WIDER Working Papers
December 15, 2018
Top income under-coverage in developing countries not only leads to downward biased inequality indicators but might also affect the ex-ante evaluation of progressive tax reforms.
We propose a simple adjustment to top incomes for formal employees (e.g. employees affiliated with social security) in survey data from Ecuador based on information from administrative tax records. ECUAMOD, the tax-benefit microsimulation model for Ecuador, is then used to compare income inequality, total tax revenue, and work incentives with and without top income adjustment under the baseline tax-benefit system and for a hypothetical income tax reform.
Our study shows that adjusting top incomes in survey data allows us to capture 98 per cent of income tax revenue in our simulations. Moreover, income inequality measured by the Gini coefficient increases by 3 points following the adjustments of top incomes. The evaluation of our progressive tax reform shows that our baseline results with unadjusted top income data capture well the percentage change in tax revenue but underestimate the additional tax revenue amount by 22 per cent.
We conclude with a discussion of the challenges to combine survey data and tax records in developing countries.