Abstract:

This paper reports on a survey experiment designed to test whether support for aid is more readily increased by emphasising aid’s benefits to the donor country, or by emphasising aid’s role in helping people in developing countries. The survey experiment was pre-registered and involved a large, nationally-representative sample of over 4000 Australians. Information based on a real Australian government aid project was presented in the form of a newspaper article. Participants were randomly assigned to one of five groups: a control group, which was given no information on the project; a basic treatment group, which was given basic information on the project; an ‘enlightened national interest’ group, which was told about benefits the project would bring Australia; a ‘global leader group’, which was told how the project would make Australia a global leader; and an ‘altruism’ group, which was given information on how the project would help people in need in developing countries. All of the treatment groups including the basic treatment group outperformed the control group in eliciting increased approval of aid, and in reducing the share of people who thought Australia gave too much aid. Both the altruism and enlightened national interest treatments out-performed the control in increasing the share of people who thought Australia did not give enough aid. However, only the enlightened national interest treatment outperformed the basic treatment group in this area. Results were complicated by the altruism treatment’s limited success in convincing people that the aid project was actually being run for altruistic ends. When we limited our analysis only to those people who appeared to accept the narrative used in the treatment, altruism actually proved to be the most effective treatment.