Employer Size and Supervisor Earnings: Evidence from Britain

Using British linked employer-employee data, we show that the establishment size effect for supervisors is approximately twice that for non-supervisors. This difference is routinely statistically significant, not explained by other controls and is an important determinant of the difference in earnings between supervisors and non-supervisors. Moreover, we use separate British longitudinal data to confirm both the statistically different effect and that it is not explained by worker fixed effects. Event study evidence and information on skill match suggest that the larger return to supervisors reflects, in large part, match specific returns supporting the view that talented supervisors receive a return on that talent only with larger employers.