Materialistic, pro-social, anti-social, or mixed – A within-subject examination of self- and other-regarding preferences

• Elicit materialistic, pro-social, and anti-social preferences in a within-subject design.
• One third of the subjects are materialistic and one third are pro-socials.
• One third of the subjects exhibit mixed social preferences, i.e. pro-social and anti-social.
• Very few individuals are anti-social without also being pro-social to some degree.
• Experimenter demand seems to increase pro-social but not anti-social behavior.

We introduce an experimental setup to elicit subjects’ materialistic, pro-social, and anti-social preferences. We find one third of the population exhibits mixed social preferences, choosing to give, to destroy, and to keep some payoffs. Most others are either materialistic, keeping all payoffs, or pro-social, giving some and keeping some, but not destroying payoffs. For individuals with mixed social preferences, giving and destruction are positively correlated, but do not seem to be influenced by payoff comparisons. We find that full information and experimenter demand may increase the extent of pro-social preferences, but do not affect the extent of anti-social preferences or the distribution of social types in the population.