Our results have two broad implications. First, unlike the prevailing impression based on the existing literature, the presence of bidder collusion need not mean that the seller can do no better than textbook monopoly pricing. Our seller can do as well as if there is no collusion in a broad set of circumstances, and do generally much better than simple monopoly pricing. Second, an auction rule different from standard may be more desirable when bidder collusion is a serious issue. We have identified an auction with a group-based sale price as being unsusceptible to collusion