Carnegie Mellon University
Skip navigation and jump directly to page content

July 9: Study Finds Minimal Racial Bias in Charitable Giving to Katrina Victims


Jonathan Potts

Carnegie Mellon, Harvard Study Finds Minimal Racial
Bias in Charitable Giving to Victims of Hurricane Katrina

People Are More Generous Toward Poorer Victims

PITTSBURGH—The perceived neediness of Hurricane Katrina victims is a better determinant of charitable giving than the victims' race, according to study by Christina M. Fong, research scientist at Carnegie Mellon University, and Erzo F.P. Luttmer, associate professor at Harvard University's Kennedy School of Government. The paper was published this month by the National Bureau of Economic Research (NBER).

The study explains that, on average, charitable giving to Katrina victims is not affected by the perceived race of the recipients. Racial bias does, however, exist within certain subgroups of whites. Whites who say they identify with their ethnic or racial group give significantly less to blacks, and whites who say they do not identify with their ethnic or racial group give significantly more to blacks.

The authors concluded that perceptions of recipient income affect charitable giving — people give more to needier recipients. Additionally, support for government aid to Katrina victims is affected by both the perceived race and perceived deservingness of victims. Overall, people support more government aid to law-abiding and helpful Katrina victims, and whites support less government aid to blacks.

Fong and Luttmer conducted an online field experiment to investigate the factors that influence generosity toward Katrina victims. In the experiment, a representative sample of the U.S. population was shown a short presentation about Hurricane Katrina victims in a small city — Slidell, La., or Biloxi, Miss. Following the presentation, respondents decided how to split $100 between themselves and a charity helping Katrina victims in that city. For 10 percent of these respondents, the split was actually implemented, with the respondent receiving the amount of money he or she decided to keep and Katrina victims receiving the rest.

To influence the respondents' perceptions of the race of the victims in each city, the experiment showed pictures of black victims to some respondents and pictures of white victims to other respondents. The narrative with the pictures was also varied to manipulate the perceptions of the income and deservingness of Katrina victims in each city. The authors made several significant findings:

  • The income manipulation had a significant effect on charitable giving: respondents gave more money when they believed the victims were poorer.
  • On average, race had virtually no effect on giving, but the averages masked significant racial bias among subgroups of the sample. In particular, whites who identified strongly with their ethnic or racial group gave significantly less to blacks, while whites who did not identify strongly with their ethnic or racial group gave significantly more to blacks.
  • Race had a significant effect on support for government aid to victims. When white respondents were shown pictures of black victims they supported significantly less government assistance to Katrina victims.
  • Audio manipulations of the perceived deservingness of victims had significant effects on support for government assistance to Katrina victims. When victims were described in flattering terms, such as "being helpful" and "law-abiding," respondents were more likely to support government assistance to Katrina victims.

"This study helps advance the understanding of the determinants of charitable giving," Luttmer said. "Findings from previous field experiments have identified several factors that affect how much people give in charitable donations. We show that in addition to these factors, donors are sensitive to the level of need of the recipients." 

"Our results also contribute to the knowledge about redistributive politics," Fong said. "Past studies have shown that race and perceptions of fairness affect public support for and opposition to governmental redistribution. The results of our experiment confirm these findings." 

The experiment was funded by the National Science Foundation and conducted through Knowledge Networks, a commercial survey and marketing research firm. Respondents were drawn from a nationally representative panel maintained by Knowledge Networks.

The paper has been released as NBER Working Paper No. 13219, and is also available on the authors' Web sites: and