Analysing the gender wage gap Empirical evidence from Germany and the United States

Abstract

The main objective of this thesis is to investigate the potential sources of the gender wage gap in Germany and the United States. For that purpose, we employ high quality microdata sets from the German Socio-Economic panel (SOEP) and IPUMS CPS. We use cross-sectional data for the years 1989 and 2019 to study a representative sample of full-time employees between the age of 25 to 64. By employing the Kitagawa-Oaxaca-Blinder (KOB) decomposition method (Kitagawa 1955; Oaxaca 1973; Blinder 1973) we estimate how much of the unadjusted gender wage gap in the countries is attributed gender differences in measured characteristics. Furthermore, by applying a Juhn, Murphy, and Pierce (JMP) (1991) decomposition we investigate how relative improvements in terms of characteristics and rising return to these characteristics affects the gender wage gap over time. The aim is to understand if increasing overall inequality counterbalances women’s progress in the labour market and has a widening effect on the gender wage gap. Lastly, Blau and Kahn (1996) find that countries with more compressed wage distributions have smaller gender wage gaps. We investigate this by a JMP decomposition of the U.S.-German difference in the gender wage gap to understand if differences in return to characteristics is the most important contributor to international differences in the gender wage gap. The results show that the gender wage gap declines in Germany and the United States between 1989 and 2019. The results from the KOB decomposition show that gender differences in observable characteristics in total no longer explain the gender wage gap for these countries in 2019. Gender differences in distribution by industry however continues to explain malefemale wage disparities over the period studied. The JMP results show that changes in the return to characteristics negates some of the progress made by female workers over the period. Lastly, the most important factor for explaining the U.S- German difference in the gender wage gap is the relative differences in return to characteristics. This effect was however stronger in 1989 and reflects that the wage distribution in Germany have become more dispersed over the period of study and is thus more similar to that observed in the United States.