How Can Bill and Melinda Gates Increase Other People’s Donations to Fund Public Goods?

How to cite

Karlan, Dean; List, John. 2019. "How Can Bill and Melinda Gates Increase Other People’s Donations to Fund Public Goods?." AEA RCT Registry. AEARCTR-0004204. https://doi.org/10.1257/rct.4204-2.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Dean Karlan* , John List

* Primary investigator

Registry ID
AEARCTR-0004204
DOI
https://doi.org/10.1257/rct.4204-2.0
First registered
Last updated
Status
Completed
Trial dates
2009-12-01 – 2017-05-01
Countries
United States of America ()
JEL codes
D12, D71, D82, H41, O12
Registry record
http://www.socialscienceregistry.org/trials/4204

Abstract

We conducted two matching grant fundraising experiments with an international development charity. Using a sample of non-prior donors, the primary experiment finds that a matching grant offered by the Bill and Melinda Gates Foundation raises more funds than a matching grant offer from an anonymous donor. The effect is strongest for donors who previously gave to other poverty-oriented charities. This is consistent with a quality signal mechanism, although alternative mechanisms are not ruled out. The results help clarify why people give to charity, what models help to describe those motivations, and how practitioners can leverage large donors to increase their fundraising potential.

Experimental design

All participants in our primary study received a letter in the mail requesting donations to TechnoServe, a poverty-alleviation nonprofit. The treatment group received a letter that mentioned that individual donations would be matched in the ratio of $2:$1 by a grant from the Bill and Melinda Gates Foundation (BMGF). The letters sent to the control group mentioned that individual donations would be matched $2:$1 by a grant from an anonymous donor. The sample for the primary study consisted entirely of individuals who had not previously donated to TechnoServe (i.e., ‘cold list donors’).

In the secondary study, the treatment group received letters that specified that individual donations would be matched by BMGF. Control letters requested donations to TechnoServe without mentioning a matching donation. The sample frame for this experiment consisted entirely of prior donors, or ‘warm list donors,’ to TechnoServe.

Design details

Randomization method
Randomization performed by a direct marketing firm using a computer.
Randomization unit
Individual
Clusters
N/A
Observations
61,483 (primary experiment); 52,988 (secondary experiment)
Arms
Primary experiment: 30,731 treated individuals; 30,735 control individuals. Secondary experiment: 25,993 treated individuals; 25,995 control individuals
Primary outcomes
Number and amount of individuals' donations to TechnoServe

Cited by

No citation data available for this registration yet.

Keywords