Comparing charitable fundraising schemes : evidence from a natural field experiment and a structural model

We present evidence from a natural field experiment and structural model designed to shed light on the efficacy of alternative fundraising schemes. In conjunction with the Bavarian State Opera, we mailed 25,000 opera attendees a letter describing a charitable fundraising project organized by the opera house. Recipients were randomly assigned to six treatments designed to explore behavioral responses to fundraising schemes varying in two dimensions: (i) the presence of a lead donor; (ii) how individual donations would be matched using the lead donation, using either linear, non-linear and fixed-gift matching schemes. We develop and estimate a structural model that simultaneously estimates individual responses on the extensive and intensive margins of giving, and then utilize the structural model to predict giving behavior in counterfactual fundraising schemes. We find that charitable donations are maximized by simply announcing the lead donation rather than using it to match the donations of others in some way. If lead donors insist their gifts must be matched in some way, we find the fundraiser is best off announcing the existence of a lead donor and using a nonconvex scheme to match the lead donation with individual donations. We conclude by providing evidence from a follow-up natural field experiment designed to probe further the question why lead donors are so effective in inducing others to give.