Some doubts about measuring self-interest using dictator experiments: the costs of anonymity

Abstract

Traditionally, economists have assumed self-interest governs economic choices. Recently, some social scientists and economists, especially those working in game theoretic and experimental areas, have begun to treat self-interest as a testable hypothesis. One important vehicle for evaluating self-interest has been a class of experiments called ‘dictator’ experiments. We believe that these experiments may have a flaw in their design which leads researchers to overstate, systematically, the role of self-interest in individuals’ motivations. Double-blind experiments, designed to create conditions of privacy and anonymity, may engender doubts in subjects regarding the existence of pairings and the disposition of any money they share. Moreover, subjects may view the experiment as a game. We test these conjectures using both traditional and modified dictator experiments.