You are on page 1of 337

Markets, Information and

The Internet bubble, which peaked in size in 2000, has now well and truly
burst. As with all bubbles, there are varying explanations for its occurrence, but
the hype that surrounds the Internet has shouldered a lot of the blame. There is,
however, no doubt that the Internet has significantly changed the way people live,
think and do business.
This impressive volume presents the Austrian school of thought and their
considered response to the “Internet economy” with clarity and insight, and counts
amongst its contributors several brilliant young academics and important figures
such as Peter Boettke, Richard Aréna and the late Don Lavoie (to whose memory
this book is dedicated). Topics include:

• “Austrian” theories of the firm and the Internet economy
• entrepreneurship and e-commerce
• private lawmaking on the Internet
• Hayek and the IT entrepreneurs

Markets, Information and Communication presents an accurate picture of the extent
to which Austrian economics can help us to understand the Internet economy and
the extent to which it vindicates Austrian economics. It will be welcomed by
researchers in economics, Austrian economists and all those wanting a better
understanding of the economic and legal ramifications of the Internet.

Jack Birner is Research Professor at University College Maastricht, The
Netherlands, Professor of Economics at the University of Trento, Italy, and
permanent visiting Professor at BETA, Université Louis Pasteur, Strasburg. He is
author of The Cambridge Controversies in Capital Theory also published by Routledge.
Pierre Garrouste is Professor at the University of Lyon II, France.

Foundations of the market economy
Edited by Mario J. Rizzo, New York University and
Lawrence H. White, University of Missouri at St Louis

A central theme in this series is the importance of understanding and assessing the
market economy from a perspective broader than the static economics of perfect
competition and Pareto optimality. Such a perspective sees markets as causal
processes generated by the preferences, expectations and beliefs of economic agents.
The creative acts of entrepreneurship that uncover new information about
preferences, prices and technology are central to these processes with respect to
their ability to promote the discovery and use of knowledge in society.
The market economy consists of a set of institutions that facilitate voluntary
cooperation and exchange among individuals. These institutions include the legal
and ethical framework as well as more narrowly “economic” patterns of social
interaction. Thus the law, legal institutions and cultural and ethical norms, as well
as ordinary business practices and monetary phenomena, fall within the analytical
domain of the economist.

Other titles in the series

The meaning of market process Expectations and the meaning of
Essays in the development of modern institutions
Austrian Economics Essays in economics by Ludwig
Israel M. Kirzner Lachmann
Edited by Don Lavoie
Prices and knowledge
A market-process perspective Perfect competition and the
Esteban F. Thomas transformation of economics
Frank M. Machovec
Keynes’ general theory of
interest Entrepreneurship and the
A reconsideration market process
Fiona C. Maclachlan An enquiry into the growth of
Laissez-faire banking David Harper
Kevin Dowd

Economics of time and Microfoundations and
ignorance macroeconomics
Gerald O’Driscoll and Mario J. Rizzo An Austrian perspective
Steven Horwitz
Dynamics of the mixed economy
Toward a theory of interventionism Money and the market
Sanford Ikeda Essays on free banking
Kevin Dowd
Neoclassical microeconomic
theory Calculation and coordination
The founding of Austrian vision Essays on socialism and transitional
A.M. Endres political economy
Peter Boettke
The cultural foundations of
economic development Keynes and Hayek
Urban female entrepreneurship in The money economy
Ghana G.R. Steele
Emily Chamlee-Wright
The constitution of markets
Risk and business cycles Essays in political economy
New and old Austrian perspectives Viktor J. Vanberg
Tyler Cowen
Foundations of
Capital in disequilibrium entrepreneurship and economic
The role of capital in a changing development
world David A. Harper
Peter Lewin
Markets, information and
The driving force of the market communication
Essays in Austrian economics Austrian perspectives on the Internet
Israel Kirzner economy
Edited by Jack Birner and Pierre Garrouste
An entrepreneurial theory of the
Frédéric Sautet

Time and money
The macroeconomics of capital
Roger Garrison

Markets, Information
and Communication
Austrian perspectives on the Internet

Edited by Jack Birner and
Pierre Garrouste

Jack Birner and Pierre Garrouste. New York.tandf. published 2004 by Routledge 11 New Fetter Lane. the contributors All rights reserved. mechanical. without permission in writing from the publishers. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www. or other means. including photocopying and recording. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record has been requested ISBN 0-203-18041-0 Master e-book ISBN ISBN 0–415–30893–3 (Print Edition) . now known or hereafter individual chapters. or in any information storage or retrieval system. London EC4P 4EE Simultaneously published in the USA and Canada by Routledge 29 West 35th Street. NY 10001 Routledge is an imprint of the Taylor & Francis Group This edition published in the Taylor and Francis e-Library. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic.” © 2004 Editorial matter and selection.

This volume is dedicated to the memory of Don Lavoie .


AND WILLIAM TULLOH.Contents List of figures and tables xii List of contributors xiii Preface xv Acknowledgments xvi Introduction 1 JACK BIRNER PART I Digitally connected networks 19 1 Subjectivism. WITH COMMENTS BY HOWARD BAETJER. MILLER AND MARC STIEGLER PART II Some history 89 4 High-tech Hayekians 91 DON LAVOIE. AND PIETRO TERNA . MARC STIEGLER. HOWARD BAETJER. entrepreneurship and the convergence of groupware and hypertext 21 DON LAVOIE 2 Open source software and the economics of organization 47 GIAMPAOLO GARZARELLI 3 The digital path: smart contracts and the Third World 63 MARK S.

FOSS 7 The new economy and the Austrian theory of the firm 169 PHILIPPE DULBECCO AND PIERRE GARROUSTE PART IV Networks and communication 187 8 The small world of business relationships 189 GUIDO FIORETTI PART V Markets and market failure 199 9 Some specific Austrian insights on markets and the “new economy” 201 RICHARD ARÉNA AND AGNÈS FESTRÉ 10 Turning lemons into lemonade: entrepreneurial solutions to adverse selection problems in e-commerce 218 MARK STECKBECK AND PETER BOETTKE 11 Big players in the ‘new economy’ 231 ROGER KOPPL AND IVO SARJANOVIC PART VI The monetary sector in the Internet 247 12 Bubble or new era? Monetary aspects of the new economy 249 ANTONY P. MUELLER 13 Possible economic consequences of electronic money 262 JEAN-PIERRE CENTI AND GILBERT BOUGI .x Contents 5 The new economy as a co-ordinating device: some Mengerian foundations 126 ELISE TOSI AND DOMINIQUE TORRE PART III The organization of the firm 141 6 “Austrian” determinants of economic organization in the knowledge economy 143 NICOLAI J.

Contents xi PART VII The legal framework 287 14 The emergence of private lawmaking on the Internet: implications for the economic analysis of law 289 ELISABETH KRECKÉ Index 308 .

2 Rescaled range analysis 243 Tables 2.2 Dunbar’s finding of a correlation between neocortex ratio and mean group size 194 11.1 Features of electronic money.5 Separation of duties 74 3.1 A graph connected according to a small-world topology 191 8.7 Layered games 78 8.2 Low-trust tragedy 68 3.1 Hubs of high trust 66 3.1 Estimated values of the Hurst coefficient 244 12.6 Covered call option 76 3.1 US economic expansion 1991–9 in historical perspective 252 13.3 Top Linux server vendors 48 11.Figures and tables Figures 3. and debit cards 279 .2 Active sites 48 2. checks.4 Simple negotiation game 73 3.3 Remote-trust bootstrapping 71 3.1 Wheat prices from 5 January 1970 to 22 June 2000 240 11.1 Top developers 48 2. currency.

it) Nicolai J. Buchanan Center for Political Economy. . USA (pboettke@gmu. George Mason University. Dipartimento di Teoria Economica e Metodi Quantitativi per le Scelte Giampaolo Garzarelli. Fairleigh Dickinson University.cnrs. Department of Economics and Finance. CNRS. France (gibo005@hotmail. Foss. ATOM. James M. Department of Management. Rome. CERDI. Centre d’Analyse Economiqué. Copenhagen Business School. France (galau@club-internet. Sophia Antipolis.ivs@cbs. USA (koppl@fdu. Italy (fioretti. France (arena@idefi. Denmark (njf. Towson University. Villa Jack Birner. University of Nice.Contributors Richard Aréna.cae@univ. Roger Koppl. Italy (jbirner@ gelso. LATAPSES. Sophia Antipolis. Université d’Aix-Marseilles III. Université d’Aix-Marseilles III. France (festre@idefi.u_3mrs. Paris 1 and University of Lyon II. University of Agnès Festré.dk) Pierre Garrouste. France (secret. Department of Gilbert Howard Baetjer. LATAPSES. Università degli Studi di Roma “La Sapienza”.guido@unimore. CNRS. Centre d’Analyse Economique. Italy ( Peter Boettke. University of Trento. LINK. IODE. Department of Sociology. Jean-Pierre Centi. International Centre for Economic Research. France ( Guido Fioretti. Turin. USA (hbaetjer@ erols. Philosophy and Philippe Dulbecco. Department of Economics. University of Nice.

cnrs. Università di Torino. France ( Marc Stiegler. USA Mark S. France (torre@idefi. former David Don Lavoie. Italy ( . Brazil (Antonypmueller @aol. USA (msteckbe@ gmu. CH-1225 Chene Bourg. Inc. George Mason University.sarjanovic@bluewin. Prato. Avenue des Vergys. Université d’Aix-Marseilles III. Dipartimento di Scienze economiche e finanziarie G. Mueller. Sophia Antipolis. Koch Professor of Mark Steckbeck.terna@unito. and Charles G. CTO and COO (markm@caplet. LATAPSES-IDEFI. University of Nice. LATAPSES-IDEFI. School of Public Policy. Sophia Antony P. EAItech-CERAM. James M. France ( Pietro (bill@erights. George Mason University. Miller. University of Ivo Sarjanovic. Switzerland (i. Erights. Buchanan Center for Political Dominique Torre. ( Elise William Tulloh. 32.cnrs. Department of Economics. University of Caxias do Sul.KRECKE @wanadoo. Combex.xiv Contributors Elisabeth Krecké.

who provided valuable advice for the selection of the conference papers for this volume. president of the Fondazione Dino Terra. We were the guests of the University of Pisa and the Fondazione Dino Terra of the City of Lucca. to both of whom we are very grateful for their financial and physical support. Its title. We should further like to thank Dr Flavia Monceri for managing the daily chores of the organization with flair and good humor. We also want to thank the participants from the United States of America. the main reason for its foundation. A last word of thanks is due to the referees.” indicates that the Association does not limit its activities to historical research but is also forward looking. Pierre Garrouste–Jack Birner Paris–Venice 8 March 2003 .Preface From 24 to 26 May 2001 the Association des Historiens de la Tradition Economique Autrichienne held its third annual conference in the Tuscan cities of Pisa and Lucca. Raimondo Cubeddu deserves our special gratitude for making the generous offer to host the conference in the Association’s first official venture outside France and for actively involving himself in the local organization. and Antonio Masala and Christian Swan for assisting her. Indeed. the United Arab Emirates. “Austrian perspectives on the new economy. Special thanks are due to Professor Raimondo Cubeddu of the Department of Political Science of the University of Pisa and Professor Pierluigi Barotta. South America. The location of the conference illustrates that the Association has international cooperation written large in its charter. South-East Asia and Europe for making the conference a lively global market place of ideas. was the conviction that Austrian economics is a largely unexplored treasure trove of resources that may be used to shed light on current economic phenomena. in 1998.

Howard Baetjer and William Tulloh. . 8.Acknowledgments The editors are grateful to the Mercatus Center (formerly Center for the Study of Market Processes) for permission to reproduce “Prefatory note: the origins of ‘The Agorics Project’” by Don Lavoie and “High-tech Hayekians: some possible research topics in the economics of computation” by Don Lavoie. Vol. pp. Spring 1990. 116–46 of Market Process.

Now that they are available. Researchers who were used to spending many days in dusty library archives now have the possibility to access (to use a neologism that is one of the consequences of the Web) many databases. while at the time of the conference. in May 2001. phenomena which are assuming an increasingly important role in our daily lives. two years later.” because the matter is far from clear. the “new economy” of the newspapers and glossy business magazines.Introduction Jack Birner The first hype of what was commonly referred to as the “new economy” now seems to be over. The matter is not limited to a market environment. expectations ran very high and investors in dotcom companies made loads of money while most of the companies themselves kept accumulating losses. at least for a first screening. This conclusion. however. would overlook the fact that without modern information technology most of these sources would not have been created in the first place. then we would have to conclude that search costs in the past were very much higher. however. scepticism about the new economy was on the increase. information seekers face a problem of selection and its costs. This may be derived from the fact that. the financial sector has now come under suspicion for having generated the exaggerated optimism that characterized the new economy sector. Initially. In the mean time. If we project the great number of sources of information that is now available on the Internet back into a recent past without them. now. The issues addressed by the chapters in this volume are not narrowly linked to the first flirt between the “traditional” economy and the Internet. I say “may. It may also have increased their search costs. In addition. none of the chapters has lost any of its actuality. Even to the uninitiated it is clear that computer-stored databases and the Internet have changed the speed with which information travels. This has increased the opportunities for consumers and producers to acquire information. public opinion seems to have swayed to the other opposite and many commentators and disillusioned investors are now sounding the death bell for the new economy. As is the case with all bubbly booms. an important part of the explanation lies in the disappointment of investors with alternative opportu- nities (which for a short while were disdainfully referred to as belonging to the “old economy”). This may also be taken as one of the indications . This cyclical change of views on the surface hides.

1 In order to take that into account and to dissociate ourselves from the excesses surrounding the concept of the new economy. to talk about the network(ed) society is a very fashionable thing to do nowadays. information and . The Internet just makes people more aware of this. information spreads at very high speed and the availability of many different sources of information makes it easier than before – at least in principle – to obtain accurate information. Their terrae incognitae are also the result of the lack of imagination and knowledge about the history of their own disciplines of social scientists themselves. That only two of the chapters in this volume explicitly go into the network-theoretical background of the Internet economy testifies to the fact that the international research community itself is an unconnected network. Many such links exist between the Internet economy and the Austrian school of economics. it cannot be denied that much of its expansion shows the features of a self-organizing process. the term we adopted for the title of this volume. including the need for protecting privacy. notably of the payments made through the Internet. it is the type of information that is generated by a network characterized by weak links (the seminal text is Mark Granovetter’s “The strength of weak ties. it is not only the lack of communication that is to blame for the large white spots that are left on the map of the social sciences. Do these matters require regulation by a central (government) organization or can private commercial and non- commercial initiatives deal with them? In a world where the next source of information is only a mouse-click away. The combination of fast traveling and undistorted infor- mation is one of the most important predictions of sociological network theory.2 Another question concerns safety.” 1973).3 However.2 Jack Birner that we are only on the verge of the Internet revolution. Even though the spontaneous character of its initial emergence is highly questionable. The fact that not even hypertexts have brought researchers from various disciplines with an interest in the same problems closer together shows that the availability of a weakly linked structure is at the most a necessary condition for efficient communication. But the way in which the expression is used is a misnomer: societies have always been networks. For instance. This raises the closely related and very complex issue of the need for regulation. The Internet is often described as an institution that has spontaneously arisen and evolved. we now prefer to speak of the Internet economy. such as the role of knowledge. One of the goals of this volume is to point to some promising links that have as yet been little explored. The most important recent example is the legal battle between the United States government and Microsoft. intellectual property rights and maintaining an anti-trust market envi- ronment. The lack of communica- tion among economists and between economists and sociologists has long been a source of wonder to me. The potential of sociological network theory for the analysis of organizations and markets has by no means been exhausted and the analysis of the Internet economy is another occasion to put the theory to work (and put it to the test). The Internet is such a network: in the Web’s impersonal environment a relation sealed by a click is one the weakest links one can possibly imagine. Characteristic features of the Internet economy.

A. courageous and creative role he played in continuing to remind economists that the Austrian tradition has much of value to offer. on 6 November. we thought that the Austrian approach would be eminently suitable for understanding the Internet economy. conversely. The editors are proud to include in this volume what is probably Don Lavoie’s last contribution to economics. the network structure of social interaction and the dynamics of competition have been studied by economists and other social theoreticians.4 It has also.) Soon afterwards it turned out that he was suffering from a disease with desperately low prospects of survival and less than six months later. Let me just say that it discusses many lines of research as possibilities for the future that in the intervening years have been taken up and developed – also by some of the contributors to this volume. which in the Internet economy seems to be assuming new features. only 50 years old. Despite his deteriorating health. The article reports on a series of interviews with young computer scientists who were interested in what the article calls the economics of computation. We leave it to the reader to judge to what extent our expectations are borne out by the contributions to this volume. He wanted to combine the conference with a holiday with his family in Florence. Hayek. Just before the conference was due to begin Don let us know that he would not be able to attend because he had to undergo medical tests. the coordination of individual activities and the spontaneous emergence of ordered structures. All of them took their inspiration from the work of F. . they might be used to put them to the test.” which he published together with Howard Baetjer and Bill Tulloh in Market Process in 1990. We also include Don’s prefatory note. One of the first academics to combine several of the conference’s themes in his work is Don Lavoie. But Austrian economics is unique in that it analyzes all these elements as interdependent phenomena. We dedicate this volume to his memory. We owe a debt of gratitude to this great pioneer for the prominent. from its very beginning. Don’s cancellation threw a shadow over the venue. The article is so rich in content that I will not even try to give the briefest of summaries. Introduction 3 communication. as it puts the article in perspective. as the phenomena that are identified with the Internet economy seem to correspond rather closely to the world described by Austrian theories. Don Lavoie’s pioneering role in the field that constitutes the theme of the conference also made us decide to include an article on “High-tech Hayekians. he gracefully accepted to cut and revise his long original paper. (His paper was read by Mark Miller. or the interface between computer science and economics. His death leaves a deep and sad void. Even though few participants knew how serious the situation was. The organizers invited him to give the keynote address and he accepted enthusiastically. Unfortunately. a job which he finished with the help of Bill Tulloh at the end of July. That is why the Association des Historiens de la Tradition Economique Autrichienne (Association of Historians of the Austrian Tradition in Economic Thought) chose this particular theme for its third annual conference. We also thought that. that was not to be. studied the interaction between the monetary or financial and real sectors of the economy. he died. the links between the economy and the legal and political framework.

He found some splendid locations for the various sessions. Between Pisa and Lucca.4 Jack Birner One of the authors and one person who was interviewed contributed a brief retrospective after the 12 years since the article appeared. an expert in this field. between one wedding and the next. It once was the holiday address of members of many European royal families. He let us stay in the beautiful Royal Victoria Hotel. enthusiasm and good taste of Raimondo Cubeddu of the Department of Political Science of the University of Pisa. patent and sell a security- code system that could make the Internet a safer place (they took their inspiration from Hayek and discovered only later the similarity of their design to Granovetter’s network theory and are now gracefully referring to their network pictures as Granovetter diagrams). For example. Most participants to the conference draw an income from teaching and writing about the ideas of the Austrian school in the relatively protected environment of the university. Stefano. CESSBA. who gracefully offered to take care of the practical and a large part of the financial aspects of the conference. We have high hopes that the conference has contributed to a cross-fertilization between these two (very loosely defined) groups. the most impressive of which was the Palazzo del Consiglio dei Dodici dell’Ordine di S. Since one of the offshoots of the “economics of computation” is the growing body of computer simulations of economic processes (shortly before this volume went to the press. and particularly from both sides of the Atlantic. for a comment. Another interviewee. Professor Pierluigi Barotta. we also asked Pietro Terna.6 A major cause of its success was the skill. presented a paper at the conference which he wrote together with Marc Stiegler. academics and people working in the Internet economy. which offers a unique cultural and artistic environment. stimulated many lively discussions and was an important factor in making the conference a success. . Raimondo hosted us in Pisa. diplomacy. On the last day we were taken by bus to the nearby wonderfully preserved fortified city of Lucca (the birthplace of Giacomo Puccini) as guests of the Fondazione Dino Terra del Comune di Lucca and its president. They have patented and sold the results in the rough (even rougher?) environment of the market. on the other hand. in the spectacular Sala degli Specchi of the town hall. Palazzo Orsetti. the entire setting was highly stimulating and conducive to informal get- togethers between researchers in Austrian economics from all over the world. Eric Drexler and Mark Miller made a highly original and creative use of network theory. The existence of these “high-tech Hayekians” prove the actuality of the conference theme. who preceded us in enjoying the view of the Arno river from its Jugendstil-decorated apartments. using it to design. He also took us to various carefully chosen restaurants with such regularity that more than one participant regretted having left his jogging shoes at home. The presence of both types of participants. opened in 1839 and still owned and managed by the same family. the Italian Center for Agent Based Social Simulations was founded5). who are both represented in this volume. Mark Miller. earn their living by creatively applying these and similar ideas. The sessions took place. The people interviewed in the article.

they increase their satisfaction level. In the process of coordinating plans. In this way. we arrive at an image of entrepreneurship as furthering the mutual orientation of individuals’ expectations of how other individuals perceive the world. Don Lavoie’s “Subjectivism. will facilitate the process of communication in which wealth is created: “Entrepreneurship . Lavoie) through a detailed organization theoretical analysis of open source software (Garzarelli) to a concrete proposal for applying smart contracts to unblock potential wealth in third-world countries (Miller and Stiegler). This intentional aspect of subjectivism is particularly present in the work of Ludwig Lachmann. perceptions of other individuals and expectations about their future activities. and of Israel Kirzner. If they succeed in realizing them. it is too early to assess the effects that computers have on the economy.8 The two types of software discussed in this chapter enable us to transcend the limits of traditional texts written on paper by combining the advantages of keeping the original text while adding annotations: reading and writing will no longer be separated. individuals discover new opportunities. and in this chapter he examines what impact they may have on economic growth. Their concept of entre- preneurship is thus more encompassing than the traditional one since everybody who does so is an entrepreneur. In the tradition of Lachmann. Lavoie further concentrates on the consequences for markets of the introduction of writing on paper as a preliminary to a tentative analysis of the effects of hyper- texts and groupware on economic growth.” This certainly captures its creative aspects. entrepreneurship and the convergence of groupware and hypertext” highlights the role of entrepreneurial imagination in the creation of wealth. All three discuss aspects of computer software engi- neering and its potential consequences. which is a major source of inspiration for Lavoie. The picture that emerges from these elements is that economic growth is the creation of value through a learning process that involves a better coordination of individual plans. according to Lavoie. By combining coordination and subjectivism. Activities are guided by perceptions and expectations. Lavoie observes that the hypertext software and groupware both serve to coordinate individual plans. An important tenet of Austrian economics is that economics in general and the market in particular is about the coordination of individual activities. and more in particular. more integrated networks of meanings are created which. what entrepreneurs do is to coordinate individual plans. Introduction 5 About this volume Digitally connected networks The first three contributions land us right in the middle of the Internet economy and its “Austrian” features. Not surprisingly to the readers of Lavoie’s work – which has a hermeneutical orientation – the role of written communication and its meaning is at the center of his analysis. a thesis known as subjectivism.7 In this perspective the entrepreneurial function of seeing new opportunities becomes a form of “reading of a meaningful situation in a language-constituted world. but they range from the highly speculative and philosophical (the creation of value.

This allows them both to benefit from the much larger part of the knowledge of other individuals that is hidden from them and to adjust and improve their own knowledge (that is. This. But contrary to what the theory of clubs implies. is a variation on Mandeville’s theme – gladly adopted later by Austrian economists – of private vices and public benefits.” The objective of Giampaolo Garzarelli’s chapter on “Open source software and the economics of organization” is very similar to that of the previous one: to examine some consequences of information technology for the economy. This would be the case if generally enforceable property rights could be established and transferred. to learn). He uses the work of Deborah Savage to argue that professions as characterized there are very similar to the spontaneous and decentralized way in which open-source software organizes itself. The “Austrian” aspect is not difficult to identify: it is the idea that a spontaneous self-organizing process is more efficient than a centrally planned structure.9 What further distinguishes the OSS story from the more traditional economic approaches is that what is maximized is not profit but reputation. which is about the supply and demand of shared goods. The modularity and its dual. This. Garzarelli takes the reader through the following variations. published in 1989.6 Jack Birner is fundamentally dependent on the media in which human communication takes place.11 That the direction is “Austrian” is not only argued explicitly. however. in the interesting cases such as OSS there are increasing returns to their increased use: the more diffused a particular type of OSS becomes. constitute exact analogues to the way in which Hayek describes the virtues of the market and competition. covers only the supply side. Miller and Stiegler propose an electronic solution to this problem. In an environment where everybody is motivated by trying to live up to or do better than the average standard. On the contrary.10 This chapter critically and creatively examines a number of existing economic theories for their applicability to OSS without delivering a finished analytical framework. He sees in open source software (or OSS. the shared good that software is does not necessarily cause congestion. Individuals monitor themselves.” that characterize software in general. of course. For the market and the price system to act as a “system of telecommunications” it is necessary that pairs of individuals have (even a very small) part of their knowledge or perceptual fields in common. The title of Mark Miller’s and Marc Stiegler’s chapter. there is no monitoring problem in the common sense of the word. the more positive network effects it will have. argues that the poor of the world would not be so poor if only they could legalize the assets over which they de facto but not de iure dispose. It also shows unintentionally where the author observes that OSS has its origins in the activities of hackers. In order to incorporate demand.12 The Other Path. Garzarelli’s main argument may be formulated as follows. “The digital path” is a deliberate variation on the title of the book with which the Peruvian economist Hernando de Soto is finally drawing attention. “information hiding. But in the course of his discussion the author is generous with indications of where to look. Garzarelli uses the theory of clubs. the most well-known example is probably Linux) a new model for economic organization that needs a new analytical tool kit. They want to avoid .

which seems like a good idea. Smart contracts are software programs that embody cooperative arrangements without requiring coercive recourse (the general legal features of this type of arrangement are discussed by Elisabeth Krecké in the last chapter). There may be still a problem. In the latter. For this to be possible. Universally enforceable contracts are necessarily formulated in very general terms. is a counterexample to this thesis. which range from banks and credit rating institutions to courts of justice. For example. According to Hayek. They may enable local communities to take advantage of existing institutions that are specialized in producing trust in more advanced countries. This is why the second instrument is introduced: video contracts. instead of dealing directly with each other. Hayek 1979: 173). Apparently. have come within reach with the widespread use of the Internet and the availability of relatively low-cost computers. One consists of so-called smart contracts. In network terms. The economic success of the “first world” can be explained by the existence of many of these trust hubs. to dodge government interference) and the creation of disconnections in networks between agents.13 Miller and Stiegler’s proposed solution relies on two instruments that. notably rating agencies. the problem described by de Soto is how to turn low-trust networks into high-trust ones. this modern global market system has evolved together with the social and legal framework that makes it work out of the pursuit of the interests of individuals themselves and of their close relatives (compare for instance. the parties to an arrangement use an escrow until both parties have fulfilled their obliga- tions. it is necessary to have a structure where local networks (such as villages) are connected by hubs that are part of a larger network (which is also the structure of the Internet). Notice that there is a strong parallel between De Soto’s and Miller and Stiegler’s analysis and the phenomenon of Italian industrial districts. particular conditions in social networks and the history of their development may obstruct a spontaneous process that Hayek says will lead towards a fully integrated market structure. Instead of writing down the terms of an agreement or the conditions of a particular property right. This risks destroying the basis of the potential wealth of the poor in the Third World: their detailed knowledge of very specific local conditions. though. These social-network aspects are not explicitly discussed by the authors but their analysis lends itself perfectly to this approach. reliable relationships are more easily formed. according to them. The chapter gives some examples of more complicated arrangements of this kind. Their discussion of contracts as games is very interesting. the lack of more numerous and longer distance connections (with weaker ties) prevents these local relationships to be turned into more stable enforceable rights. particularly in countries where the checks and balances of a developed democracy are not available (and these are not only countries of the so-called Third World!). however. However. Many of these owe their success to their location at the intersection of local networks with strong ties and the global weak- tie network that constitutes the world market. Introduction 7 involving governments. These programmes may rely on encryption techniques (for instance. . The phenomenon that De Soto describes. They are the nodes that connect local and otherwise unconnected networks in which trust relations (strong ties) may already exist.

This is also confirmed by the brief retrospectives by one interviewer and one interviewee. according to Miller and Stiegler.” Taking as their point of departure Menger’s discussion of the particular type of goods that we would now call trust-related. and by Pietro Terna. The final paragraphs of the chapter sound not so much an Austrian as an optimistic classical liberal note: smart contracts with their reliance on software- based rules will bring the ideal of “the rule of law but not of men” closer than ever before. however. an article originally published in Market Process in 1990. Its cost structure and the externalities it creates. According to the authors. the authors observe that the Internet performs the same function as the Mengerian entrepreneurs: it bridges gaps in the commu- nication network with the consequence that information is more widely available.8 Jack Birner parties may digitally record the process of negotiation. the solution of conflicts. His main thesis is that where diverse and specialized and hence more . however. It summarizes a series of conversations with young entrepreneurs in the information technology sector. Some history The chapters in this section both deal with the more recent history of the Internet economy and some more remote connections. lead to non-efficient price setting for its services and to an underutilization of its potential. I for one wonder how feasible their solution would be in a country like communist China. Their general message. The organization of the firm Nicolai Foss in his “ ‘Austrian’ determinants of economic organization in the knowledge economy” addresses an argument that is not only related to the Internet economy. The article makes for fascinating reading and many of the features that are there discussed as possibilities have in the mean time been introduced. More backward-looking is Elise Tosi’s and Dominique Torre’s “The new economy as a co-ordinating device: some Mengerian foundations. the Internet is similar to other information carriers in that it has increasing returns. To cure this particular type of coordination failure the authors recommend active outside intervention (a conclusion similar to the contribution by Centi and Bougi but quite different from the chapters by Garzarelli. In case of later disagreement. The recent history is represented in the form of a reprint of “High-tech Hayekians”. The authors admit that their proposal is a venture into largely uncharted territory and that a piecemeal introduction should help to assess its feasibility. an effect which is strengthened by its high degree of connectivity. What they all have in common is that they somehow or other were inspired by Hayek’s ideas or ideas very similar to Hayek’s. is a variation on the theme that links all three contributions in this section: trust hubs serve as the link between local strong-tie networks with global weak-tie ones. Miller and Stiegler. and Krecké). where the government effectively censures the Internet. this would facilitate.

Introduction 9 dispersed knowledge accounts for a substantial part of the value added in production (a situation that he calls a Hayekian setting). With the faster diffusion of information in more highly connected networks between enterprises. but they do so by discussing its impact on the type of a firm’s economic activity. This is because ownership and delegated rights respond to different incentives. To put this differently: their capital consists more than ever in a stock of knowledge that requires continuous investment in learning to keep it from becoming obsolete. but it would be a mistake to think that one can arbitrarily mix a central organization form with a market-type coordination. In a world where information becomes more readily available. the entrepreneur’s authority is weakened. The authors argue that information and communication technologies and organization forms coevolve. that it will. knowledge-intensive production processes rely on complementary knowledge and information. however. investing in firm-specific capabilities and their maintenance becomes more important as the basis for making a profit. The blurring of these boundaries poses a problem for the organization of the firm. with the increased interest in the Internet economy. too. begin to be recognized more generally. In their “The new economy and the Austrian theory of the firm” Philippe Dulbecco and Pierre Garrouste. which says that in order to decide rationally whether an unknown bit of information is worth its cost of acquisition one woud have to know its value in advance. Here lies a vast field of research that . This is because the workers whose activities he is supposed to coordinate are the owners of specialized knowledge that can only be kept up to date by continued interaction with other specialists. the more the firm becomes an information processor. less specific. analyze the consequences of the Internet Economy for the organization of the firm. This has consequences for its organization. a point also made by Foss. Let me add that we have here a knowledge problem twice over: not only do entrepreneurs have to decide on the correct type of knowledge and its internal organization as capital. perhaps. firms are forced to adapt by innovating continuously. The more a firm’s assets are intangible and knowledge based. Foss illustrates this in a case study of Oticon. which had to learn this lesson the hard way. less ambiguous and non-rival. as Mises argued. There is a close link with the paradox of information. The more prominent role of complementarities requires new organization forms. this is impossible by the very definition of what constitutes discovery. As Popper has observed. This is easier said than done. Where these are located is a problem that was addressed by Coase. they also have to foresee what new external knowledge will be discovered in the future. This implies the existence of influential networks that are not contained within or do not coincide with the boundaries of the firm. The point about the importance of the mix between competition and complementarity was made as early as 1960 by George Richardson in his Information and Investment (quoted by the authors) but it is only now. The decision to invest in the wrong type of knowledge is usually difficult to reverse and taking the right decisions has become more difficult in the fast-changing environment of the Internet economy (its brief history is littered with corpses). Whereas competition continues to play an important role (think only of the battles for making certain software or telecom- munications standards generally accepted).

There is a parallel with biological evolution. This is the feature on which Miller and Stiegler base their proposal.14 What they implicitly assume instead is that every economic agent can interact with any other agent. Fioretti sketches a research agenda to put his hypothesis to the test. often at great distances. He ends by hinting at the possibility that our cognitive constraints put an upper limit to the number of technologies that may be combined. What Miller and Stiegler call hubs may link many short-range networks that are thus integrated into what Hayek in 1945 called a system of telecommunications. To many economists the most surprising one will be Keynesian underemployment equilibrium: it is the consequence of a loop in the information network that causes economic agents to keep repeating the same behavior. That we nevertheless succeed in communicating and interacting with many others. but Granovetter and his followers have convincingly shown that all human relationships may be described in these terms. As economists have begun to recognize. That this link has not been noticed before is due to an almost general blindness of economists to the fact that in order to model interactive behavior one must specify the interaction structure. Networks and communication What explicit attention to networks may contribute to the progress of economic analysis is shown in Guido Fioretti’s brief but very interesting “The small world of business relationships. the human brain has only a limited capacity of handling complexity.10 Jack Birner remains largely unexplored. Information and communication technologies make distance.” As far as the Internet economy is concerned. Does this mean that Fioretti thinks the Internet has not changed anything? No! It has liberated us from the interaction constraints imposed by our physical location. To return to Dulbecco and Garrouste. they argue that one such way is vertical integration and they refer to the increasing number of mergers as a confirmation that their first steps towards an Austrian theory of the industry and of the firm lead in the right direction. its main hypothesis goes against the grain of the majority of the other contributions. less relevant. One of the implications is that we can handle only a limited number of social connections. or characteristic path length. we may expect there to be a structural invariant across the boundary of the Internet economy. As the author observes. which in its turn regulates the number of links that may evolve between firms. one of the ways of coming to grips with this set of problems is to elaborate the idea of Richardson that firms actively try to control their environment in order to keep problems of knowledge and information within manageable bounds. is due to the structure of the networks of which we are part. after Herbert Simon drew their attention to the boundaries of rationality. which together with the degree of clustering is the crucial parameter of networks. The worst manifestation of this neglect are analyses based on representative individuals. According to Fioretti. this assumption implies the absence of institutions. Fioretti gives a number of examples of what disconnections in such networks may lead to. Hayek used this expression to characterize the price system. where the positive or negative contribution of a particular .

. The theoretician that is discussed first is Hayek. The authors’ criticism of CEE takes two forms. easy access to the Internet has lowered the cost of information and almost eliminated information asymmetries. The idea of an optimal level of interaction between genes merits a critical elaboration in the direction of economic systems. particularly in the context of the fast-changing Internet economy. have in common is that they presuppose that agents operate in a social interaction structure. What is also not given are the definitions of goods and markets. For an analysis of the variety of markets and a more satisfactory treatment of what markets do Aréna and Festré go back to Carl Menger. The second criticism is more empirically orientated and uses several types of actual Internet market relationships. This is perhaps the most important difference with CEE. Hayek’s vision of the functioning of the market economy as a dynamic process of discovery seems to suit the Internet economy better. and the general use of auctions has (re-?)introduced tâtonnnement as a process of coordination. preferences and techniques have become more easily identifiable and may thus be considered to be given. Introduction 11 gene to the overall fitness of the organism is dependent on the number of other genes it affects. What the authors criticize. Friedrich von Wieser elaborated this into an analysis of the relationships between markets and property rights and different types of contracts. so this view goes. where coordination by the state partly replaced coordination by the market. What all these authors. With the Internet economy. is Hayek’s neglect of the existence of different types of markets and his recourse to the “empirical fact” that market economies show a tendency towards equilibrium. That paradigm has been criticized for its lack of applicability to the economies of the twentieth century. His emphasis on the heterogeneity of agents and the importance of their perceptions and expectations in the coordination process that also takes place though the Internet is very different from CEE’s emphasis on factors that are “given” right from the beginning. Ludwig Lachmann provided a further generalization in his analysis of the general institutional embeddedness for the variety of market forms. Menger bases his analysis on the idea that goods may have various degrees of exchangeability and he recognizes that markets are social institutions whose differences derive from differences in social rules and habits. Markets and market failure In “Some specific Austrian insights on markets and the ‘new economy. however.’” Richard Aréna and Agnès Festré criticize the idea (which they call competitive economic equilibrium or CEE) that the Internet economy signals a return to the type of free market economy that prevailed in the West in the nineteenth and early twentieth century. the context in which the foundations of neoclassical economic theory were laid. the real world has come to look more like the economy of the neoclassical microeconomics textbooks: initial endowments. one which has been addressed earlier in this Introduction. location in physical space has become well-nigh irrelevant. The first is theoretical and based on the analysis of some prominent Austrian economists. including Hayek.

they do not depend on profits or losses. by Elisabeth Krecké. they must not act upon it – unless they recognize the merits of the free-market economy. The empirical premise is that commercial relationships through the Internet are rife with such asymmetries. by Roger Koppl and Ivo Sarjanovic. “Big players in the ‘new economy.12 Jack Birner The authors see in their description of the several types of Internet markets a confirmation that the line of analysis that runs from Menger to Wieser to Lachmann is to be preferred over CEE and also over Hayek. But a lot of theoret- ical work remains to be done. Aréna and Festré are trying to construct an Austrian theory of the organization of markets that is also applicable to the Internet economy.18 Because of their power and unpredictability.16 They are likely to wreak havoc in all other cases. The last chapter of this volume. in the same vein as the chapter by Aréna and Festré. even in the face of the temporal opportunities created by asymmetric information. In financial markets their presence may cause herding. even to the point where they may cease to exist. That is the subject of the next chapter.’ ” Here we enter the domain of market failures as tradition- ally defined. The counterexample they give is the story of three Internet firms that live off the trade in second-hand and out of print books. asymmetric information between buyers and sellers sets in motion a process of adverse selection of the quality of goods that will make markets function worse and worse. This is the theoretical premise. First of all: what are big players (BPs)? They are agents that have the power to influence the market in which they operate. Its second. One of the objectives of this chapter is to test whether BPs encourage herding in commodities markets. is established by means that do not need intervention from a non- market agency. Trust. To start with the latter. According to Akerloff. say the authors. more speculative goal is to indicate what consequences BPs may have for the Internet economy. Whereas Foss and Dulbecco and Garrouste are groping for an Austrian theory of the organization of the firm.15 This is very Hayekian: individuals are not supposed to have an opinion about the system as a whole and if they do. it is the outcome of the creativity of entrepreneurs who earn money by acting as intermediaries. Conclusion: Internet markets are short-lived or show other signs of failure. Not so. Their success and that of the market they help to maintain is then explained in terms that are borrowed from Hayek: markets are self-regulating processes that are in continuous evolution. “Turning lemons into lemon- ade: entrepreneurial solutions to adverse selection problems in e-commerce” by Mark Steckbeck and Peter Boettke will not be lost on an audience of economists. The authors reject the hypothesis that individuals generally behave decently. which is fundamental for their efficient functioning. gives an analysis of this process in more general terms. The pun in the title of the next contribution. out of the type of well- understood self-interest that is equivalent to the recognition that they are all stakeholders in a trust-based market system. such as central planning or in the presence of one or a few other very influential agents.17 and their behavior does not follow rules. Its argument can be summarized as follows. Koppl and Sarjanovic argue that the Internet economy has lower transaction costs and is very much like the economy as described by Carl . BPs create uncertainty in the markets where they operate.

the more painful it will become to reverse these misdirections of investments. the author endorses the BP theory of Koppl and Sarjanovic by emphasizing that Greenspan’s measures do not fit into any of the accepted economic cubby holes and have thus become more unpredictable. velocity and virtuality. That would imply that BP intervention can be expected to be even more costly in terms of higher prices or lost opportunities for innovation than it would in the pre-electronic age economy. and this led to a loosening of investment morals. The monetary sector in the internet economy Antony Mueller’s “Bubble or new era? Monetary aspects of the new economy” applies the lessons of Mises’ business cycle theory to the Internet economy. This is further enhanced by the computer and Internet-aided globalization of financial markets. Mueller analyzes the shift in the stance of the president of the Central Bank of the world’s most powerful economy. This is the second way in which this chapter’s analysis is very “Austrian”: expectations have a crucial role to play. Second. Implicitly. the test of the BP hypothesis in the international wheat market is used as a speculative proxy for the BP–Internet economy hypothesis (the authors conclude that the outcome corroborates their theory). For one thing.20 This has created a climate of expectations that nurtures moral hazard and has set off a chain of international financial crises which according to the author is likely to continue. it created a climate in which undisciplined monetary policies in other countries were rewarded instead of being punished. it lowered the perceived risk of investing in the Internet economy. The longer this lasts. All this does not invalidate the message of Austrian business cycle theory: the huge global mechanism of money creation gives rise to malinvestments. His high expectations that the Internet economy would reinforce the United States’ leading position in the world economy led him to allow the increase in the stock of money and the level of indebtedness to continue undisturbed. Borrowing the terminology of the previous chapter. The test itself involves the application of a model for measuring herding and bubbles in terms of error duration that is of interest in its own right. as he so nicely puts it. the argument of the current chapter may be formulated thus: what started as an . which remain hidden as long as the boom continues. Introduction 13 Menger in that it relies on knowledge even more than the non-digital economy. which are characterized by greater volumes. but also as the most important financial market and member of international financial institutions have had far-reaching consequences for the world economy as a whole. The reward consists in what practically amounts to a guarantee that any country that makes a mess of its economic policies will be bailed out. Even though the New economy failed to manifest itself in productivity statistics.19 The power of the Federal Reserve Board’s president in the United States and that country’s powerful position as the “consumer of last resort” as Mueller calls it. volatility. Alan Greenspan. it affected the global economy in a more indirect and perhaps more profound way. Since prices that have not been established and innovations that have not materialized are unobservable.

Free banking and competition between currencies might very well become the rule. As to the latter. The author looks for an answer to this problem in the tradition of law and economics. led by positive expectations about the future of the Internet economy. is reinforced (or perhaps endogenized) to the extent that it creates effects in the world economy as a whole that risk getting out of control. the question is modified into the problem whether a system of legal rules that regulate trust and protect reputation can emerge spontaneously. In this perspective. and protects individual freedom. Electronic money looks like the logical next step in the evolution of money from a means of exchange and store of value with a real value based on alternative uses to a purely trust- based medium. the question this chapter poses is what an Internet-endogenous monetary system might look like.” The legal framework Almost the exact opposite is advocated by Elisabeth Krecké in the final chapter. Krecké examines and rejects the positivist conception of law. The authors indicate that this would have far-reaching consequences for monetary policy (including the maintenance of price stability) and law enforcement (as electronic money makes tracking illegally earned money almost impossible). including central banks. and then more particularly in the half of that tradition that applies economic tools to legal contexts. however. which will thus lose their monopoly in the field of monetary policy. How this may be achieved is one of the issues addressed in the next chapter on “Possible economic consequences of electronic money. It takes ever more strenuous efforts to keep the ever more rapidly accumulating debt from forcing the real economy to a crash landing.” by Jean-Pierre Centi and Gilbert Bougi. will not necessarily signal the end of the Internet economy. and what are the features of such a rule-creating process. This is because the real economy is in continuous evolution and capable of adopting digital innovations – provided the monetary sector does not produce excessive noise.14 Jack Birner instance of BP interference. banking will not be limited to officially recognized financial institutions. the conclusion of this chapter points to the possible need for intervention: “The challenge is to develop an institutional framework that provides transparent rules for the electronic payments system.” Her point of departure is that the Internet has created such a fast evolving and complex and unprecedentedly novel system of transactions that the traditional types of centralized regulation and legislation will not be able to keep up with it. This leads to two further questions: what constitutes law. safeguards the value of money. Whereas in the previous chapter the monetary sector remained something exogenous to the Internet economy. With the introduction of electronic money. in one country’s monetary sector. Whereas most of the contributions to this volume concentrate on the spontaneous mechanisms of the Internet economy. “The emergence of private lawmaking on the Internet: implications for the economic analysis of law. The end of the boom. It has been very influential in shaping the idea that is now almost generally accepted and that identifies laws with commands emitted by a national government or a .

A concrete example of this is given by Steckbeck and Boettke in Chapter 10. The implication for the economic analysis of law is that it will become a discipline of increasing importance in the age of the Internet economy. The author might have drawn the parallel with the invisible-hand mechanism through which protection agencies are established in Robert Nozick’s Anarchy. One of the examples it gives are the railways. It regards legal rules as the result – at least in part – of a discovery process in which entrepreneurs try out new types of contracts. Introduction 15 supra-national authority that have to be obeyed in order to avoid sanctions. That the evolution of the monetary and the legal system have much in common becomes particularly clear where Krecké writes that an economic interpretation of rule formation on the Internet shows that it is difficult to distinguish between what is and is not law. This raises the largely unexplored field of property rights on the Internet. This chapter is complementary to the ones by Miller and Stiegler and Steckbeck and Boettke in that it gives a more theoretical and philosophical analysis of the type of concrete mechanism that is described there. which is that of the question of what is law. many technological revolutions started with an initial boom and bust only to become very influential later. see the chapter by Centi and Bougi) of new types of money. of course. That will only happen if they keep a good reputation. Instead of selling a product and relying on an existing legal framework. the system is copied without compensation. Unless. In order to achieve that. etc. he or she may let others make use of it against the payment of a fee (or a lump sum if the system of rules is sold). Entrepreneurs need to survive in order to make a profit (and the other way around). a process that has made the control over the stock of money by central banks an increasingly difficult task. Internet entrepreneurs often sell a package that consists of a product and a set of promises that include the safety of the payment made. . The top-down type of law that uses the crux of legal positivism has to yield more and more to a type of law that is created from the bottom upwards – largely because the first type is not satisfactory for commercial transactions on the Internet. This is more like what happens in the Internet economy. the guarantee to return the product if it does not live up to the buyer’s expectations. Here we find ourselves in the second domain the author discusses. State and Utopia. This is very similar to the invention by private banks (and not only these.21 Another mechanism she might have used as an example is the way in which Carl Menger (she only refers to him for his analysis of law) and almost every economics textbook after him rationally reconstructs the emergence of the modern banking system. they need previous customers to return and new ones to become clients: entrepreneurs need to inspire trust. The underlying chain of reasoning is as follows. If someone invents a system of rules of behaviour that is particularly conducive to this. This view is contrasted with one that the author finds in the work of some economists and which she calls polycentric. Notes 1 As The Economist (23 January 2003) rightly argues.

This potential confrontation with a more typically sociological approach. 16 This ideological bias in Hayek’s work is discussed by some of the contributors to the first conference organized by AHTEA. It is tempting to elaborate Lavoie’s analysis in terms that are more familiar to philosophers of language.16 Jack Birner 2 Microsoft’s hold over the world of the Internet is nicely illustrated by the example of a French participant to the conference who unintentionally referred to portals as “gates.” which would be more accurately described as not following rules that are known or knowable to other market parties. Positive network effects are very similar to ecological niches. the Austrians’ emphasis on the importance of knowledge and its communication logically presupposes the existence of a communication structure. is not discussed in this chapter. He is right in the sense that the network type of analysis by Austrian economists has remained largely implicit. It merits. 8 Apparently. where networks have a very prominent role. 15 Yet another explanation would be that humans are basically decent social beings who by nature show a tendency to behave cooperatively. which facilitates communication. 11 And the notes that I added indicate that I at least find this chapter very inspiring. and to Guido Fioretti’s contribution. where Hayek’s 1937 “Economics and knowledge” is singled out as one of the earliest network analyses of markets and competition. further research. 9 If this were not an introduction. he agrees with Robert Solow’s well-known quip that “you can see the computer age everywhere except in the production statistics” (Solow 1987: 36). we could perhaps speak of groupware and hypertexts as world-1 objects that link world-2 psychological states and in the process contribute to the discovery of world-3 content. 7 For a non-hermeneutical attempt to deal with the role of language in an economic context. one of the examples of BPs the authors give. 5 http://eco83. 14 Alan Kirman is a conspicuous exception. 10 Here lies an interesting potential link between economics and the sociological literature on the internalization of norms from which both disciplines would stand to benefit. However.unito. see the brief discussion in the next 12 After risking his life for it. Hayek has already been mentioned. cp. The fact that this article is discussed by several contributors to this volume without going into its network features strengthens my amazement. a more widespread agreement on meaning may be established. Birner 1999.” the authors add. Ludwig von Mises’ analysis of entrepreneurship is the other example. for example. 6 As one of the organizers I am hardly an independent witness. That seems unduly restrictive. Garrouste and Aimar 2001. In the process. as the saying goes. Kirman 1983. 17 “They are funded by a pool of fiscal resources. Apart from my previously mentioned article I refer the reader to the first section of this volume. for the description of which network analysis is an excellent instrument. by Birner and Ege 1999. The two pieces of software that he discusses can be seen as means of connecting private-language arguments (à la Wittgenstein) or local networks of meaning (in the sense of Ryle) to more encompassing structures of meaning. In a Popperian perspective. in order to . as I intimated in the previous note.” 3 Compare Birner 1999. Let me therefore hasten to add that we received unsolicited compliments from many participants. Few social scientists enjoy the dubious satisfaction that their ideas are thought to be so important that someone or some group – in this case the terrorists of Sentiero Luminoso – wants to eliminate their author. 18 It is “discretionary. 13 That Hayek’s purely economic approach is too poor to describe modern market societies has been argued. 4 My fellow-editor Pierre Garrouste commented that what I write here is in apparent contradiction with my earlier remarks on the neglect of network models. But this would not fit the behavior of central banks. Compare Birner. amongst others. however. Compare. I would be tempted to elaborate the parallel with group selection in neo-Darwinian context. For instance.

M. (1999) “Two views on social stability: An unsettled question. a problem that is addressed by the author. if there was any literature on BPs in the art market. The J.. (1983) “Communication in markets: A suggested approach. But then.” in S. fit the description (if we delete the limitation to fiscal resources). (2001) F. A. and Aimar. though. Its scope is wider.s: where they have taken their degree in economics and when is essential information for knowing the sort of rules they will follow. Hayek as a Poltical Economist. Garrouste. Birner. J. 20 The fact that the explanation that the United States backs this up in order to ward off communism has lost its footing might lead one to give more weight to endogenous economic factors. for instance. (1960) Information and Investment. P. I asked Neil De Marchi. R. Granovetter. Oxford: Blackwell. Bibliography Birner. Introduction 17 predict the decisions of directors of central banks.” American Journal of Economics and Sociology. 19 This is also an implicit criticism of what I observed in note 18. it takes more fantasy to imagine the same taking place on the Internet. . Nozick.” New York Times Book Review. State.B. despite the fact that it is one of the few instances in reality of a market where auctioneers make prices. London: Routledge. G. (1987) “We’d better watch out. Extending the domain of application of BP theory would be very much in the Popperian spirit to which the authors subscribe. Solow. R. Economic Analysis and Values. J. Dow and P. (1999) “Making markets.” American Journal of Sociology 78: 1360–80.” Economics Letters 12: 101–8. T.M. J. Cheltenham: Edward Elgar. (1973) “The strength of weak ties.E. Paul Getty Trust has the means to influence the art market decisively. Birner.A. What sort of violence could be exerted there? Spam attacks perhaps? The more relevant problem is how property rights can be established. Earl (eds) Economic Organisation and Economic Knowledge: Essays in Honour of Brian Loasby. Very wealthy charities. but to his knowledge there is not. Utopia. and Ege. Kirman. it is sufficient to have a look at their c.v. Oxford: Oxford University Press. who writes about the art market and its history. which would be a very Marxian conclusion. 21 The difference is that Nozick argues that the monopoly of violence is established by the strongest protection agency. R. (1974) Anarchy. 12 July. the art market is not a typical market. Richardson.C. The characterization of a BP and the discussion in the chapter strongly suggests that BPs coincide with government agencies and central banks. it would make the theory more falsifiable by increasing its content.


Part I Digitally connected networks .


. by contrast. The Austrian school’s view of economic development identifies the entre- preneurial process. When Ludwig Lachmann refers to an orientative approach to economics he . in making and revising their plans they will take account of these facts. The production of larger amounts of physical stuff seems to be the very definition of wealth. But the Austrian school. entrepreneurship and the convergence of groupware and hypertext Don Lavoie Entrepreneurship as mutual orientation There are at least two possible types of social process. and not as determining forces. . Lachmann 1994: 204–5) What are the chief causes of the extraordinary economic prosperity in the so-called knowledge economy of the past couple of decades? Our models of economic growth have emphasized objective. And our ability to coordinate with one another in turn may depend fundamentally on our capabilities for mutual orientation. whatever men do within a period depends on the position they have reached. a process for the mutual coordination of plans. In the first. In it men’s actions are neither determined by what happened in a past period nor by the distance of their present position from an imaginary equilibrium. contends that it is mainly our ability to coordinate with one another that enables us to increase the objective standards of living of people around the globe. No doubt. the very process the Austrian school has identified as key to economic prosperity. leaves ample scope for divergent expectations. to see not the things themselves. but to see how one another are seeing things. (Ludwig M. Tools that support the way we orient ourselves to one another directly aid in the entrepreneurial coordination of plans. as a key factor in the creation of wealth.1 on the basis of what it calls its “subjectivist” approach. A ‘feedback’ mechanism in which each subsequent step depends on ‘distance from equilibrium’ is a special instance of it . We may describe the first as ‘mechanical’. material productivity. (There may be more). The other kind of social process. the second (for want of a better term) as ‘orientative’. It could be that among the important effects information technology is having on the new economy are direct enhancements to this mutual orientation process. But the latter serve them as points of orientation. The theory of entrepreneurship studies the way the multiplicity of plans of market participants get oriented to one another.1 Subjectivism.

is an important contribution. Just as the successful market participant learns to orient him/herself to the articulations of the market. but as referring to the capacity of the economy to coordinate its uses of all the factors. or prices) that have been “written down” in some specific medium. interpreted. the creation of real wealth. The realm of meaning is best seen as a public realm of expressions.22 Don Lavoie is suggesting that economics ought to be seen as an interpretive study. and the texts that result from such processes.2 However. worked upon. and that these meanings are fully in the real world. Subjectivism should try to understand the processes for the articulation of economic meanings. The point Kirzner makes about entrepreneurship as not merely a factor of production to be optimized along with all the others.) through which market participants attempt to communicate their meanings to one another. written. The term “subjectivism” may be unfortunate. whether on clay tablets. focusing on the ways in which market participants articulate meaning to one another. The word seems to imply some sort of turn inward into the realm of the mind. and the ability of market participants to get more of what they want for less. but to the public articulations (verbal. etc. papyrus scrolls. In his important work. charts. graphs. Israel Kirzner developed a perspective on the fundamental nature of entrepreneurship according to which it is seen as involving alertness to opportunities to improve on the use of priced factors of production. or even rearranged. The diverse plans of separate individuals come to “dovetail” with one another through the entrepreneur’s coordinating actions. Indeed. increasing the market’s coordinative capacity. and through that action to improve on the market’s coordination process. they are what make all the difference in economic development. what Austrians are talking about is not inaccessible mental thoughts but human expressions of meaning. Taking the idea of meaning seriously obliges us to pay more attention in the theory of entrepreneurship to the way market participants engage in acts of articulation and of the reading of articulations. or relational databases. Entrepreneurship. texts are also available for future return. It is the ability . electronic.3 The entrepreneur’s alertness to market opportunities is what impels him or her to act. to be reread. but as not itself a factor of production in the ordinary sense. especially of articulations in language. so the economist needs to attend to the communicative processes through which market participants orient themselves to one another. Texts in this sense are words and other articulations (including drawings. in this perspective. It is understood to be the essential driving force behind the process of economic development in that it improves the allocation of scarce resources. quoted. The point of subjectivism is not to turn our attention to some kind of inwardness. But the concept of alertness as he developed it suggests that what is involved in entrepreneurship is primarily a matter of directly seeing something that phys- ically exists in the world. paraphrased. is a sort of meta-factor of production: it is the alertness to how best to use factors of production. Competition and Entrepreneurship. Having been written down. not the communication of meanings among market participants within specific technological and institutional contexts.

The point is that the entrepreneur is not merely seeing what is there to be perceived. and acts to rearrange factors of production into a pattern that tries to reap profit from the situation. Even these price signals are somehow taken to already have their meaning ready-made. The elaborations of the entrepreneurship in Kirzner’s work are too abstractly described. whether oral. but simply spotting numbers. The Kirznerian entrepreneurial act might seem to be the engagement by an isolated. He may be guilty of underemphasizing the important role of language that is involved in the entre- preneurial process. They depict the entrepreneur as an intermediary in a world of (already somehow meaningful) price signals. where the social relation par excellence is not market exchange but the exchange of words. and other articulations. before anything like markets can come upon the scene. social.5 and they work through concrete acts of articulation. but as a kind of reading of a meaningful situation in a language-constituted world. as Mises was claiming in his famous challenge to socialism. The “seeing” of an entrepreneurial opportunity is best understood not as perception. contracts. .4 The entrepreneur is the maverick who simply “sees” objectively what others have so far overlooked. There is no explicit place in the analysis where human articulations of meaning happen. that this is the friendliest or the most useful way to read what Kirzner and other Austrians are trying to do. price discrepancies. without which. advertising. almost as if these are peculiar agents who only communicate in money bids and offers. with which we silently “signal” to one another. and making profits from exploiting differences between costs and revenues. I do not think. however. but we do not need to ignore that element of the process. but now there are these other kinds of texts such as prices circulating in our conversations that take numerical form and contribute to certain cognitive processes we call accounting. cultureless individual to pre-existing objective circumstances. written or electronic. Or rather it is as if the only things we ever say or write are the numbers we call prices. Subjectivism and entrepreneurship 23 to “see” opportunities to take advantage of (and unintentionally improve upon) the market’s pricing of productive inputs. prices. and technological environment. modern technology could never have evolved. You could say that markets simply add whole new kinds of texts. to the conversation. etc. We are already engaged in acts of verbal articulation. The entrepreneur “notices” inputs that are underpriced in relation to their (perceived) potential to produce outputs. These articulations take place in a given political. Markets are an extension of language. but rather opening up whole new ways of seeing. where we are trying to say what we mean to one another. This is the kind of approach to entrepreneurship that has been usefully elaborated in Spinosa et al.6 The words are still a key part of market communication. as if they were not involved in linguistic behavior. The Kirznerian entrepreneur is properly understood to be thoroughly immersed in the already linguistically constituted world. profits. This very effort to gain pure entrepreneurial profit has the effect of increasing market coordination. (1997) as a process of disclosure of new worlds. and a linguistically constructed and institutionally contex- tualized world.

language. our action is in each case orientated towards a complex network of human action of which we know enough to make it serve our ends. He also includes such phenomena as law. as orientation schemes of the second order. we may regard institutions. Whether we post a letter. It is impossible to elucidate such meaning until we realize that the mode of existence of institutions corresponds to. Although Lachmann did not explicitly work out a distinct theory of entre- preneurship. Unlike many who seem to only see institutions as constraints on the freedom of action. but rather how we are oriented to one another. We need to remember that Lachmann uses the term “institutions” to cover a remarkably wide set of phenomena. and prices. to read the “market signals. the one that is implicit in his work depicts it as essentially a process of mutual orientation. He thought that key to the entrepreneurial function is the ability to interpret. wait for a train.” Anybody can do the calculations of potential profit or loss but the profit/loss signals guide the market process . Such a mode of orientation is an element of culture. to which planners orientate their plans as actors orientate their actions to a plan . Like Menger and Mises he saw institutions as integral parts of economic processes. though we may know next to nothing about the internal working order of these institutions. If the plan is a mental scheme in which the conditions of action are co-ordinated. They are immersed in history.24 Don Lavoie Lachmann (1994: 283) pointed in this kind of direction when he called for an orientative or an interpretive economics. or draw a cheque. a web of thought – open to interpretation but not measurable. He includes organizations. the mode of orientation of those who participate in them. how we are oriented to the physical world. money. By “orientation” he meant not only. our obser- vations cannot disclose to us what meaning their objects have to those enmeshed in them. It enables them to co-ordinate their actions by means of orientation to a common signpost. or even primarily. . or the use of mailboxes for posting letters. and varies with. but also regularized practices such as profit/loss accounting. He describes the orientative function of institutions in his essay “On institutions” (1971: 49–50): An institution provides means of orientation to a large number of actors. They are “signposts” which facilitate the mutual adaptation of plans. a meaning that varies from group to group and over time. not that of nature. Although we can observe their operations. Institutions belong to the realm of culture. and other evolved structures by which human action gets mutually oriented. Lachmann saw them as indispensable aids to human action. as it were. . He points to the “mode of orientation” through which individual actors coordinate their plans with one another.

Lachmann ([1951] 1977: 102–3) put the point this way in a review of Mises’ Human Action: Profits. The market process . . An opportunity is “seen” if the entrepreneur is sufficiently alert. Entrepreneurship is not an individual act of perception. As Lachmann suggests. those temporary margins between today’s cost of complementary factor resources and tomorrow’s product prices. a matter of looking at an already existing physical reality. and the elimination of those who cannot read the signs of the market. promotes the rise of those better equipped than others to wrest economic meaning from the happenings of the market- place . In Kirzner the question can sometimes seem to be: how do entrepreneurs discover. Subjectivism and entrepreneurship 25 through the selection of those who prove to be better readers of market signals. In a symbolic form they convey knowledge.” What we are really talking about here is not that mysterious. Lachmann (1956: 22–3) makes this point about the importance of the process of interpretation this way: “In a market economy success depends largely on the degree of refinement of one’s instruments of interpretation. are signposts of entrepreneurial success. . The essence of the matter is that the market process promotes the spreading of knowledge through the promotion of those capable of interpreting market data and of thus transforming them into market knowledge.”7 The entrepreneur gets at what is going on in the economy not by directly confronting . . it is an interpretive and communicative process involving efforts of different persons to communicate meaning about the future to one another. every act is a source of knowledge to others. and in particular to one another’s articulations? The philosophical framework of Kirzner’s entrepreneur seems to be one that takes “seeing” to be central to the process of mutual coordination.” But looking and seeing are misleading metaphors for what is better rendered as a transaction of articulated meanings. or create. What we have here is a coordination process that is taking place among persons who are oriented toward the future. And crucial to these articulation processes are the institutional and technological circumstances in which market participants act. the degree of refinement of the “instruments of interpretation” is key to market success.” the single individual who is more or less alert to opportunities. it is simply the reading of the texts of market- relevant articulations. . but the symbols have to be interpreted . or share objective knowledge of the physical world? A better way to put the question might be: how do they orient themselves to one another’s perspectives on the world. Among the institutional forms that Austrian economists ought to pay attention to are the articulation processes and the resulting texts through which human actors communicate their meanings. or if not it is “overlooked. On the other hand. It would be better to understand what is going on here as a matter of what could be called a dialectic of “seeing and saying. . It is perhaps misleading to even present this coordinating process in terms of the activity of “the entrepreneur. .

Producers of capital goods several stages removed from the final consumer have no way of knowing what is more valuable other than by the information contained in relative prices. as numbers. This is what I want. of contracts. of production possibilities.26 Don Lavoie the world itself. of descriptions of products. What tools like these do. The Misesian idea of economic calculation points to the necessity of quantitative profit/loss calculation in order for the remote meanings of consumers to be translated. Permeating the market process are millions of efforts to communicate meaning. The pricing of goods and services is a set of additional acts of articulation inserted into a world already buzzing with acts of talking and writing. entrepreneurship should be conceptualized as a social process of mutual orientation in which tools for improving our ability to make and comprehend articulations would be expected to have a significant impact. exactly. into a readable form. to orient their actions to one another. in some manner say what they mean to their suppliers and customers. We are still trying to get at what opportunities for profit really exist. The question shifts from how or why the entrepreneur sees an opportunity others have missed. by way of an effort of interpretation of what is being said about the world. In other words. through the process of imputation of value. write. The communicative media. A fundamental change in the tools for writing and reading. Since entrepreneurship fundamentally involves attempts to communicate. and this is how I could pay. in which such communication can take place. is make it easier for people who see the world differently from one another. we need to inquire into the communicative-technological environment. able to coordinate with consumers only because of this price information. their coworkers. this is what I can do. but indirectly. to the question. within which ideas are brought into articulation and reinterpreted. To get a handle on entrepreneurship we need to ask about the prevailing capabilities of the communicative media through which entrepreneurs are trying to orient themselves to the plans of consumers and producers. Producers of capital goods are suspended in this relative-price world. to nevertheless make sense to each other. the media. and this could be expected to have significant effects on our orientative capacities. if you will. such as were provided by the printing press a few centuries ago and by the Internet today. etc. through interpreting the efforts of others to say what they mean. . But the point is not that prices alone. where the market value of intermediate capital goods is a necessary guide to production decisions. how do market participants talk. of alternative technologies. I will claim. do market participants manage to convey their meanings to one another? An opportunity to improve the coordination of plans is not so much “noticed” with the eyes as it is gleaned from conversations. or stock- holders? The entrepreneur is situated within the context of a complex capital structure. but the way to get at these opportunities is indirectly through texts. show. How. have changed dramatically over the past several decades. potential investors. are sufficient to accomplish this coordination. could utterly transform the workings of this sort of process for knowledge creation and sharing.

the consensus of entrepreneurial minds. “by clear and unambiguous concepts. Its arrangement is not arbitrary. The entrepreneur’s function as regards capital . . by in some sense “fitting” into an evolving structure of capital. Early in that book Lachmann (1956: xv) says that he is not pointing “towards the ‘objective’ and quantifiable. This does not mean that the phenomena of capital cannot be comprehended by clear and unambiguous concepts. The order consists in more or less intricate patterns of complementarity among diverse. Beer barrels and blast furnaces. Lachmann sees the process by which entrepreneurs adapt the structure of capital to the changing demands of consumers as central to the process of economic development.” The process of economic production involves the continuous adaptation of heterogeneous parts of the capital structure to one another. specialized capital goods. The stock of capital used by society does not present a picture of chaos. Like others in the Austrian school. and expectations. We can comprehend capital. This idea of fit is what Lachmann called “complementarity. but towards the subjective interpretation of phenomena. . remain hidden. are for the Austrians what drive improvements in standards of living in growing economies. is to specify and make decisions on the concrete form the capital resources shall have. It is in this context that he sees the essential role of the entrepreneur as one of figuring out how to specify particular ways of using heterogeneous capital goods that in principle can be used in a multiplicity of alternative ways. Lachmann (1956: 16) identifies the key role of the entrepreneur in terms of deciding how initially to select (or later adapt) particular patterns of use of capital goods. it reflects the entrepreneurial appraisal of such objects. . As long as we disregard the heterogeneity of capital. The generic concept of capital without which economists cannot do their work has no measurable counterpart among material objects. . . In a homogeneous world there is no scope for the activity of specifying.8 . which the Austrians sometimes mis- leadingly called a “lengthening” of the structure of production. Lachmann insisted. the true function of the entrepreneur must . from a subjectivist point of view. .” Then he gives us an idea what he means by a structure.” even while we insist that it is built out of subjective orientations. regards it as capable of yielding an income. Something is capital because the market. Subjectivism and entrepreneurship 27 Text and its structure Lachmann’s work Capital and Its Structure depicts the entrepreneur as involved in trying to mesh his/her plans with those of the ultimate consumers. Enhancements to capital. Capital is a structure in the sense that it has a meaningful order. . There is some order in it. values. harbour installations and hotel-room furniture are capital not by virtue of their physical properties but by virtue of their economic functions.

How Buildings Learn. Like capital. they are not understandable as an amount of something. . we should look at texts in these terms. The adaptation and complexification of the capital structure can be thought of as a sort of social learning. a specialization of individual capital items. The new capital must fit with the old (and other new goods) in order to be useful. texts cannot be adequately appreciated apart from the institutional context within which they are used. Lachmann is saying.9 One can say that our capital structure “learns” in the sense that Stewart Brand (1994) meant when he wrote the fascinating book. Texts should be thought of as possessing a meaningful. in which parts have definite relationships to other parts. For the same kinds of reasons Austrians have wanted to discuss capital in terms of heterogeneity. recipes. A text’s parts stand in definite meaningful relationships to other parts. . The subjectivity of the structure. born in the ongoing growth and division of knowledge. just as it gets embodied in our institutions. so we can question those who tend to see “information” as a homogeneous quantity and media as nothing more than carriers of information. an increasing intricacy of the pattern(s) of complementarity among increasingly specialized capital goods.” We might call it a “complexifying” of the capital structure. and need to be taken into account. The lengthening of the capital structure involves what Lachmann (1978: 79) calls a “‘division of capital’. people working with these material things keep adding adaptations to them.28 Don Lavoie Howard Baetjer (1998: 22) refers to this notion of lengthening but I think he has a better way of putting it with the term “complexifying. but as a heterogeneous structure with complex internal relationships.” When new capital goods are developed . prices. texts can “learn” in the sense that those who work with them can add on improvements over time. where new knowledge is not only held in people’s heads but comes to be embodied in our capital structure. structured reality. and to adapt elements of the structure so they fit into the whole. The reality in which entrepreneurial coordination has to take place is a complex. and to other texts. does not change the fact that the structural relationships are real. more complex patterns of complementarity. When that entrepreneurial process is working well social learning in the form of rising standards of living can be the result. and structure. Capital goods learn in the sense that people incrementally adapt them so that they come better to fit the environment into which they can be fit. say information. Just as Austrians challenge those mainstream economists who mistake capital for a homogeneous quantity. Entrepreneurs face the challenge of navigating within this complex structure of interconnected meanings. and documents. that is. complementarity. context. evolving structure of this sort. Like capital goods.10 It . Capital goods are appraised as having the potential to earn a future flow of income because they are thought to be able to fit into the structure. Like capital. laws. practices. what is generally involved is the development of new. and ultimately to the desires of the participants.

We do not just possess information. There is at least a minimal amount of structure in any text. it is necessary to have not only an exchange of unstructured content. The difference these added structural elements have made historically to the effectiveness of writing is very suggestive as we try to think about the knowledge economy. What exactly the “structure” of a text consists of will vary dramatically with different technological media. articulations can become structured in more complex ways that enrich our possibilities for conveying what we mean. to draw attention to the relationship between different parts of itself. What meaning a reader gets from a text will often depend greatly on the form in which the text is fixed as well as the institutional context in which it is used. and thus what we can do. or structures. Different presentational structures can make the same underlying content convey meaning very differently. and working together with. In order to work with production plans that reach a certain level of complexity. Structure does things like this. Subjectivism and entrepreneurship 29 can be fixed in many different kinds of forms. with questions in mind. to present the content of a message to an audience. but also to work with it. to see one part of it in terms of a larger whole of which it is a part. We find ourselves giving structure to the articulations in order to help guide the reader in how to read it. with distinct purposes guiding our interest.11 . and these can make it more readable. to take the text apart and put it back together. If the prevailing technologies for working with text permit. however. In such situations we are trying to help to provide the reader with a context in which the content of the message can be interpreted. To characterize structures only as reading aids. at least in the way the sentences are broken up into paragraphs and larger groupings. the data. And yet the structure of a text is not strictly separable from its content. Understanding the content in part involves understanding that structure. The key to the knowledge economy. would be to understate the scope of what they do. we come to specific articulations in definite contexts. it helps us to do things with texts. especially by allowing us to increase the complexity of what we can say. but an exchange of complexly structured articulations of content. An articulation by one market participant to another is not just a description of a part of a complexly structured reality. sequential string of words. an exchange of a linear. It may be that more complex kinds of production plans could not have evolved without accompanying structural features that made it easier for readers to work with texts. Thus we can differentiate the content of a text with the structure or form in which it is presented. it is more the revolution in ways of working with. The structures in a text can include things that help the reader not just to read it. Improvements to our ability to add structures to texts may have had an important effect on our ability to communicate what we mean. Structural elements we now take for granted like spacing between words and page numbers were not always available to aid readers. to place it into some kind of larger context. I will be arguing. it may also be a complexly structured description of a complexly structured reality. Complexity enters in two ways here. is not primarily a matter of getting “access” to quantities of data.

What the history of writing shows us is the profound changes that occurred as the result of the shift from a purely verbal to a written mode.13 As Walter Ong and others have elaborated. I think. could all be called. going from verbal. where transactions could only be conducted verbally and sealed with a handshake. The meanings that subjectivism says we should pay attention to are not some kind of myste- rious. and many types of market transactions today (informal deals between corporate executives. How we are able to “converse” makes an enormous difference in our ability to use. of course. and the contexts in which they are used. to electronic modes. and generate economically relevant knowledge. by fundamental improvements in the media in which we have been able to conduct our conversations. without recourse to preserving in text aspects of the transaction that might be repeated in the future. that today look to us like awkward ways to make up for the fact that nobody could preserve the “texts” in written form. “old wives tales. and how they have helped in the process of mutual orientation. and for interpretation. But the fact that we have changed from an oral to a literate culture has. and many of the differences have to do with the challenge of preserving knowledge when texts cannot be written down. to an increasingly various set of electronic texts. and are available for working with. and thus in our wealth and well- being. to a typed document printed. but that particular forms in which they have been fixed. convey. In a strictly oral culture a deal essentially needs to be transacted face-to-face. There is a need for complex procedures for the memorization of ritualized and formulaic verbalizations. even though an “utterance” could mean anything from a word or sentence verbally spoken.30 Don Lavoie The theory of entrepreneurship ought to attend more to the technological media through which our articulations can be constructed and used. The fact that texts have a material reality to them tells us that they cannot adequately be treated as pure content. Entrepreneurship has been strengthened. epoch stories. from talking to writing to printing. Exchange thus tends to . We could refer to three overall modes of conversing. utterly changed the way market coordination takes place. an exclusively oral culture is strikingly different from the kind of verbal-plus-written culture we are in today. The work of the historians of writing can help us to appreciate what textual structures have been in pre-electronic media. to a whole variety of electronic tools. The nature of entrepreneurship must have been completely different in such a world.” etc. yard sales) continue to occur primarily in verbal discursive modes. inaccessible realm of the mental.12 From the very beginning of the human species there has been an important connection between conversational media and economic advancement. sayings. auctions. To get some perspective on the possible impact the latest technological media may be having on entrepreneurship it might be useful to review the kinds of transfor- mations of media that have already taken place. In all these media we are exchanging utterances. Market processes first emerged in strictly oral cultures. to written. can matter. narratives. in a broad sense. Our media technologies. they are all around us in the material form of spoken words and written texts.. alternative ways of holding “conversations” with one another.

A key aspect of this transition was the improvements in our ability to add structure to texts. The transition from cuneiform and hieroglyphics to the alphabetic form of writing introduced new possibilities for ordering texts. is important. the book broken into pages. clearly. Oral conversation is a far more integrated process where the talking and listening are in a concise and continuous dialectical interplay with one another. Prices and other components of offers for exchange are not just written but posted in public places to attract buyers. The evolution of writing. are quite different from their precursors in oral culture. reworked. Writing and reading as a new form of discourse comes on the scene after. The earliest forms of writing seem to have been accounting records. The age between the rise of writing in the fourth century BC through the codex in the second century AD. and the printing press in the fifteenth. The fundamental advantages of writing over talking stem from the fact that things “written down” can be returned to. The cleavage is both the source of the greatest strength of writing/reading and also the source of its greatest weakness. writing down transactions was . By contrast. the writing and the reading of paper-based discourse are thoroughly separated components. The transition from the roll or volumen to the codex. I would argue. or in some way worked upon. has had profound effects on the workings of market institutions. etc. the cleavage between writing and reading. where I deliberately express the first as one thing and the other as two distinct things separated by a slash. reinterpreted. and all the various forms of monetary instruments are products of writing. I like to think of the comparison Ong elaborates of oral and written culture in terms of a contrast between verbalization and writing/reading. Money itself is writing. but it is a weakness in that we lose the immediate interplay of the saying and the hearing that is taken for granted in verbal conversation. At first. accounts that can be recorded. Subjectivism and entrepreneurship 31 take the form of bazaar trading. as opposed to increasing the likelihood of future. as a result of. contractual arrangements. Perhaps most importantly. has enjoyed the same kind of dynamic of increasing complexity through writing. It is a strength in that the word cut loose from its original context can reach a wide range of disconnected readers. and permits far more complex types of ownership arrangements. with a premium on squeezing surplus out of the individual deal. and these gradually complexify into written accounts in words of market trans- actions. written contracts and the whole apparatus of contract law become possible. Economic production. repeated deals. The possibility of written law stabilizes and clarifies property rights. These new technologies took off because the alphabetic form of writing and the structure of the book in the form of numbered pages permitted a more complex ability to manage the word. provided similar powerful structuring capabilities. Terms of incredibly complex transactions can be written out and consulted later in case of alleged breach of contract. The persistence of the written text allows us in principle to make improvements upon what we are saying by revision. was a gradual transition where western culture was increasingly shaped by written texts. and according to many interpretations of the prehistorical record. markets. The slash. Prices that can be written down.

the different strengths of oral and written modes of conversation. There are situations. if one wanted to have a conversation that picked two from column A. and three from column B. Structural hooks such as running heads. paragraphs. which writing is good for. Miller et al. In an article entitled “The open society and its media. We are used to such trade- offs and routinely find ways to bounce back and forth between modes to make up for the shortcomings of each.” Miller et al. where we might like to engage in a relatively quick interpersonal engagement of the kind talk is good for. it has also dramatically improved our ability to share these more complexly structured texts. Not only has the printing revolution brought about a significant increase in the kinds of structures available to writers. supply in their paper to illustrate the relative advantages of verbal and written (under paper technology) discourses. but it has the advantage of drawing attention to some of the essential issues. page numbers. (1995) summed up. tables of content. What Miller and his collaborators are really getting at is that the possibilities of electronic media can be usefully thought about in relation to trying to reach new combinations of these kinds of strengths. Miller et al. say. In the pre-electronic age. and in turn led to more complex forms of texts to guide these activities. are good for different kinds of communicative tasks. Verbal and written modes of communication. Strengths of each mode are as follows: Verbal Written Visibility of arguments (can hear Persistence of arguments an absence) (cumulative) Revision after publication Revision before publication Small expressions Large expressions Small audiences Large audiences Fast publication Freedom of entry Immediate feedback Complex reputation systems This formulation is of course a simplified one that passes over many of the subtle distinctions one finds in the literature on the history of writing. Conversations about a book could be precisely linked to texts in a manner that allowed a much more precise engagement with the ideas. were really tools for sharing text. are trying to provoke us to think about the combinations of strengths that are not easy to get in either mode. but also to make what we say persistent. indices. and so forth. By setting up this kind of contrast between the different strengths of alterna- tive pre-electronic modes of conversation that have been familiar for centuries. but having economically relevant texts available gradually permitted more and more complex types of transactions and production plans to be organized.32 Don Lavoie only an aid to memory in a fundamentally oral process. . are saying.14 I reproduce (with some slight rewordings) a chart Miller et al. in a particularly sharp fashion. it was difficult.

Or to put it in terms we have been using. These persistence-based advantages involve the various kinds of intertextual structures that can be built into any one speaker’s utterance. of the paper-based writing medium have been those related to persistence.” then many different knowledge views may be derived based on the particular content. and medium by which it is presented to the user . however. If the repository is conceived as a knowledge “platform. but rather are trying to offer practical advice to practitioners. the writings in knowledge management are not typically trying to offer social scientific explanations of how the world works. by and large. as well as the process of making the knowledge explicit . and afterwards in order to make improvements for the next edition. are too big to do justice to here. There is controversy about exactly what it is that these information technologies are doing that contributes to wealth. Their question is not . re- categorization. . standardization. and thus to use the feedback of interaction to revise what we are saying after we have made it public. A well-structured repository allows for a high degree of viewing flexibility. Knowledge structures therefore provide the context for making sense of information . on the other hand. which is essentially our ability to easily alternate among speakers. abstracting. What I would like to do here is to illustrate one aspect of the change in media. and I think we are too early in the media transformations to be able to take a critical perspective on all the different ways computers and telecommunications are changing the way the economy works. I will review briefly the themes of the business literature on “groupware” and “hypertext” that claim that these key capabilities of information technology are the force that is transforming the new economy. while content represents the information itself. Like most business literature. indexing. integration. have been those of interpersonal interaction. . That knowledge is then subjected to value-adding processes such as labeling. the ability to make flexible the kind of shared textual structures that were emphasized by the historians of writing. what is it about making text electronic that enhances our ability to orient our plans to one another? These questions. . The advantages of the oral medium. (Michael H. Zack and Michael Serino 2000: 307–9) It is commonly supposed within the business literature that the computer and telecommunications revolutions are somehow ultimately the engine driving economic development. and the ability to return to and revise what we are saying before we make it public. . Information and knowledge can be acquired from many different internal and external sources. The convergence of groupware and hypertext Knowledge comprises two components: knowledge structure and information content. . Structure represents a framework used to arrange information to make it meaningful. but of course the question is how. sorting. . format. Subjectivism and entrepreneurship 33 One could sum up the differential advantages of these media this way: the advantages.

