I haven’t read these papers carefully yet to see if they can control for this, but wouldn’t it also be true that promotion is very often into managerial roles where marginal value product is less measurable to begin with—which would make it easy to under estimate productivity?
Generally yes, which is precisely what makes their paper awesome. They're able to calculate the value-add from having someone be the manager of a salesteam, and since the output is very precise their value add can be very precise too.
Nice.
I haven’t read these papers carefully yet to see if they can control for this, but wouldn’t it also be true that promotion is very often into managerial roles where marginal value product is less measurable to begin with—which would make it easy to under estimate productivity?
Generally yes, which is precisely what makes their paper awesome. They're able to calculate the value-add from having someone be the manager of a salesteam, and since the output is very precise their value add can be very precise too.