Essays on Economics of Sharing Economy

Abstract

Enabled by Information Technology (IT), the Sharing Economy business models have emerged in different industries and been growing rapidly. Popular examples are Uber, Lyft, and Didi in the transportation industry and Airbnb in the lodging field. These sharing economy businesses differ from the traditional business models in that they do not contract employment relationships with their suppliers and the suppliers thus have the flexibility to choose whether and where to serve. All my dissertation chapters are motivated by the distinct phenomena in sharing economy contexts and aim to provide economical insights and business implications. My Chapter 2 investigates platform competition in sharing economy, in which the platforms compete on supply side as well as demand side. As a contrast, firms in the traditional industry (e.g., taxi firms) have relatively fixed supply and focus on the competition on the demand side. I find out the platform competition is softened in sharing economy contexts than in traditional contexts if and only if the transaction volume is larger than a threshold, regardless of the wage scheme employed by the platforms. Given the larger amount of transactions on ride sharing platforms in current days, this finding explains why prices charged on ride sharing platforms are typically lower than those charged by taxi firms. Further, any of the three wage schemes (i.e., dynamic-commission rate. fixed-commission rate, fixed-wage) can be the best for the platforms and the worst for riders and drivers, depending on the market characteristics. In markets where the competition on the demand side is more fierce than on the supply side, the fixed-wage scheme results in the highest profits for the platforms and lowest surpluses for riders and drivers. In contrast, in markets where the competition on the supply side is more competitive, when the supply is highly (mildly) more competitive, the fixed-commission-rate (dynamic-commission-rate) scheme generates the highest profits for platforms, leading to the lowest surpluses for riders and drivers and the lowest social welfare. Chapter 3 studies the coopetition between a ride-sharing platform and rental firm, in which the sharing platform cooperates with the rental firm such that drivers can use rental vehicles to serve on its platform while still competing for riders with the rental firm on the demand side. My analysis reveals that the cooperation on the drivers’ side always intensifies the two firms’ price competition on the demand side, which benefits the riders. Albeit the intensified competition, both the two firms want to cooperate under some conditions because of the platform’s decreased wage offered to drivers personally owning eligible vehicles. I further show that the cooperation can never occur if the platform does not have the flexibility to adjust its wage, which can be more easily done in a sharing economy market because of the non-contracted relationship between the platform and the drivers. Further, I investigate the conditions where the firms want to cooperate and show that the cooperation is more likely to occur when the ride sharing platform has a larger market differential advantage over the rental firm. In Chapter 4, I derived the efficient mechanism for trading differentiated service with information asymmetry on both sides. This chapter is motivated by the non-uniform pricing on some ride sharing platforms (e.g., Didi). Under the efficient mechanism that I derived, the price that a rider pays for the service does not directly depend on her reported time sensitivity but the highest among all the time sensitivities lower than her reported time sensitivity. Additionally, unlike the pricing in quality differentiation literature making distributional assumptions, the rider with the lowest time sensitivity receives a positive utility rather than a zero utility. If the platform aims to extract all surpluses from the rider with the lowest time sensitivity, the incentive-compatibility constraint is then violated. On the driver side, the wage that a driver receives does not directly depend on his reported serving cost either if he is selected. Whether a driver is selected depends on if his reported serving cost is less than a certain threshold as well as what other drivers are available. Such thresholds are decided together by the platform’s announced mechanism previously and the reported time sensitivities. I identify the equations for solving the interdependent thresholds and present an example to illustrate the solution.