Knightian Uncertainty

点击翻译按钮获取中文摘要

This project is part of a large body of research on the theory of decision making under uncertainty. Industry and government decision makers benefit from the new and better methods of decision making produced by this line of research. This project develops an alternative formulation of choice under uncertainty based on the distinction by Frank Knight between risk and uncertainty. According to Knight a gamble is risky if the probabilities are known and it is uncertain if the probabilities are not known. A Knightian theory of decision making is developed in this project and used to explain incomplete markets, bargaining, entrepreneurship, business decisions made by routine rules of thumb, and asset prices. The results of this project could lead to a major breakthrough in the theory of decision making under uncertainty. The contribution of this project comes from developing an alternative theory that is more plausible and might receive more empirical support than the Bayesian theory. For example, one of the paradoxes of Bayesian decision theory is that if two individuals attach different subjective probabilities to an observable event, then they both can gain by exchanging a bet on the event at some odds, yet we observe little such betting in reality. The project develops a rigorous explanation based on uncertainty aversion and inertia. This explanation is generalized to explain the absence of some markets for insurance and forward contracts. Even if the investigator fails to develop a completely satisfactory alternative theory of decision making under uncertainty, the project will provide new insights into such important questions as: Why do strikes occur? Why are wages downwardly rigid? Why are contracts so simplified that the parties to the contract forego mutually beneficial gains? Why are security prices so volatile? Why do businesses use apparently inefficient routines such as keeping advertising expenditures a constant proportion of sales?