AUCTIONS WITH RESALE OPPORTUNITIES: AN EXPERIMENTAL STUDY

We study first price asymmetric private value auctions with resale opportunities presented in seller's and buyer's markets. We offer experimental evidence on bidding behavior, prices, and resource allocation. Building upon the Hafalir and Krishna (2008) model, we find that bidders will bid higher in an auction if the resale market is a seller's market than a buyer's market. There is a price/revenue‐efficiency trade‐off established theoretically between these two resale regimes. In equilibrium, however, final efficiency is high irrespective of the resale market structure. Evidence of bid symmetrization and higher final efficiency is found in the buyer‐advantaged resale case. (JEL D44, C92)     

SOME EVIDENCE ON LATE BIDDING IN EBAY AUCTIONS

Bidding in the last seconds or minutes of an auction is a common strategy in Internet auctions with fixed end‐time. This paper examines the three explanations of late bidding in eBay auctions that survived the first scrutiny in Roth and Ockenfels (2002) . There is no indication that late bidding could lead to collusive gains for bidders. Late bidding is a strategic response to the presence of bidders placing multiple bids. Experts protecting their private information are typically the last to bid, while collectors are often the first. As bidders gain familiarity with eBay rules, they tend to bid slightly earlier. (JEL D44)     

ROBUSTNESS OF BIDDER PREFERENCES AMONG AUCTION INSTITUTIONS

In previous work, we found that bidders strongly prefer the ascending to the first‐price sealed‐bid auction on a ceteris paribus basis, but perhaps surprisingly, they are not willing to pay up to an entry price for the ascending auction that would equalize the profits. Risk aversion was proposed as an explanation. In this study, we examine two alternative explanations for the observed behavior: loss aversion and aversion to the dynamic bidding process. We find that neither alternative explanation can account for bidders’ auction choice behavior, leaving risk aversion as the only unfalsified hypothesis. (JEL C91, D44)     

ALL‐PAY AUCTIONS WITH A BUY‐PRICE OPTION

This paper studies all‐pay auctions in which there is a buy‐price option for bidders to guarantee purchases at a seller‐specified price. We analyze symmetric increasing bidding equilibria in the first‐ and second‐price all‐pay auctions with the buy‐price option. While the optimal buy‐price in the second price is higher than are those in the first‐price all‐pay auction, both formats maintain the same expected profit. With an endogenous entry process, all‐pay auctions with the buy‐price can attract more consumers and ultimately reach a higher expected profit than does the uniform posted‐price selling mechanism. (JEL D44, L11, L81)     

SELECTION INTO AUCTIONS FOR RISKY AND AMBIGUOUS PROSPECTS

We study experimentally the selection into first‐price sealed‐bid auctions for a risky or an ambiguous prospect. Most subjects chose to submit a bid for the risky prospect, leading to thinner markets for the ambiguous prospect. Transaction prices for both prospects were equal although subjects expected the ambiguous markets to be smaller. Evidence of a positive correlation between risk and ambiguity aversion suggests that the ambiguous markets were populated by relatively risk tolerant bidders. A control experiment with selection in a simple choice task shows that subjects correctly anticipate the effects of selection on market size and risk attitudes. (JEL C91, D44, D81)     

Bidder Preferences among Auction Institutions

This study examines bidder preferences between alternative auction institutions. We seek to characterize experimentally the degree to which bidders prefer an ascending auction to a sealed bid auction. We find very strong ceteris paribus preferences for the ascending institution with bidders choosing it overwhelmingly often when entry prices for the two auctions are the same. When the entry prices of the two auctions differ, many subjects can be shown to be willing to pay more to enter the ascending auction than is explainable by their risk attitudes when accounting for their expectations about the risk preferences of their opponents. (JEL C91 , D44 )     

A LABORATORY STUDY OF AUCTIONS WITH A BUY PRICE

We present a series of laboratory experiments that examine auctions with buy prices, which are prices that allow a bidder to stop the auction and buy the item immediately. Two types of buy prices are considered, one that is available throughout the auction and one that disappears with an initial bid. Both are evaluated with and without proxy bidding. We find that the use of a buy price increases revenue, early bidding, and auction efficiency. Differences between outcomes in auctions with permanent and temporary buy prices are consistent with the observed choices in auction design made by online auction sites. (JEL C90, D44, C70, L81)     

