Incentive pay.... Joel Spolsky of Joel On Software notes the problems that come about because of "incentive pay." This obviously applies to areas outside software. Joel [from gnxp] adds: When I worked at MSFT, the lynchpin of the annual review process was writing "clear, measurable objectives," on the completion of which (in theory) your annual bonus depended. My job was a weird mix of cleaning up messes Winston Wolf-style, and repeating routine tasks, neither of which was easily measurable. So I'd have to make up phony quantifiable objectives ("reduce X by 10%, develop at least 5 new Y's") and then waste time accomplishing them, even though they didn't really add value. But I don't think the problem is with "incentive pay" per se. The problems are that good performance is hard to measure (which leads to the fetishization of metrics), and that many (if not most) implementations give the wrong incentives. (Of course, I'm biased, since I'm going to econ school to study, in part, "incentive-compatible mechanism design.")