Persistent Heterogeneity in Tax Compliance and Income Inequality: A Discrete Choice Approach

We present an overlapping generations model that examines how the dynamics of income tax compliance among taxpayers influence the evolution of income inequality. We measure income distribution using Gini coefficients for the net wage income, capital income and net aggregate income. The first two coefficients measure within-stage income inequality, whereas the third measures total inter-generational income inequality. The proportion of tax evaders is endogenously time-varying and state-dependent, as determined by adaptively rational equilibrium dynamics (ARED). The long-run equilibrium analytical results replicate (and thus provide a rationale for) several empirical regularities, as well as conceptual and intuitive predictions, related to tax compliance. First, the proportion of non-compliance depends on the tax rate, the expected cost of tax evasion, and the median of the distribution of tax morale across taxpayers, which may be strictly negative, zero, or strictly positive. Second, heterogeneity in tax compliance exhibits persistence. Third, the proportion of non-compliance increases with the tax rate and decreases with the probability of detection, the penalty rate, and the median tax morale. As shown by numerical simulations with plausible parameter values, the relationship between the median tax morale and the Gini coefficients is inverted-U shaped, with the maximum attained at a strictly positive median tax morale. A higher probability of detection reduces the Gini coefficients, albeit modestly, and only when the median tax morale is sufficiently positive, whereas a higher tax rate increases the Gini coefficients over the considered range of plausible parameter values.