Retrospectives: How Economists Came to Accept Expected Utility Theory: The Case of Samuelson and Savage

Abstract

Expected utility theory dominated the economic analysis of individual decision-making under risk from the early 1950s to the 1990. Among the early supporters of the expected utility hypothesis in the von Neumann–Morgenstern version were Milton Friedman and Leonard Jimmie Savage, both based at the University of Chicago, and Jacob Marschak, a leading member of the Cowles Commission for Research in Economics. Paul Samuelson of MIT was initially a severe critic of expected utility theory. Between mid-April and early May 1950, Samuelson composed three papers in which he attacked von Neumann and Morgenstern's axiomatic system. By 1952, however, Samuelson had somewhat unexpectedly become a resolute supporter of the expected utility hypothesis. Why did Samuelson change his mind? Based on the correspondence between Samuelson, Savage, Marschak, and Friedman, this article reconstructs the joint intellectual journey that led Samuelson to accept expected utility theory and Savage to revise his motivations for supporting it.

Effects of taxpayer compliance with taxpayer consciousness as intervening variable

“…Expected Utility Theory (EUT) is a theory of decision making in situations of uncertainty (Harless, 1988;Moscati, 2016). This theory states that when there are several alternative possibilities of a risky event or uncertain outcome, one will make a decision based on the highest expected utility (EU).…”

“…This theory states that when there are several alternative possibilities of a risky event or uncertain outcome, one will make a decision based on the highest expected utility (EU). This EU value is the sum of the weighted utility rates calculated by summing the utility values of each choice after multiplying by probability or probability of occurrence (Mongin, 1997;Moscati, 2016). Social Contract Theory Rooted in the thoughts of Socrates & Plato about the importance of avoiding coercion regarding relations between the state and citizens.…”

Economics versus Psychology. Risk, Uncertainty and the Expected Utility Theory

“…The model of subjective expected utility by Savage (1954) postulates a preference structure that permits (Karni, 2014, p.11): 4 Von Neumann and Morgenstern do not feature an assumption corresponding to what today we call the `Independence Axiom' (Moscati, 2016). 5 Samuelson not only gave a major contribution to re-axiomatizations of the von Neumann and Morgenstern's theoretical approach, but he also became a supporter of the expected utility hypothesis (Moscati, 2016). 6 de Finetti devised the same notion of probability already suggested by Ramsey (1926).…”

Choice-based cardinal utility: a tribute to Patrick Suppes

“…Some claimed that von Neumann and Morgenstern (VNM) had rehabilitated the 'cardinal utility' of the classical and early neoclassicals, others, that this was not the case but they had nonetheless obtained a 'cardinal utility' function of their own, and still others, that neither was the case and that their EU representation was in fact ordinal. Although this would contextualize Suppes' work more fully, it goes beyond the scope of this paper to review the post-war debate on 'the cardinal utility which is ordinal', to use Baumol's (1958) striking words, and we refer the reader to the existing historical work in Fishburn (1989) and Moscati (2013aMoscati ( , 2016a. But we will briefly indicate the two key technical factors of this debate, which, respectively, concern the uniqueness property of the VNM utility function and its ability to induce a measurement of preference differences.…”