This dissertation is comprised of three papers. In Chapter 1 I analyze if career heterogeneity in terms of life-cycle earnings, occupational mobility and unemployment is predominantly driven by skills acquired prior to labor market entry or by decisions made and shocks accumulated over the working life. My study is based on a Dynamic Discrete Choice model that enriches the proto-typical dynamic Roy-model with a number of potentially important sources of career heterogeneity, such as match heterogeneity and permanent shocks to skills. I find that a large fraction of life-cycle income inequality is driven by match heterogeneity among workers with the same observable and unobservable credentials. Differences in comparative advantages, though quantitatively important as well, have a much smaller impact than what has been found in research that relies on estimates from more restrictive dynamic Roy models. In Chapter 2 I estimate a flexible non-stationary variance components model of residual earnings dynamics and investigate if recent increases in residual inequality are caused by an increase of the variances of permanent, persistent or transitory shocks. My results suggest that underlying sources of increasing wage inequality are very different across education groups. Most importantly, only the lesser educated experience a large increase in earnings instability. Chapter 1 and 2 utilize a unique administrative data set from Germany that follows workers from the time of labor market entry until twenty-three years into their careers. In the last chapter I investigate empirically if a particular set of pre-labor market skills – namely university student achievement – can be fostered by assigning male teachers to male students and female teachers to female students. I find that being taught by a same-sex instructor helps students to improve their relative grade performance and the likelihood of completing a course, but the magnitudes of these effects are small.

This dissertation is comprised of three papers. In Chapter 1 I analyze if career heterogeneity in terms of life-cycle earnings, occupational mobility and unemployment is predominantly driven by skills acquired prior to labor market entry or by decisions made and shocks accumulated over the working life. My study is based on a Dynamic Discrete Choice model that enriches the proto-typical dynamic Roy-model with a number of potentially important sources of career heterogeneity, such as match heterogeneity and permanent shocks to skills. I find that a large fraction of life-cycle income inequality is driven by match heterogeneity among workers with the same observable and unobservable credentials. Differences in comparative advantages, though quantitatively important as well, have a much smaller impact than what has been found in research that relies on estimates from more restrictive dynamic Roy models. In Chapter 2 I estimate a flexible non-stationary variance components model of residual earnings dynamics and investigate if recent increases in residual inequality are caused by an increase of the variances of permanent, persistent or transitory shocks. My results suggest that underlying sources of increasing wage inequality are very different across education groups. Most importantly, only the lesser educated experience a large increase in earnings instability. Chapter 1 and 2 utilize a unique administrative data set from Germany that follows workers from the time of labor market entry until twenty-three years into their careers. In the last chapter I investigate empirically if a particular set of pre-labor market skills – namely university student achievement – can be fostered by assigning male teachers to male students and female teachers to female students. I find that being taught by a same-sex instructor helps students to improve their relative grade performance and the likelihood of completing a course, but the magnitudes of these effects are small.