Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Motivated by models of worker flows, we argue in this paper that monopsonistic discrimination may be a substantial factor behind the overall gender wage gap. Using matched employer-employee data from Norway, we estimate establishment-specific wage premiums separately for men and women, conditioning on fixed individual effects. Regressions of worker turnover on the wage premium identify less wage elastic labour supply facing each establishment of women than that of men. Workforce gender composition is strongly related to employers' wage policies. The results suggest that 70-90% of the gender wage gap for low-educated workers may be attributed to differences in labour market frictions between men and women, while the similar figures for high-educated workers ranges from 20 to 70%.