A Ridesharing Platform with an Oversupply of Drivers: Different Assigning Rates or Commission Rates for Drivers?

This study examines a ridesharing platform's choice among two quality-dependent compensation schemes for drivers when the market is oversupplied, that is, there are more drivers than riders at a given time. We examine an information asymmetric context in which drivers (high and low cost) have different costs to improve service quality, which is private information. The platform designs a menu of contracts for drivers to reveal their type. In the assigning-differentiation compensation scheme, the platform offers contract options with different assigning rates, that is, the probability of receiving an order assignment, different service-quality levels, but the same commission rate. In the commission-differentiation compensation scheme, the platform offers contract options with different commission rates, different service-quality levels, but the same assigning rate. We find that the oversupply degree, measured as the ratio of the number of drivers to the number of riders, plays a critical role in the platform's choice of scheme. The platform's expected profit is higher under the assigning-differentiation (commission-differentiation) scheme when the oversupply degree is high (low). We also determine the effect of the driver-type distribution on the platform's choice of scheme. When the difference in the drivers’ costs is small, the platform is more likely to choose the commission-differentiation scheme when there are more high-cost drivers. When the cost difference is large, first, the platform is more (less) likely to choose the commission-differentiation (assigning-differentiation) scheme, but then, it is more (less) likely to choose the assigning-differentiation (commission-differentiation) scheme when there are more high-cost drivers. Further, high-cost drivers are indifferent to either compensation scheme, whereas low-cost drivers are better off under the assigning-differentiation (commission-differentiation) scheme when the oversupply degree is low (high). In addition, we find that both the platform and (low-cost) drivers prefer the assigning-differentiation scheme, a win-win outcome under certain conditions. Finally, we show that the platform can provide a subsidy to mitigate potential misalignment with the drivers.