These authors understand the point I have been making about understanding text in its context. the interpersonal and the intertextual.15 This skepticism about information technology’s contribution to productivity has spawned in turn a large but somewhat unfocused management consulting literature such as Thorp (1998) and Lucas (1999).” But not all this literature is unfocused.34 Don Lavoie so much to estimate the effects of groupware and hypertext capabilities on wealth as to try to explain how best to deploy such capabilities to improve the chances of making your organization successful.” John Seely Brown and Paul Duguid. The strand I have in mind is perhaps best epitomized by the papers collected in the book Knowledge. co-authors of the book The Social Life of Information. Smith. and of course there are plenty of ways to waste money on information technology “investments” which do not help productivity at all but which can mess up the statistics. In an age that emphasizes individual freedoms. The knowledge management literature argues that improvements in “knowledge sharing” have come from the convergence of these groupware and hypertext capabilities. Brown and Duguid (2000: 246). respectively. This phrase “knowledge sharing” is a bit misleading. purporting to show how businesses can overcome the “paradox. author of an often-cited Harvard Business Review article “The knowledge-creating company. Nonaka (1994) depicts the knowledge creating process as a spiral of writing and reading. The strand of knowledge management literature I am summarizing here offers some real insight about what software might be doing for some businesses that could be driving the new economy. It involves tools to enhance the kinds of work that groups of persons can do together. What we have here is not a matter of sharing some fixed thing called knowledge. or in knowledge management software systems in particular. When this literature refers to software capabilities they call groupware and hypertext they are generally pointing to the aspects of media that I have been calling. The “seeing” of opportunities is enabled through the facilitation of the “saying” of what is meant in texts. they are portrayed as . both institutional and technological. That is. for example. have been insistent in their claim that we need to see institutions as complementary with communicative technologies – as enablers and not just constraints on action. edited by David E. The emphasis is on the ability of such software to help market participants share knowledge. institutions are often portrayed merely as constraints on this freedom. of alternating between rearticulation and reinterpretation. Groupware and the Internet. contribute to worker productivity. Several scholarly studies have had difficulty finding clear evidence that investment in information technology in general. In particular. the potency of these new capabilities is said to be rooted in the process of a group of coworkers bringing tacit knowledge into explicit form. and Michael Zack of the Lotus Institute. and reading off articulations to glean new tacit awareness.16 Among the contributors are Ikujiro Nonaka.

that is. tables of contents. They change as people change. so these new electronic text advances are leading to new kinds of structures. the technological issues I have been calling textual structure – the way structure facilitates the contextualizing of what is said – is widely appreciated by these authors. Groupware begins with the notion of individuals working together in a networked environment. Yet. It might be more useful in some ways to think about the use by corporations of complex relational databases that are available to coworkers over local area networks. structured. this familiar sense we have from current Web browsers can be misleading. and vice versa. modifiable. We have all had experience. hypertext markup language. Software capabilities are powerful because they allow participants to add more kinds of structure to texts and documents. The issue is how meaningful data. Hypertext begins at the other end. can be gotten from raw information. headings. because they are human creations. Hypertext is not to be equated with the Web as it is now. static. The shift from talking about “data processing” to talking about knowledge management in fact signals a fundamental shift in thinking. such as the simple jumplink where one clicks on something to go to another document. and we should expect significant advances in new kinds of structuring capabilities in the future that would make what we see now on the Web look as crude as the manuscripts of the early days of writing look to us today. Groupware is more powerful precisely when it incorporates more hypertext capabilities. Although each of these aspects can be thought of in isolation from the other. Similarly. and soon comes to take up the interpersonal context of shared access to the text. to permit different views. even stagnant constraints. But while somewhat suggestive. might enhance their work. Just as advances in writing led to structures such as paragraphs. It starts from the context of an individual confronting a structured text. And they adapt to serve social needs. The terms groupware and hypertext can be understood as emphasizing two different aspects of these tools – groupware emphasizes the interpersonal sharing of documents and hypertext emphasizes the intertextual structuring of the documents. with the basic “hypertext” functionality that is supported in HTML. footnotes. different contextualizations of the information. electronic documents. . The technologies that enable this knowledge sharing consist of tools for the writing and reading of shared. The kinds of hypertextual structuring capabilities that have become available within corporations often go way beyond the simple links most of us experience while browsing the Web. including textual structures. and indices. in the early twenty- first century. Subjectivism and entrepreneurship 35 monolithic. and asks what tools. to be sure. or document repositories with built-in specialized annotation and search functionality such as Folio Views. knowledge. they rarely are just that. the knowledge management literature is increasingly emphasizing their combination as the key to success. or how far or fast it can be transported.17 The issue is not the amount of information or data that can be stored.

generator. Microsoft Exchange. however.19 Engelbart’s work came to focus primarily on the context of groups of coworkers. as distinct from the collaborative work context of groupware. Some of the literature focuses on the individual scholar using information on his own.36 Don Lavoie Much of the literature on the possibilities of hypertext has focused on the world of science. These applications exhibit a sequential flow of information into and out of a central repository. and workflow applications. These applications directly support interaction and collaboration within and among knowledge-based teams. Although simple email may give us a general sense of what the interpersonal capabilities are evolving into. . have evolved. was more oriented to the context of film and the humanities than that of collaboration in business organizations. structured to provide flexible access and views of the knowledge. . The real potential of groupware may be more evident in the more advanced kinds of applications that permit work groups to collaborate more effectively. toward the kind of context groupware litera- ture is talking about: collaboration within the workplace and coordination in the marketplace. The focus on the intertextual comes to embrace the context of the interpersonal. networked calendaring. Their primary focus tends to be on the repository and the explicit knowledge it contains. . The team of developers that attempted to implement Ted Nelson’s vision came to increasingly emphasize practical business applications of the ideas. scheduling.18 The development of hypertext tools and of the thinking about what they can do for us. for example. it can be misleading to focus on the limited capabilities of the kind of email systems most people are using today. and others (which include but go way beyond email) are closer to what the knowledge management literature is talking about than a basic email system is. literature. Collaborative applications are focused primarily on supporting interaction among people holding tacit knowledge. and scholarship in general. Distributive applications maintain a repository of explicitly encoded knowledge created and managed for subsequent distribution to knowledge customers within or outside the organization. groupware tools and the literature about them are converging from the other end toward hypertext themes. Products such as Lotus Notes Domino. Zack and Serino (2000: 306) are getting at the issue of convergence of these kinds of capabilities when they differentiate distributive from collaborative applications. and one of Doug Engelbart’s pioneering articles on hypertext was entitled “Special considerations of the individual as a user. enabling “teams of teams” to form across knowledge communities. or the tacit knowledge they may hold. . and how to augment their ability to work together and share knowledge. rather than on the contributors.20 Gradually both the thinking about hypertext and the practical systems involving these capabilities came more and more to turn to the world of business.21 Similarly. and retriever of information” (1961). Ted Nelson ([1974] 1987. the users. who coined the term hypertext. The earliest discussion of hypertext capabilities was the 1945 article by Vannevar Bush imagining a tool to help a single scholar work with information. [1980] 1990).

we have not integrated push and pull enough to make knowledge sharing work very well. and so forth. The combination of the interpersonal and the intertextual allows organizations to create new knowledge and share existing knowledge in different ways than are supported in a strictly paper-based communication medium. mark it up. what is needed is an integration of what are called “push” and “pull” capabilities. has not been primarily the texts. As a classic exposition of the idea of groupware (Lotus Institute 1995. had argued that what gave verbal discussion its power was its flexibility. with links. and leverageable knowledge by-product than the personal notes or memories of a traditional conversation or meeting. Structure connotes fixity and persistence. The simple and now familiar act of sending an email with a URL in it constitutes a powerful kind of functionality that allows us to work with information far more effectively than we could when push and pull were disconnected. what they call the capacity for “revision after publication. Push is like email (or sending around paper memos). but the coordination of a set of tasks and responsibilities. The traditional focus of groupware. categorizations. but flexibility seems to connote the opposite. Where the power of groupware really shows itself is when these two modes can work in concert with one another. rather than the primary focus of the application. they are focusing on the way a repository within a collaborative environment becomes a different kind of thing from one conceived only as a store of information. or other tools to make it easier to “navigate” around in it. unpublished manuscript) puts it. It is the ability to send somebody information. But to some extent it may still be conceived of as a tool for an individual researcher. Pull is like the web (or networked databases. When the best we could do is tell someone in a memo that the information needed is to be found in a file cabinet on the third floor. the documents with which the group works. What tools such as these can help with is the organizing of the interdependent tasks of a workgroup. The power of electronic media is in the paradoxical-sounding notion of flexible structures. When Zack and Serino (2000: 306) use the term collaborative applications. to alert someone of something. Subjectivism and entrepreneurship 37 A distributive application in this terminology is a repository of texts with a complex and to some degree flexible structure. what they called “persistence of arguments. the repository associated with collaborative applications is a by-product of the interaction.” “fast publication.” Writing is effective in helping interlocutors to express what they mean . It is the idea of deliberately going after information. The ability to capture and structure emergent communication within a repository provides a more valuable. In contrast to distributive applications. This repository of messages is dynamic and its content emergent.” and “immediate feedback. The interpersonal give and take of verbal conversation is what gives verbal conversation several of its strengths. or file cabinets). enduring. as Vannevar Bush had in mind. on the other hand. Miller et al.” They also argued that one of the factors that has given written communication its power has been the opposite kind of strength. annotations.

38 Don Lavoie precisely because words “written down. a layer of markups or annotations of different kinds. What is written down does not only get shaped into a single form with structural elements that aid the reader. The single hierarchical structure of a table of contents of a printed text can help a reader locate something.22 Paper documents force us to make a decision. at the choice of the reader. complex electronic text. Its fixity does not prevent it from being the site of a back and forth dialogue. you can save an original document in its pure. like “Who is on this task?” we hope instead that the information is stored by task. at whatever pace one might prefer for the kind of discussion involved. have the quality of fixity. Say we are keeping folders with information about both coworkers and the tasks they are working on. but can get shaped into a “meta-structure. or restructure. It seems useful to look at what is going on now in electronic media as essen- tially an extension of the same sort of evolutionary dynamic that the historians of writing described. untouched form. and stick with it. If we come with a different question. so to speak. so that equipping those who work with text with a flexible range of alternative ways of viewing the material can aid tremendously in the text’s usefulness. What you write into the electronic space can stay. to find what we . Fixing what is written down leads to the development of textual structures that permit the strengths Miller et al. called “revision before publication” to improve on the usefulness of the text. for finding one’s way around in a large. Think in this context of the limitations of file cabinets and folders as a way of storing corporate documents. we would be hoping the filing system has ordered them by name. Paradoxically. you do not need to give up the valued capability of relatively quick interpersonal interaction.” It can be given a form in which alternative structures can be selected. or whatever. A file cabinet takes advantage of the power of alphabetic ordering. Clearly. as happened with improvements to printing. but an electronic text can offer the reader a choice of a variety of alternative hierarchical structures. but can only take advantage once. to reorder. say. and so it can be there to return to. being able to have a networked database instead of a file cabinet to store infor- mation would make a major difference in our ability to get to it. Different readers approach texts with different questions in mind. The real payoff of electronic media comes from tapping the advantages of the complex structuring of texts that had arisen in the aftermath of Gutenberg. To “buy” the fixity and structure. but now in a context where multiple persons are involved in an interactive process with these structured texts. constitutes a fundamental advantage of the new media. as opposed to a networked database with soft- ware that offers the reader alternative hierarchies.” unlike words spoken. to add things to. and yet also add to it. to settle for one hierarchy as the least bad alternative. When we come to the file cabinet to look for information about a person. But electronic text is fundamentally a new kind of writing in that it also permits flexibility. The potential for navigating. it can be fixed. We can think of the new flexible structuring capabilities in terms of navigation and annotation. in that it has to select one kind of order over other possibilities.

and then produce a new draft. to any parts of the text that . but the dialogue only rarely includes a third step. and juxtapose to each name the tasks in the project that this person is responsible for.” This capacity for restructuring is not only an issue of how we manage to find information we are specifically looking for. Silvio Gaggi (1997: 103) points to the potential for hypertext to remove the “slash. readers can append their own comments and responses. The structuring of the content makes a difference. Electronic texts can make it so easy to follow links of this kind that in effect the reader’s experience undergoes a qualitative shift. this helps the reader to focus on how the labor time of any particular person is allocated. and so forth. or whatever. conveys quite different meaning. noting page numbers worth comparing. from beginning to end. For a long time people have been inserting annotations into the margins. or in order by name. a big difference. If I am trying to convey something about the personnel working on several tasks. Subjectivism and entrepreneurship 39 are looking for. or date. Printed materials can of course get annotated. Convenience is not mere convenience. in how it can be read. leading to a proliferation of parallel annotations. The need to go through such a process reflects the need to bring the interpersonal and intertextual closer together. Printed works often contain internal cross-references such as when we say something like “see page 5 on this point.” but the jumplink represents a dramatic reduction in the costs of following such connections. ongoing interchange. The annotations in most scholars’ personal copies of books and journal articles constitute a single further step of dialogue. It shapes our ability to convey infor- mation to others in a contextualized manner that helps us to orient ourselves to one another’s meaning. The typical experience in a book culture is that of the linear working through of a text in the author’s order. But the awkwardness of doing this in paper documents shows the potential of electronic media. It is with annotation capabilities that the distinctive differences of electronic media really become clear. Electronic media open up the possibility for a wide variety of other navigational capabilities. Printing had the effect of separating the writing from the reading process. and is useful in different situations. as Landow and others describe it. a response to the original argument. and to present it to others in ways that reflect our own meaning. if I order the information by the last names of the person. applying highlighter pens to key passages. What I have been calling the slash between writing and reading cuts off the process from what might have been a fruitful. collect the marked-up copies. The typical experience on the Web is one of following the connections that look promising to the reader. a rejoinder by the original author. Now we can say: “let’s view this list of documents in task order.” to reconnect what paper writing had separated: In hypertext. with names appearing under their tasks. add new nodes or lexias. In a corporate context it is common to duplicate a document. encouraging a group of people to separately mark it up with comments. But the same information in order by task. Annotations in paper represent an attempt to introduce interactivity into an inflexible medium.

yielding other selected representations of the document. . and yet without losing the advantages of fixity. can be marked up with a categorizing tool that allows the reader to “collect” the passages of the document that pertain to a certain issue. Might they be not only what are driving the wealth- creation process of certain organizations. We are already used to the way our marked up copies of printed text can become more useful when we have engaged with them. and they can create new links among the various lexias. say. Electronic media are beginning to transcend the limitations of most of the trade-offs identified in that table. talking and (paper) writing. Being able in electronic media to mark up a single shared text over a network opens up a crucial new kind of advantage. but also what are driving the economic devel- opment of society as a whole? These kinds of technologies are bringing about a profound transformation of the nature of organizations and management. were trying to serve with this contrast was not only to reiterate the kind of analysis Walter Ong and others have undertaken to compare the two great pre-electronic discourse media we have had up to now. That view of selected quotations can be shared with others looking at the same document. the markups can themselves become additional navigational aids. the distinction between reader and writer is attenuated. or adding some marginalia. by marking the key passages with a red pen. perhaps even dissolved entirely. With electronic media it becomes possible to use the capacity of the computer to search the markups themselves. or to reorganize information in terms of the accrued layers it has accumulated from active readers. and they may be primary causes of the new economy’s prosperity. The main purpose Miller et al. A hypertext document. One can view the original version without the annotations and other markups. It seems natural to ask what kinds of effects such technological changes could be having on entrepreneurship. In effect it makes the text into a living forum. and they can add other categories to it. as many authors in the knowledge management literature believe. and then at a click. but a communication system in which all participants can contribute to and affect the content and direction of the conversation. or that raise certain questions. but to point the way toward the future. for comparing talking with writing. Their point was that adding a third mode to our discursive repertoire would make an astonishing difference in society’s capacity to work with text. something more like words in a verbal dialogue.40 Don Lavoie they are interested in. Moreover. to view only those passages juxtaposed to one another. or one can view a selection of markups that come from certain participants. The advantages of adding groupware and hypertext capabilities to our repertoire of kinds of articulation can be summed up by reference to the table I borrowed from Miller et al. for example. The text is no longer a one-way communication system in which information and ideas proceed only from author to reader. Thus. a difference on the order of what happened with the inventions of writing and the printing press.

may be what underlies the very real. in turn. Entrepreneurship is not a direct “seeing” of opportunities. however. has been viewed through the lens of the Austrian school’s notion of subjectively meaningful structures. This means that we need to attend to the economy’s conversations through which we try to explain what we want. That literature. I see an opportunity mainly through coming to an understanding about what somebody is “saying” about the world. . what we are trying to do. and I am arguing that these kinds of changes may be propelling the entrepreneurial process. Subjectivism and entrepreneurship 41 Conclusion: mutual orientation and wealth This paper has tried to view the new electronic technologies through the lens of the literature on the history of writing and the printing press. The entrepreneurial process involves the mutual orientation of different persons to one another. Significant changes in our ability meaningfully to structure what we are saying have transformed the workings of entrepreneurship in the past. it could be that the recent general prosperity is being driven as much by improvements in tools for the exercise of the general coordinative function of entrepreneurship as it is by measurable increases in the productivity of specific factors of production: labor. One of the fundamental changes that information technology may be bringing about is enhancing the ability of market participants subjectively to orient themselves to one another and to their world. but may represent one of the main driving forces of wealth. We make ourselves wealthier by making ourselves better understood. the way people orient themselves to one another. Somehow a producer’s perspective of what possible production methods might work needs to come into some degree of coordination with a consumer’s perspective of what goods are most desirable. The way coworkers and market participants are able to share articulations with one another has been profoundly transformed. objective improvements in living conditions in the new economy. in a manner that may be highly suggestive for thinking about the transformations we are undergoing today. The theory of entrepreneurship ought to be taking these changes into account. something that sounds utterly subjective and mental. Computers and modern communications media have undoubtedly contributed to the productivity of labor and other factors. The claim of the knowledge management literature is that certain innovations in information technology are transforming the contemporary workplace and marketplace. In other words. This kind of orientative contribution of the tools would be difficult to measure in productivity studies. especially over the past decade. Creating and sharing electronic information over a network is a funda- mentally different communications medium from that of the creation and sharing of paper documents. and in particular to one another’s articulations. Entrepreneurship is fundamentally dependent on the media in which human communication takes place. etc. capital and land.

see Lewin (1999). There is no way that non-price language can convey the relative value of higher-order goods to entrepreneurs who are several stages away from those goods in the capital structure. all the way to the work of the contemporary American Austrians such as Kirzner. 2 The Austrian school has contributed to misunderstanding here. length. the writers and readers.42 Don Lavoie Notes The author would like to thank Bill Tulloh for editing this draft from a longer version of the paper. 10 As with capital. but I would insist that it can be very use- ful to think of market processes as a particular kind of linguistic communicative process whose similarities to verbal and written discourses are as important as the differences. It has been thought to be misleading in that it suggests that there would be a single dimension. a significant and growing part of the capital structure of the economy is in the form of software tools whose function is precisely the structuring of articulations. Two agents who talk to one another are not actually doing anything. in a textual structure everything is shaped by the meanings things have to the participants. whereas two agents who engage in a market transaction are accomplishing something. For a useful contemporary elaboration of Austrian capital theory. 11 There is more going on here than an analogy between capital and texts. whereas market exchange is a form of action. Friedrich Wieser called subjectivism a psychological method. Friedrich Hayek. and Boettke. But can we really maintain this distinction between communicating and acting? From the Austrian point of view a “merely verbal” exchange is every bit as much a matter of doing something as an exchange of ownership. in which to measure the extension of the capital structure. 7 Brice Wachterhauser (1999) puts the point this way in his recent book. Of particular importance is that prices make possible a process of economic calculation and the imputation of value appraisements across different stages of the structure of production. If taken seriously this psychologism would lead to the position Laslo Csontos (1998) calls methodological solipsism. through the writings in Europe of Ludwig Mises. I agree that there are important differences here. “The discovery and interpretation of profit opportunities: Culture and the Kirznerian entrepreneur” (1991b) I developed this point about the cultureless- ness of Kirzner’s notion of the entrepreneur. 1985). I do not want to slight such differences as these. 4 In my paper. 8 Böhm-Bawerk ([1884] 1959) deployed this term in his classic work on capital. see my “The progress of subjectivism” (1991a). 1979. in this case. and Ludwig Lachmann. See also Lavoie and Chamlee-Wright (2000). 3 See Kirzner (1973. 1985b). For an attempt to sort out the legitimate from the illegitimate meanings of the words subjective and objective. in particular they are altering property ownership and thus shifting the whole framework in which future action can take place. 6 The point is not to belittle the differences between markets and other communicative processes. and Mark Miller and Virgil Storr for helpful comments on the earlier draft. Rizzo. 5 One objection to this treatment is that ordinary discourse is merely talk. Garrison. 1 I refer to the contemporary Austrian school in the sense elaborated in Karen Vaughn’s book Austrian Economics in America. but the structure is nevertheless real. and there are lapses to psychologistic language in all the Austrian writings on subjectivism. and I have tried to elaborate the idea in my books on the critique of central planning (Lavoie 1985a. as did some other Austrians. As Baetjer’s book Software As Capital points out. spanning the period from the work of Carl Menger in the 1870s. a summary of the work of Hans-Georg Gadamer’s philosophical position. . 9 I get the idea of social learning if not the phrase from Hayek.

such as Martin. 21 When Tim Berners-Lee (1999). who could be called the inventor of the Web. links related concepts and areas of knowledge to allow a problem to be viewed from many angles. 20 The Miller et al. they are more practitioners than academics. when they do talk about contemporary technological media. He draws attention not only to intertextual structures of text. Henri-Jean Martin. The literature on the history of writing has said little explicitly about the impact of writing on economic processes per se. 14 These authors are taking up the same questions the historians of writing were raising about the differences between oral and written culture. 13 See Lavoie 2001. and the rise of literacy have transformed human society.strassmann. describes the early inspiration for the hypertext markup language. 15 See. . Jack Goody. and the neglect goes both ways. he is emphasizing this convergence. 18 See also Eric Drexler (1991) for a summary of the implications for scientific work of a full hypertext publishing system along the lines of Nelson’s vision. but also the interpersonal advantages of sharing structured text: Hypertext . and Roger Chartier on the way the evolution of writing. the printing press. Sloan Management Review. Elizabeth Eisenstein. This literature has been largely neglected by economists. although often it is easy enough to see the implications of broader changes on economic phenomena. for example. they are not historians. Like many of the writers in the knowledge management literature whom I will discuss in the next section. Brynjolfsson’s often cited paper “The productivity paradox of information technology: review and assessment” (1993). coauthors I have cited here were among the developers who were trying to implement Nelson’s vision. 22 Or we duplicate copies of information into different filing locations in order to improve the odds of finding what we are looking for. 16 The book includes a set of influential articles that were published in the 1990s in some of the leading organizational journals. 19 Thus when Nonaka (1994: 35) coins the term the “hypertext organization. 17 Several of the authors discussed earlier who have elaborated on the history of writing. Subjectivism and entrepreneurship 43 12 I will be drawing from scholars such as Walter Ong. The core feature of the hypertext organization is the ability to switch between various “contexts” of knowledge creation to accommodate changing requirements from situations both inside and outside the organization. this is analogous to the ability of individuals to relate stories in different ways according to the nature of the audience. “Subjective orientation and objective wealth: Entrepreneurship and the evolution of the structure of ‘text’” for a more in-depth treatment of these issues. he places emphasis on the needs he and others at CERN had to try to facilitate collaborative work. but that leads to other serious problems. and California Management Review. such as Organizational Science. making it easier to draw relationships between different sets of information. however. The same knowledge might be used but in different formats.” trying to draw out aspects of hypertextual organization of text that have analogies in the organization of organizations. See also the articles and books at the knowledge management web site: http://www. . In many ways. . in trying to maintain changes in two different They are a group of software engineers who have worked extensively on developing groupware and hypertext capabilities. use the language of information and data processing.

CA: Stanford University Press. Princeton. NJ: Princeton University Press. L. Sausalito. E. CA: IEEE Computer Society. (1995) Forms and Meanings. J. [1987] (1993) The Interface Between the Written and the Oral. (1993) “The productivity paradox of information technology: Review and assessment. L.H. (1998) “Organizing knowledge.” Social Intelligence 1(2). Stanford. Kirzner. Böhm-Bawerk.M.S. Decentering the Subject in Fiction. S.) Knowledge. (1956) Capital and Its Structure. E. V. and Libraries in Europe between the Fourteenth and Eighteenth Centuries. D. Engelbart.” California Management Review 40(3) (reprinted in D. Drexler. Opportunity.44 Don Lavoie References Baetjer. Brand. (2000) The Social Life of Information. and Electronic Media. Kansas City: Sheed. IL: Libertarian Press. Chicago: University of Chicago Press.M.S. and Duguid. Cochrane. 1912] (1959) Capital and Interest.” Economica 18: 412–27 (reprinted in L. Philadelphia: University of Pennsylvania Press. Los Alamitos. generator. Smith (ed. Chicago: University of Chicago Press.” in R.” Communications of the ACM 36(12): 66–77. Lachmann. I. von [1884. New York: Cambridge University Press. trans. (1998) “Subjectivism and ideal types: Lachmann and the methodological legacy of Max Weber. L.C.M. and Fischetti. M. Mongiovi (eds) Subjectivism and Economic Analysis. the Visual Arts. R. . and the Market Process: Essays on the Theory of the Market Economy. (1999) Weaving the Web: The Original Design and Ultimate Destiny of the World Wide Web.G. and retriever of information. Eisenstein. R. (1979) Perception. 1990). and Audiences from Codex to Computer. (1998) Software As Capital: An Economic Perspective on Software Engineering. Film. Kirzner. J. Communications and Cultural Transformations in Early-Modern Europe. New York: Routledge. Philadelphia: University of Pennsylvania Press. South Holland. and the Internet. (1997) From Text to Hypertext. Koppl and G. E. (1994) How Buildings Learn: What Happens After They’re Built. Brown. Bush.” American Documentation 12(2) (April). and Duguid.” Atlantic Monthly 176(1): 641–9 (reprinted in T. Nelson.M. Expectations. Brown. (1961) “Special considerations of the individual as a user. Csontos. P. I. J. Goody.) (1989) The Culture of Print: Power and Uses of Print in Early Modern Europe. Kirzner. Performances. (1985a) Discovery and the Capitalist Process. T. (ed. New York: Penguin. R. S. Berners-Lee. New York: HarperBusiness. H. Literary Machines. Authors. (1945) “As we may think. 1889. (1973) Competition and Entrepreneurship. K. Capital. Lachmann. 1977). Chicago: University of Chicago Press. (1979) The Printing Press as an Agent of Change. Brynjolfsson. 2000).E. Cambridge. MA: Harvard Business School Press. London: Bell and Sons. P. New York: Cambridge University Press. Chartier. I. Boston: Butterworth-Heinemann. L. (1994) The Order of Books: Readers. Chartier. Andrews and McMeel. Lachmann. and Profit: Studies in the Theory of Entrepreneurship. Chartier. (1951) “The science of human action. (1991) “Hypertext and the evolution of knowledge. Groupware. Gaggi.L.M. Texts. CA: Mindful Press.

Subjectivism and entrepreneurship 45 Lachmann. Smith. Cambridge. 3rd edn. D. UK: Edward Elgar.) (2000) Knowledge. (ed. Lachmann. NY: Cornell University Press.P. H.’” working paper. (1982) Orality and Literacy: The Technologizing of the World. Groupware. edition 90. Ong.J. Boston: Butterworth-Heinemann. Cochrane. G. George Mason University. and Chamlee-Wright. and Dreyfus.I. and the Internet. and Morality of Business. K. New York: Wiley. L. Mises. (1985a) Rivalry and Central Planning: The Socialist Calculation Debate Reconsidered. revised. (1999) Beyond Being: Gadamer’s Post-Platonic Hermeneutical Ontology. Sausalito. B. (1995) “The open society and its media. E. D.H. Lucas. Martin. [1974] (1987) Computer Lib/Dream Machines.M. Ithaca. Lavoie.) Knowledge. (1991b) “The discovery and interpretation of profit opportunities: Culture and the Kirznerian entrepreneur. into French as Esprit d’Entreprise. Lavoie.G. Baltimore: Johns Hopkins University Press.R.J. New York: Routledge. Evanston. New York: Routledge. J. (1994) Hyper/Text/Theory.. Berkeley: Glendessary Press. Tribble. Spinosa. (1985b) National Economic Planning: What Is Left?. (1977) Interfaces of the Word: Studies in the Evolution of Consciousness and Culture. Blaug and N. (1998) The Information Paradox: Realizing the Business Benefits of Information Technology. and the Internet. Boston: Butterworth- Heinemann. de Marchi (eds) Appraising Modern Economics: Studies in the Methodology of Scientific Research Programmes. T. New York: Cambridge University Press. Cambridge. trans. L. pp. Lavoie. 470–86. Lewis and M. (1971) The Legacy of Max Weber. New York: McGraw-Hill. Smith (ed. Nonaka. W.E. I. F. Nelson. San Francisco: Institute for Contemporary Studies (trans. and the Cultivation of Solidarity.C. Miller.. Lavoie. H. (1993) Austrian Economics in America: The Migration of a Tradition. Chicago: Henry Regnery Company. Pandya. H. (2000) Culture and Enterprise: The Development. D. Berger (ed. (1999) Information Technology and the Productivity Paradox: Assessing the Value of Investing in IT.H. . Paris: Maxima.J. T. L. School of Public Policy. WA: Microsoft Press. (1997) Disclosing New Worlds: Entrepreneurship. Ong. Lewin. (2001) “Subjective orientation and objective wealth: Entrepreneurship and the evolution of the structure of ‘text. London: Routledge. Democratic Action. P. E. Flores. Representation. D. Groupware.) The Culture of Entrepreneurship. D. London: Routledge. Lavoie. (1994) Expectations and the Meaning of Institutions. D. Wachterhauser. and Stiegler. Lavoie. M. (1994) “A dynamic theory of organizational knowledge creation. (1999) Capital in Disequilibrium: The Role of Capital in a Changing World.1. edited by D.” Organizational Science 5(1) (reprinted in D.” in J. D. New York: Cambridge University Press. Redmond. Cheltenham.M. Nelson.” in M. 1994). MA: MIT Press. [1980] (1990) Literary Machines.. Krummenacker (eds) Prospects in Nanotechnology: Towards Molecular Manufacturing. CA: Mindful Press. Landow. (1991a) “The progress of subjectivism. Jr. M. C.. Cultures et Societes. MA: Ballinger. W. Chicago: University of Chicago Press. Lavoie. IL: Northwestern University Press. Thorp. R. von [1949] (1966) Human Action: A Treatise on Economics.S..D.L. L. New York: Oxford University Press. Vaughn.” in B. 2000). (1995) The History and Power of Writing.

” in D.H.) Knowledge.L. Groupware. Boston: Butterworth-Heinemann. and the Internet. M. Boston: Butterworth- Heinemann. Smith (ed. and Serino. Zack. Smith (ed. (2000) “Knowledge management and collaboration technolo- gies” in D.46 Don Lavoie Zack. M. and the Internet.) Knowledge. Groupware. . M.H. (2000) “Social context and interaction in ongoing computer-supported management groups. and McKenny. J.

To give but a few recent statistics on the phenomenon – although these figures are subject to frequent fluctuations – as of March 2001 the top web server is Apache. it is seldom acknowledged that there is a complementary. Chapter 1. Sendmail and Linux. and to an increase in the thickness of most markets (e.76 per cent of the market. For instance. For instance. (Coase 1937: 397.1 and 2. but the Counter also estimates the worldwide Linux .549 registered users and 95. Dolan and Meredith 2001)>). Cox and Alm 1998. In such cases. aspect of this technological innovation: its software counterpart (some exceptions are Baetjer 1998 and Lavoie. the one generated by open source software development. as of 30 April 2002 there are 125. Open source includes such software success stories as And according to the Linux Counter. Perl.834 registered machines (<http://counter. this volume). This work attempts to do some justice to this shortcoming by describing some elements of a new type of organization of work.37 per cent of all active sites (see Tables 2. and at least equally important. (von Mises 1981 [1933]: 43) It should be noted that most inventions will change both the costs of organizing and the costs of using the price mechanism. for a total of 64. leading to an increase in the modular nature of most products and organizations.2 Open source software and the economics of organization Giampaolo Garzarelli [T]he productivity of social cooperation surpasses in every respect the sum total of the production of isolated individuals. if the telephone reduces the costs of using the price mechanism more than it reduces the costs of organizing. consumers are said to be increasingly participant in the production process. then it will have the effect of reducing the size of the firm. note 3) Introduction It is often remarked that innovation in computer technology is profoundly affecting the organization of production and of consumption of contemporary society. whether the invention tends to make firms larger or smaller will depend on the relative effect on these two sets of costs.2). with 53. However.

Microsoft is the sum of sites running Microsoft-Internet-Information-Server. Q4.g.netcraft. Microsoft-IIS. Microsoft is the sum of sites running Microsoft-Internet-Information-Server.1 Top developers Developer February 2002 Per cent March 2002 Per cent Change Apache 22.14 29. Table 2. Rosenberg 2000). DiBona et al.968>).76 –4.147.3 Top Linux server vendors Vendor Market share % Compaq 25 IBM 10 HP 7 Dell 7 Fujitsu Siemens 3 Others 48 Source: IDC. Netscape decided to develop an open source browser.33 –0.3).82 265. Netscape-Communications. Netsite-Commerce and Netsite-Communications.15 0. Microsoft-PWS-95. Netscape-Commerce.43 20.857 2. Netscape-FastTrack.67 Microsoft 11. Microsoft-IIS-W. Release 1.198. Microsoft-IIS-W. IBM adopted Apache as a web server for its Websphere product line.01 Source: Netcraft (<http://www.193 26. while . and Microsoft-PWS. Netscape-Communications. 1999. Netsite-Commerce and Netsite-Communications. Note: iPlanet is the sum of sites running iPlanet-Enterprise.06 Source: Netcraft (<http://www. Netscape-Commerce.. 1999 unit shipments.81 Microsoft 4.81 0.59 Zeus 837.225 1.13 12. Table 2.0 1998: 8ff. in 1998.1 The influence of open source both as a business and as a software development model has been vast (e. It is also interesting to notice that some of the most successful PC makers have recently also become some of the top Linux server vendors (see Table 2.2 Active sites Developer February 2002 Per cent March 2002 Per cent Change Apache 10.743 26.023 1.24>). Netscape-FastTrack.701 2.92 889.18 9. Mozilla. Note: iPlanet is the sum of sites running iPlanet-Enterprise.02 4.67 iPlanet 283.968.103 2.37 –0. Microsoft-PWS-95.18 855.069.522. For example. users to be 18 million (<http://counter.402 65. and Microsoft-PWS. Netscape-Enterprise. 2000. Microsoft-IIS.netcraft.48 Giampaolo Garzarelli Table 2.088 53.80 –0.14 3. cited in West and Dedrick (2001: Table 2). iPlanet 1.492.02 Zeus 177.112 1.966.462.954 64.php>).777 58.123.826 1.860 34.

it is first of all (and primarily) suggested that the organizational economics of open source software development is so complex that a theory that has the ambition to explain it needs to begin by looking at the nature of the knowledge involved in the production and consumption of open source software itself. innovation usually proceeds at a faster pace than most other industries because numerous individuals simultaneously try multiple approaches to solve the same problem. But the multiple approaches and the numerous individuals also create a great variety of potential improvements. Clearly. What renders all this possible and at the same time makes software so supple is its peculiar nature. Perhaps the most interesting point that shall emerge is that this ‘atypical’ organization is at least as good as a firm in sharing rich types of information in real time.H. innovation is not discrete and monolithic. such process is imbued with uncertainty. Brown Associates. and (b) that software itself is a capital structure embodying knowledge. To this end. but also the non-rivalrous network products being shared. its modularity. namely. software development is very complicated and involves a substantial amount of experimentation and trial-and-error learning. That is to say that the approach gives solid foundations to the eclectic organizational theory that this exploratory essay proposes. the pages that follow will attempt to describe how this organization can exist. Contextualization In general. Open source software 49 Apple ships Apache along with their operating system. thus increasing the possibilities for success. In addition.2 is looking into the possibility of going open source in some products (perhaps because of the pressures originating from the ongoing antitrust litigation) by launching so-called shared source: seemingly. Modularity is one method to manage . in this regard.4 The main advantage of following such a cognitive approach is that it lends itself well to explaining the self-organizing as well as the self-regulating properties of open source economic organization. seen by many open sourcers as the ultimate enemy. 2001). This renders it a cumulative process where improvements are incremental rather than radical. In turn. but often builds on previous software. By drawing on the recently developed organizational theory of professions and on the classic theory of clubs. And Microsoft. Inc. I go so far as to suggest that the distinction between input (knowledge) and output (software) is in some ways amorphous because knowledge and software are not only the common (spontaneous) standards. arguably more than any single individual.3 The open source philosophy assures a ‘self-correcting spontaneous’ organization of work that is ‘more elaborate and efficient than any amount of central planning could have achieved’ (Raymond 2001: 52). the variety leads to new problems and to new trial-and-error learning. Indeed.5 I submit that the two reasons for why this is so are (a) that constituents have symmetry of absorptive capacity. pharmaceuticals. for example. a hybrid of proprietary and open software (D. Contrary to.

. shared source code. emphasis removed) . allowing productive knowledge to converge to its most valued use. Conversely. The source must be available for redistribution without restriction and without charge. it is not rare for an open source project to be terminated in the absence of a meritocratic management structure organizing the development process. . especially more modern ones of the object-oriented type. Open source is a term that has recently gained currency as a way to describe the tradition of open standards. Open source software development.50 Giampaolo Garzarelli complexity. and the license must permit the creation of modifications and derivative works. and Python languages. a company that publishes many books on open source. this does not at the same time necessarily imply that all open source software projects lack a sometimes even rigid organizational structure. including Bind (The Berkeley Internet Name Daemon servers that run the Domain Name System). 1985: 260). are per se composed of different. compatible. [But] open source (which is a trademark of the Open Source Initiative – see <http://www. it is also apparently not rare for a very interesting project to fail to create momentum because of organizational rigidities. and collaborative devel- opment behind software such as the Linux and FreeBSD operating systems. Originally introduced by Parnas (1972). offers a concise definition. information hiding stimulates the division and specialization of knowledge. and many other programs. This is possible because through modularization a program hides information among modules while at the same time allowing for their communication – this principle is known as information hiding.6 According to this principle in fact ‘system details that are likely to change independently should be the secrets of separate modules. The decision of what software to develop. is practically based on the absence of hierarchy. the Sendmail mail server. Consequently. corporate approach to software development is centred on hierarchical relations. And all this entails that the only benchmark to assess the ‘efficiency’ of a particular software is not so much its ability to perform its tasks as its ability to evolve in order potentially to perform its tasks even better (Baetjer 1998). and much of the Internet infrastructure. means more than the source code is available. (O’Reilly 1999: 33–4. Tcl.8 Apparently.7 But as others have pointed out. and reusable. the only assumptions that should appear in the interfaces between modules are those that are considered unlikely to change’ (Parnas et al. in contrast. and it is possible to change a part of a module or an entire module without knowing all information about the program that the module belongs to and without altering other modules or the overall purpose of the program (Baetjer 1998). What exactly is open source software then? Tim O’Reilly. test or improve comes from the top of the hierarchy. and must allow those derivatives to be redistributed under the same terms as the original>). founder and CEO of O’Reilly & Associates. Programs. information hiding assures that soft- ware is extendable. .opensource. the Perl. The traditional. the Apache Web server. interacting modules.

11 Rather. Accordingly you release often in order to get more corrections. you assume that bugs are generally shallow phenomena – or. The keyword here is competence. In the cathedral-builder view of programming. Thus the long release intervals. and. that they turn shallow pretty quickly when exposed to a thousand eager co-developers pounding on every single new release. and as a beneficial side effect you have less to lose if an occasional botch gets out the door. lies the core difference underlying the cathedral-builder and bazaar styles. Eric Raymond. Or. at least. and the inevitable disappointment when long-awaited releases are not perfect. . emphasis removed) Let us try to identify some of the necessary ingredients for an organizational theory of bazaar-style software development.9 The origins of open source go back to the so-called hacker culture. in fact. bugs and development problems are tricky. the participation to open source projects is voluntary (there is strong self- selection) and supervision is assured on a peer review basis. it dematerialized the need for concentration of hackers in specific laboratories. almost every problem will be characterized quickly and the fix obvious to someone. moving their concentration to cyberspace. ‘Given enough eyeballs. Given a large enough beta-tester and co-developer base. In effect. it is a complex . performing some manipulations f (•).L).’ I dub this: ‘Linus’s Law’. deep phenomena. Stanford and Carnegie-Mellon). summarizes the fundamental philosophy underlying the open source community in the context of his discussion of Linux. In Linus’s Law . Open source software 51 Notably. The attributes of professions Deborah Savage. 10 Hackers are very creative software developers who believe in the unconditional sharing of software code and in mutual help. all bugs are shallow. (Raymond 2001: 30–1. production is not as simple as making a soup: it is not so much a question of putting some inputs together (K. . less formally. voilà. proposes the following economic definition of a profession: a ‘profession is a network of strategic alliances across owner- ship boundaries among practitioners who share a core competence’ (Savage 1994: 131). It takes months of scrutiny by a dedicated few to develop confidence that you’ve winkled them all out. on the other hand. in an innovative piece. and it multiplied digital linkages. obtaining some output X. insidious. In the bazaar view. hacker and author of the very influential open source ‘manifesto’ The Cathedral and the Bazaar (2001). The advent of the microcomputer diffused this ethos beyond the narrow confines of the academic environments where it originally developed (MIT. As the literature beginning with the resuscitated contributions by Penrose (1995 [1959]) and Richardson (1998: chapter 10) has made clear.

To change viewpoint on the matter.g. a fortiori. • Professionals are autonomous and authoritative in their fields for their competencies allow them. semiconductors. and it provides for their cohesion and coherence – in a way. the point is that the capabilities share a rudimentary (esoteric) knowledge base – the core competence noted above – that affords the absorptive capacity.13 In the specific case of professions. These characteristics do not necessarily mean. for coordination beyond mere price and quantity (compare Malmgren 1961). because integration of ownership is not a conditio sine qua non. Rather. for Hayekian (qualitative) coordination. we have ‘esoteric knowledge’ (Savage 1994: 135–6). – around which other capabilities and routines evolve and revolve. In sum. In brief.e. software. then. some similar capabilities – is virtually always present. it manages to couple competencies. It therefore offers the rationale for the existence of professions.12 The capabilities involved in producing goods and services are often based on tacit knowledge in the sense of Michael Polanyi (e. professionals and. reputation. absence of ownership structure16 – as well as its external economies. ‘to solve routine problems easily . etc. absent esoteric knowledge. This knowledge represents the elemental component of professions. a cognitive-based process encapsulating different routines and capabilities evolving through time (Nelson and Winter 1982). we have a knowledge that is highly tacit and specialized or. The most germane implications for our purposes seem to be the following: • The theory allows to define narrowly the area of operation of a profession because of its emphasis on core competencies – for example.52 Giampaolo Garzarelli process involving abilities. their coupling would not exist.14 An important corollary is that the fairly symmetric nature of capabilities present in professions assures that the ‘absorptive capacity’ (Cohen and Levinthal 1990) – i. most important for our discussion (as we shall see shortly). and learning – it is. for their boundaries as well. the general organizational implications of Savage’s theory of professions are considerable. incentive alignment. 1966). trust and. and so on. to coordinate. professional capabilities are a form of capital representing the productive essence of the network. however. what boils down to the same thing. on the one hand. • It allows to distinguish professions from other forms of organization. notably. such as firms. experience. For example. mutual monitoring. that is. and more generally coevolving with the network itself. pharmaceuticals. that each professional has exactly the same capabilities. to put it in more general terms. to use a catch-all wording. We apparently face a situation where the division of knowledge (Hayek 1948: chapter 2) drives the division of labour. 15 Indeed. a necessary condition for spontaneous self- organization. such as uncertainty reduction. it is the spontaneous orchestration of knowledge generated by the symmetry of absorptive capacity that creates a profession’s complex self-organization – with. otherwise we would not be in the presence of complex self-organization. Professions – like most other organizational forms – then exist for epistemic reasons or.

Open source software 53
and non-routine problems routinely’ (Savage 1994: 140) and, on the other,
enable them to evaluate, and only be challenged by, other professionals. More
concretely, they are independent yet interact in a coordinated and fertile
• Professions are decentralized networks in that there is not a central authority
in command.17 The ‘organization’ of a profession is guaranteed by the
exchange of knowledge that reduces uncertainty and stimulates trust among
members. Professions are thus self-organizing.
• Relatedly, there is the role played by reputation as a signalling of quality, viz.,
reputation is a positive externality. Thus, professions can be interpreted as
self-regulating organizations (a point we shall return to below).

The organizational workings of professions seem to well approximate, I think,
some of the characteristics of the bazaar-style market for ideas that Raymond
(2001) depicts in his descriptive analysis of open source. Similarly to open source,
in fact, a profession is a capital network investing in network capital. Interestingly,
the workings also seem to accord with the observation by Lerner and Tirole (2000:
8–9) that the core of the entire open source community seems to lie in the
sophisticated individuals who compose it.
Yet, the theory of professions allows us to illuminate mostly one facet of open
source: the supply-side. In order to offer a more complete story of open source,
we need to extend the theoretical framework of professions to incorporate more
explicitly demand-side considerations. We also need, moreover, to endogenize
technology. To this end, it seems necessary to bring together the theory of
professions with that of clubs, and to consider the role played by technology.

The additional dynamics of a club
In a seminal article published in 1965, ‘An economic theory of clubs’, Buchanan
described and formalized the institutional properties of a new category of good
(or product) lying between the public and private polar extremes, conventionally
called shared good. The good is usually enjoyed only by members participating in a
voluntary association – i.e. a club – whose membership may be regulated by some
dues. The theory of clubs, in a nutshell, studies the different institutional
arrangements governing the supply and demand of the shared good.
Since then, the vast literature on clubs has mostly devoted itself to the study
of positive and normative issues at the macro-level – for example, decentralization
of government and fiscal federalism. But there have also been a few studies
concerned with the firm. In particular, Antonelli and Foray (1992) propose a theory
of technological cooperation among firms called ‘technological clubs’. By means
of a simple comparative static model, they suggest that firms will cooperate in
technological endeavours only if the benefits of cooperation outweigh the costs.
This is the traditional result we would expect under familiar club models, where
the amount of shared good decreases as the number of users increases (cf. Buchanan
1965: 2–6).

54 Giampaolo Garzarelli
But, interestingly, Antonelli and Foray also underline that this logic is reversed
in the case of network products, namely, when ‘the performance of the product as well
as its utility increases with the increase of the community of users’ (Antonelli
and Foray 1992: 40). If there are, for example, possible network effects generated
by the output, by the process of production or by the technology of production
(or all of these), familiar exclusion/congestion effects caused by increased club
membership may not hold.
Because of the necessary standardization that a network product requires, the
possible exclusion/congestion effects generated by increased membership may
be overcome by ‘the overall growth of the aggregate demand for the production
induced by network [effects]’. Therefore, ‘the trade-off of the technological
cooperation is reversed and now [a] firm [may choose whether] to enter a [tech-
nological club] and to standardize its own products according to the ratio between
marginal costs of standardization and the marginal revenues of standardization’
(Antonelli and Foray 1992: 43).
This begs the question of what the shared good is in our case. In the open source
world, the shared good seems to be more than one: the software as well as the
capabilities of production and of consumption. In the light of our discussion so far,
this claim should not be too surprising because, first, software per se is an ever-
evolving capital structure that embodies knowledge (Baetjer 1998) and, second,
because in the open source community both software and capabilities are non-
rivalrous (Raymond 2001).18 Indeed, if compared with proprietary software, open
source would seem to assure an even more productive capital structure because of
the free availability of the source code. Knowledge and software are then not only the common
(spontaneous) standards, but also the network products.19
Now, were we in the presence of a more traditional organizational structure –
such as one with a non-network product – we would have a congestion problem
arising from the difficulty of capability transfer (compare Ellig 2001). But because,
as we noted, for open source esoteric knowledge is in reality common, the
congestion is actually determined by the technological state of the art. (We shall
return to technology presently.)

Reputation and shared context
These observations lead to another interesting issue. Open sourcers, we saw, are
not in the trade to maximize profits. Although their first motivation to modify a
program may originate from sheer need,20 their utility functions for sharing, as
Raymond repeatedly emphasizes, exhibit maximization of reputation; that is,
attempting to deliver an ever better product maximizes reputation.21
Algebraically, we can illustrate the process in terms of quality improvements as

qt (S) = f (qt −1(S), Kt , Ht , Ct )

Open source software 55
where q(S) is the quality of software S, K is knowledge, H is the complementary
hardware, C is complementary software, and t is a time index. The utility (U)
function of the open sourcers is:

Ut (Sq ) = f (qt , Nt (Sq ))

where N is the number of users of software S, q is quality, and t is a time index.
The qualitative property that this trivial illustration is trying to convey is the
following. The endogeneity of reputation captured by the quality of software
increases the user base (positive externality) and the ‘utility’ of the open sourcers
on both sides of the demand and supply equivalence.
This implies that many traditional organizational stories centred on, for example,
incentive alignment, monitoring, opportunism, and ownership structures are,
at best, incomplete for they neglect true Marshallian external economies (or, if
you prefer, knowledge spillovers) that act as, for example, self-regulatory monitoring
and coordinating systems. The ‘poor beleaguered conventional manager is not
going to get any [succour] from the monitoring issue; the strongest argument the
open source community has is that decentralized peer review trumps all the
conventional methods for trying to ensure that details don’t get slipped’ (Raymond
2001: 59, original emphasis).
Interestingly, the discussion has brought us back to Hayek and to the problem
of knowledge and its dispersion (Hayek 1948: chapters 2 and 4; Jensen and
Meckling 1998 [1992]). That is to say, sometimes shared context may count more
than hierarchy for the ‘efficient’ organization of production and exchange (Ghoshal
et al. 1995).

On the role of technology
The reader will have noticed by now that I have not yet said much about
information and communication technology. We have talked about it in a standard
comparative static fashion. But I have been deliberately vague about its endogenous
role. This topic shall be briefly considered here.
Arguably, the role played in our story by information and communication
technology is one of ‘technological convergence’ (Rosenberg 1963). Or, to update
Rosenberg’s notion somewhat, hardware and software represent a ‘general purpose
technology’ (GPT) (e.g. Bresnahan and Traijtenberg 1995). Put simply, a GPT
usually emerges to solve very narrow problems. Yet, in time its purposes diffuse to
many other areas of application. For example, we have passed from the MIT PDP-
1 minicomputer in 1961 to the Defence Department’s Advanced Research Projects
(DARPA) that created ARPANET, the first computer network, to today’s personal
computer and the Internet.
It would seem that even though GPT has greatly facilitated collaboration among
a great variety and number of individuals, some of the economic interactions are
very similar to what other forms of organization, such as professions and clubs,

56 Giampaolo Garzarelli
already do. But thanks to online interactions in real time, it would also seem that
GPTs might ultimately give a comparative advantage to a profession/club mode
of organization over one of hierarchy. Indeed, a fundamental reason why classical
markets often do not work well resides in the need to share rich information in real
time (compare, e.g., Kogut and Zander 1992; Langlois and Robertson 1995).22
In the specific case of open source it appears that the transition to cyberspace
to share rich information in real time was so, as it were, smooth because there
already existed a more-or-less well-defined core competence and culture.

Policy considerations
The first point to underline is probably that open source spontaneously solves
the two fundamental organizational problems defined by Jensen and Meckling
(1998 [1992]: 103): ‘the rights assignment problem (determining who should
exercise a decision right), and the control or agency problem (how to ensure that
self-interested decision agents exercise their rights in a way that contributes to
the organizational objective)’. When specialized knowledge is symmetric, we
saw, it spontaneously solves the agency problem by means of external economies
(Savage 1994, 2000). And, just as Coase (1960) taught us, the ultimate result of
this spontaneous interaction is a ‘collocation’ of production (and consumption)
knowledge and decision power in the hands of those who most value it ( Jensen
and Meckling 1998 [1992]; Savage 2000).23 If this is so, then open source is not
only a self-organizing organization, but also a self-regulating one where, as Savage
(2002: 19) points out, ‘self-regulation . . . means coordination of economic activity
through voluntary association in an interdependent network, without interference
from the government, and without resort to hierarchy’.24
Whenever organizational forms present rapid change because of their strong
ties to technology, public policy issues are always thornier than usual. Indeed,
historically, it seems that every time that there is the development of a new
technology or production process, the government has to intervene in some fashion
to regulate it or to extract rents from it. This point is well-encapsulated in the
well-known catch-phrase attributed to Faraday. After Faraday was asked by a
politician the purpose of his recently discovered principle of magnetic induction in
1831, he replied: ‘Sir, I do not know what it is good for. However, of one thing
I am quite certain, some day you will tax it’.
Since open source successfully developed in an environment of little government
presence has generated benefits well beyond its organizational boundaries,25 the
implications for policy are quite clear. The government must be sensitive to
economic activity that is spontaneously productive, and one way to guarantee
this is to preserve spheres of autonomy.26 Indeed, any intervention may suffocate
the very motivation that drives these types of organization. At the same time,
however, this is not to say that we should not think about the possibility of defining
some Hayekian abstract rules for the interaction among new organizations as well
as among new and more traditional types of organization (see Savage 2002).27

Open source software 57
Conclusions and suggestions for future research
In a somewhat desultory fashion, I have attempted to describe some general
organizational characteristics of bazaar-style software development. During the
description a substantive lesson emerged. It appears that the emergence of
spontaneous organization of work is facilitated mostly in those cases where the
constituents at least to some degree already share productive knowledge or, if you
like, where knowledge is already a standard. In the case of open source this is
a fortiori so in that not only is the input (the esoteric knowledge) a sufficient statistic
because of common absorptive capacity, but also because the output (the software,
a network product) is itself essentially a standard-setting knowledge structure whose
use is vast and whose reorganization is infinite. Relatedly, this suggests that in our
story new information and communication technology – if now virtually
indispensable – mostly performed the role of propagator rather than that of
Without wanting to make too much of the point, we should also notice that
thanks to new technology the organizational economics of open source now seem
to be closer to the putting-out system.28 The critical organizational difference
between the putting-out and the bazaar being that in the latter there is no external
authority controlling production.29 That is to say that the organization of work of
open source is one where the labour force is dispersed but connected by means
of new technology, and whose product supervision is (spontaneously) assured
by reputation effects.30
The economics of open source is very complex. My analysis has scratched the
surface. Future studies should more closely scrutinize the relation between the
structure of software, namely, its modularity, vis-à-vis the organizational structure
delineated in the previous pages.31 Further, they could explore the impact of type
of licensing agreement on organization form, and study the relationship between
the legal quandaries linked to traditional software32 and possible implications for
open source and its organization. All these avenues of investigation would naturally
lead to the very interesting issue of the origin and evolution of organizational form33
as well as to the one about the trade-off between coherence and flexibility of

I thank Cristiano Antonelli, Metin M. Cosgel, Steven R. Cunningham, George N.
Dafermos, Roger Koppl, Jorge L. Laboy-Bruno, Yasmina Reem Limam, and Mark
S. Miller. My intellectual debt to Richard N. Langlois is beyond the few references to his
work. The paper was written while I was visiting the Economics Department of the
University of Connecticut, Storrs, USA.
1 To be more precise, ‘Linux’ refers to the kernel of the operating system; while the
entire project is called ‘GNU/Linux’ because it contains parts of the GNU project,
started by Richard Stallman in 1984. See, for example, the chapter by Stallman in
DiBona et al. (1999: 53–70).
2 See, for example, cWare (nd); but compare Eunice (1998).
3 For Microsoft’s reaction to the open source phenomenon see the ‘Halloween

58 Giampaolo Garzarelli
documents’ (some internal Microsoft memoranda that try to assess competition from
open source that later became public): <>
(accessed 10 February 2000).
4 Take note that by ‘consumption of open source’ I refer to consumption on the supply-
side: consumption by open source producers. I do not consider, in other words,
‘downstream’ consumption of open source, the one made by individuals using open
source who are not at the same time involved in its production; even if these
downstream consumers may suggest to the open source community about how to
improve software.
5 ‘Real time’ in the computer science sense of being able to do, evaluate or react to things
as they are happening, without (much) delay. Two classic examples of real time
behaviour are a telephone conversation and software that tries to constantly track
weather conditions to attempt to offer forecasts. For an organizational application of
this notion see Langlois and Robertson (1995: chapter 3).
6 But compare Brooks (1975: 78ff.).
7 In some cases there still is some authority, however. In the case of Linux, for example,
Linus Torvalds (or a close collaborator) decides which software code to accept into the
Linux kernel.
8 Notably George N. Dafermos in private communication with the author.
9 There are several licenses governing open source. Analyzing these in detail necessitates
a study of its own. See especially DiBona et al. (1999: appendix B), Rosenberg (2000:
chapters 6 and 7 and appendix A) and Raymond (2001: 73ff.).
10 See in particular Raymond (2001: 1–17, 169–91).
11 The image is Leijonhufvud’s (1986: 203).
12 To clarify, next to competence, the literature also speaks of ‘routines’ (Nelson and Winter
1982), ‘capabilities’ or ‘dynamic capabilities’ (Langlois and Robertson 1995). Stricto sensu,
routines are what an organization does, they are the economic equivalent of the
biological genes or economic memory; capabilities/dynamic capabilities are what an
organization can do, e.g. if circumstances change and redeployment of resources takes
place – they are directly complementary to competencies; competencies are the core
abilities that an organization possesses, i.e. what an organization specializes in depends
on its competence (although, in time, competencies may change). To schematize: routines
∈ capabilities ∈ competencies. Yet, these categories are not mutually exclusive as the
(illustrative) classification might suggest; in fact, all the notions are quite slippery.
13 Polanyi’s tacit dimension that is opposite to the explicit one, is akin, in many ways, to
Ryle’s (1971 [1946]) dichotomy between ‘knowledge that’ (explicit) and ‘knowledge
how’ (tacit); the distinction made by de Solla Price (1965) between technological (how)
and scientific (why) knowledge is also relevant here.
14 A point of view, incidentally, compatible with Coase’s (1937) original story; see, for
example, Langlois and Robertson (1995) and Garzarelli (2001).
15 Although not widely remarked upon (an exception is Langlois and Robertson [1995]),
this is, in effect, the flip-side of competencies, at least in normal periods of production
and exchange, i.e. those involving little radical innovation.
16 Contra Hansmann (1996).
17 See especially Langlois and Robertson (1995: chapter 5).
18 The suggested idea of sharing of capabilities shares some of the properties of user-based
innovation and user-to-user assistance described in Lakhani and von Hippel (2000).
19 On open standards and networks compare for example, Garud and Kumaraswamy
(1993) and West and Dedrick (2001).
20 Indeed, this is the first lesson offered by Raymond. ‘Every good work of software starts
by scratching a developer’s personal itch’ (Raymond 2001: 23, emphasis removed).
21 For example, the ‘ “utility function” Linux hackers are maximizing is not classically
economic, but is the intangible reward of their own ego satisfaction and reputation
among other hackers’ (Raymond 2001: 53).

Open source software 59
22 As Raymond (2001: 224, note 10) observes in a related context, the ‘open source
community, organization form and function match on many levels. The network is
everything and everywhere: not just the Internet, but the people doing the work form
a distributed, loosely coupled, peer-to-peer network which provides multiple
redundancy and degrades very gracefully. In both networks, each node is important
only to the extent that other nodes want to cooperate with it’.
23 Miller and Drexler (1988), in a classic essay that greatly influenced Raymond (2001:
225–6), make a similar point. Compare also Baetjer (1998).
24 Compare Raymond (2001: 57ff.).
25 The ‘greatest economic contribution’ of open source technologies ‘may be the new
services that they enable’ (O’Reilly 1999: 34). The ‘Red Hat Wealth Monitor’ keeps
track of the profits made by Red Hat thanks to open source software. This is an effort
undertaken to encourage reinvestment in the community in order to try to generate
even more wealth. Visit: <> (accessed 21
March 2000). See also Release 1.0 (1998); and compare Lavoie, Chapter 1, this volume.
26 For instance, ‘[w]eb computing fundamentally depends upon open access because
more contacts lead exponentially to more potential value creation. For example, Bob
Metcalfe, inventor of Ethernet technology, asserts [that] the value of any number of
interconnections – computers, phones, or even cars – potentially equals the square
of the number of connections made’ (D.H. Brown Associates, Inc. 2000: 8). This
generalization should be readily contrasted to ‘Brooks’s Law’ (Brooks 1975). As
Raymond acknowledges, in ‘The Mythical Man-Month, Fred Brooks observed that
programmer time is not fungible; adding developers to a late software project makes it
later. He argued that the complexity and communication costs of a project rise with the
square of the number of developers, while work only rises linearly. This claim has since
become known as “Brooks’s Law” and is widely regarded as a truism. But if Brooks’s
Law were the whole picture, Linux would be impossible’ (Raymond 2001: 49–50; see
also 220–1, note 4). Compare Langlois (2001).
27 In many ways, this is the classic problem of increasing complexity as division of labour
increases; but in a new guise. See especially Leijonhufvud (1989) and Baetjer (1998).
28 Compare Leijonhufvud (1986) and Langlois (1999).
29 Or, to put it more precisely, authority is still present; yet it is internal in the sense that
it is among peers, i.e. its legitimacy is mostly achieved by reputation.
30 This impression should be contrasted to more traditional comparative-static stories
assuming exogenous inputs and outputs – hence constant technology and no know-
ledge combinatorics or growth (e.g. Gurbaxani and Whang 1991; Malone et al. 1987;
Picot et al. 1996). These approaches are mostly influenced by the work of Williamson
(e.g. 1985) for both method (i.e. the trichotomy among market, hybrid and hierarchy)
and core variables (the transaction, asset specificity, contractual safeguard and oppor-
tunism); but compare also Williamson (1996).
31 Kogut and Turcanu (1999) emphasize the importance of modularity. For theories of
organizational and technological modularity see, for example, Garud and
Kumaraswamy (1993), Langlois and Robertson (1995: chapter 5) and Langlois (2001
and 2002).
32 On which compare Samuelson et al. (1994).
33 An issue that is relevant to the whole of this chapter. The Mozilla project by Netscape
Communication, Inc. – that released the code of Netscape Communicator on 31
March 1998 (the first proprietary code to become open) – would be an interesting case
study. See, for example, Rosenberg (2000: 33–8) and Raymond (2001: 61–3; 169–91).
Also see: <> (accessed
31 May 2001).

60 Giampaolo Garzarelli
Antonelli, C. and Foray, D. (1992) ‘The economics of technological clubs’, Economics of
Innovation and New Technology 2: 37–47.
Baetjer, H. Jr. (1998) Software as Capital: An Economic Perspective on Software Engineering, Los
Alamitos, CA: IEEE Computer Society.
Bresnahan, T.F. and Traijtenberg, M. (1995) ‘General purpose technologies: “engines of
growth”?’, Journal of Econometrics 65(1): 83–108.
Brooks, F.P. Jr. (1975) The Mythical Man-Month: Essays on Software Engineering, Reading, MA:
Buchanan, J.M. (1965) ‘An economic theory of clubs’, Economica, N.S. 32(125): 1–14.
Coase, R.H. (1937) ‘The nature of the firm’, Economica, N.S. 4(16): 386–405.
Coase, R.H. (1960) ‘The problem of social cost’, Journal of Law and Economics 3 (October):
Cohen, W.M. and Levinthal, D.A. (1990) ‘Absorptive capacity: a new perspective on
learning and innovation’, Administrative Science Quarterly 35: 128–52.
Cox, M.W. and Alm, R. (1998) The Right Stuff: America’s Move to Mass Customization. Federal
Reserve Bank of Dallas Annual Report. Online. Available HTTP:
<> (accessed 14 April 2001).
cWare (nd) The Linux Storm. Online. Available HTTP:
<> (accessed 19 March 2000).
D.H. Brown Associates, Inc. (2000) Technology Trends Monthly, May newsletter. Online.
Available HTTP: <> (accessed 8 July 2000).
D.H. Brown Associates, Inc. (2001) Technology Trends Monthly, June newsletter. Online.
Available HTTP: <> (accessed 9 August 2001).
de Solla Price, D.J. (1965) ‘Is technology historically independent of science? A study in
statistical historiography’, Technology and Culture 6(4): 553–67.
DiBona, Ch., Ockman, S., and Stone, M. (eds) (1999) Open Sources: Voices from the Open Source
Revolution, Sebastopol, CA: O’Reilly & Associates, Inc. Also online. Available HTTP:
<> (accessed 21 July
Dolan, K.A. and Meredith, R. (2001) ‘Ghost cars, ghost brands’, Forbes (April 30). Online.
Available HTTP:
TRQFIAGWCFFA> (accessed 7 May 2001).
Ellig, J. (2001) ‘Internal markets and the theory of the firm’, Managerial and Decision Economics
22(4–5): 227–37.
Eunice, J. (1998) Beyond the Cathedral, Beyond the Bazaar (May 11). Online. Available
HTTP: <>(accessed
19 March 2000).
Garud, R. and Kumaraswamy, A. (1993) ‘Changing competitive dynamics in network
industries: an exploration of Sun Microsystems’ open systems strategies’, Strategic
Management Journal 14: 351–69.
Garzarelli, G. (2001) ‘Are firms market equilibria in Hayek’s sense?’, Centro di
Metodologia delle Scienze Sociali working paper No. 74, Rome: Luiss-Guido Carli.
Ghoshal, S., Moran, P., and Almeida-Costa, L. (1995) ‘The essence of the mega-
corporation: shared context, not structural hierarchy’, Journal of Institutional and
Theoretical Economics 151(4): 748–59.
Gurbaxani, V. and Whang, S. (1991) ‘The impact of information systems on organizations
and markets’, Communications of the ACM 34(1): 59–73.

Open source software 61
Hansmann, H. (1996) The Ownership of Enterprise, Cambridge, MA: The Belknap Press of
Harvard University Press.
Hayek, F.A. von (1948) Individualism and Economic Order, Chicago: University of Chicago
Jensen, M.C. and Meckling, W.H. (1998) ‘Specific and general knowledge and
organizational structure’, in M.C. Jensen (ed.) Foundations of Organizational Strategy,
Cambridge, MA: Harvard University Press. Originally published in W. Lard and
H. Wijkander (eds) (1992) Contract Economics, Oxford: Blackwell.
Kogut, B. and Turcanu, A. (1999) ‘The emergence of e-innovation: insights from open
source software development,’ Wharton School paper, University of Pennsylvania
(November 15). Online. Available HTTP:
<>. (accessed 25 February
Kogut, B. and Zander, U. (1992) ‘Knowledge of the firm, combinative capabilities, and the
replication of technology’, Organization Science 3 (3): 383–97.
Lakhani, K. and Hippel, E. von (2000) ‘How open source software works: “free” user-to-
user assistance’, MIT Sloan School of Management working paper 4117 (May).
Langlois, R.N. (1999) ‘The coevolution of technology and organization in the transition to
the factory system’, in P.L. Robertson (ed.) Authority and Control in Modern Industry:
Theoretical and Empirical Perspectives, London: Routledge.
Langlois, R.N. (2001) The Vanishing Hand: The Changing Dynamics of Industrial Capitalism
(August 7), Centre for Institutions, Organizations, and Markets working paper
2001–01, University of Connecticut, Storrs. Online. Available HTTP:
<> (accessed 12 August 2001).
Langlois, R.N. (2002) ‘Modularity in technology and organization’, Journal of Economic
Behavior and Organization 49(1): 19–37.
Langlois, R.N. and Robertson, P.L. (1995) Firms, Markets, and Economic Change: A Dynamic
Theory of Business Institutions, London: Routledge.
Leijonhufvud, A. (1986) ‘Capitalism and the factory system’, in R.N. Langlois (ed.)
Economics as a Process: Essays in the New Institutional Economics, New York: Cambridge
University Press.
Leijonhufvud, A. (1989) ‘Information costs and the division of labour’, International Social
Science Journal 120 (May): 165–76.
Lerner, J. and Tirole, J. (2000) ‘The simple economics of open source’, NBER working
paper 7600 (March), Cambridge, MA: NBER.
Malmgren, H.B. (1961) ‘Information, expectations and the theory of the firm’, Quarterly
Journal of Economics 75(3): 399–421.
Malone, Th.W., Yates, J., and Benjamin, R.I. (1987) ‘Electronic markets and electronic
hierarchies’, Communications of the ACM 30(6): 484–97.
Miller, M.S. and Drexler, K.E. (1988) ‘Markets and computation: agoric open systems’,
in B.A. Huberman (ed.) The Ecology of Computation, Amsterdam: North-Holland. Also
online. Available HTTP:
<> (accessed 5 June 2001).
Mises, L. von (1981). Epistemological Problems of Economics, New York and London: New York
University Press (originally published 1933).
Nelson, R.R. and Winter, S.G. (1982) An Evolutionary Theory of Economic Change, Cambridge,
MA: The Belknap Press of Harvard University Press.
O’Reilly, T. (1999) ‘Lessons from open source software development’, Communications of the
ACM 41(4): 33–7.

. Williamson. R. (1998) The Economics of Imperfect Knowledge: Collected Papers of G.gsm. Samuelson. P. (1963) ‘Technological change in the machine tool industry.tuxedo. (1971) ‘Knowing how and knowing that’. and Weiss. M. Inc. CA: M and T Books.H. (1972) ‘On the criteria to be used in decomposing systems into modules’.. N. Sebastopol. (1985) ‘The modular structure of complex systems’.A. 19 November).cfm> (accessed 19 April 2002). Copenhagen.) Collected Papers. 1929–1968.edventure. Williamson. Business and Economic History 23(2): 129–60.A. D. G. O.. Ryle.E.M. IEEE.L. Columbia Law Review 94 (December): 2308–431. 2000). open standards in the network era: an examination of the Linux phenomenon’. Polanyi.62 Giampaolo Garzarelli Parnas. J. West. (1996) ‘The fading boundaries of the firm: comment’. (1994) ‘A manifesto concerning the legal protection of computer programs’. Raymond. New York: The Free Press. The Open-source Revolution (1998. B. (ed. D. D.. Inc (originally published in Proceedings of the Aristotelian Society 46 (1946)). E. UK: Edward Elgar Publishing. (1994) ‘The professions in theory and history: the case of pharmacy’. E. (1996) ‘The fading boundaries of the firm: the role of information and communication technology’.pdf> (accessed 5 June 2001). (2001) ‘Proprietary vs.E. (1995) The Theory of the Growth of the Firm. D. in Ryle. Richardson..K. 1840–1910’. and Dedrick. Journal of Economic History 23(4): 414–43. Online. Available HTTP: <http://www. Rosenberg. CA: O’Reilly & Associates. Penrose. D. Journal of Institutional and Theoretical Economics 152(1): 65–79.0. (1966) The Tacit Dimension.. Rosenberg. Savage. Proceedings of the Fourth Triple Helix Conference. P. Foster City. M. Ripperger. revised edn. London: Routledge. J. (1985) The Economic Institutions of Capitalism. (2000) Open Source: The Unauthorized White Papers.uci. Volume II: Collected Essays. Clemens. New York: Barnes and Noble.S. Available HTTP: <http://www.B. now available in PDF format from HTTP: <http://release1.. Musings on Linux and Open Source by an Accidental Revolutionary. Davis. and Wolff. Communications of the ACM 15(12): 1053–8.L. O. Richardson. Picot. (2002) ‘How networks of professions regulate themselves’.D. Kapor. . J. Cheltenham.C. and Reichman. G.B.T. Oxford: Oxford University Press (originally published in 1959). with a new Foreword by the author. A. Also online. Dead HTTP: <http://www. IEEE Transactions on Software Engineering 11(3): 259–66. Proceedings of the Hawaii International Conference on System Sciences (HICSS-34) (January 3–6). Release> (accessed 5 February 2000).edventure. D. (2001) The Cathedral and the Bazaar. Journal of Institutional and Theoretical Economics 152(1): 85–7.html> (accessed February 10. 3rd edn. Savage. T. November 6–9. Maui. Parnas. G.

despite their distance from your community. The recognition and sense of legitimacy in your local community of your claim to the house makes this asset effectively your property. In one study. A mortgage on that house would be capital. De Soto hopes to bring the power of capital to the poor by strengthening their property rights in accord with historical precedents in the advanced economies – that is. they are nevertheless quite real. in The Mystery of Capital (de Soto 2000). the ability to engage in binding contracts such that the new owners could be confident they could indeed evict you as part of the contract. However. De Soto’s focus is on the informal sector – that sphere of economic activity that occurs outside the official formal legal system. Despite the non- official status of the informal systems of laws and property in this sector. in his terminology.) But just because no one can evict you from your house. and form the foundations on which these informal economies function. the house you live in. assets vastly in excess of their capital. As a simple example. seeding many family businesses. the formal and informal sectors are not otherwise equivalent. from which no one would attempt to evict you. Miller and Marc Stiegler Introduction Hernando de Soto. This is no easy task: the benefits of reform flow to . (In countries that have become rich.3 trillion – more than half the combined value of all publicly traded US companies. this does not mean a bank dares accept it as collateral for a loan. by reforming and extending the legal systems and bureaucracies of their various national governments. de Soto’s associates surveyed neighborhoods in various poor countries. The poor pay a great price for informality – most of all in the difficulty of capital formation. Most of the economic activity of the Third World’s poor occurs in the informal sector. shows that the poor of the world have.3 The digital path Smart contracts and the Third World Mark S. In identifying a crucial mystery – the failure of these assets to serve as capital for their owners – de Soto has identified a great opportunity for economic betterment.e. i. mortgages in particular have been a major source of highly decentralized investment. assessing the value of buildings which were not formally titled. The extrapolated value of just the informally owned buildings in the Third World amounted to $9. The distinction is one of credibility of property rights transfer at a distance. is an asset.

explores another path to de Soto’s objective. In a series of steps. dramatically increase capital liquidity. one opened by new technologies of the Net. however. And by video contracts to bridge between the above abstract world and local systems of largely unwritten arrangements. often without need for legal recourse. a . but the benefits of inertia rest with the power holders. accessible over the Net. through the nature of contract- created derivative rights. made possible by new technologies. or trust hubs. This chapter. in forming working large-scale trust networks. • Smart contracts are contracts as program code. We propose an alternate jurisdiction-free digital path. from the basic metaphor of contracts as board games. the cost of the technology itself should rapidly become a non-issue. and we concur. • Limitations and hazards: What other novel problems will we encounter on the digital path? First. spawning a flood of new wealth in the poorest areas of the world? The rest of the chapter is organized as follows: • Networks of trust synthesizes ideas from de Soto and Francis Fukuyama to suggest the strong role played by widely trusted intermediary institutions. where the terms of the contract are enforced by the logic of the program’s execution. We explain de Soto’s program. Miller and Marc Stiegler the now-marginalized property owners. to address the absence of these institutions by bringing the informals into governmental systems of formal law and property. Second. leading to techniques for combining the two kinds of contract elements into split contracts. we explain how smart contracts can resolve the conflict – gaining the benefits of global transferability without sacrificing local knowledge. smart contracts will be unable to express the subtle richness of contracts written in natural language. How may the non-coercive digital path address these issues instead? By ratings – independent estimates of the likelihood that a title listing would be honored. An example shows how the outcome of a smart contract may gain local legitimacy. that could leverage the existing wide recognition of first world institutions to short circuit the slow growth process of the other paths. We make explicit the conflict faced on this path between local knowledge and global transferability. like credit reports for systems of local law. Because binding contracts for ownership transfer lie at the heart of capital formation.64 Mark S. which we call the governmental path. even for the world’s poorest. De Soto argues powerfully that property-system reform should be a primary political objective. Could such a jurisdiction-free contracting mechanism. what if traditional contracts were supplemented and/or supplanted with smart contracts? Smart contracts will enable cooperation among mutually suspicious parties. Given the exponential rate at which the cost of electronics and wireless communications are falling. especially the institutions of title and law. • Backing and legitimacy: Why would a change of electronic title be locally honored as a transfer of control of the actual assets? The governmental path provides backing by coercive enforcement. to compositions of games to turn assets into capital.

Fukuyama’s high trust societies are the ones best able to generate vast amounts of wealth. is not between the other two on a spectrum. In low-trust societies they are not. • The rule of law and not of men: In one respect at least. it is a different stable pattern characterized by dense networks of high trust within families. so not surprisingly. markets would bloom and take care of the people. members of a particular culture come to trust each other. Because the first world has already paid the homogenization costs – it has already lost the diversity the digital path could have preserved. Many people have tried. the Third World’s success may cause the First World to follow along. In the high-trust societies. mutually trusting relationships are easily formed. The universal desire for capitalism was not enough. which could lead to the leap- frogging seen with cell phones in Eastern Europe. we can be sure the digital path will encounter further problems. but rather low trust between families. over which they exert even less influence. the third major category. Besides being first through this transition. . held the naïve view that once oppression was out of the way. and under what conditions. This has proven tragically wrong. This section just scratches the surface. including ourselves. we must first solve this riddle. Fukuyama’s Trust (Fukuyama 1995) makes many perceptive observations of how the world’s cultures differ regarding attitudes and proclivities towards trust – how easily. familial-trust societies (or Confucian societies). • Why the Third World first? Why may the Third World lead the first in the transition to smart contract-mediated global commerce? Because of differences in the nature of legitimacy in these two worlds. indeed more problems than we can anticipate. the character of life on the digital path may be more familiar to the past than the present. Networks of trust Why are some societies so much better able to generate wealth than others? During the great oppressions of the twentieth century. Fukuyama shows how these different patterns of trust seem to explain some of the observed differences in the patterns of businesses that arise in these cultures. Being untested. and the answers proposed by Hernando de Soto and Francis Fukuyama are especially insightful and complementary. Rather. Complex cooperative arrangements require trust. The digital path 65 naïvely deployed smart contract system could exacerbate rather than diminish the dangers of regulatory capture – the participants could end up with less vulnerability to local governments but more vulnerability to distant govern- ments. And because the Third World lacks a working installed base of law and property. many people. In Fukuyama’s taxonomy. The years since then have shown that the absence of oppression was not enough. In order to successfully help. It would be an almost literal realization of the classical liberal ideal. The end of socialism was not enough. Free speech was not enough. This section explains how both of these problems may be addressed.

