Email: joseph.podwol "at" atr.usdoj.gov
I am an economist with the US Department of Justice Antitrust Division. Outside of my official duties, I conduct independent economic research on auctions and trade on the Internet.
Below are links to some recent working papers. Please contact me for updated drafts. The views expresses and documents attached are my own and do not necessarily reflect those of the US Department of Justice.
"Testing for Nonstandard Behavior in Auctions in the Presence of Unobserved Demand," with Henry Schneider, 2011, under review. (PDF)
Abstract: Empirical work on auctions has found that bidders deviate from rational behavior under standard preferences in important ways. In the current paper, we investigate a range of these behaviors, including nonrational herding, auction fever, quasi-endowment effect, escalation of commitment, and irrational limited attention. Our innovations are to use new data from a field experiment on eBay and to examine the identifying assumptions of tests used in previous work. With these innovations, we now find that there is currently only limited evidence that bidders deviate from standard behavior in the field.
"Sequential English Auctions," 2011, under review. (PDF)
Abstract: Cassady (1967) describes the tactic of "fishing" for an opening bid, whereby an auctioneer calls out lower and lower amounts until an opening bid is eventually placed. The current study incorporates this behavior into an auctioneer's strategy set by considering the sale of an indivisible good via English ascending-price auction when the auctioneer cannot commit to a predetermined sequence of starting prices in advance. A key insight of the paper, which departs from McAfee and Vincent (1997) in a study of the second-price auction analogue of this game, is that the well-known strategy equivalence between the English auction and the second-price auction fails to extend to the sequential setting. This difference has important implications for the optimal starting-price path, giving rise to a Coase conjecture on the initial starting price. The well known revenue equivalence of the two auction formats in static settings is shown to extend to the sequential setting. A bidding outcome resembling "auction fever" as described by Cassady (1967) is explained within the model and serves to further highlight the differences between the two auction formats.
"Informational Shill Bidding," 2010, working paper. (PDF)
Abstract: A number of high-profile cases highlight the importance of shill bidding, the practice of a seller bidding on his own item, in selling fraudulent works of art through online auctions. In such cases, bidding by the seller serves to influence the perception of the item's authenticity by buyers who believe that competing buyers may have superior information. This paper rationalizes the observed behavior in a common-value ascending-price auction model. An equilibrium is established in which fraudulent sellers engage in shill bidding, yet these bids are informative to buyers. Necessary conditions for this equilibrium are that non-fraudulent sellers are uninformed of the item's true quality, so that signalling or other forms of information revelation are not possible; buyers are asymmetrically informed; and fraudulent sellers comprise a sufficiently small proportion of all sellers in the market. An interesting comparative static result shows that increasing the proportion of fraudulent sellers can actually increase total surplus as shill bids lose their power to deceive.
"Explaining Price Dispersion in Online Auctions with Simple Frictions," with Henry Schneider, 2011, working paper.
Abstract: We incorporate insights from the literature on price dispersion in fixed-price settings in an equilibrium model of simultaneous ascending-price auctions, following Peters and Severinov (2006). A fraction of "unaware" consumers are imperfectly informed of the existence of all auctions taking place concurrently. Price dispersion arises as unaware consumers bid above the prices in competing auctions without realizing it. An intuitive prediction of the model is that price dispersion is decreasing in the fraction of bidders who bid in multiple competing auctions. Using an observational data set on the sale of DVDs on eBay, we find results consistent with the model. Further, using variation in the wording of listings among competing auctions as a proxy for search costs, we show that greater variation in listings gives rise to greater variation in prices.