Auctions with Dynamic Populations: Efficiency and Revenue Maximization

We study a setting where objects and privately-informed buyers arrive stochastically to a market. A seller in this setting faces a sequential allocation problem with a dynamic population. We derive both efficient and revenue-maximizing incentive compatible direct mechanisms. Our main result shows that the sequential ascending auction is a simple indirect mechanism that achieves these desirable objectives. We construct equilibria in memoryless strategies where, in every period, bidders reveal all private information. These equilibria are outcome equivalent to the direct mechanisms. In contrast to static settings, sequential second-price auctions cannot yield these outcomes, as they do not reveal sufficient information