Standardized Pay Rates and the Gender Wage Gap

A substantial share of the gender wage gap can be explained by gender differences in wage-negotiation behavior (for instance: Leibbrandt and List, 2015, Biasi and Sarsons, 2021, Roussille, 2024). A simple solution for this problem may be an objective mechanism which bypasses bargaining to assign wages based on objective job characteristics. Many firms already engage in (some) pay standardization through the assignment of job titles to wage ranges. However, standardization is ineffective if classification is subject to bargaining. This may explain why gender gaps persist even in countries with a high degree of wage standardization: Women may be systematically assigned to lower job titles, creating an “assignment gap”. We study wage standardization through sectoral bargaining in Germany. Under sectoral bargaining – Europe’s dominant wage setting regime (Jäger et al., 2024) – unions negotiate binding contracts for an entire industry. These agreements specify job titles and respective minimum wages. We identify job titles in the structure of earnings survey (2010, 2014, 2018), following Busch and Weil (2024). The data contains a random subset of > 1, 000, 000 employees in > 23, 000 establishments. We observe circa 3, 000 job titles which explain 79% of the variation in wages. To test whether there is a gap in job title assignment, we proceed in two steps: First, we run a detailed fixed effects model which predicts the rank of a job title (by wage) with gender and a proxy for the assignment mechanism (bargaining unit x [age, tenure, 5 digit occupation code, job level, part time status, establishment size]). Second, we test whether gender gaps are smaller if there is less discretion in assignment, for instance in the public sector or the presence of works councils. We find substantial gender gaps in the range of 0.2 to 0.4 job title ranks, both withinand between-establishments. Differences in job title rank explains 25% of the overall gender gap in full-time wages. We find that publicly owned establishments and those with a works council have a substantially smaller rank gap.