Experimental economies and tax evasion: The order beyond the market

Abstract

Research on tax evasion will probably never get old. As long as there are taxes, there will also be policy-makers all over the world eager to tackle deviant conduct in the most efficient and efficacious way. To fill this purpose a number of theoretical and empirical frameworks have been developed in economics over the last decades, starting from the classical models of Allingham and Sandmo (1972) where individuals were assumed to be perfectly rational following a pure cost-benefit logic. Today, however, we look at a body of literature which has opened up to a number of new and interdisciplinary findings, also thanks to the inclusion of behavioral aspects that do not necessarily follow the paradigms of the homo economicus. To this end, the discipline of Experimental Economics has developed numerous ways to overcome the distance between economic theory and human behavior. The aim of this survey is to take the reader on a tour through some of these methodologies applied to the analysis of tax evasion, arguing that further research should focus on integrating multi-agent simulation models with outcomes from human subject experiments in order to create useful and necessary tools to administer, consolidate and represent the complex theoretical, empirical and experimental panorama of tax evasion research.