Fertility Dynamics across Occupations in Germany

Christian Schmitt, University of Rostock & DIW Berlin
Sandra Bohmann, DIW Berlin
David Richter, DIW Berlin

We investigate fertility across groups of educational and vocational attainment and across occupations in Germany. Building on Hoem and collegues’ (2006) analyses of childlessness in relation to educational field in Sweden, we study fertility differentials across occupations and educational groups. We identify occupations with outstandingly high and low levels of family size and childlessness, respectively, to extend our understanding how the employment nexus shapes work-family compatibility.

Firstly, we provide an overview on different levels of childlessness and number of children across educational and vocational fields, and across occupations. As age profiles differ in specific types of employment, we rely on longitudinal data and Kaplan-Meier based estimation procedures. Secondly, we apply event history methods to longitudinal data for selected employment groups in order to identify whether a large family size in a specific occupation is consequence of high work-family compatibility or whether individuals with large family size ideals “select” themselves into compatible occupations.

For the first step of our analysis, we use three waves of the German Socio-economic Panel Study (SOEP, 2013-2015) and the German Classification of Occupations (KldB). The KldB is compatible with ISCO-08, but provides a more detailed perspective on occupations with 1,286 different classifications. We originate from the lowest, most precise level of occupations and merge occupations onto the respective higher level if case numbers provide less than 100 observations. Doing so, we can rely on a total of 260 occupational categories. For the differentiation of educational and vocational attainment, we primarily rely on the ISCED indicator.

We find that cleaning workers and housekeepers have the highest number of children across occupations. Retailers, programmers, and academic researchers, in contrast have small family sizes and excessively high levels of childlessness. We aim to include these groups in our longitudinal analyses, which utilizes a piecewise constant approach to parity progressions.


Presented in Session 1137: Fertility