Econometric Analysis of the German Wage and Earnings Distribution

The thesis comprises three empirical studies on the German wage and earnings distribution with a particular focus on the application of modern econometric estimation methods. The first study investigates changes in the German wage distribution among men over the period 1995-2010, i.e. a time when Germany experienced a sharp rise in wage inequality. The analysis uses a unified framework which considers an extensive set of explanatory factors including personal characteristics, measures of internationalization, task composition, union coverage, industry, region and firm characteristics. Among other things, the results show that previous studies on German wage inequality most likely underestimated the role of de-unionization as important parts of de-unionization occurred within establishments. The second study provides a detailed decomposition analysis of rising lifetime earnings inequality in Germany using individual employment biographies constructed from high-quality administrative data. It shows that significant parts of rising lifetime earnings inequality among West German men born between the years 1955 and 1974 can be attributed to a lower labor market participation as well as the educational expansion among later cohorts. The analysis reveals similarities with the development in the U.S. in the sense that the cohorts studied did not only face an increase in inequality, but also a stagnation in earnings for a major part of their career. The third study evaluates the importance of selection effects in the context of the gender wage gap. The econometric approach enhances the Albrecht et al. (2009) method by providing a practical solution to the problem outlined in Huber and Melly (2015). The suggested transformation effectively eliminates violations of conditional independence. In addition, the functionality of the approach is illustrated from both a theoretical and an empirical point of view. The results also reveal that a positive female selection on unobservables potentially leads to an underestimation of the gender pay gap in Germany