Paper title: Biases in the Measurement of Labour Market Dynamics

This paper analyses worker transitions on the German labour market derived from different data sources. These include the two German micro data sets which provide high-frequency observations on workers’ employment and unemployment histories: the German Socioeconomic Panel (SOEP), a representative household survey containing retrospective results on monthly labour market histories, and the Employment Sample of the IAB (IABS), a process-induced administrative data set covering 2% of workers employed in social security employment. We first compute the stocks of and the flows between labour market states from the two data sources. This yields a comprehensive overview of German labour market dynamics. We then analyze the differences between the results obtained from the SOEP and the IABS. This highlights the importance of differences in data collection and sample selection methods. Our analysis also shows which groups of the labour market are particularly affected by measurement error, and which role measurement issues play when establishing the stylised facts about the cyclicality of labour market dynamics. Finally, we apply several external validation procedures to the computed stocks and flows. In particular, we compare the results derived from the two micro data sets to official figures published by the German Employment Office and the German Statistical Office. We are thus able to make a statement about which data set fits the aggregate evidence best. We argue that our results are crucial for any study related to labour market transitions, job tenure, and unemployment duration.