Published online by Cambridge University Press: 05 January 2013
The problem sets should have been graded. I had no incentive to do them, and as a result did poorly on the exams.–comment from anonymous teacher evaluation, undergraduate game theory course
We investigate the design of incentives for people subject to self-control problems in the form of a time-inconsistent taste for immediate gratification. Because such present-biased people may not behave in their own long-run best interests, there is scope for firms, policymakers, friends and family, and the people themselves to create incentives for “better” behavior. Moreover, whereas for standard agents only ultimate well-being from different courses of action matter, for present-biased people the detailed structure of incentives becomes of critical importance. Optimal incentive design, therefore, will attend to details that the conventional modelwould say are essentially irrelevant. We describe some general principles that have emerged in recent and ongoing research on incentives, highlighting the importance of heterogeneity among agents and providing for flexibility, and illustrate these principles with some simple examples.
Ever since receiving the above comment, one of us (Rabin) has presented it to students on the first day of class as the first exercise for his microeconomics courses: In the above situation, find the incentive! This exercise is a good – albeit somewhat easy – introductory economics question. If the student wanted to do well on the exams, and if doing the problem sets would have helped him to do well on the exams, then he should have had incentive enough to do the problem sets.