Exploring the Causal Mediation Effects of Public Transit Ridership on the Relationship Between Ride-Sharing Services and Traffic Congestion: An Empirical Investigation of UberX in the United States

Abstract

Despite the multiple prior studies examining the consequences of ridesharing platforms on traffic congestion, we still have limited understanding of causal relationships underlying this outcome. Using data collected from 48 metropolitan areas in the United States, we report the results of a causal mediation analysis investigating the role of city bus ridership, which is thought to lie in the causal path between the introduction of UberX and traffic congestion. Our results suggest that these platforms do increase congestion, but only 5% of the increased traffic congestion can be explained by a decrease in city bus ridership caused by these services. This implies that Uber has other effects independent of bus ridership, such as service cars roaming the streets, that would account for a much larger portion of Uber’s detrimental effect on traffic congestion.