The preference reversal phenomenon was believed to be inconsistent with the ransitivity axiom of decision theory. However, recent papers have demonstrated that previously observed preference reversals could be explained by subject violations of the independence axiom or the compound lottery axiom. The present paper reports the results of experiments in which a substantial proportion of subject responses violate the asymmetry axiom. These results are inconsistent with expected utility theory and its generalizations.

Distributional implications of current models, 159. — Reformulation, 162. — Shadow pricing with budget constraints, 163. — Conclusion, 164.

I. Specification of an objective function, 497. — II. Ranking a set of feasible programs, 499. — Conclusion, 501.

We report an experiment on effects of varying institutional format and dynamic structure of centipede games and Dutch auctions. Centipede games with a clock format unravel, as predicted by theory but not reported in previous literature on two-player tree-format centipede games. Dutch auctions with a tree format produce bids close to risk neutral Nash equilibrium bids, unlike previous literature on clock-format Dutch auctions. Our data provide a new, expanded set of stylized facts which may provide a foundation for unified modeling of play in a class of games that includes centipede games and Dutch auctions.

This paper develops a nonparametric theory of preferences over one's own and others' monetary payoffs. We introduce ?more altruistic than? (MAT), a partial ordering over such preferences, and interpret it with known parametric models. We also introduce and illustrate ?more generous than? (MGT), a partial ordering over opportunity sets. Several recent studies focus on two-player extensive form games of complete information in which the first mover (FM) chooses a more or less generous opportunity set for the second mover (SM). Here reciprocity can be formalized as the assertion that an MGT choice by the FM will elicit MAT preferences in the SM. A further assertion is that the effect on preferences is stronger for acts of commission by FM than for acts of omission. We state and prove propositions on the observable consequences of these assertions. Finally, empirical support for the propositions is found in existing data from investment and dictator games, the carrot and stick game, and the Stackelberg duopoly game and in new data from Stackelberg mini-games.

The Winner's Curse and Public Information in Common Value Auctions: Comment by James C. Cox, Samuel H. Dinkin and Vernon L. Smith. Published in volume 89, issue 1, pages 319-324 of American Economic Review, March 1999