However. In all cases. for example. and sign contracts. trade. tells a complementary story. or reasons to trust in. as does the highway system. almost as if there were flights between every pair of airports. The economies of scale available Figure 3. The Net itself has mostly the same architecture. in the First World.1). and their lack. but it is built with a backbone architecture due to the economics of the system. by virtue of their reliance on a mutually recognizing backbone of widely trusted intermediary institutions. No matter what the culture. or trust hubs. Although this pattern is a partial centralization. Many small local networks are interconnected on a wider scale through major hubs. more than a very small fraction of the members of our societies (Hayek 1937). no central hub of hubs. and that we normally take for granted in the First World. it is not a hierarchy – there is. do business. from any airport you can fly to any other airport. in the first world massive numbers of strangers meet. negotiate. a sparsely connected actual network acts for most purposes like a densely connected network. despite lack of any prior knowledge of. Nevertheless. Similarly. Rather. it is the painful lack of the various widely trusted intermediary institutions that catalyze commerce at a distance. de Soto’s emphasis is not culture but institutions. Logically. simple cognitive limitations prevent any of us from knowing. two strangers can meet and conduct business as if they had prior knowledge of and trust in each other. Miller and Marc Stiegler But to understand the success of the first world requires something more than Fukuyama’s analysis. For many purposes. much less trusting.1 Hubs of high trust. Trust relationships can be thought of as analogous to the airport hub and spoke pattern (Figure 3.66 Mark S. it is peer to peer. . De Soto’s portrayal of the poor within a Third World village is not one of culturally based low trust. De Soto can also be understood as explaining differences in economic organization according to differences in the possibilities for trust. How is this possible? De Soto’s earlier book. each other. it acts like the network shown on the right of the figure. These hubs are in the business of securing these relationships to minimize the risks their customers face from each other. For example. this often requires them to absorb some of these risk on to themselves. The Other Path (de Soto 1989).

and each individually of minor extent. The list is endless. escrow. consumer reports. The digital path 67 to a hub can help tremendously with these risks. courts and cops. he has developed methods of utilizing more knowledge and resources than any one mind is aware of. underwriters. But should not this situation be an ideal growth medium for hubs? If there is great market need for them. The widespread trust needed by these institutions can be seen as a form of capital that takes a long time to accumulate. Indeed. they build slowly over time. However. insurance. we can analyze the Fukuyama low-trust world (in which the fanout from each node is small) and the de Soto missing- hub world. without understanding it. with each . London 1997.1 forms one large market with great extent. The virtual network of Figure 3. personal communication). Historically. and immediately recognize which effect has the greater impact on a society’s effectiveness: it is the absence of the hubs that ultimately prevents large-scale complex cooperative arrangements from forming. the resulting virtual network would at best form small islands of densely connected networks. arbiters. Roger Ebert. but also because they are experts in loan and risk management. From a simple graph-theoretic point of view (inspired by Granovetter 1973. etc. money. With the absence of government oppression we should indeed expect it to grow here as well. only loosely connected to each other. One of the depressing features of the pictures painted by both Fukuyama and de Soto is that the only hope they see for these societies is home-grown. Other examples of familiar trust hubs include title companies. western societies have developed specialized hubs that bundle trust with other expertise: one trusts Citibank not only because Citibank has a demonstrated history of reliably backing their loans. in the absence of hubs. the West’s backbone of hubs are the result of slow growth processes. The resulting picture resembles both Fukuyama’s portrayal of familial-trust societies. barely connected to each other. exchanges and auction houses. Even with high cultural proclivity for individual-to-individual trust. which are necessary elements of reliability in that field: an organization that attempted to engage in banking without expertise in these fields could not be trustworthy no matter how honorable the employees and executives of the organization might be. enabling a great division of knowledge and labor.2 is one of many separate markets. The division of labor is limited by the extent of the market. and de Soto’s portrayal of networks of villages of informals. notaries. then surely there is great demand and great opportunity. this is the situation from which the hub backbone grew spontaneously in the West. (Friedrich Hayek 1978) The virtual network of Figure 3. (Adam Smith) Modern civilization has given man undreamt of powers largely because.

. If they must recapitulate our path. or the reform and trans- formation of each society’s national government into a system that may be widely trusted. it took a long time for the West. It is through widely trusted title registries that banks can become confident that the collateral against which they make a loan will indeed become theirs in the event of default. but credibility that a transfer of title on the books will be honored as a transfer of ownership in reality. including. Eleven years later.68 Mark S. Well. these do not currently seem able to provide the credibility at a distance needed for this transition. a time during which desperate poverty will remorselessly prevail. What enablers are available now that were unavailable when the West made this transition? Might these enablers be used to help accelerate today’s poor through this part of the process of capital formation? The special roles of title and law In The Other Path. de Soto explained the informal economy and its lack of institutions in general. after much investigation both of the phenomenon of persistent poverty. are credible systems of title transfer. This requires not just trust in the title company itself. individual Third World nation bootstrapping itself through all these steps.2 Low-trust tragedy. in The Mystery of Capital de Soto has narrowed his focus to the crucial role played by the institution of title transfer. While all these hubs are valuable for the formation of wealth-creating societies. and an extraordinary uncovering of the history of how the west overcame these problems. not all are equally crucial. Title registries with this level of credibility enable rights transfer at a distance: people who have never met one another and probably never will can engage in asset transfers and capital formation with the confidence that they will acquire the goods specified in the contract. for de Soto. and which are most painfully absent in the informal sectors of the Third World. the evolution of their own hubs. De Soto’s analysis suggests that the institutions most needed to get the ball rolling. Miller and Marc Stiegler Figure 3. Although de Soto documents the creation of extralegal title companies in the informal sector. it will take them a long time as well.

Why did these previous attempts all fail? Because the formals did not appreciate that the informals already had worked out systems of law. at considerable cost in time and effort. and negotiated arrangements in force in each village. the informals proceeded to ignore the formal title registries and trade assets in the way they always had. versus the need to move the governance of title transfer to hubs. This tension is acute on the governmental path. and formal title registries backed by the formal legal system – all the obviously necessary ingredients. These systems are sometimes called the people’s law. De Soto documents previous well-intentioned efforts. negotiated over time and idiosyncratic village by village. converting extralegal assets. Working with the government of Peru. and producing $2. Even with the best of intentions. de Soto’s organization helped a quarter of a million people formalize many of their assets. the legal system governing these title listings did nothing to reflect the complex informally negotiated arrangements needed to understand what rights someone actually held to a particular asset. and work out. but it is the first such attempt in the Third World to work on a large scale. This is the strategy de Soto has adopted. De Soto’s is not the first attempt to title the informals’ property and bring them into the formal sector. but is rather the official legal system itself. and powers of enforcement. The strategy has been wildly successful in bringing the poor into the modern world. into officially acknowledged parts of the formal economy. the formals’ approach to the situation was We have a legal system. as the homogenized set of rules is not even per title company. a way to integrate those local laws with the national systems. if nothing else. Here’s ours. whose wide scope would seem to require them to operate from a more homogenized set of rules. with surveyors.1 billion of new tax revenues for the government of Peru. Governmental legal systems are hardly the wonders of adaptability de Soto’s program would seem to require. perhaps the only organizations that can fill this crucial titling role in a nation are formal ones backed by government itself. You don’t. De Soto’s special insight in this situation has been that the government must discover and respect the local laws. interviews of informals to ascertain who owns what. some sense of legitimacy. Instead. and obligations. One may hope and expect that these demonstrated tax revenues. The digital path 69 The governmental path Among currently existing choices. will tempt other governments to follow suit. The conflict: local knowledge versus global transferability The difficulty comes from an inherent conflict between local knowledge and global transferability – local knowledge of the idiosyncratic people’s law. and so rapidly became even more irrelevant. rights. over 4 years. Although the title listings reflected a snapshot of who owns what. an accommodation between the two must . A government bundles together widespread recognition. creating new capital. Given this mismatch. conven- tions. geographic information systems. village by village. The title registries were not updated to reflect changes of actual ownership.

1973. and begin to bootstrap themselves out of their poverty by participating in the global networks of commerce (Figure 3. However. Shapiro 1999) able to run hostile code safely and flexibly. and capability-secure platforms – languages (Hewitt et al. absorbed the informal Wild West. National borders aren’t even speed bumps on the information superhighway. they could easily use a title registry run by Citibank in New York to execute the transfer. (However one may feel about this process. notably bureaucracies and lawyers within the national sphere that see this as an assault on their prerogatives. For example. they could reach across the Net to use these services. (Tim May) Due to the Net. Miller et al. The process never becomes easy: each step of progress is another major upheaval in the perceptions and preferences of entrenched groups dedicated to protecting the status quo. those in need of such widely trusted intermediary services could escape as well – escape from the crushing assumption that such services can only be provided by institutions backed by their own governments. if a person in village A wants to sell a tractor to a person in village D. Although difficult. 2000. so we may as well put it to good use. We describe these possibilities in mostly abstract terms until the section on video contracts. 1995) and operating systems (Hardy 1985. Instead. Miller and Marc Stiegler rapidly turn into a procrustean bed. Many First World trust hubs are already widely known and plausibly trusted in the Third World because of the pervasive spread of western media: shows ranging from CNN to Dallas and Baywatch have granted name recognition and an aura of respectability to First World organizations that most governments can only envy.70 Mark S. purely electronic goods and services can now be purchased from across the world as easily as from next door. a couple of villages away.3). Tribble et al. villages on a global scale could become part of a global trust network. as de Soto also documents. If the functions normally provided by trust hubs were offered by well-known First World trust hubs as purely electronic services. and he documents how it did work when formal US law. As if this path were not difficult enough. it is occurring. de Soto offers no alternative. in which we introduce another technological ingredient – one for connecting this abstract world to the concrete lives of the poor. this whole process faces enormous obstacles from many different factions. Describing how existing incentives and new technologies could give rise to new wealth-creating institutions may help us coordinate our activities to move the world in this direction. Rees 1996.) Using First World hubs. he is somehow successfully making this path work. slowly and painfully. In a similar . cryptography. Consumers of these goods and services have already escaped old limits of geography and jurisdiction (Johnson and Post 1996). The digital path Can we sidestep this brutally painful process? Perhaps with the Net.

3 Remote-trust bootstrapping. it will be much easier for them to grow their own high-trust hubs as well: an entity becomes widely trusted by consistently and visibly performing in accordance with various contracts – contracts being managed by hubs that are already widely trusted. ironically. fashion. And in a state such as Russia. Szabo introduces the concept with the example of a drink vending machine. Traditional contracts are understood to be backed by a coercive enforcement system made of courts and cops. the tractor may be securitized. a software program is the operational embodiment of a contract (Szabo 1997). transforming it into capital. Smart contracts How might such hubs deal with the idiosyncrasies of each village’s people’s law – the idiosyncrasies that sabotage traditional governmental attempts to capitalize village assets – without taking on the impossible burden of learning all this local knowledge itself. In what sense is it a contract? . This combination of contract and contract host is a partially trusted intermediary between the drink manufacturer and the purchaser. However. With a working trust backbone. Once the villages of the world join this global village. and without imposing the costs of homogenization? By the use of smart contracts. and performs an exchange of those goods when both have been presented. A drink vending machine is a very primitive example of a smart contract being executed on a contract host – the vending software executing on the vending machine hardware. In smart contracts. have more legitimacy because the titling institution is beyond the reach of their own government. the vending machine does not have the option of such recourse following a breach. It escrows drinks and money. in which it returns the money if the drink cannot be delivered. where the program’s behavior enforces the terms of the contract. highly trustworthy behavior gets the visibility it needs to more rapidly accumulate its own reputation-capital. a title listing with Citibank would. There is even a rollback process. or at least raises the cost of violating the contract. The digital path 71 First World Third World Figure 3.

they are jointly designing the rules of a game they would both be willing to play. which portrays its possession by Alice at this time. the players may then make moves. A. which helps lower barriers to entry. as both parties know that it is ultimately an agent only of the drink manufacturer. representing money. In this paper we explore escrow-based smart contracts. personal communication). in that a participant effectively secures their good performance with the value of their reputation capital. Figure 3. Chapter 14. but they do substantially reduce capital costs. When two people negotiate a contract. but because their logic is vastly easier to explain. neither of the pieces is on the board. not because we expect this form to dominate. Each move potentially changes the board state. Although delayed settlement may substantially reduce capital costs (Selgin 2001. but reputation feedback and credit (Friedman 2001. For concreteness. Contracts as games A basic metaphor for smart contracts is the board game. this volume). Bob might offer a certain amount of stock to Alice by placing it on the right square of the board. In the initial board state. for the electronic systems of title (called issuers). Instead of enforcement. by escrowing both drinks and payment before dispensing either. they would turn smart contracts into explosions of complexity. but only moves judged legal by the rules given the current board state. This has a similar logic.4 shows the six possible board states of a simple negotiation and exchange game (Tribble and Farmer 1991). Instead. in this chapter we assume systems that provide for instant settlement (e-gold). Net businesses have been engaging in rich and rapid experimentation with cooperative arrangements that require no coercive recourse (Krecké. Such arrangements are messier and less amenable to automation than escrow. but a customer who walks away from a contract in progress leaves behind any assets escrowed by the contract at that point (Miller et al.72 Mark S. Miller and Marc Stiegler The vending machine as contract would indeed require separate enforcement if it dispensed the drink first and then demanded payment. and so will occur sooner. 2000).) Although conventional coercive recourse is still often possible on the Net. Steckbeck and Boettke. Other smart contract scenarios involve more symmetric mutually trusted third parties. changing which moves are legal during the next turn. this volume). The most common arrangements involve not actual escrow. taking us . is off the board on the left. Likewise. The gold bar. and most of all because they apply to participants with no prior reputation. and the jurisdictional issues potentially too messy. It cannot prevent the customer from walking away before the game is over. for more and more Net commerce these costs are too great. However. because they are easier to build. let us say the knight represents stock. (The vending machine is partially and asymmetrically trusted. it also dispenses with the need for separate enforcement. Let us say Alice is playing the left side of the board and Bob the right. Chapter 10. Once they commit to playing this game. Both kinds of arrange- ments have their place and will compete in the market. For example. the contract creates an inescapable arrangement.

From here. leaving the game in terminal state F. bringing us back to state A. by placing it on the left square of the board. The contract host serves the same role as our vending machine – it is the third party mutually trusted to execute the contract/program faithfully. Bob may pick up the money offered by Alice.4 Simple negotiation game. a board manager for chess is a program that enables two people to play with each other. the only possible move is for Alice to pick up Bob’s knight. and re-enter the loop of bidirectional arrows (at B or C). we must start by explaining what is happening on whose computer. written to run on a capability system – a secure programming language or operating system suitable for writing smart contracts (Miller et al. while in board state D. maintains the board state. The contract can be any program Alice and Bob mutually agree on. This is the irreversible commitment step shown by the bold unidirectional arrow. and takes us to state E. taking us to state D. At this point. withdraw their piece. and only allows legal moves. The digital path 73 Figure 3. The execution of this game actually involves five parties. This contract/program functions as the board manager for the game they have agreed to play. We assume that each player trusts their own computer (a dangerous assumption. accepting Alice’s offer. (For example.5. We also assume that player X cannot trust player Y’s computer any more or less than they trust player Y. A board manager does not itself play the game. in which case Bob may withdraw his offer by taking back the knight.) . That is why the first transition arrow is shown as bidirectional. Under these assumptions. Or. Alice might not respond soon enough. but we cannot proceed without it). either party may still decide they are unsatisfied. The two players of course. we can treat a computer and its proprietor as a single unit for purposes of analysis. Or Alice may respond to Bob’s offer with a certain amount of money. as illustrated in Figure 3. How is this contract self-enforcing? What prevents cheating? Who needs to trust whom with what? To answer these questions. 2000). Alice and Bob. to board state B.

The contract can embody the knowledge of acceptable arrangements local to Alice and Bob. To the extent the contract host can be trusted at all. Or perhaps they will use a simplified contract construction kit. the simplified story of the custom contract still shows well the logic by which the system operates. or they may be created by specialists. and what they can manage to express in this new medium.74 Mark S.5 Separation of duties. Such an interface may even enable some players to overcome hurdles of language and literacy. they upload the board manager to the contract host. either speculatively or for hire. the contract host need not have any prior knowledge of the contract. These rights are shown as the arrows pointing at the respective chairs. whose user interface might resemble a drawing package specialized for drawing our board-state-transition diagrams. limited only by what Alice and Bob can agree on. and dispenses to each the right to play their respective sides of the game. despite its ignorance of the local knowledge that gives this contract meaning. Most players will not program up their own custom contract. These customizable contracts may have been contributed by earlier players who did write their own contract. etc. Miller and Marc Stiegler Figure 3. prior handshakes.) . it can be trusted to run this contract faithfully. In any case. without needing to speculate on the possible organization of the market for smart contract creation. which then verifies for them that they have agreed on the same contract. Unlike the vending machine. but will instead select a “boilerplate” contract/program off the shelf and fill in the blanks. (Telling the tale this way hides a further division of labor.. Once Alice and Bob agree on the text of a board manager and on a mutually trusted contract host. local custom. This is the first step in resolving the conflict between local knowledge and global transferability.

can determine which issuer backs this eright. let us say stock. they can transact with each other as if they fully trusted each other. without needing to understand those conditions or those erights. is effectively a title company for money.” transform the movement of pieces into a transfer of third-party-assayable transferable electronic rights. The digital path 75 The two remaining players. the $-issuer. Even if Alice and Bob are both use-once pseudonymous identities with no apparent physical location (Vinge 1984). the stock-issuer. Assets + Contracts × Time + ?? = Capital The smart contracts explained so far – the vending machine and the exchange game – cannot turn assets into capital. Bob. the $-issuer. but not the obligation. which can be picked up (transferred to the possession of a player) according to whatever may be the rules of the game. To explain how such unfolding creates ever more abstract forms of property. Alice has a call option when she has the right to buy some agreed asset. through the contract host. and the contract host each need not have any prior knowledge or trust of any of the other four players. For the game to be meaningful. We refer to this as oblivious escrow – the contract host. escrows up front the stock Alice may decide to purchase. but not in each other. The hubs – the issuers and the contract host – thereby succeed at providing virtual trust connectivity between their spokes: Alice and Bob. merely by running the contract. under these conditions. Alice and Bob must have prior knowledge and trust in both issuers and the contract host. but these are beyond the scope of this chapter. Even though Bob does not currently possess this eright. and what the value of this eright is according to this issuer.) The money is an eright in part because it is third-party assayable. to engage in some action at some agreed price before some deadline. at some agreed price before the deadline. A dishonest contract host could abscond with the money instead. like a mortgage. or erights. or more conventionally a bank. and the contract host engage in a three-way cryptographic transaction that brings about the transfer of title. To do so requires contracts that unfold over time. the rights to the money changes hands by the transfer of quantity between accounts – shown in the figure as purses within the issuers. Bob. which is why contract hosts need to be widely trusted. at the $-issuer. . A $-issuer. we step through a simpler example contract. With such assayability. the covered call option. When Alice places the gold bar on the board. A widely trusted contract host presumably has a valuable reputation at stake. of that much money from Alice to the contract host. An honest contract host would consider this money to be only a piece on the board. For money on record at the bank. which helps secure its honest behavior. her computer. ensures that erights in escrow can only be released under the agreed conditions. Alice has an option when she has the right. The option is a covered call option when Alice’s counter- party. the “$-issuer” and the “stock-issuer. (More sophisticated cryptographic protocols are possible which further limit a player’s vulnerability to a dishonest contract host or issuer (Beaver and Wool 1998).

Unlike the previous game. After this move. In the initial board state A. Alice has no ability to trade this new right. Alice has the option to buy this stock for a given price. has created a new valuable right. before the deadline expires.6. that only permits that transition after a deadline has expired. During this interval. This is a very different kind of value than the value of stock or money themselves. In de Soto’s terminology. then abstract rights derived by contracts about these instruments are a step towards being capital. owned by Alice. Should the deadline expire while the game is still in state A. this is the irrevocable commitment step shown as the bold unidirectional arrow. and for Bob to pick up the money. Bob could then pick up his stock and go home. If Alice places this amount of money on the board. Miller and Marc Stiegler Figure 3. Through the system so far depicted. But in the picture so far. in order for these new rights to truly be capital. The value of this new right derives from the value of the stock and the money. in this game the accept- able amounts are predetermined by the rules. this new right is somehow more abstract. What is so different about this contract? During the interval of time from the start of the game until the game leaves state A (whether by expiration or exercise) Alice has something valuable. She may place a gold piece on the left square. A new element in the game is the game clock.76 Mark S. the new right is a derivative of the more literal underlying rights. attached to a transition arrow. We may visualize this as the game shown in Figure 3. despite its complete ignorance that it has done so. and in order for yet more abstract forms of capital to be derived . neither Alice nor Bob may pick up the stock.6 Covered call option. there is something missing from this picture. However. Alice may decide to exercise the option. the stock is already on the board. if the literal instruments are physical assets. The contract host. leaving the game in terminal state B. Alice can trade those erights managed by issuers: money and stock. This needs to be repaired. The left square only accepts the amount of money agreed on when the game was constructed. the only remaining legal moves are for Alice to pick up the stock. taking us to state C. While in this state. In Wall Street terminology. Or. but whereas they are very simple literal kinds of rights.

who would also find this situation valuable. we need to turn this right – the right to continue playing the left side of this ongoing game – into an eright.7. But wait. Not that modern finance is directly relevant to the needs of the poor. this structure decouples knowledge in a way quite different than anything in the financial world. whose five participants are enclosed in the horizontal rectangle. besides being an abstraction built from widely tradeable rights. In order to resolve our conflict. since these rights are produced by games it runs. contract host 1 has no idea if the right to sit in this chair of this game means what Alice claims it means. enabling further abstraction – the creation of yet more forms of capital. game 2. The digital path 77 from them. perhaps our original negotiation game. were he convinced that Alice’s chair sitting rights mean what Alice says they mean. We have the contract host issuing rights it does not understand. De Soto’s capital. To do so requires an issuer for this new right. so how can widespread trust in the contract host translate into credible global transferability of derived rights? How can this derived right be assayable if even its issuer does not understand it? Let us walk through the example depicted in Figure 3. the current state of the board. Hence the tilted overlap of the rectangles. in which this right. would appear as a movable piece. Fortunately. but does not understand. Contract host 1 also appears in both these games – it appears as the contract host in game 1. a third-party-assayable transferable electronic right. The contract host would then revoke Alice’s access to the chair and issue fresh access to the other player. is also itself widely tradeable. we can enable Alice to tell the contract host to transfer her eright to sit in this chair to someone else. Fortunately. Fred can ask contract host 1 for the text of the contract (the source code of the program). would be willing to play a different game on contract host 2. Fred does have prior knowledge and trust of contract host 1. and it appears as an issuer in. Since the contract host is already managing access to this chair. much as the $-issuer would with Alice’s money. of the right to sit in the left chair of game 1. Unfortunately. but it is a good test of the generality of our framework. . Fred and Alice do not trust each other any more than Alice and Bob do. Alice appears in the same role in both these games – as a player. Just as Alice could tell the $-issuer to transfer some of her money from her purse to someone else’s. Networks of games To make this new right Alice holds widely tradeable. with Alice’s consent. having just met. Unfortunately. Alice starts out simply as a player of the original options game. With this ability to compose networks of games. While Alice finds herself in the resulting valuable situation – while the options game is in state A – she encounters Fred. Fred. The contract host is not in a position to vouch for any meaningful property of the rights it is issuing. hosted by contract host 1. it seems we have the ability to express the full range of contract layering used in modern finance. issued by contract host 1. we may as well have it double as the issuer for the eright to sit in this chair. or anything else.

and Fred can accept Alice’s help. without any trust required between Fred and Alice. so third-party assayability has been provided in theory. including some Fred trusts. Should the contract truly be idiosyncratic to local knowledge shared by Alice and Bob. deriving yet further erights. and the assays of all the pieces presently on the board (i. Miller and Marc Stiegler Contract Host 1 Stock $ Issuer Issuer Op Iss tion ue s r Alice Bob Co nt ra ctH os t2 $I ssu er Fr ed Figure 3. can help Fred figure this out. . With this step. Alice. he may not find the derived rights comprehensible at reasonable cost and move on.7 Layered games.e. derived rights become erights about which other contracts can be written. if the contract is understandable to some number of others. who presumably understands the game she is playing.78 Mark S. escrowed by the game). these should be sufficient for Fred to figure out what these derived rights are. knowledge to which Fred has no access. but not yet in practice. More commonly. In theory. Fred may turn to them for advice on the contract’s meaning – the analog of legal advice.

there will remain pressures for homogenizing capital-creating contracts. Although the conflict is resolved in this scenario. could resolve the conflict de Soto explains. well after starting on the digital path. Backing and legitimacy For purely electronic assets. as if these had once been separately owned – then the smart contract will have preserved this local knowledge in a form that can be uploaded to widely trusted contract hosts. applied to informally owned assets. the separation of the contract from the contract host. these pressures can gradually work themselves out over time. and in competition with the benefits provided by custom contracts. Our new technological enablers – the ability of the contract host safely and faithfully to run rights-handling code it neither trusts nor understands – allows these two specialties to be combined without conflict. This allows the second step to be applied yet again to create yet more abstract erights. Fortunately. only standardized contracts can create fungible assets tradeable on large exchanges. and therefore also a major barrier to starting on that path. Further. This step allows these three steps to be applied repeatedly. as recorded at the time of titling. The first step. like mortgage. to model the base rights as if they were derived from prior simpler rights to more literal physical objects. Our second step is contracts that unfold over time. while the contract host specializes in providing the trust connectivity needed for global transferability. creating complex networks of contracts that build on each other. The title listing for these assets is the reality of ownership – there is no separate issue of physical control that may or may not follow . but these are normal market pressures. The third step turns the local-knowledge embodying rights created in the previous steps into widely tradeable erights. allows the contract to specialize in capturing local knowledge. By contrast. Informally owned assets are not literal physical objects themselves. In the above scenario. creating derived rights. on the governmental path pervasive rule homogenization is the necessary first step. but the rights to the property held by the original property holder. according to customary law in that community. they are rights derived from these objects according to informal local laws and negotiated arrangements. like fiat money and stock. To the extent the logic of these arrangements can be successfully expressed in the new language of smart contracts – with more literal physical objects represented as underlying assets. The digital path 79 Resolving the conflict The above steps. This step would be applied twice. But she is still in danger of losing the sale through Fred’s difficulty understanding the contract’s meaning. setting loose the power of capital formation. at this point we have. not the physical property itself. providing global transferability without resort to procrustean homogenization of legal systems. First. perhaps. an adequate picture. Alice will no longer lose the sale to Fred through Fred’s lack of trust in Alice. The collateral would be.

How may we compensate for this lack? For concreteness.80 Mark S. such as eviction following a title transfer. This should be an area ripe for entrepreneurial invention. Rather. which takes these likelihoods into . Bond-rating agencies provide the market with an estimate of the likelihood that a business or government will actually meet the obligations represented by its outstanding bonds. but as a way to get started it has a fatal problem – it requires a crippling up front capital investment. On this issue. the digital path has the huge advantage over the governmental one explained previously. who have no knowledge of a particular governing village. in order to unlock the potential capital in the poor’s $9. not by their title listing. Unlike government-based title transfer. we could perhaps stop here.3 trillion dollars of extralegal buildings. But we have not yet succeeded at the mission we set out on – to establish credibility of title transfer at a distance. even with this advantage. in order to cover the massive potential liabilities. but by consensus of the governing community in question. even governments mostly understood that coercive enforcement would have had too terrible a cost. and any new system can only be seen as legitimate if it incorporates and builds on the old. But what of the disadvantage? The Net is a purely non-coercive medium. this solution may be ideal. the previous attempts failed when title listings not locally seen as legitimate were not honored. to nevertheless have some reliable basis for judging the credibility of particular title listings. The rating agency does not attempt to estimate what the price of the bond should be – that is left to the market. To explain how smart contracts may be applied to Wall Street. These are the source of the legitimacy governing current control of the property. etc. it is adequate for distant traders. Ratings The issue of credibility does not require all title listings to have high credibility. but we do not presume to foresee what the actual outcome of the market discovery process will be. Actual control of such physical assets is determined. another familiar form of intermediary may be adapted to this situation. Under these circumstances. An obvious answer is to introduce another trusted inter- mediary institution into this market – title insurance. Smart contracts can change their electronic records which claim to be about the world – such as title – but they cannot force the world to follow along (Friedman 2001). Governments employ a vast coercive apparatus to enforce the outcomes they claim are legitimate. we propose here two complementary techniques. Why wouldn’t the digital path fail in the same way the pre-de Soto governmental attempts failed? Why would these communities consider these title transfers legitimate and honor them? The governmental path has an advantage here. Still. Miller and Marc Stiegler along. The missing ingredient was the need to accommodate pre-existing local arrangements. it transmits only information – effectively speech – but cannot transmit force. De Soto documents well the repeated victories of squatters over governmental law. these changes of title are not backed by a coercive enforcement apparatus. land. in order to establish possibility. Rather. Once the digital path matures.

and store it with the contract for future reference. but complex lawyer-written contracts on paper no longer plausibly meet this standard. Ideally. A bond-rating agency does not put its money where its mouth is – it does not issue bond insurance to back its likelihood estimates. and it is more than just incentives or force. Rather. Simply recording each village’s track record of honoring past title transfers. providing it an incentive to treat these titles as legitimate claims. based on the kinds of evidence you are used to taking into account. probably including the power to evict. At the time it was a necessary trade. However. Why should Bob believe them? Absent coercive power to evict. Bob believes that if this were true. much less verify. Why would they take an outsider’s word over his? The incentives produced by rating agencies are far too weak an answer. We can think of this as a credit report. Perhaps Bob is told the veracity of these records can be verified by cryptographic means. a title listing could store . contracting parties could now record their conversation about the contract’s meaning. full of all the conscious and subconscious cues we use to understand what each other understands – so they plausibly have a good sense of what they jointly mean to agree to. we can imagine a market of third-party raters that post judgments of the likelihood that transfer of a given title will actually be honored (Stanley 2001. Verbal contracts often do at the moment of the handshake – both parties have just had a rich conversational interaction. also rapidly dropping in price. Similarly. which can become quite valuable. Into this situation. let us introduce Szabo’s Video Contract (Szabo 1998). One day a distant voice on the phone tells Bob to vacate his family home because. Video contracts But legitimacy is more than just preserving local arrangements. it claims. Bob would have heard about it from Sam. As Szabo explains. memory is fleeting. And it places each village in an iterated game with the system as a whole. if there is sufficient local consensus on the legitimacy of the claims of the outsider. not for an individual. However. The evidence presented by these disembodied strangers are impersonal electronic records Bob can hardly understand. is a low overhead procedure that is plausibly adequate. personal communication). The digital path 81 account. It is knowing in your bones what the right thing is. it backs its estimates with its reputation. sincerity. and vividness in order to get permanence. in our situation. his late father Sam took out a loan from a distant bank Bob has never heard of. but is still cheaper than issuing insurance. so people turned instead to paper to record agreements. Using video cameras. trading away richness. why should Bob even take them seriously? Why not just stay in his home? Bob’s community does have power of various sort over him. and assuming the future will be like the recent past. subjecting them to the local tradition’s means of enforcement. but for a village and its system of local law. But no longer. Bob is a known member of the community. a contract is supposed to represent a meeting of minds.

Such recordings should also help overcome barriers of language and literacy. For most communities. Conventional contracts make use of the rich expressiveness of human language. to both Bob and his community. and subject to the blind spots of wishful thinking. From a game-centric perspective. and dangers we can anticipate. adequate if a dispute does not arise. the video of the conversation with Sam about the meaning of the contract. Miller and Marc Stiegler the entire chain of videos for the chain of contracts by which the property changed hands – starting with the initial video interview at the time the initial title record was created. • an agreement on which person or institution should read the text and arbitrate the outcome of a dispute. and the arbiter is only another player. has limitation. seeing Sam explain clearly what rights he’s trading away will be enough to establish legitimacy. After all. problems. and judgment Many pre-existing and desired arrangements will not be expressible purely as smart contracts. having been previously navigated.82 Mark S. a split contract could consist of: • an automated game. perception. relevant only during a dispute. but many more issues remain. and so is the “real” contract – the game is only a lighter- weight approximation for typical non-disputed cases. all of which are vastly more subtle and sophisticated than any currently automatable alternative. not some outsider making dry claims about Sam’s past intentions. For example. Between the poles of fully human contracts and fully automated contracts is a spectrum of arrangements we call split contracts – the contract is split between automated and non-automated parts. that’s Sam talking. From a paper-contract-centric perspective. The digital path is untrod and mostly unknown. the ability to declare the outcome to be in dispute is only another move in the game. enabling us to take advantage of the strengths of both. and how one presents the results of such integrity checks as evidence properly credible to non-computer people. The outsider could show. demonstrates some of the ways to design split contracts so the two parts can play together well. In this section we take a first stab at some of the problems lurking ahead of us. Open areas to explore include technological means to inhibit forgery (like time-stamping digital notaries). Incorporating human language. and judgment. given general confidence the video is not fake. perception. Amix (Miller 1999. the text as interpreted by the arbiter is the outcome of last resort. . Walker 1994). Limitations and hazards The governmental path. • natural language text expressing what the game could not. The first smart contracting system.

and the identity of the agreed arbiter. contract host 1. Fred would then take these into account in assessing the value of Alice’s chair – the rights Alice would like to use as collateral. On de Soto’s governmental path. The contract needed to represent these latter may often remain mostly non- automated. and subject to its decrees. have we not just transferred this vulnerability from their own governments to those of the First World. The architecture of the Net has dramatically turned this around. These concentrations made economic activities of various sorts subject to regulatory capture. over which they have even less influence? The Net treats censorship as damage and routes around it. so far it has not provided the same escape for producers. On the digital path. would provide Fred as well with the text or video. such as the banks. The digital path 83 Another form of split would be by layers. Whereas the logic of a mortgage game may be fully automated. Regulatory capture versus regulatory arbitrage Historically. themselves competing to establish a reputation for operating honestly. despite its benefits. also dramatically lowered the cost of regulation.7. the growth of widely trusted large-scale institutions in the West – and the corresponding partial concentration of economic activity into the trust backbone – made possible the rise of the large regulatory state. The architectures of the first generation of electronic media – radio and television – amplified censorship and diminished free speech (de Sola Pool 1984). but will cause a flight of electronic business towards climates expected to remain freer. Any one government going bad would endanger many contracts and much property. (John Gilmore) The nature of the dangers depends on the nature of the architecture (Lessig 1999). Although the Net has allowed the consumers of electronic goods and services to escape limitations of geography and jurisdiction. spread across competing jurisdictions. the informals come to be dependent on the integrity of their own governments. the rights being put up as collateral are limited to those held by the original property holder according to the governing village’s law. These remain tied to some government. . as there were far fewer places in the economy that needed monitoring. especially those with worldwide name recognition. Might a decent architecture for distributed smart contracting treat regulation as damage and route around it? The most powerful answer is already implicit in the architecture of the digital path – a diversity of contract hosts. In Figure 3. as issuer of these locally defined rights. creating actual freedom of speech more absolutely than even the best constitutions. by having their contracts rely on the trustworthiness of First- World trust hubs. in danger of local regulatory capture. This concentra- tion.

such as a voting protocol in which a quorum of. Virtually all progress to date towards the digital path (Johnson and Post 1996. As explained above. but a reliable virtual computer may be synthesized from several actual computers by comparing the outcome of each step in a kind of voting process. For a business in these countries to be judged legitimate by the culture. Making such technologies work is tricky. we expect the Third World to overtake and then lead the First in making this transition. can seriously impede these transitions.84 Mark S. have enjoyed great wealth. an actual First-World trust hub may be considered an analogously unreliable component. among the informals the formal legal system has no monopoly on legitimacy – many extralegal institutions enjoy widespread popular legitimacy. and builds on pre-existing systems of legitimacy. The character of legitimacy in the first world is quite different to the legitimacy we have been discussing among the Third World’s informals. Due to dangers of regulatory capture as well as internal corruption. having made the transition to the governmental path long ago. five out of seven actual contract hosts have to agree on an outcome in order for it to be considered an outcome of the synthetic virtual contract host. In most rich First-World countries the issue of legitimacy is inextricably coupled to legality. For example. let us say. Krecké. this volume. These issuers themselves can be virtual reliable issuers in this same sense (Szabo 1999). Chapter 14. Why? Comparative legitimacy Primarily because. incorporates. to express these rules in the language of smart contracts. By contrast. . but have paid a subtle price in flexibility. once again. of the issue of legitimacy. the business must be legal – it must operate within the formal legal system. so we should not try to achieve trans- jurisdictional fault tolerance before we get started. a new system of law will only be seen as legitimate if it accommodates. once technology costs become inconsequential. and in order for that game’s issuers to honor the outcome. and for others to consider their dealings with this business to be legitimate. for certain demanding applications an indi- vidual actual computer may be considered unreliable. Nevertheless. From a set of these we may synthesize a reliable virtual trust hub in a variety of ways. First-World societies. Why the Third World first? This new world of Net-based jurisdiction-free coercionless smart contracting – the digital path – is an option for the First World as well as the Third. Both groups stand to gain tremendously by this transition. Lessig 1999) has been in the First World. but we should also make sure not to paint ourselves into a corner – we need to understand how a simpler working system could incrementally grow to support such fault tolerant protocols. and the effort to accommodate this complexity. Miller and Marc Stiegler The field of fault tolerant computing studies how to build reliable systems from unreliable components. The pre-existing formal and informal systems each have a very different kind of great complexity.

the very informality of these systems allows them to compromise. causing markets to bloom. because of the formality with which this law is administered. cell phones offered a huge advan- tage over their prior situation. Homogenization costs The costs of rule homogenization discussed above. and the absence of competitive pressures. not in rich high-tech societies. Those without an adequate prior system were able to more quickly leapfrog over to a better system. In the West. One of the more effective sources of pressure is that which stand to gain from large-scale commerce with the rest of the world. Likewise. (My own observations of Prague versus Silicon Valley in 1998 corroborate this – cell phones were everywhere in Prague. Saving the First World Should the Third World be the first to succeed at the digital path. and an improvement over a system many consider imperfect but adequate. The people’s law of each individual village. and imperfectly. each is a vast growth of complexity that no one even pretends to understand. They offered a much more minor improvement in societies where the land lines work. so the digital path’s option to avoid paying these costs is not a selling point there. By contrast. More important. There is no necessary sequence of telecommunication systems that each society must separately recapitulate. might emerge so slowly as to be a non-issue. creating vast wealth. village by village. being largely unwritten. This is a high enough bar that a smart contract system. However. to be legitimate by this standard. to be paid on the governmental path.) Not having a working phone network. may be simpler than formal law. and should this in fact unleash their potential capital. The First World has already lost this great source of diversity. the informals have no access to working global networks of trust and commerce. it provides a smaller improvement. Once a significant part of the world’s economy . Cell phones in eastern Europe Cell phones first became society-wide hits in poor countries with terrible telecommunication. The digital path 85 Among the informals the great complexity comes from the sheer number of local arrangements that need to be expressed. imperfect expressions can often be judged to be good enough. although there are far fewer separate systems of formal law. is a cost that has already been paid and largely forgotten in the First World. how would this effect the First World? Formal laws in the First World do change under political pressure. this system brooks no compromise except through politi- cally driven change. The digital path offers them tremendous new opportunities. As long as an adequate spirit of the law is uploaded. This lets the transition get started incrementally.

Cambridge. K. Miller and Marc Stiegler is occurring in the digital First-World businesses would then face a choice – trade with these networks or stay legal-legitimate.htm> (accessed May 2001). not the judgment or skill. A key means of enabling cooper- ation was the original right of contract. Greg Burch. This competition forms a vastly stronger and fully decentralized system of checks and balances. the character of the digital path may best be described as a pure form of the classical liberal ideal – the rule of law and not of men. With smart contracts. Terry Stanley.springer. of a diverse market of competing contract hosts. (1989) The Other Path. D. and the members of the e-lang mailing list.” in Lecture Notes in Computer Science. de Sola Pool. Once such unregulatable trade is made legal. Zooko (Bryce Wilcox-O’Hearn). Bill Tulloh. for which we thank Darius Bacon. Don Lavoie. where almost any mutually acceptable arrangement could be made binding. MA: Harvard University Press. Ted Nelson. Indeed. . (1998) “Quorum-based secure multi-party computation. (1984) Technologies of Freedom. The Third World could rise on an enhanced version of the principles on which the West grew rich. The pressures will be great to legalize trade with these jurisdiction-free networks of commerce. Eric Drexler. Charles Evans. References Beaver. Online. Michael Glenn. Ka-Ping Yee. the encoded rules themselves become the logic of their own enforcement. This is not just a cheap play on words. Nick Szabo. Acknowledgments These ideas have formed over much time and many valuable conversations. Doug Jackson. Ken Kahn. A. subject only to the honesty. Robin Hanson. What will be the character of the resulting world? The rule of law and not of men Surprisingly perhaps. Ian Grigg. enforced impartially and justly.86 Mark S. E. Springer Verlag. and Wool. This is an unpleasant choice both for them and their governments. the dam will have burst. Jonathan Shapiro. John Gilmore. New York: Harper and Row. de Soto. Jack Birner. The ideal they were describing was of a neutral simple framework of rules. Robert Gerrard. Chris Peterson. Gayle Pergamit. Available HTTP: <http://link. with the law serving as the mutually trusted intermediary for securing the arrangement. I. providing for cooperation without vulnerability – protecting individuals from each other while enabling them to cooperate with each other. H.-Dean Tribble. the digital path could more literally realize the meaning of those words than anything the original classical liberals could possibly have conceived.

The digital path 87
de Soto, H. (2000) The Mystery of Capital, New York: Basic Books. Chapter 1 online.
Available HTTP: <> (accessed May
Friedman, D. (2001) “Contracts in cyberspace,” The Berkeley Law and Economics Working
Papers vol. 2001, no. 2. Online. Available HTTP:
<> (accessed Feb 2003).
Fukuyama, F. (1995) Trust: The Social Virtues and the Creation of Prosperity, New York: Free
Granovetter, M. (1973) “The strength of weak ties,” American Journal of Sociology 78:
Hardy, N. (1985) “The KeyKOS architecture,” Operating Systems Review September: 8–25.
Updated version online. Available HTTP:
<> (accessed May 2001).
Hayek, F. (1937) “Economics and knowledge,” Economica (reprinted in J.N. Buchanan (ed.)
Essays on Cost, LSE (1973), Weidenfeld and Nicolson). Online. Available HTTP:
html> (accessed May 2001).
Hayek, F. (1978) New Studies in Philosophy, Politics, Economics and the History of Ideas, Chicago:
University of Chicago Press.
Hewitt, C., Bishop, P., and Stieger, R. (1973) “A universal modular actor formalism for
artificial intelligence,” International Joint Conference on Artificial Intelligence (August), San
Francisco: Morgan Kaufmann, pp. 235–45.
Johnson, D.R. and Post, D.G. (1996) “Law and borders – the rise of law in cyberspace,”
Stanford Law Review 1367. Online. Available HTTP:
<> (accessed May 2001).
Lessig, L. (1999) Code, and Other Laws of Cyberspace, New York: Basic Books. Excerpts online.
Available HTTP: <> (accessed May 2001).
Miller, M.S. (1999) “Observations on AMIX, the American information exchange.”
Online. Available HTTP:
<> (accessed May 2001).
Miller, M.S., Morningstar, C., and Frantz, B. (2000) “Capability-based financial
instruments,” Proceedings of Financial Cryptography 2000, Springer Verlag. Online.
Available HTTP:
<> (accessed May 2001).
Rees, J. (1996) “A security kernel based on the lambda-calculus,” MIT AI Memo No. 1564,
MIT, Cambridge, MA. Online. Available HTTP:
<> (accessed May 2001).
Shapiro, J.S. (1999) “EROS: A capability system,” unpublished Ph.D. thesis, University of
Pennsylvania. Online. Available HTTP:
<> (accessed Feb 2003).
Szabo, N. (1997) “Formalizing and securing relationships on public networks,” First Monday
2(9). Updated version online. Available HTTP:
<> (accessed May 2001).
Szabo, N. (1998) “Video contracts,” Online. Available HTTP:
<> (accessed May 2001).
Szabo, N. (1999) “Secure property titles with owner authority.” Online. Available HTTP:
<> (accessed May 2001).
Tribble, E.D. and Farmer, R. (1991) Derived from work done for AMIX, The American
Information Exchange California: Palo Alto.

88 Mark S. Miller and Marc Stiegler
Tribble, E.D., Miller, M.S., Hardy, N., and Krieger, D. (1995) “Joule: Distributed
application foundations.” Online. Available HTTP: (accessed May 2001).
Vinge, V. (1984) True Names, Bluejay Books. Online. Available HTTP:
<> (accessed Feb 2003).
Walker, J. (1994) “Understanding AMIX,” in J. Walker (ed.) The Autodesk File, 4th edn.
Online. Available HTTP:
<> (accessed May 2001).

Part II
Some history

4 High-tech Hayekians
Don Lavoie, Howard Baetjer, and William Tulloh,
with comments by Howard Baetjer, Marc Stiegler,
and Pietro Terna

Original text of “Prefatory note: the origins of ‘The Agorics Project’ ”
by Don Lavoie published in Market Process (1990) Vol. 8, Spring, pp. 116–19.

Readers of this journal are accustomed to cross-disciplinary explorations from
economics to a number of other sciences, but until now, to my knowledge, there
have not been any attempts to communicate with the field of computer science.
In September of 1989 at George Mason University there began something that
is being called the “Agorics Project” in which graduate students from the
Artificial Intelligence Laboratory at GMU’s Department of Computer Science
joined several economists at the Market Processes Center to investigate a number
of topics of mutual interest. The name “agorics” is borrowed from some research
that moves in the opposite direction across that disciplinary boundary, trying to
use economics in computer science, but the aim of our group is to explore some
ways that economics might benefit from looking into certain developments in
computer science.
The accompanying article is the product of several months of conversations
among computer scientists and market process economists who have begun to
delve into the five distinct research areas the article describes. The substance of
the topics we have been discussing is summarized there, but in this note I would
like to supply a bit of historical background that might help to explain how it
happened that several of us here at the Market Processes Center have suddenly
found ourselves looking into these new research areas. I will tell the story in a
rather autobiographical manner, not so much to try to claim all the credit for
bringing computer science issues to the attention of market process economists as
to try to share the blame. Although I had always wanted to find a way to take
advantage of my background in computer science, it was only as a result of
numerous discussions I have had with the various individuals I will be mentioning
that I finally decided to return to the computer field. Only time will tell what gains
may result from this sort of intellectual interchange. Computers are my first love.
The field of computer science in which I have been the most interested is artificial
intelligence. When I was first introduced to computers while still a high-school
student, I was trying to design a program to play poker. As an undergraduate
majoring in computer science at Worcester Polytechnic Institute, I did a senior

92 Don Lavoie, Howard Baetjer, and William Tulloh
project with two other students that produced a program to “simulate” the process
of music composition. Based on discrete event techniques, the program wrote
“music” step by step according to the rules of the Baroque fugue. I presented the
results with my co-authors at a professional meeting of computer scientists, who
seemed to like it, but we were acutely aware that if it were a meeting of musicians
it would have not been such a hit. Whether trying to play card games or compose
music on computers, the experience of trying to simulate what human minds do
was humbling. There is probably no better way to realize the serious limitations of
computers and the amazing subtleties of human intelligence than to attempt to
reproduce the thinking process. It was easy to get the programs to mechanically
obey specific rules, but it was something else to try to replicate Bach.
While at college, I first became interested in economics. Ever since, I have
wanted to find a way to legitimately use computers in economics. My attraction to
the market process school came from the way it addresses the questions about
the nature of human knowledge which I had been thinking about from the stand-
point of artificial intelligence. Most of the ways computers have been used by
economists have not interested me in the slightest, because of the way they treat
action as purely mechanical. But I was not convinced that computers were
necessarily “mechanistic” in the relevant sense. I began to explore the possibility
of designing a non-mechanistic kind of simulation with which market process
economics could be comfortable. My exposure to discrete event simulation
techniques as an undergraduate suggested to me that such aspects of choice as
creativity and divergent perceptions, which have not been modeled well by
mainstream economics, might be “simulatable” with such methods. It might be
possible to construct computer simulations that could be used as mental
experiments for developing market process theory.
When I began graduate school in economics at New York University I con-
tinued to think about the possibility of such computer simulations, and intended
to write my dissertation on the topic. The seminar I gave on the idea to the weekly
Austrian colloquium at NYU was not exactly met with enthusiasm, but it was
greeted with enough openness and tolerance to encourage me to keep thinking
about it. I continued working on it for about a year, getting so far as to design the
main structure of the program. Eventually, however, I talked myself out of the
project. I was sufficiently concerned that I would get nothing useful out of it that
I decided not to risk my Ph.D. on it, and chose a less risky topic in the history
of thought. (The first economics paper I had written was for a computer science
class arguing, rather unpersuasively, why computers could not answer Mises’s
1920 challenge to socialism, and I had always wanted to make that case more
complete.) So until recently, computer programming has been merely the way
I worked myself through graduate school, and has had nothing to do with my
economics research.
A few years later, as a more secure, soon-to-be-tenured faculty member at
GMU, I began thinking about the idea again. I started discussing it with a
graduate student, Ralph Rector, who also had computer programming experi-
ence, and who took an immediate interest. We had regular sessions for a couple of

and Kevin Lacobie. whose networks of contacts and enthusiasm for our work has helped to further fuel this interest. Salin had introduced Miller and Drexler to Hayek. Indeed. and the number of ways in which contemporary computer science appeared to be closely related to Hayekians themes was astonishing. and decided to take them up on their offer to visit them so that we could talk over the simulation idea. I had found someone to talk with at length about the kind of project I had in mind. The understanding these software engineers exhibited of Hayek seemed to me to be far more sophisticated than that of most economists. as well as Howard Baetjer. I think I can say that Howie. and I was reinforced in my belief that in principle it might be feasible. which renewed my interest in my first field of scholarship. and suggested that a few of us come out to Silicon Valley to visit them. but I was able to bring Bill Tulloh and Howie Baetjer with me. Now I am rejuvenating the idea for a third time. Bill also introduced me to dozens of interesting articles in Artificial Intelligence and other areas. and the simulation idea in particular. For the first time. Bill alerted me to two path-breaking papers by Miller and Drexler. as well as several other possible research topics their papers hinted at. It turns out that Ralph was already far along on another dissertation topic and unable to join us. He also introduced me to Bob Crosby. in our judgment. Drexler. exploring some of the possibilities as well as the apparent obstacles and methodological objections. Bill and I had one of the most intellectually stimulating experiences of our lives. A group of graduate students at the Center. an interest stem- ming from a paper he wrote for Jack High’s graduate course in market process economics. Above all I have one of my graduate students. Fred Foldvary. a resounding success. I began a series of lengthy phone conversations with the three of them. As a result of these discussions my thinking advanced considerably about how such simulations might be constructed. I wrote an enthusiastic letter to the co-authors. which used Hayekian ideas to attack certain classic problems within computer science. secretary of the Washington Evolutionary Systems Society. They also introduced us to a third person. and our discussions helped to resolve several of the conceptual difficulties I had originally worried about. I ended up talking him out of the simulation project on the grounds that it was too high a risk for a dissertation topic. The trip was co-sponsored by the Market Process Center and the three hosts. along with two computer scientists from Palo Alto. were also impressed by Miller and Drexler’s work. Bill became interested in the Miller and Drexler articles as a result of his explorations of recent developments in the study of complex phenomena and self-ordering systems. to thank for getting me back into computer science in general. High-tech Hayekians 93 months. who responded in kind. including Ralph Rector. and never really turned my research back in the direction of computer science. Miller. Although Ralph had not really given up. The trip was. Ralph developed some of the main components of an algorithm for decision- making under radical uncertainty. Yet I still had serious doubts. Bill Tulloh. Mark Miller and Eric Drexler. Phil Salin (see his article on “The Ecology of Decisions” elsewhere in this issue of Market Process). . and Salin.

Kurt Schuler. The Agorics Project.94 Don Lavoie. The process of trying to design programs that articulate the logical structure of our economic theories has been a wonderfully educational experience. and in the spring of 1991 I will be teaching a new special topics graduate course on “Economics and Computer Science. and Hugo de Garis and Pawel Stefanski. Conversations I have had with Mark Miller after our trip have convinced me that it is an idea worth exploring more seriously.” in the hopes of investigating the various topics more systematically. The main body of the text was mostly put together by Howie and myself. Many of the reasons that had twice dissuaded me from attempting to develop the simulations no longer seem so daunting. was launched at the Center for the Study of Market Processes in the fall 1989 semester. and William Tulloh The accompanying paper is a product of that stimulation. The group is continuing to meet on an informal basis throughout the spring 1990 semester and this summer. of the GMU Economics Department. David Burns. For one thing. The last of the five possible research topics described in the co-authored article sums up this idea of using computer simu- lations as mental experiments for economic theory. put a whole new slant on the simulation idea. and one by Stefanski and Baetjer. as we point out. Bill Tulloh. We don’t feel so much that we are learning computer science. although that must be happening too. who almost single-handedly wrote the bibliographic appendix and collected the references. one by de Garis and myself. of the Department of Computer Science. Kevin Lacobie. Although building such simulations for the purpose of doing economic theory was certainly not the point of the Miller and Drexler papers. The team has so far been composed of Howard Baetjer. as reinterpreted in our own words. though so far the one we have talked about the most is the simulation idea. and it has. Artificial Intelligence (AI) research has matured a great deal since the days I studied it in college. We have had on our discussion agenda all five of the topics mentioned in the co-authored article. we have been encouraged by the results – and even more by the process. The underlying ideas. . for me. Howard Baetjer. We welcome any suggestions our readers may have as we begin to explore the possible gains from intellectual exchange in this new form of interdisciplinary research. their work has obvious implications for this kind of research. an informal discussion group that is looking into ways computer science might be useful to market process economics. Although as a result of our remarkable visit I have added several new research areas to my interest in the computer idea. and myself. but that we are learning Mengerian economics. Although these simulations are still at a rather rudimentary stage. The literature the paper surveys was mostly unearthed by Bill. they have not supplanted but only reinforced that original interest. are mostly those of our gracious hosts. So we have begun to do just that. and the techniques being used now in machine learning seem much more promising than anything I had known about in my earlier flirtations with the simulation idea. Two distinct programming simula- tion projects aimed at the replication of a Mengerian process for the evolution of money have been initiated by the team.

We visited them in their homes and in their offices. Mark S.A. Miller.” This was no accident. These academic interests.” the two software design companies where Miller and Salin work. programmer. Salin et al. have made major advances in what might be called the Economics of Computation. Miller and K. entrepreneur. but at the same time they almost always involved computer science. especially the work on property rights economics and the spontaneous order theory of F. pp. but they have academic interests. are focused on practical. or “AMIX. Our time in Palo Alto was so replete with ideas. two computer scientists. What makes this research especially interesting to several of us at the Center for the Study of Market Processes is that the kind of economics Miller and Drexler are using to overcome their own problems in computer science is “our kind. In August of 1989. Hayek. The organizations they are involved in are start-up companies that are developing software to help people manage information in ways that enable knowledge to evolve more rapidly. Drexler. real-world appli- cations. High-tech Hayekians 95 Original text of “High-tech Hayekians: some possible research topics in the economics of computation” by Don Lavoie. consider market process ideas to be of enormous practical usefulness in the design of commercial software. Eric Drexler. discussion and new insights that we can hardly begin to summarize them all. We were extremely impressed. we spoke to some twenty to thirty young computer scientists who are energetically applying market process insights to some very ambitious software ventures.A. We came hoping to find one or two possible research topics. and philosopher. even in the cars on the way to restaurants. we were struck by the depth of understanding these computer scientists have of our literature. The success of these pioneering papers suggests that there may be some interesting research possibilities in computer science for the application of market process ideas. and William Tulloh published in Market Process (1990) Vol. for reasons to be described shortly. Throughout the visit the conversations always involved Hayekian ideas. 8. In a thoroughly intriguing set of papers recently published in a book edited by B. Each has a firm understanding of market principles and a deep appreciation for the general social principle of non-intervention. however. In all. and for whom Drexler is a consultant. Spring. We returned with too many possible topics for the three of us to handle. largely unexplored subdiscipline on the interface of economics and computer science. Howard Baetjer. 119–46. In this note we would like to introduce the reader to the most important ideas we encountered. and indicate why we think market process economists should look into this fascinating work. How shall we describe this group? Each of them seems to be a rich brew of scientist. Reading Miller and Drexler’s papers. . the three of us made a trip to Palo Alto to visit Miller and Drexler at the Xanadu Operating Company and the American Information Exchange Corporation. a new. Huberman entitled The Ecology of Computation. They are not primarily academics. The animation of our discussions rarely slackened. It turns out that Miller and Drexler were specifically alerted to the economic writings of Hayek by Phil Salin.

an investigation of the nature and implications of startling advances in technology toward which we are moving. He brings to Xanadu a constant reminder of the practical steps that need to be taken on a daily basis to reach the company’s ambitious goals. K. and with his wife Chris Peterson runs the Foresight Institute. and now chief architect for Xanadu. All shared an expectant outlook on the future. Phil Salin: an entrepreneur. since the whole is a summary of our discussions with them. co-author of four of the essays in the Huberman book. Howard Baetjer. in a sense. of this paper.96 Don Lavoie. The development of the research ideas outlined below was very much a team effort. With serious interests ranging from molecular engineering and nanomachinery to Austrian economics.” especially the approach called “neural networks. formerly in the telecommunications and space transportation industries and now on Xanadu’s board of directors. Any given project may easily overlap these research areas. It would be useful to distinguish five different kinds of research areas that our discussions hit upon. We also spoke with Marc Stiegler. and author of Engines of Creation: The Coming Era of Nanotechnology. however astounding they may be. Drexler defies categorization in customary terms. an organization devoted to preparing the world for nanotechnology. but it will help to keep them distinct. He came to see the importance of Hayek’s thought for understanding the business world. We also got the chance to visit with Ted Kaehler. now with the Department of Computer Science at Stanford University. vice-president of Software Engineering at Xanadu and a professional writer of science fiction. Eric Drexler: formerly with the artificial intelligence laboratory at MIT. He was instrumental in bringing the Xanadu group to Palo Alto. As usual with such efforts. And yet these individuals are software engineers focused on practical concerns. but busy with the details of building systems that are immensely practical. He might best be described as a 1990s version of a Renaissance thinker. Mark has derived many insights about how to design computer programs from studying Hayek.” Ted is very interested in using market principles in the development of AI systems. We spoke to many other programmers on the Xanadu and AMIX teams. within each of which we found several specific possible projects. separating them by what each takes as its primary focus. a software engineer with Apple Computer Corporation and an expert on the field of Artificial Intelligence known as “machine learning. They are not just speculating. Miller: software engineer formerly with Xerox Palo Alto Research Center. These individuals are coauthors with us. flowing fountain of ideas and enthusiasm. the total output exceeds the sum of the inputs. and William Tulloh The three people most intensely involved with market process ideas are as follows: Mark S. the ability to achieve technological control over matter at the molecular level. He assists with the Xanadu design work. We are grateful to have been on the team. . Nanotechnology refers to the next step beyond microtechnology. for computer science and for society.

A given CAD–CAM (computer aided design– computer assisted manufacture) system. There is a qualitative difference between the computer as a medium of expression and clay or paper. High-tech Hayekians 97 Process-oriented case studies of the computer industry As with most media from which things are built. Studying software as capital may help us illuminate the market process approach to capital theory. but they are written in many different. a fugue or a word processor. and it views markets as disequilibrium processes. may require a particular computer with particular specifications. To understand clay is not to understand the pot. It is reproducible at negligible cost. performing tasks useful only in certain kinds of endeavors. architecture dominates material. It is to see in pots thrown by beginners (for all are beginners in the fledgling profession of computer science) the possibility of the Chinese porcelain and Limoges to come. Few kinds of tools today are more important than software. since it concentrates on the way knowledge is used in society. Software is a special kind of capital good with some economically interesting characteristics. not only do different programs accomplish entirely different tasks. Not the least important reason economics might study computation is that a growing and increasingly important part of the economy consists of the computer industry. Increasingly computers play a central role in the process of knowledge transmission. incompatible languages. But software does not fit the assumptions made about capital goods in mainstream theory. involving time and uncertainty. and a particular manufacturing device. its use by one agent does not preclude its use by another. it does not wear out when used. may in fact find this industry especially intriguing. Hence the task for someone who wants to understand software is not simply to see the pot instead of the clay. whether the thing is a cathedral. In the software industry the detailed timing of events is particularly important. the major costs associated with it are information costs (often neglected in neoclassical theorizing). The capital goods we call software are especially heterogeneous. and much of the capital “equipment” that counts today is in the form of instructions to computers. The patterns of complementarity of software-capital are exceptionally complex: most programs can run only on certain kinds of machines. Like the genetic apparatus of a living cell. What a pot is all about can be appreciated better by understanding the creators and users of the pot and their need both to inform the material with meaning and to extract meaning from the form. Such studies would be in keeping with the overall thrust of the research combining theoretical and empirical study that has been going forward at the Market Processes Center over the last several years. Change . Market process economics. Market process economics empha- sizes that capital is a complex structure. write and follow its own markings to levels of self-interpretation whose intellectual limits are still not understood. It is a cliché that we are entering the information age. a sonnet. One example of the kind of topic we have in mind here is considering software as capital. for software increasingly directs our “hard” tools. for example. a bacterium. (Alan Kay 1984: 53) One type of research project which frequently came up was simply to use the tools of market process economics to examine the specifics of the computer industry. the computer can read.

As. industry standards develop. It is essentially a pattern of on and off switches in the computer. As such. This is often the case whether or not that operating system is really the best available for the program’s purpose. new capital in a given period is defined as the unconsumed product of the previous period. Information technology and the evolution of knowledge and discourse Knowledge evolves. it is pure knowledge. “Interoperability” means making it possible for software that is built to one standard to be used with another standard via “software adapters. except perhaps for the Coca-Cola (aka “programming fluid”) and Twinkies the software engineer consumes as he generates the code. What are the economic causes and consequences of the evolution of a particular industry standard? What would be the economic consequences of more widespread interoperability? These are interesting questions worthy of research. They are additionally the embodiment of careful plans and much accumulated knowledge.” With interoperability. studying the computer industry seems likely to inform our understanding of economic processes from a new. to improve their chances of acceptance. Old programs and machines created at great cost and bought at great price become obsolete in months. and capital destruction. the first versions of new programs tend to be written for that operating system. for example. some particular operating system appears to be outstripping others in popularity. and William Tulloh is so rapid that a product not released in time is doomed. Conceiving of software as capital goods calls into serious question the common treatment of capital in economic models as accumulating solely out of saving. software developers have less at stake in using some less-preferred programming language or operating system that better suits their needs. In this way. now available in limited degree. the knowledge aspect of capital reaches its logical extreme. and media are important to the evolution of knowledge. In sum. The new capital is a combination of the physical wood and steel with this essential knowledge. as MS-DOS did in the microcomputer industry. Beer barrels and blast furnaces are more than the saved wood and steel from a previous period. With software. What becomes an industry standard is of immense importance in determining the path an industry takes. The knowledge and creativity of the programmer are crucial. say.98 Don Lavoie. Decisions to choose on the merits rather than by projections of others’ preferences will tend to influence the evolution of industry standards. occurs constantly. for instance. illuminating perspective. Hypertext publishing promises faster and less expensive means for expressing new . In many growth models. Other studies that fall within this research category include studies of the evolution of standards and of the development of interoperability. Software has virtually no physical being. This model implies that the most important thing new capital requires is previous physical output. as software developers bet on an uncertain future. No physical output from a previous period is required to produce it. through the advance of knowledge. Howard Baetjer. It ignores what the Austrian view emphasizes: the role of knowledge and creativity.

Professor Clearsighted can link her comments directly to Mistaken’s original article. One of the crucial questions which they focus on is the evolution of ideas in society. In this regard. original work may be linked downstream to subsequent articles and comments. They are particularly interested in the evolution of knowledge. to see it or an abstract of it. The improved cross-referencing of. Hypertext publishing should bring emergent benefits in forming intellectual communities. Suppose few see the flaws. making criticism more effective by letting readers see it. transmitting them to other people. in particular. References thus go upstream towards original articles. all sides of an issue should foster an improved evolution of knowledge. This refutation may be of great interest to those who read Mistaken’s original article. Anyone encountering a part of this literature will see references to Mistaken’s original article. A potential problem with this system of multidirectional linking is that the user may get buried underneath worthless “refutations” by crackpots. It will. which is one reason Hayek’s work is so attractive to them. Links. for instance. and be able. so that readers of Mistaken’s article may learn of the existence of the refutation. and extending the range and efficiency of intellectual effort. they hope. at the touch of a button. In a hypertext system such as that being developed at Xanadu. but with our present technology of publishing ideas on paper. The Xanadu system will include provisions for filtering systems whereby users may choose their own criteria for the kinds of cross-references to be brought to their attention. One important aspect of the limitations on information dispersal is the one-way nature of references in scholarly literature. But it may be that Mistaken’s article also provokes a devastating refutation by Professor Clearsighted. and indeed to the whole literature consequent on his original article. The supportive literature following Mistaken will cite Mistaken but either ignore Professor Clearsighted or minimize her refutations. Suppose Professor Mistaken writes a persuasive but deeply flawed article. High-tech Hayekians 99 ideas. while so many are persuaded that a large supportive literature results. building consensus. The market for scholarly ideas is now badly compartmentalized. there is no way for Mistaken’s readers to be alerted to the debunking provided by Clearsighted. and in the second place be able to see both (or more) sides of it with ease. (Drexler 1987: 16) Both the Xanadu and AMIX groups pay much attention to evolutionary processes. These devices would seem to overcome the possible problem of having charlatans . provide a market for information of a scientific and scholarly kind that will more closely approach the efficiency standards of our current markets for goods and services. The refutation by Clearsighted may similarly and easily be linked to Mistaken’s rejoinder. In our example. due to the nature of our institutions for dispersing information. and access to. the hypertext publishing system under development at Xanadu is of great importance. and evaluating them in a social context. will enable critics to attach their remarks to their targets. Scholars investigating this area of thought in a hypertext system would in the first place know that a controversy exists.

and projecting how advances such as hypertext might reshape it in the future. (This very sentence. In the first place. guided by price signals and constrained by general rules. was all but effortlessly inserted in the fourth draft of this article. Clearsighted’s responses can be available literally within hours of the publication of Mistaken’s original article. could more rapidly gain the support of fair-minded thinkers. most users would choose to filter out comments that others had adjudged valueless and comments by individuals with poor reputations. Third. before Clearsighted can get her responses accepted in a journal and then published. through word processing. Another difficulty of today’s paper publishing environment is that turn-around times are extensive. memory space. and computational services inside computers? Steel mills. insurance . Refutations of bad ideas could immediately become known. Second. it would provide a medium through which any number of economists might easily carry on open-ended written discussions with one another.100 Don Lavoie. In the early days of economics as a profession. one would have to pay a fee for each item published on the system. for instance. The research program for economists that is suggested by hypertext publishing falls within the field of the philosophy of science. In more recent years the advent of number-crunching computers has enabled the development of more complex mathematical modeling and econometric tech- niques. But today’s microcomputers. it would facilitate research by drastically reducing time spent among library shelves. The former attempts to make economic tradeoffs in a rational. It involves interpreting and explaining the effects of technology on the shape of economic research in the past. Mistaken’s persuasive but flawed article can be the last word on a subject for a year or so. it would enable easier reference to. Should one expect markets to be applicable to processor time. and to implement that plan through detailed central direction of productive activity. and more rapid dissemination of. Complexity. and accordingly most economics consisted of verbal arguments. have also begun to dramatically enhance economists’ ability to present verbal arguments.) The advent of hypertext might revitalize the rich literary tradition of economics in at least three ways. In the second place. Thus a hypertext system seems able to inhibit the spread of bad ideas at their very roots. with no necessity for retyping or cutting and pasting. very few will ever see his work. In a hypertext publishing environment. The latter allows economic tradeoffs to be made by local decision makers. book printing was the leading edge of technology. if sensible. by contrast.1 In other words. centrally- formulated plan. verbal arguments. with no inconvenience of time and distance. and unorthodox new ideas. though anyone could publish at will on a hypertext system. and William Tulloh clutter the system with nonsense. farms. Howard Baetjer. coordination and the evolution of programming practices Two extreme forms of organization are the command economy and the market economy. if one develops a bad reputation. Even then. First. many of those who have read Mistaken may not have ready access to the journal in which Clearsighted is published.

software firms – even vending machines – all provide their goods and services in a market context. a mechanism that spans so wide a range may well be stretched further. and using them to develop a better understanding of economics. Lionel Robbins. there is much that could be done in the direction of elaborating on these analogies between economics and programming practices. High-tech Hayekians 101 companies. Programmers are faced with difficult choices of how to make the best use of scarce computational resources. of computer programming practices as illustrative of economic principles. through property rights. explaining exactly what they are could be exceedingly valuable to both fields. and the possible future. Coping with complexity seems to depend on decentralization and on giving computational “objects” property rights in their data and algorithms. Programmers are now facing similar problems of complexity. The extensive use of subroutines and structured programming enhanced the . Coordination is achievable in complex economies only through decentralized decision-making processes: through specialization and the division of labor. the application of scarce means to valued ends. would involve examining the history. Hayek has recast the economic problem in terms of how societies may make effective use of the knowledge that is so widely dispersed among all the people they comprise. but with the evolution of programming. Many of the problems that arise in human economic systems have their analogs in well-known problems in computational systems. As the following paragraphs indicate. Gary Becker) as applicable in general to the study of choice. The division of labor and modularity An appreciation of the advantages of the division of labor is embodied in the programming principle of modularity.g. the making of decisions. Hayek has argued that coordination in complex systems such as human economies exceeds the capabilities of central planning and direction. undivided sequences of instructions. As programs and distributed computation systems grow larger. Ludwig Mises. The earliest programs were linear. A. Even if the differences are enormous. (Miller and Drexler 1988b: 137) The type of research that would be most directly a follow-up of the paper “Markets and Computation: Agoric Open Systems” by Miller and Drexler. Perhaps it will even come to depend on the use of price information about resource need and availability that can emerge from competitive bidding among those objects. and through the price system. Economics has been conceived by many of its practitioners (e. practical considerations forced a dividing up of the problem into discrete modules. they are outrunning the capacity of rational central planning. Inside computers things are going on which have some amazing similarities – and of course some significant differences – to what goes on in human economies. It doesn’t matter where one stands on the question of how similar computational and market processes are for one to see the possible value of this research program. F.

we might say. . In this setting the likelihood of mutual interference would be very great – analogous to the tragedy of the commons – without the “property rights” structure of OOPS. because the boundaries between objects are clear and respected. The integrity of the various data structures and algorithms in OOPS provides especially welcome clarity in very large programs written by teams of program- mers. they tend to be very dubious about the ability of government planners to develop successful systems in human society. Property rights and object-oriented programming The practical advantages that property rights give the economy can be provided for computer programs by what is called “object-oriented programming” (the common acronym in the computer language literature is OOPS. Another advantage is that if an object is directed to do something it cannot do. they need a certain amount of autonomy – if other parts of the program interfere with them in unexpected ways. The program’s “labor” is thus not only divided up among many parts. and William Tulloh ability of programmers to solve their problems. centrally planned systems to work properly. Programmers spend many frustrated hours debugging their own work. even though the programmer has absolute control over every aspect of the system.102 Don Lavoie. the more difficulty they have with bugs.” essentially autonomous sections of code whose workings cannot be interfered with by other parts of the program. For the various subroutines to operate effectively. They broke down the whole into manageable chunks. The bigger and more complex their own programs. One advantage to the programmer is that he need not hold in his mind all at once the myriad possible options and combinations of calculations and data. for object- oriented programming systems). Howard Baetjer. One subroutine’s data cannot. The reason is that programmers. Accordingly. The use of knowledge in computation At one point Roger Gregory. a founder of Xanadu. The programmer need not know how an object works. but the “rights” of these parts are respected. “intentionally” or by accident.” instead of acting on the bad instructions in a senseless or destructive way. interfere with data “belonging to” another subroutine. Or. cannot help but learn the severe difficulties in getting large. it simply returns a message that it “does not understand. The analogy to property rights is very close. In object-oriented programming. whose activities were known and clearly bounded. Virtually never does a large program work as intended the first time. where the complexity is far greater and the ability to control is far less. as it were. in their day-to-day experience. made a comment which helped us understand why his group is so interested in market process economics. the subroutines’ “rights” to what is “theirs” need to be respected. the result is a crashed program or nonsense. only what it does. the different kinds of tasks that the program must carry out are assigned to “objects.

High-tech Hayekians 103 What this suggests of course. programs should be designed so that their different parts would “bid competitively” for. both within particular programs running on single computers and across extended computational networks over large geographical areas. As in the economy. As Mises and Hayek have emphasized. but in computational systems: markets are an abstraction that need not be limited to societies of talking primates.” Derived from the Greek word for marketplace. The nature of complexity is arguably the central issue in computer science today. the more necessary it becomes that the orderliness of the system grow out of the interaction of relatively autonomous parts. it is necessary also that they be able to communicate their special knowledge of time and place. is Hayek’s line of thought that the more complex a system is. the various firms. Miller and Drexler wish to persuade the computer community to drop this central planning model for allocating computational resources. such as that between the speed at which the program completes its task with the core space it takes to do so. Programmers are beginning to realize that “central planning” of computational systems is fraught with the same difficulties. research centers and so on would bid for the computational goods they need. say. which would be more expensive per millisecond than disk space. Likewise. in large. Miller and Drexler are concerned about the efficiency of computer resource use. The most innovative aspect of Miller and Drexler’s papers is their introduction of “agoric systems. Presumably this bidding and asking . as it is in Hayekian economics. just as downtown property rents at a higher rate than rural land. albeit not in the same terms economists use. it is not enough that agents have property rights. individuals. Traditional programming practices have been based on a central planning approach. distributed systems. they are robust in the face of experimentation. In both cases they envision allocation of scarce computational resources such as disk or memory space and processor time – being determined by a market process among compu- tational objects. deliberately deciding on tradeoffs. Instead. So far we have been describing economic insights that are already widely appreciated by computer scientists. Computer time has traditionally been allocated on a time-share or first-come- first-served basis. agoric systems aim at solving the problem of maintaining coordination in complex computational systems by the same means as in complex economies: by a price system. they encourage cooperative relationships. They integrate knowledge from diverse sources. the reason that central planning of economies does not and cannot work is that human society is too complex. and they are inherently parallel in operation. or by some other fixed prioritizing system. All these properties are of value not just in society. the “rental” of memory space. As Miller and Drexler (1988b: 163) put it: Experience in human society and abstract analysis in economics both indicate that market mechanisms and price systems can be surprisingly effective in coordinating actions in complex systems.

the complexity of resource allocation problems would grow without limit. These firms would thus under- produce their services if they feared that their products could be easily copied by any who used them. The challenge here is to develop secure operating systems. given a secure operating system. Owners of computational resources. The box is inaccessible to anyone. the system in effect would create a virtual black box in Gamma’s computer. which might be used in a variety of applications. and it self-destructs once the desired results have been forwarded to the George Mason researcher. Royalties would be paid to the owners of given objects. We might find certain computational resources in extremely high demand. In their articles. Norm Hardy. Such open systems. and William Tulloh would be carried out by the computers themselves. In independent work. Suppose. changing their prices several times a day – or several times a second. Programmers would be able to develop software by adding their own algorithms on to existing . whom we met at Xanadu. Secure property rights are essential.104 Don Lavoie. that Beta cannot steal Alpha’s data set. Imagining computation markets of the future Miller and Drexler envision the evolution of what they call agoric open systems – extensive networks of computer resources interacting according to market signals. To return to the above example. Howard Baetjer. on a superfast computer owned by Gamma Processing Services. need to be able to sell the services of their agents without having the algorithm itself be copyable. senior scientist of Key Logic Corporation. free of the encumbrances of central planners. such as agents containing algorithms. In the sort of agoric open systems envisioned by Miller and Drexler. would most likely evolve swiftly and in unexpected ways. which could be sold on a per- use basis. but it would also facilitate the discovery of new knowledge and the development of new resources. has developed an operating system called KeyKOS which accomplishes what many suspected to be impossible: it assures by some technical means (itself an important patented invention) the integrity of computational resources in an open. part of the problem apparently has already been overcome. or in volatile markets. Given secure property rights and price information to indicate profit opportunities. for example. The operating system needs to assure that Alpha cannot steal Beta’s algorithms. and that neither Gamma or the George Mason researcher can steal either. there would be a vigorous market for computational resources. in which Alpha’s data and Beta’s algorithms are combined. entrepreneurs could be expected to develop and market new software and information services quite rapidly. that a researcher at George Mason University wanted to purchase the use of a proprietary data set from Alpha Data Corporation and massage that data with proprietary algorithms marketed by Beta Statistical Services. interconnected system. Within vast computational networks. according to pre-programmed instructions. Miller and Drexler propose a number of ways in which this problem might be overcome. Not only would a price system be indispensible to the efficient allocation of resources within such networks.

The traditional approaches to AI. precisely the aspects of intelligence that market process economists consider the most important. 1962).” takes intelligence to be an algorithmic. however. later Wittgenstein. however. and Rosenblatt were right in thinking that we behave intelligently in the world without having a theory of that world. Worse. there must be elements and principles in terms of which one can account for the intellegibility of any domain.” Indeed. This AI does not suggest that the computer be given rules to follow but tries to set up a system of independent elements within a computer from whose interactions intelligence is expected to emerge. a revolution has been occurring in AI researchers’ thinking about thinking. mechanical process. sometimes called Emergent AI. and Frank Rosenblatt (1958. Neural networks may show that Heidegger. conceive of mental processes as complex. whom Hayek cites favorably in The Sensory Order. spontaneous ordering processes. but they are today making a dramatic comeback. Husserl. They would not need to understand all the workings of what they use. Emergent AI traces its origins to early contributions to neural networks such as those of Donald O. These efforts had at one time been discredited by the more rationalistic approaches. one may have to give up the more basic intuition at the source of philosophy that there must be a theory of every aspect of reality – that is. have proven to be the most difficult to produce artificially. they have been extremely disappointing in terms of their ability to exhibit anything that deserves the name “intelligence. The ideas about machine intelligence that it puts forward are not so much about teaching the computer as about allowing the machine to learn. Most important. this would save the tremendous amount of time now used by programmers in the trivial redevelopment of capabilities that have already been well worked out. As Sherry Turkle (1989: 247–8) put it in an article contrasting “the two AIs”: Emergent Al has not been inspired by the orderly terrain of logic. and early Wittgenstein that the only way to produce intelligent behavior is to mirror the world with a formal theory in the mind. . Over the past decade. and imagination. Although there are many commercially successful applications of these traditional AI systems. researchers will have to give up the conviction of Descartes. (Dreyfus and Dreyfus 1989: 35) The field of computer science that market process economists would be apt to find the most fascinating is Artificial Intelligence (AI). creativity. Mind as a spontaneous order: what is (artificial) intelligence? If multilayered networks succeed in fulfilling their promise. Among other advantages. such as learning. only the results. still dominant in the specialization area known as “Expert Systems. Hebb (1949). is the increased rapidity with which new products could be developed. The newer approaches. High-tech Hayekians 105 algorithms.

