Less is more : the search for expert advice
Lee, Su Ann
Nguyen, Thi Minh Hue
Date of Issue2016
School of Humanities and Social Sciences
This paper considers a credence goods market in which each expert carries out a conditionally independent test before providing a recommendation to the customer. For instance, a patient with medical needs can consult multiple doctors, each providing a recommendation after a running a medical test. Based on the recommendations received, the customer decides whether to undergo surgery or not in the end. We analyze the Bayesian Nash equilibrium of this game, and find that the experts distort their recommendation strategy knowing that the customer may seek others’ opinions. Our analysis allows us to identify the relationship between the number of doctors consulted and the patient’s search cost, his prior belief regarding the suitability of the surgery, as well as medical test accuracy. Finally, we compare the customer’s equilibrium payoff with that of an alternative scenario in which there is only one available expert. We find that the patient can be worse off if he has access to multiple opinions. The reason is that the experts’ distortion in recommendation strategy drastically worsens the quality of information transmission, resulting to an increase in the search cost that the customer incurred.
Final Year Project (FYP)
Nanyang Technological University