Search results

We present results of several experiments that deal with endogenous entry in auctions and auction valuation. One observation that is constant across all the experiments we report is that laboratory subjects have a difficult time evaluating potential gains from auctions. Even after they are given some experience with particular auctions, the uncertainty inherent in the auctions (the probability of winning as well as the potential gains from winning) makes it difficult for subjects to compare different auction mechanisms. This highlights the need for new experimental procedures to be used for testing theories that involve endogenous auction entry in the laboratory.

When there are few treated clusters in a pure treatment or difference-in-differences setting, t tests based on a cluster-robust variance estimator can severely over-reject. Although procedures based on the wild cluster bootstrap often work well when the number of treated clusters is not too small, they can either over-reject or under-reject seriously when it is. In a previous paper, we showed that procedures based on randomization inference (RI) can work well in such cases. However, RI can be impractical when the number of possible randomizations is small. We propose a bootstrap-based alternative to RI, which mitigates the discrete nature of RI p values in the few-clusters case. We also compare it to two other procedures. None of them works perfectly when the number of clusters is very small, but they can work surprisingly well.

We compare allocation rules in uniform price divisible-good auctions. Theoretically, a “standard allocation rule (STANDARD)” and a “uniform allocation rule (UNIFORM)” admit different types of low-price equilibria, which are eliminated by a “hybrid allocation rule (HYBRID).” We use a controlled laboratory experiment to compare the empirical performances of these allocation rules.

We conduct three-bidder uniform price divisible-good auctions varying the different allocation rules (standard, uniform, or hybrid) and whether or not explicit communication between bidders is allowed. For the case where explicit communication is allowed we also study six-bidder auctions.

We find that prices are similar across allocation rules. Under all three allocation rules, prices are competitive when bidders cannot explicitly communicate. With explicit communication, prices are collusive, and we observe collusive prices even when collusive agreements are broken. Collusive agreements are particularly fragile when the gain from a unilateral deviation is larger, and an implication of this is that collusive agreements are more robust under STANDARD.

We do not find conclusive evidence of differences in performance among allocation rules. However, there is suggestive evidence that STANDARD may be more vulnerable to collusion.

Divisible-good uniform price auctions are used in financial markets, but it is not possible to use naturally occurring data to test how alternatives to the standard format would perform. Using laboratory methods we provide an initial test of alternative allocation rules.

Objective – This chapter examines the performance of the market to discover efficient equilibrium under alternative auction designs.

Background – Auctions are increasingly being used to allocate emissions allowances (“permits”) for cap and trade and common-pool resource management programs. These auctions create thick markets that can provide important information about changes in current market conditions.

Methodology – This chapter uses experimental methods to examine the extent to which the predicted increase in the Walrasian price due to a shift in willingness to pay (perhaps due to a shift in costs of pollution abatement) is reflected in observed sales prices under alternative auction formats.

Results – Price tracking is comparably good for uniform-price sealed-bid auctions and for multi-round clock auctions, with or without end-of-round information about excess demand. More price inertia is observed for “pay as bid” (discriminatory) auctions, especially for a continuous discriminatory format in which bids could be changed at will, in part because “sniping” in the final moments blocked the full effect of the demand shock.

Conclusion – Uniform-price auctions (clock and sealed-bid uniform-price, and continuous uniform-price) generate changes in purchase prices that are reasonably close to predicted changes. There is some evidence of tacit collusion causing prices to be too low relative to predictions in most cases. The worst price tracking was observed for discriminatory auctions.

Application – Uniform-price auctions appear to perform at least as well as other auction designs with respect to discovery of efficient market prices when there are unexpected and unannounced changes in willingness to pay for permits.

Many jurisdictions fine illegal cartels using penalty guidelines that presume an arbitrary 10% overcharge. This article surveys more than 700 published economic studies and judicial decisions that contain 2,041 quantitative estimates of overcharges of hard-core cartels. The primary findings are: (1) the median average long-run overcharge for all types of cartels over all time periods is 23.0%; (2) the mean average is at least 49%; (3) overcharges reached their zenith in 1891–1945 and have trended downward ever since; (4) 6% of the cartel episodes are zero; (5) median overcharges of international-membership cartels are 38% higher than those of domestic cartels; (6) convicted cartels are on average 19% more effective at raising prices as unpunished cartels; (7) bid-rigging conduct displays 25% lower markups than price-fixing cartels; (8) contemporary cartels targeted by class actions have higher overcharges; and (9) when cartels operate at peak effectiveness, price changes are 60–80% higher than the whole episode. Historical penalty guidelines aimed at optimally deterring cartels are likely to be too low.

Livingston (2002) shows that bidders in Internet auctions are easily convinced of a seller’s trustworthiness: they bid large amounts even if sellers have barely established a reputation for performance, suggesting that they believe that typical sellers usually perform. This study reinforces this conclusion by looking at how bidders choose which auction to bid in when there are several that are selling the same item. The analysis shows that so long as a seller has some history, bidders consider bidding in the seller’s auction. They then choose auctions that offer the best chance to obtain the good at the lowest price.

Supplier selection is a complex and strategic activity needed in every organization, involving many stakeholders and different attributes as price, delivery performance, and product quality. Globalization, in the last decades, increased the competitiveness between vendors, enhancing the use of decision models to support the best choice based on optimizations and bidding variations due to specific needs. This chapter presents three models of multi-dimensional auctions to improve an international humanitarian NGO process procurement efficiency by reducing procurement costs and the decision-making process time. These models have the advantage to be easily implementable in typically complex environments where there is a large number of categories, suppliers, and other features.

The first proposed model uses combinatorial auctions and is suited for procurement, where suppliers can benefit from cost complementarity. The second one uses volume discount auctions and is suited for volumetric purchases, where discounts for large quantities are common. The third one is a multi-attribute model, which computes the best possible solution considering several criteria and can be used in case of complex purchases that involve various categories and trade-offs and are subject to spot prices.

Several design considerations for this type of auctions are reviewed, as well as the mathematical formulation to determine the best alternative (i.e., winner) that can be solved using simple tools like Microsoft Excel. The models are optimized by a mixed-integer programming, and the multi-attribute one is developed using multi-criteria decision analysis (MCDA). All three models developed in this research showed superior results compared to the baseline, being between 9% and 20% more efficient than a regular supplier selection (singly choosing the lowest price) and improving the bidding compliance.

Traditional auction theory assumes that bidders possess values defined solely on the auctioned object. There may, however, be cases in which bidders possess preferences over the revenue achieved by the auctioneer. We present here a comprehensive framework of price-preference valuations, unifying several phenomena ranging from preference for charitable giving to shill bidding. We compare expected efficiency and revenue of first- and second-price auctions for some specific cases of key interest. We also incorporate heterogeneous bidder preferences and examine the effects of mis-specified beliefs and show that both are crucial for understanding these situations.