Hagen (Eds.), The expected utility hypothesis and the Allais paradox

However, many important neo-classical economists employed more sophisticated psychological explanations, including Francis Edgeworth, Vilfredo Pareto and Irving Fisher. Economic psychology emerged in the 20th century in the works of Gabriel Tarde, George Katona and Laszlo Garai. Expected utility and discounted utility models began to gain acceptance, generating testable hypotheses about decision making given uncertainty and intertemporal consumption respectively. Observed and repeatable anomalies eventually challenged those hypotheses, and further steps were taken by the Nobel prizewinner Maurice Allais, for example in setting out the Allais paradox, a decision problem he first presented in 1953 which contradicts the expected utility hypothesis.

(Eds.), The expected utility hypothesis and the Allais paradox…

...nty through revealed preference. As evidence of non-linearity in perceived utility increased (People are found to underweight high probability events when certainty is not guaranteed (Allais Paradox (=-=Allais, 1979-=-) ) and inflate the larger gain when facing alternatives with small probabilities), a large body of literature has been dedicated to apply prospect theory in route choice by investigating the value fu...


Expected Utility Hypotheses and the Allais Paradox

The pattern can actually be explained through models of subadditive discounting which distinguishes the delay and interval of discounting: people are less patient (per-time-unit) over shorter intervals regardless of when they occur. Much of the recent work on intertemporal choice indicates that discounting is a constructed preference. Discounting is influenced greatly by expectations, framing, focus, thought listings, mood, sign, glucose levels, and the scales used to describe what is discounted. Some prominent researchers question whether discounting, the major parameter of intertemporal choice, actually describes what people do when they make choices with future consequences. Considering the variability of discount rates, this may be the case.