Model Event History of Car and License Availability

Abstract

Sound forecasts of car and driver's license availability are crucial for accurate estimates of future mobility trends and the development of planning strategies. Often these forecasts ignore dynamic trends and spatial influences. Cross-sectional analysis reveals that in areas with good accessibility by alternative modes, many households live with no or only one car even if they can afford a vehicle. This study focuses on that dynamic perspective by analyzing changes in car and license availability by event history models. It uses data from the German socioeconomic panel that observes the same households for up to 25 years to estimate the hazard rates of changes between different states of license and car availability. The results show a detailed picture of socioeconomic factors such as age, sex, household characteristics, and income. In addition, the impacts of accessibility on the probability of acquiring or disposing of a car are presented.