Title: The Value of Data Records (with Simone Galperti & Aleksandr Levkun)

Abstract: Many online platforms intermediate trade between sellers and buyers using data records of the buyers' personal characteristics. How much value do such intermediaries derive from each record? Is this value higher for specific buyers? What are its properties? We find that an important component of the value of a data record is a novel externality that arises when a platform pools records to withhold information from the sellers. Ignoring this externality can significantly bias our understanding of the value of data records. We then characterize a platform's willingness to pay for more data, thereby establishing a series of basic properties of the demand side of data markets. Our analysis combines modern information design with classic duality methods and applies to a large class of principal-agent problems.