The Value of Bosses
Replacing a bad boss with a good one increases productivity of each subordinate's output by more than 10 percent.
One extreme view of the labor market holds that bosses are irrelevant and that worker productivity is unaffected by the choice of supervisor. Anyone can do the supervisor's job because the supervisor has little direct impact on worker output. At the other extreme, is the view that workers are indistinguishable and that the output of the firm depends only on how well bosses use labor. Whichever is the case, we know that a significant fraction of resources is devoted to supervision: among manufacturing workers, front-line supervisors comprised 10 percent of the non-managerial workforce in 2010; among retail trade workers, front line supervisors comprised 12 percent of the non-managerial workforce.
In The Value of Bosses (NBER Working Paper No. 18317), authors Edward Lazear, Kathryn Shaw, and Christopher Stanton use data from a large service oriented company to examine the effects of bosses on their workers' productivity. They estimate the daily productivity for 23,878 workers matched to 1,940 bosses over five years and find that bosses vary greatly in productivity, with the difference between the best bosses and the worst bosses being significant.
The average worker at the firm they study produces about 10.3 units of output per hour. Bosses in the top performance decile increase each worker's output by about 1.3 units per hour, or more than 10 percent of average output, relative to bosses in the bottom decile. Given that the typical boss supervises about nine workers, this amounts to a change in total productivity that is larger than the amount produced by the average worker. Based on what the authors believe is a conservative estimation method, the average boss adds about 1.75 times as much output as the average worker, which is in line with the differences in pay received by the two types of employees.
The authors further determine that the boss's primary job is teaching, defined as providing skills that persist. Contemporaneous motivation of workers is secondary. They also observe that the worst bosses are unlikely to be retained. Over a one-year period, bosses in the lowest 10 percent of the quality distribution are 67 percent more likely to leave the firm than bosses in the top 90 percent of the distribution.