My dissertation consists of three chapters in theoretical auction analysis. The first chapter considers optimal sequential auctions with new bidders arriving in each period. The second chapter examines how resale affects bidding strategies and auction outcomes in an auction environment with costly entry. The third chapter investigates how resale affects bidding strategies and auction outcomes in a sequential auction setting where the values of items auctioned in different periods exhibit synergies.
The first chapter gives a full characterization of the optimal sequential second-price (or ascending English) auctions with sequentially arriving bidders. There are n bidders in the first period and m new bidders arrive in the second period. Based on the auctioneer’s commitment power, we study two cases: full commitment and noncommitment. In both cases, we establish the existence of a symmetric equilibrium characterized by a threshold strategy – a bidder does not bid in the first auction when her valuation is below this threshold and bids according to an increasing function otherwise. In the noncommitment case, the auctioneer chooses an optimal reserve price to maximize the expected revenue from the second period; thus her decision of whether to include previous bidders as potential buyers is endogenously determined by the reserve price in the first auction. This might create multiple equilibria depending on the beliefs of the auctioneer and the bidders. We apply a fairly intuitive rule to establish the uniqueness.We also extend our analysis to allow for opportunities for resale, where the winner in the first auction can opt to resell the item to new bidders.
The second chapter, joint with Dan Levin and Lixin Ye, studies how resale affects auctions with costly entry in a model where an arbitrary number of bidders possess two-dimensional private information signals: entry costs and valuations. We establish the existence of symmetric entry equilibrium and identify sufficient conditions under which the equilibrium is unique. Our analysis suggests that the opportunity of resale induces motivation for both speculative entry and bargain hunting abstentions. By following the uniform distribution for numerical analysis, our results suggest that while the entry probability and efficiency are always higher when resale is allowed, the auctioneer’s expected revenue is lower when resale is allowed for almost all parameter values. We also compare this model to one where bidders may follow “strong” or “weak” distributions in terms of valuations.
The third chapter, joint with Dan Levin and Lixin Ye, studies a sequential second-price auction of two objects with two bidders, where the winner of the package obtains a synergy from the second object. If reselling after the two auctions occurs, it proceeds as either monopoly or monopsony take-it-or-leave-it offer. I find that a post-auction resale has a significant impact on bidding strategies in the auctions. When seller makes a take-it-or-leave-it offer in resale, there is no equilibrium where at least one bidder reveals her type with positive probability. When buyer makes the offer instead, there exist symmetric increasing equilibrium strategies for both items. While allowing resale always improves efficiency, I demonstrate that the effect of resale is ambiguous on expected revenue as Ill as the probability of exposure. I also extend this model to allow for three bidders and provide the equilibrium analysis.