THE ROLE OF THE BIDDING PROCESS IN PRICE DETERMINATION: JUMP BIDDING IN SEQUENTIAL ENGLISH AUCTIONS

This paper uses data collected from a series of public auctions of used cars in New Jersey to examine how strategic bidding affects price determination in open‐outcry English auctions. “Jumps” in the bidding, which occur when a new offer exceeds the old offer by more than the minimum bid increment, are highly pervasive and consistently related to the item’s presale estimated price. The size of the jumps is not affected by the selling order, however. This jump bidding pattern suggests that open‐outcry auctions are more appropriately interpreted with models that assume common‐item valuations rather than models assuming private valuations. (JEL D44)     

PRIZE‐BASED MECHANISMS FOR FUND‐RAISING: THEORY AND EXPERIMENTS

We study the optimal design of mechanisms for the private provision of public goods in a setting in which donors compete for a prize of commonly known value. We discuss equilibrium bidding in mechanisms that promote both conditional cooperation and competition (i.e., the lottery and the all‐pay auction with the lowest‐bid payment rule) and rank their fund‐raising performance vis‐à‐vis their standard (pay‐your‐own‐bid) counterparts. The theoretically optimal mechanism in this model is the lowest‐price all‐pay auction—an auction in which the highest bidder wins the prize and all bidders pay the lowest bid. The highest amount for the public good is generated in the unique, symmetric, mixed‐strategy equilibrium of this auction. In the laboratory, the theoretically optimal mechanism generates the highest level of donations with three bidders but not with two bidders. (JEL D44, D64)     

DESIGNED TO FAIL: THE MEDICARE AUCTION FOR DURABLE MEDICAL EQUIPMENT

We examine the theoretical properties of the auction for Medicare Durable Medical Equipment. Two unique features of the Medicare auction are (1) winners are paid the median winning bid and (2) bids are nonbinding. We show that median pricing results in allocation inefficiencies as some high‐cost firms potentially displace low‐cost firms as winners. Further, the auction may leave demand unfulfilled as some winners refuse to supply because the price is set below their cost. We also introduce a model of nonbinding bids that establishes the rationality of a lowball bid strategy employed by many bidders in the actual Medicare auctions and recently replicated in Caltech experiments. We contrast the median‐price auction with the standard clearing‐price auction where each firm bids true costs as a dominant strategy, resulting in competitive equilibrium prices and full efficiency. (JEL D44, I11, H57)     

RESERVE PRICE EFFECTS IN AUCTIONS: ESTIMATES FROM MULTIPLE REGRESSION‐DISCONTINUITY DESIGNS

We present evidence from 260,000 online auctions of second‐hand cars to identify the impact of public reserve prices on auction outcomes. We exploit multiple discontinuities in the relationship between reserve prices and vehicle characteristics to present causal regression‐discontinuity estimates of reserve price impacts. We find an increase in reserve price decreases the number of bidders, increases the likelihood the object remains unsold, and increases expected revenue conditional on sale. We then combine these estimates to calibrate the reserve price effect on the auctioneer's ex ante expected revenue. This reveals the auctioneer's reserve price policy to be locally optimal. (JEL D44, L11, L62)     

RISK AVERSION AND OVERBIDDING IN FIRST PRICE SEALED BID AUCTIONS: NEW EXPERIMENTAL EVIDENCE

Bidding above the risk‐neutral Nash equilibrium in first price sealed bid auctions has traditionally been ascribed to risk aversion. Later studies, however, offer other explanations and even argue that risk aversion plays no or a minor role. In a novel experimental design, we directly test the relationship between risk aversion and overbidding by systematically varying the distribution of risk attitudes in auction markets. We find a significant relationship between our measure of risk aversion and overbidding. (JEL D44, C91)     

FIXED REVENUE AUCTIONS: THEORY AND BEHAVIOR

In this paper, we study auctions in which the revenue is fixed but the quantity is determined by the auction mechanism. Specifically, we investigate the theory and behavior of English quantity clock, Dutch quantity clock, last‐quantity sealed bid, and penultimate‐quantity sealed bid auctions. For theoretically equivalent fixed quantity and fixed revenue auctions, we find that fixed revenue auctions are robust to all the previously observed empirical regularities in fixed quantity auctions. (JEL C9, D4, L2)     

REWARDS, EXPERIENCE AND DECISION COSTS IN FIRST PRICE AUCTIONS

Consistent with a model of asymmetric risk aversion, subjects in first price auctions consistently bid above the risk neutral Nash prediction. It has been argued that this is due to low opportunity cost of deviating from the risk neutral bid. We increase opportunity cost (and payoff levels) by factors of 0, 1, 5, 10, and 20, from the normal levels generating payoffs up to $250 for risk neutral subjects and observe an insignificant increase in the slope of individual bid functions. The RMSE of bids declines significantly. This is consistent with a reward/decision cost model of bidding behavior.     

CAPACITY CONSTRAINTS AND INFORMATION REVELATION IN PROCUREMENT AUCTIONS

We utilize laboratory experiments to study behavior in sequential procurement auctions where winning an auction round increases a bidder's future costs. The game admits competitive as well as bid‐rotation style collusive equilibria. We find that (a) bidders show some propensity to account for the opportunity cost of winning an auction, but underestimate its magnitude; (b) revealing all bids (instead of only the winning bid) after each round leads to dramatically higher procurement costs. The rise in procurement costs is accompanied by an increase in very high (extreme) bids, a fraction of which appear to be collusive in nature. (JEL C91, D44, L44)     

THE BEHAVIOR OF INEXPERIENCED BIDDERS IN INTERNET AUCTIONS

In Internet auctions, bidders alter their strategies as they gain market experience. While inexperienced bidders bid the same high amounts regardless of the seller’s reputation, experienced bidders bid substantially less if the seller has yet to establish a reputation and raise their bids as reports are filed that the seller has treated bidders well in the past. Experienced bidders also wait until much closer to the end of the auction to place their bids, although it takes very little experience to learn that waiting to submit one’s bid is a superior strategy. (JEL L14, L15, D83, D12)     

RESERVE PRICE VERSUS ENTRY FEE IN STANDARD AUCTIONS

Analysis of standard auction rules when bidders are risk averse is usually carried out under the assumption that the seller is able to set an optimal reserve. The role of entry fees has been generally overlooked in that analysis. We consider bidders with constant absolute risk aversion and show that reserve price is an essential tool in the second price auction while entry fee is essential in the first price auction. Furthermore, setting a reserve price and entry fee combination optimally may change some of the rankings of the standard auctions that hold under optimal reserves. (JEL D44)     

Synergies in Recurring Procurement Auctions: An Empirical Investigation

This article examines the impact of synergies on bidder behavior in recurring road construction procurement auctions. The study reveals that projects are spatially correlated. When bidders with potential synergies participate, the results indicate that their probability of bidding and winning increases and they bid more aggressively. Finally, the study shows that a firm that is capacity unconstrained will bid more aggressively than one that is capacity constrained.     

PRICE ADJUSTMENT POLICIES AND FIRM SIZE

A number of U.S. State Departments of Transportation have adopted a price adjustment policy designed to limit cost fluctuations of oil‐based inputs in government procurement. Similar policies are common in defense contracting, and have been used to offset financial losses of health insurance companies in Medicare and the Affordable Care Act. We show that while all bidders submit lower bids after the policy is introduced, the extent of bid reduction diminishes with firm size. Small new firms are able to compete more frequently, promoting auction competition and efficiency. (JEL H4, H57, D44)     

ENTRY COORDINATION AND AUCTION DESIGN WITH PRIVATE COSTS OF INFORMATION ACQUISITION

Auction design with endogenous entry is complicated by entry coordination among bidders due to multiple entry equilibria issue. This article studies auction design when information acquisition costs are private information of bidders. We show that this problem can be resolved by sufficient dispersion in these costs. First, we find that a simple second‐price auction with no entry fee and a reserve price equal to the seller's valuation is ex ante efficient, while a revenue‐maximizing auction involves personalized entry fees, which are determined by the hazard rates of their information acquisition cost distribution. Second, we show that sufficient dispersion in the information acquisition costs (more dispersion than a particular uniform distribution by the Bickel‐Lehman dispersive order) can coordinate bidders and implement uniquely the desirable entry. The dispersion in information acquisition costs is also necessary for this “unique implementation” result. (JEL D44, D82)