Relationships among access to bus rapid transit, urban form, and household transportation outcomes: Evidence from a quasi-longitudinal study in Bogota, Colombia

Bus Rapid Transit (BRT) has become popular as a means to provide reliable, non-automobile-based mobility and alleviate impacts of rising traffic congestion in cities around the world. To date, however, there is little empirical evidence supporting BRT's potential to meet these objectives, and limited understanding of the urban form conditions necessary to support BRT. This research improves knowledge of BRT's potential as an alternative to vehicle ownership at the household level and provides new evidence of the role of urban form in supporting transit investment.;I use a difference-in-differences research design to examine the changes in vehicle ownership and non-car mobility from before to after implementation of Bogota, Colombia's TransMilenio BRT system. With respect to vehicle ownership, my results indicate access to TransMilenio's main trunk system is negatively related to the odds of vehicle ownership for higher wealth households. Among lower wealth households, access to the main trunk system is not related to vehicle ownership except in neighborhoods with urban forms that are supportive of walking, bicycling, and transit use. Furthermore, I find access to TransMilenio's feeder system (which brings passengers from peripheral neighborhoods into the main trunk system) is associated with an unexpected increase in the odds of vehicle ownership. This increase may be due not necessarily to the introduction of the feeder service itself, but to concurrent policies of upgrading roadways and investing in housing stock in feeder served neighborhoods. Regardless of the cause, however, the increase in vehicle ownership among lower wealth, feeder-served households appears to be reversed in neighborhoods where urban form supports transit and non-motorized travel. This finding suggests the importance of a concerted effort to coordinate transit interventions with urban development policies that support those interventions, particularly in lower wealth neighborhoods.;I found no evidence TransMilenio access was significantly related to non-car mobility (tour frequency, travel purpose diversity, and vehicle independence) among lower wealth car-less households. Relationships between mobility and urban form were ambiguous and inconsistent. I attribute these results to limitations in the available techniques for measuring mobility. I describe to improve the validity and reliability of these techniques in future research.