Does Price Matter in Charitable Giving? Evidence from a Large-Scale Natural Field Experiment

This paper presents the results of a large-scale natural field experiment testing changes in revenue per solicitation and response rates given different types and levels of matching grants/gifts conditional on charitable giving. This is in order to provide insight to research in fundraising models and measures of non-market valuation for cost-benefit analyses. Of a sample of 50,083 individuals who have given to the organization since 1991, randomized and assigned 67% into \"match\" treatment group and 33% into control group, of which the \"match\" treatment group was offered a matching grant conditional on their donation. The \"match\" group was divded evenly between various sizes of the matching ratio ($3:$1, $2:$1, and $1:$1), the maximum size of the matching gift across all donations ($25,000, $50,000, and $100,000), and the example donation amount suggested to the donor (the individual's highest previous contribution, 1.25 times their highest previous contribution, and 1.5 times their highest previous contribution). All individuals received a four page letter identical in all respects except two: (a) the treatment let ters included an additional paragraph inserted at the top of the second page that announced that a \"concerned fellow member\" will match their donation, and (b) the reply card included in bold type the details of the match. For the control group, the reply card match language was replaced with a large logo of the organization. Outcome measures were donation response rates (donated or did not donate) and dollars donated.