Howard Baetjer. This approach tries to introduce market bidding processes and monetary cost calculations into a neural network model in order to generate artificial learning. It is too early to tell whether these approaches to AI will be more successful in replicating learning.106 Don Lavoie. and the connectionist or neural networks approaches. and contemporary phenomenology and hermeneutics – have been used by the newer AI researchers as the basis of their critique of Cartesian rationalism. Even if one believes many current approaches to AI are utterly unworkable. For example. competitive process. and payments. A significant faction of the AI community thus finds itself arriving at essentially the same conclusions about the nature of the human mind as Austrian economists. “To do this we treat each rule as a ‘middleman’ in a complex economy” (Holland et al. John Holland makes explicit use of market processes for reinforcing a program’s successful rules of action. for example in his essays “Rules. creativity. in a system they named Derby. and so forth than the tradi- tional approaches. This type of research involves using economic principles to advance AI research. the later Wittgenstein. inductive thinking. and as such may display some interesting analogies to market orders.” Some of the interesting work that has been done within Al directly introduces market principles into the design of models of learning. In his work on classifier systems. Hayek draws significant methodological conclusions from his view of the nature of the mind. and William Tulloh The critiques in the new AI literature of the failings of the rationalist approach to AI sound remarkably similar to Hayekian economists’ critiques of the rational choice model. The process of mind is a decentralized. but to see what it can teach us economists about the nature of human intelligence. and Terry Winograd and Fernando Flores’s Understanding Computers and Cognition. Books critical of mainstream AI. No CPU exists in the brain. consumers.” he says. “We use competition as the vehicle for credit assignment. Understanding how the human mind works is not only of interest in that the mind is conceived as a specific instance of spontaneous order processes. there are the “genetic algorithms” and classifier systems approaches developed by John Holland. such as Hubert Dreyfus’s What Computers Can’t Do. the point is not so much to help AI achieve its aspirations. A great deal of work has been done in the new Emergent AI literature that simulates complex ordering processes. Perception and Intelligibility” and “The Primacy of the Abstract. There is also the . Even the very same philosophical traditions market process economists have used – post-Kuhnian philosophy of science. one from which we could borrow in our own efforts to challenge mainstream economics. bids. but our interest in it is not primarily altruistic. He speaks of suppliers. Some initial research along these lines has been done by Ted Kaehler and Hadon Nash at Apple. their very failures might be worthy of closer study. The fact that mainstream AI research has not yet been able to reproduce certain key aspects of human intelligence may be highly significant. Marvin Minsky’s conception of the way the mind works in The Society of Mind is remarkably similar to Hayek’s in The Sensory Order. 1986: 380). present a powerful critique of the rational choice model. but it is already clear that the emergent AI approach is able to do some things the older approaches could not. capital.

and what properties do they share? Computation as a discovery procedure: possibilities for agoric mental experiments Underlying our approach to this subject is our conviction that “computer science” is not a science and that its significance has little to do with computers.” Up to now we have been talking about applying more or less standard market process economics to some computer-oriented topics. and to market processes. market process economists think an alternative approach to the subject matter is necessary precisely because neoclassical economics has tried to subsume action into a mechanistic model. as it were. is an interplay of purposeful human minds. for a subjectivist approach to economics. as opposed to the more declarative point of view taken by classical mathematical subjects. To what extent have genuine learning and creativity been simulated in this research? How much does it matter that natural language simulation has not yet gotten very far? Just how different are these different levels of spontaneous order.2 The idea here is that we could try to improve our understanding of the market order by developing spontaneous order simulations on a computer. The essence of this change is the emergence of what might best be called procedural epistemology – the study of the structure of knowledge from an imperative point of view. is wonderfully informative on these issues. Even the skeptic about AI research.” (Abelson and Sussman 1985: xvi) Perhaps the research direction we talked about which will prove to be most controversial involves not so much using economics to study computers as the other way around. in its failures and its achievements. and computer modeling in general. it might be asked. High-tech Hayekians 107 point that. Mathematics provides a framework for dealing precisely with notions of “what is. The computer revolution is a revolution in the way we think and in the way we express what we think. the direct use of computer modeling techniques to develop economic theory. We might be able at least to illuminate existing ideas in market process economics. could a school which stresses the non-mechanical nature of human action find itself trying to simulate action on electronic machines? After all. and we might conceivably develop substantive new ones. what the market order is. This would be a matter of expanding on ideas sketched in another of Miller and Drexler’s papers. as we advance from biological to animal and human cog- nitive processes. by doing mental experiments within artificial minds. AI. But what we mean by “mechanical” has been shaped by the age when machines . We need to know as much as we can about how human beings think and communicate for our substantive economics. Those first topics could be done in words. could see that these simulation methods raise some provocative research questions for market process economics.” Computation provides a framework for dealing precisely with notions of “how to. How. “Comparative Ecology: A Computational Perspective. but this one could also involve actually doing some computer programming of our own.

because it can deal with shapes rather than quantities. Thus it could constitute a useful new complement to the traditional procedures of theorizing that market process economists now employ. Market process oriented economists have often pointed out that the mathematics of differential calculus that has played such a central role in mainstream economics is not the appropriate mathematics for studying the economy. and William Tulloh were extremely crude and rigid things.108 Don Lavoie. As the previous section pointed out. As computers advance. has suggested that game theory. they increasingly challenge our ideas about what “machines” are capable of. we suspect that the appropriate formalization for economics might not be a mathematics at all. If the aspects of human action that the school considers the most important cannot be handled with the modeling tools of the mainstream. Computer programming may constitute the kind of formalization most conducive to the development of market process theory (see note 2). or of the overall economy. Mathematical modeling has often distracted economics from paying attention to the processes that market process economics emphasizes. such as creativity and learning from experience. It is a formal medium for articulating the “how to” of dynamic processes. Without disagreeing with the reasons game theory and topology might be useful formalizations. The market process approach has never been against models as such. and see what can be done with them. and the “agents” would be only “artificially” intelligent. it may be necessary to devise better tools. We have to admit at this point that we are not at all sure ourselves whether market process economics can usefully deploy computer simulation methods. which is to say. Howard Baetjer. Computer “simulations” of spon- taneous order processes might prove to be the kind of modeling approach that is process-oriented enough to help rather than obstruct economic theorizing. It certainly has nothing to do with efforts to build direct simulations of specific real world markets. since it can deal with the interplay of strategic choices through time. they would be rather stupid. In principle. James Buchanan. in which market process economics is interested. but has only objected to modeling that loses sight of certain non-mechanical aspects of human choice. But these agents may still be more like humans than the optimizing agent of mainstream neoclassical theorizing: they may be equipped with the capacity to learn from experience. there is no reason why computers themselves could not become sufficiently “non- mechanistic” to be of interest to market process economics. But there would be no pretensions of capturing the complexities of reality within a model. for example. rather the “what is” of timeless end-states with which mathematics is concerned. Kenneth Boulding has pointed to topology as an alternative mathematics. But the only way to tell if computer simulations can serve as such tools would be to try to build some. . It is important to be clear about just what is meant by “simulation” here. would constitute a more appropriate mathe- matics. at this point in AI research. It would not be empirical but theoretical research. The “worlds” in the computer would be radically simplified. research in AI aspires to reproducing on electronic machines exactly the sorts of phenomena. The notion here is rather of using computers as an aid for conducting complicated mental experiments. or of being able to derive predictions about reality directly from the simulation exercises.

There.” Economists get preoccu- pied with the game of modeling itself. And we. for example. and in theoretical ecology. Business cycle effects might be studied: could we. Imagine. and how industries evolve as markets and technology change. than most theoretical research is today. The mental experiment was not just a game played for its own sake. in a sense. This kind of simulation would differ from most contemporary theorizing. in that the purpose of the modeling exercise would not be to devise a whole deterministic mechanism. but a heuristic vehicle. the programmers. the iterative game theory work of Robert Axelrod is very much the kind of research we are thinking of here. but by the spontaneous interplay of its component parts. High-tech Hayekians 109 On the other hand. In industrial organization. that are being used in AI. could be reinterpreted as referring to economic insti- tutions or economies instead of brains or species. What different emerging properties of the economic order would we observe? Of course we need to be careful about the unrealistic nature of these mental experiments. for example. trying to contrive Mengerian simulations in which a medium of exchange spontaneously evolves. the computer tournament was couched in a substantive interpretive effort. and not let the modeling become an end in itself. observe changes in a simulated capital structure as a result of an injection of credit? We might probe constitutional economics by running a series of parallel simulations differing only in certain fundamental rules such as property rights and contract sanctity. and then to run the simulation in order to see what happens. which we might critically examine as illuminating the properties of spontaneous order processes. The various evolutionary process modeling strategies mentioned in the previous section. One of the first projects that needs to be undertaken is to see just how much there is in existing computer modeling. . or free banking simulations that evolve clearinghouses and stable monies. it is to run the simulations as mental experiments. it would be more experimental. it might be possible to investigate how firm sizes vary with different industry characteristics. Although most game theory in economics suffers as much from formalism as general equilibrium theory and macroeconomics do. Rather. Or completely new programs could be developed with markets more directly in mind. inside or outside of economics. specifying institutional environments or decision rules for agents. The crippling vice of most economic theory today is its “model-fetishism. Existing program designs could be examined from a Hayekian perspective and modified in order to illuminate selected properties of spontaneous order processes. or general equilibrium theory in microeconomics. a mental experiment to help us to think about the evolution of human cooperation. The order would emerge not by the programmer’s design. would not know how the process was going to come out until we ran the mental experiments. Rather. in theoretical evolutionary biology. the aim would be to set up constraining conditions. and forget the need to interpret the mental experiment. such as is the goal of IS/LM and Rational Expectations models in macroeconomics. where what is of interest is not what the end results are so much as how the process works. The idea is not to create a mathematical model that already implies its conclusions in its premises.

as does most economic experimentation. Permission to reproduce this text is gratefully acknowledged. Conceivably. it is “a novel formal medium for expressing ideas about methodology. and only incidentally for machines to execute. a computer program is not just a way of getting a computer to do something. As Ted Kaehler pointed out. programs must be written for people to read. Notes Copyright 1990. but there are other advantages of computer experiments which suggest that some interesting theory development might be possible along these lines. and we certainly encourage computer scientists to follow-up directly on their work. and William Tulloh Other than Axelrod’s work. probably the closest thing in the existing eco- nomics literature to this kind of simulation would be what is called experimental economics. we economists might be able to introduce theoretical principles to computer scientists that could help them address their problems. will undoubtedly have many useful ideas for us along these lines. such as Vernon Smith and Ron Heiner. this is a step down in many dimensions of intelligence. but otherwise we believe this kind of theoretical spontaneous order simulation is unexplored territory for market process economics. and for royalty payments per use to all authors. 1 Plans for Xanadu also include provisions for reader feedback as to the value of different articles or notes. Thus. we can expand the interplay between theoretical insights and empirical praxis. But a more likely benefit of our taking up these questions is that. Howard Baetjer. As Abelson and Sussman (1985: xv) put it in their classic textbook on programing.110 Don Lavoie. 2 This is not necessarily as radical a change from writing “in words” as we are implying here. we could use artificial agents. even of short comments. Computer programs are written in order to be read by people. and not merely to be run on machines. what we are really interested in.” Appendix: annotated bibliography The interface of computer science and market process economics provides a wealth of exciting research opportunities. we . By studying both the application of market process ideas to computer science and the ways in which developments in computer science enrich our understanding of market processes. the use of non-human agents would in many instances allow for greater flexibility in the design of spontaneous order simulations. Drexler. after all. and Salin deserve our thanks for introducing Hayekian ideas to the computer science community. Whereas our simulations would not use human subjects. Miller. the experimental economists’ focus on the design and functioning of market institutions is very much in the spirit of what we have in mind. lnstead of using rats. by applying our existing economics to the study of computational processes. Moreover. In this bibliographical note. The more Hayekian contributors to the experimental literature. which is. we might help improve our economics. and that may help us think more clearly about human economies. Mercatus Center (formerly Center for the Study of Market Processes).

provides fertile soil for studying such traditional market process concerns as the role of entrepreneurship (Kirzner 1973). For many theorists. and legislative solutions to such problems as the assignment of property rights to intangible goods. but also is one that is ushering in new forms of market interaction in a broad variety of industries. Mink and Ellig 1989. Process-oriented case studies of the computer industry The computer industry. the blurring of the distinctions between the various communications media (print. In addition. whether as the outcome of the market process. A further example of the important dynamic interactions between technical and policy responses to change in the computer industry can be found in the emergence of industry standards. or imposed by legislation. as evidenced by the recent breakup of AT&T. The interplay of technological. An additional area of interest to market process economists is the role that computers have played in the socialist calculation debate. is an exciting and relatively unexplored research opportunity. Katz and Shapiro 1985). the production and use of knowledge and information (Hayek 1948). the rapid evolution of information and communication tech- nologies and the peculiarities of information as a good underscores the important relationship between the legal environment and technical change (Liggio and Palmer 1988). and market competition (Huber 1987. however. the process of standard formation is constantly being threatened by the entrepreneurial discovery process (Hayek 1978a). public policy. The problems of compatibility and interoperability between various software and hardware components highlights the role of market dialogue in the shaping of expectations and the formation of consensus in reaching an industry standard. is not only an increasingly important sector in its own right. The emergence of electronic markets (Malone et al. the broadcast spectrum and intellectual property (Palmer 1988). Moreover. for example. Pool 1983). we hope that it will provide a useful starting point for those wishing to pursue these research opportunities. Gould 1987). as well as computer-aided buying and selling through home-shopping systems. In addition. and the peculiarities of software as capital (Lachmann 1978). has become an important research and policy topic. highlights the relationship between innovation. as the dominance of the technologically inferior QWERTY keyboard demonstrates (David 1986. broadcasting and common carriers). with its fast growth and rapid technological change. program trading on increasingly interlinked electronic exchanges. The computer industry. While this list is far from exhaustive. High-tech Hayekians 111 hope to alert interested readers to some of these sources. leading to the search for technical adapters and innovations which would make such standards obsolete. More broadly. In light of recent trends in computer . 1987. or through agreement forged in standard committees (David. this process may exhibit a high degree of historical path dependency. the computer was seen as the answer to the challenge of Mises and Hayek to the workability of central planning (Lange 1967). 1989) in such areas as computerized airline registration systems. common-law. contractual.

a central problem that emerges is how to cope with the rapidly expanding complexity of software systems. The techniques used to grapple with the intellectual complexity of large software systems (Brooks 1975) could be usefully studied in comparison to techniques in economics. McCloskey 1985. the bold claims of future computopians takes on an ironic twist (Lavoie 1990). 1987. as computers become more widespread and interconnected. and computer systems more powerful. These developments could have important implications for the spread of economic ideas. and use of local knowledge have their counterparts in each system. . Complexity. Such issues and techniques as modularity. Howard Baetjer. coordination. While the complexity of an economic system is obviously much greater than even the most complex software system (Hayek 1967b). Insights from the economics of property rights could perhaps be usefully applied to the development of software systems (Anderson and Hill 1975. the use of computer-aided dialogue and communication could have important repercussions on the evolution of knowledge and scientific discourse (Bush 1945).112 Don Lavoie. they could enhance the evolution of scientific knowledge through more rapid dissemination of knowledge. These new knowledge media (Stefik 1988) could have an effect not seen since Gutenberg. As hypertext systems emerge (Drexler 1986. Hayek 1989). Nelson 1973). Miller and Drexler have usefully applied the analogy between economic institutions and programming practices through their insight that the latest development in pro- gramming languages (object oriented programming) can be viewed as being the reinvention of the concept of property rights in the domain of computer science (Miller and Drexler 1988b). Computer programming is properly seen as a medium of expression of ideas. specialization. Never- theless. As Alan Kay expresses it. a product of the human mind (Ableson and Sussman 1985). the methods used to maintain intellectual coherence and to improve the use of knowledge may have analogs in each system. embodied knowledge. In computer systems the knowledge embodied in the software is more important than the physical hardware upon which it runs. “architecture dominates material” (Kay 1984). and the pattern of discourse within economics (Colander and Coats 1989. and the evolution of programming practices As software applications become more advanced. and more efficient means of criticism and debate. and William Tulloh science towards decentralized means of coordination. the reality seems to move much slower than the promised dramatic effects. Demsetz 1967. Information technology and the evolution of knowledge and discourse While it is commonplace to hear how computers will transform our society. Tullock 1965). A recent attempt to implement a version of a computopian plan in Chile (Beer 1975) could be usefully examined by market process economists.

Perhaps the best known are the connectionist or neural network approaches. Object-oriented programs perform computations by passing messages between various objects. Agoric open system can take advantage of object oriented programming’s ability to provide opportunities for easy reuse and recombination of components. as well as the coupling of data and procedures. An important aspect of these open. Simon 1983. decision-making process has led to a shift towards a view of intelligence as being an emergent property of a complex learning process (Drexler 1989. The mind is viewed as being subject to competitive and cooperative pressures like other complex. Miller and Drexler 1988b). As Miller and Drexler point out. The inability of humans to fulfill the demands of the optimizing agents has become increasingly recognized (Kahneman et al. Thomas 1989). Hewett 1985. much has been learned by the attempt. their complexity is sure to increase as computation becomes more distributed across networks of hetero- geneous computers with different users pursuing their own particular goals. and has been instrumental in bringing to the attention of economists the nature of the computational demands placed upon their perfectly optimizing agents. An additional aspect of the combination of large distributed systems and object oriented programming is the promise it holds for the more rapid evolution of software applications (Drexler 1988. While current software systems can be very complex. 1989. Miller et al. and incremental improvements. 1985). 1987). However. The work of Herbert Simon. they have had little success in creating anything resembling intelligence. While these attempts have yielded many interesting applications. decentralized systems will be the need to maintain the security of the proprietary data and software of the different agents (Hardy 1988. central planning techniques are no longer adequate for coordination of these open systems. 1982) as well as the implications that these less than perfect agents have for economic theory (Heiner 1983). which can be viewed in terms of their real-world analog (Cox 1986. High-tech Hayekians 113 Object oriented programming techniques such as encapsulation that allows for the separation of internal state from external behavior. which attempt to mimic the . and a more decentralized coordinating mechanism is needed (Bond and Glassner 1988. promise to expand the range of a programmer’s conceptual control. Kahn and Miller 1988). evolving systems. Shriver and Wegner 1988. has been influential in both economics and computer science (Newell and Simon 1972. Graubard 1989). of course. A variety of metaphors has been explored in attempting to create an emergent approach to artificial intelligence. Mind as a spontaneous order: what is (artificial) intelligence? A further area of interest for market process economists refers to the insights regarding the nature of rationality that have been achieved through the attempt to create artificially intelligent computers. Huberman 1988. The mind is seen as a spontaneous order process in which the resulting intelligence is greater than is possible by design (Hayek 1952). Economists have much to gain from the successes and failures of artificial intelligence. The limitations of attempting to design intelligence as a mechanistic.

and William Tulloh neural process of the brain itself. Hillis 1985. in capturing the tacit and context-dependent nature of knowledge (Polanyi 1962). computer scientists have attempted to apply social metaphors. 1988). These critiques have emphasized the important role that language and social interaction play in intelligence. One of the closest examples to the approach being suggested here is the work of Robert Axelrod on the evolution of cooperation. and empirical case studies has proved to be both influential and illuminating (Axelrod 1984). and an economic model for the problem of credit assignment (Booker et al. De Jong 1988.1987). to the neural metaphor.114 Don Lavoie. Miller 1989). as well as the nature of intelligence. a computer tournament. 1989. In addition. His work has inspired a wide range of further empirical work and theoretical insights (Axelrod and Dion 1988). In addition. These approaches include the Eurisko project of Doug Lenat (Lenat 1983. Reeke and Edelman 1989. has attempted to build a parallel rule-based learning system that combines the rule discovery properties of genetic algorithms. 1986. and the genetic algorithm approach pioneered by John Holland (Booker et al. Another area that highlights the importance of the dialogic interplay between theory and empirical observation is the fast-growing field of experimental . McClelland and Rumelhart 1986. A wide variety of connectionist approaches are currently being attempted (Cowan and Sharp 1989. The goal of this type of modeling is not a predictive model that tries to simulate reality. The failure of traditional Cartesian rationalist approaches to artificial intelligence has prompted a critique similar to much of the market process economists’ critique of neoclassical rationality (Dreyfus 1972. Howard Baetjer. Winograd and Flores 1986). including the use of genetic algorithms to generate strategies that in certain situations improved on the performance of “tit for tat” (Axelrod 1987. Kornfield and Hewett 1981. 1989. Computation as a discovery procedure: possibilities of agoric mental experiments The advances in artificial intelligence and computer programming suggest that these techniques could be usefully applied to experimental modeling the complex processes of interaction that occur in economic systems. Holland 1986. the classifier system. 1989. as well as the limitations of the knowledge as representation approach (Dascal 1989). including an attempt that applies some of the agoric insights to the problem of attributing success to various “neurons” (Kaehler et al. Holland et al. Edelman 1987. Lenat and Brown 1988). Genetic and evolutionary analogies from biology have also been influential (Langton. recognizing the greater social intelligence (Lavoie 1985) that emerges out of the interaction of less intelligent parts (Campbell 1989. Minsky 1986). These developments in computer science may improve our understanding of both a wide variety of spontaneous order processes. Dreyfus and Dreyfus 1985. also pioneered by John Holland. Holland 1975). but rather mental experiments to help us better understand spontaneous order processes. J. Axelrod’s mixture of theoretical insights. 1986). Schwartz 1989). Goldberg 1989.

the experimental results have demonstrated the importance played by the exchange institutions – the system of property rights in communication and exchange. Further work in developing “agoric mental experiments” can begin by examining the current work in computer science that uses market principles. While the majority of experiments to date has focused primarily on the relatively straightforward auction-type institutions. Anderson et al. a group at the Sante Fe Institute has already combined Axelrod’s computer tournament model with the experimental focus on market institutions. Smith 1982). Huberman and Lumer 1989. Computer-aided experimental markets. Market-based models for computation have also been explored by Tom Malone at MIT (Malone 1988. have developed a computerized market allocation mechanism called Spawn (Waldspurger et al. not on the dynamic adjustment and learning processes (Heiner 1985). Nelson and Winter 1982. Huberman and Hogg 1988. 1989). Day 1987. Horwitz 1989). and Ferguson (1989). High-tech Hayekians 115 economics (Plott 1982. 1986. Bernardo Huberman and his coworkers at Xerox PARC. The development of order analysis. This increasing interest in evolutionary approaches to economics (Allen 1988. such as the computerized double auction mechanism (PLATO). 1989). The complexity of keeping track of these adjustment processes suggests a fruitful role for computers. and the nature of the rule systems and institutional order that help guide the ordering processes (Brennan and Buchanan 1985. “it is not possible to design a resource allocation experiment without designing an institution in all of its detail” (1982: 923). and on static outcomes. b). 1989a. Prigogine and Stengers 1984). Motivating much of this interest in computer modeling of spontaneous order processes is a dissatisfaction with traditional equilibrium approaches to capturing the evolutionary and self-ordering aspects of the market process. Malone et al. . building on the work of Drexler and Miller (1988). 1989a. Hayek 1973. b. Holland 1988. have already helped to illuminate these dynamic processes (Smith et al. 1989). as an alternative to equilibrium-bound theorizing can be enhanced by our better understanding the working of spontaneous order processes (Boettke et al. 1988). and by the Agoric Research Group at George Mason University (De Garis and Lavoie 1989). Buchanan 1986. by running a computerized double auction tournament (Rust et al. Buchanan 1982. Nascent attempts to apply these computer techniques to economics have been attempted by members of the Sante Fe Institute (Marimon et al. 1988. Silverberg 1988) has been fueled in part by the developments in the new views of science (Lavoie 1989. 1988). Furthermore. As Vernon Smith notes. Kephart et al. This detailed focus on the institutional constraints is perhaps the most valuable aspect of the experi- mental approach. Huberman 1988. The focus to date in this young field has been on relatively simple market institutions. Langlois 1986). 1989. and have explored the dynamic properties of distributed computer systems (Cecatto and Huberman 1989.

Stoneman (eds) Economic Policy and Technological Performance. New York: Basic Books. L. T. (1986) Liberty..J. Cambridge. MA: Addison- Wesley. Market and State. Booker. J. Arrow. D.” Market Process 4(2). V. MA: Addison-Wesley. May. New York: Printers Publishers.B. and Prychitko.L. innovation. (eds) (1988) Readings in Distributed Artificial Intelligence. 5(4). A. David. H. and Buchanan.. Axelrod. Beer. Jr. April.J. Cambridge: Cambridge University Press. and economics. J. and Hill. G.” in W.W. Reading. P. Dosi et al.116 Don Lavoie. (1986) “Beyond equilibrium economics: Reflections on the uniqueness of the Austrian tradition. G.) Genetic Algorithms and Simulated Annealing. (1989) “Classifier systems and genetic algorithms. CA: Addison-Wesley.D. (1945) “As we may think.J. and Huberman. and Sharp.” Proceedings of the National Academy of Science 86.R. (1988) “The further evolution of cooperation. (1975) The Mythical Man-Month. 9 December. Winter. B. Buchanan. Dasgupta and P. and William Tulloh References Abelson.” in G.A.) The Artificial Intelligence Debate: False Starts.L. Ceccatto. (1975) Platform for Change. D. P. and Pines. July. Redwood City. Cambridge: Cambridge University Press. Reading. (1986) Object Oriented Programming: An Evolutionary Approach. New York: Wiley. Fall. P.M. L. Boettke.H.E.M.H. CA: Morgan Kaufmann. Axelrod (1987) “The evolution of strategies in the iterated prisoner’s dilemma.” Systems Research 6(1). and Dion. D.” in L. and Glasner. (1989) “Artificial intelligence and philosophy: The knowledge of representation. and Coats. Colander. Anderson. Liberty. New York: New York University Press.M. Graubard (ed. Dascal. reprinted in J. P. Cowan.) Economic History and the Modern Economist. D.W. J. K. (1984) The Evolution of Cooperation. (eds) (1988) The Economy as an Evolving Complex System. Allen. (eds). S. (eds) (1989) The Spread of Economic Ideas. David. P. (1989) The Improbable Machine: What the Upheavals in Artificial Intelligence Research Reveal about How the Mind Really Works. F. Parker (ed.” Atlantic Monthly. (1985) Structure and Interpretation of Computer Programs. D. Cambridge: Cambridge University Press. Bush. S. Oxford: Basil Blackwell. Horwitz. and Sussman. R. D. New York: Simon and Schuster. Los Altos. and Holland. Brooks. Brennan. H.” Literature of Liberty.P. Goldberg.H. Market and State: Political Economy in the 80’s.” in P. .A. J. (1989) “Neural nets and artificial intelligence” in S. Cambridge.” Journal of Law and Economics. Economic Theory and Technical Change. New York: New York University Press. Davis (ed. Anderson. M. Bond.D. MA: MIT Press. Campbell.. A. Buchanan. Cox. (1988) “Evolution. (1985) The Reason of Rules: Constitutional Political Economy.. P. (1987) “Some new standards for the economics of standardization in the information age. J. 1986. MA: MIT Press.J. Axelrod. San Mateo.” Science 242.. (1986) “Understanding the economics of QWERTY: The necessity of history. (1989) “Persistence of nonoptimal strategies. (1975) “The evolution of property rights: A study of the American West. J.. Howard Baetjer.M. B. CA: Morgan Kaufmann.” Artificial Intelligence 40. (1982) “Order defined in the process of its emergence. Real Foundations. Buchanan. R.

Drexler.” in B. G. Hayek. and Dreyfus.” European Journal of Operations Research 30. (ed.M. Real Foundations. (1978a) “Competition as a discovery procedure.E. H. Redwood City. Optimization. F.” chapter 2 in Law. New York: Doubleday. CA: Addison-Wesley.A. (1988) “Biological and nanomechanical systems: contrasts in evolutionary capacity. Chicago: University of Chicago Press.A. New York: The Free Press.E.) (1989) The Artificial Intelligence Debate: False Starts. S. (1989) The Fatal Conceit: The Errors of Socialism. Philosophy. and Dreyfus.” Ph. perception and intelligibility. Hayek. Drexler. K. Dreyfus. (1978b) “The primacy of the abstract. (1973) “Cosmos and taxis. S.” Expert. R. New York: Basic Books. (1967a) “Rules. Chicago: University of Chicago Press. thesis. Hayek.” in Individualism and Economic Order. De Jong. CA: Foresight Institute. Chicago: University of Chicago Press. Cambridge. Philosophy and Economics. Hayek.A. Chicago: University of Chicago Press. June.J.” in New Studies in Politics. Reading.E. Langton (ed.” manuscript.A. (1989) “Making a mind versus modelling the brain: Artificial intelligence back at a branchpoint. Philosophy and Economics. (1987) Hypertext Publishing and the Evolution of Knowledge.” in S.A. Columbia University. and Dreyfus. K. Hayek.” American Economic Review.A. H. (1989) “Evolution of the money concept: A computational model. D. MA: Addison-Wesley.A. S. and Machine Learning. Drexler.” Natural History. Santa Clara. Hayek.E. F. Legislation and Liberty: Rules and Order. K.” Machine Learning 3. D. (1988) “Learning with genetic algorithms. K. MA: MIT Press. Hayek.S. F. (1948) “The use of knowledge in society. Huberman (ed.” in Studies in Politics.E. K. Ferguson. and Lavoie. Chicago: University of Chicago Press.A. Amsterdam: North- Holland. (1987) “The panda’s thumb of technology. F. M. Edelman. Chicago: University of Chicago Press. K. Graubard. K.E. George Mason University.D. May. Real Foundations. Drexler. K. De Garis. S. (1988) “Incentive engineering for computational resource management. (1972) What Computers Can’t Do: A Critique of Artificial Reason. (1985) Mind over Machine. (1986) “Why expert systems do not exhibit expertise. (1989) “The application of microeconomics to the design of resource allocation and control algorithms. Drexler. (1987) Neural Darwinism: The Theory of Neuronal Group Selection.A. Drexler. Hayek. (1986) Engines of Creation: The Coming Era of Nanotechnology.E. D.L. F. Goldberg. Hardy. S. Drexler. Chicago: University of Chicago Press. (1988) “Key KOS architecture. (1967b) “The theory of complex phenomena. Gould. N.) The Ecology of Computation. CA: Key Logic.” in C. Drexler. Economics and the History of Ideas. 1. High-tech Hayekians 117 Day. Graubard (ed.” in New Studies in Politics.) The Artificial Intelligence Debate: False Starts. (1967) “Towards a theory of property rights.E. Chicago: Universityof Chicago Press. Palo Alto. New York: Harper and Row. Economics and the History of Ideas.” Foresight Update 5.) Artificial Life.E. MA: MIT Press. F.E. March.E. (1952) The Sensory Order: An Inquiry into the Foundations of Theoretical Psychology.R.” in Studies in Politics. (1989) Genetic Algorithms in Search. vol. (1987) “The evolving economy. and Miller.D. . F. H. January.E. K. (1989) “AI directions. Cambridge. F. Philosophy.” unpublished manuscript. Demsetz.

(1988) “The behavior of computational ecologies.” Washington. Holland. Kornfeld. and Miller. Learning. P. I. Huberman (ed. Ann Arbor: University of Michigan Press. Huberman.O.A. July/August. (1985) “The challenge of open systems. Los Altos.O. J. Nisbet. B. George Mason University. D. and William Tulloh Hebb. PARC. and Lumer.A. C.O. Anderson. Cambridge. E.118 Don Lavoie. New York: Addison-Wesley. B. J. B.” Byte. September.” in B. R. (1989) “Dynamics of adaptive controls. J. . New York: John Wiley.. B. (1988) “The global economy as an adaptive process. May. Huberman (ed. J. P. Howard Baetjer. Michalski et al. M. K. A.A. (1973) Competition and entrepreneurship. (1986) “Escaping brittleness: The possibilities of general-purpose learning algorithms applied to parallel rule-based systems.A. (1987) “The geodesic network: 1987 report of competition in the telephone industry. and Huberman. Huber. (1989b) “The performance of cooperative processes. and bankruptcy: A simulated economy learns to predict. (1989b) “Collective behavior of predictive agents. and Thagard. D. (1985) “Network externalities.A. Kahneman.E. Huberman. Amsterdam: North-Holland.” manuscript. Hogg. R. (eds) Machine Learning II.” American Economic Review.” in R. and Discovery. Holyoak. DC: Government Printing Office.A.W. (1989a) “Dynamics of computational ecosystems: Implications for distributed artificial intelligence. K. C. (1982) Judgement Under Uncertainty: Heuristics and Biases. R. L.M. Kay. (1989a) “An ecology of machines: How chaos arises in computer networks. Huberman.. Kaehler.” unpublished manuscript. Slovic.” The Sciences.A. competition. Kephart. T.) The Ecology of Computation.A. D. T.S.” unpublished manuscript. A.” unpublished manuscript.. and Hogg. (1986) Induction: Processes of Inference. Lachmann.A..” American Economic Review.. M. B. Chicago: University of Chicago Press.) (1988) The Ecology of Computation. March. Horwitz.R. S. Amsterdam: North-Holland. C.. Kephart. Xerox. Kansas City: Sheed Andrews and McMeel.” in P.) The Ecology of Computation.A.J. K. J. R. (ed. Xerox PARC. and Tversky. Kirzner. and Shapiro. T. B. Man. Holland. S. Nash.L. (1983) “The origin of predictable behavior.” in Transactions on Systems.” Ph. and Miller. MA: MIT Press. Hillis. April. bribery.. T. (1989) “The private basis of monetary order: An evolutionary approach to money and the market process. Hewitt. MA: MIT Press. Katz. and Cybernetics. Hogg. Xerox PARC. September. (1978) Capital and Its Structure. Holland. Amsterdam: North-Holland.D..M. M. Huberman. J. CA: Morgan-Kaufmann. (1981) “The scientific community metaphor. and P. dissertation.” unpublished manuscript. and Huberman. Pines (eds) The Economy as an Evolving Complex System.” Scientific American. Holland. Cambridge.” American Economic Review. Huberman. Kahn. (1984) “Computer software. and Hewett.. (1949) The Organization of Behavior. Arrow. (1985) The Connection Machine. Xerox PARC. compatibility. Heiner. New York: Cambridge University Press. P. W. (1988) “Betting. (1988) “Language design and open systems” in B.S.A.. (1975) Adaptation in Natural and Artificial Systems. (1985) “Experimental economics: Comment. B. Heiner.

M. and Jacobs.) The Ecology of Computation. DC: Citizens for a Sound Economy Foundation. Bobrow.” in B. Capitalism and Economic Growth: Essays Presented to Maurice Dobb. O. D. J.” unpublished manuscript. and the future of telecom- munications. Liggio. D. MA: Ballinger Publishing Co. Miller. Nelson. New York: Simon and Schuster. (1988) “Enterprise: A market- like task scheduler for distributed computing environments. McCloskey. Malone. and Benjamin.E. D. and Howard.G. D. R. and Winter. Huberman (ed. incentives. K. J. T. and Drexler.G. R. Nelson. San Mateo.P.S. T.S. M. (1989) “Money as a medium of exchange in an economy with artificially intelligent agents.” unpublished manuscript..D. CA: Morgan Kaufmann.A. Huberman (ed. L. Sante Fe Institute. Minsky.R. 1 and 2. Tribble. Amsterdam: North-Holland. Huberman (ed. Amsterdam: North-Holland.H. January. (1988) “Modelling coordination in organizations and markets. S.” Cato Journal 8(3). MA: MIT Press. P..R.” in C. New York: Cambridge University Press. C. vols. Winter. R. M. (1983) “The role of heuristics in learning by discovery: Three case studies” in R.G. (1988b) “Markets and computation: Agoric open systems. Cambridge. MA: MIT Press. (1988) “Freedom and the law: A comment on Professor Aranson’s article. Feinstein (ed. and discovery: The cognitive function of markets in market socialism. T. (1986) Economics as a Process: Essays in the New Institutional Economics. (1967) “The computer and the market. T. CA: Tioga Publishing. Michalski et al. et al. Lavoie. (eds) Machine Learning: An Artificial Intelligence Approach I.T.) Concurrent Prolog: Collected Papers. T. and Sargent. Mink. McClelland. Glassner (eds) Readings in Distributed Artificial Intelligence. Lavoie. E. (1985) The Rhetoric of Economics. McGrattan. D. Miller. D.E. Amsterdam: North-Holland. K.. (1987) Artificial Life: The Proceedings of an Interdisciplinary Workshop on the Synthesis and Simulation of Living Systems. self-published. (1989) “The courts.” in E. Palo Alto.H.) The Ecology of Computation.” in B.E. Cambridge: Cambridge University Press.A.) Socialism. Cambridge. (1986) Parallel Distributed Processing: Explorations in the Microstructure of Cognition. Malone. Lavoie. (1985) National Economic Planning: What is Left?.) The Ecology of Computation.W. May–June.” The Annals. Fikes...S. Marimon. J. CA: Addison-Wesley. .W.” Harvard Journal of Law and Public Policy. K. R. Malone.” Harvard Business Review.” Washington. D. Redwood City.A. Miller. (1989) “Economic chaos or spontaneous order? Implications for political economy of the new view of science. (1989) “The logic of electronic markets.L. and Drexler. Lenat. Bond and L. Cambridge.. Madison: University of Wisconsin Press. Amsterdam: North- Holland.” in B. D.S.J. Grant. (1982) An Evolutionary Theory of Economic Change. L.W. MA: Harvard University Press. T. and Palmer. (1990) “Computation. Lenat.B. the congress. Shapiro (ed.E.N.G.) The Ecology of Computation. (1988a) “Comparative ecology: A computational perspective. (1986) The Society of Mind. and Ellig. High-tech Hayekians 119 Lange. J.” in A. (1989) “Coevolution of automata in the repeated prisoner’s dilemma. Langton.S.B. Huberman (ed.. Miller. Langlois.. and Brown.. (1988) “Why AM and Eurisko appear to work. E. R.” in B. J. Sante Fe Institute. Cambridge. and Rumelhart. M. M. (1973) Computer Lib. (1987) “Logical secrets. Yates.

T. MA: MIT Press. Stefik. Elmsford. de Sola. T. and Stengers.. T. Cambridge. J. MA: MIT Press.” in G. (1989) “Natural and artificial intelligence. NY: Pergamon Press. B. Rust. Fall. August. Chicago: Chicago University Press. Polanyi.A. New York: Printers Publishers. V. (1985) Sciences of the Artificial. H. Simon. Englewood Cliffs. Schwartz. Palmer. and Flores. and Wegner.) The Artificial Intelligence Debate: False Starts. Cambridge. (1989) “The new connectionism: Developing relationships between neuro- science and artificial intelligence. G.. (1988) “Modelling economic dynamics and technical change: Mathematical approaches to self-organization and evolution. (1989) “Artificial intelligence and psychoanalysis: A new alliance. R. (1984) Order Out of Chaos: Man’s New Dialogue with Nature. (1983) Technologies of Freedom. I. Smith. Huberman (ed.R. F.” Byte. I. Jr. and endoge- nous expectations in experimental spot asset markets. A. Shriver. Graubard (ed. and Williams. P. Thomas. Rosenblatt.. and Miller.” unpublished manuscript. Durham. NJ: Prentice Hall. C. (1988) “Bubbles. and Edelman.L. (1958) “The perceptron. Turkle. Suchanek.O.” in S. Grauman (ed.” in S. (1989) “A double auction market for computerized traders. crashes. Cambridge. Cambridge. New York: Bantam Books. a probabilistic model for information storage and organization in the brain.) The Artificial Intelligence Debate: False Starts. (1989) “What’s in an object. Xerox PARC. (eds) Economic Theory and Technical Change. Huberman. (1988) “The next knowledge medium.R. (1986) Understanding Computers and Cognition. (1982) “Microeconomic systems as an experimental science. Silverberg. Pool. S. B.” Market Process. Simon. Hogg. D. S. Graubard (ed.N. (1965) The Organization of Inquiry. Sante Fe Institute. (1962) Personal Knowledge: Towards a Post-Critical Philosophy. New York: Addison-Wesley. F. Prigogine.A. Cambridge.) The Ecology of Computation. Sokolowski.M.L. G.” in B. Real Foundations. Real Foundations. Cambridge.. Tullock. and Stornetta. March.” in S. Amsterdam: North-Holland. G. MA: MIT Press. MA: MIT Press. (1983) Reason in Human Affairs. von Foerster (ed..L. G. Dosi et al. (1962) “Strategic approaches to the study of brain models. A.J. V.” in S. Reeke. Smith. Palmer.120 Don Lavoie. (1988) “Intellectual property.. M. G.” Econometrica. J.A. (1989) “Real brains and artificial intelligence.) Principles of Self Organization.) The Artificial Intelligence Debate: False Starts. (1972) Human Problem Solving. M. MA: Harvard University Press.. Cambridge. Real Foundations.” Journal of Economic Literature. F. (eds) (1988) Research Directions in Object Oriented Programming.” in H. Plott. . MA: MIT Press. and William Tulloh Newell.R.” Psychological Review 62: 386. Rosenblatt. Kephart.” American Economic Review.W. (1982) “Industrial organization theory and experimental economics.) The Artificial Intelligence Debate: False Starts. J.” unpublished manuscript. R. CA: Stanford University Press. NC: Duke University Press. J. MA: MIT Press. Waldspurger. Stanford. I.T. Real Foundations. and Simon. September. Winograd. H. C. Graubard (ed. H. (1989) “Spawn: A distributed computational economy. Howard Baetjer.R.

” because many of its recommendations for further study I soon after developed into my dissertation. my book argues that capital is best understood as embodied knowledge: the knowledge is the key aspect of any capital good. An Economic Perspective on Software Engineering. it does. “Process-oriented case studies of the computer industry. to those of us enchanted by the vision of Xanadu. a classic Hayekian spontaneous order. We say in the article. Never have I had so much fun – such excitement – in developing courses. Yet. the physical steel or copper or silicon in which that knowledge is embodied is secondary. Don Lavoie and I (with Don in the lead as always) investigated the capabilities of hypertext in scholarly learning pretty far. We would put our syllabus. The exposure to both the ideas and the people involved has strongly shaped my career. High-tech Hayekians 121 Comment after twelve years by Howard Baetjer On rereading “High-tech Hayekians. then and now. From the high-tech Hayekian viewpoint. the ability for a reader to see links to (rather than from) a document (with appropriate filtering. everyone uses it. this is wrong- headed: technology is in the capital goods (including “human capital”) we use. we were correct about the tremendous importance of hypertext to the evolution of knowledge and society. we were clueless that the Web was just around the corner. In both the book and more recent work I have been developing this idea into a critique of mainstream growth theory. space for student work. we did not imagine what we should have – that world-wide hypertext publishing would not be consciously planned and built.” twelve years after we published it. Software as Capital. course readings (including whole books . “Studying software as capital may help us illuminate the market process approach to capital theory”. Now. My book develops this idea at length. I must have written the first section. Few knew about hypertext then. independent variables. I cannot help smiling that when we wrote the article. We had high hopes that Xanadu would be built as envisioned by Mark Miller and others. frustrating tool. Our comments about how such a capability would enhance scholarly exchange (in the second paragraph of the section) are as important today as they were then. in our hypertext-based graduate courses.” thinking about software as a category of capital good very usefully puts our attention on the knowledge embodied in all capital goods. to separate technology and capital is to obscure one of the main processes that drives economic growth. While we were mistaken in the course we expected hypertext development to take. and from there into my book. that Xanadu promised. In particular. I feel very lucky to have been involved in the agorics project. One very important capability it lacks. is backlinking. Indeed. The second section was about the promise of hypertext for improving human discourse and the evolution of knowledge. software demos. but instead would evolve in a highly distributed way. because “software has virtually no physical being. The hypertext software we used was Folio VIEWS. which treats technology and capital as entirely separate. the Web seems a primitive. of course).

The last section proposed Austrian. students could create and see links to any part of any course document. immensely valuable demonstrations might be developed. As I write this note. While we correctly anticipated that object-oriented programming would greatly enhance software development by giving software the benefits of a kind of property rights.” These are operating systems for computation that would use internal pricing of computational resources such as processor time and space in RAM. With more time. . I look forward to further evolution of the Web. We each gave the effort a couple of terms. Kevin Lacobie and I. The third section deals with the idea that most fascinated the economists among us – Miller and Drexler’s anticipated “agoric open systems. Howard Baetjer. and William Tulloh – VIEWS is a marvelously efficient platform) – everything – into a Folio “infobase. valuing one good in terms of another. and learning from experience. . I believe the general enterprise of developing Austrian economic simulations is full of promise. and others we never imagined. . I had ten students literally around the world. programming expertise. Consider preference at the margin. I found the experience of working on the simulations extremely valuable (as well as great fun). and money certainly never emerged. In the most successful graduate course I ever taught. In this area. From within this infobase. process-oriented computer simulations of economic processes. and to comments by other students. reciprocating. I wish I had the time to investigate why not .” one per course that all the students shared. too. in hopes that the kinds of tools we developed (inspired by Xanadu). Nevertheless. or why not yet. It spoiled me. Neither our agents nor the environments in which they interacted ever developed very far. actually built a couple of programs that aimed at simulating Carl Menger’s story of the evolution of money from barter. and carry out effortlessly. so far the further step to networked markets for computational resources has not emerged. anticipating what A might want that B offers. will become generally available.122 Don Lavoie. I am frustrated that I cannot give my readers context by linking them directly to particular passages in the text of “High-tech Hayekians. That experience was real proof-of-concept. devoting most of our time to developing simple agents capable of barter. I have a strong hunch that it can complement the experimental economics being done by Vernon Smith and his team. yes? But try to embody them in code! I found myself awed by the fantastic complexity of both the individual decision-making and the interpersonal institutions that underlie human action we take for granted. I had the privilege of working under Don Lavoie’s guidance. Why? Because it rubbed my nose in the marvelous complexity of even the simplest aspects of everyday exchange. and the hypertext-based discussion was as rich and illuminating as any I have experienced. Despite missing our goal of seeing money emerge. with expert guidance from Mark Miller and Dean Tribble. Commonplace human action.” as I could have in my Lavoie-style hypertext course.

but by the spontaneous interplay of its component parts. could suffice. And we failed. In our arrogance. And we. we believed that we could be. Rather. painfully right. For this reason it may be that. they will succeed. the world was immensely enriched by the HTML browser and EBay. Though we have tried ever harder to give up the “great leaps. We would become “great leapers. as Hayek himself would have predicted. And though the HTML browser/Web server is a cartoonist’s caricature of Xanadu. High-tech Hayekians 123 Hayek’s road by Marc Stiegler Hayek was. The order would emerge not by the programmer’s design. Indeed. it can be argued that we still have not completed our difficult journey. he would also predict that. it is good enough that it transformed society. (Lavoie et al. to some extent this is how Hayek himself succeeded. These attempts to leap far ahead are perhaps a result of seeing a bright future clearly. and then to run the simulation in order to see what happens. the elegant complete solutions to the problems of global hypertext and electronic markets. although our leaps may improve in style and grace. Yet clarity of vision and a certain impatience are trademarks of the high-tech Hayekian. 1990: 135) . specifying institutional environments or decision rules for agents. and will continue to be. After all. outliers in the system: we would exploit the fact that occasional great leaps could be taken in a Hayekian world even though most progress is made in tiny steps. it would seem. would not know how the process was going to come out until we ran the mental experiments. However. where what is of interest is not what the end results are so much as how the process works. It has taken a full decade for some of us to align our efforts with the intellectual knowledge we had even then. Instead. the high-tech Hayekians knew that the people of the former Soviet Union had a long hard road to travel to reach a future in which they worked with the power of free markets. just once in a very rare while. it is to run the simulation as mental experiments. and of being too impatient to take a thousand little steps where just one. Fortunately. in our own way. is. at the end of a long hard road. they will never end. the programmers.” And so we built Xanadu and AMIX.” you can still see large leaps trying to break free from our projects at the slightest encouragement. The idea is not to create a mathematical model that already implies its conclusions in its premises. just as Hayek would predict that most of these leaps will fail. A comment on “high-tech Hayekians” in the perspective of agent-based simulation of economic realities: sound results urgent nos by Pietro Terna [T]he aim would be to set up constraining conditions. Twelve years ago. we had no idea just how difficult a road we ourselves would have to walk to align our efforts with those same free markets.

an unknown author in 2028 (when an asteroid crashed on Earth). look at . but also a third way: computer simulation. it is now absolutely clear (Gilbert and Terna 2000) that. as reported recently by Tesfatsion (2001). However. aside from a few interesting anecdotes. three different “symbol systems” are now available to social scientists: the familiar ones of verbal argumentation and mathematics. despite the quantity and the quality of the work done in the last 10 years? In the mean time a lot of work has been done. Next to the skills required to develop this kind of modeling technique and the need to find workable applications to the real world there is the important problem of the interdisciplinary methodology The key question now is: which tool is the most suitable for this “third way. unito.124 Don Lavoie. and the Electronic Journal of Evolutionary Modeling and Economic Dynamics (e-JEMED) (www. also for social scientists. involves representing a model as a computer program. Is the lack of this common language a cause of the slow (terribly slow) progress in obtaining results in our field.pdf). In either case.econ. we need a common language to aid economists to employ simulations as one of their tools of analysis: a sort of lingua franca that is widely understood and relatively easy to but the real problems are the results. and computer programs can be used to model either quantitative theories or qualitative ones. there is some phenomenon that we as researchers want to understand better. useful for the real world.” and what skills should researchers possess? Pace Luna and Stefansson (2000). The logic of developing models using computer simulation is not very different from the logic used for the more familiar statistical models. we have also two specialized electronic journals: the Journal of Artificial Societies and Social Simulation (JASSS) ( As a positive conclusive remark it is worth noticing that we are now working toward a unification goal.surrey. or computational modeling. where the complexity paradigm confirms Hayek’s theory of spontaneous order. “easy” is a term that covers a lot and in our case it does not eliminate the necessity of excellent programming skills. observes “that in a typical complexity book in the late 1990s . and William Tulloh I am this kind of programmer and I was beginning to work in this perspective in the early 1990s.” We urgently need to obtain sound results. A strong internal critique is that by Pryor (2000) in an auto-ironic paper: looking backwards. and agent-based simulations of enterprise behavior may help. Howard Baetjer. In this field it is possible to move from pure abstract theory to concrete problems of the organization of the firm and knowledge problems (for an outline of some work in progress. as Ostrom 1988 first observed. when the article was published: what are the most important changes that can we can observe after more than a decade? First of all. . almost all of the essays have no real empirical applications. as well as his laissez-faire economic philosophy (Kilpatrick 2001): a good goal for programmers working on agent-based simulations! . It is well-known that interdisciplinary and exploratory studies have less impact than monodisciplinary mainstream ones. This is another reason why finding concrete results become more and more necessary.soc. Computer simulation.e-jemed.

Ostrom. P. <http://www. Spring: 120–47. Cheltenham: Edward Elgar. 1. L. (2000) “How to build and use agent-based models in social science. F. H. Tesfatsion.. Pryor. (1990) “High-tech Hayekians: Some possible research topics in the economics of computation. Jr. spontaneous order and Friedrich Hayek: Are spontaneous order and complexity essentially the same thing?” Complexity 6(4): 16–20. and Stefansson. Colander (ed.pdf>. B. (2000) “Looking backwards: Complexity theory in 2028.” Market Process 8. Dordrecht and London: Kluwer Academic.L. Lavoie.” in D.iastate. (2001) “Complexity.econ. Luna.” Mind & Society 1: 57–72.. (eds) (2000) Economic Simulations in Swarm: Agent-Based Modelling and Object Oriented H. W. High-tech Hayekians 125 References Gilbert. and Terna. T. Kilpatrick. (2001) “Agent-based computational economics: Growing economies from the bottom up.” Journal of Experimental Social Psychology 24: 381–92.E. N.” ISU Economics Working Paper no. . pp. 63–9. F. and Tulloh. (1988) “Computer simulation: The third symbol system. D. Baetjer.) The Complexity Vision and the Teaching of Economics.

1963. without convention. 1892b). They cope rather well with the line of analysis selected by the more advanced version of monetary search models introduced by N. Bianchi 1994. but more saleable. A few years after the first edition of Jevons’ Theory of Political Economy. without legal compulsion. These works have developed without excessive difficulty the Mengerian principles of analysis. 1994. even without any regard to the common interest. then extended to different applica- tions of the monetary form of exchange. Garretsen 1994. consists in focusing on the self-reinforcing process of emergence of money from the set of commodities. O’Driscoll 1986). . under different circumstances. credit and remunerated assets. . motivates a correlated research on . The current stage of development of what could be considered as one of the more active testimonies of the Mengerian tradition in contemporary economic theory. . 1892a. Wright (Kiyotaki and Wright 1989. Hodgson 1992. but without explicit reference to Jevonian developments on barter and monetary exchange. Arena and Gloria 1997. Kirzner 1997). each by his own economic interests. 1991). he then provides a long-lived solution for the problem of the ‘double coincidence of wants’. according to its initial properties of marketability: Men have been led. . (Menger 1892a: 248) The microeconomic foundations Menger provides for the use of money are free from any assumption on the necessary disposal of some intertemporal reserve of value.5 The new economy as a co-ordinating device Some Mengerian foundations Elise Tosi and Dominique Torre Introduction Recent historical works have stressed the relevance of Hayekian suggestions to analyse the co-ordination patterns of market economies (Aimar 1999. standard views on market co-ordination have largely been influenced by the renewed interpretation of Mengerian theory of money. Over several decades. Kiyotaki and R. Simultaneously. applying them to a world with fiat money. nay. His nowadays well- known methodology (see Aimar 1996. to exchange goods destined for exchange for other goods equally destined for exchange. Menger has repeatedly exposed his genetic conception of the ‘origin of money’ (Menger 1985.

and more generally of interindividual arrangements involving law and positive institutions. When consumption only is considered. more in harmony with Menger’s point of view. variable. (Menger 1963: 158) The remainder of this chapter is devoted to investigate the possibility of founding the analysis of the new economy following Mengerian suggestions. The new economy as a new form of production A basic form of analysis of the new economy develops the incidence of the characteristics of immaterial goods from the consumer’s and the producer’s points of view. (Menger 1963: 155–6) The present-day system of money and markets. In the hierarchy of wants. in other words. The first unit has a positive cost but the following are free of cost. present day law. The fourth section. The economy of the Internet and other communication technologies appear at this point as forms of production of immaterial goods that deepen some . focusing on the production of these new goods. but costless to reproduce’ (Shapiro and Varian 1998: 9). indeed. conversely. The third section deals with information economics approach. ‘information is costly to produce. the modern state. These likewise serve the common welfare. applied to the analysis of the nature and the consequences of information goods. When Menger refers to these social links. immaterial goods are submitted to a principle of subordination in the Gorman–Lancaster consumption- set. More fundamentally. The final section presents the main conclusions. Our intuition is that this new form of economic activity cannot be reduced to a simple evolution in techniques and goods’ content. without being regularly the result of an intention of society directed toward advancing this welfare. even cause it. etc. Fixed costs are very important. The new economy as a co-ordinating device 127 Menger’s conceptions of other forms of co-ordination. develops the impact of the specific co-ordination device corresponding to network architecture in the new economy. Therefore.. the second section explains the essential arguments of the technological approach of the new economy. they emerge at an advanced stage of economic development. he explicitly refutes all holist or historicist interpretation of the role of the State. or. then marginal costs are zero. of ‘organic’ and ‘positive’ factors. Whenever the nature of the good is information. An essential part of the Problems of Sociology (Menger 1963) is devoted to present an organic conception of these other economic ‘insti- tutions’: The question of the origin of a number of social structures can be answered in a similar way. one could point out their specificity in terms of costs. offer just as many examples of institutions which are presented to us as the result of the combined effectiveness of individually and socially teleological powers.

thus economies of scale. differentiation has a relatively low cost and prevents the exit of the major part of competitors. Conversely. and helps companies remove layers of bureaucratic fat’ (Litan 2001: 16). The immaterial nature of the goods of the new economy displays consequences for activity location. the introduction of a new fixed cost-saving method is sufficient to dismiss certain competitors. In this case. cuts the costs of delivering many services and entertainment. the compatibility of the different systems becomes a key issue and leads to the redefinition of producers’ strategic behaviour. ‘the Internet makes it cheaper to design products remotely. the IBM-generation tech- nologies could only be considered as a sort of transitional pattern between the old ‘Fordist’ model and the new Microsoft/AOL standard. But when one of them currently operates under decreasing costs. and service customers. High fixed costs render competition actually violent. hardware and software.128 Elise Tosi and Dominique Torre previously existing properties of modern production technologies. Empirical evidence nevertheless confirms that the location of activities is maintained in the production sectors concentrating on the use of Internet connections. there would be no major discontinuity between the IBM model and the Microsoft/AOL one. suppliers have a strong incentive to exploit technological flexibility. Generally. One of the most immediate consequences of the immaterial nature of the Internet is that transportation costs do not matter in the setting up of a new firm. This is a source of increasing returns too. Moreover. communicate with. One usually distin- guishes two possible market structures. differentiated products may be supplied. . The technologies of the new economy also rely for a good part on technological complementarities: for instance. When nobody is able to produce a sufficient scale to enjoy a constant average cost. reduces the need for vast inventories. Then. dominated by the largest (but not necessarily the most efficient) firm. A monopolistic market may emerge. This form of industry would thus radicalize a more general tendency to develop newly adapted methods of production free from variable costs and mainly submitted to development– research and other fixed costs. In this line of reasoning. The relevance of cluster emergence in activities free from variable costs challenges the approach purely founded on the importance of increasing production returns (Quah 2000) and suggests that extra-technological reasons should be required to explain the geography of the Internet. provides a better means to target. The new economy would precisely be defined as the sector generating purely input-saving methods of production. in order to increase their market shares. The former takes advantage from the generalization of constant returns to scale while the latter radically improves the costs-minimizing strategies at work by generating new needs able to be satisfied by adequate immaterial goods. he can enjoy a monopoly situation. they produce different versions of the same good and by the way enlarge the demand basis. This specific structure of costs and returns has important consequences for the competitive process between firms in the new economy. From this point of view. hardware and processors.

there are two different reasons to refer to information economics when dealing with the new economy: information is the content of the produced goods and imperfect information governs the relationship between producers and their economic partners. in order to account for these possible forms of lack of information. they must spend long years of negative returns before reaching their breakpoint. the new economy is a world of ‘crooks and heroes’ (Volle 2000). This excessive impredictability of the start-ups future seems to be the consequence of two phenomena. then bearing the cost of information. Information is symmetric when the level of uncertainty is the same for all agents. The so-called new economy can indeed be considered as the economic sector producing information. While this kind of uncertainty is initially symmetric. Now. Then. Banks and stock markets do not offer relevant solutions whereas new financial intermediaries as venture-capitalists or business angels substitute for them more or less efficiently. At the first stage of its development. certain agents or certain classes of agents possess information of better quality than others. according to their cost structure. Risk and uncertainty in the activities of the new economy are at the origin of imperfect (symmetric) information. the relevant typology distinguishes between symmetric information and asymmetric information. These creditors have generally no real possibility of distinguishing between the risk level of the project and of evaluating the quality of their investment. Moreover. the relationships between start-ups of the new economy and their fund-providers are largely constrained by information asym- metry. The new economy as a co-ordinating device 129 Information economics and the new economy Information economics provides some apparently adequate reference for the analysis of the new economy in the most advanced textbooks (see Shapiro and Varian 1998) as in the research programmes applied to the Internet strategies. As other new firms involved in new sectors. it was observed that the amount of information available could be increased. At this stage. . Moral hazard and adverse selection are the well-known consequences of this kind of uncertainty. The new economy presents relevant examples of these two kinds of informational contexts. Information is relevant when economic performance exhibits some imperfections. information economics only considered information as given data of the economic context. Their own protection is first a higher interest rate and second an adequate level of risk diversification. The result is the difficulty of funding their activity during the first years of their existence. the difficulties of raising funds induce start-up entrepreneurs to hide some part of their information. Information has an individual utility as soon as all agents do not share the same knowledge or all the knowledge related to both the economic goods that are supplied and their economic partners. A large part of the start-ups have a very short life duration. At this point. a cost being incurred as a counterpart. The new firm must take high risks and seeks to make them supported by creditors. Internet’s actors have to face important uncertainty about future activity outcomes. In a sense. In the case of asymmetry.

Mobile telephony presents a very small number of operators. For these reasons. depends upon the . This enables to mass-produce basic components. For the consumer. Experience goods are. there is then no reason to suppose that this information is of poor quality. those goods such that their qualities cannot be disclosed by inspection but only after a precedent consumption. in order to increase the level of their reputation on particular sections. the informational content of experience goods takes a long time before becoming publicly available. Private experience of a good is then not really useful on the Net since public informa- tion generally reveals the level of quality of the available goods and services. infrastructure suppliers using reputation as a strategic weapon. ‘The modularization of the industry is one of the essential replies to the increasing demand for diversified and cus- tomised products. uncertainty finds its roots in the capitalistic form of production: uncertainty with respect to the quantity and the quality of product one has at one’s disposal through possession of the corresponding goods of higher order is greater in some branches of production than it its in others . During that year. The specialized literature frequently refers to the episode of Intel’s Pentium processor in 1994. Economic considerations on uncertainty and information are not absent from Mengerian developments. Reputation is not only the consequence of a good circulation of informa- tion between agents on the network. able to increase the identification of the brand by consumers. that are assembled along different ways to get a wide variety of products and services’ (Brousseau 2000: 3). hardware producers. In the case of search goods.130 Elise Tosi and Dominique Torre Information goods are usually allocated to two categories. The greater or less degree of uncertainty in predicting the quality and quantity of a product that men will have at their disposal due to their possession of the goods of higher order required to their production. When the market is imperfect. If the produced good has an informational content. suppliers would be immediately penalized by a bad reputation effect. Modularization is another feature of the Internet economy. . . When interindividual relationships are established in a network. one generally considers that reputation implies imperfect compe- tition. Sufficient level of specialization can be desired by competitors on the Internet. search goods only require low information costs measurable in terms of search time. The reputation phenomenon is enhanced. the quality of the good is directly observable and it is disclosed after simple inspection. Reputation frequently requires specific investments. Private experience – the processor’s failure experience – led consumers to react on the Internet and Intel was forced to improve its components. If that were the case. positive profits and a large sized firm. in contrast. the opportuni- ties to publicly reveal information are multiplied. The quality of the information provided by Internet sites is not enforceable before a time lag. Intel commercialized this processor that was supposed to be highly reliable. The information content of goods in the new economy has other well-identified consequences on competitive processes. According to Menger.

Menger already suggests a somewhat academic distinction between material and immaterial goods. (iv) command of the thing sufficient to direct it to the satisfaction of the need’ (Menger 1995: 35). the higher the uncertainty that arises when the output depends on stochastic components of the economic context. trade licenses. The fact [that these things] are actually goods is shown. Moreover. Of special interest are the goods that have been treated by some writers in our discipline as a special class of goods called ‘relationships’. goodwill. his own illustra- tions concerning immaterial goods have something different from the traditional examples. The most interesting line of analysis of the new economy therefore insists on the network form of co-ordination induced by production and circulation of information goods. No fundamental distinction related to the nature of the good would be relevant except for the four prerequisites which ‘must be simultaneously present: (i) a human need. . religious and scientific fellowship. by the fact that we often encounter them as objects of commerce. it reveals that for Menger. a hunter and even a pearl-fisher’ (Menger 1950: 51). etc . An analysis that was only founded on the information content of the goods of the new economy would not insist sufficiently on the specific nature of the link these goods create between agents. copyrights. (iii) human knowledge of this causal connection. The new economy does have specificities related to the form of the costs. the new goods introduced in many sectors of the new economy present a level of intrinsic uncertainty as regards their marketability. In this category are firms. . (Menger 1995: 38–9) . . the levels of uncertainty and information are linked with the form of the production process. patents. in a way comparable to Menger’s examples of ‘a grower of hops. even without appeal for further proof. and upon the greater or less of control they can exercise over these elements . Nevertheless. The longer the roundabout in production. family connections. These specificities enhance uncertainty in the same way the time of production does in Menger’s examples. friendship. This uncertainty is one of the most important factors in the economic uncertainty of man. . (ii) such properties as render the thing capable of being brought into a causal connection with the satisfaction of this need. and also. authors’ rights. The new economy as a co-ordinating device 131 greater or less degree of completeness of their knowledge of the elements of the causal process of production. The new economy as a new form of co-ordination Prior to developments relative to the role of information goods. love. according to some writers. (Menger 1950: 51–2) This form of production has no apparent similarities with the production of information goods. Nevertheless.

132 Elise Tosi and Dominique Torre One of Menger’s goals is to depart from a view whose main disadvantage would be its inadequacy to his gestating organicist conceptions. They have. Therefore. the growing division of labour causes the development of a special professional class which operates as an intermediary in exchanges and performs for the other members (shipping. we do not consider it as a good. Information which is the major content of the services supplied by the interconnections is able to render more easily other economic activities as Mengerian intermediaries would have done. ). Menger introduces a new topic really in harmony with the paragraphs consecrated to ‘relationships’. moreover. as is true of customer good-will. currently at the disposal of the various peoples and nations whose trade they mediate. he does insist on the ability of the notion of good to integrate such ‘relationships’: If. . or even entire countries or continents) according to the position the intermediaries in question occupy in commercial life. we observe that a specific class of people has a special professional interest in compiling data about the quantity of goods. etc. the network belongs to the category of ‘relationships’. These lines are devoted to the importance of middlemen. . . but we are interested in the Mengerian . but we will have occasion to discuss this at a later point. The data they compile cover trading regions that a smaller or larger (single counties. distribution. transportation costs and other trade imperfections. these useful actions or inactions of such a kind we can dispose of them. without finding it necessary to resort to the obscure concept of ‘relationships’. etc. there is no reason why we should not classify them as goods. Thus. Now. firms. (Menger 1995: 39–40) However. but stressing the informational role of these intermediaries: Operators of this ‘sector’ could thus correspond to some members of this ‘special professional class which operates as an intermediary in exchanges and performs for the other members’. (Menger 1995: 69) The place these intermediaries occupy in the production and trade processes are not very far from the mission of our Internet. but also the task of keeping records of the available quantities. the storing of the goods. . . . . these ‘actions’ are of the range of things that we now consider as means of co-ordination between individual decisions. and without bringing these ‘relationships’ into contrast with other goods as special category. As intermediaries. so-called stocks in the widest sense of the word. without considering transaction costs. (Menger 1995: 69) As soon as a society reaches a certain level of civilisation. after the developments concerning the need of information in the roundabout production process. A few pages later. monopoly rights. an interest in many other general kinds of information. provinces.

and also to avoid inefficient demand lock-ins. the competitors may be engaged in a rather complex strategy. The new economy as a co-ordinating device 133 distinction between direct and indirect disposal of some good that displays important consequences in terms of information: A person with consumption goods directly at his disposal is certain of their quality. The value of any item is enhanced by the increased availability of complementary goods. The more members of a network. . from the benefits for each individual to be able to communicate with others. the larger the number of subscribers. . At the first stage. But a person who has only indirect command of them. the more users on the network. stemming from the existence of complementarities and intrinsically linked with network organization. Considering the challenge represented by the size of the firm. cannot determine with the same certainty the quantity and quality of the goods of first order that will be at his disposal at the end of the production process. the weakness of increasing returns would be unable to prevent the network imploding. . indeed. supply positively acts on demand through feedback effects. Therefore. This ‘specific class of people [that] has a special professional interest in compiling data about the quantity of goods’ is now a network composed of complementary nodes and links. In the same order of ideas. co-operation rather than competition is the means to ensure the network’s performance and reliability. a monopolist exclusive holder of a technology may have an incentive to invite competitors and even subsidise them’ (Economides 1995: 22). (Cave and Mason 2001: 3) A direct consequence of these externalities is the so-called Metcalfe’s law: the value of a network increases like the square of the number of its participants. a minimum level of customer base is required to ensure the profitability of the network. This network generates externalities as the higher the value of a subscription. the Internet has to improve the information of the agents whose position could be assimilated to the ‘persons who have only indirect command’ on some kind of good. . The value of the connectivity arises . through possession of the corresponding goods of higher order. the network growth is fed by positive feedback effects. (Menger 1995: 51) As Mengerian specialists. In other words. the more likely it is that new services will be offered over the network. When the initial supply of all competitors is insufficient. the feedback effects result from the existence of scale economies on the demand side: The driving force behind the Internet is Network externalities – the fact that the value of a set of computers increases with the number of computers that are interconnected. ‘In the presence of strong network externalities. There are also indirect benefits associated with a large network. it is the relevant solution in order to provide the necessary compatibility between standards. . the more benefit .

both backbones face a demand reduction.134 Elise Tosi and Dominique Torre Then. gains a competitive advantage. ‘Innovation requires openness. the quality of interconnection is governed by the preferences of the backbone which values interconnection the least. performance). and when they are of equal size. (Cremer et al. When the network reaches a sufficient size. (Cremer et al. the degree of control of the existing standard must increase in order to maintain the leadership position. The larger backbone. In order to stimulate innovation. In the case of the Internet. 1999: 2) Recent literature extends the analysis of such a competitive process. as a counterpart. Second. 1999: 2) When the backbones are not too few. some trade-offs emerge between flexibility/system openness and control. the existing institutional framework has to be flexible and incomplete to be able to welcome new practices’ (Brousseau 2000: 4). Multihoming strategy from the customers’ part becomes the essential response of demand to the rational choices of backbones in terms of standard. This quality of interconnection is a strategic variable. firms begin competing in order to control the whole system. the co-operative stage induces competitors to adopt adaptive technologies which increase their compatibility potential. and its position deteriorates according to the installed base of the other backbones. which relies relatively less on access to the other backbone’s customers. a similar quality differentiation effect handicaps both backbones relative to the other ones. . and according to the Katz and Shapiro model of sponsorship. The resulting conclusions put into balance the advantages for the competitors gaining a larger share of the network and the penalties induced by the simultaneous withdrawal of some customers as a consequence of a lesser compatibility of existing standards. innovation is responsible for standards differentiation. it is motivated by ‘network shares’ seeking activity: When connectivity between the two networks is degraded. each backbone has an installed base and otherwise competes for unattached customers. Thus. when other backbones are present. However. as their customers’ access to each others deteriorates. A backbone that would degrade the quality of its interconnection with an equal-sized backbone would not gain a sufficient advantage over this backbone. and because ‘it takes two to tango’. In this setting. a degradation of connectivity creates a quality differentiation between the two networks. less flexibility is needed but. none of them has an incentive to degrade interconnection. when the network is installed. During the different stages of a network’s life (adoption. and competition between the two backbones is softened. The benefit derived by a customer from joining a backbone is an increasing function of the size of his or her backbones and of the quality of interconnection with the other backbones.

Consumption goods generally have characteristics that are independent from the way they are allocated. some adequate price-war behaviours develop among competitors. but they lose from the lack of variety and from the remaining imperfections in technological compatibility. Although marginal cost may be zero. but use them in order to get the goods they have decided to purchase or to access the services they need. Their value is essentially determined by the form and the size of the network of which they constitute a fundamental component. But the Internet and more generally information and knowledge-based networks reveal a fundamentally different content. Consumers do not consume search engines. financing. between two temporary equilibria. it becomes very difficult to implement a differentiation strategy. Standard microeconomic analysis elaborates a widely accepted distinction between activities and forms of reallocation or co-ordinating processes. in fact. they lose all relevance. fixed costs have to be incurred: advertising substitutes to final consumers’ payments and oligopolists may be incited to compete in their call for advertising. network analysis could only offer some incremental developments of imperfect markets chapters in industrial organization handbooks. Information goods have just the opposite property. . Information and knowledge-based networks are another form of economic connection between decentralized agents. Consumption. As Mengerian ‘relationships’. sites. production. The new economy as a co-ordinating device 135 Prices are not the only way to compete in the new economy. The representation of the market is founded on the interactions of the two opposite forces of supply and demand which contribute through different equilibrium categories to define the level of co-ordination. as a singular co-ordination device. supply and demand cannot generally be isolated. When the network stands in a transitory position. are elementary activities which contribute both to the individual satisfaction of agents and to national wealth and growth. while the incurred costs would be paid back by the exploitation of more profitable segments related to the same standard. this kind of network can be considered as a specific configuration of the market. price differentiation is bounded by a tendency to supply free products. Market. contractual agreements are the more representative forms of co-ordination. Special low prices may be decided to capture extra demand share. etc. This competition shows contrasting effects on consumers’ well-being: consumers gain from network externalities. and even more. Within information and knowledge-based networks. price war remains the single opportunity. Then. Indeed. But this form of co-ordination seems to break the usual distinction between activities and co-ordinating processes. at this stage of the reflection. State. etc. The emergence of a standard puts limits on differentiation and increases price competition: when compatibility problems are solved and when network externalities are enjoyed. Relationships of this kind can be considered as specific intermediaries that form a sort of bijection with the network as middlemen do with the market. Flat prices may be the ending of sequential prices contraction. Whatever the originality of the competitive behaviours within the new economy. access providers. Relationships are the essential component of these interindividual connections.

‘The combination of . (Cave and Mason 2001: 26) The other pricing patterns are economic responses to deal with the inefficiency of flat-pricing. In a world characterized by the wide diffusion of Internet services. access is unrestricted for customers. in the transaction-based model.136 Elise Tosi and Dominique Torre If real markets were of the Walrasian kind. An intermediary may subsidy the participation of some agents in order to increase its attractiveness for other participants’ (Caillaud and Julien 2000: 2). This solution is hard to implement but copes well with the tendency of considering the Internet as a free access co-ordinating . . ‘The main role of intermediaries is to gather information on users . Mackie-Mason and Varian (1997) suggest solving the problem of pricing Internet services by introducing a ‘smart market’ principle. . . . 2000: 411). Finally. asymmetric network externality and third degree price discrimination opens the possibility of cross-subsidisation among the others . middlemen exist because of market imperfections. In the case of the Internet. Usage- sensitive pricing makes the price depend on the degree of congestion on the demand side. the total decentralization of subscribers requires the intervention of a large class of operators which guarantees the reliability and the efficiency of the network. . missing information or poor quality information (Rubinstein and Wolinsky 1987). Conversely. leaving little room for [other] match-making middlemen. the marginal cost of the connection is zero and there is no way to avoid congestion. The network’s configuration has important effects on pricing in the new economy. These middlemen services are being replaced by capital-intensive services that rely on large databases and sophisticated search engines’ (Wimmer et al. ‘the introduction of advanced information technologies greatly reduces the cost of search. there would be no room for middlemen. The activity of middle- men involves specific competitive relationships between users and intermediaries. But Internet intermediaries exist as constitutive elements of the Web architecture. ignoring the social consequences of their actions. With flat-rate pricing. Flat-rate pricing corresponds to the payment of a fee to connect the network. Indeed. to find each other’ (Caillaud and Julien 2000: 1). . in the US as in Europe: The underlying economic problem is an old one. This solu- tion involves a zero usage price when there is no congestion. pricing is a function of the char- acteristics of the transaction. This pricing method is rather well established in a large part of the Internet. like transaction costs. Only with the congested part of the network is a priority order introduced according to bids attached to the messages by users. . free access is a sort of norm but takes different modalities. . Users of any common and freely-accessed resource have a tendency to over-exploit: each will use the resource until the private (marginal) cost of doing so equals the private (marginal) benefit. without considering the connection time. Recent literature evokes three categories of prices that McKnight and Bailey (1995) have dealt with. known as the tragedy of the commons . so as to help different classes of agents. in particular buyers and sellers of one specific good.

the network generates increasing returns when it can be used for the systematic diffusion of large-scale messages like professional correspondence. When the firm already runs an old technology and faces adjustment costs. the fee paid by advertisers is directly connected to the second component of the site. the former determining only the demand size. its natural decision is to postpone the adoption of the new technology. apparently easily reaches Pareto optimality. it also generates special motives of co-ordination failures. 1997 and Mason 2000). Some of them stem from the free disposal. The new economy has special kinds of intermediaries. Every site could be considered as being composed of two elements: the information needed by the consumer and the advertising message which covers the cost. The new economy as a co-ordinating device 137 scheme. thus. both because of flat marginal cost and because of the neglect of utility resulting from the former component of the site. Conversely. able to test the properties of these different principles. . In such a situation. broadcasting services’ (Cremer 2000). When we get close to the universal access of the network. phone bills. during the process of access diffusion. Cremer focuses on a special consequence of network externalities. electricity invoices. the adoption returns are initially decreasing when the newly connected people attach less and less value to the network. Thus. such as the ‘Paris Metro pricing’ proposal of Odlyzko (1997) which suggests differentiating between two types of access: a zero price for congested access and a non-zero price for less congested access. Then. Recent development of mathematical models. it has some tendency to pre-empt the adoption of the new technology in order to avoid some subsequent adaptation costs. In a recent work. individual optimality of the advertisers’ choice does not correspond to social efficiency. Other pricing formulas have been introduced in the literature. or rather there exist new opportunities to deliver . ‘the current Internet pricing structures are unlikely to be optimal. While the network structure is compatible with a high degree of connectivity between agents and. the decision to adopt a new technology only depends on the relative weights of fixed and variable advantages generated by the adoption. the structure of costs and even more the effect of externalities induces underexploitation of the network’s capacities. excepted in the more congested (attractive?) part of the network. other kinds of co-ordination failures are likely to emerge. The large heterogeneity in uses of the Internet means that equal treatment of all sources (the current Internet practice) is inefficient’ (Cave and Mason 2001: 26). For a technology consumer. While the total utility of each consultation is the sum of the utility resulting respectively from the consumption of the informational content and of the advertising message. When the technology has already been adopted. have not yet provided a definitive solution to the problem of selecting among these pricing rules the optimal one (compare Gupta et al. . when the firm enters the network. while sunk costs are to be borne by advertisers. None of these decisions are socially efficient. Pre-emption and replacement are the more representative effects occurring during the adoption stage. More generally. ‘there are new services which can be offered. without any other differentiation of the supplied services. .

Conclusion This chapter investigates the relevance of a Mengerian methodology to analyse the microeconomic foundations of the new economy. Finally. This is essential to enable innovators. On the other hand. The process of price determination in the new economy is quite different from the one at work in the ‘old economy’. recalled in the second section. This view. The fourth section tries to base an alternative approach of the new economy on Mengerian developments related to relationships. Operators of this ‘sector’ could thus correspond to some members of this ‘special professional class which operates as an intermediary in exchanges and performs for the other members’ (Menger 1950: 69). this currently accepted microeconomic approach presents the new economy as the expression of more general concepts as experience goods or search goods. the kind of intermediation generated by the new economy is undoubtedly the more relevant one. to efficiently exchange an recombine those intangibles’ (Brousseau 2000: 4–5). that create intangibles. special forms of intermediation that could be as much relevant when markets co-ordinate economic actions as when networks provide the essential economic link between agents. Another way of rationalization of new economy is provided by infor- mation economics. underlining the relevant consequences of uncertainty and asymmetry on the properties of these goods. . and the enforceability of contracts. This new configuration of economic activity can be considered as an extreme consequence of technological properties of new methods of production. On the one hand. according to Menger. such policies must ‘guarantee the consistency of the property rights system.138 Elise Tosi and Dominique Torre Increasing returns are at the origin of multiple equilibria and subsequent inefficiencies. As money and middlemen are the way to co-ordinate market transactions. In adequacy with its previous developments. or contracts Menger would have considered as a way to improve the co-ordination of economic activities. special kinds of goods. networks representative of the new economy as markets of the ‘old economy’ are imperfect co-ordination devices. they have to guarantee the Pareto efficiency of the network. Both systems are compatible with the existence of other ‘organically created insti- tutions’ (Menger 1963: 57) like State. These co-ordination problems call for active policies. cannot distinguish new economy from other forms of increasing returns production processes. However. If there are new relationships in our modern economies. the Internet structure is the appropriate form of co-ordination of network relations. it may be suggested that the so-called new economy is an adequate solution to the information problem arising both from the increasing complexity of the roundabout production processes and from the enlargement of qualities of consumption goods.

(1994) ‘The relevance of Hayek for mainstream economics’. van Zijp (eds) Hayek. (1994). J. and Wright. 94–108. Garretsen. ‘E-economy. They are particularly grateful to B. and Tirole. Aréna. R. S. Cremer. 3–5 September.B. A.O. P. Lebanon. History of Economic Ideas VII(1–2): 137–65. GREMAQ. (2000) ‘Competing cybermediaries’. Stahl. (1997) ‘Entrepreneurial discovery and the competitive market process: An Austrian approach’. (1996) ‘The economics of networks’. . (1991) ‘A contribution to the pure theory of money’. (1995) ‘An introduction to Internet economics’. A. B. N. http://www. Bailey (eds) Internet Economics. in Caillaud. 60–85. McKnight and J. Hodgson. Foreign Policy 16–24. Gupta. R. Bristol. M. Litan. and Whinston. Journal of Economic Dynamics and Control 21(4–5): 697–722. International Journal of Industrial Organization 14(2). Kiyotaki. History of Economic Thought Conference. and Mason. September. European Economic Review 44: 1045–56.. and Julien. Aimar. Mason. (2001) ‘The Internet economy’.W. MA: MIT Press. R. N. Cambridge. D. J. (2001) ‘The economics and regulation of the Internet’. Economides. May. ‘Time. European Economic Review 44. mimeo. P. Schmidt (ed. McKnight.W. (1996).umich. in J.P. March. E. (1997) ‘A stochastic equilibrium model of Internet pricing’. ‘Hayek’s spontaneous order: The correct versus the corrigible society’. The usual disclaimer applies. B.P. 15–16 May. pp. paper prepared for the OXREP issue on economics and the Internet. (1997) ‘Economic FAQs about Internet’. J. (2000) ‘Network externalities and Universal Service Obligation in the Internet’. Co-ordination and Evolution. Journal of Political Economy 97(4): 927–54. MIT Workshop on Internet Economics. References Aimar. The United Nations Economic and Social Commission for Western Asia (ESCWA). G. London: Routledge. (1999) ‘Connectivity of the commercial Internet’. Birner. Journal of Economic Literature 1. and Bailey. Review of Political Economy 4(4). Bianchi. Caldwell for his comments on a first version of this chapter. and Gloria. pp.M. March. in C. mimeo. in J. Rey. I. A. N. and Wright. Beirut.. (1992) ‘Carl Menger’s theory of the evolution of money: Some problems’. Co-ordination and Evolution. R. Aldershot: Edward Elgar.) Uncertainty in Economic Thought. co-ordination and ignorance: A comparison between Hayek and Lachmann’. J. R.W. (2000) ‘Simple competitive Internet pricing’. Cremer. Birner and R. and Varian. Kiyotaki. Centi and C. van Zijp (eds) Hayek. (2000). Kirzner. new growth regime and public policies’.K. (1989) ‘On money as a medium of exchange’. London: Routledge. Cave. M. R. T. T. R. ‘Money and uncertainty in the economic thought of Carl Menger’. (1997) ‘Menger and Walras on money: A comparative view’. The new economy as a co-ordinating device 139 Acknowledgments The authors would like to thank J. L. 232–51. Schmidt for their helpful critiques and suggestions. (1999). Brousseau. J. Birner and R. Journal of Economic Theory 53: 215–35.

