Essays on Electricity and Sequential Auctions

57 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2007.In the essay on electricity, we model the restructured wholesale electricity markets as oligopolies facing uncertain demand and "capacity constraints" in which each firm chooses as its strategy a "supply function" introduced by Klemperer and Meyer. To facilitate the computation of optimal strategies, we consider piecewise linear bid (or supply) functions, and we impose restrictions to allow for just a few knots. We implemented simulations to compare the performance of various market structures. The simulations consist of various combinations of firms. The simulation results show that the firms' equilibrium supply functions are steeper with fewer firms, smaller variance of demand, and more severe asymmetry between small firms and a big firm. We also performed comparisons with the Cournot equilibrium assuming storability and Cournot schedule for every demand realization. The results show that Lerner index of supply function equilibrium with capacity constraints is far smaller than that of Cournot equilibrium assuming storability and Cournot schedule. It decreases with the number of firms and it increases with the asymmetry of firm sizes. The essay on sequential auctions studies how to set the optimal starting price in online auctions. The distinctive characteristic of online auctions is that a seller can resell an item that failed to sell at earlier auctions to a new set of buyers. The existing literature on sets of future auctions of unsold items does not consider changes in the set of buyers. As a result, existing explanations for the declining dynamic path of reserve prices cannot be directly applied to online auctions. I consider auctions in which sellers are uncertain of the distributions from which valuations are drawn. Therefore, sellers update beliefs about the distribution itself instead of beliefs about the realizations of valuations through the sequence of auctions. First, considering the listing fee structure explicitly, I find the optimal reserve price for an informed seller who has knowledge about the valuation distribution and show that reserve price does not depend on the seller's salvage value. Then I consider an uninformed seller who is uncertain about the value distribution. In this case, I find, using a numerical example, a way to set a reserve price based on the belief about the distribution and show that the dynamic path of reserve prices for the relisted auctions decline. Additional empirical works support well the theoretic implication providing a reasonable description of seller behavior in real online auction markets.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD