Research
Working Papers
1. Zhewei Song. (2022). “Sellers’ Defensive Behavior in Credence Goods Markets with Uncertain Outcomes – Does a Reputation System and/or a Behavioral Nudge Improve Efficiency?” (Job Market Paper)
In credence goods markets such as healthcare service markets, when a buyer encounters a problem that needs treatment, only sellers have the expertise to determine which type of treatment is sufficient to address the buyer’s problem. Although sufficient treatment maximizes buyers’ expected utility, sometimes it cannot guarantee a 100% success rate. When a treatment failure happens, buyers cannot determine whether it is caused by insufficient treatment or bad luck after sufficient treatment, and they may express their dissatisfaction with sellers in costly ways by engaging in “crying behavior” that results in compensation. To avoid the costly aftermath of “crying behavior”, sellers will “defend” themselves by overtreating to minimize the probability of treatment failure. However, the high cost of overtreatment incurred by both sellers and buyers will result in a Pareto-inefficient outcome, as compared to the situation where buyers do not “cry” and sellers choose sufficient treatment. This study investigates whether the market inefficiency stemming from sellers’ overtreatment and buyers’ crying behavior can be alleviated through a reputation system and/or a behavioral nudge. I show that when there is a reputation system which makes sellers’ treatment history and buyers’ reactions publicly visible, there exists a perfect public equilibrium in which the seller and buyer frequently play the Pareto-efficient strategy profile. I also predict that a behavioral nudge, which makes salient that sufficient treatment and not crying lead to a Pareto-efficient outcome, encourages them to play the Pareto-efficient strategy profile. My laboratory experiment results show that most sellers overtreat while most buyers “cry” when neither the reputation system nor the nudge is present. Buyers are significantly less likely to “cry” when the reputation system is introduced. When both the reputation system and the nudge are present, sellers are significantly more likely to choose sufficient treatment and significantly less likely to overtreat in the late stage of the game, and market efficiency is weakly improved.
2. Zhewei Song & Ulrike Vollstaedt. (2022). “How to Make Better and/or Cheaper Products Accessible to Buyers through an Optimal Product Testing Mechanism”
In markets where reliable information about product qualities is not available to buyers, a product quality testing organization has expertise in finding out and revealing true qualities of products to buyers. However, the quality testing organization often has limited testing capacity, and many existing testing mechanisms are unable to provide quality information of products that are most preferred by buyers. In this study, we design a product testing mechanism which not only makes full use of the limited testing capacity to only provide quality information of the best and cheapest products on the market, but also incentivizes enough sellers to produce products with high qualities and at a price equal to the marginal cost. We show that with our proposed mechanism, all weak perfect Bayesian equilibria maximize consumer surplus, and thus our proposed mechanism (weakly) dominates any alternative testing mechanism. We also consider a generic benchmark mechanism in which the testing organization randomly selects products to test and reveal their qualities. Our experimental results show that the consumer surplus is significantly higher when the testing organization uses our proposed mechanism than when it uses the benchmark mechanism which randomly tests products.
3. Zhewei Song & Erin Krupka. (2022). “When Group- and Self-Esteem Lead To ‘We-Thinking’: When Does Social Identity Motivate Group Behavior?” Under review at Experimental Economics.
In this paper, we take as given that social identity motivates individuals to make group contributions and focus on examining the determinants of when it motivates these contributions. We test whether “we-thinking”, group-regarding behavior in the presence of an individual-group tradeoff, is predicted by a specific relationship between group- and self-esteem. We define group- and self-esteem as having positive feelings about the relative performance of one’s group and self. By extending Akerlof’s model (2016), we predict that engaging in “we-thinking” is positively correlated with group-esteem and negatively with self-esteem. We proxy for group-esteem and self-esteem using rank-based measures and self-reported measures. Using a laboratory experiment, we manipulate subjects’ group-esteem and self-esteem through intergroup and inter-personal competitions. We measure their engagement in “we-thinking” through a modified dictator game in which they allocate tokens to their group at their expense. Using the self-reported measures our predictions are supported: We find that subjects’ self-reported group-esteem (self-reported self-esteem) is significantly positively (negatively) correlated with engagement in “we-thinking”. Our results using the rank-based measure partially support the model’s predictions: Individual rank is significantly negatively correlated with engagement in “we-thinking” when group rank is high. The findings have implications for when individuals are likely to adopt group-regarding behavior and for how to measure the psychological concept of group/self-esteem.