IZA DP No. 783: Match Bias in Wage Gap Estimates Due to Earnings Imputation
published in: Journal of Labor Economics, 2004, 22 (3), 689-722
About 30% of workers in the CPS have earnings imputed. Wage gap estimates are biased toward zero when the attribute being studied (e.g., union status) is not a criterion used to match donors to nonrespondents. An expression for “match bias” is derived in which attenuation equals the sum of match error rates. In practice, attenuation can be approximated by the proportion with imputed earnings. Union wage gap estimates with match bias removed are presented for 1973-2001. Estimates in recent years are biased downward 5 percentage points. Bias in gap estimates accompanying other non-match criteria (public sector, industry, etc.) is examined.