Two-Sided Dynamics in Duopolistic Markets with Ride-hailing and Ride-pooling

Abstract

Network effects in driver-traveller matching suggest that ridesourcing markets with multiple platforms may be less efficient than monopolistic ones. The practice of pairing travellers in ride-pooling introduces additional network effects, suggesting that market segmentation costs may be further elevated in ride-pooling compared to ride-hailing. We examine the evolution of a duopolistic ridesourcing market, with service providers offering solo or pooled rides, using an agent-based model that considers day-to-day processes and within-day operations. Results reveal winner-takes-all scenarios in markets with two solo providers, but not with two pooling providers offering ex-ante discounts. Coexistence is also possible when one platform offers solo ride-hailing and the other pooling, with both services targeting different users. Our study provides insights into market segmentation costs when platforms co-exist, as well as how ridesourcing vehicle mileage efficiency and modal shifts depend on the service type (solo or pooled) offered by each provider.