How to Avoid Intrapersonal Strategic Conflicts in Game Theory

Summary

If the same economic agent has to decide at different stages, (s)he may well suffer from her own behavior. This is demonstrated with the help of durable monopoly markets where — in spite of the monopolistic market structure — one may have fierce price competition. Since game theory is not prepared at all for such intrapersonal decision conflicts, it should transform intrapersonal into interpersonal decision conflicts by relying on local players, e.g. by analyzing games in agent normal form. Experimental results, however, indicate that behaviorally such an extreme position is incorrect: human decision makers often are committed by their previous behavior. The conflict between normative or prescriptive game theory and the behavioral theory of game playing, as part of economic psychology, is further illustrated by outside option games to which forward induction is applicable, principal agent models with incomplete information, and repeated games. Throughout the paper we refer to experimental results to contrast normative game theory with real decision making.

The author would like to thank Hermann Brandstatter, Klaus Ritzberger and Friedrich Schneider for helpful comments.