On Self-Serving Norm Beliefs and Motivated Processing of Norm Information
25 Pages Posted: 28 Dec 2018 Last revised: 13 Oct 2019
Date Written: November 26, 2018
I experimentally study settings where an individual may have an incentive to adopt a negative belief about the honesty norm, which means whether others lie in similar situations, in order to justify her own past or future lying behavior. I elicit incentivized beliefs both from subjects with such an incentive and from neutral third parties without such an incentive, before and after each time they receive a signal indicating whether a randomly drawn pilot subject lied. I find no difference between the two sets of subjects in their prior beliefs, but after receiving the same random signals, the subjects with the incentive end up believing in a significantly higher prevalence of lying than the bystanders, although the difference in the posterior belief does not lead to more lying behavior subsequently. Specifically, subjects with and without the incentive both under-update signals of lying behavior, but only subjects with the incentive under-update signals of honest behavior while subjects without the incentive process those signals in a Bayesian manner. Indeed, such a different way to process information of those with the incentive reduces a positivity bias in norm beliefs in comparison to the actual lying behavior. Finally, within-subject analysis shows that subjects with the incentive who receive mixed signals, deviate more from Bayes' rule when receiving a signal of lying behavior than when receiving a signal of honest behavior; They also lie more subsequently compared to before receiving the signals --- suggesting asymmetric patterns of information processing and social contagion for honesty and for deception.
Keywords: moral wiggle room; information processing; peer effects; norm compliance; lying behavior
JEL Classification: D83, D84, C91
Suggested Citation: Suggested Citation