Richard Rogerson // Princeton UniversityTo the profile
Occupations, Life Cycle Wage Growth and InequalityResearch Seminars
In the first part of this paper presented in this ZEW Research Seminar the authors document several novel facts about occupations, the life cycle and inequality using cross-section data from the CPS. First, a large portion of mean wage differences across occupations reflects differences in mean life cycle wage growth across occupations. Second, occupations that feature higher mean life cycle wage growth tend to experience a higher increase in cross-sectional wage dispersion over the life cycle. That is, differences in mean life cycle wage growth differences across occupations are driven by higher order moments of the life cycle wage distribution for older individuals; occupations with high mean life cycle wage growth tend to have a small mass of individuals who earn very high wages when old. These patterns hold for both males and females.
For the second part of the ZEW Research Seminar, the authors present and estimate a simple two-occupation life cycle model that features learning by doing. Showing that it can account for the key features of the data that were documented even when the only source of comparative advantage in the model is with respect to learning ability across occupations. A key feature of the estimates is that the variance of learning ability is much greater in the high wage occupation than in the low wage occupation, though mean learning ability is modestly higher in the low wage occupation. Importantly, the estimates imply large selection effects with regard to life cycle wage growth across occupations. That is, although the observed life cycle wage growth for individuals is much greater in the high wage occupation, a randomly chosen individual from the low wage occupation will not achieve this rate of wage growth if switched into the high wage occupation.
In the last part of the ZEW Research Seminar, the authors use the estimated model to carry out counterfactuals to assess how changes in the occupational employment structure affect overall inequality. A key finding is that mean occupational wages do not provide useful guidance for inferring the effects of changes in the employment distribution on overall wage inequality.