A. Good). (2000) E-conomie. H. A. B. 1883). Rubinstein. Streissler and M. M.P. B. Menger. Streissler and D.E.F. D. Nock of Untersuchungen über die Methode der Socialwissenchaften und der Politischen Oekonomie insbesondere. A. C. C. Wimmer. (1997) ‘A modest proposal for preventing Internet congestion’. Hoselitz of Grundsätze der Volkwirsthschafslehre. Menger.140 Elise Tosi and Dominique Torre Menger. Streissler. Quah. (1892b) ‘La monnaie.R. (1987) ‘Middlemen’. I) (English translation by J. (1892a) ‘On the origin of money’. A. Aldershot: Edward Elgar (1876. Boston: Harvard Business School Press. Odlysko. (1994) Lectures to Crown Prince Rudolf of Austria. European Economic Review 44: 1032–44. The Quarterly Journal of Economics. New York and London: New York University Press. mesure de la valeur’. IL: Free Press (reproduced in I. G. C. (1950) Principles of Economics. Revue d’Economie Politique IV: 159–75. and Varian. London: William Pickering. AT&T Labs.) Classics in Austrian Economics. 1871). German translation by M. C. Shapiro. Paris: Economica. Urbana: University of Illinois Press (English translation by J. C. ed. and Chezum. (1986) ‘Money: Menger’s evolutionary theory’. Vol. Menger.W. research mimeo. (1963) Problems of Economics and Sociology. . C. History of Political Economy 18(4): 601–16. Glenoe. Journal of Labor Research XXI(3) Summer: 407–18. Vienna: Wilhelm Braunmüller. (1998) Information Rules: A Strategic Guide to the Network Economy. Volle. 1995. by E. Economic Journal 2(6): 239–55. Kirzner (ed. and Wolinsky. Menger. (2000) ‘Internet cluster emergence’. O’Driscoll. Dingwall and B. C. August.. (2000) ‘Information technologies and the middleman: The changing role of information intermediaries in an information-rich economy’. Townsend. (1985) Investigation into the Method of Social Sciences. Menger. with Special References to Economics.S.F.

Part III The organization of the firm .


This assumption is similar to the market socialist assumption of “glass walls” mysteriously arising under socialism. 2000). Mendelsson and Pillai 1999. exchange and build information rapidly in response to changing contingencies. 2000. an assumption that was harshly criticized by Hayek in the context of the socialist calculation debate. Ellig 1993. Ichniowski et al. since the crucial revelation principle implies that even with asymmetric information. Austrians and non-Austrians alike. OECD 1999. that firms need to harness the ability of markets to utilize. centralization will (weakly) dominate decentralization. the overall conclusions that emerge from a number of these studies are that that hierarchy and planning methods are as problematic inside firms as they have proved to be outside firms. Osterman 2000). Foss 1999. may these drivers produce changes in internal organization towards more decentralized forms? From the perspective of. a number of economists.6 “Austrian” determinants of economic organization in the knowledge economy Nicolai J. Milgrom and Roberts 1990). Hayekian) ideas on the need for decentralization fostered by the presence of dispersed knowledge (Cowen and Parker 1997. However. 1998. It is well-known that this result depends on a highly unrealistic zero communication cost assumption. 1996. for example.g. Why. Ellig and Gable 1993. it is not completely evident how to understand the nature of the relevant causal forces. for example. Factors such as increased differentiation of tastes on the demand side (e. Briefly. acceleration of innovation and technological development on the supply side (D’Aveni 1994). and changes in the composition of labor on the input side (Tomlinson 1999) are argued to be important drivers behind the current dynamics of economic organization. Hodgson 1998. standard incentive theory. who have addressed economic organization in the knowledge economy have drawn upon Austrian (more precisely. not only with respect to the firm’s vertical and horizontal boundaries (Helper et al. Jensen and Wruck 1994). this is something of a puzzle. It is easy to see how . Ghoshal et al. 1995. 1997. and that extensive delegation of decision rights and the use of high-powered incentives to support this are imperative. Foss Introduction There is evidence that firm organization is currently undergoing profound change as a result of the emergence of the knowledge economy. In fact. but also with respect to internal organization (Gittleman et al. Jensen and Meckling 1992.

and that commercially useful knowledge becomes increasingly distributed and needs to be accessed from several sources. In particular. In his critique of market socialism. Misesian arguments are used to criticize arguments derived from Hayekian insights that firms should emulate markets to the largest possible extent. but also in central Austrian contri- butions such as Mises (1936. economists should . I focus on internal organization issues. Jensen and Wruck 1993). an increasing number of which lie outside the boundaries of firms (Liebeskind et al. 1944. In other words. present not only in Coase (1937) (and most of post-Coasian organizational economics). Mises 1936. 1995. Although they have so far had little to say here. 1949) and Hayek (1973). Mises argued that the economic institutions of capitalism are strongly complementary so that (unhampered) capitalism is a stable system. Ellig and Gable 1993. Jensen and Meckling 1992. Mises sides. Mises (1949) pointed to the folly of “playing markets. as well as on his insights in property rights and ownership (Foss 1994a. This chapter links up with recent applications of Austrian ideas to issues of organizational design. as it were. The entrepreneur con- trols the factors of production” (Mises 1949: 306). I argue that – contrary to some of these contributors – it does not follow that firms should emulate markets as far as possible. I use this fundamental notion to argue that firms are also systems of complementary elements and that this fact places constraints on the extent to which firms may be made “market-like. Thus. Matusik and Hill 1998). I also draw on Mises’ (1949) related insight that the mixed economy is inherently unstable. In modern parlance. Foss this may imply a skeptical attitude to the analytical dichotomization between planned firms and unplanned markets. I argue with Mises that “[t]he function of the entrepreneur cannot be separated from the direction of the employment of factors of production for the accomplishment of definite tasks. Ghoshal et al.” In particular. with the Coasian notion that markets and hierarchies are indeed different mechanisms for resource allocation.” and I draw on his overall argument that bringing coordination mechanisms characteristic of market organization into a planned organization is inherently problematic. However. such that changes away from pure capitalism will result in serious allocative inefficiencies.144 Nicolai J. I discuss the implications for internal organization of the Hayekian notion that the dispersed and subjective character of knowledge is a strongly binding constraint on the use of planned coordination (Cowen and Parker 1997. Like a number of recent contributors. The “direction” and “control” undertaken by “the entrepreneur” is qualitatively different from allocation by means of the price mechanism.1 The knowledge economy: a challenge to economic organization The concept of the “knowledge economy” is used in a loose sense to refer to the overall tendencies that many industries become increasingly “knowledge- intensive.” that an increasing share of the workforce is constituted by “knowledge workers” (Tomlinson 1999). since it relies on authority that is backed up by the entrepreneur’s ownership of the alienable (non-human) means of production. consisting of interlocking elements. 1949). 1995.

Such networks typically cut across the legal boundaries of the firm. are particularly useful organizational arrangements for sourcing and transferring knowledge because the costs of pricing knowledge (in a market) or transferring it . some distinct themes are discernible. a consensus seems to be emerging that tasks and activities in the knowledge economy need to be coordinated in a manner that is very different from the management of traditional manufacturing activities. The clusters are largely autonomous and engage in decentralized decision-making and planning . . uncertainty. and that traditional answers are no longer valid for the knowledge economy (see e. so the argument goes. Prusak 1997). . because of the growing importance in knowledge-intensive industries of being able to access knowledge from multiple sources. For example. with profound transforming implications for the authority relation and the internal organization and boundaries of firms. Networks. knowledge- based networks (Harryson 2000) increasingly become the relevant dimension for understanding the organization of economic activities. and pace of change of the market-place . Overall. They facilitate organizational flexibility. this is because recent discussion goes right to the heart of the crucial and perennial issues in the theory of economic organization. Specifically. contributions to Myers 1996. In cluster of network working. Organizations are adopting new forms of decentralization to cope with the instability. “Austrian” determinants of economic organization 145 face the challenge of examining economic organization in the context of the emerging knowledge economy. . Fundamentally. What are the limits to resource allocation by means of authority? What do we mean by authority? What defines the boundaries of firms? How do we distinguish an independent contractor from an employee? The need to revisit such fundamental questions is prompted by the numerous writers who claim that the knowledge economy will fundamentally change the answers we provide. . long-batch production runs. employees of undifferentiated rank may operate temporarily on a certain task or tasks in teams. low variety. They are conducive to individual initiative (“intrapreneurship”) and faster decision-taking. Recent claims about economic organization in a knowledge economy: review Although a number of fields and sub-fields are involved in the ongoing discussion of efficient organization in the context of the emerging knowledge economy. relevant to Taylorist methods. The following section presents a brief review of these arguments. Cowen and Parker (1997) explain that Market changes are moving manufacturing farther and farther away from steady-state. to high variety and small runs . challenging us to rethink issues such as.g. . . (Cowen and Parker 1997) A number of writers also claim that not only the internal organization of firms but also their boundaries will be profoundly affected by the emerging knowledge economy.

. The combined effect of the increased importance of knowledge assets that are controlled by knowledge workers themselves and of the increasingly specialist nature of knowledge work is to wreck the traditional economist’s criterion of what distinguishes market transactions from hierarchical transactions (Zingales 2000). (Cowen and Parker 1997: 15) . Powell 1990: 304) often exceed the costs of transmitting knowledge within an informal network. Finally. Helper et al. but when an external authority market can provide information that leads to greater effectiveness. and not on the authority of office. then authority tends to migrate into the market. This . expert authority depends on the information resources available to an individual. Thus. with respect to theories rather than phenomena. because authority increasingly shifts to expert individuals who control crucial information resources and may not be employees of the firm. the increased reliance on knowledge networks tends to erode authority-based definitions of the boundaries of the firm. and partly a result of the increasingly specialist nature of knowledge work (Hodgson 1998).146 Nicolai J. Simon 1951. 1995: 7. . a number of writers are quite explicit that the advent of the knowledge economy increasingly questions the relevance of Coasian organizational economics. Williamson 1985) and backed up by the ownership of alienable assets (Hart and Moore 1990) obtains or not is increasingly irrelevant for understanding the organization of economic activities in a knowledge economy (Grandori 2001). albeit means that do not differ substantially in kind. . but rather they differ in empirical terms. (Zucker 1991: 164) To the extent that important knowledge assets are increasingly controlled by employees (“knowledge workers”) themselves. . . traditional authority relations are fading into insignificance. with its rather strong dichotomization between allocation of resources by means of authority and by means of prices (e. 2000). Furthermore. Demsetz 1991. They refer to different means of organizing economic activity. This is partly a result of the increased bargaining power on the part of knowledge workers (stemming from the control over critical knowledge assets). . authority may be located within the organization . Cowen and Parker argue that firms and markets are not exactly the same. . Rather the difference between the firm and the market as a resource allocator involves what might more usefully be viewed as subtle differences relating to contracting. Boisot 1998. Foss (in a hierarchy) (Liebeskind et al. The specialist nature of knowledge work implies that principals/employers become ignorant about (some of) the actions that are open to agents/employees.g. Thus. thus making the exercise of authority through direction increasingly inefficient. As Zucker argues: While bureaucratic authority is by definition located within the firm’s boundaries. Thus. view does not seek to find a clear-cut distinction between firms and markets. whether direction by means of order giving (Coase 1937.

commercially relevant knowledge is becoming increasingly distributed or dispersed in the sense of Hayek (1945) (e. Economic organization in the Hayekian setting In the interpretation adopted here. tacit and subjectively held (O’Driscoll and Rizzo 1985) but which it may nevertheless be necessary to somehow mobilize (e. Interpretation: organizing in Hayekian settings Hayekian knowledge settings It is no coincidence that so many of those who write on economic organization in the emerging knowledge economy cite Hayek’s work.” This is because the basic propositions (or generalizing statement) about the role of knowledge in production that drive recent debate about organization in the knowledge economy have a strongly Hayekian flavor. I interpret recent debate to assert that the phenomena of knowledge becoming increasingly dispersed and important in production influence authority relations. “Distributed” or “dispersed” knowledge” is knowledge that is not possessed by any single mind and which may be private. The second proposition that appears to be at the heart of much discussion of organization in the knowledge economy is that because of the increased importance of sourcing specialist knowledge. traditional authority relations vanish (Hodgson . particularly his 1945 paper. Coombs and Metcalfe 2000. Hodgson 1998). the two Hayekian propositions above drive recent debate on economic organization in the knowledge economy in the specific sense that the phenomena they describe are claimed causally to produce a host of other phenomena that relate to economic organization (Hodgson 1998).g. it is necessary to consider the Hayekian insights that lead some to conclude that there is not much of a difference between firms and markets. The first one is a claim that because of the increased need for diverse. through a price system) for the carrying out of a productive task or a complex of such tasks. more is involved in the choice between firms and markets than “subtle differences relating to contracting. specialized knowledge in production. 1995. Specifically.g. I shall say that “Hayekian settings” obtain. the boundaries of the firm and the ways in which various mechanisms of coordination may be combined inside firms in regular and predictable ways. in the emerging knowledge economy. When these two propositions are met. It is possible to discern at least two such propositions in recent debate. as. before this argument can be presented. Hayek (1945) famously argued. fleeting. contrary to Cowen and Parker. For analytical purposes.” However.g. Boisot 1998. knowledge assets controlled by individual agents (“knowledge workers”) are becoming increasingly important in production (e. Ghoshal et al. as approximated by the two Hayekian propositions above. Hodgson 1998) in the sense of accounting for a greater part of the value-added of goods. “The use of knowledge in society. “Austrian” determinants of economic organization 147 Much of the following is essentially an argument that. of course.

It is convenient to begin with a discussion of the meaning of authority. discrete forms. of course. 2000). In the present context. More or less authority is then simply defined as making the set A0 larger or smaller. and coordination mechanisms (i. supplemented with a later contribution by Simon (1951). Coase initiated the tendency to see the employ- ment contract and the authority relation as the defining characteristic of the firm. but that coordination mechanisms can be combined in a multitude of ways (e. which is defined as obtaining when a “boss” is permitted by a “worker” to select actions. Helper et al. etc.e. four such puzzles are particularly relevant: 1 What is ultimately the source of the employer’s authority? In other words. Foss 1998. the boundaries of firms blur because of the increasing importance of knowledge networks that transcend those boundaries.) will increasingly be combined in new. the employment contract is explained as one whereby the factor. Coase’s understanding. norms. and high-powered incentives (Miles et al. teams.148 Nicolai J. for a certain remuneration (which may be fixed or fluctuating) agrees to obey the directions of an entrepreneur within certain limits. I present and discuss some of these arguments. prices.g. entrepreneurial control over resources. where A is the set of the worker’s possible behaviors. still provides most economists’ working definition of authority. Authority: Coase and Simon Although Max Weber had much of interest to offer on the topic of authority in the beginning of the twentieth century. authority. A later paper by Herbert Simon (1951) provided a formalization of Coase’s notion of the employment relationship and a clarification of the notion of authority. authority. firms may adopt coordination mechanisms that we normally think of as characteristic of the market rather than of planned coordination. In the following. Moreover. In particular. A0 傺 A. The essence of the power is that it should only state the limits to the powers of the entrepreneur. he can therefore direct the other factors of production. In Coase (1937). Zucker 1991). why .” This final claim implies that organizational forms do not come in a few rigid. it was not until Coase (1937) that an economic conceptualization of authority was offered. (Coase 1937: 242) This contractually agreed upon right to “direct the other factors of production” is. Some puzzles relating to authority The notion of authority in the context of an employment contract as the defining characteristic of the firm raises several puzzles. in particular pricing. Within these limits. Grandori 1997. innovative ways (producing what is often referred to as “new organizational forms. 1997).

e. To some extent it does so by changing the terms of the debate: whereas Weber. while the first question asks about the sources of the employer’s bargaining power over the employee. an implication of this view is that the distinction between the authority-based and the price-based modes of allocation emphasized by Coase (1937) is superficial. The . the firm/market boundary is unclear and the notion of authority elusive at best. As Alchian and Demsetz (1972) argue. Only the first question has been given extensive treatment in the economics of organization. human) assets. human capital) cannot be traded? 2 What happens to the notion of authority in the sense of Coase and Simon if the employer does not possess full knowledge of the employee’s action set (i. it has been one of the classic points of contention in a long-standing debate. In fact. “Austrian” determinants of economic organization 149 exactly is it that the employee accepts to be directed when non-alienable assets (i. Note that this “nexus of contracts” position is remarkably close to the position that in a knowledge-based economy. 4 What happens to the notion of authority if employees control knowledge assets that are “within their heads” and which may give them substantial bargaining power? The latter three questions concern the limits of authority in Hayekian settings. and presumably better. The sources of authority Thus. 1996. between an inter-firm and an intra-firm transaction. namely alienable (i.2 The work of Oliver Hart and others (Hart 1995. while an employee does not. Two kinds of assets are distinguished. it does not make such sense to speak of firms as well-defined entities at all (Cheung 1983). the actions that he can take when uncertainty is resolved). now turns on who owns the alienable assets that an agent (whether independent or employee) utilizes in his work. there is an implicit assumption that the employer is at least as well informed. Simon. In fact. The basic distinction between an independent contractor and an employee. it is not meaningful to assume that an employer can force an employee to do what the employer wants in the absence of coercion. Hart and Moore 1990) – called the incomplete-contracts literature – provides a response to the Alchian and Demsetz/Cheung view. and others focused on direct authority over (non-alienable) human assets.e. there is no economic difference between “firing” one’s grocer and firing one’s secretary. the incomplete-contracts literature rather explains authority over human assets as something that is indirectly acquired through ownership of alienable assets. etc. about the efficiency implications of alternative actions. that is. An independent contractor owns his tools.e. Coase. to writers such as Alchian and Demsetz (1972) and Cheung (1983).. so that the employee can take actions about which the employer has no knowledge? 3 What happens to the notion of authority if the employee is better informed than the employer with respect to how certain tasks should (optimally) be carried out? In the works of Coase and Simon. non-human) and non-alienable (i.e.

this is. Thus.e. However. and operating with rigid routines and operating procedures. Given this. However. Efficiency considerations then suggest that authority (i. since they do not directly concern the issue of the sources of authority. the fact that firms exist is prima facie evidence that they can somehow cope with the problem and/or that there are offsetting benefits of firm organization. exertion of effort or investment in human capital). and the implication that the entrepreneur assumes his directing role because his inputs to the venture are those that matter the most for the total monetary surplus. the argument that the boundaries of the firm lie where the entrepreneur’s ownership of alienable assets stops. bound to lead to disaster. They are of a different nature. the incomplete-contracts approach is a neoclassical approach. which is specific to the asset. The incomplete-contracts perspective does not provide an immediate answer to these.3 In turn. the planning problems posed by Hayekian distributed knowledge have become increasingly pressing for firms (see Cowen and Parker 1997).4 Distributed knowledge and delegated rights How may the Hayekian knowledge-problem be handled in firms? One way is to suppress distributed knowledge as far as possible by discouraging local initiative. we are still left with puzzles (2) to (4). In one interpretation. notably the emphasis on ownership as backing up authority. indoctrinating employees harshly. not an Austrian one. since the pattern of ownership will influence the parties’ outside options. The parties to a relation (whether customer and grocer. As Mises (1949: 303) emphasized. ownership to the alienable assets) should be allocated to the agent who makes the most important (non-contractible) relation-specific investment. depends on who owns the asset. In a dynamic economy. The Hayekian challenge to authority However. I shall rely on these basic ideas in criticizing some of the claims put forward by writers on economic organization in the knowledge economy. or employer and employee) are seen as being in a bargaining situation.150 Nicolai J. Foss importance of asset ownership derives from the fact that the willingness of an agent to undertake a non-contractible investment (say. Something else must be done. the expectation of this division feeds back into the investments that the parties are willing to make. they are about the reach and efficiency of authority in Hayekian settings. coping with the problem posed by Hayekian distributed knowledge has moved from being a problem for socialist managers and dirigiste bureaucrats to also being a problem confronted by managers of (at least large) firms in capitalist economies. “entrepreneurs . however. each having an outside option. Of course. some interesting aspects in it make it complementary to a Misesian perspective. the division of the surplus from the relation will depend on who owns the alienable assets in the relation. because of the increased importance of specialist workers and the increased knowledge-intensity of production. rather.

“molecular forms”. Boisot 1998). the implication is that ownership over such assets is an increasingly ineffective source of bargaining power and that. and other manifestations of organizational delegation and decentralization.g. to better serve customer preferences) and structure reward schemes in such a way that optimal tradeoffs are reached. such as team-organization. Of course. They cannot themselves attend to the manifold tasks which are incumbent upon them. assets. These are prompted by a market-driven pressure to delegate decision rights (e. This provides a useful perspective on many of those new organizational forms that are argued to be characteristic of the knowledge economy (cf. Thus.. in the Misesian scheme. implement employee stock-ownership programs. Adding to the puzzle is that authority in the sense of Coase or Simon appears to play at best a very limited role under Hayekian dispersed knowledge. or perhaps even irrelevant. firms increasingly rely on high-powered incentives. Moving teams out of firms would appear to yield net benefits.” try to price corporate resources to the largest possible extent. “Austrian” determinants of economic organization 151 are not omnipresent. since incentives would be further strengthened. and since such assets are of declining economic and commercial importance. we are still left with the puzzle why such teams are organized inside firms. as numerous writers have emphasized. it is obvious that the very notion of the firm’s boundaries is becoming increasingly fuzzy. resort to other coordination mechanisms is necessary. therefore. This is because the assets that in Hart’s scheme confer authority are physical assets. an important aspect of the knowledge economy is precisely that physical assets are of strongly waning importance (e. An . invest in building “corporate cultures. Finally. Thus. through delegation of decision rights to managers (Mises 1949: 305). authority must wane as bargaining power increasingly becomes more symmetrically distributed over the owners of knowledge assets. Hayekian settings and economic organization However. Cowen and Parker 1997). an organizational equilibrium obtains where decision rights are delegated in such a way that the benefits of delegation in terms of better utilizing local knowledge are balanced against the costs of delegation in terms of agency losses (as in Jensen and Meckling 1992). because authority declines in importance as knowledge becomes distributed and knowledge inputs increase in importance. This is because the Coase/Simon notion of authority assumes that a directing principal is at least as knowledgeable about the relevant tasks as the agent being directed. Since the bound- aries of the firm are (also) defined in terms of legally recognized ownership to the firm’s alienable. primarily physical. etc.” so that coping with distributed knowledge leads in the direction of decentralization (cf.g. The ownership-based notion of authority developed by Hart also seems to play only a limited role under Hayekian distributed knowledge. However. but argued that the system of double-entry bookkeeping and other control measures may partly cope with such problems. etc. being subject to the exercise of authority. also Hayek 1945: 83–4). Mises also recognized that dele- gation leads to agency problems. while an Austrian perspective is useful for understanding why firms adopt team organization.

and the combinability of coordination mechanisms. and hybrids (e. Foss outcome of this is the emergence of “new organizational forms. the boundaries of the firm.e. Nobody denies that in the emerging knowledge economy. and that his venture can rely on all sorts of combinations of coordination mech- anisms. The following section discusses the reach of these arguments. Authority in Hayekian settings In this section. One way of doing so is to focus on “hidden knowledge” (Minkler 1993) in relations between a principal (e. hires the managers and “technicians. Since I later discuss the importance for economic organization of the distinction between physical and knowledge assets. where economic activities are normally assumed to be organized across three discrete governance structures. I assume that Hayekian settings obtain and discuss the implications of such settings for the Coasian themes of authority. 1997). where knowledge is distributed and knowledge inputs are more important than physical inputs.). in particular that he can offer employees incentives that in terms of their strength (i. boundaries and malleability of coordination mechanisms seriatim. effectively mimicking the effects of market pricing. Williamson 1996). and acquires ownership of the firm’s alienable assets. arguments can be made that Hayekian settings. make the bound- aries of firms blur. On the contrary. In sum. The underlying perspective is Misesian. people who have the ability and skill to perform definite kinds and quantities of work” (ibid. legal sense. Discussion In this section. that the boundaries of his venture will become ill-defined (if not in a formal. many recent writings on the knowledge economy very strongly stress entrepreneurship (e. then in an economic and commercial).152 Nicolai J. here I only concentrate on the distributed knowledge aspect of Hayekian settings.g. Miles et al. I discuss the three issues of authority. present real problems for the exercise of authority in firms. the strategy is to examine the role of authority in Hayekian settings. there will still be a need for such enterprising agents.g. and remove many of the constraints on the malleability of coordination mechanisms. firms.g. the Misesian .” The theoretical implication is that various mechanisms for coordinating resources are combined to a much larger extent than hitherto assumed in. in the sense that I shall throughout assume the existence of a speculating entrepreneur who is ultimately in charge of the business venture in the sense that he determines “the general plan for the utilization of resources” (Mises 1949: 303). organizational economics. for example. markets. determines “the expansion and contraction of the size of the business and its main sections” as well as “the enterprise’s financial structure” (Mises 1949: 307). i. What is being claimed is rather that the entrepreneur will no longer be able to exercise much authority.e. the way in which they link effort and rewards/punishment) are very close to the incentives provided under market contracting.

or both. hidden information). the more attractive the central planning solution is” (1990: 805). as in the usual agency model (Holmström 1979). Loosely.e. information is (strongly) decisive if – in a setting involving many cooperating individuals – a decision can reasonably be made on the basis of this information without involving other pieces of information (Casson 1994). a hired manager). Moreover. The general principle is that decision rights . because doing nothing is worse. Bolton and Farrell conclude that “the less important the private information that the planner lacks and the more essential coordination is. In the context of a specific model of this trade-off. “Austrian” determinants of economic organization 153 entrepreneur) and an agent (e.” “decisive information. Coordinated adaptation or action may be required when actions or activities are complementary (K. it will be assumed that the problem facing a principal is not just that he is uninformed about what state of nature has been revealed or of the realization of the agent’s effort (i. he may in many cases still hold the information that is decisive. As I shall argue. it is possible to explain the presence of authority in such a setting. Centralization is assumed to not involve delay and therefore is a good mechanism for dealing with emergencies. Foss 2000. That is. when it is important to make some urgent choice (possibly highly inefficient).” and “defining incentive systems. or the agent may be better informed than the employer with respect to how certain tasks should (optimally) be carried out. According to Casson (1994). but that the agent’s knowledge is superior to that of the principal with respect to certain production possibilities (i. hidden knowledge). The principal may be ignorant about some members of the set of possible actions open to the agent. Milgrom and Roberts 1990). it may be better to have somebody pick a strategy and make everybody play this strategy. a conclusion Bolton and Farrell argue is consistent with the observed tendencies of firms to rely on centralized authority in cases of emergencies. In such cases.g.” “economies of scale in decision-making. if the inefficiencies from picking a bad strategy are smaller than the inefficiencies from delaying a coordinated solution. the extent to which a problem involving the knowledge of several individuals has decisiveness features and the cost at which knowledge can be communicated helps to explain the allocation of decision rights.e. Decisive information Even under distributed knowledge.” The need for urgent coordination While Hayek (1945) did much to identify the benefits of the price system in the context of alienable property rights in coping with distributed knowledge and unexpected disturbances. for example. he arguably neglected those situations where efficiency requires that adaptation be “coordinated” rather than “autonomous” (Williamson 1996). the decentralized solution performs poorly if urgency is important. where the centralized decision-maker per definition does not possess (at least some) local information. I discuss them under the headings of “the need for urgent coordination.

Economies of scale in decision-making Demsetz (1988) argues that economies of scale in managing are a neglected factor in the explanation of the existence of firms and the understanding of authority. and if it is costly to communicate the entrepreneur’s knowledge. In fact. the more likely is it that the entrepreneur will choose to rely on multiple incentive instruments to influence employee behavior (Henderson 2000). there may also be substantial learning economies. Both Knight (1921). they did assume that the entrepreneur can rationally delegate decisions to managers and control these. If the knowledge possessed by. None of them assumed that entrepreneurs would have full knowledge of their managers’ action set. then overall decision rights should be concentrated in the hands of the entrepreneur.. However. suppliers. if the knowledge possessed by the entrepreneur is decisive. Minkler 1993). In a dynamic economy. Foss will tend to be concentrated in the hands of the individual who has access to the decisive information. Hidden knowledge does not imply that subjective performance measurement becomes impossible. managers is not decisive. it may be conjectured that the more we depart from simple settings where employees are very easily monitored.” However. clearly allowed for this possibility. even under hidden knowledge the principal may be able to form conjectures of the financial results that result from the agent’s activities. discussing the entrepreneur delegating responsibilities to managers.154 Nicolai J. using the authority mechanism also seems to become “pointless. This provides a further argument for authority under hidden knowledge. etc. there may still be a role for authority. and the more complicated the control problem becomes. and compensate him accordingly. still. he does not explain the underlying reasoning. learning about potential suppliers. the relevant economies may relate both to managing the internal relations between agents inside the firm and managing relations to outside agents (customers. discussing business “judgment” and Mises (1949: 303). Economies of scale . maintaining coherence between such instruments may be a recurrent task. that is. Not only may there be scale economies in such activities. even under hidden knowledge. Defining incentive systems It is hard to deny that Hayekian settings pose special problems for the use of monitoring mechanisms and incentive pay (Aghion and Tirole 1997. govern- ment agencies) (Hermalin 1998). Other agents may be happy to let a central agent incur the effort costs of negotiating. However. he should assume ultimate authority in the firm. For example. Minkler (1993: 23) argues that “if the worker knows more than the entrepreneur. and particularly so the more costly it is to communicate this information. He can check whether these conjectures are actually confirmed using the control systems of the firm.” which implies that to the extent that monitoring is a precondition for the exercise of direction. Foss 1999. for example. it is pointless for the entrepreneur to monitor the worker.

the purpose of the present section is to go more into ownership issues. an insight that is not present in Coase (1937) and Simon (1951). but they are delegated as means to an end (Hayek 1973). The argument that will be critically discussed is that. decision rights are delegated in firms. as Mises emphasized. It is possible to use the incomplete contracts framework (Hart 1995. it is often efficient to centralize ownership to alienable assets. As I shall argue next. 1999). I did. hint that ownership would play a key role. when there is such a need. and therefore the issue of the boundaries of the firm. to get an under- standing of the implications of knowledge assets for the boundaries of the firm. the boundaries of firms will blur. however. is an increasingly unrealistic conceptualization of authority.5 Thus. In turn. as a relation in which the principal has superior knowledge and can observe all contingencies that require a response by some employee. “Austrian” determinants of economic organization 155 in this task may dictate that this activity is centralized. Recall that in this approach. Per implication. Both arguments point towards the centralization of decision rights To sum up. authority as understood by Coase and Simon. Ownership and the boundaries of firms In the previous section. their use is monitored (Jensen and Meckling 1992). This is not to say that authority relations will remain unaffected by the arguably increasing importance of Hayekian settings. Hart and Moore 1990). even in “knowledge-based” firms. it has been argued that it is possible to give efficiency explanations of authority in the context of Hayekian settings. approximated by hidden knowledge. This suggests that authority in the sense of direction and centralized decision-making – which. Mises clearly recognized that in many firms decision rights are allocated by the entrepreneur (and the board of directors) to lower levels. I did not say much about what backs up authority. However. summarized very briefly earlier. centralization is required to the extent that externalities arise when the instruments are controlled by separate firms and transaction costs hinder the internalization of these exter- nalities. and authority may have important efficiency implications. Hart 1996. Moreover. Mises also understood that such rights are circumscribed in an attempt to cope with the control problem that follows from delegation. as argued earlier. at least to the extent that these are defined in terms of legally recognized ownership of the firm’s alienable assets. To be sure. I argue that when there is a need for centralized coordination. and top-management reserves ultimate decision rights for itself (Baker et al. this suggests that centralized coordination is a feature of firms rather than markets. this kind of authority was already too descriptively narrow with respect to the business firms that Mises (1949) discussed. presumably in order better to cope with distributed knowledge. there may be a need for centralized coordination. asset ownership is central because it provides the bargaining lever that backs up authority. as knowledge assets become relatively more important in production. does not require detailed knowledge about a subordinate’s knowledge or available actions – may persist in Hayekian settings. efficiency considerations often suggest a need for also . Actually.

aggregates information from the messages of the scientists and directs their efforts. This will strengthen the entrepreneur’s incentives (the scientist cannot hold him up anymore) and it will leave the scientist’s incentives unaffected. Given this. In fact. the entrepreneur can hold-up the scientist by threatening to withdraw from the relation. Given these assumptions. Therefore. for example. owns the only other asset in the relation which we may assume to be a patent. who is again equipped with a non- alienable entrepreneurial idea in which he may invest further. One can show (details in Hart and Moore 1990 and Brynjolfsson 1994) that because of the externality problem that the hold-up threat creates. It is prohibitively costly to communicate the knowledge embodied in the entrepreneurial idea from the entrepreneur to the scientist. The conclusion is that it is possible to speak of the boundaries of the firm in terms of ownership – even in a situation where all relevant productive assets are knowledge assets. governing the use of the assets in all contingencies. assume that two agents interact and that one of these. so that the one is of zero value without the other. the entrepreneur. Foss concentrating asset ownership. The entrepreneur may improve on this decisive knowledge. has decisive information (in the sense discussed earlier). Both assets are necessary to create value in the relation. this ownership arrangement should be chosen. all actions of the other agents produce zero value. by assuming that one of the agents. since the latter needs access to the patent to create value (and the contract is incomplete). the entrepreneur. it is not possible to write a comprehensive contract. makes a spin-off patent. should he also be an owner? Consider a bigger “knowledge-based” firm where there is a group of scientists who each owns a patent. owns a knowledge asset that is “inside his head” (e. and they are (strictly) complementary. it is possible to dispense entirely with physical assets. every agent underinvests. We can address this issue. His knowledge is decisive in the sense that without it. so that if the scientist makes an effort investment. the scientist can effect a hold-up on the entrepreneur. specifically. Moreover. Of course. the scientist. While efficiency may require that this agent should have decision rights amounting to authority (as argued earlier). an entre- preneurial idea) and the other agent. and discuss a purely knowledge-based firm.g. Each agent needs access to his own patent and to the entrepreneur’s direction in order to be productive. that is. each party invests to the point where the marginal cost of effort investment equals one-half of the marginal value (because they are assumed to split the extra surplus 50:50). this does not yet demonstrate the point made earlier. we again have the hold-up . namely that concentration of coordination tasks produces a need for concentration of ownership. the reverse also holds. elabo- rates on his idea and creates extra value. The key to this argument is to introduce knowledge assets explicitly.6 For simplicity. however.156 Nicolai J. if the entrepreneur makes an effort investment. so it is effectively non-alienable. Suppose now that the entrepreneur owns both the patent and the entrepreneurial idea. The entrepreneur. In this setting. However. we may ask who should own the (alienable) patent which – in terms of incomplete contracts approach – is the same as asking who should own the firm.

notably in their investment decisions. since managers could not be sure that they would not be overruled by the planning authorities. they were not likely to take a long view. would not make efficient decisions. it was argued that there is a connection between authority and ownership. As he (1949: 709) explained.” However. the patents. rely on the prices of outside markets. a good deal of recent analytical energies have been devoted to the commitment problems of delegation in firms (e. in addition to his “pure” calculation argument. However. Mises argued that these schemes would be unworkable. To an important extent this is a matter of the sheer impossibility of having rational calculation when crucial markets are eliminated. to a much larger extent. in the hands of a central planning board would dilute the incentives of socialist managers. introducing pricing in the context of hierarchy. the hold-up problem disappears. since managers were not the ultimate owners. Moreover. In particular. But Mises also placed much emphasis on property rights and ownership issues (particularly Mises 1936). Williamson 1985). leading the entrepreneur to choose inefficient investment levels. Moreover. While planning authorities could (and according to the schemes of the day. With reference to various socialist schemes of his day that tried to preserve some market relations while eliminating capital and financial markets. does not imply that firm organization is “impossible. some of the property rights insights into socialism also apply to firms. Thus. they were not the full residual claimants of their decisions and. since reneging on a promise to delegate will in many cases be extremely tempting and those to whom rights are delegated anticipate this. if the entrepreneur is given ownership to the alienable assets. the Misesian calculation problem.” that is. First. and that this link also exists in Hayekian settings. that is. Miller 1992. while constraining the efficient size of firms (Klein 1996). Thus. Baker et al.7 . A main conclusion is that credible delegation may be very hard. 1999. hence. there are inherent contradictions involved in “playing market. he was aware that the concentration of ultimate decision- making rights and responsibilities. Firms have the great advantage relative to socialist planning boards that they may. Any one of the scientists can hold up the entrepreneur on his investment. This prompts the question of whether there are other necessary “links” between organizational elements. The combinability of coordination mechanisms Earlier paragraphs have established that it is possible to give efficiency reasons for authority as well as legal/ownership-based notions of the boundaries of the firm in Hayekian settings. these rights could not be used efficiently. Thus.g. should) delegate rights to make production and investment decisions to managers. “Austrian” determinants of economic organization 157 problem. Mises also put forward property rights argu- ments why the attempt to “play market” under socialism would only lead to inefficiencies. and therefore ownership. it is pertinent to turn once more to Mises’ work. Thus. At this stage. this ownership arrangement should be chosen.

the problem is stated in the following way (cf. Ultimately. But if he overrules the subordinate. this is because authority and ownership will continue to be important in the knowledge economy. An implication is that mixing very different coordination mechanisms may lead to efficiency losses. However. one cannot simply take a subset of these away.” because the formal right to ratify is still in the hands of the manager and because that right cannot be allocated to the subordinate through a court-enforceable contract (cf. reducing product . and how much the manager values his reputation for not reneging relative to the benefits of reneging on a bad project (for details and extensions.158 Nicolai J. The particular equilibrium that emerges will be determined by the discount rate of the manager. and it is the inherent tension between ownership and delegated rights that create the incentive problem. These insights suggest. The basic problem is that emulating market organization inside firms amounts to “playing market. he may be tempted to renege on a promise to delegate decision authority. Oticon became one of the best-known and admired examples of radical organizational turnaround. Foss In a recent treatment. The turnaround aimed at increasing employee empowerment and responsibility. Assume that a subordinate initiates a project. unhampered capital markets. This means that those who possess ultimate decision rights can always overrule employees. Williamson 1996). will he lose trust in the manager for all future periods if he is overruled?). on the most basic level.8 Assume further that the manager has information that is necessary to perform an assessment of the project. this will induce more effort in searching for new projects (Aghion and Tirole 1997). this amounts to full informal delegation of the rights to initiate and ratify projects – “informal. 1999). The expected benefits of these increased efforts may overwhelm the expected costs from bad projects that the manager has to ratify. and may not be feasible for this reason. that is. this is a sort of application on the level of the firm of Mises’ demonstration that the various elements that make up the capitalist market economy are complementary ones. holding back on effort. Thus. Effectively. In a sense. Oticon (now William Demant Holding A/S) is a world leader in the hearing aid industry.9 Illustration: trying spaghetti in Oticon Founded in 1904 and based mainly in Denmark. They are not full owners. but that he decides upfront to ratify any project that the subordinate proposes. 1999). there are incentive limits to the extent to which market principles can be applied inside firms. and substitute them with elements that are characteristic of a different system. Baker et al.g. corporate employees never possess ultimate decision rights. see Baker et al.” Unlike independent agents in markets. the specific trigger strategy followed by the subordinate (e. say.10 For a number of years in the beginning to the mid-1990s. intervene in a “selective” manner. in this game a number of equilibria is feasible. the latter will lose trust in him. Clearly. that coordination mechanisms are not simply combinable in an arbitrary fashion. the problem is that because the manager has information about the state of a project (“bad” or “good”). Because the subordinate values are being given freedom.

employees now had a choice to decide which projects they would join. etc. Thus. In the Oticon simulation of the market. Project groups were self-organizing in much the same way that. Rather than being assigned tasks from the above. first. It should be capable of combining and recombining skills in a flexible manner.11 The project team was required to undertake all the tasks connected with product development until the product was successfully introduced in all markets.12 Interpreting spaghetti From an Austrian economics point of view. For example. the intention was that the organization should mimic the market in such dimensions as flexibility. The setting of salaries was decentralized to project leaders. All projects were to be announced on an electronic bulletin board. Development projects could be initiated by. As Hayek (1945) explained. for example. that employees were not restricted in the number of projects they could join. although project teams were self- organizing and basically left to mind their own business once their projects were ratified. “Austrian” determinants of economic organization 159 development cycles. employees. that employees were actively encouraged (and in the beginning actually required) to develop and include skills outside their skill portfolio. machines. Most hierarchical levels were eliminated and formal titles done away with. Finally. autonomy. any employee.) that broke with the boundaries imposed by the old departments. they received the right to negotiate salaries. they were still to meet with a “Project and Product Committee” once every 3 months for ongoing project evaluation. the immediately noticeable aspect of the spaghetti organization is the importance of the market metaphor in the design of the new administrative structure (Lyregaard 1993). where skills and other resources would move to those (new) uses where they were most highly valued.g. employees were given many and far-reaching decision rights. Departments gave way to “competence centers” (e. almost 100 percent project-based organization. and building new knowledge.” yet based on “free market forces” (Lyregaard 1993). The new organization amounted to breaking down the earlier functional department-based organization into an almost completely flat. audiology. These goals would be reached by means of a radical project-based organizational structure that should be explicitly “knowledge-based” (Kolind 1990) and “anthropocentric. in principle. fleeting . the main problem is that much of this knowledge is transitory. in mechanical engineering. increasing contact to customers. and accordingly received a considerable amount of decision-making power. partnerships are self-organizing. where employees who would like to join them could sign in. or customer preferences. etc. A major problem that besets centralized decision-making systems – in large firms as well as in centralized economies – is that they have difficulties efficiently mobil- izing and utilizing important local knowledge. just like entrepreneurs in a market setting. flatness. mobilizing dispersed and “hidden” existing knowledge. and. such as the precise characteristics of specific processes. The much noted “multi-job” principle meant. Project managers were free to manage the project as they preferred. second.

in other words. were the real holders of power – they possessed ultimate decision rights. making preliminary plans. last but certainly not least. the market- like spaghetti organization was to be kept together by a shared set of values (Kolind 1994). namely that of making sure that decision rights are utilized efficiently. rather initiation and implementation rights may be controlled by one person (or team) while ratification and monitoring rights are controlled by other persons. etc. Thus. staffed by Kolind and three other managers. This committee. Several of the components of the spaghetti organization may be seen as responses to this fundamental agency problem. Fama and Jensen 1983). In general. Markets are not plagued by these type of problems to the same extent. By giving project teams extensive decision rights. There was no a priori guarantee that project leaders and other employees would act in the interest of the firm. making contacts. The rights to allocate resources to a particular project may be broken down into the rights to (1) initiate a project. the spaghetti organization stimulated a co-location of decision rights with local knowledge. advanced information technology. its use of “free market forces” (Lyregaard 1993) was fundamentally a simulation. However. The Oticon spaghetti organization was very much an attempt to mimic the market in these dimensions. the charismatic leadership of CEO Lars Kolind himself. firms usually do not concentrate these rights in the same hands. Jensen and Meckling 1992). research or development task. For reasons of efficiency. Those who held the relevant knowledge were also to have the authority to decide over the use of company resources. the primary purpose of which was to approve of or reject proposed projects (the Projects and Products Committee). markets tend to economize on the costs of transferring knowledge by instead allocating decision rights to those who possess the relevant knowledge (Hayek 1945. (3) implement projects. a basic problem in the old organization had been that commercially important knowledge simply did not reach the relevant decision-makers. Foss and/or tacit. by a committee. In lieu of a distinct price mechanism that could coordinate actions. Oticon remained a firm. . usually hierarchical superiors. doing the required calculations. for the full decentralization of decision rights that characterizes market organization never took place in Oticon (and neither could it). and (4) monitor and evaluate projects (cf.160 Nicolai J. Rather than involving the transfer of costly-to-transfer knowledge to those with decision rights (as in a command economy or a centralized firm). firms confront a problem that markets confront to a smaller degree. (2) ratify projects. in the sense of sketching. at least within limits. making ideas for projects public and requiring that teams/project groups possessed the necessary complementary skills for a particular marketing. Thus. as well as the strongly overlapping top management committee. and.13 This allocation of control rights corresponds to that of the Oticon spaghetti organization. and therefore costly to articulate and transfer to a (corporate) center. anybody could initiate a project. Decision rights It is the same co-location that takes place in a well-functioning market. the problem of moral hazard.

projects had to be evaluated by the Products and Projects Committee (PPC) that was staffed by Kolind. Retreating from spaghetti A retreat from the radical spaghetti organization that Kolind had implemented in 1991 began long before he resigned as CEO in 1998.” “team technology. the new CEO. the marketing manager and the support manager – The PPC was the real holder of power in Oticon. Many of the decision-making rights held earlier by project leaders have now been concentrated in the hands of the Competence Center. or the managers of the business teams. Project leaders are appointed by the Competence Center. a “Competence Center” is in charge of all projects and their financing and of an operational group controlling administration. It is one of the successors to the abandoned PPC. The internal market was. many of the crucial elements of the spaghetti organization have been left. The team leaders and the head of the Competence Center comprise. These teams refer directly to Niels Jakobsen. Thus. its style of managing the projects is very different. logistics. as under the spaghetti organization. IT. however. What happened? Some basic design problems Insights from organizational economics and Austrian economics suggest that although the spaghetti organization was characterized by substantial coherence obtaining between its complementary elements. sales. “Austrian” determinants of economic organization 161 However. In 1996. In addition to the business teams. In particular. and exports.e. the “Development Group. the right to be a project leader is not something that one grabs.” and “team high volume”) which function as overall administrative units around projects. Each business team is managed by two team leaders. namely a technician and a person with marketing or human resource skills. Although Oticon is still characterized by considerable decentralization and delegation of rights. together with the CEO. Frequent intervention on the part of the Committee ex post made that clear to everybody. project evaluation and monitoring). Much of the initiative with respect to starting new projects is taken by the Development Group.” which essentially is the senior executive group and is in charge of overall strategy making. Although a considerable amount of variety was indeed allowed to evolve. Project teams were required to report to the Committee on a 3-monthly basis. and its prevalence has been much reduced. the selection over this variety was very much guided by the visible hand of the PPC (Lovas and Ghoshal 2000). very much a managed one.e. and the Committee could at any time halt or close down projects. the development manager. Oticon headquarters was divided into three “business teams” (“team advanced. care is taken to avoid the erratic behavior with respect to intervening in already approved projects that characterized the PPC. Although multi- jobs/multi-tasking are still allowed. this practice is no longer directly encouraged. something which happened quite frequently. decision management (i. it was still beset by a number of . initiation and daily project management) was separated from decision control (i. in actuality.

knowledge tended to be held back within projects. . this was taken by the PPC to be a quite natural feature of a flexible. plans would have to be adjusted or remade in an ever-continuing process. 1998). Second. because of the widespread. and correct. given that anybody at the project could leave at will. Third. having been more carefully researched beforehand. . contrary to the aim of making Oticon a knowledge-sharing environment. since it abolished tournaments between managers. Hierarchy could not be used anymore as a sorting mechanism for allocating skills so that those with more decisive knowledge would obtain authority over those with less decisive knowledge.15 However. influence activities (Milgrom 1988) were important under the spaghetti organization. Selective intervention was partly motivated by the fact that the “PPC does not make general written plans. the official rhetoric of a flexible market-based structure. if this were done. The present Oticon organization is characterized by a much more consistent approach towards projects on the part of the Competence Center (one of the descendants of the PPC).e. which are accessible to the rest of the organization . from an incentive perspective.14 Fifth. if noticing a superior opportunity in the internal job market. it also led to diluted incentives and strongly harmed intrinsic motivation (as documented at length by Eskerod 1997. the multi-job led to severe coordination problems. leading to a problem of the allocation of managerial competence. Thus. perception that projects were essentially in competition over resources. with substantial autonomy and the management team (i. with the “dynamics of interventionism. because project leaders had very little guarantee that they could actually carry a project to its end. Foss problems that may arguably have been among the causes of its partial abandon- ment about 5 years later. However. Thus. Second.162 Nicolai J. was increasingly at odds with the frequent selective intervention on the part of the PPC. the PPC preferred to intervene directly in projects. In fact. projects now rest on generally more secure ground. instead of drafting and continuously revising plans under the impact of changing contin- gencies. Lyregaard 1993). Personal relations to those who staffed the committee became paramount for having a project ratified by the committee (Eskerod 1998: 80). Selective intervention in Oticon It is arguable that one of the reasons why the spaghetti organization was changed into a more hierarchical organization has to do with the sort of problems described by notions of selective intervention. because the old plans had become outdated” (Eskerod 1998: 80). the PPC) acting as little more than facilitator and coordinator (Kolind 1990. the extremely flat spaghetti organization sacrificed an incentive instrument. Monitoring systems apparently could not cope satisfactorily with these problems. Fourth. and there is a stated policy of sticking to ratified projects. Apparently. the spaghetti organization eliminated most hierarchical levels. First. perhaps the most important incentive problem related to the behavior of the PPC. project-oriented organization (Eskerod 1998: 89). reputation mechanisms were not sufficient to cope with this problem. . First. Projects are rarely stopped or abandoned.” if you like.

delegation increases. and it may be argued that there is a certain imbalance in the above writings because of their neglect of these insights. one of the main problems of the old spaghetti organization was that Kolind and the PPC never committed in this way. I have argued that Mises’ insights in entrepreneurship. diluting incentives) by overruling ongoing projects is strongly stressed. provides some useful insights. property rights and the complementarity of elements in economic systems are useful ones for claiming a role for authority and the boundaries of firms. In particular. while Austrian insights are useful for interpreting recent claims. and necessitates the use of multiple coordination instruments to utilize this knowledge efficiently. . they were likely to possess the decisive overall knowledge. Foss 1999. as a number of writers have already pointed out (Cowen and Parker 1997. is a task of almost forbidding complexity. Yet. Ellig 1993.e. these insights are also useful for understanding their reach. as well as for helping to uphold the notion that there are discrete organizational forms (e. However. Conclusion The understanding of the dynamics of economic organization. Kolind’s view appears to have been that in important respects and in many situations. neither. In contrast. markets. To the extent that increasingly firm hierarchies do flatten. 2001). the boundaries of firms and firms’ use of distinct coordination mechanisms.g. did they intend to do so. and examine these in the light of organizational economics and Austrian economics. apparently. such as what will happen to authority relations. Apparently. Notes Some parts of this chapter draw on material in Foss (2000. On the other hand. and hybrids). and that efficient utilization of resources dictated intervening in. firms. Misesian insights are helpful here. Ellig and Gable 1993. Thus. Ghoshal et al. projects. a com- bination of organizational economics and Austrian insights. 1995). “Austrian” determinants of economic organization 163 the wish to avoid harming motivation (i. that view clashed on a basic level with the rhetoric of widespread delegation of decision rights. Thus. functions are spun-off in an attempt to improve incentives. makes the boundaries of firms blur. and sometimes closing down. The approach of this paper has been to try to distill some key assumptions and propositions that characterize much of this literature. and that coordination mechanisms cannot be combined arbitrarily. etc. much of this may be interpreted using insights originally put forward by Hayek. it has been argued that much of the recent discussion of economic organization in the knowledge economy may be distilled into a basic assertion that the kind of knowledge that Hayek (1945) talked about represents an increasingly binding constraint on the exercise of authority. primarily represented by the works of Hayek and Mises. The comments of Pierre Garrouste are appreciated. the present Oticon management has realized the need to credibly commit to a policy of non-interference with ongoing projects.

10 The history of Oticon prior to the introduction of the spaghetti organization is extensively covered in Poulsen (1993) and Morsing (1995). 9 See also Milgrom and Roberts (1990) for an important discussion of complemen- tarities. Boisot 1998) not whether assets are material or immaterial.g. 1996). . Hart and Moore 1990. because they are unenforcable. e. Kolind introduced an employee stock program. use the asset in the service of a competitor firm. 15 Foss (2000) discusses whether these design mistakes were remediable.g. and Klein and Klein (2000) treat corporate governance issues in a Misesian manner. This is because the option to intervene “can be exercised both for good cause (to support expected net gains) and for bad (to support the subgoals of the intervenor)” (Williamson 1996: 150–1). and he invested 26 millions DKK of his own funds in Oticon. concluding that they were not. the Austrians anticipated ideas that have become prominent in the modern economics of organization. 12 Complementary measures were taken to back up these initiatives. Williamson argues. 3 For example. 8 This should be understood in a broad sense: a “project” may refer to many different types of decisions or clusters of decisions. 5 For example. which was motivated by the need to raise needed additional money for the transformation.” The main problem is that incentives are diluted. the most crucial variable with respect to determining salary changes in the present organization is the degree to which an employee contributes to knowledge-sharing. however. 14 Possibly as a reflection of these problems.164 Nicolai J. if the employer owns all the alienable assets. it is understood that that right does not entail the right to. Baker et al. but he cannot take the assets with him. Klein (1996) is an application of the Misesian calculation argument to the issue of the boundaries of the firm. 13 Exceptions may occur when giving subordinates more extensive rights (e. Promises to only intervene for good cause can never be credible. 6 This is because the crucial issue is (contrary to the thrust of some contributions. 1999 for analyses of this). and the employer can ensure that if the employee leaves. a package of initiation. The arguments developed in the present chapter go beyond those in Foss (1994b) by putting more stress on Misesian arguments. average salaries do not appear to have changed. the employee can still quit if he dislikes the employer’s orders (as in Alchian and Demsetz 1972). 2 And see Cowen and Parker (1997) for a statement of the Alchian and Demsetz/ Cheung view in this context. in Gould (1994) and Lovas and Ghoshal (2000: 877–8). but whether they are alienable or non-alienable. and more briefly. 11 Although the variance on the distribution of salaries was increased as a result of the new reward schemes that characterized the spaghetti organization. ratification and implementation rights) strengthens employee incentives (see Aghion and Tirole 1997. For example. Williamson 1985. Foss 1 Foss (1994b) argued that in many important respects. for example. 4 Such as the superior ability of firms to organize transactions characterized by high- levels of relation-specific investments (Grossman and Hart 1986. 7 Transaction cost economist Oliver Williamson has referred to these kinds of problems with his concept of the “impossibility of (efficient) selective intervention. the right to use an asset in certain ways may be delegated. somebody else can take over the job.

J. duplication. Sønderborg: Handel- shøjskole Syd. (1994) “Information assets. (1993) “Internal pricing for corporate services. (1998) Knowledge Assets: Securing Competitive Advantage in the Information Economy. M. “Austrian” determinants of economic organization 165 References Aghion. Available HTTP: <http://www. and Farrell. (1994a) “Ludwig von Mises: Precursor of property rights economics. London: The Institute of Economic Affairs. and Metcalfe. Indianapolis: Liberty Press. Alchian.H. Morsing and K. and Tirole. Oxford: Oxford University Press. (1993) Introduction to Market-Based Mangement. Foss and V. Oxford: Oxford University Press. II.” International Journal of the Economics of Business 1: 47–76.A. M.” Journal of Private Enterprise 13: 133–46.A. (1994) “Why are firms hierarchical?. (1997) “From Austrian economics to market-based management. vol. E. P. J.” in A. W. J. (1937) “The nature of the firm. Fama. K. Hellerup: Oticon. J. S. and Organization 4: 141–61. P.” Journal of Law. Online. and organization. Brynjolfsson.C. A. Casson.) The Austrian School and Modern Economics: Essays in Reassessment. Ellig.” Journal of Law and Economics 26: 301–25. Demsetz. D’Aveni. (2000) “Organizing for innovation: Co-ordinating distributed innovation capabilities”. Ellig. Economics and Organization 15: 56–73. Eskerod. Eiberg (eds) (1998) Managing the Unmanageable For a Decade. H.” Journal of Political Economy 98: 803–26.” in M. Governance and Entrepreneurship. and Parker. Mahnko (eds) Competence. information costs. (1997) “Nye perspektiver på fordeling af menneskelige ressourcer i et projektorganiseret multiprojekt-miljø.J.S. Cheung.” Management Science 40: 1645–62.” working paper in Market-Based Management. K. and Demsetz. (1997) “Formal and real authority in organization. (1994) Hypercompetition: The Dynamics of Strategic Maneuvering. N. J. Copenhagen: Munksgaard.doc> (accessed 15 January 2002).J. and economic organization. dissertation. Foss.. (1999) “Informal authority in organizations. R. Fairfax. Cowen.J.” Journal of Law and Economics 26: 1–22. (1988) “The theory of the firm revisited. Eskerod. Coase. H. T. Bolton. and Gable. (1983) “The contractual nature of the firm. M. . and delay.) (1999) The Theory of the Firm: Critical Perspectives in Business and Management. London: Routledge. Economics. J.” Ph. Foss.J. Gibbons.” Journal of Law. (1998) “Organising by projects: Experiences from Oticon’s product development function. technology.D. and Jensen. Coombes. VA: Center for Market Processes. P.” in N. G. and Murphy.” Journal of Political Economy 105: 1–29. Foss (ed. (2000) “Organizing technological interdependencies: A coordination perspective on the firm. R. Ellig. (1997) Markets in the Firm: A Market Process Approach to Management. New York: Basic Books. P. Centre for the Study of Market Processes. R.” forthcoming in Industrial and Corporate Change. R. Alchian (1977) Economic Forces at Work.” in N. Boisot. (1990) “Decentralization.N. Foss (ed. J.S. Baker. (1983) “Separation of ownership and control. (1972) “Production. in N. George Mason University.

. (1996) “What works at work: Overview and assessment.A. (2001) Organizations and Economic Behavior. vol. and Sabel. Moran. S.doc> (accessed 15 January 2002).” in S. Hart. Jensen.. A. M.” Online. and Wruck. T. D. N.” Journal of Law.P. P.” Rationality and Society 8: 371–86. (1998) “The firm as a non-economy: Some comments on Holmstrom. Werin and H. and Moore. Hart. O.A. N.G. Foss. N.. Available HTTP: <http://www. (1999) “The use of knowledge in firms. London: Prentice-Hall.” in L. C. Levine.C. (1995) “The essence of the megacor- poration: Shared context. MacDuffie.J. Grandori. Hart. (2001) “Economic organization in the knowledge economy: Some Austrian insights. Helper. 1: Rules and Order. (1996) “An economist’s view of authority. (1995) Firms. M. Foss.” Journal of Management and Governance 1: 29–42. A.A. J. M. (1990) “Property rights and the nature of the firm.. O. Economics and Organization 15: 103–5. S. Cheltenham: Edward Elgar.-F.” in F. (1998) Economics and Utopia. specific knowledge and total quality .” Review of Austrian Economics 7(1): 31–65.A.” Journal of Institutional and Theoretical Economics 151: 748–59.J.. Olson. M. von Hayek (1948) Individualism and Economic Order. (2000) Managing Know-Who Based Companies.. Manzoni (eds) (1999) Process Re-engineering. London: McGraw-Hill. and Liberty.. Aldershot: Edward Elgar..” in N.166 Nicolai J.” Journal of Political Economy 98: 1119–58. S. and Almeida-Costa. G. Dutta and J.J. (1994) “Science. Gould. B.H.” Industrial and Corporate Change 9: 443–87. Hodgson. von Hayek (1948) Individualism and Economic Order. F.” Bell Journal of Economics 10: 74–91. O. Hermalin. Foss. C.” Journal of Institutional and Theoretical Economics 155: 458–86. Wijkander (eds) (1992) Contract Economics. and Financial Structure.A. R. K. L. Contracts. Grandori. Ghoshal. Chicago: University of Chicago Press.” Industrial Relations 35: 299–333. (1997) “Governance structures.A. R. Oxford: Oxford University Press. (1994b) “The theory of the firm: The Austrians as precursors and critics of contemporary theory. Jensen. and Strauss. Legislation. Gittleman. von (1945) “The use of knowledge in society. M.I. Chicago: University of Chicago Press. and Joyce. Harryson. Holmström. Organizational Chnage and Performance Improvement. London: Routledge. (1998) “‘Flexible’ workplace practices: evidence from a nationally representative survey. coordination mechanisms and cognitive models. Klein (eds) (2001) Entrepreneurship and the Firm: Austrian Perspectives on Economic Organization. London: Routledge. (2000) “Pragmatic collaborations: Advancing knowledge while controlling opportunism. Ichniowski. Henderson.C.. B. (1992) “Specific and general knowledge and organizational structure. (2000) Compensation Management in a Knowledge-Based World. von (1973) Law. G.J.” Industrial and Labor Relations 52: 99–115. J. Chicago: University of Chicago Press. N. and Meckling W. Foss Foss. (1979) “Moral hazard and observability. Hayek. not structural hierarchy. (1994) “Revolution at Oticon A/S: The spaghetti organization.J. (2000) “Internal disaggregation in Oticon: An organizational economics interpretation of the rise and decline of the Oticon spaghetti organization.” in F.J. von (1946) “The meaning of competition. Oxford: Blackwell. Foss and P. Kochan. F. Hayek. Hayek.M. C..

W. Ankerhus.J. Lyregaard. C.A. von (1944) Bureaucracy. and Profit. H. (1990) “Think the unthinkable.. (1993) “Oticon: Erfaringer og faldgruber. Cambridge: NBER working paper no. G.” in M. Indianapolis: Liberty Press. (1992) Managerial Dilemmas. L. and Hill. P. (1995) Omstigning til Paradis? Oticon i processen fra hierarki til spaghetti. Zucker. Learning. dynamics. Copenhagen: Schultz. L. Alken (eds) På vej mod helhedssyn i ledelse. M. Knight.” Strategic Management Journal 21: 875–96. Uncertainty. Milgrom. Jr (1997) “Organizing in the knowledge age: Anticipating the cellular form. (1993) “Knowledge and internal organization. Copenhagen: Copenhagen Business School Press. J. Cambridge: Cambridge University Press. Mathews.” Academy of Management Review 23: 680–97. (2000) “Strategy as guided evolution.T.L. Lovas. L.” Industrial and Labor Relations Review 53: 179–96. strategy and organization. B. S. R. Hellerup: Oticon. San Francisco: Fox and Wilkes.H. Poulsen. Morsing and K. (1999) “Information age organizations. J.L. Matusik. (1998) “The utilization of contingent work. “Austrian” determinants of economic organization 167 management. Jensen (1998) Foundations of Organizational Strategy. Eiberg (eds) (1998) Managing the Unmanageable For a Decade. Mises. Milgrom. and Flexibility: Sourcing Scientific Knowledge in New Biotechnology Firms.” Journal of Economic Behavior and Organization 21: 17–30. Powell. and competitive advantage. and Klein. (1994) “The knowledge-based enterprise. Boston: Butterworth- Heinemann. and Coleman. and Ghoshal. P. and Rizzo. nor hierarchy: Network forms of organization.E. Kelley. Osterman. knowledge creation. (1921) Risk.” American Economic Review 80: 511–528.. Snow. L. Klein. Kolind.” in M.. H.” in M. and performance. Boston: Butterworth-Heinemann.P. S.H. (1993) Tænk det utænkelige – revolutionen i Oticon. M.” in S.P. (1995) Social Networks.B. Miles.F.” Journal of Political Economy 96: 42–60. L. P. (1985) The Economics of Time and Ignorance. (1990) “Neither market. Oliver.R.P. Hellerup: Oticon.. Myers. Kolind. von (1936) Socialism.C. and Brewer. W5320. Morsing. (1990) “The Economics of modern manufacturing: technology.” Research in Organizational Behavior 12: 295–336. Morsing and K. Klein. S. Mises. New Haven: Yale University Press. (1988) “Employment contracts. J.G. C. M. A. O’Driscoll. P.. G. P. Liebeskind. (1997) Knowledge in Organizations. influence activities and efficient organization design. and Pillai.W. L. P.. Oxford: Basil Blackwell. R. F. (1996) Knowledge Management and Organizational Design. . Hildebrandt and L. Minkler. and Roberts.” Review of Austrian Economics 9: 3–28. (1996) “Economic calculation and the limits of organization. A.S. P. Eiberg (eds) (1998) Managing the Unmanageable for a Decade. Mendelson. von (1949) Human Action. (2000) “Work organization in an era of restructuring: Trends in diffusion and effects on employee welfare. Paris: OECD.C. (2000) “Do entrepreneurs make predictable mistakes? Evidence from corporate divestitures. P. Cambridge. Mises. OECD (1999) Employment Outlook. Miller. New York: Augustus M. G. L.” unpublished paper. Prusak. 1965 reprint. Miles.” Academy of Management Executive 11: 7–20.. MA: Harvard University Press.” Journal of Economic Behavior and Organization 38: 253–81.-E.

. L. Foss Simon. (1985) The Economic Institutions of Capitalism. Oxford: Oxford University Press. (ed. Cambridge.A. (1999) “The learning economy and embodied knowledge flows in Great Britain. (2000) “In search of new foundations. Tomlinson. M. New York: Free Press.168 Nicolai J.” forthcoming in Journal of Finance. O. O.E.” Research in the Sociology of Organizations 8: 157–90. Zucker. Williamson.” Models of Bounded Rationality.) (1951) “A formal theory of the employment relationship. Zingales. MA: MIT Press. (1996) The Mechanisms of Governance. L.E. (1991) “Markets for bureaucratic authority and control: Information quality in professions and services.” Journal of Evolutionary Economics 9: 431–51. Williamson. H.

Our purpose here is to highlight some challenges to firms attributable to the emergence and expansion of the new information and communication technology (ICT) paradigm. The analysis of the information systems of firms and the study of the conditions for consistency between these systems and the coordination mechanisms implemented by firms should yield an understanding of the behavior of the firm in an environment characterized by a strong information content. from our point of view. the key challenge to the firm posed by the new economy: that is its ability to transform information into knowledge in order to innovate. the transformation pressure induced by the new economy seems to reward organizations that adapt and innovate. both internal and external. What is needed is an analytical framework where the focus is on how economic transformation results from a combination of external pressure. any study of the exploitation by the firm of a wider and wider range of information must include an analysis of the process by which information turns into knowledge in order to innovate. internal innovative capabilities and coordination mechanisms. Indeed. We argue that such challenges cannot be reduced to informational aspects. The first section reviews the main characteristics of the new economy. However. The second section evaluates theories of the firm according to their ability to provide instruments for the analysis at the firm level of the effects of this new ICT paradigm. As far as the behavior of the firm is concerned the question addressed by the new economy affects the kinds of coordination forms. We defend the idea that most theories . The aim of this chapter is to make use of the analytical framework provided by the Austrian analysis of production and knowledge (Dulbecco and Garrouste 1999) in order to propose a theory of the firm that may frame a study of such a transformation process and consequently offer an answer to the question of the coordination of economic activities in a knowledge-based economy. which may permit exploitation of information henceforward relatively both richer and less expensive in order to innovate.7 The new economy and the Austrian theory of the firm Philippe Dulbecco and Pierre Garrouste Introduction So-called “new economy” approaches highlight the role played by innovation and information in all economic activities. Indeed. such an analysis is unable to provide an answer to what represents.

contributing to the development of interfirm networks. Thus. which articulates the notions of information and capabilities. science has become more efficient. owing to a technological wave based on new applications such as the World Wide Web or the Navigator. by focusing mostly on the informational aspects of the new economy. and ICT diffusion. narrow. at most. and become more and more motivated to intensify their actions (Carter 1994): change gives rise to change. less than 10 percent of GDP and around 4 percent of employment. ICT first contributed to the rise in productivity. broad acceptance.170 Philippe Dulbecco and Pierre Garrouste of the firm. ICT has reduced externalization costs.2 Investments related to ICT have considerably increased these last years. by means of their diffusion in other industries. and more intertwined with enterprise. A new regime based on innovation and permanent change substitutes for the old one. By following a very short technological cycle. Lundvall and Nielsen 1999). the latter being characterized by a short construction period of new capacities articulated with a relative long exploitation period. the evolution is particularly obvious for activities related to the processing. Those phenomena combine to make change and innovation the most important elements of any economic activity. a new productive pattern. Indeed. changes initially induced by the use of ICT are amplified by the actions of agents who benefit from previous changes. storage and exchange of information (Steinmuller 1999). if the first. it appears that because of ICT. ICT represents the main source of a kind of permanent upheaval (Foray 2000. broad. Finally. Moreover. the second. it appears more and more obvious today that technical innovations coming from the information and communication industries1 have generated. if not the main economic activity itself (Carter 1994. The cumulative character of the introduction of ICT inside the firm (Zuscovitch 1983). Such a theory offers a detailed analysis of coordination over time of the plans of the firms. What’s new for the firm in the new economy? The term “new economy” immediately points to a newly arising sector – the information and communication industries – and a new way of driving the whole economy. forces firms into a permanent process of adoption and adaptation (Benghozi and Cohendet 1999). A final section concludes. ICT has played a major role both in the acceleration of innovation processes and in the decrease of life-cycle durations. meaning of the new economy represented. are unable to develop the analytical categories required for appreciating the whole information exploitation process. have speeded up since the mid-1990s. Third. definition embraces all the using sectors and hence the whole economy (Cohen and Debonneuil 1998).3 Second. Foray and Lundvall 1996). The third section sets out what we agree to call an Austrian theory of the firm. combined with the rapid evolution of ICT. Recall that . The new economy as an economy in permanent change The consequences for firms of the development of the new economy relate mainly to this last.

it is then possible to identify the main disruptions now introduced by innovation and change into the economic process. it is not the case when innovation and change become the rules of the game. and mobility. The objectives followed by firms which chose to confront this challenge are today clearly identified (Askenazy 1999). that is. On the one hand. Innovation and knowledge in the new economy The new economy is often labeled a knowledge-based economy (OECD 2000). quality – that is. However. and the causality works both ways (Lundvall and Nielsen 1999). (3) destabilizes productive organizations and complicates the co- ordination issue. On the other hand. The new economy and the Austrian theory of the firm 171 in Europe more than 30 percent of manufacturing turnover comes from new products (OECD 2000). changes impose learning for all agents affected by those changes. . and (4) exacerbates uncertainty concerning the quality of products and consequently the information asymmetries of any transaction. flexibility. Langlois and Garrouste 1997. According to Foray (2000: 33). when firms have to adapt at any time. the necessity continuously to adapt to demand. The role played by knowledge in the new economy must be appreciated in light of the innovation constraints introduced by ICT. Permanent innovation then requires: (1) an increased level of training. Steinmuller 1999). Learning and change are indeed closely related (Antonelli 1999). the innovative regime explains the key role henceforward played by knowledge in society. the supplement of cost induced by innovation from current revenues (Dulbecco et al.4 a growing number of works have recently permitted a distinction between the two concepts. If economics has for a long time conceived of knowledge as information. 1995. Knowledge is thus conceived as a particular output – linked to a simple informational input – that may contribute to the creation of wealth (Dibiaggio 1999). The point is that these four elements all refer to the very notion of knowledge. (2) some specific competencies favoring adapt- ability. (2) declares most of the equipment obsolete and depreciates competencies. learning is an important and necessary input in the innovation process. It is often argued in this perspective that the last decade has been characterized by acceleration in both knowledge creation and knowledge destruction (Lundvall and Nielsen 1999).5 In this way knowledge refers mainly to the capacity for generating. whereas information remains a set of more or less formatted and structured data. in advance. Gaffard 1995). flexibility. the behavior of the firm is today fundamentally oriented toward objectives such as reactivity. and (4) the elaboration of complex coordination procedures covering all economic functions (Foray 2000). it is easy to demonstrate that firms can adapt by deducting. If cost reduction and the optimization of the production process represent major issues. extrapolating and inferring both new knowledge and information. Innovation: (1) cancels the technical links built in the past. unable to engender by itself new information (Foray 2000. (3) investments in systems that permit good access to information. When changes are infrequent and of small magnitude.

Knowledge and coordination in the new economy None of the results associated with the new economy would have been accom- plished without a profound transformation of the organization8 of the firm. have described changes as a “cultural and social-economical revolution” (Freeman 1987). but has also made skills and competencies obsolete. Brousseau and Rallet (1999) divide this last causation process into three phases. The study of the relationship between the development of ICT and the organization of firms is not new. by the increase in the share of intangible assets in the real capital stock (Kendrick 1994). the issue becomes understanding how firms. characterized by their specific capital asset base. by the continuous expansion of knowledge industries7 (OECD 1998). In other words. This last movement is finally limited again by firm organization and environment. first. process and organization. In this way the organizational component is going to play a major role. ICT is not confined to communication processes. Today. and third. Indeed. It particularly constitutes the heart of the “new techno-economic paradigm” theory (Perez 1985). The ICT revolution is not only a technical revolution. however.172 Philippe Dulbecco and Pierre Garrouste ICT has made a lot of information more easily accessible to a lot of people. empha- sizing the profound transformations involved in large but also small firms. From a dynamic perspective. by the growth of highly qualified manpower in total employment (OECD 1996). Indeed. . changes in the techno-economic paradigm are based on combinations of radical innovations in product. Several commentators. as is now well known. but contributes to the whole production and decision system of the firm (Benghozi and Cohendet 1999). Knowledge is con- nected to information. it is also an organizational revolution: ICT diffusion requires an evolution of firms’ organization in order to implement and exploit new technologies and hence to benefit from these innovations. and knowledge is also connected to firms’ physical assets. The second is associated with changes induced by this introduction in organizations’ efficiency. The first relates to the introduction of ICT into the firm inasmuch as this introduction is fundamentally constrained by the prevailing organizational scheme. as capital constitutes the foundation for all innovative strategies (Amendola and Gaffard 1988). Changes induced by the development of ICT give rise to major changes in the organizational structure of firms if their potential is to be fully exploited. These analyses are based on implicit causality: from the technology to the organization. as information represents a principal input into knowledge. Such an important evolution is confirmed. Some other works that highlight the existence of a reverse causation refute this technological determinism: from the organization to the technology. may transform information into knowledge in order to innovate.9 In most cases. information and production.6 second. the importance given to knowledge may not lead firms to underestimate the role played both by information and production inside the new regime. innovation constraints contribute to the settlement of complex interactions between knowledge.

Between 1991 and 1999 the value of international mergers increased six-fold. diffuse or exploit knowledge and not only information becomes one of the main sources of competitive advantage. today it is more and more obvious that those interactions are more complex and cannot be reduced to simple causation. that is whose organization squares with that implicitly required by those technologies. For example. it is the whole coordination10 issue which is addressed by the devel- opment of the knowledge-based economy. but must be correlated with a wider knowledge perspective. transmission and utilization of different kinds of information (Stiglitz 1985. (OECD 2000) is it not necessary to conceive the firm as a “knowledge processor”? The new economy and the behavior of the firm conceived as a processor of information The economics of information is typically the part of our discipline that is centrally concerned with the emergence and development of ICT. Indeed. As innovation requires knowledge and knowledge results mainly from particular coordination mechanisms. in Austria 61 percent of firms creating new products are associated with one or more partners. This diffusion contributes to the development of a new economic paradigm based on innovation and continuous change. Indeed. because individuals operate under imperfect and asymmetric information. This complex co-evolution process confirms. the exploitation and creation of knowledge exhibit a rather collective dimension. the challenge posed by the new economy to the theory of the firm is to provide an analysis which articulates the informational and knowledge components of innovation along with the productive one.000 in 1989 to 7. and because it is costly for .000 in 1999. if needed. What is new for the firm in the new economy results from the very wide and intensive diffusion of ICT into the whole economic system. It is rare indeed that enterprises innovate in isolation. ICT is consequently efficient for firms capable of integrating it. ICT poses a new question: is it possible to analyze the impacts of ICT by using the idea that firms are only “information processors” (Cohendet and Llerena 1999)? If the capacity to create. while ICT simultaneously opens new solutions that transform the organization (Benghozy and Cohendet 1999). One may notice in this perspective a significant expansion of the numbers of both cooperative agreements and mergers during the 1990s. The idea is that the organization and ICT co-determine each other: the organization adapts ICT to a prevailing function. this ratio reaches 83 percent in Spain and 97 percent in Denmark (OECD 2000). In this context. The expansion of so-called “collaborative technologies” (Foray 2000) contributes to the emergence and development of new industrial organization patterns. 2000). that the challenge to the firm posed by the development and diffusion of ICT cannot be reduced to an informational one. Finally. the basic idea of the economics of information is to treat the central part of economic analysis as the study of how individuals deal with the acquisition. the latter being both internal and external to firms.12 These data must be linked again with the innovation constraint. The new economy and the Austrian theory of the firm 173 However.13 In fact.11 while the number of new cooperative agreements rose from 1.

174 Philippe Dulbecco and Pierre Garrouste them to acquire information. the main responses information- based theories of the firm give are the following. Our concern here is neither to present an exhaustive analysis of information based theories of the firm. What can be called information-based theories of the firm are also concerned with those kinds of problems inasmuch as they all more or less assume. the characteristics of those information-based assets (in terms of their specificity) strongly determine the existence and the boundaries of the firm. the model captures very nicely the essence of market competition: the right to exit the relationship and the information and incentives that the ensuing bargaining process generates. because information is not free”14 (Barzel 2001). and this feature is a central point of the Hart–Moore’s model. Our point is rather to deal with the following question: if the acquisition. individual as well as organizational behavior is conditioned by their ability to solve those problems. In fact. or any other forms of organization. needed? The answer is. Concerning the evolution of competition. Brynjolfsson and Hitt 2000). (Stiglitz 2000: 1444) It can be added that because assets are becoming more and more information based and intangible. basically. “Why are firms. what can the emergence of ICT imply for the way organizations function? Even if they do not all yield the same results. even if they do not consider it to be equally central. Indeed ICT lowers the cost to the buyer of acquiring information about seller prices as well as product offerings. transmission and utilization of information strongly conditions the existence and functioning of organizations. but it seems plausible that if information were perfect – if all contingencies could have been anticipated – all important contingencies (at least where there is the ability and desirability of transferring risk) would have been taken care of in this original contract. Stiglitz also assumes that the existence of incomplete markets and contracts is due to imperfections in knowledge. This kind of result is . according to Holmström. some authors show that there is a redistribution of rents from suppliers to buyers (Bakos 1997. nor to discuss further the relevance of its definition. including asymmetries of information. (Holmström 1999: 85) However. this information is costly and someone has to invest efforts into finding the information to reach an agreement. the firm can be conceived as a processor of information (Cohendet and Llerena 1999). In a different way. Consequently. bargaining generates information. ICT reduces the inefficiency inherent in search costs (Bakos and Brynjolfsson 1997). that the existence and the functioning of the firm is linked to some problem of information. and he adds that transactions costs provide the major alternative explanation for incomplete contracts. Indeed.

that is to say the ratio of the number of informed consumers to the sum of the number of informed and less-informed consumers is low and. The costs of external coordination are decreasing. It is. The tendency would be for those hybrid forms to disappear. this result is very controversial. many works (Black and Lynch 1996. “it is widely believed that IT [Information Technology] lowers the costs of inter-firm coordination” (Bakos and Brynjolfsson 1997: 3). it is more beneficial to cooperate than to merge. In fact. The problem is that it is also possible to show that if coordination costs are decreasing. Those two phenomena are due to the fact that. there is a tendency for firms to develop organizational innovations. if the redistribution of rents is really buyer oriented. problematic (see above). Greenan and Mairesse 1996) show that at the firm level productivity increases when ICT is introduced. the merger process cannot be explained. This technological determinism is. Even if the importance of this result is often debated. Third. unreasonable to assume that ICT originates outside the economic system and that its introduction inside the firm obliges this firm to adapt its organization structure. However. this kind of evolution is explained by the effect of the acceleration of the circulation of the information inside the firm (due to the ICT) on the whole organization of the firm. in fact. there is no tendency for competition to increase – quite the opposite. Such an evolution of coordination costs is attributable to the fact that “investments in IT are less likely to be relationship specific than other investments designed to reduce coordination costs between firms” (Bakos and Brynjolfsson 1997: 4). reducing the risks of opportunism and holdup. More problematically. such an evolution can explain both the tendency for new firms to emerge and for firms to develop cooperation. It seems more relevant to consider that there is some co-evolution of technology and organization. since the reduction in transaction costs (due to the decrease in asset specificity) normally provokes a de-integration process. As a consequence. Fourth. second. if coordination costs are decreasing and if the possibilities for opportunism to be effective are reduced. “A significant component of the value of IT is its ability to enable complementary organizational investments such as business processes and work practices” (Brynjolfsson and Hitt 2000: 4). In other words. if the cost of price search is high. the number of firms on the market has to increase and prices decrease. on the basis of different assumptions (constant returns to scale) Janssen and Moraga (2000) show that the evolution of prices as well as their dispersion is strongly sensitive to the maturity of markets (the number of informed consumers relative to less-informed ones) and to the cost of price search. the productivity of firms implementing ICT is increasing. In fact. first. then the introduction of ICT places firms in a more competitive environment. In fact. This kind of evolution is interesting because it opens new possibilities in terms of de-integration and partnership. it is less necessary for firms to cooperate because opportunism as well as asset specificity are decreasing. The new economy and the Austrian theory of the firm 175 important inasmuch as the evolution of competition depends on the possibility for consumers to increase their own market power. In fact. if the market is not mature. As a matter of fact. however. One of the theoretical bases for this result is basically that internal coordination costs are decreasing .

the productive one and the learning one. in a different way. its cost is decreasing. to consider that information asymmetries are decreasing and that the efficiency of markets and organizations is increasing. as we have seen. This process makes knowledge less specific and endows it with all the characteristics of information. Another consequence is that the process of generating knowledge is part of the whole production activity. if it is one thing to recognize the necessity of this articulation. One consequence of such a distinction is that the transfer. The quantity of information circulating in the firm is thus increasing. in particular. It then becomes possible to explain. Information and investment in the new economy As we have previously explained. production and knowledge. and appropriation of knowledge cannot be reduced to simple exchanges of informational signals (Johnson and Lundvall 1992). It is precisely the aim of the remainder of this chapter to propose an analytical framework dedicated to this objective.176 Philippe Dulbecco and Pierre Garrouste because information is at the root of communication inside the firm and conse- quently. However. since information is unambiguous. But. the distinction between information and knowledge refers to the difference between a flow of data and a structure that organizes the data and reorganizes itself. ICT makes information less firm-specific and reduces imperfections in the distribution of information. at the firm level. in light of the underlying assumptions. the possibility for firms to develop partnerships. The point is that knowledge is here conceived as not being different in its very nature from information. it is difficult. In other words. it becomes unambiguous and non-rival. it is quite another to have available the analytical structures that yield an understanding of the complex relationship existing between information. Lundvall (1998) shows. and finally increased competition on the market. that learning in connection with production is fundamental for success in process and product innovation. if not impossible. In this way it is natural. some important things. the two dimensions of the innovative process. this communication is less costly. those explanations do not seem able to capture all the evolution of coordination inside and outside firms. Our aim is now to show that an Austrian perspective of the firm is able to propose a complementary and interesting framework. All those results are based on the following idea: even if information is not a public good à la Arrow (1962). diffusion. or both. Even if knowledge is distinguished from information. the idea is that the “natural” evolution of knowledge is to become less informal and then more codified. to identify the specificity of the knowledge base of any firm without taking into account its production structure. the productivity of firms to increase. Any analysis of the behavior of the innovative firm in the new economy environment must consequently articulate. for example. such as the tendency for coordination costs to decrease. .

and which allows it to undertake a number of activities. by defining its orientations and its content. The point is then to consider that any investment decision implemented by any enterprise that chooses to modify its production process is not only constrained by the composition of the stock of capital that prevails in the economy. although the problem of coordination depends mainly on the latter (Dulbecco and Garrouste 1999). The firm is then quite logically understood.17 In Richardson’s words. Such training is then oriented and supervised by these capabilities. within the production process. the properties of knowledge and capabilities of the firm enable collective training and on the other hand they limit its very scope. production and knowledge: lessons from the Austrian theory of the firm We argue elsewhere that the generic object of the Austrian theory of the firm consists in studying the nature of the relationship between the structure of capital and the structure of knowledge in an ever-changing world (Dulbecco and Garrouste 1999). defined as factors of transferable inputs. The capabilities of any enterprise thus result from the configuration. must take into account the fact that this investment is at the same time complementary to and in competition with other investments that have been. are. at any point of time. capabilities arise from investment decisions taken by the firm. thus become the support for collective routines and serve as the basis for a training process. This results in a better adjustment of the firm to its environment. be it in quantitative or qualitative terms. However. The great achievement of the capabilities-based approach is hence to represent “a real-time account of production costs in which knowledge and organization have as important a role as technology” (Langlois and Robertson 1995). These capabilities. itself collective. In this framework. The new economy and the Austrian theory of the firm 177 Information. but as the place where the capabilities are built and modified.16 Unlike resources. On the one hand. of its own structure of production. but it will also affect the whole complex network of relations that connects the firms to each other (Richardson 1990). and thus likely to be freely obtainable on markets. the concept of capabilities is able to give an accounting for elements of specific subjects as diverse as the configuration of productive activities or the constitution of expectations and the transmission of knowledge. viewed as a knowledge and a know-how that are more or less incor- porated into the equipment. it is important here to notice that. Any understanding of the role played by capabilities in solving the coordination issue must consequently integrate an analysis of the investment decision. this means that any firm that modifies its capacity of production. knowledge and experience available to each enterprise at any time. but at the same time creates difficulties in facing radically new changes in this very environment. at any moment in time. capabilities represent firm-specific tangible or intangible assets because they are created over time through the combination. of resources available to the firm.18 Hence the challenge is to determine . not only as the place where factors of production are combined. or will be made within the economy by other enterprises. The idea of capabilities was introduced by Richardson (1972)15 to designate the set of abilities.

