Interrelationships between traditional taxi services and online ride-hailing: empirical evidence from Xiamen, China

Abstract

• The study establishes a bivariate Tobit model to examine the interrelationships between traditional taxis and ride-hailing using real-field data. • Parameter estimates suggest that both competitive and complementary relationships exist between traditional taxis and ride-hailing. • Relationships between travel demands and built-environment factors are revealed. • Suggestions are given for the sustainability of the personal mobility market. This study aimed to examine the underlying interrelationships between traditional taxi services and online ride-hailing. To this end, demand data for these two mobility modes were collected from Xiamen Island, China, for three typical days in June 2020. By considering censoring and the potential interdependence and endogeneity of the demand variables, a bivariate Tobit model was established. Through parameter estimation with real-field datasets, the conjecture about the interdependence was verified, and three types of significant interrelationships were identified: two-way competitive, unilaterally competitive and complementary. Under most spatiotemporal circumstances, ride-hailing acts as a formidable competitor for traditional taxis, whereas the latter only demonstrate significant competitiveness in half of the cases. It is noteworthy that the western residential area of Xiamen Island is found to be a blue ocean during the early morning, in which these two parties are complementary to serve an unsaturated market. From empirical analyses, it is concluded that traditional taxis will co-exist with ride-hailing for the foreseeable future, and reasonable resource allocation is needed on both sides for the sustainability of the personal mobility market. The methodology and findings should provide valuable insights for practitioners and decision-makers in framing strategies and regulations towards sustainable urban mobility.