Measuring Performance and Other Illusions

Blogger Joel Spolsky has argued that it is not possible to meaningfully quantify performance in knowledge-based professions because the measuring system can be "gamed" by workers to create the illusion of improved productivity while in reality increasing dysfunctionality. As Spolsky puts it,

Software organizations tend to reward programmers who (a) write lots of code and (b) fix lots of bugs. The best way to get ahead in an organization like this is to check in lots of buggy code and fix it all, rather than taking the extra time to get it right in the first place. When you try to fix this problem by penalizing programmers for creating bugs, you create a perverse incentive for them to hide their bugs or not tell the testers about new code they wrote in hopes that fewer bugs will be found. You can't win.

Similarly, Spolsky observes, rating customer service representatives by the number of calls taken leads to frequently disconnected calls as employees try to maximize the measurement criterion. (This happened to Amazon.com, among others.) Stock options for CEOs lead them to work to inflate the stock price even at the cost of corporate profits (Enron is only the most egregious example of such abusive "gaming.") As Robert Jackall (a professor at Mark's alma mater) described in his Moral Mazes: The World of Corporate Managers, plant managers for major chemical corporations will often "milk" the plant, cutting costs by avoiding needed maintenance, hoping to get promoted before everything collapses.

Is Spolsky right? Is it impossible to measure knowledge-based work? If not, how is it possible?