Two problems (at least) with this argument. There's no objective or consistent measure of performance. A programmer may be super productive in one context but

Two problems (at least) with this argument. There's no objective or consistent measure of performance. A programmer may be super productive in one context but average or even a dud in another. Productivity is heavily influenced by factors beyond the individual, especially co-workers and the work environment. Performance does not necessarily follow a bell curve distribution. For some skills a power law better describes the distribution of high and low performers. See http://www.forbes.com/sites/joshbersin/2014/02/19/the-myth-of-the-bell-curve-look-for-the-hyper-performers/ http://www.forbes.com/sites/joshbersin/2014/02/19/the-myth-o...