Empirical tests of information aggregation

This paper proposes tests to empirically examine whether auction prices aggregate information away from the limit. These tests are applied to a unique data set containing winning bids and measures of the value and dispersion of information signals in eBay auctions for computers. Results suggest that prices partially aggregate information but do not converge to the common value. Even away from the limit, more bidders and stronger information have a significant effect on the convergence of prices and information aggregation. Even partial information aggregation may represent a potential efficiency gain over one-to-one trade of used goods with uncertain common values.     

PENNIES FROM EBAY: THE DETERMINANTS OF PRICE IN ONLINE AUCTIONS*

This paper presents an exploratory analysis of the determinants of prices in online auctions for collectible United States one‐cent coins at the eBay web site. Starting with an initial data set of 20,000 auctions, we perform regression analysis on a restricted sample of 461 coins for which we obtained estimates of book value. We have three major findings. First, a seller's feedback ratings, reported by other eBay users, have a measurable effect on her auction prices. Negative feedback ratings have a much greater effect than positive feedback ratings do. Second, minimum bids and reserve prices have positive effects on the final auction price. In particular, minimum bids appear only to have a significant effect when they are binding on a single bidder, as predicted by theory. Third, when a seller chooses to have her auction last for a longer period of days, this significantly increases the auction price on average.     

Competition in a Dynamic Auction Market: Identification,Structural Estimation,and Market Efficiency

The model incorporates an infinite series of auctions for identical items (or close substitutes) ordered over time and bidders with unit demand. The participants of each auction are drawn from a dynamic pool, with losing bidders remaining in the pool of potential bidders. The number of bidders in each auction is unobservable. Risk‐neutral, forward‐looking bidders submit bids below valuation. A novel identification and estimation strategy is used to estimate the valuation distribution from an order statistic of the bids. The model is used to evaluate the efficiency of online auctions for a 60GB Apple iPod Video, compared to a perfect‐competition benchmark.     

Mechanism choice and the buy-it-now auction: A structural model of competing buyers and sellers

A striking feature of many online sales platforms is the coexistence of multiple sales mechanisms. Items on eBay, for instance, are frequently offered through auctions, posted prices, and buy-it-now auctions. In this article, I study how this mechanism multiplicity influences the welfare of buyers and sellers. I specify and estimate a structural model of mechanism choice in online markets, in which I consider both sides of the market: On the demand side, buyers' choices among available listings are equilibrium outcomes of an entry game. On the supply side, sellers make equilibrium decisions when choosing sales mechanisms and prices. I estimate this model using data from sales of baseball tickets on eBay and calculate consumer and seller rents in three markets: the actual market with all three sales mechanisms and two counterfactual markets with auctions and fixed prices or only fixed-price listings, respectively. I find that the addition of auctions to fixed-price markets hurts sellers and risk-averse buyers but benefits risk-neutral buyers. Additionally, the consumer surplus increases when buy-it-now auctions are offered but the seller surplus is reduced further. I discuss the intuition for the cause of this result.     

Entry and Bidding in Common and Private Value Auctions with an Unknown Number of Rivals

In many procurement auctions bidders do not know how many rivals they face at the time that they incur the cost of preparing their bids. We show in a theoretical model that regardless of whether the procurement is characterized by private or by common values an increase in the potential number of bidders may lead to higher procurement costs. This raises potential policy questions of whether and how entry should be encouraged or limited in public procurement auctions. We use evidence from auctions of construction contracts to estimate the effect of an increase in the pool of potential bidders on entry and auction prices when entry and bidding decisions are made sequentially with no knowledge of the number or identity of the actual competitors.     

Inferring ascending auction participation from observed bidders

Participation in internet auctions goes well beyond those who place a bid. Participants arrive in random order, and if the auction’s standing price has already exceeded a participant’s valuation, she will not bid. Even so, her unreported valuation is a relevant part of demand for the item, and in an alternate random order, her bid would have been registered. I provide a method to estimate the average number of participants from the average number of bidders per auction. This enables estimation of the distribution of participant valuations from either the distribution of closing prices or the distribution of all observed bids.     

ADVERSE SELECTION IN ELECTRONIC MARKETS: EVIDENCE FROM ONLINE STAMP AUCTIONS

We analyze adverse selection costs in online stamp auctions, based on a comparison of prices on eBay with those of matched stamps at a specialty stamps auction site in the U.S., Michael Rogers, Inc. (MR), which we know a priori has low quality uncertainty. We find that buyer prices are 10–15% lower on eBay as compared to MR, and the price difference, increases with the value of the stamps. Consistent with this adverse selection discount we find that the seller reputation mechanism on eBay has an economically modest, although statistically significant, effect on auction price and probability of sale.     

Common Values and Low Reserve Prices

I show that the benefit of a high reserve price in a common‐values ascending auction is lower than in the observationally equivalent private values setting. Put another way, when bidders have common values, empirical estimation based on a private‐values model will overstate the value of a high reserve price. Via numerical examples, I show this same ranking typically applies to the level of the optimal reserve price as well, and often to the benefit of any reserve price, not just high ones. With common values, the optimal reserve can even be below the seller's valuation, which is impossible with private values.     

Auctioning Horizontally Differentiated Items

This paper analyzes competition between two sellers that offer horizontally differentiated items at competing auctions. Three possible outcomes can arise in equilibrium. For substantially differentiated items the monopoly reserve prices form an equilibrium, and the market is not entirely covered. When products become closer substitutes, a non-empty interval of types becomes valuable to both sellers. When transport costs are low, and virtual valuations are high, sellers compete for these bidders and set their reserve prices below the valuation of the marginal bidder. In equilibrium all types enjoy participation rents. For intermediate levels of transport costs it is not worth competing for runaway bidders. In this case, there is a continuum of equilibria. In each equilibrium sellers adjust their reserve prices so that the marginal bidder gets no rents.     

On the Nature of Self‐Assessed House Prices

In models of optimal household behavior, the value of housing affects consumption, savings and other variables. But homeowners do not know the value of their house for certain until they sell, so while they live in their home they must rely on local house price data to estimate its value. This article uses data from the recent housing boom and bust to demonstrate that changes in households' self‐assessed home values are strongly consistent with the predictions of a model in which households optimally filter available house price data. Specifically, we show that self‐assessed house prices did not increase as rapidly as house price indexes during the boom and did not decline as severely during the bust. A Kalman filter model nearly perfectly replicates these data. These findings have direct implications for economists studying asking prices during booms and busts, optimal default decisions and other key housing‐related phenomena.