Timothy C. Salmon

An Evaluation of Econometric Models of Adaptive Learning

This paper evaluates the effectiveness of four econometric approaches intended to identify the learning rules being used by subjects in experiments with normal form games. This is done by simulating experimental data and then estimating the econometric models on the simulated data to determine if they can correctly identify the rule that was used to generate the data. The results show that all of the models examined possess difficulties in accurately distinguishing between the data generating processes. Copyright The Econometric Society.

An experimental test of alternative models of bidding in ascending auctions

The theory and behavior of the clock version of the ascending auction has been well understood for at least 20 years. The more widely used oral outcry version of the ascending auction that allows bidders to submit their own bids has been the subject of some recent controversy mostly in regard to whether or not jump bidding, i.e. bidders submitting bids higher than required by the auctioneer, should be allowed. Isaac, Salmon & Zillante (2005) shows that the standard equilibrium for the clock auction does not apply to the non-clock format and constructs an equilibrium bid function intended to match with field data on ascending auctions. In this study, we will use economic experiments to provide a direct empirical test of that model while simultaneously providing empirical evidence to resolve the policy disputes centered around the place of jump bidding in ascending auctions.

Revenue Equivalence Revisited

The conventional wisdom in the auction design literature is that first price sealed bid auctions tend to make more money while ascending auctions tend to be more efficient. We re-examine these issues in an environment in which bidders are allowed to endogenously choose in which auction format to participate. Our findings are that more bidders choose to enter the ascending auction than the first price sealed bid auction and this extra entry is enough to make up the revenue difference between the formats. Consequently, we find that both formats raise approximately the same amount of revenue. They also generate efficiency levels and bidder earnings that are roughly equivalent across mechanisms though the earnings in the ascending might be slightly higher.

Social Insurance and Income Redistribution in a Laboratory Experiment

Why do some voters support income redistribution while others do not? Public assistance programs have two entangled effects on society: they equalize wealth, but they also cushion people against random catastrophes (like natural disasters). The authors conduct a laboratory experiment to determine how individuals’ responses to the environment are related to their self-expressed political ideology and their self-interest. The findings support the hypothesis that ideology is associated with a person’s willingness to use redistribution to reduce income inequality that is caused by luck, but it is not related to preferences for inequality that are not related to luck.

Maintaining Efficiency While Integrating Entrants From Lower-Performing Groups: An Experimental Study

Efficiently growing a group or firm often requires integration of individuals from lower-performing entities. We explore the effectiveness of two policies intended to facilitate such integration, using a laboratory experiment that models production as a coordination game with Pareto-ranked equilibria. We initially create an efficient group and an inefficient one. We then allow individuals to move into the high-performing group and vary by treatment whether movement is unrestricted, limited to one entrant per period, or subject to an entry exam. We include two additional treatments that combine the two restrictions in different ways to help understand why the institutions are effective in maintaining coordination. We find that both restrictions work to maintain efficient coordination but they are effective for different reasons.

An Investigation of the Average Bid Mechanism for Procurement Auctions

In a procurement context, it can be quite costly for a buyer when the winning seller underestimates the cost of a project and then defaults on the project midway through completion. The average bid auction is one mechanism intended to help address this problem. This format involves awarding the contract to the bidder who has bid closest to the average of the bids submitted. We compare the performance of this mechanism with the standard low price mechanism to determine how successful the average bid format is in preventing bidder losses as well as its impact on the price paid by the buyer. We find the average bid mechanism to be more successful than expected because, surprisingly, bidding behavior remains similar between the average bid and low price auctions. We provide an explanation for the bidding behavior in the average bid auction that is based on subjects having problems processing signals near the extremes of the distribution. This paper was accepted by Teck-Hua Ho, behavioral economics.

Revenue from the Saints, the Showoffs and The Predators: Comparisons of Auctions with Price-preference Values

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.

The Incentive Effects of Inequality: An Experimental Investigation

The effect of inequality on economic growth and efficiency is often debated. Our study investigates a behavioral phenomenon through which inequality might have adverse effects on economic growth. In particular we investigate whether or not individuals exhibit a discouragement effect in the face of inequality that leads to lower work effort. If such an effect exists, it provides a mechanism for converting even idiosyncratic inequality into sustained inequality with adverse consequences for the individuals being affected by the inequality and the economy as a whole. We investigate this phenomenon using an economic experiment to allow us to cleanly vary the nature of inequality and to allow us to directly observe several characteristics of the workers. We find robust support for the existence of a discouragement effect lending credibility to the claims that such an effect would exist in external situations among workers confronted with disadvantageous inequality.

Continuous ascending vs. pooled multiple unit auctions

Abstract We examine a seldom studied multiple unit auction format known as a pooled auction. We use both theory and economic experiments to compare the pooled auction to the continuous simultaneous ascending auction on the grounds of revenue and efficiency generation. The results show that due to very aggressive bidding in the pooled auction, it generates substantially more revenue than the ascending auction while achieving equivalent efficiency levels. We then attempt to explain the overbidding in the pooled auction and find that neither risk nor loss aversion can explain it. We present instead a model with an attentional bias that is related to models of probability underweighting and show that it can explain the data.

The high/low divide: Self-selection by values in auction choice

Most prior theoretical and experimental work involving auction choice has assumed bidders only find out their value after making a choice of which autcion to enter. In this paper we examine whether or not subjects knowing their value prior to making an auction choice impacts their choice decision and/or the outcome of the auctions. The results show a strong impact. Subjects with low values choose the first price sealed bid auction more often while subjects with high values choose the ascending auction more often. The average numbers of bidders in both formats ended up being on average the same, but due to the self-selection bias the ascending auction raised as much revenue on average as the first sealed bid auction. The two formats also generate efficiency levels that are roughly equivalent though the earnings of bidders are higher in the ascending auction.