Probability and Uncertainty in Economic Modeling

S ince the early days of probability theory, there has been a distinction between probabilities that are given, as in a game of chance, and probabilities that are not given, but reflect a subjective degree of belief; Hacking (1975) and Shafer (1978) offer historical surveys. In economics, Knight (1921) is typically credited with the distinction between situations of “risk” and of “uncertainty.” In his formulation, “risk” designates situations in which probabilities are known, or knowable in the sense that they can be estimated from past data and calculated using the laws of probability. By contrast, “uncertainty” refers to situations in which probabilities are neither known, nor can they be deduced, calculated, or estimated in an objective way. The Bayesian approach minimizes the importance of this distinction by introducing the notion of “subjective probability.” According to this approach, when objective probabilities are not known, they can be replaced by subjective ones, so that problems of decision under uncertainty are reduced to problems of decision under risk. The standard practice in economics when modeling situations of uncertainty is to follow the Bayesian approach and to assume that people have probabilistic beliefs over any source of uncertainty, that they update these beliefs in accordance with Bayes’s rule, and that they use these probabilistic beliefs in decision making,