Optimal Information Transmission in Private Value Auctions∗

[Preliminary, do not quote without permission from the authors] We study a situation in which a seller wishes to sell an object to one of N risk-neutral bidders. Before the auction takes place, the seller can provide information to bidders. Formally, bidders receive private random signals related to their valuations. The seller controls the accuracy of bidders ’ private signals, but he cannot observe their realizations. We characterize the auctioneer’s optimal information release and the optimal mechanism in this setting. We show that the optimal selling strategy is to allow to bidders to learn their valuations with lower precision than would be efficient. We also show that in standard auctions the amount of information released by the auctioneer increases with the level of competition in the market (as measured by the number of bidders) and reaches the efficient level with perfect competition. Further, we prove that if we allow the auctioneer to introduce a reserve price, that reserve price will not be independent of the number of bidders: as the number of bidders increases the reserve price will become increasingly restrictive