Self—Knowledge and Self—Regulation: An Economic Approach1

In forecasting how Homo Economicus would evolve into Homo Sapiens, Thaler (2000) predicted that he would gradually " begin losing IQ " , but warned at the same time against simply " making [him] dumber ". He recommended instead that economists pay closer attention to the actual processes of human cognition, which psychologists have shown to be subject to a number of specific imperfections and biases. The research summarized in this chapter can be seen in the light of this overall agenda. Focussing on the links between self—judgement and self—regulation, our approach has been to develop an analytical framework with three main objectives: (i) to unify a number of findings from separate areas of psychology into a parsimonious model of cognition and motivation; (ii) to draw out their main economic implications; (iii) conversely, to derive from the theory potential explanations for the lack of consensus among psychologists on certain empirical and policy—relevant questions, and suggest alternative experimental directions. We introduce three " grains of sand " (or humanity) into the well—oiled mechanics of the ultra— rational economic agent: imperfect self—knowledge, imperfect willpower, and imperfect recall. Incorporating these three ingredients —sometimes even only a subset of them— yields a surprisingly richer account of human behavior than that of traditional Homo Economicus. We are first able to give formal content to individual traits such as self—confidence, intrinsic motivation, dependence or autonomy, and power of will, as well as to cognitive processes such as wishful thinking or selective memory, self—monitoring, and the setting of personal rules (diets, moral precepts, etc.). The resulting framework then allows us to address the following array of questions. Why do individuals value self—confidence, both for themselves and for others towards whom they are not necessarily altruistic? At the same time, why do they sometimes sabotage their own performance , or deprecate their own accomplishments? Is it possible for a rational, Bayesian individual to deceive himself and hold self—serving beliefs? And if so, is such " positive thinking " ultimately beneficial or harmful? How can we simultaneously account for undersaving and miserliness, procrastination and workaholism, overeating and anorexia? Why do extrinsic rewards (incentives) work well in some contexts, but appear counterproductive in others? Why do agents sometimes undermine the ego and self—confidence of others on whose effort and motivation they depend? II.1 Imperfect self—knowledge The agents who usually populate economic models have little doubt about " who they are …