Cumplimiento fiscal y bienes públicos: ¿Son realmente compatibles?

This work provides experimental evidence of the determinants of tax compliance in Mexico, and compares them to the results of an experiment administered in the United States reported by Alm, Jackson and McKee (1992). The results for both countries are contrasted to the predictions of the Alligham-Sandmo model, whose basic assumption is that the probability of being detected cheating taxes is exogenous and independent of the level and type of income reported. Specifically, in both experiments, increasing the probability of audit always increases compliance; higher penalty rates are ineffective to increase compliance; there is a negative relationship between tax rates and compliance; but there was free riding in the public goods sessions in Mexico, even when the government is efficient providing them.