It is because the production processes are subject to complementarities over time that the question of knowledge becomes so important. (Lachmann 1978: 3) At this point.20 and on the other hand. on one hand. on the contrary. Naturally. Indeed. but rather as a structure (Langlois and Garrouste 1997). The second reason is that the mere ability to capture information in a situation of uncertainty does not solve the problem faced by the firm. which refers to the problem of being able to capture infor- mation. and differ from one market to another. in a subjectivist way. it is necessary. it is possible to show that.178 Philippe Dulbecco and Pierre Garrouste what Lachmann calls “the best mode of complementarity” (Lachmann 1978: 6). and when the economy is subject to a large number of simultaneous changes. the question of the coordination of investment plans takes an informational turn. the “best” mode of complementarity is not a “datum. depending on whether one is interested in the competitive or the complementary aspect of investments.e. Even where he succeeds quickly he will not enjoy his achievement for long. which renders the ability to process information essential. not as a receptacle. In the first case. i. as it has just been formulated. so that the latter is of relatively little use to underpin action. Because production . So. has a true meaning only when referring to a diachronic analysis of production: delays in transmitting and processing information are accounted for in conjunction with delays in construction. When the transmission delays are too long. or more exactly. Firms are basically unaware of the evolution of their own production process. it is important here to stress how much the problem of knowledge. not to limit this distinction to an opposition between a stock and a flow.19 But the transmission of both types of information “is often delayed and sometimes faulty” (Lachmann 1978: 22). because the latter itself depends on the evolu- tion of those implemented by others facing the same difficulties.” It is by no way “given” to the entrepreneur who. the aim is to obtain market information. to distinguish between knowledge and information. informational and cognitive. information relative to activities planned by the others within the system – mainly consumers and competitors – whereas in the second case the point is to acquire technical information relative to the feasibility of projects implemented by enterprises linked together through technological complementarities (Richardson 1990). the type of information looked for and exchanged is different. Knowledge is understood. Besides. the optimal combination that enables us to reconcile the different plans implemented by enterprises with interrelated activities. First. the objective becomes to bring out the modes of access to information that are likely to improve the entrepreneur’s expectations. as sooner or later circumstances will begin to change again.e. it becomes difficult to reconstitute the chronology of events that gave rise to the obtained information. and this for two main reasons. i. as a rule has to spend a good deal of time and effort in finding out what it is.

Indeed. innovation is not an instan- taneous process. success depends directly on its ability to access an informational advantage in its field of production with a view to rendering its production plans consistent vis-à-vis those of other firms over time. a delay in the transmission of the information to the set of actors involved in implementation of the new production process (Langlois and Robertson 1995. Competitive uncertainty deals with relations between the firm and its market environment. It is due. its capabilities. Malmgren 1961).e. that is to the issue of the interaction between information. to the inability of the firm to forecast the evolution of supply and demand on its new markets and. Second. and because these delays are specific to each firm. the harmonization of plans assumes such a complex form: the time of production generates ignorance. The second element refers to a fundamentally temporal conception of innovation and change. i. Uncertainty is two-sided: it is first technical/technological. impose high dynamic transaction costs during the period of development of the new activity. The first element which must be taken into account when studying the behavior of any firm is that it more or less deliberately confronts economic uncertainty by implementing investments that deal with previously unknown options. a delay in acquisition and/or development of the abilities required by the new activity. The objective of any enterprise thus becomes to develop its technological and informational abilities. The new economy and the Austrian theory of the firm 179 is subject to delays.e. its consequences for the firm in terms of internal as well as external organization become more and more important. by the construction of the new capacity (Foss 1993). to its vulnerability during the period of development of the new activity relative to competing firms that have chosen a “passive” strategy of adapting to the environment (i. Third. which is delimited by the configuration of their respective routines). second. as we previously mentioned.e. The economic output of the firm quite naturally depends on its very ability to manage constraints imposed by these delays. first. The problem posed to the firm by the new economy is that the gap between the time available . production and knowledge in an ever-changing environment. Inasmuch as uncertainty is increasing in the new economy. First. But this uncertainty also concerns competition. The risk for the enterprise is that the continuity of construction and of use might be impeded or even arrested before the process enters the phase of use. i. as such.21 The firm’s main problem hence becomes to find complementary assets that permit it to organize a process of creation of new knowledge when the speed of renewal of those assets is very high. a delay in construction of the production capacity before it can be actually used for the production of a new output. The new economy and the organization of industry Let us return to the analysis of the behavior of the firm in the context of the new economy. rather it is subject to delays. Such factors are indeed generally non-contestable in the sense of Langlois (1993) and. in the sense that nothing guarantees ex ante that the firm will be in a position to develop and/or to acquire the technical and human factors required by the new activity. in order to face such ignorance.

maintained. knowledge which permits. its main drawback is the sacrifice of adaptability. The third element that we have retained as a main attribute of the behavior of the firm in the new economy context relates to the irreversible nature of its commitments – this is indeed the idiosyncratic character of the assets sought after by the firm. We mean by relevant knowledge. an analysis in terms of dynamic transaction costs (Langlois 1992). 1996). i.180 Philippe Dulbecco and Pierre Garrouste for capacities to be built up and the period during which those capacities generate useful output is becoming more important. the firm then becomes a place where knowledge is generated. The external resources integrated into the firm also acquire an idiosyncratic dimension since they partake of a collective and cumulative learning. whereas adaptability is required by uncertainty and the risk intrinsic in the involvement in innovative investments. replicated and modified (Cohendet et al. The merger wave that seems to go with the development of the new economy contributes to rendering this thesis attractive. integration is supposed to allow the development of . the choice of a coordination mode naturally takes on a crucial dimension. Indeed. and (3) the degree of instability of final demand. two a priori contradictory time horizons must here be reconciled: that of the operation or the optimal mobilization of existing resources. the coordination of investment over time and. (2) the learning capacities relative to firms and markets. on the one hand. in particular on the theory of the innovative firm (Lazonick 1992). The selected mode of coordination must indeed enable firms to transform information into relevant knowledge. But if specialization leads to comparative advantage and high profits. Arbitration between these different modes of coordination depends basically on: (1) the nature of investments brought into play. Managing this problem is essential inasmuch as there is a tendency in the new economy for investments in R&D to have decreasing returns (Foray 2000). on the other hand.23 Thus. Specialization hence carries irreversibility in so far as it calls for a specific articulation over time of the investments that are themselves specific. This point is crucial. 1996). and that of the exploration or the development of new production capacities (Cohendet et al. the firm is basically subject to what we agree to call the specialization–adaptability dilemma: specialization is represented by firm’s involvement in the construction of new and specific production processes carrying profit opportunities. This feature originates in the accumulation over time of specific assets dedicated to the construction of new production capacity. and the approach of the firm in terms of competence (Foss 1993). it alone crystallizes an important dimension of the problem faced by the firm in the new economy environment. access to information that is relevant and necessary to the development of specific production processes. Moreover. Indeed.22 In such a context. This phenomenon makes the manage- ment of the dilemma specificity/adaptability very difficult (see below).e. our consideration of knowledge leads to a wider definition of the organization of industry as a system for coordinating ever changing production processes. The integration/internalization thesis was recently the object of a number of exemplary studies. justification of revisions and redeployments in an economic world characterized by continuous change.

indeed requires that cumulative processes be initiated in which the organizational capabilities of the enterprise never stop improving. The problem here is to transform high fixed costs during the period of development into low unit costs when the new process enters the phase of utilization. It is the fact that new knowledge needs to be organized in a new way that drives the creation of new little firms. i. By providing tools for the management of technical and competitive uncertainty. cumulative. This phenomenon is due to the lack of adaptability of the big firm and its inability to react rapidly to new opportunities. tacit and idiosyncratic nature of the knowledge involved in innovative activities makes it impossible to resort to the market. In other words. The new economy and the Austrian theory of the firm 181 capabilities required by the new activity. in this way. As an example. is not only due. both of which are based on an internalization of resources and an increase in the production scale. first by developing the production scale and second by integrating suppliers and distributors to avoid any disruption in the production process and to take advantage of synchronization economies (Lazonick 1992). unless one is ready to bear prohibitive dynamic transaction costs. The main problem is that the price paid for integration is likely to be prohibitive for a firm that has chosen to fight against the innovative constraints of the new economy. This is partly due to the existence of a close relationship between the range of productive opportunities perceived by enterprises and the origin of investments undertaken: organizational capabilities are increas- ingly likely to allow for a number of options as their implementation comes from various sources. the fact that IBM was initially unable to develop the personal computer. to react to new opportunities. we are able to explain why competition is increasing in the new economy without making reference to the idea that if information is becoming less costly this . one can note that the type of integration proposed in this set of analyses assumes relatively high costs of capital redeployment (sunk costs) which thwart the firm’s potential for flexibility and adaptability all the more. One of the important consequences of the logic of integration is then that the enterprise becomes less “responsive” to its markets (Teece 1992). integration represents a mechanism of coordination of innovative investments over time. Indeed. New knowledge sometimes needs new methods of organizing it. This price is the price of the sacrifice of adaptability. to wrong expectations concerning the evolution of demand (such an explanation can be given by an entrepreneurship-based conception of the firm) but also to the diffi- culty reorganizing the production process. It then becomes possible to explain why the number of firms in the new economy is increasing. The collective. the firm runs a risk of inertia and a lack of organizational flexibility.24 Vertical integration may also help to achieve some kind of cost rationalization. reflecting a variety of opinions concerning coming threats and opportunities (Loasby 1996). the fact that the number of mergers is increasing. competencies and capabilities.e. By the way. we obtain one of the results concerning the boundaries of the firm. Moreover. integration secures the development of new production capacity and the transition from development to utilization. because of the internal accumulation of knowledge. and then the firm. The logic that prevails here. in view of the search for economic efficiency. Yet. and that Apple was.

But as we have already mentioned. for the firm. The emergence of new firms is also linked to new activities that arise and are undertaken by new firms (Bailey and Bakos 1997) before a possible process of merging takes place. . Second. The commonly accepted idea that information and knowledge have the same nature. and therefore new productive opportunities. can particularly be explained more satisfactorily in this Austrian theory. as a consequence of the diversity of investment sources (Loasby 1994) and of the permanent interaction between the production decisions of different firms grouped together within a cooperative process (Imaï and Baba 1994). The make or buy choice. and the development of hybrid forms of coordination (the partnership). a theory of the firm based on the idea that knowledge is not a stock of information and on the importance given to the production structure yields some interesting consequences. thanks to a permanent flow of technical and market information between cooperating firms. does not seem to be able to capture the most interesting characteristics of the new economy. firms. Conversely. when it is a coordinated behavior of partner-enterprises over time in view of common tasks and long-run objectives. especially when it comes to the management of sunk costs. which is at the basis of the information-based theory of the firm. The division of knowledge is necessary for the growth of knowledge (Loasby 1996). of ICT developments. because cooperation allows an efficient management of technological and competitive uncertainty. is also said to provide an advantageous solution to the problem of implementing innovative investments undertaken by firms in the new economy context (Warden 2001). Conclusion In this chapter we show that the new economy is modifying the organization of industries as well as of firms. It is also efficient in the sense that it creates information. First. and the coordination of the growth of knowledge and the coordination of the specialized activities that result from the division of knowledge both require the development of the firm’s external organization (Loasby 1993).182 Philippe Dulbecco and Pierre Garrouste reduces the possibility for opportunism and the need to integrate. It is the fact that new knowledge needs to be organized differently than old knowledge that produces the necessity to create new firms. because cooperation represents an efficient form of managing the specialization–adaptability dilemma (Dulbecco 1998). Cooperation agreements concluded between vertically related. it is efficient in the sense that it allows the coordination of specific capabilities without bringing the disadvantages of integration. the development of the new economy is also associated with the development of interfirm cooperative agreements. are one of the very possible embodiments of market connections implemented by firms. Cooperation between firms. but also competing. Such a process is reinforced in the new economy because in-house and ex-house knowledge needs to be more and more connected to be efficient and because relational activities are emerging or arising (Bailey and Bakos 1997).

in what follows. knowledge-based industries contributed. for example. 7 The contribution of these industries to GNP surpassed 50 percent for the OECD countries as a whole in 1985. 11 Rising from 85 to 538 billion dollars (OECD 2000). 17 The application. A survey of the numerous recent studies on this subject is provided by Askenazy (1999). 8 Including.” 4 Except for Machlup’s work (1984). that obtaining information can be costly. or which oblige these units to modify their plans over time” (Ménard 1990: 120). it is not sufficient to say that information is specific. By the way. Greenan and Mairesse (1999). innovative sector. Caroli and Van Reenen (1999). Then the specificity of information is the consequence of the existence of some path dependency that makes the process non-ergodic. One needs to explain this phenomenon. 15 The author says he drew his inspiration from Penrose (1959). See. without cost. The new economy and the Austrian theory of the firm 183 Notes Thanks to Cristiano Antonelli. Jack Birner and Nicolai Foss for helpful comments and to the International Center for Economic Research (Torino) for its financial support. and is now dominant. has had profound implications for the wisdom inherited from the past and has provided explanations of economic and social phenomena that otherwise would be hard to understand” (Stiglitz 2000: 1441). 5 Hayek is one of the first who showed that information is not given by the external world and that the mind is organizing and reorganizing information in some specific ways depending on the whole history of the individual. 12 In 1999 the increase of the number of cooperative agreements was around 40 percent (OECD 2000). Usual caveats apply. and the difference between information and knowledge is a possible means to accomplish that. 3 Even if this evolution has been particularly slow. 2 One sometimes speaks of a “Third Industrial Revolution” (Cohen and Debonneuil 1998). training. 16 In Richardson. 6 Kendrick’s work established that the intangible capital stock (education. giving rise to the famous “productivity paradox. 14 That is. 1 The ICT sector is one of the primary.” The specificity (and the character idiosyncraticity of data) of information is then the consequence of its embeddedness in specific theories. Osterman (2000). 18 Investments are said to be competitive when “the profitability of investment made by . 10 Following Ménard we mean by coordination “the procedures which render compatible the plans decided by elementary economic units. of Richardson’s analytical categories is entirely our own. 35 percent of firms’ total value-added. the implementation of new work practices. R&D. and that the extent of information asymmetries is affected by actions of firms and individuals. for example. ICT represented 31 percent of the incremental patents issued in the United States between 1992 and 1999. in 1997. and knowledge is at the very origin of this process. activities refer to different stages of an elementary production process. OECD (1999). We can find something similar in Feyerabend and Lakatos who show that information (facts or data) is produced by theories and is thus “theory-laden. 13 “The recognition that information is imperfect. that there are important asymmetries of information. of course. a quarter of the US enterprises consulted have implemented some organizational transformations in order to adapt to the changes induced by the Internet (OECD 2000). if not the primary. 9 According to a recent study. health) became equal to the capital stock in 1973.

Black. (1999). C.” Management Science 43: 12. P. L.184 Philippe Dulbecco and Pierre Garrouste one producer will be reduced by the implementation of the investment of others” (Richardson 1990: 3). 24 Moreover. organizational transformation and business performance. J. if taken by itself” (ibid. relatively few have considered carefully the problem of knowledge. and Van Reenen. and Gaffard. (2000) “Beyond computation: Information technology. Paris. and Cohendet.B. (1999) “Innovations technologiques et organizationnelles.: 7). (2001) “The role of measurement.Y. A. often dearly bought. (1996) How to Compete: The Impact of Workplace Practice and IT on Productivity. (1999) “The evolution of industrial organization of the production of knowledge. but also their physical counterpart. Bailey. April. . E. (1999) “Synthèse des travaux du groupe. and Rallet. An Economic Analysis of the Dynamics of Technology. Cepremap 9917. (1997) “An exploratory study of emerging role of electronic intermediaries” International Journal of Electronic Commerce 1(3): 7–20. persuasion. Antonelli. Brynjolfsson. Oxford: Basil Blackwell. (1988) The Innovative Choice. (1999) “L’organization de la production et de la décision face aux TIC. “Skill biased organizational change? Evidence from a panel of British and French establishments.F. M. E. organization et performances économiques. Malinvested capital is still capital that can be adapted to other uses” (Lachmann 1978: 25). exceeds the sum of profits to be obtained from each of them. Commissariat Général du Plan. internationalisation et inégalités. EHESS. and Lynch. Commissariat Général du Plan. Organization et Performances Économiques.Y.” mimeo. J. and Bakos. 20 “Many economists have professed to analyse information. P. May. E.” in E. Caroli.” in E. Rallet (eds) Technologies de l’information. 22 The formulation of this dilemma was inspired by Richardson’s work (1990). Y. J.” Couverture Orange. Kemerer (ed. MA: Harvard University Press. J. training and coordination costs that arise from resorting to an external supplier. May. E. Brousseau and A. but to provide resources. (1997) “Reducing buyer search costs: Implications for electronic marketplaces. when taken simultaneously. Dordrecht: Kluwer Academic Publishers.E. and Brynjolfsson. of past mistakes (our own and others) to learn from. Brousseau.-L.” unpublished thesis. (1997) “Organizational partnerships and the virtual corporation. Bakos. Askenazy.M. References Amendola. P. Y. and Hitt. guarantee capital and enforcement in the formation of firms and other organizations. Hayek is pre- eminent” (Loasby 1986: 38). Among those who have. 19 This distinction between two types of information is used again by Malmgrem through the concepts of “controlled information” and “secondary information” (1961: 408). the delays and costs of information transmission are minimized. In revising our expectations we not only have the knowledge.” mimeo. P.M. L. Rallet (eds) Technologies de l’Information. Investments are said to be complementary when “their combined profitability. 21 A non-contestable factor is a factor that one cannot obtain on a market unless one is ready to spend more than the cost of its internal development.” Cambridge Journal of Economics 23. S. Dynamic transaction costs cover the prospecting. Bakos. malinvested capital. Brousseau and A.” in C. Barzel.-J.) Information Technology and Industrial Competitiveness: How Information Technology Shapes Competition. Cambridge. 23 Lachmann explains in a similar perspective that: “The results of past mistakes are there not merely to provide lessons. Benghozi.

-L. (1999) “Towards an Austrian theory of the firm.” Industrial and Corporate Change 1: 1. L. and Garrouste. (1999) “Apprentissage. La Découverte. K. L. and Debonneuil.W. Arena and C. Greenan. (1994) “The evolution of modern economic accounts: review article. J. Dibiaggio. B. B. Paris. and Lundvall. D. The new economy and the Austrian theory of the firm 185 Carter. redundancy and learning in organizations. R. (1987) Technology Policy and Economic Performance.-A. P. Cohendet. Freeman. M.A.” Rapport de recherche 2660 de l’Institut National de Recherche en Informatique et en Automatique. B. (1993) “Theories of the firm: Contractual and competence perspectives. Kansas City: Sheed Andrews and McNeil. P. (1996) “Learning and organizational structure in evolutionary models of the firm. N. Dulbecco. Foss N. R. (1994).” in R. (1997) “Cognition. A. (1999) “Organizational change in French manufacturing: What do we learn from firm representative and from their employees ?. Paris: Repères’.” Revue d’Economie Industrielle 88(2). Rapport du Conseil d’Analyse Economique 28. et al. Johnson. Cohen. (1998) “Inter-firms cooperative agreements. Department of Economics.” mimeo. Langlois. (1989) “Systemic innovation and cross-border networks. P. Janssen. Economics and Organization 15: 74–102. D. Université Louis Pasteur. La documentation française.” working paper 333. .” mimeo.N. (1978) Capital and its Structure.N. Holmström.A. Langlois. Lundvall (eds) Employment and Growth in the Knowledge-based Economy. (1995) “De la substitution à la complémentarité: propositions pour un réexamen de la théorie de la firme et des marchés. Cohendet. (2000) “Pricing. (1992) “The learning economy. Langlois.L. Strasbourg. The University of Connecticut. J. (1996) “The knowledge-based economy: From the economics of knowledge to the learning economy. développements récents. coordination et organization de l’industrie – une perspective cognitive.” Journal of Evolutionary Economics 3. and Mairesse. P. (1993) “Capabilities and coherence in firms and markets.” in D. P. consumer search and the size of internet markets. Kendrick. Foray and B. C. and Lundvall. P. (1999) “La conception de la firme comme processus de connaissance.” Review of Income and Wealth 40(4): 457–59. D. Llerena. P. (1995) “Un modèle dynamique de comportement d’une entreprise. R.J.” Economics of Innovation and New Technology 4. and Moraga. Foray. OECD. and Marengo. (1999) “The firm as a subeconomy. New York: Springer. B. “Change as economic activity. Imaï. L. numéro exceptionnel: Economie industrielle.” Revue d’Economie Industrielle 88(2). J.” Review of Austrian Economics 12.” mimeo. Lachmann. Department of Economics. P. and Baba.” Revue d’Economie Industrielle. Lessons from Japan. la Technologie et la Croissance Economique. (1992) “Transaction-cost economics in real time. J.” Journal of Industry Studies 1(2). P.” Journal of Law. September. Paris: OECD. Longhi (eds) Markets and Organization. London and New York: Pinter Publishers. and Llerena. and Garrouste. Brandeis University. Dulbecco. M. (2000) L’économie de la connaissance.” NBER Working Paper 7285. Foray. (1998) Nouvelle économie. Gaffard. Y.. BETA. Dulbecco.P. 5–8 June.” paper presented at the Séminaire International sur la Science.

” Quarterly Journal of Economics 75. B. . Markets and Economic Change. vol.-A. (2000) “The contributions of the economics of information of twentieth century economics.” in N.E. E.” Journal of Economic Behavior and Organization 18.” The Economic Journal. Paris. (1994) “Organizational capabilities and interfirm relations. Stiglitz.J. expectations and the theory of the firm. (2001) “Organizational form. Teece. and innovation. Richardson. E.” Metroeconomica 45: 3. long waves and world structural change: New perspectives of developing countries. “The organization of industry. and Nielsen.” Economic Journal 95: 21–41. (1985) “Micro-electronics. Steinmuller. 2.J. (1999) “The many shapes of knowledge. Loasby B.-B. G. Foss and C.” in G. Loasby. (1999) Net-worked Knowledge and Knowledge-based Economics. OECD (1999) L’Économie Fondée sur le Savoir: Des Faits et Des Chiffres. and Robertson. Perez. (eds) Technical Change and Economic Theory. Oxford: Oxford University Press. Stiglitz. W. Paris.” mimeo. Cheltenham: Edward Elgar. (1983) “Informatisation. H. P. Osterman. Dosi et al. 2nd edn. Technologie et Innovation dans la Nouvelle Économie. La Découverte. Cambridge: Cambridge University Press.E.” Revue d’Economie Industrielle 88(2). J. Knudsen (eds) Towards a Competence Theory of the Firm. (1988) “Innovation as an interactive process – from user-producer interaction to the national system of innovation.” Revue d’Economie Industrielle 88(2). London: Routledge.186 Philippe Dulbecco and Pierre Garrouste Langlois. Productivity and Job Creation. Oxford: Clarendon Press. F. D. Distribution and Economic Significance. L’observateur OECD.” Quarterly Journal of Economics 1441–78. P.J. Loasby. incentives and the management of information technology: Opening the black box of outsourcing. P. Lundvall. cooperation. Zuscovitch. Carlson School of Management. Analytical report. Simon H. Paris. C. (1992) “Competition. Lazonick. (1995) Firms. October.E. (1984) Knowledge. flexibilité et division du travail.-J. B. OECD (2000). G.” World Development 13(3). (1985) “Information and economic analysis: A perspective.-A. (1959) The Theory of the Growth of the Firm. B. September. Lundvall.J. Warden. Machlup. vol. III. OECD (1996) Technology. Malmgrem. Princeton. J.” Industrial and Labour Relations Review 47(2). Paris: Repères. Synthèses. Delft: Telematica Instituut. Richardson. (1986) The Mind and Method of the Economists. London: Pinter Publishers. (1972). (1990) L’économie des Organizations. E. (1961) “Information. (1992) The Innovative Business Organization and the Myth of the Market Economy.A. (1996) “The organization of industry. Ménard C. a Dynamic Theory of Business Instituions. (1990) Information and Investment. Penrose. (1999) “Competition and transformation in the learning economy – illustrated by the Danish case. NJ: Princeton University Press. Science. January.” Revue d’Economie Industrielle 25. Paris. (2000) “Work reorganization in an era of restructuring: Trends in diffusion and effects on employee welfare. London: Routledge. de la Technologie et de l’Industrie. OECD (1998) Perspectives de la Science.-B. R. its Creation. B. W.

Part IV Networks and communication .


this approach is assuming institutions away (Birner 1999). because investments are low. on the contrary. due to insufficient effective demand (Keynes 1936). the existence of structures is assumed away. organizational arrangements that gather command lines according to functions (e. • Chandler’s discovery of the organizational shift from multifunctional to multidivisional form that took place in some large American companies in the 1930s and subsequently diffused over most large companies in the world (Chandler 1962). Modes of interaction. In systems theoretic terms. and that a few economists who pointed to concrete cases actually highlighted some of the most important issues ever raised in this discipline: • Keynes’ discovery of the possibility of underemployment equilibria. because demand is low. all this is neglected by a fantastic jump from the microeconomics of an isolated utility maximizer to the macroeconomics of a single representative utility maximizer. because unemployment is high. According to Keynes. In this way. Interactions are thought to take place in a homogeneous space where different agents are uniformly distributed. historical accidents that shaped the habits of peoples along centuries of agreements and quarrels. am claiming that structures do matter for the generation of collective behaviour. if not a single one. theorizations and models produced within economics generally assume that any agent may interact with any other. According to Chandler. organizational arrangements . equilibrium involuntary unemployment arises when investments are low.g. Rather. and so on.8 The small world of business relationships Guido Fioretti Introduction Curiously. a loop where information can circulate forever causing a series of economic agents endlessly to repeat the same sequence of actions. this chicken-and-egg situation can be char- acterized as an information circuit. I. production technologies) are not viable for large companies that produce a number of differentiated goods for a number of different markets. Ultimately. organisms for collective governance. so in the end the outcome of interactions can be averaged and complex economies populated by multitudes of different agents can be subsumed by the behaviour of a few representative individuals.

neuronal cells. structures that aggregate information for particular decision-makers. 2000). • Axelrod’s discovery of the feasibility of islands of cooperation in a sea of competition. Along years of observation. The present contribution suggests the possibility that a widespread structure of interactions among the components of distributed systems (including human societies. Just like in the case of Keynes’ ‘effective demand’. Austrian economics distinguishes itself for highlighting the need of taking account of communication. In organizational terms. the Marseille fish market displayed a certain degree of stability of customers around vendors. 1945). Among theoretical viewpoints. as well as of its effects on the firm as a whole. . Friedrich von Hayek has been a forerunner in this field.g. i. Chandler was discovering information sinks. a very important and very ubiquitous kind of information structure. The chapter is organized as follows. The fourth section proposes methodologies for highlighting a small-world topology in the structure of business relationships and speculates about the event that this structure is actually found and the final section concludes. Among the many aspects of Axelrod’s research. the third section highlights the conditions that make a small-world topology arise in a distributed system. American psychologist Stanley Milgram discovered the surprising ability by distant people to connect to one another (Milgram 1967). In structural terms. if the prisoner’s dilemma can be repeated by a population of individuals (Axelrod 1984). namely the small world topology.190 Guido Fioretti that gather command lines according to divisions (e.e. In other words. who eventually may be unaware of being part of a sequence that endlessly repeats itself. • Kirman’s investigation of Marseille’s wholesale fish market. Subsequently. which is possibly the most detailed investigation of transactions in a non-financial market (Weisbuch et al. I would like to draw attention to the fact that he was highlighting information clusters. knowledge formation and cognition. we could say that vendors act as information stars in the net of relationships that take place. • Nelson and Winter’s work on routines. vendors are nodes where infor- mation converges and from where it radiates. A small world During the 1960s. regulates business relationships as well. The next section explains the basics of small world structures. information structures. mutation and selection (Nelson and Winter 1982). Routines can be seen as sequences of operations carried out by individual workers. sectorally or geo- graphically distinct markets) are more apt to channel relevant information to management. modes of behaviour that are peculiar to specific firms and that eventually reproduce and propagate in the process of business replication. writing a treatise that anticipated modern connectionist models (Hayek 1952) and calling for consideration of information flows in economic theory (Hayek 1937. the internet and many others). we are dealing with an information circuit.

Nodes are arranged in tightly connected clusters. highlighted the same clustering feature around seven. weak ties are generally responsible for important breakthroughs in individuals’ lives. In this way. . because they describe friends that are all acquainted with one another) are linked by a few long-distance connections (weak ties. weak ties can easily explain the occurrence of the small world phenomenon. he can ask people in his clique whether any of them has it.1 illustrates an example of a small-world topology. The small world of business relationships 191 Milgram assigned individuals living in Kansas and Nebraska the task of making an envelope arrive at an individual located in Massachusetts with whom they were not acquainted. information concerning the person who was receiving the envelope and shipping it forth had to be inserted in the envelope itself. by means of personal contacts. In fact. At each passage. namely six. rather than six (Korte and Milgram 1970). Since this human ability of finding connections with distant people is reminiscent of anecdotes of people finding unexpected common acquaintances. Milgram could track the geographical and social milieus that were crossed. Notably. since radical change often originates from unexpected connections. the net of human acquaintances is such that clusters of localized connections (strong ties. that are linked to one another by a few long-distance edges. Rather. To his surprise. it states that the net of human acquaintances entails long-distance shortcuts. irrespective of geographical and social distances. ‘small’ world does not mean that connections between any two individ- uals have the same length. Clearly. completed chains clustered around quite a small number of steps. Later investigations. Figure 8. According to Granovetter. because they originate from occasional acquaintances) that bridge between cliques. although any single individual is unlikely to have the proper long-distance connection in order to reach any other individual by means of a single jump. Milgram called it the ‘small world’ phenomenon. aimed to evaluate the effect of changing physical distance and racial groups. Figure 8. More specifically. Sociologist Mark Granovetter (1973) was the first to understand the topology underlying the small-world phenomenon.1 A graph connected according to a small-world topology.

before examining the possibility that small-world structures regulate economic life it is sensible to ask on what conditions small-world topologies arise. • Clustering coefficient C. Since market economies are precisely like that. Watts 1999. However. defined as the average number of edges that must be traversed in the shortest path between any two pairs of vertices. networks whose connections have been drawn at random) will exhibit a low clustering coefficient and a low characteristic path length. mainly by physicists (Marchiori and Latora 2000. small-world networks will exhibit high clustering coefficients and low characteristic path length. simultaneous occurrence of low L and high C identifies a small-world topology. one may expect that business relationships are organized according to the same principle as well. Interestingly. small worlds seem to be ubiquitous in distributed systems. defined as the average of the ratios of actual immediate neighbours to maximum immediate neighbours for each vertex. Newman 2000. In fact. These are. However. In its simplest formulation. the Massachusetts Bay underground transportation system. Small-world topologies have been found in the collaboration graph of feature films extracted by the Internet Movie Database (where links have been defined as actors working in the same movie). Watts and Strogatz 1998). autonomous and (to some extent) intelligent agents. Watts and Strogatz 1998). On the contrary. small-world topologies seem to be a general property of the structure of connections between a large number of interacting. The intuition behind the above magnitudes is that characteristic path length measures the ability of a node to link to a distant one. elegans. researchers are looking for small-world structures in the most diverse settings. clustered networks having only local connections will exhibit a high clustering coefficient and a high characteristic path length. random networks (i. An instance of bounded rationality Small-world topologies are so widespread because they arise for simple reasons. Strogatz 2001. The above pair of magnitudes characterizes small-world networks. the English language (where links have been defined as co-occurrence of words) and the structure of hyperlinks connecting Internet sites (Adamic 1999. whereas clustering coefficient measures the amount of local structure in the network.e. Ferrer i Cancho and Solé 2001. the Western Power Grid of the US. the neural network of the nematode worm C.192 Guido Fioretti Formalization of the notion of small world was not provided until recently. and whether these conditions are likely to hold within economic systems. Thus. Since tools for identifying small-world networks are available. Marchiori and Latora 2000. essentially: (1) a generic tendency for each node to establish connections . it relies on two magnitudes: • Characteristic path length L. Thus.

Possibly. rather than optimizing behaviour. However. that is balanced by (2) a constraint on the number of connections that can be entertained. who carried out extensive analyses of circles of acquaintances. Models of bounded rationality assume satisficing. meaning that friends of our friends are likely to be our friends as well. the problem lies in separating stable acquaintances from occasional ones. the relevant variable was actually the ratio of neocortex volume to total brain volume. Derek De Solla Price.2. Since Simon’s pioneering work on bounded rationality (Simon 1982). who inspired the creation of the Science Citation Index. unpublished letter). if optimizing behaviour is simple but false while satisficing behaviour is correct but difficult. Correlation between neocortex ratio and group size turned out to be high. Point (3). In other words. rather than striving for the best possible performance. On the contrary. point (2) is more problematic and will be the subject of this section. 1997). this problem does not exist if we begin with primitive. simple societies. The idea is that decision- makers face cognitive limits to their information-processing abilities. which he called invisible colleges (De Solla Price 1965). satisficing behaviour is modelled by assuming that decision-makers are content to attain certain levels of performance. human groups must have counted 150 individuals. a limitation of models of bounded rationality is that no general rule is available to calculate the threshold where cognitive limits put a halt to optimization. models of bounded rationality do not supply a ready-made decision rule. deemed that scientists gather around informal groups of about 100 people. this correlation allowed an inference to be made on the size of prehistoric human groups. Apparently. whose localization arises out of (3) greater ease of establishing links with nodes that can already be reached through indirect paths. In order to avoid spurious correlation due to the need of larger animals to have larger brains just in order to control a larger number of muscles. Torsten Hägerstrand. a leading figure in cultural and economic geography. As a rule. According to Dunbar’s calculations. approximately. However. one should look for regularities in satisficing behaviour that would make it easy to employ. Most interestingly. It has long been suspected that cognitive limitations set an upper bound to human circles of acquaintances. evolutionary psychologist Robin Dunbar started to look for a correlation between neocortex size and group size for various species of primates (Dunbar 1996). Point (1) is an obvious feature of human societies. rather than sticking to wrong optimization assumptions. The small world of business relationships 193 with any other. . economists have a conceptual alternative to the idea that economic agents are able to make use of all information they get. is equally obvious (Granovetter 1973). whereas utility maximization does. attempted unsuccessfully to discover them by counting the number of references in commemoration books of prominent people (Hägerstrand 2001. and a valuable ability in the world of business (Burt 1992. as illustrated in Figure 8. By assuming that intelligence developed in order to keep groups of hominids together.

where people typically entertain relationships with many more than 150 fellows? If one counts the number of people with whom each individual interacts. However. Dunbar supported his findings with many examples taken from observation of actual human societies.194 Guido Fioretti MEAN GROUP SIZE 100 10 1 0 0. cares to keep in touch. In its turn this might . He found out that clans of contemporary primitive societies average almost exactly 150. and that these clans are much less variable in size than any other grouping. Hutterites live in groups whose mean size is a little over 100. the most interesting evidence collected by Dunbar concerns religious communities in North America. When Mormon leader Brigham Young led his followers out of Illinois into Utah. at any given point in time. Possibly. Furthermore. where individuals only interact with the members of the group to which they belong. he reports that archaeologists have suggested that the villages of the earliest farmers of the Middle East (5000 BC) typically numbered 150 people. with a more precise definition. one finds numbers that vary greatly according to profession and can be up to the order of the thousands (De Sola Pool and Kochen 1978). This is because they always split as soon as they reach a size of 150. he chose groups of 150 people as the ideal size.2 Dunbar’s finding of a correlation between neocortex ratio (the ratio of neocortex volume to total brain volume) and mean group size. In fact. it becomes increasingly difficult to control its members by peer pressure alone! Another example is provided by the Mormons. the elders claim that once a community exceeds 150 people. just like today’s horticultural villages in Indonesia. Dunbar’s anthropological constraint eventually holds only for a core of stable acquaintances. Thus. By courtesy of Robin Dunbar ©. How is it in modern societies. all these examples regard simple societies. only for the number of people with whom an individual.1 1 10 NEOCORTEX RATIO Figure 8. or. Latin America and the Philippines.

a research agenda could be set out. spreading of business relationships all over the globe is a salient feature of the ‘new economy’. Economists Franco Malerba and Francesco Lissoni. On the contrary. since physical distance may bear advantages of its own (Felsenstein 2001. eventually blurring any distinction between stable acquaintances and occasional ones. spreading over continental areas of free trade and eventually. and even quite different from one another. Furthermore. according to profession and inclinations. Thus. independence of information clusters from physical distance would be the hallmark of globalization. Kilkenny 2000). in the ‘old’ as well as in the ‘new’ economy. since one could speculate that in modern societies human relationships may take any degree of depth. it is interesting to speculate what would happen if business relationships really conform to a small-world topology. Namely. Thus. inventors form clusters of a nearly constant size of eighty people (Lissoni. The global network Admittedly. this figure is likely to be different for different kinds of people. personal communication). all over the globe. Rather. However. if bounded rationality means – inter alia – that businessmen cannot entertain relationships with infinitely many people. then this structure should be invariant with respect to technological paradigms and availability of information and commu- nication technologies. we should not expect that structural invariance is in any way related to physical location. One could reasonably think of gathering data on business relationships with respect to geographical location. it is not even necessary to resort to improving information technologies and falling transportation costs in order to justify this development. there exists some anecdotal evidence suggesting that even in modern societies and businesses. These numbers are much lower than 150. . 150 should be discounted for friends and relatives before yielding an upper limit to the number of business relationships that one can entertain. at a later stage. we should expect the clusters of the supposedly small-world network of business relationships to become ever less dependent on physical distance. In many cases. The small world of business relationships 195 be a problematic concept. while carrying out research on the structure of co-authored patents. the little evidence presented in the previous section cannot be deemed to be conclusive in any sense. humans are subject to constraints as far as it regards the number of people with whom they can interact. Psychologist Nigel Nicholson reports cellular organizational forms where a large number of semi- autonomous units are kept at an average of fifty employees each (Nicholson 1998). we should expect structural invariance of business relationships across time and space. Possibly. If business relationships are arranged in a small-world structure because of businessmen’s bounded rationality. However. then even a global new economy can be expected to retain certain structural features of the old one. Nevertheless. discovered that apart from researchers who are working for large firms. None the less.

Clearly. In general. this is precisely the way natural evolution proceeds. Possibly. might shape the set of innovations that can be carried out at any given point in time. mutation of a gene affects overall fitness to a greater extent the farther-reaching its interactions with other genes are. beyond that threshold a favourable mutation of a gene causes overall fitness to fall because of the negative influence that it exerts upon other genes. besides industries. the overall fitness of an organism generally does not result from simple summation of the fitness of its genes. one would need extensive interviews with managers across industries. Stuart Kauffman proposed to extend these concepts to the economic system. and (2) the changing relation of this structure to physical space. Up to a certain threshold. in the hope that the structure of business relationships did not get blurred in the process of aggregation from managers to firms and from firms to industries. one may try to use a series or set of regional input–output tables. there exists an optimal level of interaction between the components of an evolving system (Kauffman 1993). but rather depends on the extent of interactions between genes as well. In this case. input–output tables should be discretized. there exists an optimal level of genes interaction or. one could state that the net of inter-firm connections evolves according to precise psychological laws that inhibit combination of too many technologies at a time. Possibly. where technologies would take the role of genes and products would take the role of organisms (Kauffman 1988). translating their entries into zeros if they fall below a certain threshold. greater interaction means that a favourable mutation of a gene increases overall fitness to a greater extent. Thus. According to this scheme. understood as the existence of a limit to the number of relationships that can be entertained with repositories of particular technologies. . In fact. this limit lies in the cognitive inability to handle infinitely many business relationships at the same time. particularly because of the requirement to span time and space.196 Guido Fioretti looking for: (1) the existence of a small-world structure. ones if they are above it. Weird as it might seem at first sight. if this metaphor makes sense. In fact. In order to do this. Bounded rationality. Threshold values need not be fixed arbitrarily. then the path of technical progress could no longer be conceived as exogenous. However. innovations would arise out of mutation and recombination of existing technologies (Schumpeter 1911) and. to speak in broader terms. space and time. one could claim that there should exist a limit to the number of technologies that can be recombined at any given point in time. a panel of this kind is very unlikely to be realized. Note that. since they can be chosen to maximize information entropy (Schnabl 1994). if one would find a small-world structure and if this structure would be found to be invariant with technological regimes.

psychology might command variables that most economic theories take as exogenous. and Action. New York: Columbia University Press. (1999) ‘The small world Web’. Eugene Garfield. an anony- mous referee for comments and suggestions. (1992) ‘The Social Structure of Competition’. certain structural invariants are likely to persist.J. If this would turn out to be true. in S. Acknowledgments I wish to thank Jack Birner. Cambridge. R. September. Yet the main message of this contribution is that. namely from exogenous technologies towards production and consumption according to exogenous tastes. paper presented at the Third European Conference on Research and Advanced Technology for Digital Libraries. In particular. July. Burt. D. Administrative Science Quarterly 42: 339–65. Turin. D. in spite of this prejudice. A. Paris.C. R. In particular. this contribution focused on a possible psycholog- ical invariant. De Solla Price.S. Bart Nooteboom. Boston: Harvard Business School Press. Big Science. Nohria and R. Chandler. Gossip and the Evolution of Language. Axelrod. psychology is seldom accepted in economics. New York: Basic Books. (1999) ‘Making markets’. then the chain of causal links to which economists are accustomed. and Robin Dunbar for permission to reproduce a picture from his book. I. Form. R. Pierre Garrouste. Generally speaking. (1965) Little Science. Eccles (eds) Networks and Organizations: Structure. . paper presented at the IGU Conference on Local Development.G. London: Faber and Faber. Cesare Marchetti. like technology (Sen 1989). References Adamic. The small world of business relationships 197 Concluding remarks Although this short chapter does not present definitive results but rather hints and suggestions for future research.S. Burt. and Kochen.A. Social Networks 1: 5–51. Torsten Hägerstrand. namely a constraint on the number of stable relationships that humans can entertain. (1962) Strategy and Structure. De Sola Pool. MA: MIT Press. (1996) Grooming.E. Felsenstein. (2001) ‘New spatial agglomerations of technological activity – anchors or enclaves? Some evidence from Tel Aviv’. would close into a double ring where both producers and consumers innovate technologies and habits along anthropological and psychological guidelines. Cheltenham: Edward Elgar.D. J. influencing one another in a never ending evolutionary spiral. Earl (eds) Economic Organisation and Economic Knowledge: Essays in Honour of Brian J. its leading theme was that beyond all possible differences between ‘old’ and ‘new’ economy. Dunbar. Loasby. Francesco Lissoni. Dow and P. (1997) ‘The contingent value of social capital’. in N. (1978) ‘Contacts and influence’. (1984) The Evolution of Cooperation. it is never accepted when one deals with ‘hard’ issues. M. R. L. Birner.

R. S. . Kirman. V.A. (1998) ‘How hardwired is human behaviour?’.A. (1967) ‘The small-world problem’. The Economic Journal 110: 411–36. Kauffman. Physica A 285: 539–46. (1998) ‘Collective dynamics of “small world” networks’. Journal of Personality and Social Psychology 15: 101–8. The American Economic Review 35: 519–30.V. (2000) ‘Harmony in the small world’. Berlin: Duncker and Humblot.J. Amsterdam. Nature 410: 268–76. F. Weisbuch.W. Hayek. H. Harvard Business Review Jul–Aug: 134–47. and Herreiner. D. Pines (eds) The Economy as an Evolving Complex System. Arrow and D. Online. (1982) Models of Bounded Rationality. (1911) Theorie der wirtschaftlichen Entwicklung. S. Nicholson.. London: Routledge and Kegan Paul. MA: MIT Press.A. in P.J. F.B. The American Journal of Sociology 105: 493–527. (1982) An Evolutionary Theory of Economic Change.H. and Winter. (1952) The Sensory Order. Psychology Today 1: 60–7. (1999) ‘Networks. (2000) ‘Market organization and trading relationships’. and Solé. ICT and Local Policy Initiatives: Comparative Analyses and Lessons. (1973) ‘The strength of weak ties’. Hayek. (2000) ‘Models of the small world’. analyzed by a multi-layer procedure’. Keynes. Schumpeter. M.A. Nature 393: 440–2.H.> (accessed 4 April 2002). Economic Systems Research 6: 51–68. Milgram. M. Watts. Journal of Statistical Physics 101: 819–41.A.R. A.J. S. Anderson. Radner and R. S. M. Schnabl.G. Social Research 56: 299–329. Watts. R. Newman.198 Guido Fioretti Ferrer i Cancho. London: Macmillan. Economica 4: 33–54. K. S. Available HTTP: <http://www. Cambridge.P. (1989) ‘Economic methodology: Heterogeneity and relevance’. and Strogatz. (1945) ‘The use of knowledge in society’. Simon. Kilkenny. Marchiori. Granovetter. Interest and Money. D. D. F. and Latora.A. (1993) The Origins of Order.M. (1936) The General Theory of Employment. Santa Fe Institute working paper 01-03-016. H. J. G. Korte. C. Strogatz. (1988) ‘The evolution of economic webs’.A. Oxford: Oxford University Press. Cambridge. Hayek.J. (1970) ‘Acquaintance networks between racial groups: Application of the small world method’. (1937) ‘Economics and knowledge’. Redwood City: Addison- Wesley. S. (2001) ‘Exploring complex networks’. A. (2000) ‘Community networks for industrial recruiting’. Sen. (1994) ‘The evolution of production structures. MA: The Belknap Press of Harvard University Press. dynamics. S. and the small-world phenomenon’. and Milgram. Kauffman. Radner (eds). (2001) ‘The small world of human language’. The Tinbergen Institute. C. J. June.S. The American Journal of Sociology 78: 1360–80.santafe. M.K. R. paper presented at the conference Entrepreneurship. Nelson.E.

Part V Markets and market failure .


the probable opportunity of quasi- monopolies or oligopolies also requires the necessity of adequate anti-trust laws and policies. information generates important externalities and therefore substantial decrease of consumption marginal cost: let us think of the . On the one hand. information appears to be a non-rival good: consumers can use a good or a service without any possibility of excluding other consumers from this use. Other authors (see. Some of them. especially Shapiro and Varian (1998). Finally. Gensollen 2001).” On the one hand. They indeed considered that markets for information goods cannot be analyzed with the usual tools of microeconomic analysis. are more eager to combine the analytical tools of public economics with those of “new microeconomics. for example. however.9 Some specific Austrian insights on markets and the “new economy” Richard Aréna and Agnès Festré Introduction Economists disagree today on the elaboration of an adequate analysis of the markets which emerged with the development of the “new economy” (NE). firms are assumed to adopt price differentiation strategies. Two main interpretations. requiring the assumptions of perfect information and perfect competition. The NE would imply the necessity of abandoning this theory and of replacing it with the so-called “new microeconomics. This does not mean that the traditional theory of pure competition in a private good economy provides the best analytical framework. the usual tools of marginal analysis continue to fit for the study of the new realities which have to be considered and coped with. On the other hand. On the other hand. who focus specifically on the nature and regulation of markets. the existence of network externalities. According to a first view. what is argued is that information technologies (IT) and Internet permit an increasing independence of information as regards its usual means of storage and transportation. stressed the interest of using the concepts and mechanisms provided by the modern theory of information and tried to elaborate on them guidelines for firms or policy-makers. of uncertain quality and of different types of increasing returns imply the existence of substantial information asymmetries. seem to prevail. Finally.” This view has been developed by various authors even if they do not always agree on the specific tools which are the most relevant to cope with markets in the NE.

present economics editor of The Economist: Economic theory has never described the real world completely. Market coordination was supposed to prevail in the “competitive capitalism” of the nineteenth century. In both these areas. neither IT nor globalization overturns the basic rules of economics. therefore. economists have been doing some serious rethinking in recent years. advances in telecommunica- tions have brought down transaction costs by slashing communications costs . the development of business-to-business (B2B) e-trade would foreshadow a future general substitution of market coordination to hierarchical coordination among firms. like the development of large production economies of scale or the increase of the part of information in the production costs. to scan images or to reproduce video tapes with rather cheap equipment. in turn. economic mechanisms which are close to those analyzed in public economics. Perfect competition does not exist. product. often described as the modern materialization of the model of pure competition in the real world. It is often associated with the view of the emergence of a third period in the history of market economies. and the Internet in particular. Hierarchical coordination is assumed to have become progressively predominant in the “trustified capitalism” of the late nineteenth and of the twentieth centuries. and profit opportunities abundant. by making economies work rather more as the textbooks say they should. in the following text. This. This interpretation can be found. seen as “the leading edge of the digital economy” (see US Department of Commerce 2000: 7). makes information on prices. they also require new types of market regulation. The empirical support of this view – either explicit or implicit – is generally based on the increasing importance of elec- tronic commerce among producers or between producers and consumers. and many question marks remain over the precise role of technology and human capital in growth. for instance. A second view concerning the NE rather stresses the idea of a revival of market coordination. makes markets more transparent. written by Pamela Woodwall. IT. it probably never will.1 The NE would mean the end of the prevalence of the modern managerial firm and the return of market coordination. the future prevalence of direct relations between suppliers and demanders. allowing buyers and sellers to compare prices more easily. Indeed. However. The theory of perfect competition. serving it up faster and reducing its cost. the development of business- to-consumers (B2C) e-trade is supposed to generate a global process of desinter- mediation and. zero transaction costs and no barriers to entry. Computers and advanced telecommunications help to make these assumptions less far-fetched. At the same time. optimistically assumes abundant information. Moreover. On the other hand. therefore. This importance is frequently associated with two consequences. IT does the opposite. On the one hand.202 Richard Aréna and Agnès Festré possibility of individual consumers to get copies of an electronic text. They imply. a basic building-block of conventional textbook economics. All these phenomena are combined with drastic changes on the supply side.

initial individual endowments. On one side. This purpose is characterized in the second section. Information is less and less costly and more and more shared by the participants to markets: information asymmetries seem to decrease thanks to the generalization of electronic transactions. the basic assumptions of perfect competition are starting to become true. The CEE model view of the “new economy” Three arguments are generally put forward to support an interpretation of the impact of IT on e-markets based on the concept of CEE. In the final section we will refer to some concrete examples taken from the reality of electronic markets. After having stressed the limits of these arguments. and lower barriers to entry all add up to a more efficient and competitive market. Now. In other words. It is predominant today in mass media and in part of the empirical economic literature. the usual neoclassical assumptions seem to apply in our modern real world. and this is the reason why we will focus on it in the present chapter. cutting out the middleman. in order to emphasize how these concepts and developments allow a better understanding of the working of markets in the NE. even if some of the devel- opments which follow are also relevant for an assessment of the first view.” i. what is generally argued is that these “fundamentals” are more easily identified in the NE. In the first section we will recall the usual arguments which are put forward to analyze the NE as the empirical realization of the CEE model.e. consumers preferences and the “blueprint” of techniques. Austrian insights on markets and the “new economy” 203 between far-flung parts of the globe and by allowing direct contact between buyers and sellers. in a CEE model. Transaction and search costs are also supposed to diminish with the reduction of intermediaries on markets. the very notion of economic equilibrium – either partial or general – presupposes the existence of a distinction between “given” and “unknown” economic magnitudes. On the other side. “given” magnitudes corre- spond to “fundamentals. As we know. Consumers’ tastes are explicitly revealed in accordance with the requirements of the CEE model. through the use of Internet. even if they exhibit some limitations. consumers are supposed to be able to better express their preferences while producers can take them clearly into account thanks to the interaction implied by the IT. Thus. better information. low transaction costs. IT has also lowered barriers to entry by improving the economics of smaller units. we will try to show that some Austrian concepts and developments better fit to explain the economic impact of IT on present markets. (The Economist 1999: 93) This view of the markets of the NE implies that the schemes used by conven- tional ‘old’ microeconomic theory in characterizing a competitive economic equilibrium (CEE) are more and more relevant to analyze the set of transactions which is daily taking place in modern market economies. Spatial distortions also tend to disappear since any agent can buy or sell from .

Inventories can be reduced through an electronic management of the supply chain of firms: therefore. Therefore. even if some unsolved difficulties remain to explain the persistence of price dispersion on e-markets (ibid. Supply is also concerned by the use of IT. Hayek contested the possibility for a central decision-maker to have an explicit and codified knowledge of the parameters of calculation. as those which are assumed to appear on the Internet. expectations and decisions must belong to the “same set” (Hayek 1937: 37). On the one hand. a list of standardized commodities as well as a list of suppliers and demanders that would qualify for a sufficient characterization of any given market. the “subjective” data. leads to some substantial objections. From here. two conditions. which are related to their own specific positions within the mechanisms of social interaction of a market economy. interpretations of the real world might differ but agents have to refer to a unique and common real world. it means that the data on which agents base their decisions are no longer limited to the ‘fundamentals’ of the CEE model. referring to Austrian concepts. the apparatus of the general .: 104–5). however. A more careful investigation of the markets of the NE. However. Disputing the choice of a general equilibrium framework as a guide to take rational decisions in a socialist planned economy. he could not conceive of a central planner able to define. As we know. The first objection we will raise is “Hayekian. Under these circumstances. confirms this view. In 1940.” Hayek indeed always insisted on the importance of individuals’ heterogeneity on markets. Now. The debate on social calculation.204 Richard Aréna and Agnès Festré any location in the world through the use of the Internet. the assumption of full utilization of the productive capacities appears to become a better approximation of economic reality. They also include. The generalization of auction markets is also put forward in order to show that tâtonnement processes are no longer theoretical devices but tend to emerge within the real world. Hayek gave a convincing example of this problem. The use of electronic data interchange (EDI) and Internet also contributed to decreasing the importance of transaction and search costs for suppliers. his approach is based on the differences between individual preferences as well as those related to the subjective perceptions of the environment. therefore. showing that it was hardly possible to have a codified and explicit knowledge of real markets and activities. This situation. Some empirical studies seem to confirm this general picture (see Smith et al. indeed. “external events” on which individual agents found their perceptions. which Hayek participated in. According to Hayek. They must also include among their decision parameters some forecast of the future behaviors of other agents. at least. He indeed argued that part of the information on the blueprint of productive techniques in the economy were only available in the form of tacit knowledge related to “circumstances” (Hayek 1935/1949: 155). if these two conditions are fulfilled. a priori. On the other hand. agents cannot base their plans on purely external and objective facts or information. agents’ heterogeneity does not exclude some mechanism of economic coordination. involves the existence of heterogeneous subjective plans as well as strategic uncertainty. 2000). are required in order to obtain this result.

Thus. This type of objection is all the more relevant for the markets of the NE. The definition of the use-value of commodities (or services) is rendered even more complicated by the fact that. Therefore. the interactions between suppliers on one side. as early as 1928. Now. by arguing that information is a good related to experience.” When a new “day” or a new “season” begins. permanent changes affect what the CEE model calls “fundamentals. which are emerging today. within the NE. the use-value of a good is often imprecise a priori since it emerges from the interaction of consumers and producers.” called “day” or “season. in each sub-period. Indeed. Hayek did not accept the static framework of GE theory but preferred to substitute it with what he called an “intertemporal price equilibrium” framework. in the NE. which constitutes the horizon of agents’ decisions (Hayek 1928/1984: 72). the final transaction is often only indirectly linked with the electronic connection between one demander and one supplier (or intermediary). as well as consumers’ preferences (ibid. Finally. the conception of market coordination implied by the CEE inter- pretation of the NE is clearly contradictory to Hayek’s conception of the working . This illustrates why. time is described as a sequence of “flows” of “individual processes. in the field of e-commerce.” namely. in his conception of intertemporal equilibrium. for instance). Within this concept. on the other side. For instance. Obviously. where the life cycle of goods tends to be shorter and shorter and new commodities and services are appearing and disappearing at a quicker pace. it includes a flow of new economic “processes” of production.” They form the “economic period” (the “year”). Austrian insights on markets and the “new economy” 205 equilibrium (GE) theory could not be relied upon if one is looking for a satisfac- tory explanation of the mechanisms of demand and supply that prevail in real markets (Hayek 1940/1949: 188–9). Moreover.: 73). preferences and goods themselves. so that its utility cannot be appraised ex ante. Hayek accepts the possibility of real disequilibria related to persistent changes of techniques and preferences (see the “voluntary saving” case in Hayek’s business cycle theory). some of the services associated with electronic markets are free although they might contribute to the realization of final transactions. but also between consumers. By the way. Each temporal “flow” corresponds to a subdivision of the “economic period. it becomes more and more difficult to define precisely what a good is and what a market is. a computer or a computer system has no a priori use-value till the consumer interacts with the producer or intermediaries to define it a posteriori. are constantly adapting the changing environment implied by the adoption and diffusion of IT. production techniques. This feature of IT goods has often been stressed in the literature (especially by Shapiro and Varian 1998. according to Hayek. in which it is difficult to distinguish the commodity itself from the bundle of services that are related to it. Commodities are often supplied according to a “bundling” context. which contrast with the virtual disequilibria of the Walrasian and neo-Walrasian theories of tâtonnement. the objections put forward by Hayek are also valid for a market economy: it is impossible to define an objective list of “fundamentals” independently of the subjective perceptions of agents. It is also difficult to conceive of static “fundamentals” in a world in which changes affect every-day techniques.

a market economy does not refer to a “taxis” but rather to a “kosmos. According to the former interpretation. and in which market failures or disequilibria might be characterized as simple frictions.206 Richard Aréna and Agnès Festré of a market economy. Gloria-Palermo (1999) has convincingly pointed out the detail of the analytical origins of this divergence. Now. Part of their knowledge is indeed tacit or related to specific “places” or “circumstances. and Lachmann’s view. I: 16): each member of the society only knows a very limited part of “global” knowledge and any of them ignores most of the facts on which the working of the economic system rests. In such an economy. for Hayek. This kind of approach better fits with the realities of the NE. Our basic idea. Our . undiscovered profit opportunities or individual misperceptions. here. Some specific Austrian insights on the “new economy” We will try now to develop a different conception of market economies. from the unintended consequences of individual subjective plans. showing the existence of a major difference between a “Kirzner–Hayek conception” which assumes “(without really demonstrating) that disequilibria signals are sufficient to move the system towards equilibrium” and “[derives] from this assumption the conclusion that the market process constitutes an efficient coordinating device” (ibid.” It is closer to Lachmann’s treatment of markets as “institutions” (Lachmann 1986: xi) and therefore implies the analysis of the institutional features of the various types of markets. a market economy is indeed the outcome of voluntary behaviors. the intended consequences of which confirm the a priori individual objectives resulting from optimization behaviors. as it is in a CEE framework characterized by formal utility.: 126). This is why the image of the “discovery process” better describes the NE than the abstraction of a complete set of interdependent markets related to objective mechanisms and purely codified information.: 81–2). however. Their knowledge of economic activities is not entirely codified and explicit. divergent views on the theory of market processes. Hayek’s analytical time (ibid. rooted in the Austrian tradition and able to provide a better framework for understanding market realities in the NE. In this framework. according to which “the possibility of inconsistency of plans challenges the traditional view of a tendency towards equilibrium” (ibid. the “Kirzner–Hayek conception. demand and supply functions. agents again are heterogeneous.” namely.: 125). is indeed to give up the view of a universal model of market which would express the essential features of any kind of market. Austrian economists of the past and the present offer. different markets indeed imply various market processes. to some extent.” The situation of agents in their relations to knowledge might be described as a situation of “division” (Hayek 1937/1949: 50) or a “fragmentation of knowledge” (Hayek 1973/1980. our viewpoint does not only exclude the framework of the CEE model but also. a self-organized order that results. which allows the dismissal of what O’Driscoll and Rizzo (1985) called Newtonian time. where self-emerging markets are the rule and their characteristics prevent agents from understanding what is going on at level of the global society. production. imperfections. vol. namely. on the contrary. Therefore.

. according to Menger’s expression of market economies characterized by organized markets. The first factor is related to the various forms of trade organization. its inefficiency prevents its generalization: temporal distortions indeed appear in supplying as well as in delivery. which depend on four main factors. This failure of the production on order system then paves the way to the “institutional arrangements” (ibid. For Menger. our dissatisfaction with Hayek’s theory of markets mainly concerns. The second stage of the process consists in the introduction of a craftsmen system where producers use inputs belonging to consumers in order to provide them with outputs in counterpart of a material levy. as well as the prevailing legislation (ibid. These mechanisms do not always imply flexible prices. In accordance with Menger’s evolutionary approach of the emergence of institutions (Menger 1871/1976: 232–86. however. therefore. 1883/1963: 127–61).: 251–2). Austrian insights on markets and the “new economy” 207 preference for a Lachmann-type conception does not exclude. In a third stage. This process can take on the most various forms and.2 The technical division of labor is present but “self-sufficiency” is prevailing. However. These constraints do not mean that Menger always assumes market clearing. however.: 252–3). the utilization of Hayek’s contribution to the analysis of the relations between social rules and individual behaviors. production on order is introduced. He explicitly considers the occurrence of inventories and this possibility obviously depends on the flexibility of prices. explain the diversity and specificity of market and market organization types.: 248 and 249). cannot be assimilated to competitive imperfections. For Menger. which ignores market mechanisms (Menger 1871/1976: 236). the means of information circulation. their periodicity and to the purchase rate. The third factor concerns the mechanisms of auction and the habits and methods of bargaining.: 251). The variety of the forms of market organization thus precludes the existence of a unique model of market economy.: 238). The origin of our view is to be found in Menger’s works. the origin of this process is located in the existence of a production economy. The fourth factor refers to the length of the period during which transactions are allowed. In these “isolated domestic economies. The self-organized and self-reinforced aspect of this evolutionary process explains why initial conditions and followed paths are different according to the country or the culture which is considered. They correspond to different degrees of exchangeability (Menger 1871/ 1976: 241). his belief in the existence of an “empirical” tendency of market economies toward equilibrium. as it is in the CEE model. These forms. the accessibility of markets and their working mechanisms.” production is not directed towards exchange transactions between anonymous agents. These forms differ according to the size of supply and demand (ibid. These elements correspond to the temporal constraints of transactions (ibid. in fact. The second factor is con- nected to the location of agents and the spatial constraints of transactions (ibid. this kind of economy is the result of a slow process of self-organization and self- reinforcement. intermediaries between producers and consumers and monetary institutions (see Aréna 1999: 24). Menger does not exclude the possibility of sticky or rigid prices (ibid. it is clear that a market economy is not a universal and unchanging system of agent coordination.: 243).

Thus. Far from conceiving the first type of market as an anomaly. emphasizing what he called “exchange institutions. which might explain the existence of panics. market diversity is therefore assumed. If Menger’s criteria play a major role in Wieser’s typology of markets. decentralized social . which permits producers to impose supply prices on consumers. The second criterion is related to the forms of auctions and bargaining methods. The mention of these rules or habits is important since it shows that. for instance. Lachmann rather considers that it results from the process of commodity stan- dardization. the presence of arbitrageurs or speculators on a specific market implies very different market mechanisms (Lachmann 1986: 125). Menger combines them with natural rules or habits. taking this context into account. namely. Wieser resumed Menger’s criteria of market diversity and used them to define a real institutional typology of markets. sticky prices and inventories. in each specific market.208 Richard Aréna and Agnès Festré In Menger’s approach. These rights are essential since. market transactions are “embedded” in a specific social and cultural context and have to be studied. for instance. they determine what is. Wieser characterizes “customs” as permanent social rules. Therefore. in his explanation of market diversity. The third criterion put forward by Lachmann is related to the nature of market intermediaries: for instance. However.: 3). These “institutions” first refer to the various forms of property and contract rights. The first one refers to the organizational and spatial specificity of markets (ibid. which Menger also identified. as in Menger. Lachmann inherited this type of approach from the Menger–Wieser Austrian tradition. They are similar to what modern economic analysis would characterize as “routines. Lachmann indeed opposes consumers’ markets and producers’ markets. These preceding developments show how Menger. Others are “disorganized” (ibid. Exchange institutions also refer to agents’ “customs” (ibid. they shape the nature of market transactions in accordance with law and. Combining the two already mentioned by Menger. “social institutions” (ibid. The fourth criterion derives from Menger’s seminal distinction between exchange economies and production economies.: 179). Some of these markets are “organized” according to some permanent and specific rules (Wieser 1927/1967: 173–6). Wieser however took a step forward. and what is not allowed. He also emphasized the fundamental diversity of markets (Lachmann 1983: 3) and his criteria are close to Menger’s.: 172). beyond it. Within this typology. The four factors we just mentioned are related to the degree of exchangeability of commodities.” Routines or customs are especially significant to understand why market conditions are always changing gradually (ibid. but the latter can be rendered more flexible by the introduction of an intermediary. Lachmann opposes fix price markets and flexible price markets.: 175) and represent a kind of economic pathology.” defined as a supplementary element of market diversity. substantial productive capacities and strong technological complementarities tend to imply sticky prices. which agents follow when they make transactions on markets.). for Menger. Wieser and Lachmann all contributed to put an institutionalist typology of markets in the place of the idea of a universal model of market.

then it is repeated. two main types might be distinguished. finally. Agents also unconsciously adopt some social rules. If the rule makes the agent better off. The second type of social interaction corresponds to the innovative attitude. shows how decentralized social interaction is important to understand the working of markets. we will define a typology of the markets of the NE which combines Menger’s criteria of market diversity and Hayek’s focus on decentralized social interactions. for instance. we however welcome his conception of “dispersed information. Then. Therefore. What is more important is to note that they always suppose some type of social interaction. The empirical foundation of this typology is obviously rooted in the two main types of markets. the agent tries to imagine and to introduce a new kind of behavior and. In this case. Agents are not able to acquire a complete knowledge of past actions of other participants to markets. Others are the legacy of culture. If the agent realizes that the innovative behavior makes him better off. Market coordination then requires an indirect way of knowing and understanding the various strategies of other individuals. he assimilates the rule(s) that govern(s) it. which appeared with the emergence of the NE. Rule following does not always derive from mimetic or innovative attitudes. considering the market realities of the NE. this part of individual knowledge cannot be transferred to another agent. It consists in the imitation of other individuals’ observed actions. then he reiterates this conduct and. A process of “trial and error” is often necessary in order to find the behavioral rules that fit with the social context.” Now. The markets of the “new economy” in the Austrian perspective In this last part of our chapter. Consequently. are imposed by law (see Hayek 1973/1980: 52). Some are inherited conventions. he observes how it performs. the existence of rules. one of the major consequences of this “division of knowledge” is the fact that a substantial part of agents’ knowledge is strictly tacit and private (see on this point the influence of Polanyi (1966) on Hayek). namely. however. this imitating behavior is revised according to how the rule that has been chosen actually performs. Hayek strongly contributed in stressing this aspect when referring to the social division of knowledge: if we disagree with Hayek’s belief in the existence of a tendency of market economies towards equilibrium. in accordance with Wieser and Lachmann’s approaches. This way belongs to the realm of what Hayek called “unorganized knowledge” or “knowledge of the particular circumstances of the moment and the place” (Hayek 1949: 80). According to Hayek. Burton-Jones 1999 and Currie 2000) and . It is not worth analyzing here the forms of these rules. The first is mimetic.3 It is now time to check it. Agents accumulate this type of knowledge through the use of some persistent behavioral rules. Austrian insights on markets and the “new economy” 209 interactions between agents play a major role. here again. Others. neither are they able to forecast their future actions. electronic markets (see. little by little. Beyond this diversity. Innovative attitudes are not always successful however. the mimetic attitude gradually endogenizes the rules governing the behavior of observed and imitated agents.

in our typology. Generally. limiting the time of bargaining to an a priori fixed period (Raisch 2001: 136–7). even sometimes. for instance. to the following four kinds of markets. product quality or delivery guarantee. however. sticky prices that are revised from time to time. they combine a search engine. or information on prices or quality of new products. Very often. quick auctions can also occur. Producer or consumer auctions also exist. consumers are also interacting between each other through “peer-to-peer” communities (such as Napster or Gnutella). These two empirical types of markets will give birth. it can be seen that new intermediaries introduce a kind of ancillary market. Finally. namely. Usual intermediaries are indeed replaced with “infomediaries. Moreover. Quite the reverse. secure payments. Moreover. E-market bundling indeed confirms the impossibility of defining an a priori list of identified consumer preferences and standardized goods. In this case. which offers to consumers a bundle of services and information dedicated to help them to make choices. Strong objections can. the Austrian view appears to be . a means of interactivity between intermediaries and demanders and possibilities for personalization in individual choices. This new form of intermediation also appears in the case of portals. others. the very nature of goods (Raisch 2001: 23). an organization of available information. on B2C markets. it is striking to note that today.210 Richard Aréna and Agnès Festré technological markets (see Guilhon 2001).4 Some offer fixed prices associated with a catalog. From this angle. music. they do influence the determination of the use-value of goods. Second. various auction systems are used on e-markets and most of them are not Walrasian. a process of reintermediation or substantial changes in intermediation is presently occurring on these markets. their decisions are indeed partic- ularly difficult. Thus. All these aspects stress how much the Austrian tradition is right when it emphasizes the importance of intermediaries and their impact on the formation of subjective consumers’ preferences. be raised against the idea of tendency to desintermediation on these markets. they do not only concern prices but also delivery or payment dates and. the variety of auction systems is substantial. Here again. the nature of goods and the formation of preferences emerge from the interactivity between intermediaries and consumers and social uses are created through the use of the Internet. Portals are not only entry points for purchase. in order to exchange digital goods. B2C direct e-markets are not only characterized by the existence of “info- mediaries” but also by bargaining modes. First. intermediaries tend to become information providers. Now. in accordance with Hayek’s view of the division of knowledge.” These new forms of intermediation are especially useful when consumers are confronted with complex digital producers and numerous web sites. B2C direct e-markets These markets are those to which commentators refer when they interpret the NE as the achievement of the type of market coordination analyzed in the CEE model. They also try to take into account problems related to the safety of transactions.

Therefore. while new standards are more related to information and collective soon realized. in accordance with Hayek’s approach of the emergence of rules. Spatial and temporal constraints are also important on B2C direct e-markets. lies in the fact that all standards are essentially physical and characterize tangible goods. Menger’s typology appears to be particularly relevant. In the “old economy. the difference. to the very private institutions or professional associations which contribute in defining and introducing new standards of communication and transfer of information. therefore. A second type of agent corresponds to market organizers. play an essential role in the NE. that free information or services through dynamic hypertext or discussion forums allowing internauts to interact between each other could be profitable. A consensus must thus emerge in order to allow e-consumers and e-intermediaries to communicate. This illustrates the role of learning.” the definition of standards is also playing an important role. imitation and innovation in the development of electronic markets. however. Now. what is striking in an Austrian perspective is that standards are never defined a priori by hierarchies but that they emerge from self-organized processes that market participants can only approve or disapprove (Picot 2001: 8). Standards such as htlm (hypertext markup language). also contribute to the emergence of new social-use values. new standards require a social agreement of the participants to the market in order to set a situation of common knowledge. A third type of actors is constituted by consumer communities which try to influence both the first two types of actors. i. They are thus shaping market processes on the supply side and these processes emerge as typical unintended consequences of entrepreneurs’ decisions. Production firm managers or intermediaries are exemplary Austrian entrepreneurs. It is clear. E-markets are indeed the result of spontaneous behaviors of different types of actors who affect various aspects of transactions on markets.e. for instance. some “dotcom” firms such as Amazon. For instance. The reference to exchange institutions also allows analyzing electronic markets as typical self-emerging Austrian markets. These processes are the result of innovative and imitative behaviors. Wieser’s “exchange institutions” are also present on e-markets. that e-markets cannot be analyzed with the usual tools . On the contrary. entrepreneurs are not the only relevant actors. in order to convince them to build responses in accordance with their requirements and. therefore. Austrian insights on markets and the “new economy” 211 very useful since it distinguishes markets according to the method of bargaining and the system of auctions. before others that imitated them afterwards. They indeed look for profit opportunities and contribute to the discovery of new products or new markets through a process of trial and error. A significant example of these institutions or these rules is given by standards. the idea of a generalized Walrasian auction price system is clearly disputable. Here again. These features are specific to this type of market and imply that spatial and temporal constraints do not play a major role in its working. Direct e-markets are indeed defined as those where order and delivery are both electronic and where also the products themselves are digital. However.

inventory management or financing optimization. which prevails on e-markets. While increasing returns in the upper stages of production tend to favor cooperation between firms. They are confronted with additional problems. the empirical price dispersion. In this framework. specialized in specific groups of products belonging to some precise human interest (sport. Part of the organizational change that is taking place in the NE is interpreted as a reaction to the increasing tension inside the value chain (Gensollen 2001: 7). which allows defining them as typical institutional arrangements. R&D strategic choices are crucial. such as aggregators that try to combine the electronic flexibility of transactions with an inventory management policy. Within this framework. It is easy to see that these new constraints exert immediate effects on trade organization and explain why new types of intermediaries emerge. which are often virtual. however. They are submitted to self-organizing processes and continuous changes. The emergence of standards is not. is not necessarily a market imperfection or a sign of “market immaturity.” It can derive from both a bundling strategy of intermediaries and the existence of a hierarchy of trust relations in the minds of consumers. products and services.” On the other hand. Now. social interaction is clearly important on e-markets. derives from the fact that transactions concern tangible goods even if orders are electronic. the increasing need for “one-to-one” marketing downstream seems to be more adapted to competitive strategies. They are therefore confronted by a permanent uncertainty regarding the nature of transacted goods. which appears when we consider indirect e-markets. For instance. suppliers have impor- tant means of identification of consumers’ communities but the importance of consumer learning and preference change implies the adoption of a systematic “alertness. The use of the Internet for market transactions indeed entails problems related to the anonymous nature of consumers. which were usual in the “old economy”: for instance. consumers buy new goods and services.). Another important form of social interaction is related to the role of trust on e-markets. B2C indirect e-markets The preceding remarks could be resumed and applied to indirect e-markets too. or e-market places. Therefore. cultural and social subjects. On the one hand. . etc. The only difference.212 Richard Aréna and Agnès Festré of the CEE model. They indeed tend to become a deciding factor in the context of the NE. the existence of trust seems to contradict economic rationality in some cases. scientific topics. These new forms of organization are prevalently dedicated to the decrease of spatial and temporal constraints related to good delivery and technological innovations. the only example. intermediaries cannot limit their role to the management of information flows. This is why firms or intermediaries spend important resources to acquire a reputation in order to win consumers’ trust.

firms or intermediaries can easily acquire numerous and various pieces of information related to consumers. These new possibilities allow them to develop a much more efficient marketing policy. This factor is certainly essential but does not. firms were forced to employ many employees to answer to their customers’ queries and improve their knowledge of consumers’ preferences. These assets are protected by intellectual property rights in the form of patents. however. These new types of trade organization allow firms to replace hierarchies with efficient producer markets that are dedicated to specific firm needs and requirements. B2B e-markets easily replace interfirm relations. These advantages also prevail in the realm of interfirm relations for supply as well as delivery. these relations cannot be obtained through usual market coordination.” either vertical or horizontal. Austrian insights on markets and the “new economy” 213 B2B e-markets Very often. A good example of this improvement is given by the changes occurring in marketing activities within e-trade. This emergence is comparable to what is happening on B2C e-markets. To some extent. we might indeed note that B2B e-markets allow a more efficient “discovery process” on markets since they contribute in improving information between firms. However. they are now able to aim at precise targets corresponding to specific communities of consumers. From this perspective. In usual markets. imply a tendency to perfect competition. Finally. A significant example is given by subcontracting relations where the generalization of e-trade is sometimes equivalent to a selection process among small and medium firms devoted to find those which are the most reliable. The use of the Internet substantially changed this situation. standardization on B2B e-markets also implies a separation between two kinds of markets. New forms of marketplaces indeed imply the utilization of “hubs. Seen from a Hayekian angle. For instance. the relatively increasing role of B2B e-markets is explained by the reduction of transaction and search costs (see Aréna 2001: 17–18). licenses and patterns. When they are more complex. When firm needs can be easily defined and give birth to standardized products. On the other hand. through the Internet. it is then necessary to come back to firm agreements or to introduce “technological markets. they shape relationships between instrumental knowledge and activities that . B2B e-markets also help to develop social interactions. These markets are likely to transfer knowledge already established or on the way to be. Discovery processes might also be improved by the reduction of strategic uncertainty. Another example could again correspond to the emergence of standards.” Technological markets Technological markets concern transactions related to scientific and intangible assets. or e-procurement marketplaces (Raisch 2001: 211–14). copyrights. which decrease the degree of dispersed knowledge.

which entails externalities and proximity effects. to a national system of innovation. (Guilhon 2001: 11) These products first require strong temporal constraints: most of the time. a priori. sets of preferences or catalogs of techniques. to some extent. Quite the contrary. for instance. forms of social interaction that cannot be reduced to price coordination. and spatial and temporal constraints rather contributes in reviving the old Austrian tradition. Technological markets also provide a significant example of the limits of the CEE model definition of fundamentals. Firms are looking for instrumental knowledge. They sometimes also imply geographical constraints when. technological markets imply relations of trust and mutual knowledge. the introduction of a typology which includes direct and indirect B2C and B2B e-trade as well as technological markets substantially weakens this interpretation. even if the statistical importance of this trade still remains very modest. development. It is for instance the case when the market is related to an industrial district or. However.214 Richard Aréna and Agnès Festré represent the firm’s value chain: research. but excludes the deterministic tendency towards a market order characterized a priori. marketing. However. This is why our Austrian reference only concerns its Mengerian origin and .” while Guilhon referred to them as “quasi-markets. In these markets. it is indeed perfectly impossible to define. Finally. The observed diversity of trade organization forms. suppliers and buyers belong to a network. According to us. and which they contribute in creating during both stages of conception and production of goods. This reference to the Austrian tradition does not mean. Technological markets also exist because of the existence of a Hayekian division of knowledge in the economy. they are perfectly compatible with the Menger– Wieser–Lachmann line of interpretation. Some conclusive remarks and limitations The rise of direct B2B e-trade during the last decade led some observers to interpret it as a confirmation of the relevance of the CEE model view of market coordination. the NE exhibits the emergence of a multiplicity of market types. however. This is why Lundvall labelled these markets “rather organized markets. however. conception. the use of which is partially ignored by them. that the NE confirms Hayek’s conception of the market order. bargaining and auction systems. namely. They indeed allow transactions of codified and explicit knowledge between firms.” This mixture of electronic hierarchies and e-market coordination does not fit so well with Hayek’s view of the market order. all the more so since an essential role is played by social coordination mechanisms on markets. pro- duction. firms do not acquire this knowledge for itself but rather to complete the tacit and private knowledge that forms the basic resource of firms. producers who buy or sell in technological markets have to build long-term relations based on trust and mutual knowledge. which does not contradict the idea of a spontaneous order.

