Robots and Inequality Between and Within Occupations

We examine the impact of industrial robots on wage inequality both between and within occupations. Analyzing two decades of U.S. data, we find that moving from the 25th to the 75th percentile of industry-level robot adoption reduces the within-occupation wage inequality by 16% and increases the between-occupation inequality by 7% of their respective standard deviations. These results are robust to confounding factors, different inequality measures, and instrumental variables. Our evidence shows that robots compress within-occupation wage inequality by lowering wages at the top of the distribution while increasing between-occupation inequality by impacting the middle of the distribution. We show that these findings are consistent with a standard wage-posting model incorporating endogenous robot adoption. Overall, our results suggest that robots exacerbate wage polarization while reducing the rents of the highest-paid workers in the occupations most vulnerable to automation.