Employment discrimination based on gender and ethnicity and its impact on gender segregation in the labor market: Evidence from an experimental study in the Swedish labor market
Author(s): Magnus Bygren, Moa Bursell
Thursday 15 | 10:40-11:00
Room: TP52
Session: Frontiers in segregation research
Our aim with this study is to contribute to knowledge about the process behind gender segregation
in the labor market, i.e., the process whereby individuals sort by ascribed characteristics – here
gender – into job positions in the labor market. Segregation may be seen as generated by the
combined outcome of decisions located at the supply side and the demand side of the employeremployee match. Previous research has almost exclusively focused the supply side, because this is
where data have been easily accessible. Consequently, we do not know much about whether or how
employer hiring decisions affect gender segregation.
Discrimination research usually focuses on one dimension in isolation, although theoretically there is a relatively obvious intersection between gender and ethnicity that is relevant in this context. For example, discrimination may depend on niche-specific priors on gender and ethnic differences in job competence, and these may, consciously or subconsciously, affect employer beliefs on how prospect employees will perform.
Our specific focus in this study is therefore to analyze how existing structural gender segregation combined with employer discrimination based on gender and ethnicity gives rise to processes that reinforce gender segregation in the labor market.
In order to estimate the impact of employer discrimination on gender segregation, we use experimental correspondence audit data. Correspondence audits consist in constructing applications for fictitious individuals and use these to apply for real jobs in the labor market, and then observe the employer responses to these job applications.
We applied for 16,121 jobs in 28 different occupations in the Swedish labor market, in 2021-2023. The equally merited applicants’ ethnicity and gender were signaled by Swedish sounding names and names typical for individuals with an origin from former Yugoslavia and the larger Middle East. Equal proportions of 'women', 'men', 'natives' and 'foreigners' applied for the jobs. We created a sharewomen variable at the occupational level, drawing on annual data from Statistics Sweden’s occupation register 2023.
We report results in Figure 1. We can see that 1) ‘Swedish’ applicants (as expected) have higher chances of receiving employer callbacks, 2) that ‘Swedish men’ are favored over ‘Swedish women’ in male dominated occupations, and 3) that ‘foreign women’ are heavily favored over ‘foreign men’ in female dominated occupations.
These patterns indicate that employers in general contribute to reinforcing gender segregation in the labor market, and that employers in particular contribute to reinforcing gender segregation for people with a foreign background. The results strongly suggest that gender and ethnicity should be analyzed in combination to get the full picture of how employers' discrimination contributes to gender segregation in the labor market.