Tax morale has emerged as one important determinant of tax behavior in general, and hence of undeclared paid activities too. The research aims to measure the effect of tax morale on participation in undeclared work, using the data from the 2013 Eurobarometer survey. If, on the one hand, this survey is a useful tool since it offers a lot of information on individual behaviors and choices towards several aspects connected to tax morale, on the other hand the data are affected by two methodological issues characterizing microdata on participation in shadow activities: non-randomness of non-responses in the dependent variable, that induces sample selection bias in parameter estimates, and misclassification of the response variable.
Our goal is to specify and estimate a model that can address both problems, while measuring the effects of tax morale on undeclared work.
The research is undoubtedly innovative. To our best knowledge it represents the first proposal in the literature for the following goals:
- to measure the effect of tax morale on undeclared paid activities taking into account censoring mechanism and misclassification of the response variable;
- to make the probabilities of misclassification depending on individual characteristics.