Agent-Based Computational Economics (ACE)
Research Area: Market Design

Last Updated: 9 January 2009

Site Maintained By: Leigh Tesfatsion Department of Economics Iowa State University Ames, IA 50011-1070 http://www.econ.iastate.edu/tesfatsi/ tesfatsi AT iastate.edu Online ACE Course (Self-Study eBook) ACE WebSite

Table of Contents: Basic Issues Readings Software, Toolkits, and Demos Resource Sites, Groups, and Individual Researchers

Basic Issues

• To what extent are market outcomes the result of market protocols vs. the result of the rationality and learning capabilities of the market participants?
• Can even zero-intelligent agents do well in strongly structured market contexts (e.g., in double auctions)?
• Market efficiency, market power, security, and reliability concerns

Readings

• Andrew Byde (2002), Applying Evolutionary Game Theory to Auction Mechanism Design (pdf,8pp.), Hewlett-Packard Company, Report HPL-2002-321, November.
  Abstract: "In this paper we describe an evolution-based method for evaluating auction mechanisms, and apply it to a space of mechanisms that including the standard first- and second-price sealed bid auctions. We replicate results known already in the auction theory literature regarding the suitability of different mechanisms for different bidder environments, and extend the literature by establishing the superiority of novel mechanisms over standard mechanisms, for commonly occurring scenarios. Thus this paper simultaneously extends auction theory, and provides a systematic method for further extension."

  Andy Clark (1998), Being There: Putting Brain, Body, and World Together Again, MIT Press, Cambridge, MA, Chapter 9: "Minds and Markets."

  Dan Friedman and John Rust, eds. (1993), The Double Auction Market: Institutions, Theories, and Evidence, Santa Fe Institute Studies in the Sciences of Complexity, Proceedings Volume XIV, New York: Addison-Wesley.

  Steven Gjerstad and John Dickout (1998), "Price Formation in Double Auctions" (pdf preprint,283K), Games and Economic Behavior, Vol. 22, 1-29.

  Abstract: The authors develop a model of information processing and strategy choice for participants in a double auction. Sellers form beliefs that an offer will be accepted by some buyer, and buyers form beliefs that a bid will be accepted. These beliefs are formed on the basis of observed market data, including frequencies of asks, bids, accepted asks, and accepted bids. The traders choose an action that maximizes their own expected surplus. The trading activity resulting from these beliefs and strategies is sufficient to achieve transaction prices at competitive equilibrium and complete market efficiency after several periods of trading.

  D. K. Gode and Shyam Sunder (2004), Double Auction Dynamics: Structural Effects of Non-Binding Price Controls, Journal of Economic Dynamics and Control, Vol. 28, No. 9, July, available from Science Direct.

  Abstract: The authors build a simple dynamic model of a double auction market with "zero intelligence" (ZI) computer traders that accounts for many, though not all, of the discrepancies between human-subject experimental data and theoretical competitive equilibrium (Walrasian tatonnement) predictions. They focus in particular on the effects of non-binding price controls (i.e., price floors below and ceilings above the competitive equilibrium).

  D. K. Gode and Shyam Sunder (1993), "Allocative Efficiency of Markets with Zero Intelligence Traders: Market as a partial substitute for individual rationality" (pdf,1.4MB), Journal of Political Economy, Vol. 101, No. 1, 1993, 119-137. ON-LINEJournal of Political Economy, Vol. 101, pp. 119-137. Published article available at JSTOR.

  Abstract: Gode and Sunder report on continuous double auction experiments with computational traders. They find that high market efficiency is generally obtained even when traders randomly select bids and offers from within their budget sets as long as these "zero intelligence" traders abide by certain protocols restricting the order of executed trades. The authors conclude that the high market efficiency typically observed in continuous double auction experiments with human subjects is due to the structure of the auction and not to learning. Their seminal work has highlighted an important issue now being actively pursued by many other researchers: what are the relative roles of learning and institutional arrangements in the determination of economic, social, and political outcomes?

