Abstract
This paper studies the effect of personality on performance by gender. We investigate this relationship in the academic labor market for Economists. We build a unique and novel data set from individual CVs of 2,471 faculty members employed at the Top 100 US University Economics Departments and use the facial Width -to-Height ratio (fWHR) to proxy for personality traits. We are interested in the marginal effect of personality on performance after accounting for the common drivers of variations in academic performance. We find that non-dominant females and dominant males achieve better outcomes. This effect, to some degree, is driven by the accumulation of non-cognitive skills, e.g. networks. While our data does not allow to investigate whether this is a result of choice, our results are in line with discrimination based on violation of gender norms.