Initial Industry and Long-Term Earnings Growth

We test whether employment growth in a male worker's initial industry influences earnings growth using the 1979 National Longitudinal Survey of Youth. We follow workers for 20 years after reporting their first industry, finding that lower employment growth in their initial industry implies substantially lower earnings growth. Notably, after controlling for observable skills, controls for family background and region have no impact on estimates. Effects appear larger for initial occupations that involve more routine or manual tasks as well as for occupations that involve less abstract tasks, but these differences are not statistically significant.