Netting Down Gross Earnings Data in the LIS Database: An Evaluation of Two Procedures

LIS researchers who seek to perform country-comparative and / or trend analyses have to account for the fact that in some LIS datasets income variables were reported net of taxes and social security contributions, while in other datasets income variables were reported gross of taxes and social security contributions. In this technical paper we discuss, develop, and evaluate two ‘netting down procedures’ that help reduce bias that would be introduced by directly comparing net and gross datasets. Results of evaluating the performance of these netting down procedures indicate that the validity of the comparison of net and gross datasets can be greatly improved when netting down procedures are applied. In several cases, however, substantial amounts of bias remain. Institute for Innovation and Governance Studies (IGS), University of Twente. [email protected] LIS. [email protected] LIS & Graduate Center, City University of New York. [email protected]