Experiments in Economics.

Experimental economics, a relatively new approach, helps economists evaluate the assumptions routinely used to make predictions about the ways markets operate and how individuals respond to financial incentives. By constructing and observing simple experimental situations under controlled conditions, economists can test and measure the impact that different institutional arrangements have on economic performance. Experimental economics has demonstrated that the study of simple structures opens up new ways to improve and refine economists' analyses and predictions.

Hey touts his book as the first textbook designed to provide an introduction to experimental economics, but it really isn't a textbook. It is actually a reference work which covers the research taking place in the field of experimental economics. The prerequisites needed to read this book are an understanding of economic theory, econometrics, and economic methodology. Experiments in Economics would be an appropriate text for the advanced undergraduate taking an elective course in experimental economics, the graduate student adding an experimental component to his or her thesis, or the general economist seeking to learn about the rapidly growing field of experimental economics [p. x]. This book will help fill the gap as the profession waits for the Handbook of Experimental Economics which is to be edited by John Kagel and Al Roth.

Hey is well qualified to write this text since he is co-director of the Centre for Experimental Economics at the University of York (EXEC). It is his intention to provide a practical guide to experimental economics that the reader can use to carry out his/her own experiments [p. 3]. It is also Hey's goal to enable economic experiments to conform to the methodology of experiments in other disciplines. This methodology allows the scientist to carefully test alternative theories, to control for extraneous factors and to isolate the key determinants of scientific process [p. 81].

For those just becoming acquainted with experimental economics, chapter 3 may be one of the most useful chapters in the text. This chapter concentrates on the practical details of conducting both computer based and non-computer based economic experiments. Hey argues that all experiments take on a common framework:

First, we decide what it is a theory of - usually it is a theory of how one or more economic agents choose the value or values of one or more economic variables in the pursuit of some goal. This then defines the decisions that the subjects in the experiment will be asked to take. Second, we decide what the nature of the goal is - usually the maximization of some objective function. This then defines the payment or incentive mechanism - payment for participating in the experiment should be directly related to the achieved value of the objective function [p. 24].

In Parts II (chapters 4-6) and III (chapters 7-9) Hey shows his expertise in experimental economics. These chapters are primarily devoted to reporting the results of Hey's own research and the research of his colleagues at EXEC. The research being conducted at EXEC involves individual decision-making under risk and uncertainty. More specifically, the research has to do with the modern theory of subjective expected utility theory and its relationship to such things as the St. Petersburg paradox and the Allais paradox [p. 71 ]. Hey is careful to provide both the theoretical underpinnings behind the subjective expected utility theory [chapter 4] and the results of experiments testing the theory [chapters 5-7]. In many experiments subjective expected utility models do not provide an adequate prediction of economic behavior [pp. 51 and 110].

The most interesting section of the book is Part IV which deals with interactive economic behavior. In this research area experimental economics is used to develop a theory that explains games and bargaining as seen in the operation of auctions and markets. This type of research is synonymous with the name of Vernon Smith, one of the pioneers of experimental economics in the United States. Much of the work presented by Hey in this area is the work of others since his research program is only beginning to head in this direction [p. 161]. What is most interesting about this section of the book is that it provides a history of how experimental economics has matured over the last decade [p. 178]. This presentation is important because the policy implications of experimental economics are beginning to emerge. As Hey reports:

The results that have emerged are important for a number of reasons: first, they have indicated which theoretical results seem to be robust to experimental investigation and which results are sensitive to changes in institutional detail; second, they have indicated potentially fruitful lines of theoretical inquiry in areas in which existing theory is inadequate or nonexistent; third they have indicated important policy implications which are potentially of considerable value for governmental regulatory practice; fourth, they have confirmed the strength and power of experimental methodology [p. 208].

Experimental economics is a relatively new tool that gives economists the ability to test their assumptions and theories in ways that were not previously available. In the past economists tried to learn how markets operate by observing real world markets, but were hindered by an inability to observe or control the many factors at work. Experimental economists get around this problem by setting up small markets which they are better able to control. The greatest contribution of experimental economics is "the ability to provide reliable data under controlled conditions" [p. 222].

This text presents in summary form the current research strain in experimental economics while the profession awaits the publication of a Journal of Experimental Economics.