Book-or-Drive: The Impact of Ridesharing on the Automobile Industry

Abstract

This paper empirically examines the impact of Uber, an online ridesharing platform, on automobile registrations, a sales performance indicator in the traditional, mature automobile industry in the United States. We leverage the sequential entry of Uber in different locations in a natural experiment setting and use the staggered difference-in-differences (DID) econometric model to estimate this impact. Our dataset includes yearly observations of Uber’s entry and automobile registration that span more than 600 counties from 2010 to 2017. The results suggest that Uber’s entry has an overall negative effect on automobile registrations, and this effect sustains for another year after entry. The negative effect manifests significantly for counties with larger values of GDP, population, and population density but can become insignificant for counties with smaller values. Our results are robust to matching, placebo tests, and the consideration of Uber’s endogenous entry decision, and are reinforced by the analysis of Uber’s search intensity on Google. Our findings have important managerial and policy implications and contribute to the growing stream of research on the social and economic impacts of sharing economy.