Defaults and dishonesty – Evidence from a representative sample in the lab

Abstract

Unethical behavior is a massive societal problem that occurs in almost all aspects of our everyday lives. A prominent theory in behavioral economics is that the decision about whether to behave dishonestly involves a balance between securing greater gains, on the one hand, and maintaining a good self-image, on the other. In the present study, I explore whether simple manipulation of default behavior can influence this balance. In a laboratory experiment using a sample of the general population, the impact of default answers on dishonesty is studied. In a task of reporting the outcome of die rolls in private, I randomly impose default answers. The defaults are either high and favorable, low and unfavorable, the expected mean of the die rolls, or an empty condition with no default. I find that the high default answers are effective at increasing reporting, whereas the low defaults do not result in a corresponding reduction, compared to no default or the expected mean as the default answer. As a reference group, the experiment is repeated among a student sample with the results revealing that students react differently to the defaults. In particular, students did not increase their answers with high defaults, but including the mean as a default answer resulted in a significantly higher degree of cheating.