Occupational Complexity, Experience, and the Gender Wage Gap

Abstract I explore the role of individuals' skills and work experience in explaining the gender wage gap across occupations. I use the O*NET dataset to build an index of occupational complexity: the ratio of abstract to manual tasks. The ratio of female to male wages is U-shaped across occupations ordered by increasing complexity. The U-shape flattens over the lifecycle and across successive cohorts. I develop an occupational choice model with male and female individuals who are heterogeneous with respect to their level of skill. An individual's skill at a point in time depends on his/her exogenous initial level of skill and his/her work experience. Individuals decide how much time to spend in the labor market. Occupations differ by two features in my model: 1) the skill required to perform, and 2) the marginal product of skill. If occupations involve simple tasks, output and wages vary little across workers of different initial skill levels. Also, acquired work experience influences wages only slightly, since little can be learned by performing simple tasks. I discipline the model with data on occupational complexity, occupational choice, labor supply and male wages. The model reproduces the gender wage gap across occupations for cohorts born between 1915 and 1955. The little work experience of females relative to that of males is a key factor behind the U-shape. It decreases female wages disproportionately across occupations and it influences female occupational selection. I find that 69% of the lifecycle gender wage gap is attributable to work experience. Removing differences in work experience between genders results in a larger fraction of females choosing occupations for which the gender wage differential is smaller. JEL: J1, J2, J3