  Atakelty Hailu and Steven Schilizzi (2003), Learning in a `Basket of Crabs' An Agent-Based Computational Model of Repeated Conservation Auctions (pdf,27pp), Discussion Paper, School of Agriculture and Resource Economics, The University of Western Australia. Accessed on 5/19/03. The final published version of the paper ("Are Auctions More Efficient than Fixed-Price Schemes When Bidders Learn?", Australian Journal Of Management, Vol. 29, No. 2, 2004) is available here.

  Abstract: "Auctions are increasingly being considered as a mechanism for allocating conservation contracts to private landowners. This interest is based on the widely held belief that competitive bidding helps minimize information rents. This study constructs an agent-based model to evaluate the long term performance of conservation auctions under settings where bidders are allowed to learn from previous outcomes. The results clearly indicate that the efficiency benefits of one-shot auctions are quickly eroded under dynamic settings. Furthermore, the auction mechanism is not found to be superior to fixed payment schemes except when the latter involve the use of high prices."

  ADDITIONAL NOTE: This study begins with careful description of the empirical auction problem at hand, and the difficulties encountered in trying to use the existing theoretical auction literature (constrained by analytical tractability concerns) to investigate this problem. It also provides an exceptionally thoughtful discussion of the potential role of human-subject and computational-agent experiments in helping to advance the understanding of real-world auctions.

  Charles A. Holt (1995), Industrial Organization: A Survey of Laboratory Research, in: John H. Kagel and Alvin E. Roth, Handbook of Experimental Economics, Princeton University Press, Princeton, N.J., Chapter 5, pp. 349-443.

  Abstract: This excellent survey by a well known experimental economist provides a careful thoughtful look at human-subject laboratory experiments focusing on market efficiency effects of alternative auction mechanisms. The chapter is organized around four themes: (1) monopoly regulation and potential entry; (2) concentration and market power; (3) conditions that facilitate cooperation; and (4) product differentiation.

  Paul Klemperer (2000), Auction Theory: A Guide to the Literature (pdf,681K). Other auction writings by Klemperer [e.g., What Really Matters in Auction Design (1999)] are are available for downloading at Klemperer's Papers Site.

  Paul Klemperer (2004), Auctions: Theory and Practice (on-line version), Princeton University Press, Princeton, NJ.

  Abstract: "(This book) provides a nontechnical introduction to auction theory, and emphasizes its practical application. Although there are many extremely successful auction markets, there have also been some notable fiascos, and Klemperer provides many examples. He discusses the successes and failures of the one-hundred-billion dollar `third generation' mobile phone license auctions - he, jointly with Ken Bilmore, designed the first of these. Klemperer also demonstrates the surprising power of auction theory to explain seemingly unconnected issues such as the intensity of different forms of industrial competition, the costs of litigation, and even stock trading `frenzies' and financial crashes."

  Deddy P. Koesrindartoto (2004), "Treasury Auctions, Uniform or Discriminatory?: An Agent-Based Approach", Economics Working Paper No. 04013, Department of Economics, Iowa State University, July.

  Abstract: This study develops an agent-based computational economics (ACE) framework to explore experimentally how a Treasury should auction its securities. Buyers are modeled as profit seekers capable of submitting strategic bids via reinforcement learning. Two distinct auction pricing rules are considered, uniform and discriminatory. The author shows that these two rules result in systematically different auction outcomes under different treatment conditions for relative capacity and for price volatility in a secondary security market. In particular, which auction pricing rule generates greater Treasury revenues varies systematically with these treatment factor specifications. These findings help to explain why previous Treasury auction studies attempting to determine "the" best Treasury auction pricing rule have reached contradictory conclusions.

  Robert Marks (2006), "Market Design Using Agent-Based Models", in Leigh Tesfatsion and Kenneth L. Judd (editors), Handbook of Computational Economics, Vol. 2: Agent-Based Computational Economics, Handbooks in Economics Series, North-Holland/Elsevier, Amsterdam, Spring 2006.

