Association for Public Policy Analysis & Management

In this paper, we compare estimates of the black-white male earnings gap using self-reported earnings from the Census Bureau’s Current Population Survey (CPS) Annual Social and Economic Supplement (ASEC) and reported earnings from the Social Security Administration’s Detailed Earnings (DER) File. We find that using self-reported earnings data from the CPS masks the magnitude of inequality between black and white earnings and that research using imputed earnings with self-reported data has the potential to skew earnings gap estimates generated from self-reported household survey data compared to administrative records. Overall, our results provide informative and relevant information for understanding the extent to which self-reported earnings data and reported earnings from administrative records influences the estimation of the black-white earnings gap. The results will also help inform the potential for improvements to official national inequality estimates and measurements.