The role of reputation systems in digital discrimination

Abstract

Reputation systems are commonplace in online markets, such as on peer-to-peer sharing platforms. These systems have been argued to be a solution to (ethnic) discrimination on such platforms. This argument is based on empirical studies showing that ethnic disadvantages are smaller for users with ratings than for users without ratings. We argue that this conclusion may be premature, because minorities have a harder time accumulating ratings. The greater benefit of ratings to minorities may be offset by their troubles acquiring any, thereby diminishing the potential for reputation systems to reduce discrimination. We tested this counterargument using a unique data set that contains information on all interactions on a peer-to-peer motorcycle rental platform. We find that the reputation system does not reduce initial inequalities between otherwise comparable renters of different ethnicity. Platforms that wish to reduce discrimination should not only make their reputation system more effective, but also help users collect ratings.