Recent articles in the front page of the Washington Post 8/22/2008 and the Business Day section of the New York Times 8/20/2008 both are talking about paying students to perform. So results so far in NY to incentivize performance on AP exams are mixed. I suspect there is not too much surprise there. It looks like the incentives were offered after students had chosen to take the classes and they get paid once. I remember this was an old debate when I was in graduate school between economists and psychologists on this topic. Economists want to design incentive compatible mechanisms that either use selection or shift goals to improve performance. Psychologists wonder how this affects the incentives that are already in play. Well recently there has been a lot of attention on this using contingent management mechanisms to treat addictions. A nice article by Higgins and Petry looks at alcoholism. In these contexts paying people to stay sober and attend meetings seems to work. I thank Warren Bickel for introducing me to this research a few years back. In some ways the DC program looks more like contingent management paying students for attendance as an attempt to reinforce good behavior. This is an important element that has been mostly lost on economists. That frequency of reward matters. They get the contingency part. But more important, once you get kids in school what do you do with them? If they aren't engaged how has this helped much? In fact they may quickly learn in school strategies that are more rewarding than paying attention. I like the quote by Benjamin Franklin that sits in my office. "Tell me and I forget. Teach me and I remember. Involve me and I learn." So this raises a nice question how is involvement reinforcing? So in the end I think both economists and psychologists need to work together on this one. Contingent management, or monetary incentives, can help fix incentives in the short run, but involvement fixes them in the long run.