Bo Chen (Shenzhen MSU-BIT University) - Revenue-Maximizing All-Pay Auctions: Disclosure vs Discrimination

We study all-pay auctions with two bidders and one-sided private information. To sharpen the competition, the auction organizer can either design an information disclosure policy through Bayesian persuasion about the bidder with private information, or discriminate against a bidder by systematically discounting his bids. Depending on the bidders' relative strengths, the revenue-maximizing disclosure policies can take the form of partial disclosure, full disclosure, or no disclosure, while the revenue maximizing discrimination either completely or partially level the playing field for the bidders. A comparison of the two instruments shows that information disclosure is more effective in improving the organizer's revenue if significant information asymmetry is present, while discrimination is more effective if the bidders' expected abilities are sufficiently different.