The Rich Underreport their Income: Assessing Bias in Inequality Estimates and Correction Methods using Linked Survey and Tax Data

Do survey respondents misreport their income? If so, how does misreporting correlate with income, how does this affect estimates of income inequality, and how well do existing methods correct for bias? We use a novel database in which a subsample of Uruguay's official household survey has been linked to tax records to document the extent and distribution of labor income underreporting and to assess the performance of various existing methods to correct inequality estimates. Individuals in the upper half of the income distribution tend to report less labor income in household surveys than those same individuals earn according to tax returns, and underreporting is increasing in income. Using simulations, we find that this leads to downward-biased inequality estimates. Correction methods that rely only on survey data barely affect the biased inequality estimates, while methods that combine survey and tax data can lead to over-correction and overestimation of inequality.