dc.description.abstract
This dissertation consists of four chapters that contribute to the literature on gender and labor economics. More specifically, this dissertation contributes to exploring the role of women as corporate leaders in the labor market and in firms, examining the causes and consequences of women's (under)representation in management, and understanding the interaction of these factors with gender stereotypes.
Chapter 2 documents the underrepresentation of women as leaders of German companies by presenting data from the DIW Women Executives Barometer, the most comprehensive data collection on women's representation on supervisory and executive boards in Germany, which was recently made available for scientific research. The data shows that, despite progress in recent years, women remain underrepresented on both supervisory and executive boards in Germany, with shares ranging from 33 to 38 percent on supervisory boards and 14 to 23 percent on executive boards of the largest publicly listed companies as of the end of 2023. Women's representation is even lower among chairs of executive or supervisory boards of these companies, with values below 10 percent in all groups assessed. The DIW Women Executives Barometer is then used as a data source for further work in Chapter 3.
Chapter 3 examines how women and men on German company boards are portrayed in the media and explores the relationship between this coverage and gender stereotypes. By providing initial evidence on gender differences and stereotypes in newspaper coverage of board members through quantitative text analysis of more than 45,000 newspaper articles, this chapter expands our understanding of stereotypes as barriers to women’s career advancement and shaping public perception of female leaders through media coverage. In regularized regressions, terms related to family and social interaction predict articles about women, while terms related to power and competition, as well as to failure, scandals, and adversity, predict articles about men. An association of women with family and men with careers is further shown using the term frequency-inverse document frequency (tf-idf) and word embeddings. Additionally, agentic language depicting male stereotypes like success-orientation is more prevalent for men, while communal language related to female stereotypes like caregiving is more associated with women. Economists' view of gender stereotypes as a form of statistical discrimination suggests that stereotypes in the portrayal of female managers in newspapers should decrease over time as more women attain board positions and information asymmetries about their characteristics diminish. However, I find no clear-cut changes in newspaper coverage in the analyzed articles over the period from 2010 to 2022. Further, in the context of statistical discrimination, gender differences in newspaper coverage would arise if they correctly reflect gender differences in aggregate distributions of managers' characteristics. I assess demographics and previously unanalyzed psychological traits of managers from representative data of the German population, finding that female managers are less likely to be married or have children than male managers. There is no evidence for gender gaps in agency among managers. Thus, the stereotyped representation of board members in newspapers to a large extent does not seem to reflect aggregate distributions of characteristics of women and men in these positions, providing little support for an explanation based on statistical discrimination.
Chapter 4 then asks to what extent gender stereotypes in newspaper coverage of company board members affect perception and economic decision-making of newspaper readers. Taking the result from Chapter 3 that more family-related language is used in newspaper articles on female than male company leaders as a starting point, we assess this question in a randomized online experiment. We show participants articles consisting of elements from real newspaper coverage on a real company and its CEO, varying whether and how information about their family is presented. We then ask participants incentivized questions on their perception of the CEO's competence (measured by expected `survival' in the firm and their rating by employees on Glassdoor), the firm's performance on the stock market, and to make a decision on an investment in the firm. We find that expected firm performance substantially differs by CEO gender with participants being less likely to believe in better stock performance in the year after compared to the year before CEO appointment for female CEOs. Although expected stock performance does not differ by CEO coverage, investments in the firm's stock are significantly lower for female CEOs when their family is neutrally mentioned. However, highlighting the successful management of family and career as a trade-off for female CEOs does not result in less favorable investment decisions. The treatment effects are most pronounced for female respondents and parents. Further, we find that women expect female CEOs to be rated worse by their employees, in particular for the treatment highlighting a trade-off between family and career, while male participants perceive an employee bonus for female CEOs in the trade-off treatment compared to female CEOs without a family mention. There is no effect of stereotypical coverage on CEO `survival' over a two or five year horizon. Through quantitative analyses of free-text questions, we find that considerations about family and gender play a role in respondents' reasoning.
Finally, Chapter 5 focuses on spillover effects of female leadership by assessing the causal impact of women's representation in management on labor market outcomes of their direct subordinates on the establishment level. We investigate the influence of women's representation in first- and second-level management on the gender pay gap among employees in German establishments. To this end, we estimate a panel model with establishment fixed effects and industry-specific time dummies based on Linked-Employer-Employee data from the years 2004 to 2018. Our results show that a higher share of women in management significantly reduces the gender pay gap within the establishment. An increase in the share of women in first-level management from zero to above 33 percent decreases the adjusted gender pay gap from a baseline of 15 percent by 1.2 percentage points, i.e. to roughly 14 percent. The effect is stronger for women in second-level compared to first-level management, indicating that female managers with closer interactions with their subordinates have a higher impact on the gender pay gap than women on higher management levels. Notably, the results are very similar for East and West Germany, despite the lower gender pay gap and more gender-egalitarian social norms in East Germany. From a policy perspective, our findings suggest that increasing the presence of women in management positions has the potential to reduce the gender pay gap to some extent. However, further policy measures will be necessary to fully close the gender gap in pay.
en
