We study the effectiveness of leaders for inducing coordinated organizational change to a more efficient equilibrium, i.e., a turnaround. We compare communication from leaders to incentive increases and also compare the effectiveness of randomly selected and elected leaders. Although all interventions yield shifts to more efficient equilibria, communication from leaders has a greater effect than incentives. Moreover, leaders who are elected by followers are significantly better at improving their group’s outcome than randomly selected leaders. The improved effectiveness of elected leaders results from sending more performance-relevant messages. Our results are evidence that the way in which leaders are selected affects their legitimacy and the degree to which they influence followers. Finally, we observe that a combination of factors—specifically, incentive increases and communication from elected leaders—yields near-universal turnarounds to full efficiency.

This paper was accepted by Uri Gneezy, behavioral economics.