Three findings stand out. First, the choice of boss matters. There is substantial variation in boss quality as measured by the effect on worker productivity. Replacing a boss who is in the lower 10% of boss quality with one who is in the upper 10% of boss quality increases a team’s total output by about the same amount as would adding one worker to a nine member team. Using a normalization, this implies that the average boss is about 1.75 times as productive as the average worker. Second, boss’s primary activity is teaching skills that persist. Third, efficient assignment allocates the better bosses to the better workers because good bosses increase the productivity of high quality workers by more than that of low quality workers.