  Abstract: This chapter explores the state of the emerging practice of designing markets by the use of agent-based modeling, with special reference to electricity markets and computerized (on-line) markets, perhaps including real-life electronic agents as well as human traders. The chapter first reviews the use of evolutionary and agent-based techniques of analyzing market behaviors and market mechanisms, and economic models of learning, comparing genetic algorithms with reinforcement learning. Ideal design would be direct optimization of an objective function, but in practice the complexity of markets and traders' behavior prevents this, except in special circumstances. Instead, iterative analysis, subject to design criteria trade-offs, using autonomous self-interested agents, mimics the bottom-up evolution of historical market mechanisms by trial and error. The chapter highlights ten papers that exemplify recent progress in agent-based evolutionary analysis and design of markets in silico, using electricity markets and on-line double auctions as illustrations. A monopoly sealed-bid auction is examined in the tenth paper, and a new auction mechanism is evolved and analyzed. The chapter concludes that, as modeling the learning and behavior of traders improves, and as the software and hardware available for modeling and analysis improves, the techniques will provide ever greater insights into improving the designs of existing markets, and facilitating the design of new markets.

  R. Preston McAfee and John McMillan (1987), "Auctions and Bidding", Journal of Economic Literature 25, June, 699-738.

  For a summary and commentary on this article, see the online Notes on First Part of McAfee-McMillan, prepared by Leigh Tesfatsion (October 1999) and Notes on Second Part of McAfee-McMillan (WordDoc), prepared by Alexei Kroujiline (October 1999).

  Kevin McCabe, Steven Rassenti, and Vernon L. Smith (1990), Auction Design for Composite Goods: The Natural Gas Industry, Journal of Economic Behavior and Organization 14, 127-149.

  Paul Milgrom (2004), Putting Auction Theory to Work, Cambridge University Press, Cambridge, UK.

  Abstract: "This book provides a comprehensive introduction to modern auction theory and its important new applications. It is written by a leading economic theorist whose suggestions guided the creation of the new spectrum auction designs. Aimed at graduate students and professionals in economics, the book gives the most up-to-date treatments of both traditional theories of `optimal auctions' and newer theories of multi-unit auctions and package auctions, and shows by example how these theories are used."

  Alvin Roth (2002), "The Economist as Engineer: Game Theory, Experimentation, and Computation as Tools for Design Economics" (pdf,412K), Econometrica, Vol. 70, No. 4 (July), 1341–1378

  Abdolkarim Sadrieh (1998), The Alternating Double Auction Market: A Game Theoretic and Experimental Investigation, Lecture Notes in Economics and Mathematical Systems, Vol. 466, Springer, Berlin.

  Gerald Thompson and Sten Thore (1992), Computational Economics, The Scientific Press, New York.

  Note: Chapters of particular relevance for auction design are Chapter 1: Transportation Problems (9-21) and Chapter 2: Discrete Auctions (23-32).

  Steve Widergren, Junjie Sun, and Leigh Tesfatsion (2006), "Market Design Test Environments" (pdf,136K), Proceedings, IEEE Power Engineering Society General Meeting, Montreal, June.

  Abstract: "Power industry restructuring continues to evolve at multiple levels of system operations. At the bulk electricity level, several organizations charged with regional system operation are implementing versions of a Wholesale Power Market Platform (WPMP) in response to U.S. Federal Energy Regulatory Commission initiatives. Recently the Energy Policy Act of 2005 and several regional initiatives have been pressing the integration of demand response as a resource for system operations. These policy and regulatory pressures are driving the exploration of new market designs at the wholesale and retail levels. The complex interplay among structural conditions, market protocols, and learning behaviors in relation to short-term and longer-term market performance demand a flexible computational environment where designs can be tested and sensitivities to power system and market rule changes can be explored. This paper discusses the use of agent-based computational methods for the study of electricity markets at the wholesale and retail levels, and explores distinctions in problem formulation between these levels."

  Robert Wilson (2002), "Architecture of Power Markets" (pdf preprint,75K), Econometrica, Volume 70, No. 4 (July), 1299-1340.

  Abstract: Liberalization of infrastructure industries presents classic economic issues about how organization and procedure affect market performance. These issues are examined in wholesale power markets. The perspective from game theory complements standard economic theory to examine effects on efficiency and incentives.