Schlör are specially and gratefully acknowledged. if the Menger–Wieser– Lachman line of interpretation provides a broad framework for the analysis of various types of price formation mechanisms. suggests is that the Austrian approach is useful to explain some important features of the markets of the NE. What our contribution. 8–11 November 2001). single and multiple round sealed bid auctions and Dutch auctions. other contributions might be utilized. this theory does not seem to provide a ready-made analysis of supply phenomena. to “cognitive economics. therefore. N. The defenders of the CEE view often argue that IT makes markets more transparent. This is permitted by the introduction of new forms of marketing and advertisement. P. Belussi. On the other hand. Neither does it mean that Austrian theory is the only one able to offer a proper theoretical framework for understanding the emergence of new markets in the NE. See Kumar and Feldman (1999). for instance. which appear on the markets of the NE. Kirznerian entrepreneurship is certainly not sufficient from this standpoint. Austrian insights on markets and the “new economy” 215 its post-Mengerian developments. but also that it must be completed and included in the broader perspective of institutionalist economics. they are organized in a way which allows a tighter control of producers on demand.” which focuses on the role of social interactions in the working of markets. it is necessary to use other lines of contributions. Useful comments on these drafts by F. the absence of control of supply by demand. This assertion is strongly dubious. Jansson and H. for instance (see Aréna 1999. which have little in common with the Austrian tradition. Internet markets are not more transparent than traditional markets (see Brynjolfsson and Smith 1999). Notes Preliminary drafts of this contribution have been presented to the third conference of the Association of Historians of the Austrian Tradition in Economic Thought (held in Pisa. N. 3 The role of rule emerging and rule following on Hayekian markets has been especially stressed in the literature these last years: see Birner (1999). On the one hand. Hayek’s “empirical” tendency towards equilibrium cannot offer a solution. From this standpoint. it does not however go very far. vol. Gunning. Garrouste (1999a. such as the Marshallian one. Foss. by the emergence of new standards and institutional arrangements and by the ability of firms to generate the creation of consumer communities which individuals are induced to belong to. 24–26 May 2001) and to the annual conference of the European Association for Evolutionary Political Economy on “Comparing Economic Institutions” (held in Siena. Rizzello (1999). . P. 2001 and 2002). namely. in some cases. Quite the contrary. Schmidt and Versailles (1999). Ioannides (1999). The usual disclaimer applies. 1999b). I). 2 This analysis shows how the assumption that the Austrian School coped with pure exchange economies is superficial. 1 The expressions “competitive” and “trustified capitalism” are due to Schumpeter (1939. Now. Here again. 4 The commonly used auction types are the open-cry (or English) auctions. We could refer. We are here confronted by a usual shortcoming of the Austrian tradition. its neglect of organization and planned institutions. Garrouste. one of the problems of the NE is indeed.

London and New York: Routledge. Hayek. Dordrecht: Kluwer Academic Publishers. firme et direction d’entreprise: Une perspective hayekienne. MIT.” Revue d’Economie Industrielle 97(4). and Feldsman.A. (1986) The Market as an Economic Process. P.” London: The Economist Newspaper Ltd. Hayek (1949) Individualism and Economic Order. I: Rules and Order. F. (1999) The Evolution of Austrian Economics – From Menger to Lachmann.” in F. Garrouste. Hayek. L. (1999) “Frictionless commerce? A comparison of Internet and conventional retailers. Hayek. F.” contribution to the conference Nouvelle Economie: Théories et Evidences. Chichester: John Wiley and The Economist (1999) “Economics – making sense of the modern economy. Birner. et des institutions dans l’analyse économique d’Alfred Marshall: Une interprétation évolutionniste. . (2001) Technology and Markets for Knowledge. London: Routledge.A. London: Routledge and Kegan Paul. Vol. London: Routledge and Kegan Gloria-Palermo. (1999) “Making markets.A. F.” History of Economic Ideas 7(1–2). (1935) “Socialist calculation: The present state of the debate. M. (1999) “Internet auctions. Hayek: Money Capital and Fluctuations. Guilhon.” in B. Early Essays. M. New York: Oxford University Press. F. London: Routledge.A. (1999) Knowledge Capitalism. (2001) “A propos de la place de l’organisation.” in F. Vol. Paris.216 Richard Aréna and Agnès Festré References Aréna. R.” in F. University of Paris-Sud. (1937) “Economics and knowledge. (1999b) “La firme ‘hayekienne’ entre institutions et organisation. London: Routledge and Kegan Paul. Bellon. Brynjolfsson. Loasby. Hayek: Money Capital and Fluctuations. London: Routledge and Kegan Paul. (1928) “Intertemporal price equilibrium and movement in the value of money. R. Dow and P. R. (1999a) “Is the Hayekian evolutionism coherent?. Aréna. Rallet (eds) La Nouvelle Economie en perspective. (1949) Individualism and Economic Order. I. Hayek. and Profile Books Ltd.E.” in S. 1980). Ben Youssef and A. McCloughry. Earl (eds) Economic Organisation and Economic Knowledge: Essays in Honour of Brian J. edited by R.” in S.” Revue d’Economie Politique 6. P. (2001) “L’avenir des marchés: Planification ou écosystèmes. McClaughry. (1999) “Austrians and Marshallians on markets: Historical origins and compatible views.A. E. 1. Hayek.” Revue d’Economie Politique 6. Hayek (1984) F. 1. Hayek (1949) Individualism and Economic Order. 33–56. W. F. F.C. marchés technologiques: Nouvelles réalités et nouveaux outils d’analyse. Currie. Burton-Jones. Lachmann. Paris: Economica.A. Hayek. London: Routledge and Kegan Paul. Vol. S.” mimeo. (1973/1980) Law. Ioannides. edited by R. Hayek.A. 17–18 May 2001. (2003) “Marchés électroniques. (1999) “Marché. Aréna.A. Earl (eds) Economic Organization and Economic Knowledge: Essays in Honour of Brian J. Early Essays.A.A. Dow and P. London: Routledge and Kegan Paul pp. S. Garrouste. Kumar. http://ebusiness. (2000) The Global Information Society.A. M. A. and Smith. Hayek (1949) Individualism and Economic Order. J. PUF.M.A.C.E.A. (1984) F. (1940) “Socialist calculation: the competitive solution. F. Loasby. London: Routledge and Kegan Paul (French translation. A.” PDF. Legislation and Liberty. Gensollen.” in F. Oxford: Basil Blackwell. B.

A. (1985) The Economics of Time and Ignorance. (1871/1976) Principles of Economics. . E. (1999) The Economics of the Mind. Tools and Research. Washington. Bailey. (1967) The Tacit Dimension.P. S. Wieser. (1927/1967) Social Economics. and Brynjolfsson. J. G. and Versailles.J. New York: McGraw-Hill.” Problèmes Economiques 2697. C. (1999) “Une théorie hayekienne de la connaissance économique?. Rizzello. Economics and Statistics Administration – Office of Policy Development. (1998) Information Rules. Kelley (first English translation: 1927). Shapiro. Raisch. London: Routledge and Kegan Paul. Urbana: University of Illinois Press (first German edition: 1883. M.” in E. (2001) “Un facteur-clé. New York: McGraw-Hill. Kahin (eds) Understanding the Digital Economy – Data. Picot. C.R. A. Polanyi. C. M. H. and Varian. Schumpeter.. Harvard: Harvard Business School Press. (1939) Business Cycles (2 vols). English translation: 1963). and Rizzo.” Revue d’Economie Politique 6. New York: Augustus M. US Department of Commerce (2000) Digital Economy 2000. (2001) The E-marketplace – Strategies for Success in B2B E-commerce. O’Driscoll. MA: MIT Press. F. Menger. Schmidt. C. D. Brynjolfsson and B. M. W. les standards. (1883/1963) Problems of Economics and Sociology. Austrian insights on markets and the “new economy” 217 Menger. English translation: 1976). (2000) “Understanding digital markets: Review and assessment. Smith. J. Cambridge. A Strategic Guide to the Network Economy. Oxford and New York: Basil Blackwell. London: Edward Elgar. New York: New York University Press (first German edition: 1871.

a function that markets have shown to be more efficient at accomplishing than regulations (Coase 1960). Prospective buyers accordingly lower their expectations of average quality. Akerlof offers a subsequent caveat stating “these institutions are nonatomistic. Contrary to these private means of ameliorating information asymmetries being nonatomistic. if it materializes at all. therefore. Akerlof’s devotion to the neoclassical theory of competition.”1 In other words. the gains to unscrupulous sellers from misrepresenting their products are externalized on to other sellers in the market in the form of lower subsequent prices. thus leading to market power and a further misallocation of resources. markets fail due to asymmetric information and private solutions themselves lead to further failure in the market system. his otherwise insightful analysis of information problems inherent in markets notwithstanding. More precisely. the market becomes increasingly saturated with lower quality goods and the market disappears. the incentive exists for sellers of relatively low quality goods to promote their wares as being of higher quality. leads . Akerlof inevitably concludes that since “[I]n these markets social and private returns differ . . . Adverse selection. and therefore concentrations of power – with ill consequences of their own – can develop. Although he cites “private institutions” as a means of ameliorating the problems resulting from uncertainty. These lower prices force sellers of higher quality goods – unable to credibly signal the superior quality of their goods in order to exact a price above their reservation price – to exit the market. leads to declining average quality of goods in markets rife with asymmetric information. the prices they are willing to pay. As this cycle is repeated. entrepreneurs vie to minimize measurement costs and efficiently to allocate these and other verification or social costs across buyers and sellers. In the absence of an intervening mechanism to correct information problems. and consequently.10 Turning lemons into lemonade Entrepreneurial solutions to adverse selection problems in e-commerce Mark Steckbeck and Peter Boettke Akerlof (1984) postulates the failure of markets in which asymmetric information between buyers and sellers leads to adverse selection in the quality of goods sold in such markets. especially the assumption of product homogeneity.”2 In this chapter we challenge Akerlof’s inference that market solutions to resolving problems stemming from asymmetric information lead to further market failures. governmental intervention may increase the welfare of all parties.

Markets for used cars (Akerlof’s favorite example) thrive. as in most markets. Such success can be attributed to the private sector. theoretically cause both sellers and potential buyers of high quality goods to leave the market leading to its demise. We contend that. commerce conducted over the Internet continues to grow. We also witness successful exchanges in such markets as automotive repair. Entrepreneurial solutions in e-commerce 219 him to overlook the robustness of markets. that the incentive to facilitate exchange in a safe and secure environment leads entrepreneurs to discover and to experiment with new technologies and ideas in their quest to provide such a marketplace. markets consisting of experience goods in which asymmetric information is prevalent. Despite the widespread media attention about “dot-com” failures. In the second part we describe the competitive solution to market failure resulting from asymmetric information as posited by Hayek (1948). however. Attributing the success of these markets to government intervention and regulation is dubious. His response. The Akerlof problem Akerlof’s model identifies potentially serious problems in some markets: asymmetric information creates an incentive for individuals to conceal private information when they interact in markets. Perhaps an even more challenging empirical anomaly to Akerlof’s “lemons theory” is the overwhelming success of electronic commerce and Internet markets. is to give short shrift to private institutions for abating this failure. therefore. software. and the rules or norms that have evolved. In the first part we summarize Akerlof’s lemons model and describe his proposed solutions for abating market failure and the subsequent market power. it is the entrepreneurial innovations employed to ameliorate potential “lemons” problems in Internet markets – markets where exchange participants are largely unconstrained by formal legal institutions – that explain the success of electronic commerce. have actually given rise to Hayek solutions. In the last section we consider Akerlof’s problem as it pertains to electronic commerce. and home remodeling and construction. that promote trust among strangers interacting in cyberspace. notwith- standing the pervasiveness of asymmetric information in this market. and illustrate the Hayekian solutions private firms have elicited to prevent lemons markets from evolving on the Internet. not the public sector. The ubiquity of asymmetric information and the ineffectiveness of formal legal institutions to compel cooper- ation in these markets have not proven to be obstacles to the growth of electronic commerce. Such problems. the quintessential forum for such problems to arise. What Akerlof perceives as problems in market systems. if left unabated. As Hayek (1948: 103–4) argued: . It is precisely the self-governing mechanisms developed by private firms. a problem assumed to be more common with suppliers than with consumers. as private actors situated in particular time and place have adjusted their behavior to realize the mutual gains from exchange.

and formal legal enforcement is emasculated and impractical for policing and mediating Internet transactions.5 million users listing 700.5 More intriguing is the success of Internet auction markets – the online equivalent of flea markets where unknown individuals exchange with each other from around the world. the market leader with an estimated 76 percent of the online auction traffic and over 85 percent of online auction sales. Owners of used cars have this experience and. the probability of conducting subsequent exchanges is quite low. And eBay. The empirical observation of thriving markets against such a difficult backdrop should direct our attention to the multitude of ways in which market participants gain assurance and earn trust from one another (see Klein 1997 and 2000). Retailers selling over the Internet grew by one-third to 550.. exceeding $132 billion for 2000. Aggregate Internet auction sales grew from about $7.” I am inclined to argue that the need for competition is nowhere greater than in fields in which the nature of the commodities or services makes it impossible that it ever should create a perfect market in the theoretical sense.3 Revenues of business to consumer (B2C) Internet retailers – roughly 20 percent of e-commerce – are projected to exceed $125 billion by 2004. most often from different states or even countries. payment is remitted prior to shipment of the exchange good. .1 trillion by 2010.4 increasing to $1. Average revenue growth for these retailers increased by 130 percent from the previous year. have better knowledge of the car’s quality than potential buyers. They know whether it has been properly maintained. It consists of virtual markets in that exchange goods are sold sight unseen. has a current membership of over 22. adverse selection and moral hazard ought to prevail thwarting the development of markets in cyberspace. whether it has been in an accident. especially for auction market participants.000 online businesses by mid-2000. Internet commerce has. The Internet is seemingly an ideal marketplace for Akerlof’s lemons theory to be tested. up from $58 billion in 1999. Indeed.g. The success of e-commerce illustrates Hayek’s notion of competition. far from competition being beneficial only when it is “perfect.000 items daily on its web site. therefore. and continues to grow rapidly.220 Mark Steckbeck and Peter Boettke The confusion between the objective facts of the situation [e. Under such conditions. an average annual increase of 446 percent. Akerlof asymmetries] and the character of human responses to it tends to conceal from us the important fact that competition is the more important the more complex or “imperfect” are the objective conditions in which it has to operate.5 billion at the end of 2000. Determining whether a car is a lemon (poor quality) or a cream puff (high quality) requires experiencing the good – driving for a period of time in order to ascertain defects in the vehicle. Akerlof lemons and Hayek entrepreneurs The canonical example of Akerlof’s lemons model is the used car market.3 million at the end of 1996 to $6. exchange participants interact from distant geographic locations.

stating “[T]hose who can identify used cars in our example and can guarantee the quality may profit by as much as the difference between type two traders’ buying price and type one traders’ selling price.6 Economic goods are produced and sold in various forms and qualities. rather than sources. then subsequently changes his or her identity and begins anew. Sellers of poor quality goods can internalize the benefits of providing false or misleading information. hence the use of labor for anything other than production of these homogeneous goods is a wasted resource. even in online auction sites. he denounces such entrepreneurial endeavor as a waste of resources. a rarely observed occurrence. certain that they could be concealed from a potential purchaser shopping for cars online whose only information about a particular car’s quality is gleaned from a seller’s description as posted on the Internet. Akerlof’s premise is correct in that an effective and efficient mechanism compelling cooperation being absent. but in the process resources are diverted from production activities. It is certainly conceivable that defects such as these could be concealed from potential buyers who are able to inspect such a vehicle in person. imposing direct and indirect costs on others. the tendency would be for Internet markets to be inundated with malfeasants attempting to defraud potential purchasers outright by taking their money and neglecting to ship ordered goods. Akerlof (1984: 16) does address the role of the entrepreneur in abating problems with asymmetric information. From an Akerlof lemons perspective. a potential problem more severe than misrepresented product quality. But. a company founded 25 years after publication of this seminal piece. in other words. may work to minimize the problems of quality assurance in marketing. Given that goods are almost universally shipped only after the purchaser remits payment and that individuals can effortlessly remain anonymous while interacting on the Internet. Markets with product variation and the activities associated with assuring quality in such markets are impediments. After payment is remitted the miscreant neglects delivering the good(s). What Akerlof’s model tends to ignore. Offering to sell goods on eBay with no intention of shipping the expected good is easily achieved. we contend. is the dynamism of markets and the incentive mechanism driving entre- preneurs to discover ways to ameliorate problems associated with market exchange. Honesty and truth telling possess public goods characteristics in that individuals need not expend resources verifying accuracy and completeness of transac- tions when market participants act with the interests of others in the community (Hollis 1998). that in the neoclassical world of completely homogeneous goods there would be no variations in quality. human nature elicits opportunistic behavior in markets beset with asymmetric information. on the other hand. or whether there are hidden defects that appear intermittently like electrical shorts or vibrations. of economic development in the Akerlof lemons model. It is. Entrepreneurial solutions in e-commerce 221 whether the brakes are safe.” a prescient reference to Carmax. Entrepreneurship. and disagreements between exchange partners often emerge as a result of miscommunication or dissimilar characterizations regarding quality or quantity . it is an ideal environment for such miscreants to flourish.

and (4) licensing. Market processes emerge from a series of trial and error experimentations. it is how and where relevant information is made known. Additionally. estab- lishing relationships with real estate agencies to provide broker services to both buyers and sellers in exchange for providing them with referrals. Consequently.7 Indeed. markets are not creatures of conscious human planning of their entire structure. Newer. What is deemed efficient today becomes obsolete tomorrow. and knowledge of market participants in constant flux. more efficient standards. That Akerlof alludes to the role private institutions and entrepreneurship play in markets is laudable. leaving them with little experience in buying and selling a home. derived from a progression of finding a more efficient means of facilitating exchange. That he gives such short shrift to market forces for self- regulation is not. markets serve to facilitate exchange between two or more individuals in an orderly process of human interaction. In return for the referral. is a heterogeneous good and is the single greatest asset in terms of cost in which most people invest. behavior. which encompasses a never-ending chain of human actions. entails an infinite number of individual actions based on individual knowledge of time and place. the various roles that entrepreneurs take on within the economic system can be seen as providing wealth-maximizing services rather than as detractions from other productive activities. He cites as examples: (1) guarantees. Social interaction. by continually updating . With wants. including markets. practices. Real estate transactors. can never be molded from a specific point in time. Markets incorporate a dynamic and evolutionary process whereby established standards. relocation services have emerged. including means of maintaining social control. It is this knowledge deficiency on the parts of both buyer and seller that makes middlemen (who are generally lower cost producers of information than consumers) a vital resource in market exchanges. and tools deemed to be the most efficient means of facilitating exchange today. Akerlof points to private institutional forces as a means of assuring quality.222 Mark Steckbeck and Peter Boettke attributes of these goods or services. The role of the entrepreneur is. maintains a fiduciary responsibility to the seller. providing assurances to buyers and sellers of real estate. projected to carry forth into the future.8 Real estate. must rely on the recommendations of a local real estate agent who. most people engage in real estate transactions infrequently. In response. for example. by law. therefore. markets are more robust than how they are modeled in Akerlof’s lemons model. once the diffusion of knowledge in society is recognized. and generally to act in the best interest of the referred client.9 These relocation companies act as monitors in a principal–agent relationship. agents and their brokers agree to meet specific requirements. the relocation service is remunerated for their matching and monitoring service through the commission received by the selling agent. provide enumerated services. provide full disclosure statements. Contrary to Akerlof’s premise. To be considered for referrals. not of intentional deceit. (3) chains. Lack of knowledge by both seller and buyer is as common in transactions as asymmetric information. Social ordering. (2) brand-name goods. and tools subsequently replace them. are deemed inefficient tomorrow. practices.

involves itself less in the exchange process than does Alibris. payment services (Abebooks and Alibris only). and then providing a satisfaction guarantee. handling all transaction processes. many of these fears being unfounded. with networked booksellers sharing common pooled resources.12 and Alibris. The used. especially as they pertain to Internet markets. the largest network in terms of number of member merchants and sales. employs a distinctly different means of assuring the outcome of the transaction. verifying its quality relative to the merchant’s description. Alibris provides the greatest level of direct assurance. To illustrate the robust nature of markets with regard to ameliorating potential problems in realizing the gains from exchange. Abebooks. these booksellers would not be able to take advantage of the economies of scale in the production of search and. Balancing consumer wants in order to ascertain efficiency can be achieved only through a trial and error process. fully intervening in the exchange process. however. Acting independently. we look at the market for used and out of print books that has emerged on the Internet. to discover more efficient means of promoting human interaction. The three networks provide the same basic matching service pairing sellers of used and out of print books with prospective purchasers. not of human design” (Hayek 1967). Abebooks .10 The used and out of print book market is a specialty market within the broader market reflecting our literary culture. out of print. thus facilitating exchange. The market and all of its rules are products of “human action. Fears about stolen credit cards or other personal information abound. Abebooks. Consumers vary in their preferences. and Internet consumers are price conscious.13 network thousands of booksellers from around the world. Each.000 used and out of print booksellers from around the world belonging to one or more of the three major used book networks operating on the Internet. Books sought by Alibris customers are purchased outright from individual merchants.11 Bibliofind. time preferences vary. No single individual can acquire the knowledge of time and place known only to each and every individual transacting in the market in order to enumerate specific rules and procedures of market behavior. and antiquarian book market provides an appropriate example of the dynamism of Internet markets. There are approximately 10. nor would they be able to develop the credibility these networks are able to assure. to a lesser extent. taking possession of each book. Entrepreneurial solutions in e-commerce 223 information. Applications to e-commerce The Internet market for used books It has been estimated that domestic sales of new books will reach $38 billion by 2004. The entrepreneur’s job is to use local knowledge to determine how best to meet the needs of book consumers and merchants. up from $23 billion in 1994. most notably a search engine.14 consumers seek different assurances about product quality and merchant credibility.

the buyer and seller are left to mediate their own disputes. They do not involve themselves in mediating disputes. Alibris. leaving the buyer and member bookseller to negotiate payment on their own. No formal planning existed specifying the safest and most efficient means of matching used and out of print booksellers with potential book buyers. and out of print books. devising a marketplace where such exchanges can be con- summated. Should a dispute arise. offering only to guarantee the satisfaction of a book if it was ordered directly through Abebooks using its payment acceptance service. employs a different strategy to foster reputable dealings by its individual members and provide quality assurances. The incentive to bring booksellers and buyers together exists. Given the recent innovation of the Internet as an exchange medium. Bibliofind no longer accepts payments. an outcome highly unlikely were the process to be regulated by actors outside of the context of particular time and place. there exists little historical knowledge specifying which practices lead to the most effective and most efficient means of facilitating exchange. members share a common good that would be prohibitively costly and less effective to produce if acting independently. In fact. This incentive invokes creativity and entrepreneurship. These used and out of print bookseller networks regulate members’ behavior by enumerating specific rules. By being part of such a network. It is. rare. impossible to . bringing sellers and potential buyers of used and out of print books together to negotiate prices and shipping. Bibliofind. monitoring their compliance to such rules. the largest Internet seller of new books.224 Mark Steckbeck and Peter Boettke provides primarily matching services through their search engine while also offering payment remittance services for member booksellers. differentiated by their levels of intervention. Bibliofind offers matching services only.15 As more information is acquired and processed. has merged its services with Amazon. thus eliciting trust among its customers. These three used book networks have one objective: to facilitate the exchange of used. however. Finally. Each. What has emerged is the fascinating aspect of network markets. The value of the network to member booksellers is dependent on customers’ trust in dealing with members of the network. Bibliofind offers no assurances of the quality of books sold by member booksellers through its web site although they will take action to remove a bookseller if Bibliofind receives an unspecified number of complaints or if they fail to adhere to Bibliofind’s enumerated policies. the incentive is to improve continually the exchange process. but it is also imperative to elicit cooperation between sellers and buyers in order to achieve success. the three networks have all devised different means of effectuating transactions based on their level of intermediation in the exchange process. has fewer sales and fewer member booksellers than Alibris or Bibliofind. the network intervening the least in the exchange process. of course. which intervenes the most in the exchange process. enhancing successful practices and scrapping unsuccessful and sanctioning defectors. Each network employs different practices for assuring the reputation and credibility of both the network and its individual members. a result certainly attributable to higher book prices and lower fees paid to booksellers.

as well as feedback information provided by the seller’s previous exchange participants about the seller’s credibility. sellers utilize eBay’s web site bidding on desired goods in predominantly second-price auctions. and over the Internet in particular. maintaining a hands-off approach in mediating the exchange process. a cost compelling both buyers and sellers to cooperate. Prospective buyers use this information. and a founded general threat that computer hackers will unlawfully access credit card information stored on a web merchant’s database. A consumer encountering fraud or deceit can dispute related charges to the issuing bank. Sellers list their wares on eBay’s web site including descriptions of their goods – in terms of quantity and quality. and their shipping policies. their terms of trade. it is estimated they make up about 50 percent of all credit card fraud. It is only through trial and error and experimental practices entrepreneurs undertake in their quest for the greatest return on the private capital they risk that such determinations can be made. the product information provided by individual sellers. eBay takes no role in assuring the satisfaction of an exchange. Those skillful or fortunate enough to make the correct decisions are rewarded through increased profits. settling transactions.17 Using a credit card to make purchases generally. This does not imply that all exchanges. or remedying inequities. As previously stated.16 Although transactions conducted over the Internet make up just 2 percent of all purchases made by credit card. sellers who have acquired negative feedback in prior transactions receive lower prices for goods they subsequently sell on eBay. adds a layer of security for consumers. resolving disputes. but exchange participants are provided relevant infor- mation allowing them to discount. incorrect or inefficient decisions result in losses and being forced from the market (Alchian 1950). As Steckbeck (2001) finds. the dominant Internet auction site. Internet auction sites eBay. if needed. performs a service similar to that of the used bookseller networks. Credit card use on the Internet Another characteristic of electronic commerce underscoring trust and assur- ance problems is the security of credit card information when transactions are consummated over the Internet. Having determined their willingness to pay. eBay matches buyers and sellers of new and used merchandise. Entrepreneurial solutions in e-commerce 225 make such determinations ex ante. If the exchange good fails to . to assess their offer price for the seller’s exchange good(s). The information sharing employed by eBay enables sellers and potential buyers to assess each other’s integrity based on each other’s exchange histories. thus protecting him or her from charges paid to unscrupulous merchants. consummated or not. consumers have two general fears about submitting credit card information over the Internet: an unfounded fear that credit card information will be stolen in transit. result in completely satisfied participants or that deception does not occur on eBay.

Additionally. Responses are situation dependent and most . inadequacies in their security measures. have jointly developed Secure Electronic Transaction (SET). For no other reason than the securing of existing profits. either online or off. they devise and implement new practices to protect consumers and merchants from fraud. There is no single entity responsible for the practices implemented to abate credit card. or if the quality is less than what was conveyed at the time of purchase. responding to the theft of credit cardholder information from merchants’ web sites. Conclusion Individuals respond to incentives and different situations provide different incentives. experiment with diverse measures. Disputed charges are then arbitrated between the card-issuing bank and the merchant’s bank. Developing precise rules or programs to rectify all problems occurring in all situations a priori is not feasible. employing digital certificates used to accurately identify all parties to an exchange. attempting to discover the least-cost means of providing the most security for transacting over the Internet. both parties have devised a proprietary encryption technology for transferring credit card data over the Internet. they are the most adversely affected parties if credit card fraud reduces the use of credit cards for transacting online. It is also the case that paying for goods and services over the Internet by credit card reduces transaction costs and may eliminate delays in the exchange process. Various firms.19 Consumer demands for assurances to protect their privacy and security is the impetus behind these programs developed by credit card companies. for example. this added layer of security and convenience becomes defunct resulting in a financial loss to credit card companies. is requiring merchants to meet security guidelines to maintain their privilege of accepting Visa credit cards. and continue to discover.18 Credit card information unsecured on a merchant’s web site is therefore rendered useless.226 Mark Steckbeck and Peter Boettke be delivered. Visa and Mastercard. attempting to use their specific knowledge of time and place. do credit card companies devise and implement newer and more efficient security measures. As credit card companies discovered. or the acquisition of additional profits. Credit card companies have great incentives to provide online security as they gain from the use of credit cards as a means of payment for Internet purchases. Credit card companies have responded to online credit card fraud both as a means of assuaging consumers’ and merchants’ fears – thus promoting credit card usage – and to protect themselves from loss. especially delays in the shipping process. the purchaser can simply refuse to make payment on the disputed charge(s). Visa. Should credit card usage pose a security problem to either consumers or merchants. Other examples of credit card companies devising security measures include American Express providing its cardholders the ability to obtain randomly generated disposable account numbers for making transactions that expire after each use.

formal legal institutions are largely unavailable for monitoring behavior. Akerlof’s lemons model suggests the fragility of market arrangement in the face of informational asymmetries and thus the problems identified should be prevalent in Internet markets. Contrary to observing the collapse of markets under such circumstances. the apparent problems in these markets today are simply tomorrow’s profit opportunities as actors possess strong pecuniary incentives to adjust their behavior to ameliorate these problems and realize the gains from exchange. including assuring cooperation in Internet markets. perfect competition may indeed imply allocative efficiency. Entrepreneurial solutions in e-commerce 227 effective when left to the people who possess the knowledge of time and place to adjust on the margin to changes in the situation. the fact that our inadequate knowledge of the available commodities or services is made up for by our experience with the persons or firms supplying them – that competition is in a large measure competition for reputation or good will – is one of the most important facts which enables us to solve our daily problems. we can expect to provide the most satis- factory solution for whatever particular personal problem we have to face. and effectuating remedy. Businessmen have attempted to advertise their trustworthiness as much as their products to potential customers throughout the history of commerce (see Mueller 1999: 21–56. Decentralized exchange provides incentives and generates the knowledge required for economic actors to coordinate the plans of the most willing suppliers and most willing demanders so that the mutual gains from exchange are realized. We have argued that Akerlof’s “lemons” model has indeed identified some of the ways in which markets may stumble on infor- mational grounds. But underlying this process of coordination is also the various ways in which people elicit promise keeping from one another (see Klein 1997 and 2000). The function of competition is here precisely to teach us who will serve us well: which grocer or travel agency. profit incentives direct entrepreneurs toward devising solutions to overcome such problems. we have seen how various . which department store or hotel. the argument for free competition in economic affairs does not rest on the conditions defining perfect competition. 72–98). In particular.” Hayek argued. (1948: 97) The market for assurance and trust has existed throughout the history of known markets. Actors compete on a multitude of margins to satisfy the demands of consumers in order to secure profits. As Hayek (1948: 104) taught. Competition from this Hayekian perspective is an activity. Contrary to Akerlof and others. “In actual life. especially with respect to the Internet. which doctor or solicitor. but that Hayek’s theory of competitive processes highlights the resilience of the spontaneous institutions of a voluntary society that assure trust within complex and uncertain conditions. But as we have shown. In addition. not a state of affairs. but it does not imply that competition will fail to bring about the most effective adjustment of plans and use of resources in situations where the nature of the situation is imperfect. mediating disputes.

available at: http://www. and since individuals realize that without trust markets inevitably collapse. It could be argued that Hollis’ premise describes “altruistic” behavior.html. out of print and antiquarian book market.abebooks. Also see Landa (1994) for an exploration of the role of traders in regimes of contract uncertainty. Abebooks began as a brick and mortar store in Victoria. See Shearmur and Klein (1997). Available at: http://www.shtml?story=b1031090. 2 Ibid. 3 See for example. the largest relocation firm in the US.emarketer. and antiquarian books. out of print. the latter two providing such relocation services to their policyholders or account holders. If a customer seeks a certain book and the owner does not have it in stock. 4 “eAuctions bid up online revenue. cit.” eMarketer. Notes Financial assistance from the Oloffson/Weaver Fellowship at George Mason University is gratefully acknowledged. yet we have provided suggestive evidence that an array of Hayek-type solutions emerge to self-police market participants and spread trust and civility in economic affairs. Consumers face high search costs and if a bookseller is able to lower those costs of search by matching buyers with sellers. we have taken a market which on the face of it seems to be ripe for Akerlof frames/story. 6 Hollis contends that since trust promotes market exchange. 7 On the role of middlemen and speculators within a market economy see Heyne (1994: 173–92). com/body_free_newsroom. individuals tend to cooperate for communal reasons. the exchange goes unrealized. By focusing on the Internet market. This figure does not include the used. ActivMedia Research available at: http://www. The problem the company faced was the same all independent booksellers face in this market. specializing in used. Schumpeter’s focus on the “creative destruction” role of entrepreneurship in terms of innovation within the economic system is “externalist.html. and the Navy Federal Credit Union. they could realize gains from the .” Ludwig von Mises (1966: 143–76) argued that social cooperation emerges through competitive processes through which men learned how to act in their enlightened self-interest.” On these two perspectives on entrepreneurship see Kirzner (1999). 1 Akerlof (1970: 488). a home- owner’s insurance company with military affiliations. Adam Smith’s discussion of the doux commerce thesis postulates a reputational mechanism that brings self-interest in line with social cooperation. 10 See the Book Industry Study Group press release on trends within in the industry at 8 Kirzner (1973) is the classic reference on the positive “internalist” role of entre- preneurship in the economics but that would rely on a very myopic view of “self-interest. USAA.bisg.activmediaresearch. 11 http://www.html.801&level3=2848&date=20001101&inIssue= eMarketer has also published electronic commerce data. regardless of the costs they internalize from cooperating.individual. BC. both historical and projections. 17 October 2000. op. 9 Examples include Cendant.228 Mark Steckbeck and Peter Boettke practices and institutions can serve to ameliorate the problems of asymmetric information about product quality and discipline the behavior of sellers to minimize their ability to cheat their customers. 5 ActivMedia Research.

note 14. pp. M. 18 Trombly. available at: http://www. 96–105. Akerlof. 17 November 2000. C. Hayek. 2 April 1999.A. (1950) “Uncertainty. “Visa to require e-security rules. Available at: http://www. Rather. (1960) “The problem of social cost. Available www.bibliofind.” Computerworld. 12 http://www. see Markoff (2000).00.. (1948) Individualism and Economic Order. New York: Cambridge University Press.000 customers. M.A.00. appearing to steal credit card and other personal information from the customer database. First. evolution.. (1967) “The results of human action but not of human design. Hayek.00. “American express offers disposable credit card numbers for online shopping. 14 Credit card fraud is by far a greater problem for merchants.1212. rumored to be a former Soviet bloc country).18904. E. References Abreu.” Journal of Law and Economics 1. (1994) The Economic Way of Thinking.html. Credit card information is rarely gleaned while in transit in cyberspace. Heyne.” R. A. Customers must now arrange for payment directly with the sellers whom they were matched with through the Bibliofind Entrepreneurial solutions in e-commerce 229 resulting exchange. 17 See Gold. Alchian. finding refuge in another country (assumed to be an Eastern European country. See Abreau (2001) for the story on Bibliofind.NAV47_S TO49788-. “Visa invests in online battle against Available at: http://www. Online. (2001) “Bibliofind closes its books after hack.” Journal of Political Economy 58(3): 211–21.1367.. cit. “Credit card fraud bedevils Chicago: University of Chicago Press. including the posting of an endorsement by Forbes as one of the best sites to buy on the internet.” Newsbytes.html. effectuating their transactions online. 1 March 2001. The second was a breach of Bibliofind’s web site we have already discussed where hackers had been lurking undetected for about 5 months. a hacker gained access to the web site of CDNow (http://www. Bibliofind has ceased accepting payments from customers after a computer hacker compromised their security system. 15 The strategies invoked involve branding (Abebooks’ use of the Forbes endorsement and Bibliofind joining forces with Amazon. Englewood Now a subsidiary of Amazon.H.” in F. Hayek. available at: http://www. Chicago: University of Chicago Press. 19 Delio. Coase. S. S. The hacker attempted to extort money from the web merchant and has been able to remain anonymous to US authorities.html). it is generally taken from merchant’s servers who have failed to adequately protect their sites from hacking (see Gold. F. and economic theory. not consumers (see Studies in Philosophy. Politics and Economics. F.thestandard. op. .com/news/business/0.computerworld. Abebooks relies on various techniques to signal quality to potential consumers.wired.” The Industry Standard (March 6) stealing credit card and address information of over and guarantees (privacy and consumer satisfaction guarantees are explicit on Alibris).alibris.A.newsbytes. 13 http://www. 16 Two cases have received much publicity during the past year.1294.38655. NJ: Prentice Hall. (1984) An Economic Theorist’s Book of Tales. G. Abebooks created a website where customers can search the inventory of thousands of booksellers.” Wired. 7 September 2000. P. On Alibris’s ability to raise $30 million in venture capital funds in 2000 see Landry (2000).

M. Klein.nytimes.) Reputation: Studies in the Voluntary Elicitation of Good Conduct. NY: The Foundation for Economic Education.” in Daniel Klein (ed. J. reputation and trust. Ann Arbor. MI: University of Michigan Press. I. Landry. (1994) Trust. (1999) Capitalism. D. Online. (1973) Competition and Entrepreneurship. NJ: Princeton University Press. MI: University of Michigan Press. (1998) Trust Within Reason. thesis. D.” The New York Times (January 14). Chicago: Henry Regnery. by voluntary means. (2001) “The spontaneous emergence of cooperation: An empirical analysis of eBay. (1997) “Knowledge. (2000) Assurance and Trust in a Great Society. Mises. I. Online. John (2000) “An online extortion plot results in release of credit card data. Steckbeck. Kirzner. Ethnicity.” Ph. and Identity: Beyond the New Institutional Economics of Ethnic Trading Networks. New York: Cambridge University Press. Democracy and Ralph’s Pretty Good Grocery.D. . (2000) “Albris books $30 million for literature. Contract Law.redherring. Irvington-on-Hudson.” Review of Austrian Economics 11(1–2): 5–18.html. (1999) “Creativity and/or alertness: A reconsideration of the Schumpeterian entrepreneur. Available www. M. J. Ann Arbor. Princeton. 1–14. Landa. Department of Economics. Chicago: University of Chicago Press. (1966) Human Action: A Treatise on Economics. L. J.” Red Herring (April 5).230 Mark Steckbeck and Peter Boettke Hollis. George Mason University. Klein. and Gift-Exchange. Markoff. pp. Available www.

we conduct such tests on wheat futures. As we explain below. Big players induce herding in asset markets. Its founder. our study gives some empirical support to the Austrian suggestion that activist government policies may extract a higher price in the new economy than in the old economy. (Popperian methodologists would say the theory is ‘not falsified’. societies pay an ever-higher price for activist government policies. Because big players are hard to predict. political activism tends to frustrate and corrupt the market processes producing and distributing knowledge. The new economy Transaction costs are lower in the new economy The most characteristic feature of the new economy may be its low transaction costs. In financial markets. Test results fit the predictions of the big players theory. The theory of big players clarifies one mechanism for the bad consequences of government activism. Big players are privileged actors who employ discretion in the exercise of their power.) Our study thus gives theoretical and empirical support to Austrian doubts about activism in government policy. Austrian oral tradition warns us that the new economy needs a new commitment from governments to choose rules over discretion. Austrians should translate their oral tradition into testable theories about the use of knowledge in society. Using R/S analysis. In the oral tradition of the Austrian school. as knowledge grows in importance.11 Big players in the ‘new economy’ Roger Koppl and Ivo Sarjanovic Introduction The ‘new economy’ has drawn popular attention to the vital role of knowledge in economic affairs. they create ignorance and uncertainty in the markets they affect. It has not yet been tested in the commodities markets. Advances in information technology are an important cause of reduced . Carl Menger. The theory of big players helps to explain herding in financial markets. this ignorance and uncertainty encourages herding. placed human knowledge and its growth at the centre of his theory. The theory of ‘big players’ is an example. The Austrian tradition of economic thought has long recognized the crucial role that knowledge production and distribution play in economic events. Thus.

people produced goods. but of utility’ (1971: 62).’ When workers specialize.’ He elaborates on the point by noting that the invention of useful machines ‘seems to have been originally owing to the division of labour. each of which affords occupation to a peculiar tribe or class . it is subdivided into a great number of different branches. and the like’ can be classed as ‘goods’ in economic theory if they ‘are of such a kind that we can dispose of them’ (1981: 55). As the division of labour progresses. Like every other employment too. This shift is consistent with the reduced costs of information. Without a plan to guide production. We believe this distinction is mistaken. the salient features of production may be the organization of information. they ‘are much more likely to discover easier and readier methods of attaining any object’. Adam Smith listed three reasons for the increase of wealth produced by the division of labour. Menger’s four ‘prerequisites’ of ‘goods-character’ include ‘Human knowledge of [the] causal connection’ between a thing and ‘a human need’ (1981: 52). ‘Production is the creation.232 Roger Koppl and Ivo Sarjanovic transaction costs. not of matter. therefore. Carl Menger and Adam Smith wrote about the central role of knowledge in social and economic life. Knowledge is important in the new economy. be commensurate with the degree of progress of human knowledge’ (1981: 74). customer good-will and the like’ are certainly goods (1981: 54). discussions of it might be found in the history of economic thought. they organize information. In the ‘new economy’. no increase in utilities can be expected. For Carl Menger. As Say argued. the salient features of production may have been physical exertions and transformations of matter. ‘monopolies. Speculation becomes a specialism. But the intellectual dimension was always present. however. they organized matter. religious fellowship. the system ramifies. even the division of labour must take a back seat to progress in knowledge (1981: 73). For Menger. something radically new. people produce knowledge. In the past. A noteworthy shift has occurred. It was also important in the old economy. A firm’s know-how and its place in an industry network may account for a larger fraction of its value than was typical in the past. He criticized ‘the materialistic bias of [his] time which regards only materials and forces (tangible objects and labor services) as things and. ‘Nothing is more certain than that the degree of economic progress of mankind will still. It has caused an increase in the average ratio of intangible assets to tangible asset. we are told. in future epochs. copyrights. Knowledge is the key for Menger. In the ‘old economy’. But the material dimension is still present. The new economy was familiar to Menger and Smith If the new economy is not new. Before Menger was Smith. founder of the Austrian school. It is not. The new economy has merely shifted the balance a bit more toward knowledge. The facts are sometimes exaggerated. The third was that ‘labour is facilitated and abridged by the application of proper machinery. ‘Even relationships of friendship and love. Now. Lower transaction costs have led to an increase in the relative value of knowledge. also as goods’ (1981: 54).

more work is done upon the whole. . Lower transaction costs imply fewer idle resources. If the new economy is more dynamic. in the new economy. Widespread and persistent unemployment is less likely. resources would be allocated efficiently. and the quantity of science is considerably increased by it. It is easier to coordinate the actions of many dispersed actors whose cooperation is needed to launch the new idea. The overall implication seems to be that the system will be more dynamic. and saves time. or to direct investment. If our accounting is about right. and this subdivision of employment in philosophy. thus. The second tendency is somewhat inconsistent with the first. regardless of the distribution of property rights. The new economy is more dynamic than the old economy. The growth of knowledge is a necessary part of economic progress. What difference does the new economy make? Transaction costs are low in the new economy. Inventories tax our patience. Reduced inventories also encourage financial intermediation. then there will be more innovations to be discouraged by big player activity. they tie up value over time. First. however. the value of the innovations lost to discretionary policy will be greater. Advertising is cheaper. (Smith 1937: 9) The division of labour creates a division of knowledge. two implications seem to follow. Discretionary policies will throw fewer people out of work. Big players in the new economy 233 of philosophers. Lower information costs make it easier for entrepreneurs to receive and process information. It leads to greater participation in financial markets by ordinary households. Partly. if the new economy will have more coordination or less coordination than the old economy. low transaction costs should lead to a better approximation to general economic equilibrium. Because financial intermediation is cheaper. it is easier to get the word out. Lower transac- tion costs encourage innovation. Each individual becomes more expert in his own peculiar branch. They also encourage innovation. The value previously tied up in inventories can be redirected to the purchase of financial assets. the value lost to discoordination of plans caused by discretionary policy will be less. Assuming away innovation. This fact produces two competing tendencies. Prudent inventory levels are lower. but they will strangle more innovations that are useful. 1 If transaction costs were zero. The more sophisticated methods of calculation made possible by lower costs of calculation allow entrepreneurs to make more precise and reasonable estimates of profitability. These superior estimates reduce the uncertainty entrepreneurs must bear and thus encourage enterprise. Searching more of the market and searching known parts more minutely increases the chance of seeing an opportunity. Second. more of it occurs. improves dexterity. this increase takes the form of financial innovations such as deriv- atives. It probably cannot be decided. The tendency toward general equilibrium follows from the Coase theorem. as well as in every other business.

In the short run. In the presence of a big player the market process changes. a private actor (such as a fund) may show big player features. prices will no longer accurately reflect market fundamentals that were previously being generated by decentralized. but the market process under the conditions described would be orderly and error correcting. Whether the filter works well or not is an empirical question.2 But we can make an indirect test. The big player’s actions short-circuit. Now. The filter of profit and loss is corrupted when big players derange markets. A big player’s actions will affect market signals. (3) Big players are discretionary. The theory of big players A big player is defined by Koppl and Yeager (1996) as ‘anyone who habitually exercises discretionary power to influence the market while himself remaining wholly or largely immune from the discipline of profit and loss’. to the theory of big players. Big players are actors who combine three features. however. This approach gives us a kind of proxy for a more direct test. We turn. Nor do we have a useful measure of our distance from general equilibrium. We cannot observe the innovations that do not occur because of big player influence. Markets then would generate prices and other signals that tend to reflect underlying economic fundamentals. A paradigmatic case is a central bank (even if independent of political influences) not bounded by a general rule. therefore.234 Roger Koppl and Ivo Sarjanovic Our conjectures about the new economy and discretionary policy are probably not subject to direct test. We do not need to assume the attainment of fully coordinated outcomes. In the long run. Big players can disrupt order in the market process. In the absence of big players and given the appropriate institutional frame- work. self-interested entrepreneurs seeking to discover unexploited arbitrage and speculative opportunities. When they use discretion. at . it operates without discretion. if the filter works badly. (1) Big players are big because they have the power to influence the market(s) where they operate. it is difficult for a private actor to challenge the profit and loss screening mechanism. They are funded by a pool of fiscal resources. (2) Big players are insensitive to profit and loss. stable constitutional rules and competition are filtering conditions that promote the efficient production and distribution of knowledge. We can test the claim that the discretionary policies of ‘big players’ encourage herding in financial markets by creating ignorance and uncertainty. On the other hand an orthodox currency board would the case of a ‘ruled’ body. government entities are typical big players. They do not follow rules but they do not follow any strategy of profit maximization either. If empirical studies give strength to the idea that big players create ignorance and uncertainty. prices may wander freely from fundamentals values. Hence. Well-defined and enforceable property rights. then the general considerations we offered above would suggest that discretionary policy does indeed extract a higher price in the new economy than in the old economy. markets would be expected to exhibit tendencies towards coordinated outcomes.

With big players.d. Sometimes these cases are empirically evident. a price) itself becomes distorted. The big players theory has been already applied with success to financial markets. where Pt is the asset’s price. Koppl and Nardone (2001) and Broussard and Koppl (1999) apply different statistical techniques to Yeager’s data.g. a highly politicized market. apply for the first time this theory to commodities futures market prices. Following Parke. Big players may also induce or increase long memory. Big players in the new economy 235 least partially. Each error is a price movement induced by news that does . The model will produce long memory if the probability of an error enduring k periods. but that. Gilanshah and Koppl (forthcoming) study US money demand from 1950 to 1990. The bubble can be modelled as an error duration process. Koppl and Yeager (1996) study an important episode of Russian nineteenth- century monetary history using data gathered by Yeager. Other cases are more difficult to isolate empirically.i. Big players increase the survival probabilities of the errors and decrease their rate of decline. shocks having mean zero and common variance σ2. Herding and contra-herding as error duration Koppl (2002: 135–8) argues that Parke’s ‘error duration’ model (1999) gives us a simple model of the herding and contra-herding produced by big players in asset markets. big players increase volatility. Koppl models bubbles as error duration processes (2002: 135). When autocorrelations exist. (1997) have studied the behaviour of closed-end countries funds in the late 1980s. Bt is a bubble and εt is an i.2. Finally we. error term. in the rest of this chapter. the epistemic value of market signals is diminished. . their rate of decline is determined by partial sums of the survival probabilities. . t = 1. The values pk are called ‘survival probabilities’. and herding. the advantages derived from the division of knowledge. is a sequence of shocks of stochastic magnitude and stochastic duration. such as the impact of certain fiscal decisions on interest rates. as when a central bank uses its discretion to move short-term interest rates or when it tries to defend an exchange rate. The price of an asset can be described by Pt = Ft + Bt + ε t . Ft is its fundamental value. What matters in all cases is not only that the big player’s action will affect in some direct way an economic variable. Koppl and Mramor (2003) study a recent episode in Slovenian monetary history. pk. . Contra-herding is consistent with this error duration model.3 In Parke’s error duration model. } be a series of i. .i. in addition. Ahmed et al. declines slowly as k grows. in this case wheat. Thus. Now prices will also reflect the big player’s influence. . Koppl explains. let {εt. noise stays in the system longer (2002: 135). The variable observed in a given period is the sum of those shocks that survive to that point’ (1999: 632). Thus. the quality of the affected signals (e.d. ‘the basic mechanism .

Thus. p1. Koppl posits two distinct sets of traders in the market. Noise traders behave like fundamentalist except that they respond to false signals ignored by fundamentalists. εt. As d grows. p2. The realization of the process is just the sum of the errors. by Day and Huang 1990. s =−∞ As Parke notes. the survival probabilities ‘are the fundamental parameters’ of the model (1999: 632). .s + k = 1). The indicator is one as long as the error endures.t. Some of them may be movements in excess of the level properly implied by the news to which fundamentalists respond.d. a sufficiently small (i. (If 1⁄2 ≤ d ≤ 1. pk = P(gs. large. Through its effect on noise traders. He shows (Parke 1999: 632–3) that if autocovariances exist. and a short memory if d ≤ 0. negative) value for d will make any bubbles difficult to detect. Thus. The probability of an error enduring at least k periods is pk. where ∞ λ = ∑ pi . The process has a long memory for 0 < d ≤ 1⁄2.t denote ‘an indicator function for the event that error ∈s survives to period t’ (1999: 632). i=1 In this case. the variance is σ2 (1 + λ). t Bt = ∑ g s . each false signal induces a price change of ∈t.t are independent for t ≥s. In that case λ and the λk’s exist.236 Roger Koppl and Ivo Sarjanovic not convey new information about the asset’s fundamental value. p0 = 1 and the series {p0. j= k If λ exists. pk will be close to zero even for small k. The error disappears when noise traders revise their initial interpretation of the false signal. . . they are given by ∞ γ k = σ 2 ∑ p j. pk > 1. Following Parke. the first-order autocorrelation is ρ1 = λ/(1 + λ). } is monotone non-increasing.t ∈s . switches from one to zero after period s + ns and stays at zero thereafter. Each error ∈s lasts ns periods beyond s. namely fundamentalists and noise traders (2002: 136). The indicator. λ and the λk’s do not exist. where ns is a random variable. Assume the survival probabilities n→∞ k=1 have the form pk=k–2+2d with d ≤ 1⁄2. Clearly. Assume ∈s and gs.) The fundamentalists correctly estimate Ft up to an i. for instance. let gs. gs. error.i. . If d > 1. (This common assumption is used.) In this model. The process has a long n memory if lim Σ kpk is infinite (Parke 1999: 633). These errors are distinct from the εt.e.

but most of it. Big players may also cause and increase in σ2. However. The increase in d will cause an increase in λ and thus in σ2 (1 + λ). tariffs and quotas affecting the commercial flows. the bubble process is stationary with a long memory. however. big players cause an increase in volatility. Assume ∈t is the sum of several independent random variables each corresponding to a process unrelated to the asset’s fundamental value. Big players may increase the number of unrelated processes noise traders respond to. Noise traders will thus take longer. The increase of d results from the greater ignorance of traders regarding the meaning of any piece of news. Wheat is the ‘energy’ input that our body needs to work properly. Others may expect a ‘correction’. The wheat market is the one most affected by the presence of players who are more or less immune to the tight mechanism of profit and loss. as bread or pasta. In that case. the variance of the ∈t will grow. in the non-bubble part of the asset’s price. Practically no agricultural market in the world is totally free of government interventions. In Koppl’s view. ut–1 error will be added to or subtracted from ∈t according to whether sheep or contrarians prevail. If the increase is from a value below 0 to one above 0. ut–1. he will cause it to increase. If 0 < d ≤ 1⁄2. In some places this role is also performed by rice. the variance of the ∈t. |∈t| = |ut–1|. is one of the main sources of carbohydrates in the world population’s diet. that the big player will cause d to increase from one value below 0 to another value that is still below 0. etc). wheat is a very political animal. Not all the wheat. Therefore. ∈t = ut–1. There are also regulations such as export and import licenses. OECD countries spent during the year 2000 US$ 370 billon in different kinds of subsidies (minimum prices for farmers. soft credits. ∈t = –ut–1. (More generally. If the contrarians prevail. In either event. It is perfectly possible. As a basic food ingredient. A larger d means larger survival probabilities. is used as a food grain. Wheat. a share of the world’s production is used as a feed grain. governments around the globe have been and are. If long memory is already present. to revise their interpretations of false signals. As an example. The increase in d raises the autocorrelations of the bubble process. the variance of the bubble process.4 This representation of the effects of big players is consistent with contra-herding according to Koppl (2002: 137–8). One of the false signals followed by noise traders may be last period’s movement. Big players in the new economy 237 bubbles are increasingly large and persistent. on average. export subsidies. very involved in guaranteeing on the supply side its availability via mainly minimum prices and . Big players cause or increase herding. If the sheep prevail. then the big player will produce long memory.) The winning interpretation will then endure for nt periods beyond t. Some noise traders will see a trend. but they need not cause long memory. The wheat market Our hypothesis is that the wheat world market was and is inhabited by big players of different sizes but that the big player influence fell after 1990. many of the effects of big players can be represented by an increase in d (2002: 137). Thus.

India (70 myn mt) and the USA (60 myn mt). the commission in Brussels grants to exporters a subsidy called restitution to cover the price difference between local and international prices. The USDA decided how to allocate these subsidies. trying to make it available to the main sectors of the population as cheaply as possible. The main exporters are USA (30 myn mt). The exporters estimate the subsidy needed to make possible the sale abroad. Then if a Moroccan buyer decided to shop in USA. Every Thursday. via subsidies. It was available from 1985 and it was used for the last time in 1994. Canada (14 myn mt).5 myn mt). Egypt (6. The USDA then decides if they grant the subsidy or not and if so for what volume. thanks mainly to the Uruguay Round of the GATT.238 Roger Koppl and Ivo Sarjanovic on the demand side. Let us assume that the free market price of USA wheat delivered to Morocco was US$ 150/mt on a cnf basis. subsidies have been decreasing and therefore the involvement of governments in export markets is less than before. Among the different export support programs promoted by the US government (PL-480. The world trade flow (exports and imports net of domestic utilization) is about 90 myn mt. all the other countries suffer different degrees of interventions in their markets. Other than Argentina. etc. which. how big the subsidy is going . The subsidy is the price difference between the domestic price and the export price. are used to make viable exports of European origin to world markets. specially monitoring the actions of its rival the European Commission who still follows a very similar approach (but less actively than in previous years). GSM. Most of the big producers used to subsidize production heavily via minimum prices. Donations. An international buyer checks prices with different USA exporters and one day decides to give a bid for a US commodity under EEP. The 1980s saw a heavy subsidies war to gain market share especially between the USA (through the USDA – United States Department of Agriculture) and the European Union (through the Agriculture Commission in Brussels). Argentina (10 myn mt) and the European Union (10 myn mt). which is a real free market.2 myn mt). But you do not know in advance if the USDA will decide to approve the subsidy in order to compete with the European wheat.) the Export Enhancement Program (EEP) was by far the most actively utilized and most popular one. Iran (7 myn mt). Among the main importers we can list Brazil (7. Australia (16 myn mt). This decision is absolutely discretionary and it is what makes the USDA a big player in the wheat market. they should give a bid to a USA private exporter who will ask the USDA for a subsidy of US$ 50/mt to make it competitive versus the European origin. In the last decade.5 myn mt). But Morocco was buying French wheat (subsidized) by the European Union at US$ 100/mt. Japan (6 myn mt) and Indonesia (3. The mechanism of the EEP was the following. once granted. European Union (90 myn mt). USA and Europe allocate these funds from their budgets to those sellers who request the smallest subsidy to make the operations feasible. Today. The world wheat production of about 500 million metric tons (denoted myn mt) is widely spread among different nations of the world. Trading houses bid for restitu- tions. These subsidies generate excessive supplies that are sold into world markets at parities that reflect a discount relative to domestic prices. the main producers are: China (100 myn mt).

they can accept 2 myn mt with subsidies ranging from US$ 20/mt to US$ 30/mt.000 mt at US$ 20/mt subsidy. But the final say on the sale is in USDA hands. for what amount. The European Union also has the power to apply ‘negative restitution’. in Europe this is not the case. have been rare in recent history. which is a fundamental element to determine price direction. . Some days nothing is allocated. Such occasions. the other major wheat exporters. the subsidy received needs to be used to sell to a specific destination. which in fact is an export tax if the world markets are quoting at higher values than the local European markets in order to ‘protect’ local availability and to ‘avoid inflationary pressures’. they want to avoid problems with the Argentine government. One morning you arrive at the office and the USDA announces that they have decided to accept bids for 1 million metric tons of wheat to Egypt at US$ 40/mt subsidy (in order to make it competitive versus French wheat). subject to the EEP allocation. when and at what final price affecting drastically the final stocks situation. the reduced role of government subsidies since about 1990 has marked a significant decrease in big player influence on wheat futures. and exporters can decide how to use them. The impact in the futures markets is the following. they receive direct support from their treasuries. On the other hand. for how long they will subsidize sales to that destination. Nevertheless. The European Union has full discretion to decide every Thursday if they are going to grant restitutions or not. These entities behave as price discrim- inators because they sell to different destinations at different levels following how the USA and Europe price their sales to those places. If the Canadian and Australian boards lose money. not the sellers. that same morning you could find the USDA declining bids for Brazil at US$ 20/mt subsidy because. (Argentina is the traditional wheat supplier of Brazil. Through the amount of subsidies the USDA can decide how much is going to be exported. Big players in the new economy 239 to be. On still other days. Other days they accept bids for. for how long. if they decide to get rid of a good proportion of their stocks. which sectors of the economy they will decide to stimulate. We can then say that with the sole exception of Argentina (its grain industry was also nationalized until 1976) most of the wheat supply in the world is subject to political forces and therefore not sensitive to tight economic feedback. for how much volume they are ready to do it. and if so for what amount. say. none of the other big players in the market showed any interest in selling cheap wheat to those destinations. and for what volume.) And surely. will need to hedge their sales buying futures in Chicago. have their exports monopolized by grain boards that buy all the national production from the farmers and sell it into the world market. unfortunately for them. etc. however. for political reasons. it would be similar to having an official bank managed by politicians who can decide at their discretion (although given the limit of a budget allocation) whether they will lend money cheaper than the market. You know shippers who sold that amount. Canada and Australia. 100. In the case of the USA. Tracing a parallel with financial markets. the same date you will have some buyers from central America buying wheat at free market prices because. there is a considerable degree of discretion. Thus.

The timing of the purchases is very different.1 plots wheat prices from 1970 to 2000. Brazil. The 1990s show a big difference with the 1970–80s market. A quick summary of the 1970s and 1980s shows that government boards made about 80 per cent of the purchases. (A value of 1000 corresponds to $1. Purchases are spread more smoothly over time. It is linked more with the margin prospects of the industry. The trader has a different risk perception of the different members of the industry instead of a common approach to country risk. industry margins. Today less than 50 per cent of the imports are still handled by public officials. governments were mainly in charge of buying wheat. These changes have had a big impact in terms of buying decisions.00 per bushel. . China. etc. Indonesia and partially Egypt are already in private hands. Buying decisions were mainly a ‘physical’ decision.) What are the hints that market players use to figure out the potential activities of the big players in the wheat market? Among others: their fiscal capacity to grant Figure 11. Among the biggest were the USSR.5 Figure 11. Pakistan. exposure. not an economic choice. This does not mean at all that agricultural markets are not distorted by subsidies (which are still very heavy and widespread utilized) but the presence of big players is less evident. before the privatization wave. Volume is not the only driver. The demand side is more private. And so on. Private flour millers were rarely involved in international transactions. Note: Vertical bars indicate the beginning and end of 1990. There were needs and they had to be filled to preserve social peace.240 Roger Koppl and Ivo Sarjanovic A similar reduction in big player influence has occurred on the demand side. but without any relation to the industry profitability. The USDA finished their EEP programme in 1994 and Europe intervenes less because with lower production subsidies they have lower export surpluses.1 Wheat prices from 5 January 1970 to 22 June 2000. the presence of big public buyers more limited. Purchases are much more price conscious. Traded lots are smaller because buyers are more atomized. which looks at a better risk management. Iran and Indonesia. Until the early 1990s. risk management criteria. All prices are for contracts traded on the Chicago Board of Trade. Egypt. whose data were omitted from the study. certainly as cheaply as possible so as not to waste public funds. Big markets like Brazil.

the domestic political situation of different countries trying to predict their prior- ities and urgencies (guns or butter?). Big players in the new economy 241 subsidies. beans are one stage higher than wheat in a Hayekian structure of production. public statements of the public officials in charge of the exports and imports programmes. perceived by governments as a problem. In other words. embargo decisions like the USA embargo towards the USSR (the biggest wheat buyer of the world at that moment) in the early 1980s following the Afghanistan invasion. Therefore wheat is a deeper political issue than soybeans. Dealing with big players is much more complex than trying to figure out when a price will be considered high enough to induce sellers to market their crops or make buyers reluctant to move or when a price will be perceived as cheap enough to induce farmers to hold stocks and buyers to look for coverage. etc. They cannot afford to trust in the invisible hand to guarantee its supply. The oil and meal relative contribution to the seeds’ value varies in time. as a source of protein. the political relationship between producer and consumer countries (via purchases importers could decide to strengthen relationships with a specific exporter regardless of relative prices). the soybean market is much more volatile than the wheat with respect to weather. Oil. how the budget is allocated among different programmes and by whom and when those decisions are made. We can say that given that meal should be fed to animals first and only after a certain period of time are animals ready to be consumed as food by humans. how the agricultural budget is discussed. requires in the minds of politicians a more centralized solution. The soybean market is very different from the wheat one in terms of political involvement. while bread is vital for people of low income. The main consequence of this fact is that weather concerns in the soybean market affect the CBOT futures market more directly. who is the person in charge of the grain boards. Only a very minor proportion of these oilseeds is used directly as human food. These facts explain why weather risks differ for wheat and soy. oil and meat are consumed mainly by people of relatively high income. The relatively low importance of weather in governing world wheat prices gives us a greater chance of isolating the effects of politics on the wheat market. Most of the beans grown in the world are crushed to convert them into by-products: they contain about 17 per cent crude oil and 80 per cent meal. wheat has two crops a year. once refined. a winter crop and a spring crop. Not all futures markets have had the big player influence of the wheat market. Other than that. trips to key exporter nations giving the hint that some deal will be negotiated. . Moreover. The soybean world production was basically located in the USA until the early 1980s and it is still concentrated in America: USA and South America. One big difference between these two markets is that the world produc- tion of wheat is much more decentralized geographically. but in general more than 60 per cent of the soybean’s value is ‘imputed’ from the meal leg. availability of soft credits. mainly in the Far East to prepare tofu and tempe. Wheat availability. is consumed directly by people and the meal is used to feed animals (mainly hogs and poultry). the political willingness of big producers to grant food donations.

Mandelbrot and Wallis 1968. persistent dependence is the failure of the error terms to die off geometrically or quicker. the range of the interval is .E. 1969b). The theory of big players predicts an increase in herding. Our statistical test follows. positive persistence.242 Roger Koppl and Ivo Sarjanovic The picture we have painted is one of a substantial big player influence in the wheat market. (1997). Hurst (Hurst et al. The method of R/S analysis has been explained in a number of big player studies including Ahmed et al. Persistent dependence in time-series data creates ‘aperiodic cycles’. ρ( k ) ≤ Cr − k . The theory of big players. The concept was invented by H. Statistically. The condition that the error terms die off at least geometrically is expressed in the following equation. Statistical methods and results Our review of the wheat market suggests that there was a decrease of big player influence after 1990. The cumulative sum of the series. Rescaled range analysis tests for ‘persistent dependence’ in time-series data. where C is positive and r is between zero and 1 (Brockwell and Davis 1991: 520). irregular ups and downs in the data that cannot be attributed to ‘short-period’ autocorrelation. Let us consider such a series {Xt} for which we have a sample of size T. In the case of interest to us. Koppl and Mramor (2003). But it is also a picture of reduced big player influence in the 1990s. then. X*(t). and Koppl and Yeager (1996). 1972. The use of R/S analysis as a test was developed by Mandelbrot and others (Mandelbrot 1971. predicts a reduction in the amount of herding after 1990. R/S analysis tests for persistence assuming the series is ‘stationary’ in the sense that the expected value of any function of {Xt} is the same for {Xt–k}. Short-memory processes are ‘ergodic’. 1965). For any interval of length s beginning at t. Long-memory processes are ‘nonergodic’. To test for herding we ran a rescaled range (R/S) analysis on wheat data from the Chicago Board of Trade. The essence of the technique is to see if the accumulations of the series under study wander around too much to have been produced by a random walk. is just the sum of the values up to t: t X * (t ) = ∑ X u u =1 where t ≥ 1. For t = 0 we define X*(t) = 0. 1969a. Gilanshah and Koppl (forthcoming). the series moves up and down in long waves that are inconsistent with short-memory processes.

s) is called the ‘rescaled range’. R(t.’ h > 1⁄2. 0≤u≤s To rescale the range.s).’ h< 1⁄2. This ratio grows with interval length. In the figure. s ) ~ Cs h where C > 0 and 0 < h < 1. of an interval is the vertical distance between Figure 11. if it has ‘positive persistence. Mandelbrot and Wallis (1969b) report that. Big players in the new economy 243 { [ R(t . The range. divide through by the sample standard deviation. If the series has ‘negative persistence. The Hurst coefficient. in the limit. R(t .2 Rescaled range analysis. h = 1⁄2. h. If the series is ergodic. In other words.s)/S(t. s. The ratio R(t. X* is plotted on the vertical axis and time is plotted on the horizontal analysis.2 illustrates R/S analysis. S(t. s ) = max X * (t + u) − X * (t ) − (u / s ) X * (t + s ) − X * (t ) ]} 0≤u≤s { [ ]} − min X * (t + u) − X * (t ) − (u / s ) X * (t + s ) − X * (t ) . s ) / S (t . Figure 11. the rescaled range is proportional to a power of s. is a measure of persistence. Finally. . The line showing the cumulative values of Xt goes up and down because {Xt} is stationary with some values positive and some negative. of the original series {Xt}.s).

The results of our test are reported in Table 11. to omit 1990 from our sample and look only at the period from 1970 to 1989 and the period from 1991 to 2000.244 Roger Koppl and Ivo Sarjanovic Table 11. therefore. closed-end country funds in the period 1989–90.1 Estimated values of the Hurst coefficient (standard errors are in parentheses) 1970–89 1991–2000 GH (10) 0.558 (0. (1997). We chose. .596 0. Using the ‘classic Hurst’. Conclusion Our results give empirical support to the idea that discretionary policy does induce ignorance and uncertainty and. We tested for persistent dependence in the return series created in this way. we estimated the persistence in our wheat data. US money demand in the post- war period. Our data were prices of wheat as set on the Chicago Board of Trade. the rescaled range would grow rapidly with interval length. As in these other studies. The rescaled range adjusts this figure to correct for the size of the variance of the original series over the interval. therefore. s.008) (0. our results are consistent with the theory of big players (see Ahmed et al. Our result is consistent with the theory of big players. Herding would create positive persistence. We started with each day’s closing price beginning with 5 January 1970 and ending with 22 June 2000. Given the importance of the USA as a world producer and exporter in the wheat market. the Chicago prices tend to be considered good proxies of world prices. We then calculated the return for each day except the first. Recall that 1990 was the transition year from a regime of relatively large big player influence to a regime of relatively small big player influence. Koppl 2002. herding in financial markets. and the Slovenian stock market of the transition period. Thus. Our reasons for preferring the classic Hurst to Lo’s technique are the same as those given in Ahmed et al. 1997. The return is the change from the previous day’s price over the previous day’s price.1. These results are consistent with our conjecture that big players frustrate innovation in the new economy. Previous studies have tested this claim in the foreign exchange market of Czarist Russia.008) two straight lines tangent to the cumulative series and parallel to the straight line connecting X*(t) and X*(t + s). The range just measures how much the cumulative series deviates from trend over an interval. they give some empirical support to the Austrian suggestion that activist government policies may extract a higher price in the new economy than in the old economy. Koppl and Mramor 2003).

3 Parke uses his model to represent persistence in the volatility of asset prices. London: Constable Publishing Company. References Ahmed. Managerial Finance 25: 49–63. Klaus Schredelseker and two anonymous referees for comments. 4 In this case. Mandelbrot.. H. and Koppl. important. The Coase theorem deals with only one type. Gilanshah.H. Koppl. R. R. and market sheep’. 2nd edn. Koppl.M. (1965) Long-Term Storage. In other words low prices no longer constrain supply. Review of Economics and Statistics 53: 225–36. Day. If the ‘free market’ price stays below the target level the price does not act as a signal for the farmer any more because he will get the target price anyway so the selling decision gets somehow disconnected from the final income they will get. Review of Austrian Economics 16: 253–69. Brockwell. Transaction costs are.J. Koppl. Rosser. J. D. . B. R. R. and Sinaika. R. L. Koppl. R. Koppl conjectures that it would not alter any of the conclusions about big player effects (2002: 220). (forthcoming) ‘Big players and money demand’. (1971) ‘When can price be arbitraged efficiently? A limit to the validity of the Random Walk and Martingale models’. Broussard. Backhouse (ed. Black. (1991) Time Series: Theory and Methods. and Nardone. C.E. however. C. Hurst. Dropping this assumption would complicate Parke’s error-duration model. J. 5 In USA the farmer still has a guaranteed minimum price (loan level) that could be higher than the free market value. Journal of Economic Behaviour and Organization 32: 19–37. (1999) ‘Big players and the Russian ruble: Explaining volatility dynamics’. R. 1 There are many causes preventing convergence to general equilibrium. New York and Berlin: Springer-Verlag. P. farmers receive the subsidy directly. E. and Davis. an Experimental Study. and Yeager.. M. New York: Routledge. and White. Their reduction tends to bring us closer to an imaginary state of perfect coordination described in general equilibrium theory. (2001) ‘The angular distribution of asset returns in delay space’. and Mramor. New York and London: Palgrave Macmillan. Koppl. 2 It is not even clear that the economy is further from equilibrium in the bust than in the boom if the boom was a false one induced by monetary growth. If the market does not reach the target level. and Koppl. (2003) ‘Big players in Slovenia’. Explorations in Economic History 33: 367–83. Big players in the new economy 245 Notes We thank Jack Birner. Discrete Dynamics in Nature and Society 6: 101–20. the several errors of a period would all endure for the same number of periods. bears. R. (1997) ‘Complex bubble persistence in closed-end country funds’.B.) Modern Applications of Austrian Economics. Journal of Economic Behavior and Organization 14: 299–329.A.B. Y. (1996) ‘Big players and herding in asset markets: The case of the Russian ruble’. the government reimburses directly to the farmer the difference between the free market price (sold to an exporter or a processor) and the target level so we can say that instead of an EEP allocation these days.. (2002) Big Players and the Economic Theory of Expectations. transaction costs. in J. (1990) ‘Bulls. although one might lump them all into this type by definition. R. and Huang W. as in the Austrian theory of the trade cycle. R.

R. and operational hydrology’. Mandelbrot. J. Kelley. (1968) ‘Noah. Smith. . and Wallis. (1981) Principles of Economics. or the Production. and Wallis. and Consumption of Wealth. J. Mandelbrot. New York: Modern Library. Dingwell and B. J. New York: New York University Press. Water Resources Research 4: 909–17.R. Prinsep. (1969b) ‘Robustness of the rescaled range R/S in the measurement of noncyclic long run statistical dependence’.F.246 Roger Koppl and Ivo Sarjanovic Mandelbrot. C. (1999) ‘What is fractional integration?’ Review of Economics and Statistics 8: 632–8. Menger. B.R. Parke. trans. Hoselitz. Annals of Economic and Social Measurement 1: 259–90. Water Resources Research 5: 321–40. A. Water Resources Research 5: 967–88. C. (1937) An Inquiry into the Nature and Causes of the Wealth of Nations. Say. B. Distribution. (1972) ‘Statistical methodology for nonperiodic cycles: From the covariance to R/S analysis’.R. J. Mandelbrot. Joseph. J. B. W. and Wallis. (1971) A Treatise on Political Economy.B. B.R. (1969a) ‘Some long-run properties of geophysical records’. New York: Augustus M. trans.

Part VI The monetary sector in the Internet .


In a Schumpeterian perspective. The central categories of a monetary analysis of this kind are asset valuations. based on a Misesian framework (Mises 1998: 535). Such an analysis. (Alan Greenspan1) The new economy in a Misesian framework Conventional views of the American boom during the 1990s may hold that it was primarily a technological phenomenon characterized by the breakthrough of innovations and their rapid dissemination. and. is in evidence in other cycles. the boom may be interpreted as the occurrence of one of the seminal clusters of inventions and innovations that put the economy on higher levels of productivity and growth. while interpretations along the lines of the real business cycle models may note that the booming economy was brought about by major technological advances that spread about the entire economy. In the Misesian perspective money is not neutral. would detect various elements in the boom that suggest that a monetary cycle has taken place. But while non-monetary explanations do reveal some aspects of economic behavior. and rarely. if ever.12 Bubble or new era? Monetary aspects of the new economy Antony P. All of these interpretations assume an impact that will transform the economy as a whole and make economic activity more efficient and finally more beneficial for consumers. credit and the interest rate. In a monetary economy. ignoring monetary conditions makes the analysis incomplete and often quite misleading. economic calculation is done in monetary terms based on profit and loss as they appear in accounting.2 If the monetary rate of interest falls below the natural rate. as future goods have become relatively cheaper demand for them increases. Mueller Every business cycle is the same with the exception of some fundamental difference that characterizes that particular cycle.” largely ignoring the financial and monetary factors. it will deviate from the original valuation between present and future goods. prices. changes in the supply . they concentrate on the “real economy. The starting point of Mises’ monetary theory is the proposition that the monetary rate of interest may deviate from the natural rate due to money creation (or its contraction) in the credit markets. profit and loss. Methodologically. The crucial issue is to identify what that particular phenomenon is.

250 Antony P. Mueller
of money come along with shifts of the individual wealth positions. By going beyond
the Wicksellian theory (Wicksell 1898) in two important aspects, Mises, first, applies
a strict subjective valuation approach by introducing the “originary rate of interest”
as the point of reference; and, second, applies “sequential analysis” in contrast to
the “all-at-once-adaptation” that characterizes most monetary models since David
Hume and is still the trademark of models in the tradition of the quantity theory
of money. In its originary form, the interest rate is the discount that human action
must give to later available goods compared to the earlier available goods, which
may render the same service. Human action implies categorically a preference for
the more immediate over the more remote in time. Otherwise, in an imaginary
world without an originary interest rate, saving would become infinite.3
Sequential analysis of monetary effects shows that changes in the money supply
affect economic agents heterogeneously. In a prolonged process of credit expansion4
fundamental differences occur in relative wealth positions depending on who
gets the newly created money first. Money cannot be neutral because it enters the
economy not at once or at the same time nor in the same quantities for all economic
agents alike. While money creation may or may not change the overall price level,
it will always change relative prices and with it the relative fortunes of individual
economic agents. Even if the change in the quantity of money could be known in
advance, and if it were known for which kind of activities it enters the economy,
it is impossible to know ex ante how this will affect the different prices later on. Only
perfect foresight could transform the monetary rate of interest into a neutral rate.
But it is principally impossible to foresee how, when and to what degree individual
valuations will change, and thus the formation of expectations about a certain
direction of prices is disparate and must remain uncertain.
This monetary theory based on individual valuation and sequential analysis leads
to a theory of the business cycle which holds that the expansion of circulating credit
brings about deviations of valuations due to the transmission of false signals as to
the availability of real funding in terms of sustainable savings, thereby causing a
misallocation between the production of immediate and future goods. Easy money
creates an illusion of wealth and thus instigates the enlargement of the production
process, while consumers may aspire for the acquisition of goods that rank higher
in their preference scale and that previously were out of reach. While such an
artificial boom may go on for a prolonged period of time and thereby irreversibly
transform the economy, it cannot be maintained in the long run5 because it is based
on money creation and the accumulation of debt in the presence of insufficient
savings. Then, disproportionalities6 occur within the economy, which later on
require reversals brought about by a recession or a severe slump depending on the
extension of the bubble. In the meantime, apparent wealth and economic and
technological progress may have been created temporarily. But given that real
savings are lacking, more pressing needs will emerge sooner or later that force the
abandonment of the ambitious economic projects.7
The monetary trade cycle, which begins with credit expansion of fiduciary
money within the fractional reserve bank system at home or abroad, or by outright
money creation or external debt accumulation by the government, changes the

Bubble or new era? 251
pattern of investment and consumption in an economy. It is a characteristic of a
monetary business cycle that growth is concentrated in specific sectors of the
economy where clustered investments occur. This primary effect of a monetary
business cycle tends to become more obscure in the course of time when second-
ary effects take place in the form of a spillover to other economic activities. It is
the primary effect of money creation, which makes for the interpretation of a
modernization of the economy, of an innovation boom or as “economic progress”
in general; and it is the secondary effects, which usually become regarded as
“the economic boom.” But real funding has been overextended in both cases, and
the new capital structure and consumption patterns cannot be maintained in the
long run. Economic actors have overestimated the permanence of their wealth
based on financial illusion.
When applying this framework to the American boom of the 1990s several
factors, which otherwise receive little attention, are put into the center of the
analysis. First among these ranks the phenomenon that the take-off took place
within an environment characterized by ample financial liquidity brought about
by loose monetary conditions in combination with the provision of a monetary
safety net encompassing domestic institutions and the international financial arena,
while all of this was accompanied by widening financial imbalances, particularly
in terms of falling private savings. With ample financial funding available and the
lowering of the risk perception, highly concentrated investment activity resulted,
predominantly in information technology. The apparent strength of the US
economy in these areas must be contrasted with the fact of increasing debt levels,
particularly in the corporate and foreign sector of the US economy.

