The preference reversal phenomenon: Response mode, markets and incentives

摘要: This paper addresses the apparent conflict between results of experiments on individual choice and judgement market experiments. Data are reported for designed to analyze effects (a) economic incentives, repetition, feedback information (b) valuation response modes (c) subjects' decisions in paired nonmarket environments. Causes divergent behavior identified context preference reversal phenomenon (PRP). Study PRP is extended two types The observed first repetition a setting (second price auction) with immediate feedback, both without financial incentives. However, after five repetitions auction, bids generally consistent their choices asymmetry rates predicted unpredicted reversals disappears. An pricing task using BDM mechanism yields similar but which differ from second auction fifth repetition. Choice tasks produce lower than do decision making settings. We grateful support National Science Foundation (grant no. SES-8820552) California Institute Technology. Research facilities were provided by Economic Laboratory at University Arizona. Valuable computer programming was Sean Coates Shawn LaMaster. We thank Professor Jeffrey Dubin his help up software generalized Tobit model, Professors Joyce Berg, Graham Loomes, Charles R. Plott helpful comments.