Topic：Convenience vs. Pleasance: Two-Dimensional Matching and The Information Design on Peer-to-Peer Platforms
Speaker：Jinzhao Du，University of Hong Kong
Time：13:30-14:45
 Date：May 6, 2019
Venue：106B, zhonghui Building
Introduction to Speaker：
Jinzhao Du is an Assistant Professor of Marketing at the University of Hong Kong. His research interests lie in the area of multi-sided markets, including media markets and matching markets. His work focuses on understanding the strategic interaction among the multiple players in media markets: content suppliers, advertisers, media platforms, news aggregators, and consumers. He also examines how a matching platform can improve matching efficiency through limiting the cross-side information exposure.
 
Abstract:
We consider a two-sided decentralized matching scenario on a peer-to-peer platform. Each platform user on one side, called a “sender”, sends an offer to one user on the other side, called a “receiver”. A match is successful if the offer is accepted. Individuals on both sides are differentiated in two dimensions:  a vertical attribute, which reflects the individual's quality of being high (H) or low (L), and a horizontal location, which captures the extent of mismatch between a sender and a receiver. A platform user's utility from a successful match increases with the match's vertical quality but decreases with their horizontal distance.  We first derive the equilibrium when the platform fully discloses user information across sides. Interestingly, even when senders' valuation for the horizontal closeness increases, an L-sender may instead choose a distant receiver to avoid competition. We then examine a platform's three possible information designs:  withholding the senders' vertical information, withholding the receivers' vertical information, and centralized matching. Surprisingly, withholding one side’s vertical information not only always hurts the H-type users on both sides, but may also hurt the L-type users on the opposite side.  The centralized matching scheme can benefit both types of users. Lastly, we compare the transaction volume under all matching schemes. Among the decentralized matching schemes, withholding receiver’s vertical information results in the highest number of successful matches. Full information disclosure leads to the lowest number of transactions.