Drive, by Daniel Pink, is a fascinating book about work vs. play, motivation and money. Some experiments have shown that economic incentives ($) can actually lower the long-term performance of so-called "knowledge workers". That is, performing an interesting task (i.e., writing software) can become less interesting to workers if management links performance too heavily to pay. The best explanation in the book for that is that it takes away the worker's sense of autonomy: people want to work, and they want to solve interesting and worthwhile problems.
Well worth reading. The author gave a talk about his book at work, but I wasn't able to listen to the talk, unfortunately.