom 22 million would be persons diverted from private autos. At the same time, highway capacity is to be significantly increased with the completion of a major paralleling freeway. In advocating these expensive mass transit systems, some urban researchers and planners have postulated a vicious circle theory to explain the decline in transit ridership. They argue that reductions in transit use lead to reductions in service levels, which lead to further declines in transit use, still further reductions in service, and so on. This argument probably has some validity, but whether the policy prescriptions derived from it follow logically is another issue. Its proponents often recommend a short-run policy of granting large subsidies to improve transit service until a new service-usage equilibrium is reached, at which point the system becomes self-sustaining or can operate with only a small subsidy. The empirical findings of this paper raise serious doubts about such policies.