Friday Meeting - Toshiji Kawagoe (Hakodate Future University)
de 10:00 à 12:00
|Où ?||Salle SCI, ISEM|
|S'adresser à||Agnès Festré|
|Ajouter un événement au calendrier||
Vendredi 4 novembre 2016 – FRIDAY MEETING – ISEM – salle SCI de 10h à 12h
Toshiji KAWAGOE (Hakodate Future University)
Title: Reconsidering Induced Value Theory
Abstract: Recently, an extreme form of random payment system (RPS) appears in the laboratory experiment. When subjects participate in several experimental rounds, only one of the participants is chosen and the outcome at only one of the rounds is paid.
RPS is now widely used in individual decision-making and games with repeated interaction. One of the advantage of RPS is robustness to wealth effect. But RPS has been criticized by Holt (1986) and others that when a subject follow non-expected utility theory, it is not incentive compatible. In infinitely repeated games, discount factor is represented by the random termination (RT) probability. But, in such an environment, even though RT gives the same incentive compatibility constraint, equilibrium path may be different from the original game, if we use RPS. But only RPS should be criticized? According to induced value theory by Vernon Smith, if subjects show other-regarding preference, it may violate dominance precept. That is, presence of other-regarding preference should be considered as the lack of experimental control over subject’s preference. But confirming other-regarding preference in economic experiment is widely accepted project in economic science. What justification is made for it?
Recent studies show that even though all the subjects are paid for the outcomes in all the rounds, such payment system may not be incentive compatible. Then, induced value theory is no longer valid methodology in economic experiment? If so, why we should pay for subjects? We will provide a possible answer for those questions concerning the payment system in experimental economics.