Male Advantage and the Gender Composition of Jobs: Who Rides the Glass Escalator?

Abstract

Is the gender gap in pay constant across all jobs, or does the gender composition of the job affect male advantage? Using data from the NLSY and a finely detailed measure of the gender composition of jobs, I investigate gender differences in wages and in wage growth. I show how they differ between female-dominated, male-dominated, and balanced jobs. Predictions from Kanter's theory of tokenism and the Williams and Acker theory of gendered organizations are tested. Findings indicate that men are advantaged, net of controls, in both pay levels and wage growth in all jobs, regardless of gender composition. Contrary to predictions generated from Kanter's tokenism theory, men do not suffer when they are tokens, in terms of pay. Not only are predictions from Kanter's theory untrue for male tokens, they also do not hold for female tokens when it comes to wages. Rather, consistent with the Williams and Acker theory of gendered organizations, men are no more-and no less-advantaged when women are tokens; in terms of earnings, men are uniformly advantaged in male-dominated, female-dominated, and balanced jobs. Analyses of promotions data indicate that men are also not additionally advantaged whether they are the numerically dominant or minority gender; in fact, male advantage in promotions is the smallest when men are tokens.