Software, Toolkits, and Demos

AMES Market Package (Java)

The AMES Market Package, developed entirely in Java by an interdisciplinary team of researchers at Iowa State University, is an extensible and modular agent-based test bed for studying the performance of wholesale power markets restructured in accordance with a market design recommended in April 2003 by the U.S. Federal Energy Regulatory Commission and now implemented in over 50% of the U.S. The framework models strategic traders interacting over time in an ISO-managed wholesale power market operating over a transmission grid subject to congestion effects. The day-ahead wholesale power market is a double auction. Grid congestion in the day-ahead market is managed by means of locational marginal prices derived from optimal power flow solutions.

The AMES Market Package is a free open-source tool suitable for research, teaching, and training applications. It is designed for the intensive experimental study of small to medium-sized systems. A graphical user interface permits the creation, modification, analysis and storage of scenarios, parameter initialization and editing, specification of behavioral rules (e.g. learning methods) for market participants, and output reports through table and chart displays. AMES is an acronym for Agent-based Modeling of Electricity Systems.

Jasa: Auction Simulator (Java)

JASA is a high-performance auction simulator that allows researchers in agent-based computational economics to run trading simulations using a number of different auction mechanisms. The software is designed to be highly extensible, so that new auction rules can easily be implemented. The software also provides base classes for implementing simple adaptive trading agents. It was developed for research carried out at the Agent Applications,Research and Technology group of Liverpool University.

Annotated list of pointers to ACE/CAS General Software and Toolkits

Annotated list of pointers to ACE/CAS Computational Laboratories and Demonstration Software

Resource Sites, Groups, and Individual Researchers

• ACE/CAS Research Sites and Groups

• Individual Researchers using ACE/CAS Tools

• Agorics, Inc., maintains a site titled Auctions that features auction information (prepared by Kate Reynolds) covering basic definitions, key results, and illustrative examples in a non-technical, highly readable format. The entries all display a 1996 copyright date.

• Steve Gjerstad (Economic Science Laboratory, University of Arizona, Tucson): Design and tests of market performance; Microeconomic models of economic environments; Experimental specifications of exchange mechanisms; Human decision makers and models of human decision processes.

• Paul Klemperer (Nuffield College, Oxford): Industrial economics theory and policy; Competition policy; Microeconomic theory, particularly auction theory.

• Deddy P. Koesrindartoto (School of Business and Management, Bandung Institute of Technology, Indonesia): Industrial organization; Financial economics; Computational economics; Electricity markets; Market design.

• Robert Marks (Australian Graduate School of Management, University of New South Wales, Sydney, Australia): Strategic behavior in markets with small numbers of sellers; Application of economic theory to various social issues (e.g., illicit use of drugs, environmental impacts of energy use); Learning and adaptive behavior in oligopolies.

• Al Roth (Harvard University, Cambridge): Game theory and experimental economics; Market matching mechanisms in theory and practice.

• John Rust (Department of Economics, Yale University, New Haven): Econometric and computational methods for understanding and predicting human decision making over time and under uncertainty; Artificial intelligence and automata trading in double auction markets.

• Leigh Tesfatsion (Department of Economics, Iowa State University, Ames, Iowa): Market power and efficiency in a computational electricity market with double-auction pricing; Role of learning v. market protocols in determining outcomes in double auctions and other types of markets.

• M. Utku Ünver (Department of Economics, Koc University, Istanbul, Turkey): Social learning in market games using genetic algorithms; Experimental economics; Game theory; Two-sided and one-sided matching; Auctions.

• Michael Wellman (Electrical Engineering and Computer Science, University of Michigan): Computational market mechanisms and their applications to various distributed environments.

• Robert Wilson (Graduate School of Business, Stanford University) is an expert in auction design, both theory and practice. For extensive resources related to market design, check out Wilson's Market Design Course. Resources available at this course site include numerous web links on restructuring of electricity industries in the USA, UK, Canada, Mexico, Spain, Australia, Columbia, and general international comparison. He also includes numerous Web links related to auction design more generally.

Copyright © 2009 Leigh Tesfatsion. All Rights Reserved.