Optimal Auction Design with Risk Aversion and Correlated Information

In this paper, we develop a general auction model in which buyers and seller are risk averse and private information is multidimensional and correlated, and in this setting we examine the problem of optimal auction design. In particular, we consider the problem faced by someone who has an object to sell but who does not know how much prospective buyers might be willing to pay, and allowing for risk aversion and correlated information on the part of buyers, we demonstrate the existence of an auction procedure that yields the risk averse seller the highest expected utility among all the auction procedures that are rational and Bayesian incentive compatible.