Charitable Giving When Altruism and Similarity are Linked

This paper presents a model in which anonymous charitable donations are rationalized by two human tendencies drawn from the psychology literature. The first is people’s disproportionate disposition to help those they agree with while the second is the dependence of peoples’ self-esteem on the extent to which they perceive that others agree with them. Government spending crowds out the charity that ensues from these forces only modestly. Moreover, people’s donations tend to rise when others donate. In some equilibria of the model, poor people give little because they expect donations to come mainly from richer individuals. In others, donations by poor individuals constitute a large fraction of donations and this raises the incentive for poor people to donate. The model provides interpretations for episodes in which the number of charities rises while total donations are stagnant. (JEL: D03, D64, H31) ∗Harvard Business School, Soldiers Field, Boston, MA 02163, [email protected]. I wish to thank Rafael Di Tella, Erzo Luttmer, Michael Norton, seminar participants at the Toulouse School of Economics, and an anonymous referee for helpful comments. This paper presents a model that is directed at rationalizing several aspects of charitable giving. First, individuals do not appear to reduce their contributions to a charity significantly when they learn that the government or other individuals have increased the funds that they devote to the charity’s beneficiaries. Indeed, there are instances in which people increase their contributions when they hear that others have contributed more. Second, there are often several distinct charities that contribute to the same beneficiaries. For example, there were hundreds of U.S. charities devoted to fighting breast cancer in 2011. Third, a rise in the number of charities, of the sort that took place recently in the U.S., is not systematically associated with an increase in the contributions relative to income. Lastly, the extent to which individuals contribute to charity differs greatly across countries. These observations can be rationalized by supposing that people have social preferences with the properties assumed in Rotemberg (2009). These preferences are based on two human tendencies detected in the empirical psychology literature. The first is that people are happier when they learn that there is more agreement with their point of view. The second is that they have warmer feelings towards, and are more willing to help, individuals whom they perceive as sharing their beliefs or, more generally, individuals who are more similar to themselves. Rotemberg (2009) captures these properties in a utility function and shows that, in combination, they can explain why people vote. Charitable contributions are similar to voting in that they allow people to signal what they like. People who think a particular charitable cause is worthwhile can signal this attitude to others by contributing, just like voting for a candidate can signal the belief that a candidate is suitable for office. The parallel is in some ways even closer in the sense that both charitable contributions and voting involve the expression of beliefs about the best way to distribute resources to others. In the current context, it leads people who believe in a charitable cause to gain (vicarious) utility from contributing to this cause because they A search on Guidestar.com led to the display of 600 charities with the expression “breast cancer” in their name, though some of these were regional chapters of large national charities. In addition, Guidestar lists numerous charities with the words “pink” or “mammogram” in their name whose exclusive aim is to fight breast cancer.