Errors in Self-Reported Earnings: The Role of Previous Earnings Volatility

Abstract: I report the measurement error in self-reported earnings for a developing country. Administrative data from the Federated States of Micronesia’s (FSM) Social Security office are matched to the FSM census data. I find that the error in annual self-reported earnings is centered on zero but less efficient than results from the US. Additionally the error is not classical measurement error – I find evidence for mean reversion in the data. Using the earnings histories provided in the Social Security data, I construct accurate measures of past deviations of administratively recorded earnings to identify the impact of transitory income (income volatility) on current reporting of earnings. The prior earnings volatility is an important determinant of the error in earnings for the current period. However, the effect of prior shocks diminish significantly over time; suggesting that information on transitory income shocks will be helpful in evaluating the usefulness of self-reported earnings measures in applied work