On the optimality of diverse expert panels in persuasion games

• 成果类型: Article
• 署名作者: Bhattacharya, Sourav; Goltsman, Maria; Mukherjee, Arijit
• 署名单位: University of London; Royal Holloway University London; Western University (University of Western Ontario); Michigan State University
• 刊物名称: GAMES AND ECONOMIC BEHAVIOR
• ISSN/ISSBN: 0899-8256
• DOI: 10.1016/j.geb.2017.10.028
• 发表日期: 2018
• 页码: 345-363
• 关键词: Persuasion games; Panel design

摘要

We consider a persuasion game where the decision-maker relies on a panel of biased experts. An expert's preference is parameterized by his ideal action, or agenda. Common intuition suggests that more information is revealed if the panel includes experts with opposed agendas, because such experts will undo each other's attempts to conceal unfavorable information. In contrast, we show that recruiting experts with diverse agendas is optimal only if the correlation between the experts' types-i.e., whether they are informed or not is above a threshold. Moreover, if the experts' types are independent, under mild assumptions it is optimal to recruit experts who have extreme but identical agendas. These findings suggest that the diversity of preferences must be considered in conjunction with the diversity of information sources, and it is generally sub-optimal to seek diversity in both dimensions. (C) 2017 Elsevier Inc. All rights reserved.