Do the First- and Last-Mile Matter? Examining the Complementary and Substitution Effects of Bike-Sharing Platforms on Public Transit

A priori relationship between the entry of bike-sharing platforms and public transit demand is not straightforward and is still open to debate. In this study, we empirically examine the complementary and substitution effects of bike-sharing platforms’ entry on public transit, with a particular interest in city bus systems. We exploit a natural experiment using a fixed-effects econometric formulation of the difference-in-differences framework to empirically estimate the effect of bike-sharing platforms’ entry on city bus systems. We employ a panel dataset that comprises 64 cities in the United States spanning 96 months from January 2012 through December 2019. Our findings reveal that the entry of bike-sharing platforms decreases the demand for the city bus by 2.7% on average across cities, suggesting that bike-sharing platforms substitute city bus ridership. These results also indicate that the first- and last-mile complementary effect of bike-sharing platforms is not very strong for city bus services. We also conduct extensive analyses to confirm the robustness of the main results. Then, we examine the impact of the bike-sharing platforms on the demand for city bus changes depending on heterogenous factors such as bike-friendliness, pedestrian-friendliness, transit-friendliness, population, language spoken, working-at-home, and transit fares. We find that bike-friendliness, pedestrian-friendliness, transit-friendliness, and population moderate the relationship between bike-sharing platform entry and city bus usage, whereas we do not observe any moderation effects of the language spoken, working-at-home, and transit fares. We also repeat our analysis for heavy rail and light rail systems. Our findings provide guidance to bike-sharing platform owners, public transport operators, and policymakers in cities