(Un)Informed Charitable Giving  Silvana Krasteva Texas A&M University Huseyin Yildirim Duke University October 24, 2012 Abstract Evidence suggests little informed giving. To understand this behavior, we examine a model of charitable giving with costly information. We find that an individual who considers a smaller contribution is less likely to seek information, and thus the percent- age of informed giving diminishes as the population grows. We also find that a direct grant to the charity exacerbates crowding-out by discouraging information acquisition, whereas a matching grant increases donations by encouraging it. We further show that a (first-order) stochastic increase in valuations for charity can decrease donations; and that facilitating private acquisition of information can be a better fund-raising strategy than directly supplying it. JEL Classifications: H00, H41, D82, D83 Keywords: charitable giving, value of information, crowding-out “It is more difficult to give money away intelligently than to earn it in the first place.”— Andrew Carnegie (The Gospel of Wealth, 1889) 1 Introduction According to a recent survey conducted by Hope Consulting, 1 only 35% of people did any research before donating to a charity. 2 One explanation for this “disinterest” in informa-  We thank Tom Nechyba and participants of the Duke theory lunch and Public Choice Meetings (Miami) for comments. Osman Kocas and James Speckart provided valuable assistance. All remaining errors are ours. 1 The 2010 survey includes 4,000 people with incomes over $80K in the U.S, and its results are available at <www.hopeconsulting.us/money-for-good>. 2 This evidence is consistent with online fund-raising statistics. According to 2012 eNonprofit Benchmarks Study across four sectors, the average open rate for e-mail fund-raising is 12%; click-through rate is 0.47%; and the response rate is 0.08% (www.e-benchmarksstudy.com). That is, out of every 10,000 deliverable e-mail solicitations, 1,200 are opened by their subject lines; 47 are clicked through for more information about the cause, but only 8 end up generating donations. See also Chen et al. (2006) for a field experiment of online fund-raising. In direct mail fund-raising, it is difficult to track who actually opens and reads solicitation mails, but a 1% response rate is often considered a success (Bray 2010, Sharpe 2007). The survey evidence is also consistent with the experimental finding of Fong and Oberholzer-Gee (2011): although donors become more generous if they know their assistance benefits a group they like, only a third of subjects have purchased such information prior to giving. 1