Measurement error in longitudinal earnings data: evidence from Germany

Abstract

"We present evidence on the extent of measurement error in German longitudinal earnings data. Qualitatively, we confrm the main result of the international literature: longitudinal earnings data are relatively reliable in a cross section but much less so in frst diferences. Quantitatively, in the cross section our fndings are very similar to those of Bound and Krueger (J Labor Econ 9:1–24, 1991) and Pischke (J Bus Econ Stat 13:305–314, 1995) for the United States while we fnd even stronger evidence that frst-diferencing exacerbates measurement error problems. We also show that measurement error in our survey data is not“classical” as it is negatively correlated with administrative earnings and positively autocorrelated over an extended period of time. Additionally, we estimate a model of measurement error stemming from underreporting of transitory earnings shocks in combination with a white-noise component and make a number of methodological contributions. Our results are robust to the use of two diferent linked surveyadministrative data sets and various other sensitivity checks." (Author's abstract, IAB-Doku, © Springer) ((en))

Cite article

Schmillen, A., Umkehrer, M. & Wachter, T. (2024): Measurement error in longitudinal earnings data: evidence from Germany. In: Journal for labour market research, Vol. 58, accepted on May 13, 2024. DOI:10.1186/s12651-024-00366-x