University of California Transportation Center

Various 'new suburb' land use designs have recently been proposed to address a number of social and environmental problems, including the dominance of automobile travel. Transportation benefits are to be accomplished by reducing the surface street distance between locations, mixing land uses, and promoting walking, bicycling and transit via redesigned streets and street-scapes. That auto travel will fall is a largely unchallenged premise of these designs, though what little evidence exists is either weak or contrary. This paper presents a simple behavioral model to explain why. Generally speaking, driving is both discouraged and facilitated in the new suburbs, with the net effect being an empirical matter. In particular, both the number of automobile trips and vehicle-miles traveled can actually increase with an increase in access, such as a move to a more grid-like land use pattern. Whatever the merits of neotraditional and transit-oriented designs, and there are many, their transportation benefits have thus been oversold. Each development must be evaluated on a case by case basis to determine whether its net impact on auto use is positive or negative. An analytical framework for doing so is suggested.