We evaluate policies to increase prosocial behavior using a field experiment with 1,500 referees at the Journal of Public Economics. We randomly assign referees to four groups: a control group with a six-week deadline to submit a referee report; a group with a four-week deadline; a cash incentive...