【5月24日】Does Regret Explain Why People Search

报告题目：Does Regret Explain Why People Search Too Little? A Sequential Search Model with Anticipated Regret and Rejoicing
报 告 人：翁祉泉（西南财经大学经济与管理研究院副教授美国俄亥俄州立大学经济学博士）
时    间：2013年5月24日（周五）下午2:30-4:00
地    点：中山大学岭南堂汪道涵会议室
 
Abstract：
    This paper models how regret and rejoicing arise when consumers (or workers) sequentiallysearch for lower prices (or higher wages). It shows that the anticipation of regret and rejoicingcan explain why in search experiments people are found to “search too little ”as compared to thetheoretical benchmark. Anticipated regret and rejoicing are embedded into optimal sequentialsearch based on the formal regret theory of Loomes&Sugden (1982) and Bell (1982). Becausethe observed price distribution is truncated by the final acceptance price, one can only regret"search too much" but never "search too little". Due to this fundamental asymmetry, the modelpredicts: First, if people feel more sensitive towards regret than towards rejoicing, it is optimalto "search too little". Second, if additional feedback is offered so that people expect to see whatthe price would have been had they continued to search, search behaviors become observationallyindistinguishable from the benchmark case. In addition, ifpeople's sensitivities towards futureregret or rejoicing are strengthened after recently experiencing regret or rejoicing, the model alsoexplains why people occasionally recall past prices whereas standard model predicts no recall.An empirical investigation of 673 separate searches from an experimental dataset confirms that people's (latent) reservation prices do shift with the experiences of regret and rejoicing during search as the model prescribes. Within a single search task, regret about the last search attempt being unsuccessful increases the probability of stopping in the current round from 18% to 31%.Competing explanations, namely, risk aversion and satisficing behaviors, are evaluated and are rejected in favor of the anticipated regret/rejoicing model. One policy implication of the model is that enhancing post-purchase price transparency may induce consumers to search more efficiently inthe first place.
 
 
个人主页：http://riem.swufe.edu.cn/riem/info.asp?id=35