Does reputation hinder entry? Study of statistical discrimination on a platform

Résumé

This paper studies entry dynamics of users facing statistical discrimination into a two-sided market with moral hazard. Firstly, we develop a theory model, where we study effort dynamics of users coming from different populations. We show that level of required effort and the expected surplus of agents are functions of their population. We also discuss when and why a platform will decide to create a reputation system. Secondly, we bring our theoretical results to data collected on a popular ride-sharing website. We show that minority male users face statistical discrimination during first interactions. However, the reputation system resolves this issue in the longer run.