Abstract

The aim of the project is to generate synthetic datasets for the BHP. A simiar dataset was generated at Cornell University last year for the U.S. equivalent of the BHP, the Longitudinal Business Database (LBD). The software routines that were developed for the generation of the LBD should be applied to the BHP. If the direct application of the routines is possible the generated synthetic version of the BHP would satisfy highest data confidentiality requirements. The generated dataset could be disseminated at Cornell University as well. Thus, it would be possible to compare the German and the US job market based on the LBD and the BHP.