One of the most salient and relevant dimensions of team heterogeneity is ethnicity. We measure the causal impact of ethnic diversity on the performance of business teams using a randomized field experiment. We follow 550 students who set up 45 real companies as part of their curriculum in an international business program in the Netherlands. We exploit the fact that companies are set up in realistic though similar circumstances and that we, as outside researchers, had the unique opportunity to exogenously vary the ethnic composition of otherwise randomly composed teams. The student population consists of 55% students with a non-Dutch ethnicity from 53 different countries of origin. We find that a moderate level of ethnic diversity has no effect on team performance in terms of business outcomes (sales, profits and profits per share). However, if at least the majority of team members is ethnically diverse, then more ethnic diversity has a positive impact on the performance of teams. In line with theoretical predictions, our data suggest that this positive effect could be related to the more diverse pool of relevant knowledge facilitating (mutual) learning within ethnically diverse teams.