Key points are not available for this paper at this time.
An evolutionary game provides a parsimonious benchmark for repeated interactions between tax authorities and taxpayers in income-tax evasion decisions. The model relaxes the assumption of fully rational, forward-looking agents and characterizes the stationary shares of inspection and full compliance through payoff comparisons. It predicts that lower tax rates and higher penalties are associated with a larger share of fully honest taxpayers. A laboratory experiment from the taxpayers' perspective examines how tax-authority policy instruments affect decisions across repeated rounds. Consistent with the model's comparative statics, a higher tax rate is associated with a lower probability of full honesty, whereas a higher penalty rate is associated with a higher probability of full honesty. The data also show patterns consistent with experience-based adjustment, including lower compliance after an audit, a pattern commonly labeled the “bomb-crater effect”. Because subjective beliefs were not elicited, we frame these behavioral mechanisms as interpretations of the observed patterns rather than as directly identified psychological channels.
Fedeli et al. (Thu,) studied this question.