Observational Learning and Information Disclosure in Search Markets

Abstract

This paper studies the role of observational learning in search markets where buyers do not take the list price as a take-it-or-leave-it offer. Using a unique data from the Beijing housing market, we estimate a structural model in which buyers infer a seller’s reservation value from the home’s list price, time-on-market, and records of in-person home viewings by prior prospective buyers, after which they decide whether to view the home and how much to offer. We use the estimated model to quantify the welfare impact of different information disclosure rules. We find that buyer surplus is reduced, and seller surplus is increased on average if time-on-market information is disclosed. However, disclosing home-viewing information in addition to time-on-market information affects individual homes differently. We find that due to the disclosure of this additional information, buyers are slightly better off while sellers are slightly worse off on average.