Controlling monopoly power in a double‐auction market experiment

2004

We present selected results in experimental economics with relevance for market regulation and derive from them concrete insights that could be interesting for regulating authorities. For those readers that are new to experimental economics, the purposes and advantages of economic experiments are discussed and the experimental double-auction market is described in detail to serve as a benchmark example. The experimental results regarding three potential sources of market failure are then outlined: market power, asymmetric information, and externalities. Furthermore, experimental test-bedding, a promising technique for market regulators to examine a new market design ex-ante, is discussed. One important contribution of experimental economics we would like to stress is that the market institution is more important for market outcomes than economic theory has recognized so far.

Science, 1986

A laboratory experimental methodology has been developing in economics in recent years. The nature of the methodology is to integrate clearly motivated but largely subjectively determined human decisions with the organizational features of markets. The article summarizes the nature of the incentive system and how market organization can be used as an independent variable. Initial basic research results that involved the assessment of the ef fects of posted price organization demonstrated that the effect of the institution is to raise prices and lower market efficiency. The existence of such effects and the close proximity of the laboratory posted price institution and rate posting institution required by the government in several industries has led to a series of policy related experiments. The results have also led to more basic research efforts on seemingly unrelated topics.

We study experimental two-sided markets in which the information structure is endogenous. When submitting an offer, a trader decides which other traders will be informed about the offer. This setup allows both a decentralized bargaining market (Chamberlin, J. Polit. Econ. 56 (1948) 95), and a double auction market (Smith J. Polit. Econ. 70 (1962) 111) as special cases. The results show that offers are typically directed to all traders of the other side of the market, but to none of the traders of the same side of the market. Even though traders receive much less information, the resulting market institution leads to the same outcomes in terms of prices and efficiency as a double auction market. In two additional treatments we examine the robustness of these results. First, it is found that the market institution adapts predictably, but not necessarily efficiently, to the imposition of transaction costs. Second, we find that the preference of sellers to conceal offers from competitors is strict. At the same time, sellers benefit collectively when they reveal offers to each other. r

The Economic Journal, 2008

There has been a dramatic increase in the use of experimental methods in the past two decades. An oft-cited reason for this rise in popularity is that experimental methods provide the necessary control to estimate treatment effects in isolation of other confounding factors. We examine the relevance of experimental findings from laboratory settings that abstract from the field context of the task that theory purports to explain. Using common value auction theory as our guide, we identify naturally occurring settings in which one can test the theory. In our treatments the subjects are not picked at random, as in lab experiments with student subjects, but are deliberately identified by their trading roles in the natural field setting. We find that experienced agents bidding in familiar roles do not fall prey to the winner's curse. Yet, when experienced agents are observed bidding in an unfamiliar role, we find that they frequently fall prey to the winner's curse. We conclude that the theory predicts field behavior well when one is able to identify naturally occurring field counterparts to the key theoretical conditions.

The American Economic Review, 1981

Economic Inquiry, 2009

The paper studies bidder behavior in simultaneous, continuous, ascending price auctions. We design and implement special conditions, a type of "collusion incubator" environment, within which tacit collusion develops quickly, naturally and reliably. The collusion incubator environment is designed as a methodological tool that permits observation of phenomenon that has difficulty surviving in other environments, study models of its development, and then study institutional and environmental remedies that would cause it to evolve into competitive behavior. The special, collusion incubator environments are based on a type of public, symmetrically "folded" preferences together with what we call "item-aligned" preferences. The research design called for exploratory, experimental probes of possible institutional or procedural "remedies" that might destroy the tacit collusion and promote competitive behavior should tacit collusion take place. The results are as follow. (1) The collusion incubator environmental conditions do foster tacit collusion. (2) The tacit collusion corresponds to the unique buyer Pareto Equilibrium of a game theoretic model of the auction process. (3) Once tacit collusion developed, it proved remarkably robust to institutional changes that weakened it as an equilibrium of the game theoretic model. (4) The only remedy that was clearly successful was a non-public change in the preference of participants that destroyed the symmetrically, "folded" and "item aligned" patterns of preferences, creating head to head competition between two agents reminiscent of the concept of a "maverick".

Experimental Economics, 2001

This paper is the first experimental study of the effects of competition and adverse selection on the performance of market maker (MM-) markets. Information distribution may is either symmetric or heterogeneous. MM-markets are either monopolistic (the specialist markets), or competitive (the multi MM-market). Welfare comparisons are with respect to a continuous double auction (DA-) market. Informed subjects receive an imperfect signal of the true state of the world. We find three main results. First, competition among market makers significantly reduces the bid-ask spread, and increases transaction volume. Second, competition among market makers induces competitive undercutting, yielding net trading losses for market makers as a group in most periods. Third, from the perspective of uninformed traders, a competing MM-regime is optimal, since it minimizes their expected trading losses.

Previous experimental results are reviewed to address the extent to which oligopolistic equilibria are good predictors of behavior observed in laboratory experiments with human agents. Although the theory is unrealistically demanding with respect to the agents' informational and rational endowments, experimental results obtained in more realistic settings with subjects using trialand-error decision mechanisms tend to confirm predictions of simple symmetric theoretical models. However, in the presence of more complex and/or asymmetric environments, systematic deviations are observed between theoretical predictions and experimental results. Certainly, more effort should be made by all parties involved to build a bridge between theoretical predictions and the behavior observed in the lab in order to assist both the theorist and the policy maker; the former to enrich theoretical models including the behavioral factors identified by experimentalists and the latter to gain insights on the strengths and weaknesses of theory in predicting human behavior in the marketplace.

2004

Prediction markets are increasingly being considered as methods for gathering, summarizing and aggregating diffuse information by governments and businesses alike. Critics worry that these markets are susceptible to price manipulation by agents who wish to distort decision making. We study the effect of manipulators on an experimental market. We find that manipulators are unable to distort price accuracy. Subjects without manipulation incentives compensate for the bias in offers from manipulators by setting a different threshold at which they are willing to accept trades. * The authors thank Manuela Abbate for research assistance and an anonymous referee for helpful comments. We gratefully acknowledge the financial support of the International Foundation for Research in Experimental Economics.