Abstract

Using correspondence testing, we investigate if employers exert discrimination against women in Spain’s two largest labor markets, and if this discrimination is due to employers’ reliance on stereotypes about men and women’ expected or prescribed qualifications, or to a moral aversion towards women. We sent two pairs of fictitious man–woman résumés to 1,371 real job offers from a broad selection of occupations advertised on a leading national online job-posting website in 2016. The two pairs of fictitious candidates had equivalent CVs except for candidates' sex. Pairs differed by either one of two experimentally manipulated characteristics of the job applicants: qualifications (meeting standards or higher) and parenthood status (with or without children). We interpret the observed differences in favor of men in callback rates in all pairs, and in the order in which the selected candidates were called for further screening, as measures of overall gender bias in recruitment. We show that this penalty is reduced when women have higher qualifications and that it increases when they have children. We interpret these employers’ adjustments or the openness to modify their decisions when candidate’s personal characteristics differ from their group’s norm as evidence that gender bias in recruitment is largely grounded in employers' stereotypes about the potentially lower productivity of female applicants. Finally, we show that even for the least penalized group of highly qualified non-mothers, discrimination against women is significant, providing some indication that gender discrimination in these labor markets may be partially based on a moral aversion against women.