This paper focuses on the impact of historical rail infrastructure—even after it no longer exists—on current auto ownership in the Boston area. We find that zones with shorter distances to historical rail infrastructure have lower levels of auto ownership. The results are robust to controls for current rail infrastructure, built environment, and demographic variation, as well as to instrumental variables that control for endogeneity. They imply that transportation infrastructure has lasting effects on travel behavior. Excluding these long-term effects in current models can skew decision-making.
