Few would deny that public transit has an important role to play in any sensible solutions to the transportation’s sustainability problem. Yet, the consensus often dissolves at the question of how. A case in point concerns urban rail transit (URT), which has expanded rapidly in recent decades.   The ongoing debate about URT has been fueled by inconclusive, sometimes contradictory, empirical evidence reported in the literature.  Has URT consistently reduced driving and/or auto ownership to affirm its appeal to sustainability? We set out to address this question head-on in this study.

You may read the abstract below, and download a preprint here.

Abstract: Urban rail transit (URT) has expanded rapidly since the dawn of the century. While the high cost of building and operating URT systems is increasingly justified by their presumed contribution to sustainability — by stimulating transit-oriented development, promoting the use of public transportation, and alleviating traffic congestion — the validity of these claims remains the subject of heated debates. Here we examine the impact of URT on auto ownership, traffic congestion, and bus usage and service, by applying fixed-effects panel regression to time series data sets compiled for major urban areas in China and the US. We find that URT development is strongly and negatively correlated with auto ownership in both countries. This URT effect has an absolute size (as measured by elasticity) in China three times that in the US, but is much larger in the US than in China, relative to other factors such as income and unemployment rate. Importantly, the benefit transpires only after a URT system reaches the tipping point that unleashes the network effect.  Where this condition is met, we estimate about 14,012 and 31,844 metric tons of greenhouse gas emissions can be eliminated each year in China and the US, respectively, for each additional million URT vehicle kilometers traveled. We also uncover convincing evidence of cannibalization by URT of bus market share in both countries. However, rather than undermining bus services, developing URT strongly stimulates their growth and adaptation. Finally, no conclusive evidence is found that confirms a significant association between URT and traffic congestion. While traffic conditions may respond positively to URT development in some cases, the relief is likely short-lived.