Women in economics: the role of gendered references at entry in the profession

We study the presence and the extent of gender differences in reference letters for graduate students in economics and how these may affect the start of young researchers' careers. To these ends, we build a novel rich dataset covering ten cohorts of academic job market applicants to two top institutions hiring on the international market. We collect information from the application packages and conduct text analysis of reference letters using Natural Language Processing (NLP) techniques in order to measure gender differences in the style and content of the letters. We then combine the resulting measures with information on the applicants’ subsequent labor market outcomes as extrapolated from the main online repositories. Our results reveal that male and female candidates receive different support from their sponsors and are described in systematically different terms. While female advisors talk more about personal characteristics, only male advisors do so at a different extent for male and female candidates. Such differences in how candidates are talked about affect subsequent career outcomes and explain a non-negligible part (5 to 8% approximately) of the observed gender gaps.