This paper examines the wage effects of different types of career interruptions. We consider the timing and duration of non-employment spells by exploiting an administrative data set of German social security accounts (IAB employment sample) supplemented with information on the employees? entire working lives (IAB supplement sample I). These data allow us to distinguish between employment breaks due to registered unemployment, formal parental leave, training or other reasons – a distinction which can only be approximated using just the IAB employment sample. Our IV fixed effects estimation results suggest that women?s labor supply is endogenously determined, whereas men?s employment histories can be treated as exogenous. Career interruptions reduce the wage rates of both men and women. Moreover, the wage cuts resulting from unemployment, parental leave and additional home time are larger than the pure human capital effects of missing experience, hinting at a possible stigmatization of workers with discontinuous employment histories.
career interruptions returns to experience wage differentials panel estimation