Abstract:
While increasing the number of small donors, standard linear matching schemes have been shown to cause considerable crowding out in charitable giving with pronounced effects on large gifts. We propose a form of threshold matching where donations above a certain, potentially personalized, threshold are topped up with a fixed amount. We show theoretically that threshold matching can induce crowd-ing in if appropriately personalized. In a field experiment, we explore how thresholds should be chosen depending on past donations. We find that the opti-mal choice of thresholds is rather bold, approximately 60-75% above past dona-tions. Additionally, we explore how thresholds should be set for new donors as a function of their personal characteristics and demonstrate the benefits of person-alization as opposed to setting general thresholds applying to all recipients of a fundraising call.