Online Auction Demand

Online Auction Demand Online auctions represent a sizable and growing sector of the retail economy. As a result, there has been a proliferation in auctions research in recent years. However, few empirical papers consider the role of the seller in shaping the demand for auctions, a point we seek to redress. Sellers are an important component of the auctions environment because their listing behavior a¤ects buyer behaviors and auction house revenues. Using a unique set of data with all records from buyers and sellers in the Celtic coins category, we estimate a structural model of buyer and seller behavior in auctions. We estimate this model using MCMC techniques which, via data augmentation, enables us to infer latent buyer and lister costs as well as buyer valuations. Results indicate that buyer valuations are a¤ected by item, seller, and auction characteristics; buyer costs are impacted by bidding behavior; and seller costs are a¤ected by item characteristics and the number of listings. Using these results, we conduct a policy experiment to assess the e¤ect of auction house fees on auction house revenues. As a result of the structural model, we are able to compute these fee elasticities even in the absence of any variation in fees over time. We …nd that a decrease in the commissions would lead to higher revenues, while a decrease in per item fees would not (the respective elasticities are 0.11 and -0.33). Commission reductions appear to be more e¤ective because they target high value sellers and can greatly enhance the likelihood they list. By searching across a grid of alternative fee levels, we compute the optimal fee structure and …nd it has the potential to increase auction house revenue by 5% in the category considered.