On the Empirical Content of Cheap-Talk Signaling: An Application to Bargaining
提出了一个基于排序、激励相容和信念三要素的信号博弈经验分析框架，并应用于在线讨价还价中整数报价的廉价谈话信号研究。
We outline a framework for the empirical analysis of signaling games based on three features: sorting, incentive compatibility, and beliefs. We apply it to document cheap-talk signaling in the use of round-number offers during negotiations. Using millions of online bargaining interactions, we show that items listed at multiples of $100 receive offers that are 8–12 percent lower but are 15–25 percent more likely to sell, demonstrating the trade-off requisite for incentive compatibility. Those same sellers are more likely to accept a similar offer, and buyers are more likely to investigate their listings, consistent with seller sorting and buyer belief updating.