A debt-driven boom
High real economic growth rates along with very low levels of unemployment and
inflation had been thought of as being incompatible based on concepts such as the
NAIRU. The apparently miraculous performance of the US economy during the
1990s instigated the search for alternative models of explanations with the “new
economy” paradigm becoming the most prominent. According to this model, the
world economy, led by the United States, has entered a new development stage
characterized by high productivity due to technological innovation. For monetary
policy, this view implied that the US economy would be capable of maintaining
much higher levels of real growth rates and employment than previously without
risking inflation. Productivity growth, even though it might not clearly show up in
all statistics, came to be seen – particularly within the Federal Reserve System and
on Wall Street – as the prime mover of the economic transformation in order to
explain the performance of the US economy in the second half of the 1990s, when
high real growth occurred in the absence of significant inflation rates (cf. Onliner
and Sichel 2000). For US monetary policy such a hypothesis opened the prospect
of lowering interest rates below levels that were previously thought of as a sure
way to cause inflation. But according to the new paradigm, the presumed existence
of a new economy represented the pursuit of a beneficial cycle, where fast

252 Antony P. Mueller
technological advances drive productivity enhancements that account for a non-
inflationary growth potential. Low interest rates then appear as conditional for high
new capital investments that in turn bring forth productivity growth. In a
paradoxical way, the conclusion said that the ample availability of financial funding
is causal for containing inflationary risks. A change of the monetary policy bias
towards restriction would abort this beneficial circle. Higher interest rates would
reduce capital formation, and lower capital investment would reduce productivity
growth. In this case, the economy would be at risk of inflationary tendencies. The
new guidelines for monetary policy suggested that money and credit must be made
amply available if the full fruits of the technological boom should be harvested.
For this interpretation, however, the problem remained that the new economy
did not show up convincingly in the statistics, and the technological innovations
appeared rather small compared to the great inventions of the past (Gordon 2000).
Only creative interpretations made it seem possible to diagnose a productivity jump
(Nordhaus 2001). Based on standard economic indicators and official figures, the
thesis of a new economy seemed hard to justify (see Table 12.1).
While the US real growth rates have been quite high (but not exceptionally so)
and inflation moderate (but in no way absent), remarkable productivity growth
does not show up in the official statistics despite exceptionally high investment rates.
Rather, the foremost difference of the 1990s’ cycle in contrast to the earlier ones
appears to be – besides a high level of equipment investment – the exorbitant
increase of debt levels in households, businesses, and in the external accounts. An
analysis of the flow of funds accounts (Godley and Izurieta 2001) supports the
diagnosis that unsustainability has been building up. Additional figures show that
in the 1990s, US credit markets expanded at an unprecedented pace (for data see
BIS 2000: 111). In the private sector, annual international and domestic debt
securities net issues increased sharply, amounting to around 500 billion US dollars
in 1995 and increasing to 1.2 trillion US dollars in 1999, dwarfing the respective

Table 12.1 US economic expansion 1991–9 in historical perspective

1991Q2– 1995Q4– 1983Q2– 1961Q1–
1999Q4 1999Q4 1990Q4 1969Q3

GDP volumea 3.6 4.4 4.3 5.0
Productivitya 2.1 2.6 1.7 2.9
GDP deflatora 1.9 1.6 3.3 2.6
Household debt/income 89.6 94.2 74.4 63.4
Business debt/outputb,c 75.3 76.3 70.6 54.9
Current account/GDPb –1.7 –2.3 –2.4 0.5
Equipment investment/GDPb,d 8.3 9.7 6.0 3.4

Source: National Data, Bank for International Settlements, 70th Annual Report, 2000: 14.
a Annual percentage changes.
b Average of period.
c Non-financial corporate sector.
d In volume terms.

Bubble or new era? 253
figures for Europe and Japan and historical standards. While the surplus of the
central government budget received high publicity it has been largely ignored that
the other public sector, i.e. independent and quasi-public entities of the United
States, issued debt instruments rising from around 300 billion US dollar in 1995
to almost 800 billion in 1999. Consolidating these figures for 1999, new net debt
issues by the private and public sector together reached two trillion US dollars in
that year. The trend continued in 2000 and 2001, when in the first half of 2001
government-sponsored enterprises like Fannie Mae and Freddie Mac8 issued
coupon securities and bills amounting to 590 billion US dollars (Federal Reserve
Board 2001: 15). Due to a level of internal absorption, which is surpassing domestic
production, the US trade deficit has widened during the second half of the 1990s
and at the beginning of the new decade it is widely recognized that this trend is
unsustainable. The overall debt accumulation in the United States, which has
occurred in the private and external sectors of the economy and by the semi-
governmental agencies, is particularly disturbing when taking into account that
this explosion of credit was accompanied by a virtual breakdown of household
savings, where the rates turned negative in 2000 and 2001. A further element of
doubt as to the sustainability of the boom is suggested because high-tech goods,
which currently represent less than 8 percent of total manufacturing output,
contributed two-thirds of the increase in manufacturing output between 1995 and
2000 (Greenspan 2001b). Finally, figures on productivity, prices, and growth may
be biased in the official statistics due to the application of the so-called hedonic
technique, which is now applied to about 18 percent of the gross domestic product
(Aizcorbe et al. 2000), thereby boosting “real growth” and productivity figures while
at the same time reducing inflation rate figures.

Monetary policy
Shortly after announcing his concerns about “irrational exuberance” in financial
markets, the chairman of the Federal Reserve changed stance, and from 1997
onwards the Federal Reserve System seems to have become more convinced each
year that the new economy is not a chimera. In December 1996, Alan Greenspan
was still concerned about the potential effect of high equity prices for economic
stability when he said:

But what about futures prices, or more importantly, prices of claims on future
goods and services, like equities, real estate, or other earning assets? Is stability
of these prices essential to the stability of the economy? Clearly, sustained low
inflation implies less uncertainty about the future, and lower risk premiums
imply higher prices of stock or other earning assets. We can see that in the
inverse relationship exhibited by price-earning ratios and the rate of inflation
in the past. But how do we know when irrational exuberance has unduly
escalated asset values, which then become subject to unexpected and
prolonged contractions as they have in Japan over the past decade? And how
do we factor that assessment into monetary policy? We as central bankers need

254 Antony P. Mueller
not be concerned if a collapsing financial asset bubble does not threaten to
impair the real economy, its production, jobs, and price stability. Indeed the
sharp stock market break of 1987 had few negative consequences for the
economy. But we should not underestimate or become complacent about the
complexity of the interaction of asset markets and the economy. Thus,
evaluating shifts in balance sheets generally, and in asset prices particularly,
must be an integral part of the development of monetary policy.
(Greenspan 1996)

After that declaration, and particularly from 1997 onwards, by announcing
the United States as the leader in the creation of the new economy, the expectation
seems to have gained hold more strongly within the Federal Reserve Board that
the tolerance towards credit growth despite falling private saving rates and rising
external debt would pay off. Monetary and other policy-makers increasingly
began to ignore monetary and credit growth rates along with the accelerating build
up of foreign and domestic debt levels and instead turned to the growth rate,
employment, and “price stability” as guideposts.
With very few exceptions, the caveat could rarely be heard that excessive internal
and external credit growth as measured in relation to nominal gross domestic
product and relative to the savings rate along with excessive valuations of the
stock market relative to profits cause transformations in the real economy that go
beyond mere aggregate level effects. With disequilibria between demand and
sustainable production already established, instigating further consumer demand
may then disrupt the intertemporal allocation structure further.
Easy monetary conditions are imperative according to the “Greenspan model,”9
which says that productivity growth keeps inflation down. In order for that to
continue high capital investment is warranted, which in turn requires an accom-
modating monetary policy. In early 2001, the Federal Reserve lived up to this
paradigm by initiating massive cuts of the Federal Funds Rate – irrespective
of considerations that money supply had risen drastically since 1995, reaching a
growth rate of 13 percent for M3 in the first half of 2001 (Federal Reserve Board
2001: 32).
The rise of the new economy was accompanied by formidable advances of the
stock and housing market valuations raising the question about the causal link
between these phenomena. In accordance with the theory of financial market
efficiency, the conventional view would hold that stock market prices reflect
superior future profit potential of the new companies. In this view, technological
progress forms the basis for higher productivity, which in turn results in higher
profits and thus justifies stock market valuations that may seem excessive by
historical standards, but are not out of line when evaluated in terms of the new
economy. As the new technologies facilitated drastically the globalization of
business activities, the new economy paradigm gained a wider meaning. With
the emergence of new companies that became global players almost overnight, the
existence of a new economy became to be seen as a phenomenon of global reach
where companies would be left behind should they not follow the bandwagon. The

Bubble or new era? 255
wider perspective of a global new economy meant for US monetary policy the
new imperative of extending the boom beyond the domestic economy and spread-
ing it abroad by maintaining its underlying momentum. This seemed to justify
that the United States become the “consumer-of-the-last-resort” for the rest of the
world, this way additionally diminishing the importance of increasing current
account deficits as a warning light. In the second half of the 1990s, US monetary
policy began to assume center stage as the guardian of the new technologies and
its globalization.

Institutionalization of moral hazard
Making the boom continue at home and abroad became the prime focus of US
monetary policy in the second half of the 1990s. But among the unintended con-
sequences of the new paradigm ranked the broadly based lowering of perceived
risk levels in the financial markets and a global spread of careless investment
The series of international financial crises that occurred towards the end of the
1990s and continues in the new decade is not a new phenomenon when isolating
each case; it is new, however, in so far as it is a single theme, where the individual
cases represent connected parts in a chain of events. In this perspective, it is not
the lack of control of financial market but the institutionalized system of bailouts
and the credit creation that comes along with it, which causes the spread of financial
fragility. The logic of this argument says that the bailout of Mexico laid the
foundation for the crisis in Asia, which formed the groundwork for the financial
and monetary crises in Russia and Brazil, which in turn brought forth the
Argentinian and Turkish crises in 2001. In the wake of the Mexican bailout, and
reinforced by the series of IMF-sponsored financial packages for the crisis countries,
debtors and borrowers felt ever more confident when increasing their exposures.
During the 1990s,10 the presumption of the existence of an effective lender of the
last resort became a common learning experience for the financial market actors.
As with the debt-driven expansion strategies of many new economy enterprises,
risk considerations increasingly became to be substituted by the relentless aim of
The practice of bailouts on a national and international scale has established
moral hazard as a paramount effect and works towards undermining the internation-
al financial and economic structure. Bailout expectations cause an overextension by
financial actors as considerations of risk levels of debt become diminished. Excessive
liquidity creation – as part of any bailout – lays the foundation for malinvestment,
which remain disguised as long as the boom continues.11 Central banks or other
institutions as the sources of excessive liquidity are then creating these very extern-
alities that they are presumed to contain, when their policies mislead individual
investors into erroneous behavior by exposing them to wrong price and interest rate
Moral hazard in financial markets is more ample in its ramifications than the
traditional concept of moral hazard as it was developed for insurance markets.

thus making distinctions. the new economy as a phenomenon also relates to the so-called globalization. Financial flows have become more massive and more abrupt. Risk models that are based on these theories may actually have contributed to providing false assessments of risk. which form the basis of modern financial market theory – such as those between market risk. this artificial continuation of the boom tends to make existing maladjustments more rigid and later on more difficult to unravel. This process has been accompanied by rapid financial innovation and liberalization leading prominent observers to the conclusion that “today’s international financial system is sufficiently different in so many respects from its predecessors that it can reasonably be characterized as new. as they usually fail to incorporate other than normal distributions and base their risk standards on limited historical data. do not possess a deeper understanding of the economic and socio-political .” Together with new technologies. as distinct from being merely a continuing evolution from the past” (Greenspan 1998). The lowering of interest rates allows the continuation and resumption of investment projects and facilitates the beginning of new projects. ever more international transactions take place in a sphere of virtuality without personal knowledge of the counterparts and are quite frequently conducted by actors. moral hazard as a result of bailouts affects the whole monetary system. velocity. these effects are also not confined to just one economy. and virtuality that characterize current international financial market operations (Mueller 2001a). With the resources already under stress.256 Antony P. International financial markets appear to be funda- mentally transformed by a number of factors. who due to other specific requirements for their job. which are not only unpredictable or beyond the realm of standard economic modeling. markets are exposed to higher degrees of volatility. but have ramifications for the international financial system. as the provision of additional liquidity has also non-specific consequences. and liquidity risk as well as the effects of diversification – seem obsolete. credit risk. along with the application of leverage. which can be summarized in terms of volume. but also show the tendency to produce extremes of volatility and deviations from otherwise reasonable valuations. Bailouts change the expectations of the market participants and they transform financial data. there is a trade-off involved. the continuation of the boom could also end in a “soft landing. Modern communication technologies help to overcome the barriers of distance and time. and facilitate international transactions due to strongly diminished transaction costs. In the modern environment. In contrast to traditional banking. While non- bailouts and monetary restriction cause a prompt contraction almost certainly. Mueller While moral hazard in insurance markets refers to the incentives that the coverage provides for the insured.12 Taking into account that modern means of communication allow for very short reaction time and given the dimension of the assets involved. because it implies additional liquidity creation that potentially affects the economic system as a whole. While the supervisory bodies and intervening institutions may well be quite aware of this build up of misallocation of capital. volatility. which aids to forego restrictive measures and interventionist restraint. Interrelationships emerge.

when unlimited prosperity seemed feasible based on an intelligent monetary management. as the economy appeared to have access to sufficient funding. When observing the recent trends. the pattern emerges that the futile attempt of the Japanese central bank to prop up the economy by monetizing government debt and by reducing its call money rate step by step in the past years to a rate that effectively attained zero. End of the miracle? In the US. Short-term volatility afflicting stock and bond markets and of exchange rates go hand in hand with more profound deviations from fundamentals and trigger various forms of contagion among different assets classes and regions when perceptions are not sufficiently differentiated. with the IMF as the third party to provide loans in cases of emergency. disequilibria hardly mattered. when emerging economies were concerned. With Japanese bonds moving towards historical low yields. and it works counter to the assumptions of rational expectations theory. . pointing to increased speculation or irrationalism when explaining apparent excessiveness misses the point. and additional expansionary biases were introduced due to events in the international financial environment or special circumstances such as the expectation of a “year 2000 problem. As long as expectations prevailed that financial wealth creation by the stock markets could go on indefinitely. Bubble or new era? 257 specifics of the countries whose debts and currencies they deal with. By abandoning theoretical models. because excessive capital flows and asset price inflation tendencies must have a footing in a loose monetary base. However. incentives have been introduced for capital exports to search for higher yields abroad opening additional profit opportunities for private financial institutions by practicing the so-called yen-carry trade. But while both the Japanese monetary policy and the provision of additional liquidity by the US central bank had their own rationale in each of the specific cases. American monetary policy has become increasingly discretion- ary and highly dependent on the chairman’s views about the state of the economy. the economic and financial boom had been helped to continue. US monetary policy under Alan Greenspan’s chairmanship of the Board of Governors of the Federal Reserve System fits none of the dominant economic models. While the deficit reduction of the central government was widely noticed. during the late 1990s. It defies monetarism as much as Keynesianism or its variants. the debt inflation in the semi-governmental sector has hardly attracted attention – despite the speed of its accumulation and its dimension. they also produced the side-effect that domestic and international investors were taught to trust these institutions in guaranteeing ample liquidity. when monetary policy was guided by the model of a new economy. has been a constant source of ample liquidity in international financial markets. and there must be sufficient credit creation by financial institutions.” Monetary policy ignored the build up of debt and left aside considerations of massive disequilibria such as the current account deficits and the disappearance of a financial surplus in the private sector.

some of the drastic policy moves. particularly the aggressive rate cuts in 1998 and in 2001. a strong currency and high real growth.258 Antony P. the arguments for justification are deeply flawed when growth is debt-driven. and that this was also highly beneficial for the domestic economy. But with new liquidity or active demand management. more profound transformations of the capital structure will occur. the process of reassessment and correction gets postponed and the trend towards a widening of disequilibria is made to continue (cf. Mueller 2001b). Given the relatively stable inflation rates at home. temporary respite was bought at the cost of more systemic fragility. when the “Greenspan model” began to dominate US monetary policy. Is there a future for the new economy? The emergence of the new economy in terms of the rapidity of its extension is closely linked to the specific monetary conditions and the international environment of the 1990s. It is hardly conceivable that the new technologies could have spread so rapidly without loose credit conditions. By ignoring that monetary and fiscal impulses change relative prices and the capital structure of an economy. This will make the economy less and less efficient leading to a bust and finally to economic paralysis. but with each measure to safeguard the system. However. Mueller Particularly during the second half of the 1990s. particularly the International Monetary Fund – as the global lender of the last resort. the new imperative now said that the boom at home must be kept going as an essential part of the global requirements. For the domestic economy. debt accumulation became so massive that almost any measure seemed justified to maintain its structure. while the policies of easy money must also be seen as reflections of the conditions for economic policy when financial markets are globally connected. were motivated as much by international as domestic concerns and do reflect the new role of the American central bank to act – in conjunction with international monetary institutions. when the dollar exchange rate is strongly influenced by capital inflows from abroad. the policy formulations based on the analysis of aggregates leave aside the consideration of the effects that take place at the micro-level and emerge symptomatically as macroeconomic disequilibria. The readiness of the monetary authorities to provide ample liquidity in conjunction with the provision of so-called rescue packages to a series of emerging economies by the International Monetary Fund has led to the dramatic trade-off that while a stop of these measures would reveal the distortions immediately. What has been so impressive about the emergence . this monetary policy stance seemed justified. Plans and economic actions must undergo revision in order to establish a new tendency towards equilibrium. the preference has shifted towards the continuation of the policy as long as new liquidity and further credit expansion may be expected to do their job. More so. During the 1990s. when measures to “keep the boom going” are applied repeatedly. The emergence of a recession signals a misdirection of investment due to de-coordinated economic plans that need correction. and when the moderate inflation rates – besides statistical bias – are also the result of cheap imports.

creation of . when government or central bank bailout guarantees are presumed to exist. debt positions in the balance sheets constitute a future burden.e. and as such they do not simply return to initial conditions. Bubble or new era? 259 of the new economy – most notably in the Internet area – was the speed of its global dissemination. Working off that burden is a time-consuming process accompanied by lower levels of consumption. Notes 1 Alan Greenspan quoted in Woodward (2000: 180). i. and like the “planning errors” familiar from the centrally planned economies of the past. Economies are evolving systems. which may also be the result of reduced risk perception. they also did not take properly into account their customers’ willingness and ability to pay for the service. This means that while market prices for physical investment and specific know-how will fall. At the root of investment decisions lie errors due to misleading signals from a monetary rate of interest.” making his approach in this regard somewhat “un-Austrian”. precisely because the originary rate of interest cannot be removed from human valuation – whatever the monetary rate of interest might be (Mises 1998: 523). Will the end of the boom signify the end of the new economy? Not necessarily. in contrast. The existence of many projects that do not generate noticeable profits implies that capital has been squandered. Isolated malinvestments and the ensuing problems for individual companies to meet debt obligations are a common feature in a market economy. the central point is prolonged growth of circulating credit. Many of the structures that were built in the past decade will survive. is part of this performance. which has deceived economic actors about the sustainability of funding of their various projects. The value of money – a function that modern economic systems entrust to a central planning committee – has sent false signals to the whole economy. Trying to boost economic activity again when recession has struck will make the situation worse. What makes the situation different for a financial bubble is the dimension of the required adaptation. without doubt. It sug- gests the existence of a bubble economy brought about by exceptionally loose monetary and credit conditions. saving would stop in favor of the consumption of accumulated capital. for Mises. 3 The difference between the originary rate and the monetary rate of interest becomes obvious under the hypothesis of a complete elimination of interest income (by expropriation or taxation): under such a condition. wrong investments have taken place on a formidable scale. But it has become evident as well that many of these new companies not only expanded too fast and too far. 4 The term “credit expansion” as applied here refers to circulating credit. The credit expansion has brought forth erroneous economic activities. The occurrence that companies could expand in the absence of profits is a monetary and financial phenomenon. Many of the companies in this area appeared almost from the start as instant global players. But valuations must adjust. The neglect of generating profits cannot be explained by technology. which potentially affect the entire economy. Technology. 2 A somewhat different starting point is given by Hayek (1941) as his theory also contains elements of the “real business cycle. as the necessary adaptation process gets postponed.

Jerome Levy Economics Institute. (2001) “As the implosion begins? Prospects and policies for the US economy: A strategic view. Federal Reserve Board (2001) Monetary Policy Report. . 5 It is not possible to time a boom and bust sequence or to measure its extension in terms of years. by exuberantly following this advice. R. While it is usually relatively easy to ascertain the occurrence of a bust. W. submitted to the Congress on 18 July. particularly those given in early 2001 (Greenspan 2001a. Board of Governors of the Federal Reserve System. the formation of the bubble may stretch back considerably in time and constitute a prolonged learning process by economic actors as to the assessment of risks and price expectations. 2001b). it is now evident that risk models can also offer a false sense of security because they may lose their predictive powers in extreme market conditions. 7 Even the finest electronic equipment will lose its immediate value when. 8 Although these agencies are not officially counted as a part of the government sector. Mueller money and money substitutes (such as shares and stock options as means of payment) that is not backed by savings. and credit risk is also more highly correlated with liquidity risk than earlier realized. Moreover. Furthermore. June. Indeed. London and New York: Routledge. It is disproportionalities of this kind that bring about the abortion of the artificial boom. industrial output section. A. market risk is more highly correlated with credit risk than previously thought. p. energy supplies become insufficient. triggered an asset bubble whose fall-out continues to plague the Japanese economy up to the present day.. for example. because inflation numbers represent averages and do not adequately reflect specific price movements.. and the most luxurious and modern houses and cars lose their utility and turn into a financial burden when the supply of complementary goods is lacking. they receive their financial privileges due to an implicit bailout guarantee by the government and it is the common assumption among the investment community that they represent government institutions. (2000) “Constructing price and quantity indexes for high technology goods. Bank for International Settlements (2000) Annual Report. (2001) Time and Money. Basel. 9 See various Greenspan testimonies.” Strategic Analysis.260 Antony P. M. op. liquidity growth must not be diagnosed as being excessive. in addition to being exposed to various distortions ranging from special statistical devices (such as the application of hedonic calculations) to the impact of the exchange rate. 12 Compare the conclusions put forth by the Bank for International Settlement’s 69th Annual Report. cit. The Japanese authorities. 149 and passim: “Through various channels. when the US Federal Reserve Bank staged the rescue of the US stock market and – together with the government – urged global monetary expansion. since market exposures are often built on leverage. New York. Corrado. their mechanical use may actually contribute to market turbulence. Garrison. but this may be deceiving. see Garrison (2001). C. A. 11 For a modern analysis of the capital effects based on the Mises–Hayek theory of the business cycle. Godley. financial institutions including banks are becoming exposed to higher levels of market risk. and Izurieta. 10 The first stage of this learning experience may even go back to 1987. The Macroeconomics of Capital Structure. 26 July.” References Aizcorbe. and Doms. as this is a singular event in most cases. 6 It is widely presumed that as long as inflation is tame.” Division of Research and Statistics.

The Scholar’s Edition. New York: Simon and Schuster. DC. (2001) “Productivity growth and the new economy. (2001b) “Financial cycles. Eine Studie über die den Tauschwert des Geldes bestimmenden Ursachen. Mueller. . Wicksell. London: Routledge. Greenspan. Northwestern University. Jena: Fischer. R. A. Mises. Auburn. Bubble or new era? 261 Gordon.” remarks by Chairman Alan Greenspan at the Washington Economic Policy Conference of the National Association for Business Economics. Nordhaus. (2000) Maestro. business activity.D. F. S.D. (2001a) “Current fiscal issues. AL: The Mises Institute.” Federal Reserve Bank of Chicago. DC.” Quarterly Journal of Austrian Economics 4: 1–19.P. (1998) “Speech at 34th Annual Conference on Bank Structure and Competition.P.” testimony of Chairman Alan Greenspan before the Committee on the Budget. Mueller.” NBER Working Paper W 8096. Is information technology the story?. A. W. D. A. 2 March. Institutional and theoretical aspects.J. Hayek.E.” Zeitschrift für Wirtschaftspolitik 50: 15–34. (1898) Geldzins und Güterpreise. and Sichel. (1941) The Pure Theory of Capital. Greenspan. (2000) “The resurgence of growth in the late 1990s. A. (2001a) “Reforming the world financial order. A. Washington. Onliner. January.” remarks to the annual dinner and Francis Boyer Lecture of the American Enterprise Institute for Public Policy Research. Woodward.” draft of a paper for the Journal of Economic Perspectives.A. Greenspan. Greenspan. A. Washington. (2001b) “The challenge of measuring and modelling a dynamic economy. (2000) “Does the ‘new economy’ measure up to the great inventions of the past?. 7 May. 27 March. L. US House of Representatives. 5 December. von (1998) Human Action: A Treatise of Economics. K. B. DC. (1996) “The challenge of central banking in a democratic society. Greenspan’s Fed and the American Boom. May. May 2000. Washington.” Federal Reserve Board. and the stock market.

Is there something really new about such innovation? To answer this question. then came paper money. and now electronic money is a reality.13 Possible economic consequences of electronic money Jean-Pierre Centi and Gilbert Bougi What is now urgently required is not the construction of a new system but the prompt removal of all the legal obstacles which have for two thousand years blocked the way for an evolution which is bound to throw up beneficial results which we cannot now foresee. we need to place money in the context of a procedural theory of money (first section) that has unfortunately been neglected (if not thrown out) by mainstream monetary theory. on the other hand. Yet many people seem to be frightened by the harms caused by the new payments technology. its productivity within the information network not only entails changes in the mechanisms of the money supply but also a change towards a full laissez-faire in money and banking that makes obsolete the dominant monetary theory itself. and by the same token what could be a monetary system of fiat private and competitive moneys with different brand names. The characteristics of electronic money foreshadow the cashless society. this is an innovation. This chapter examines the economic consequences of changing the product money in the information age. which is based on the existence of a monopolistic high-powered money controlled by the State. Undoubtedly. which has to be proved) they are diminishing . Entrepreneurial actions are at the origin of electronic moneys. These are of course the costs of electronic money. but these kind of costs always existed in the past and we need to take into account that even though they are positive (or even if they are increasing in absolute value. Our contention is that if on the one hand. What are the economic consequences of electronic money? What are its implications from the view of economics? The change of payment practices that we are observing today owing to progress in electronics is significant enough to ask questions about the future of the monetary economy. (Hayek 1978: 130) Introduction It is frequently said that money is what money does! Gold was once the dominant medium of exchange. the emergence of electronic money is explained by the procedural theory exactly in the same way as the other previous and past innovations.

notably by Kiyotaki and Wright (1989 and 1993). This evolution will change the rules of the monetary game in a way such that they will belong entirely to the market process. the so-called models of search. once inefficiencies are taken into account. Even the most recent models that take into account transaction costs and/or uncertainty and attempt to give logical explanations of the role of money (comparatively with barter). In other words. because the subjective value that people place in the services provided by electronic money increases considerably). Such an approach does not discover anything really new regarding the benefits that money obtains. This kind of explanation of the use of money is based on non-cooperative games. and the loss of effectiveness of the usual transmission mechanisms make the monetary policy inefficient or outright useless (third section). however.e. and it is either dumb or unable to deal with the issue of the emergence and genesis of money. the subjective opportunity costs of the services provided by electronic money are lower than those related to the services from traditional means of payments (second section). the only suggestion from the comparative statics experiments is that it is possible to jump from one equilibrium to another (Pareto-superior) – that is to say from some closed universe defined by some institutions to another closed universe associated with better institutions – if and only if the latter is exogenously created in order to achieve a well-defined optimal social state of the economy. a new monetary order is being created. in concordance with the real nature of money as an institution. economists were induced to include in their analyses the objective notion of production cost and to make of it the dividing line between material and immaterial money. are always partial and more often than not biased explanations as regards the nature of money. In the standard micro-monetary theory. The whole neoclassical theory faced the difficulty of integrating such a view of money as a product and value theory. According to the general equilibrium approach. it is said that immaterial money owes its existence to some deus ex machina and the value of money could not be preserved without some exogenous constraining force. The substantial changes in the demand for money and the supply of money. then immaterial product. Economic consequences of electronic money 263 in relative value before the widening freedom released by electronic money (i. Therefore. it focuses on the acceptability of money rather than its emergence. money is viewed as a special device of exchange determined by reason. and whose value should be controlled by the State. Some recent attempts within the micro-monetary theory have been made. Implicitly or explicitly. where money will be managed according to the rules of laissez-faire banking systems and the discipline of monetary competition (final section). In concentrating on the notion of product. An institutional approach to electronic money Money as a procedure and money as a product Monetary theory focused on money as a product: initially material product. and it is not because we gain some appreciation of the general acceptability of fiat money (Kiyotaki and Wright 1989) that we have some explanation of the origin .

through an experimental study and using game theory. at least. it would be more helpful for the explanation of monetary phenomena if money were an adjective. which is viewed. This is not to say that money is irrelevant in reducing transaction costs but that such reduction is related to an institutional or procedural aspect that cannot be crowded out and should be addressed by starting from lack of confidence among individuals.264 Jean-Pierre Centi and Gilbert Bougi and the emergence of money. The latter. Therefore to trust in money is to trust in a general rule that emerges from below as a regularity that people observe but are themselves unable to rationalize because of their ignorance. First. Such an approach notably abstracts from the question of how evolved rules (i. money in its medium of exchange function is intrinsically involved in the market process. through an ongoing market process. as money is a general rule resulting from a self-imposed procedure spontaneously applicable to all people. is only able to capture parametric uncertainty but not structural uncertainty. and . regularity is far more important than the physical attributes of the goods. Some others. that is the problem of ignorance (Hayek 1945). Second. Menger (1871) emphasized the notion of saleability of goods in order to show how. . nor is it the explicit outcome of a rational agreement. which is typical of the open society in comparison with the primitive society. give rise to repeated practices whose regularity fully reflects the existence of an organic social institution. Three points. Thus money is a procedure to achieve indi- vidual exchanges when a general lack of mutual confidence prevails. attempt to deal more explicitly with the issue of emergence. this also means that no one invented it. following an ongoing and competitive process. It is Carl Menger’s (1892) contention that moneys emerge from subjective entrepreneurial perceptions. money as an undesigned social institution is a basic principle of the Austrian theory and this is how we may understand the following comment of Hayek (1978: 52): “it has been rather a misfortune that we describe money by a noun. exactly as no one invented the market.e. such as Duffy and Ochs (1999). this approach dares to push forward questionable and hazardous proposals regarding the supply and the management of the product money. Thus. By cutting the real nature of money. the Austrians developed a theory of the monetary procedure: it is a procedural theory and in focusing on the product we may forget the procedure. as Hayek explained. are to be derived from the Mengerian theory of money. . outside static contexts and within the dynamics of real time that paves the way to entrepreneurial discoveries. institutions) are. According to Austrian theory. as an exchange process of services among individuals who are ignorant of their own ignorance. interpersonal trust in trade is part and parcel of the knowledge problem. although it is widely used in evolutionary economics. because the latter. Our main skepticism here comes from the use of game theory. Third. the unintended outcome of human actions. .” More than a theory of the product money. the most easily saleable goods become “under the influence of custom” widely acceptable in trade.

they are like various shapes of the basic institution. such innovations of the product money were born through diversion of the public regulations in force.1 As a result. that is the continuation of the market process and price mechanism abiding by private property and commutative justice. Although the form of money may change according to places and times. all conveying the same message. In the past. monetary innovations do exist. the institution will not produce the best monetary services. These innovations represent themselves as changes which are not the result of human design. and electronic moneys). they will also undermine the nature of money and produce unintended side-effects which will weaken money as an institution. it does not do anything other than attempt to separate money from the market process. Once the State assigns itself the exclusive power of producing and managing the product that should be the certainty equivalent of money as social institution. the medium is being changed once again with the discovery of electronic money. That happened when we passed from commodity moneys to bank notes. That only means that there is a kind of hierarchy. then to scriptural money. the innovations in the product money have always been intermingled with the market process and. It is a great mistake to argue that money as an institution has been invented by the state and imposed on population by fiat. If policy-makers aim at controlling these spontaneous changes. although they do not cancel the basic principle of money as a social institution. As was expressed above. Second comes the functional approach of money in con- nection with which we should agree that the medium of exchange function is the major role of money. The first and basic approach is the one that introduces money as a social institution. but they . a mismanaged money produces wrong information.2 Most of the time. but by no means should the form take precedence over money as an institution. pen moneys. These various forms of money as a product are due to endogenous innovations. considerations of political expediency overturn monetary and economic soundness. Economic consequences of electronic money 265 New concepts for monetary theory and new rules for monetary discipline To give precedence to the institutional approach of money does not mean of course to give up either the functions or the forms of money. If money is mismanaged as a product. money as an institution is capitalized trust reducing complexity in social relations: it is a knowledge and information repository. and happens today while we are passing to electronic money. Nevertheless. accordingly with the procedural theory of money. but the message being conveyed is (should be) always the same. Third comes the formal approach of money. The demand for some form of money or another is nothing else than the demand for liquidity in a changing world in places and real time. The form of money may be constituted by commodity moneys (as media of exchange and stores of value) or immaterial moneys (such as currency notes. and go in the opposite direction. We should not confuse the information conveyed by the product money and its medium. that entail in succession other public regulations that economists attempt to bear out by frantic rationalism. came from spontaneous changes. which generates pernicious effects.

266 Jean-Pierre Centi and Gilbert Bougi have more or less swiftly always been seized by the State. two proposals that fit with the evolutionary approach of money and are on the opposite side of the predominating constructivist approach of money over the twentieth century. voluntary exchange and relative prices. The expansion of electronic money entails a radical change in monetary theory together with a substantial change in monetary mechanisms.e. The concepts of monetary theory have almost always (at any rate in the mainstream) leant on this principle of public control over money and the latter has also been justified in a rationalistic way. The monetary theory that takes form seems more and more to be based on a full integration of the money supply into the private sector of economic activities (i. The procedural theory of money does not teach something really new when applied to electronic money: money as a product should be the certainty equivalent of money as an institution. Electronic money provides evidence of the epistemological superiority of the market in processing information over the political decision-making process and therefore comes to the point of undermining the monetary power of the State. On the one hand. into the market process). as important is its potential for change in the payments system and its potential for extending exchanges. To the vast majority of people government interference in matters of banking have become so much an integral part of the accepted institutions that to suggest its abandonment is to invite ridicule. The event of electronic money is particulary interesting because it practically leads to a refutation of the general assertion expressed in Vera Smith’s quotation. On the other hand. This new theory is necessarily built on the basis of laissez- faire banking and monetary competition. the economic distortions induced by the statist manage- ment of money tend to vanish. This entails two consequences: first. such monetary innovation is one among others that took place in past centuries and that were the unintended outcome of the market process. We would like to lay stress on two points that seem to characterize such change. An economic overview of electronic money What is electronic money? Electronic money is broadly defined as an electronic store of monetary value based on a technical device that may be widely used for making payments without . and second. Consider the following comment of Vera Smith (1936/1990: 195): But whatever may be our verdict as to the comparative outcome of free banking and the system of central control in terms of stability it is unlikely that the choice can ever again become a practical one. the basic concepts on which monetary theory was built up in the mainstream are themselves set aside by monetary and financial innovations. the information flow and the way information is being processed owing to electronics allow elec- tronic money to reinforce the unfolding of the market process such as it is based on the rules of private property.

In other words. but acting as a prepaid bearer instrument. e-money. specifically a serial number identifying each bill and the monetary value it represents. DigiCash’s system was established by Chaum. Virtual money is actually a piece of information with intrinsic value that is stored on an individual’s personal computer’s hard drive or in an electronic wallet and could be used to make online purchases. and these are used in a wide range of applications. and its versatility opens up a host of new markets and applications. It is the first functional form of virtual currency. Mondex has received much recognition in the financial press. Mondex is a big company for bank-issued smart cards. Mondex can only transfer funds to another Mondex card. the electronic value is stored on a computer chip (or integrated circuit) embedded in the card and value is typically transferred by inserting the card into a card reader (smart cards and stored-value cards). and paper currency overseen by central banks. the electronic value may be stored on the hard disk of a computer and transferred over communication networks such as the Internet when payments are made. The smart card gives the appearance of a credit card with an embedded micro- chip that can be loaded up with e-money. is money that moves along multiple channels. the Internet. To store the prepaid value. Another virtual money form is the one issued by NetCash. Rather than a wallet filled with bills. There are two different types of electronic money: prepaid cards and prepaid software products (virtual money). i. The company has plans to create a card reader that attaches to a computer so that individuals may use these cards online. With prepaid cards. which will store electronic value on a chip on the card itself. Economic consequences of electronic money 267 necessarily involving traditional bank accounts in the transaction. The most revolutionary products developed in financial and Internet circles in decades have been the true “DigiCash” products (Chaum 1992). electronic money. Electronic money products differ in their technical implementations. but users can transfer funds among each other. The advent of Internet commerce permitted a much more widespread use of e-cash transactions. and started its experimental system in 1994. A number of smart cards are on the market today. It adds value to any service involving payment. With software products. Many companies have developed their own forms of virtual money. others aimed at making all traditional forms of currency electronic. It can also be transferred from a personal computer to a smart card for use offline. while software-based schemes use specialized software installed on a standard personal .3 partly outside the established network of banks. These are nothing more than information.e. we have a virtual wallet4 filled with virtual bills. new forms of electronic money began to appear in early 1995. some aimed at assisting Internet commerce. card-based schemes involve a specialized and portable computer hardware device. NetCash works very much like physical cash but without the physical symbols. It is primarily working on smart cards. Today. Virtual money by DigiCash is a new concept in payment systems. In response to this. It combines computer- ized convenience with security and privacy that improve on cash paper. typically a microprocessor chip embedded in a plastic card.

These unique data that define digital cash are given to the merchant previously selected by the customer. public key technology is used in combination with symmetric algorithms to protect the messages in transit and authenticate origins. as long as customers of the bank do not use it twice. there is a public key for sending and a private key for receiving. Characteristics Security: the use of encryption If the transmitted information is of a sensitive nature (i. makes instantaneously available goods and services for anyone in the whole world. which is the science of keeping digital data secure. Currently. and of course. the vendors of specialized hardware and software. the sender and receiver have the same key. There are two kinds of cryptography: symmetric key and public key. then it needs to be protected so that only those authorized to read it may do so. In symmetric key cryptography. four types of service providers will be involved in the operation of an e-money scheme: the issuers of the e-money value. As noted. In public key cryptography. The science of cryptography. The bank issues this digital cash using encryption and deducts the fund from the established account.9 Negligible transaction costs and interest earnings The high-speed communications network benefits the users of e-money.5 The digital cash is a combination of two huge integers which have a special mathematical relation. Only the bank can confirm that this digital cash is legitimate and has not been used elsewhere. the bank credits the merchant’s bank account by that amount. The bank cannot know who used the digital cash. The merchant’s bank has to make contact with the customer’s bank for confirmation. How does this work? An Internet user opens an account with real money at an Internet-based bank.6 Let us suppose that a customer (X) wants to purchase goods or services through the Internet. The key is what we use to “unlock” a message.7 Encryption is the process of scrambling data into ciphers or code so that they can only be unscrambled (decrypted) by individuals who have the key essential to accomplish this task.268 Jean-Pierre Centi and Gilbert Bougi computer. The customer asks the bank to issue a certain amount of digital cash for use on the Internet. financial data). . and the clearers of e-money trans- actions. If the bank confirms the digital cash. Moreover. On the other hand. Typically. banks themselves suffer less processing costs as people switch from currency notes and checks to e-money to make payments.e. the network operators. No other person or institutions can imitate this relation. makes this possible. other methods are being experimented with to obtain digital cash amounts through financial assets transfers. The merchant in turn sends these data to his bank for confirmation. The public key algorithm is RSA8 and the symmetric algorithm is DES.

the social benefit from the national provision of money by the State is higher than that which would result from a competitive supply of money. The vast potential of cyberspace and the correlative trans- nationality of e-moneys tend to prove that there is no reason to believe that any . Then the argument goes like this: since there is equal probability to have good money and bad money. the monopolistic supply of money is desirable for the purpose of transferring wealth (as Keynesians. the extent of its use does matter as its ill effects (inflation. Those using electronic money can purchase services and goods from any site anywhere on the Internet. Such possibility allows people to evaluate the soundness of the money they use and to move from one money to another. As the evolving internet market is international in scope. banks issuing electronic money will themselves be induced to extend their competition in the cyberspace. The banks will have tremendous incentives to promote their own moneys among cybercustomers: electronically initiated debits and credits. be artificial constraints. it is technically feasible to supply prepaid cards that can be charged against an account bearing interest. E-money is not constrained by national borders. owing to freedom to choose. Accordingly. Economic consequences of electronic money 269 as has just been described. according to some scholars. In contrast. Under competitive pressure. and since a good money at the national level concerns all the people while private competing moneys will concern fractions of the same people. will tend to generate competitive private moneys and to abolish the monopolistic supply of government money. unemployment. its use will extend. to choose alternative moneys – for all those who expect to lose wealth through transfers. Although banks currently face similar types of risks in international banking.e. Yet the potential for bad government money is also widely proven in practice and in theory as well. being more and more attractive through competition in banking. It is often argued that the monopolistic supply of money provides economies of scale at the national level. banks themselves will induce people to hold and use more electronic money by offering attractive interest rates on their electronic balances. it is important to note that these risks are also relevant to the cross-border conduct of electronic money10 (Solomon 1997). Such market expansion is not without certain risks. all the more so as geographic boundaries will. from now on. Transnationality E-money activities are based on technology that by its very nature is designed to extend the geographic reach of banks and customers. yet is clear that a necessary prerequisite for such money is strong barriers to exit – i. stagflation) are imposed on the whole nation. Perhaps. each bank will have cybercustomers everywhere in the world and this will make the denationaliza- tion of money a reality.11 The development of e-moneys together with the competition among the issuing banks pave the way for people to choose among alternative moneys. the ill effects of some bad money in a competitive framework are more limited. for instance would maintain) and increasing social welfare. while a good competitive money tends to be emulated and. But it is clear that if money is a bad one.

This technique offers the potential for a greater degree of privacy in financial transactions. regulations are only the response of governments. The origins of funds are relatively opaque and the identity of the individual or entity transferring them difficult to determine. In fact. which might induce them to implement inadequate investment policies and security measures. payer anonymity12 is a central characteristic of some proposed systems. without the assistance of a regulated financial institution. The main difference among countries lies in the firms that are allowed (or will be allowed) to issue e-money (European Central Bank 1998).14 If . Untraceability and anonymity Electronic money transactions are untraceable and anonymous. The identity of the issuer has implications for the security of the electronic money system. Chaum (1992) proposed untraceable electronic payment systems using advanced encryption technology. Potential risks for transactions Two main types of cross-border use of electronic money can be foreseen. The anonymity of electronic money will make knowing your customer much more difficult. regulation is seen as necessary to cushion the possible failure of market incentives. but at the same time they might be subject to a number of constraints. trans- nationality also means that there is no special problem regarding the size of the area of some competitive money: competition makes it an endogenous variable. the pressure on the part of share- holders to obtain high returns and to reduce costs. while the merchant is located abroad. It is important to note that this question also regards the status of the issuer of electronic money. To achieve untraceability and anonymity on the Internet. In the meantime. encryption has to be fully employed. the customer using e-money and the issuer may be located in one country. presumably in the customer’s home currency.270 Jean-Pierre Centi and Gilbert Bougi particular national boundaries constitute optimum currency areas. In fact. Therefore. There are several possible types of issuers: banks13 or other regulated non-bank financial firms and non-financial firms. issuers established in one country may implement electronic money schemes by which they offer electronic money in another country. E-money systems allow the parties to the trans- action to deal with each other directly. Failures of market self-regulation? A further aspect for consideration is that electronic money issuers may not have sufficient market incentives to implement policies aimed at promoting their financial integrity. Second. for instance. For instance. The use of e-money systems will mean fewer face-to-face financial transactions. the latter two categories of firms are allowed to issue e-money in some states in the United States. transaction anonymity could lead to law enforcement being diverted. First. For virtual money transfers. such as. It is a commonly held view that the issuers have a clear com- mercial interest in avoiding failures.

e-money can be transferred around the globe at the speed of light. a greater degree of competition could yield commensurate benefits. Fraud could also be attempted through repudiation of transactions made with an electronic money payment.e. Electronic money systems face risks that the systems may not be well designed or implemented. and confiscate. associated with electronic money schemes. and at the same time keep our liberty and privacy. or failures in the transmission of messages. Of course. This is especially so concerning traceless electronic money systems that provide payers and payees more anonymity than they would have with paper currency. In the electronic payments system. they could be increasingly used for such criminal purposes. What are the economic consequences of tax evasion? An argument against financial privacy is that it will make the collection of some types of taxes more difficult for governments. malfunction of an application (such as accounting or security functions). i. we cannot both keep the present income tax system and enforce it.15 An unsophis- ticated method of attack would be to steal consumer or merchant devices and fraudulently utilize the balances recorded on them. Containment measures are intended to limit the extent of any fraud once it has been committed. In contrast. the regulatory framework already in place can be extended to cover the new products but competition and innovation might be more limited. fraud. Tax evasion can be accomplished with a few strokes of the keyboard. This reduction in the tax base and tax rates that will be required in the digital age will make the relative size of government smaller. Data stored on devices could also be stolen via unauthorized copying16 (Committee on Banking Supervision 1998: 5–8). control. counterfeiting. and to protect against losses due to fraudulent duplication or repudation of transactions. there are different security measures in the electronic money system. In fact. Electronic money products could suffer from accidental corruption or loss of data stored on a device. an electronic bank is exposed to the risk of an interruption or slow-down of its existing network system if the electronic money system it chooses is not compatible with user requirements. or disruption of the system. Should electronic money schemes offer the possibility of executing anonymous transfers of large amount of money. measures to detect and contain fraud may also have an important deterrent function and thus serve to prevent as well. Detection measures are those taken to alert the issuer or system operator to an occurrence of fraud and to identify the source of the fraud. Economic consequences of electronic money 271 issuance of e-money is limited to banks. The more government tries to tax. the greater the incentive for business and investors to leave. It was . The emergence of e-money has sounded an alarm in government. For example. In a world of anonymity. are tax evasion and money laundering. External attacks are on electronic money products themselves. They are designed to safeguard integrity. regulate. Potential exploitation for criminal activities The criminal activities. if a greater variety of institutions can be issuers. authenticity and confidentiality.

Government authorities cannot stop this change. In the Keynesian macro- economic tradition.20 as well as complicating efforts to fight it. because technology has made it an impossible task. Note that countries like Switzerland that have maintained financial privacy for their citizens.272 Jean-Pierre Centi and Gilbert Bougi only with the rise of Keynesian political economy in the twentieth century that an ever-growing government sector was viewed as desirable. laundering diminishes government tax revenue and therefore indirectly harms honest taxpayers (a kind of prisoner dilemma). First. it makes crime pay. Reasons given by governments for increasing regulation include the need for monitoring the soundness and safety of financial institutions. Perceived ease of entry to one country attracts an undesirable element across borders. degrading the quality of life and raising concerns about national security.17 it is recommended that the government adopt fiscal policies to manage demand. This drives up the costs of law enforcement and health care. Second. in the past.18 Since electronic money is untraceable.000 US dollars. The cost of trying to enforce taxation may well exceed the revenue collected and certainly will exact a price in terms of lost economic efficiency and lost privacy rights that exceed the benefits of their continued taxation. taxation will not be easy. Regulation will not be effective because progress in developing the means of evasion will always be far ahead of those who are trying to restrict it. Moreover. supply will take care of itself. government spending had been a relatively small share of GDP in most countries. Electronic money technologies have the potential to make money laundering much more widespread than traditional methods. . It is too early to determine whether market pressures will cause products to evolve in such a manner as to become more or less attractive for money laundering. because it is a worldwide change with too many people having knowledge (Hayek 1945). The rapid expansion of unregulated and untaxed activities in cyberspace is indeed seen as a threat to state power.19 This problem may need to be solved by a whole new view of international taxation. At present. and Swiss citizens are no more subject to terrorism than American (Rahn 1999: 109–12). the combat of money laundering by governments leads to the destruction of financial privacy. Finally. most stored-value-card and electronic-purse pilot projects have established limits for consumer cards ranging up to the equivalent of 1. smugglers and other criminals to expand their operations. To avoid depressions. not leaving well-defined records for a tax authority to follow. It allows drug traffickers. do we need a smaller or bigger government sector? Remember that. This fact does not justify imposing further restrictions on legitimate money tranfers by honest citizens to the detriment of our liberties. The fact is that the fears of both sides are correct. In the same way. electronic money laundering has the potential to undermine the financial community because of the sheer magnitude of the sums involved. have lower crime rates than the United States. governments have largely given up trying to control the flow of information. Keynesians recommended an important government sector. Today.

. As for the implications of electronic money regarding the conduct of monetary policy. . Mi. This agent can use any or all these moneys to buy each type of good. . Economic consequences of electronic money 273 Implications for the conduct of monetary policy Monetary policy is defined. Effects of replacing central bank currency We will examine electronic money’s potential to replace central bank currency. in broad terms. There are K moneys. focusing on the demand for electronic money. The household seeks to maximize the profit from managing its assets over the payment period. Attention is focused on how the characteristics of these moneys affect the repre- sentative agent’s choice of transaction vehicle. he makes Zgi shopping trips to buy commodity g with money Mi. Each shopping has a cost Bgi. 2. as the apparatus according to which a central bank decides how to act in order to achieve its final objectives (sustainable growth. The return on the savings assets is larger than the return on any of monetary assets. high level of employment and price stability). . For this. 2. The savings asset earns the rate of return rs.1) i =1 g =1 i =1 i =1 g =1 . We present a simplified version of their model. transaction frequency. During the period. The representative agent receives his income Y at the beginning of a period of fixed length. . The representative household spends only a fraction of his income during one shopping trip. and the various kinds of money earn rMi. and average balance in various media. G. The following discussion is based on a model by Santomero and Seater (1996) in order to show what are the conditions and implications of an increasing use of electronic moneys. . and in money balances. the profit function of the representative agent can be written in terms of average values of the respective assets. Because all conversion and shopping trips are evenly spaced and consumption proceeds at a constant rate. S. k G k L G Π = rs S + ∑ rM i Mi + ∑ rX g X g − ∑ Ti ai − ∑ ∑ Zgi Bgi (13. it is necessary to take into account that the development of electronic money will reduce the use of central bank notes and coins (Bank for International Settlements 1996). K. A reduction in the demand for bank notes Santomero and Seater study the behavior of a representative agent facing the opportunity to choose among different generally accepted payment instruments. with i = 1.21 There are Ti conversion trips to obtain money Mi and each such trip has associated with it the conversion cost ai. we will show how a representative agent would choose among various payment instruments once the latter exist. During this period. . with g = 1. he spends the entire income by buying and consuming G different commodities g. . Unspent income is held in a single savings asset.

Suppose that A = ∑ XgEM is the total electronic money spent. In fact. it may dominate cash in the near future. Competitive pressure could force the electronic money issuer to pay interest on electronic money balances. we get the expression of the demand for electronic money:22 1/ 2  aEM A  1/ 2 G  B X  (13.24 Competition among issuers of electronic money makes it likely that consumers will pay no per transaction fees and negligible cost. If the cost per shopping trip is zero. BgEM = 0. • the cost per shopping trip. given the lower cost of transferring value into the card. A. the cost per shopping trip BgEM.2) MEM =  − ∑  gEM gEM   2(rs − rMEM  g =1  2(rMEM − rX g  According to (13.e.3) MEM =    2(rs − rMEM  It is important to note that the cost of conversion aEM is likely to be small and will decrease further as technology improves and smart card readers are more widely distributed. • total electronic money spending. consists mainly of potential fees per purchase charged by the shop owner or issuer of the smart card.e. rs – rMEM. A will grow because of all characteristics and low fees provided by an electronic money. BgEM. g =1 Operating appropriate substitution into the profit equation.274 Jean-Pierre Centi and Gilbert Bougi We suppose that Xgi is the amount of commodity g that is bought during a shopping G trip with money Mi. aEM. • the interest differential between the savings asset and electronic money balances. The cost of using electronic money to buy commodity g. the equation for demand for electronic money MEM is reduced to the standard Baumol (1952) and Tobin (1956) square-root formula: 1/ 2  aEM A  (13. Zgi and Xgi. the demand for electronic money depends on several parameters: • the cost of transferring electronic money on a smart card or a computer hard drive. The smaller this differential. i. the larger the demand for digital money. From the first order condition. Demand for electronic money depends on the total amount of electronic money A spent each period.23 It is likely that the use of electronic money will be free for consumers because competition between electronic money issuers and the competition between electronic money and costless paper money will prevent issuers raising fees on electronic money transactions. i. So what are the consequences for the central bank? . one can show how the household chooses Ti. What remains to be considered is the effect on the interest differential rs – rMEM.2).

However. Economic consequences of electronic money 275 The loss of central bank seigniorage revenue The replacement of banknotes by privately issued electronic money in retail payments implies a reduction of base money and the shrinking of the central bank’s balance sheet. The substitution means a reduction of the demand for real balances in base money and unless this kind of money is imposed by coercion-like reserve requirements. Consequently. The replacement will likely make it more difficult for a central bank flexibly to absorb a liquidity shock. Because central bank currency is by far the largest liability of central banks. i. the loss of seigniorage could become a concern to central banks. Central banks could be forced to step in and absorb these reserves by selling central bank assets. or government would lose its monopolistic status in supplying high-powered money and compete with private money suppliers.26 an extensive substitution could reduce the monetary base to the extent that it would adversely affect monetary policy implementation and eventually remove it. Both cases boil down to denationalization of money and . The effects of electronic money on the implementation of monetary policy will depend upon whether its primary impact is on the demand for bank reserves or on the central bank’s capacity to supply these reserves.e. As e-money reserves would increase. The effect on demand would result from the substitution of electronic money for reservable deposits. However. substitution for central bank currency would increase the total supply of reserves. one issue is how to face the desired balances without questioning the monopolistic power of the central bank. Since banknotes in circulation represent non-interest-bearing central bank liabilities. This revenue has been associated with central bank operating costs. Substitution of e-money for central bank currency would increase cash holdings of banks. special circumstances could arise in which the central bank might not be able to implement reserve-absorbing operations swiftly on a large enough scale25 because it would lack sufficient assets on its balance sheet. which might in consequence become more dependent on other sources of revenue. It could fall substantially and become too small to cover the cost of central bank operations. a substitution of electronic money for cash would lead to a corresponding decline in central bank asset holdings and the interest earned on these assets that constitutes central bank seigniorage revenue. if the spread of electronic money was extensive enough. It is conceivable that a very extensive substitution could complicate the operating procedures used by central banks to set money market interest rates.27 Such contingency would be nonsensical and could not last long. there would be a substantial reduction in a bank’s demand for settlement balances in base money. thereby increasing their reserves on the books of the central bank. and this is equivalent to an expansionary open market operation that provides additional reserves to the banking system. The banks would observe that their cash holdings exceeded the optimal amount and they would return cash to the central bank. That means that either central banking would lose any reason to exist and the government would have nothing to do with a central bank. Monetary policy is based on the ability of central banks to determine the conditions that equilibrate demand and supply in the market for bank reserves. on other income sources such as government subsidies (themselves covered by taxes).

The emergence of electronic money seems to strengthen the case for a strict monetary base rule (Selgin 1997).28 A large increase in velocity is troublesome. The income velocity of money is of interest to central bankers who rely on monetary aggregates either as indicators or as ultimate targets. the central banks can control the price level by choosing the appropriate level for the money stock. however. If the velocity and real GDP are known in advance. the central bank must be able to determine the money stock. Ineffective monetary policy We will consider the potential effects of electronic money on the monetary transmission mechanism. As the velocity of money would be increasing. Stable velocity of money is traditionally crucial to them. a sine qua non to determine the monetary aggregate. gives less freedom to central bankers. The velocity of money How does electronic money affect the income velocity of base money? Here. Although the recent models dealing with monetary policy are based on the game theoretic framework and are rather turned towards the time-inconsistency issue. we study the implications of electronic money for the level and the stability of the income velocity of money of the monetary base. at least implicitly. From the money multiplier to mutual fund banking The public. monetary policy itself would be jeopardized. Failures in achieving monetary targets have larger unwanted effects on nominal income. even if measured correctly. second. Yet such a prediction is not so simple. As the whole monetary base would tend to disappear. its variability would make it more difficult to maintain financial stability. the income velocity of base money would increase. Consequently.276 Jean-Pierre Centi and Gilbert Bougi the substitution of the micro-profit motive for any macro-objective funtion defined in terms of social welfare in the present as well as future periods. assuming central bank reserve-absorbing operations. To do so. the velocity of money must be predictable and stable. b the public’s desired currency notes and r the banks’ desired reserves (including required ones).4) M = kB. in substituting central bank currency by electronic money. as suggested by the quantity equation. reduce the monetary base. two requisites must be satisfied: first. Our monetary system depends on the money multiplier whose formula is: 1 (13. the stability of the demand for money function rermains. with k = b + r − br where B is the monetary base. The theoretical background underlying the use of the monetary aggregates is provided by the quantity theory. . As we suggest that electronic money could substantially replace central bank currency. this would.

it can offer a powerful mechanism for limiting (reducing) settlement in the interbank settlement process. A long-standing argument against a monetary base rule is that such a rule would not allow the central bank to adjust the base in response to unforeseen changes in the public’s desired currency ratio. banks may choose to lend to each other during the day. or at least inefficient. all retail payments are transacted in real time. In such a system. This system eliminates the risk of a large contagious end of day settlement failure. Moreover.30 Real time gross settlement is probably a source of decline in reserve holding.29 Settlements are effected continuously throughout the day. comparable to cash payment. Yet. RTGS has been adopted by the European Union countries. commercial banks will themselves be induced to economize on the use of unpro- ductive central bank reserves. Indeed. employing the same prudent standards by which they lend to each other overnight. It seems to be clear that if currency notes are increasingly rare among the public. The multiplier would then simply be the reciprocal of the banking system reserve ratio. electronic money contributes also to reducing r in various ways. The implication of real time settlement. the central bank could always appeal to unforeseen changes in factors beyond its control. First. with no need for extensions of credit to the payments system by the central bank. A major step was taken by Switzerland in 1987. One can observe that capital markets become more and more liquid and mutual funds gain direct access to the payments system. With electronic money. Banking will no longer be limited to financial . because it can effect final settlement of individual funds transfers on a continuous basis during the processing day. when it introduced a RTGS system that eliminates the risk of a large. What happens at the level of ordinary transactors is also happening at the central clearing level of interbank payments. Electronic money seems to strengthen the case for monetary base rule by helping to eliminate the public’s desired currency ratio as a factor influencing the money multiplier. the central bank would not need so much freedom to improvise. in conjunction with various forms of electronic money is that systemic risks to the payments system can be controlled without resorting to lender of last resort facilities. in which case there will be – due to the new monetary technology which saves a tremendous amount of time – no need for a whole interbank clearing system (whose final purpose is precisely to save time). Moreover. This information technology sheds light on the real nature of banking. Undesired fluctuations in nominal incomes and prices could still occur as a result of unforeseen changes in the demand for money or in bank reserve ratios. We can conceive another argument. electronic payments technology today allows real time gross settlement (RTGS) systems to operate in many countries: such a system means that the settlement is completed at the time of transaction. even the emergence of electronic money is not enough to make a strict monetary base rule work perfectly. The challenge of monetary control would be simplified accordingly: with one less variable to worry about. the holding of bank’s reserve would be useless. More recently. we have to take into account two other real possibilities. contagious end-of-day settlement failure. Economic consequences of electronic money 277 With any sort of rule (including a zero inflation rule). Second.

. Together with electronic payments. This change is also giving banks other kinds of reserve and clearing media that are more profitable than central bank deposits owing to the transfer of mutual fund shares. The value of the liabilities of a mutual fund bank will. to pay for goods or services. We can point out that the credit worthiness of the buyer can be instan- taneously verified by the seller through the immediate electronic access to the buyer’s wealth account. As a result. the development in technology allows people to collect information about the creditworthiness of the business sector (the non-financial spending units). An intermediary scenario would be overseeing and some kind of regulation exer- cised by monetary authorities. as with any mutual fund. without any privileged status. government money will tend to be abolished.278 Jean-Pierre Centi and Gilbert Bougi institutions that undertake to redeem their deposit liabilities at par. since the central bank’s leverage over the commercial banks would disappear. banks belonging to different jurisdictions with different minimum reserve require- ments (even zero requirement) intensifies pressure to reduce holdings of base money. A new monetary order would then take form. In this system. always reflect the current value of its assets. Towards a new monetary order From the foregoing points it seems to be clear that central banks are no more able to make the issue of electronic money exclusively a central bank activity and to give e-money the status of legal tender. its influence over the interest rates and exchange rates would be ineffective. Hence. base money and outside money would become ineffective concepts as well as the exogenous money supply. The more plausible scenarios are either the disappearance of the principle itself of centralization in money issuing. confirming the buyer’s payment instruction and the seller’s instruction to transfer funds to the mutual fund of his choice. Finally. the money multiplier will make no more sense. In any case. mutual funds are not subject to runs: there is no incentive for mutual fund depositors to form a queue to redeem investments whose value is continuously marked to market. There is no more need for the holding of the bank’s reserve. The seller’s point of sale terminal immediately communicates through the clearing system with the buyer’s mutual fund. and outside money will disappear together with its theoretical implications from the lender of last resort function of the central bank to the quantity theory of money. the buyer offers electronic money issued by a mutual fund of his choice. or the immersion of the central bank within the new payments technology so that it would become one money issuer among others. there are more and more purely financial transactions concerned with the direct transmission of saving from consumers to investment in business. wealth can be securitized to form liquid wealth accounts. The famous “new view” of the bank as a financial inter- mediary is out of date as securitization and mutual fund shares are developing (Fama 1980. as the transnationality of e-money is enhancing competition in banking. In backing transaction media by equity claims. By the same token. More significantly. 1985).

To transfer money in the traditional way. with the transaction completed within the Internet. Since electronic money uses the existing Internet network and the specific computers of its users. video. Electronic wealth should not be restricted to a unique. and computer software. automatic teller machines. and specific electronic transaction systems. currency. Electronic money payments can be used potentially by anyone with access to the Internet and an Internet-based bank. This ability to finally handle micro-payments might also provide a solution for the payment of fees to authors and publishers for use of copyrighted materials in electronic form. like 10 cents or 50 cents. conventional banks maintain many branches. now much higher than transfers within a given state. and debit cards32 Characteristics Electronic money Currency Check Debit card Legal tender No Yes No No Acceptability ? Widespread Restricted Restricted Marginal cost per transaction Low Medium High Medium Payment finality face-to-face transaction Yes Yes No No Payment finality non-face-to- face transaction Yes No No No User-anonymity Yes Yes No No To understand the importance of transnationality. exchanges become more open and less expensive. The cost of transfer within a state is almost equal to the cost of transfer across different states.1). only a bank in a given state can issue electronic money in that state’s currency. An electronic money token in a given amount can . close to zero.31 Table 13. the cost of digital money transfer is much lower. The cost of international money transfers. That is. digital money makes person-to-person payments possible. which in turn may encourage a new distributional system and fee structure for music. to be possible. let us assume that electronic money is completely domestic. This lower cost enables micro-payments. checks. While credit card payments are limited to authorized stores. The cash can be transferred through computer networks and off the computer network into other storage devices. clerks. Only the citizens of the state can use this electronic money and only with merchants for products and services within the state. will be reduced. Costs for all this bureaucracy is generated in part from fees for money transfers and credit card payments.1 Features of electronic money. proprietary computer network. Now. Economic consequences of electronic money 279 Increased efficiency of exchanges Transnationality provided by the use of electronic money makes international transactions more efficient in several ways (Table 13.

the digital money does not expire. trade will be executed by instantaneous and simultaneous debit- ing and crediting liquid wealth accounts. In any market economy. is how a combination of technologies 20 years later would enable and end the run around the forces of monopoly and inertia. non-earning asset. or gold. held by both banking and non-banking institutions. family. Hayek (1978). What Hayek did not foresee in 1976.33 which is a troublesome. Moreover. • global commodities futures markets: which enable anyone on the globe to know the price of any freely traded currency or any specific commodity. . Thus. effectively driving the weaker. or firm to make decisions about the things most important to them. A resident of a country with a chronically weak currency could easily shift his savings into stronger ones. there could easily be a massive flight of electronic financial assets from weak currencies to stronger ones. based on their own diverse plans and desires. Is it not a sort of Gresham’s law in reverse? The future of currency competition The arguments in favor of competing moneys are the usual arguments in favor of competition generally. • asset securitization: enabling many assets to serve as the backing for private moneys. With electronic money. The new technologies dramatically reduce foreign-exchange transaction costs. even very small businesses and individuals can use digital money for all sorts of transactions. Again. • public key encryption: which enables anyone to communicate with anyone else on the globe with a degree of privacy. The new electronic digital payments technology will enable property rights claims on real assets.280 Jean-Pierre Centi and Gilbert Bougi be subdivided into smaller pieces of cash in smaller amounts. to review. It maintains value and even yields interest until lost or destroyed. less significant currency out of existence. those technologies that will make Hayek’s ideal not only a utopia but a reality are: • the Internet: the communication system can be accessed by all on a global basis. Users of moneys are generally assumed to prefer moneys that provide a stable expected (or at least predictable) value (Klein 1974). property rights perform the crucial economic function of facilitating decentralized decision-making. Property rights enable each person. provided that the issuer has not debased the unit to nothing or gone out of business. has notably argued that competing sup- pliers of currencies would be required to provide market participants with moneys exhibiting the characteristics that are most widely desired. nor did anyone else. Multinational small businesses will become a dynamic new force in local and regional economies. Indeed. New technologies will enable people to acquire the goods they want without holding or handling money. to be utilized as the medium of exchange for virtually all transactions. such as stock and bond funds.

each group using one money.e. As Klein (1974) and Hayek (1978) maintained. brand name differ- entiation is essential. The electronic money system seems to be going to change the way monetary trust is created and main- tained. The competitive cyberbanks would have tremendous incentives to form transnational groups by establishing their own moneys and to preserve the qualities of their moneys. use more their one money). These new technologies protect each individual’s right to privacy and provide sound money. Therefore a good (high quality) money can spread as one group emulates the money that proves effective in another group and as individuals can “exit” (i. to buy and sell). the falling costs of information and communication technologies make brand identification pervasive so that it becomes increasingly easy to compare the qualities of moneys and this would encourage the issuing firms to stabilize the expected values of their respective moneys. This is indeed denationalized money. adverse clearings at the level of interbank payments would no longer be necessary. responsive monetary system which can best be provided by unbridled market competition. both for money producers and for money users. robust economic commerce depends on a flexible. Thus. As today we live in a completely artificial monetary system. Accordingly. As the current trend on the Internet demonstrates. dispense and collect electronic money (i. there would be various and different groups of cybercustomers. Economic consequences of electronic money 281 • smart cards: which provide an easy and convenient way to store. the main problem facing any producer of (base) money is to produce trust in money. flexibility of exchange rates among competing moneys is a condition to dissuade banks from overissuing because overissue by some single issuer would involve a disinvestment in its brand- name capital and a loss of customers.e. 2 The size of any monetary area is endogenously determined. has nothing to do with the national territory. can move from one group to another) or even belong to more than one group (i. As transnationality and competition among banks across the cyberspace would induce people everywhere to choose among alternative moneys. On the one hand. for private fiat competitive moneys. On the other hand. Membership in a group using the same money serves as a bond assuring others that this issuing bank is reliable. defined as individual membership in a group using the same money. The money quality is mainly observed on the markets and as Klein (1974) argued. investment in brand name is a way to produce trust in money and to build reputation. the money issuer has to invest in brand name capital. . Trust in money today depends above all on control- ling the power of the State34 (Selgin and White 1994). while repeated and voluntary dealings are con- ducive to the development of trust. in order to guarantee money quality. Some important features of this possible future monetary order deserve attention. 1 The monetary area.e. as the latter would opt out of this money into a competing currency with a more stable expected value.

This monetary innovation could allow individuals to hold all their assets in highly liquid and divisible mutual-fund shares that reflect current market values. safeguards the value of money. Electronic value exchange is mainly an institutional issue not a technical one. i. and. Many scholars believe that there should be a single centralized legal money within a national jurisdiction. each using one money. It is about developing a set of relationships among people. Why is a monopolistic money seen as desirable? Many arguments have been introduced in monetary theory and although they have been taken for granted in the mainstream. consequently. Money is as much a social norm as an object or technology. Its use value in a given form depends critically upon widespread acceptance of that form. Many points are not developed in our study. As it is the case for customary norms. each of whom possess only partial knowledge. It must be ensured that price stability and the unit of account function of money are not endangered. and protects individual freedom. A significant development of electronic money could also have implications for monetary policy strategy and control of the operational target. may geographically coexist.282 Jean-Pierre Centi and Gilbert Bougi 3 The monetary area is not necessarily defined in terms of a specific territory or geographical area. . the monetary discipline would evolve spon- taneously from bottom up rather than being intentionally imposed by legislation. and creating guaranteees of general trust. 4 One individual may belong to more than one group. The challenge is to develop an institutional framework that provides transparent rules for the elctronic payments system. 6 No money is imposed from above. would represent economically viable exchange media. Conclusion The issuance of electronic money is likely to have significant implications for monetary policy in the future. The emergence and unfolding of electronic money today seems also to show practical contemporaneous evidence that government monopoly of money is unnecessary (Hayek 1978: 107) and that money will be more and more self-regulated through the market process. What is money will be determined by what buyers and sellers accept and use as money rather than by government definitions. 5 Many groups. the necessity of governmental control over money is not theoretically proven.e may use more than one money. that is a monopolistic supply of money. Electronic money could severely affect the central bank’s position as the sole supplier of currency. Market forces are on the side of private suppliers of electronic money and those forces will make a system of private competing currencies inherently stable.

13 Credit or deposit-taking firms are defined differently in different countries. 11 The argument here is drawn from the comparison made by Osterfeld (1989) between “good” and “bad” legislations at the national and local levels.marktwain. Cyberspace is a non-physical “space” created by networked computers. the number of users on the Internet has grown more rapidly as a result of recent developments. The most commonly discussed part of cyberspace is the Internet. Economic consequences of electronic money 283 Notes The authors would like to thank the referee. there were forms of money. a combination of high- speed communications lines and computer networks. RSA. From a policy point of view. before the Berlin Wall fell. 3 The Internet is a data infrastructure that connects computers via telecommunication networks. but they had absolutely no importance (no value) because they had no message to convey from any monetary institution: no “money as an institution” existed whatsoever. For example. A related issue is whether legislation (regulation) was a source of monetary innovation. named after its inventors Rivest. since e-money is a balance-sheet liability of these institutions. It enables the formation of virtual communities and virtual corporations including shops and banks. is the most widely used primarily because of its simplicity and because it works well. We can visit Mark Twain Bank at http:// www. 4 It is a file in our computer. 9 Data Encryption Standard (DES) remains the most frequently used. look at the expansion of scriptural money after the enactment of Peel’s Act in 1844). 2 Bills of exchange and bank notes were the creation of merchants and bankers within the private sector. 5 This new payment system deserves the name of digital cash because it is almost equal to a cash payment in terms of security. 14 As we are saying. respectively. peer-to-peer payment. one cannot argue that the innovative process is accelerated owing to legislation rather than without it. historically this has often been the case. 8 There are many secure modern public key algorithms. 12 The identity of the party initiating a cyberpayment value transfer. 6 The network operators and vendors are only supplying technical services. 7 Cryptography was originally developed by the military for sending secret messages past the prying eyes of enemy forces. DigiCash’s system has been realized with the establishment of the Mark Twain Banks. but first. second. when the United States Department of Defense Advanced Research Projects Agency (ARPA) funded a small group of computer programmers and electronic engineers to redesign the way computers were operated. This resulted in the creation of the first network of computers. the most important providers are the issuers. while clearing institutions are typically banks or specialized bank-owned companies that provide a service that is no different from that provided for other cashless payment instruments. restrictions might be placed on the maximum value that consumers and retailers are . The Internet is merely a medium through which people transmit information to one another. the issuer of e-money could be a non-bank financial firm as well as a non-financial It originated in the 1960s and 1970s. Acknowledgements are also adressed to Stefan Schmitz and George Selgin for helpful comments. Since 1990. 10 Banks may face different legal and regulatory requirements when they deal with customers across national borders. and Adleman. of course. Shamir. 1 In the Eastern European socialist countries. and untraceability. So governments feel a desire to regulate not just who can issue e-money but also the types of e-money product that can be offered. fee. The authors are responsible for any remaining errors. innovations were made by the private sector to escape regulation (for instance.

19 Supposing an American software developer uses a server in France to sell his software. to sterilize the effects of large purchases in the foreign exchange markets. or scheme operators might be required to monitor transactions.” March). Australia. the actual cost of electronic money transfers will be recognized as negligibly small. as we have previously written. or insert an unauthorized software program into a user’s personal computer that enables the attacker to copy electronic notes stored or in transmission. RTGS systems are already in operation in the Czech Republic. and security. the reader may contact http://junon. . total government spending as a percentage rate of GDP rose from an average of 8. in a note-based system. storage. The use of RTGS systems is also growing outside the Group of Ten and the European union. business cycles are directly related to weaknesses in the demand side of the economy. particularly in countries with large budget deficits. it is possible for the government to stimulate demand by increasing spending. 28 Jordan and Stevens (1997). banks and individuals. 31 If the cost of connecting the Internet and personal computers is taken into account. credit cards are subject to fraud and counterfeiting losses that are estimated to be well over one billion US dollars each year. transport. and it is reported that. 21 We suppose that rs > rMi > rXg.2 percent in 1995. among others. Expected costs of loss and theft are similar to paper currency. The theory is that since government expenditure is a significant accounting component of demand. in OECD countries. and the expected cost of a card becoming unbearable. Which sales tax rate should be applied. For example. so that a very extensive spread of electronic money could shrink central bank balance sheets significantly. 25 For example. 22 For more details regarding the mathematical demonstration.284 Jean-Pierre Centi and Gilbert Bougi allowed to hold or on user-to-user transactions. 16 For example. the cost of electronic money is high. 15 For example. even a moderate loss of seigniorage could be of concern to some governments. 30 Rahn (1999: 156–7). an inter-governmental body created by the G-7 countries in 1989. an attacker could intercept messages between a genuine user and an issuer. 32 It is a summary table (European Central Bank 1998). where rXg is the rate of return on goods.” 29 That does not exclude. The authors suggest that the income velocity of base money could approach infinity: “what may be new and different about the 21st century is the possibility that the velocity of central bank money might approach infinity. 23 It also includes the expected cost of loss and theft. 26 Cash is a large or the largest component of central bank liabilities in many countries. 27 Moreover.3 percent in 1870 to 47. 17 According to Keynesian theory. which includes costs of processing and accounting of money. estimates of the amount of money laundered annually worldwide from the illicit drug trade alone range between 300 and 500 billion US dollars. and by whom? Which country should benefit from the tax? 20 According to the Financial Action Task Force. 33 The estimated annual costs of handling central bank currency by US retailers and banks are 60 billion US dollar.u-3mrs. The Group of Ten is now using RTGS systems. 18 They show China and Saudi Arabia will introduce RTGS systems in the near future (see Bank for International Settlements 1997. 24 The cost of conversion includes opportunity cost of time spent for this activity (for example. Hong Kong and Thailand. say to a customer in Prague. But with the recent explosion of the Internet and its attraction to businesses. and then use the notes to perform transactions. the constitution of bank reserves in electronic units. walking to a smart card reader) and per transaction fees charged by the issuer of electronic money and/or the provider of the telecommunication service. “Real Time Gross Settlement Systems.

Santomero. (1999) “Emergence of money as a medium of exchange: An experimental study. F. (1992) “Achieving electronic privacy. E. 70. (1985) “What’s different about banks?. W.) The Future of Money in the Information Age. Rahn. Tobin.” Journal of Money. (1952) “The transaction demand of cash: An inventory theoric approach. (1892) “On the origin of money. R. Duffy. New York: Oxford University Press. pp. and Banking 28(4). E.H. Jordan.” Journal of Monetary Economics 15: 29–39.A. ch. Hayek. Economic consequences of electronic money 285 34 Historical monetary research shows that banking regulations have increased the frequency of recent bank failures. G.” The Economic Journal II(6): 239–55.A.” American Economic Review 35(4): 519–30. and Wright. and Wright. (1993) “A search-theoretic approach to monetary economics. (1989) “A contribution to the pure theory of money.” Quarterly Journal of Economics 66: 545–56. Dorn (ed. Chaum. 97–100. and Ochs. (1936/1990) The Rationale of Central Banking and the Free Banking Alternative. Selgin. (1996) “Alternative monies and the demand for media of exchange. Selgin. (1997) “E-money: Friend or foe monetarism?. N.” Journal of Economic Literature 32: 1718–49. Washington: Discovery Institute. Fama.” Journal of Monetary Economics 6: 39–57. Credit. (1956) “The interest-elasticity of transactions demand for cash. (1871) Principles of Economics.) The Future of Money in the Information Age.” Scientific American. Fama. and Stevens.” Journal of Money. and White. Osterfeld. 149–73. Hayek. J. Kiyotaki. (1945) “The use of knowledge in society.” Review of Economics and Statistics 38(3): 241–247. (1999) The End of Money and the Struggle for Financial Privacy. In the United States. D. (1974) “Competitive supply of money. G. F. Credit and Banking 6(4): 423–54. New York: Cambridge University Press. Washington: The Cato Institute. Basel. Menger.” Federal Reserve Bank of Minneapolis Staff Report 123: 37. and Seater. New York: New York University Press. References Baumol. L. Menger. Indianapolis: Liberty Press. Brennan and L.” in J. (1997) “Money in the 21st century. Washington: The Cato Institute. (1978) Denationalisation of Money: The Argument Refined. J. R. Klein. August. New Essays in Democratic Thought. D.” American Economic Review 89(4): 847–77.” American Economic Review 83(1): 63–77. Lomasky (eds) Politics and Process. Hobart paper no. B. E. Smith. pp. 115–25. (1980) “Banking in the theory of finance. (1989) “Radical federalism: responsiveness. conflict. 96–101. J. (1994) “How would the invisible hand handle money. (1997) Virtual Money: Understanding the Power and Risks of Money’s High-Speed Journey into Electronic Space. E. C. federal deposit guarantees have undermined depositor discipline and encouraged banks to take greater risks. London: Institute of Economic Affairs (1st edition 1976). J.” in G. pp. R. 6.” in J. Solomon. A. V. Kiyotaki. and efficiency. Dorn (ed. Other reading Bank for International Settlements (1996) Implications for central banks of the development of electronic money. . C. N. J.

286 Jean-Pierre Centi and Gilbert Bougi Committee on Banking Supervision (1998) Risk management for electronic banking and electronic money activities. Basel. Frankfurt. . European Central Bank (1998) Report on electronic money.

Part VII The legal framework .


centralized forms of lawmaking. however. self-regulating market practices. In other words. and according to which positive law is the only possible form of public order. anthropology and of course legal philosophy have been invoked to study the relationships between social. but which has considerably become up to date in recent years with the develop- ment of the Internet. is a direct and obvious challenge to the tradition of legal positivism. decentralized process through which indi- vidual actors themselves develop and experiment rules. to an important extent. Can they produce anything that might be labeled law? Can they engender legal rules. practices and solutions which help them face the problems posed by electronic transactions. they are generally reluctant to consider private. an issue which for all times has constituted a fundamental and controversial preoccupation for legal theory. The spontaneous emergence of a private. the legal system. While social sciences such as sociology. alternative means to overcome legal problems as being part of what they call law. The Internet. indeed deeply challenge our common understanding of law. but. We will therefore have to invoke the problem of the definition of law.14 The emergence of private lawmaking on the Internet Implications for the economic analysis of law Elisabeth Krecké Introduction Most lawyers share the prevailing legal conception in which command. centralized systems of Internet governance? These questions will lead us to deal with more general issues such as to know who sets the moves of the legal system. Emphasizing the decentralized and private nature of the demand for legal change that accompanies the development of the new economy. and by which forces the law is shaped and adapted to change. competing with the traditional. history. As a consequence. may advance through events and decisions that are extra-legal. self-governing process within cyberspace and its astounding ability to address major problems that previously might have been thought to require a centralized form of law production. economic and technological changes and the extent . obedience and repression appear as essential attributes of law. this chapter argues that sometimes the shaping of legal evolutions emanates from sources that are outside the rationale and the doctrines which have erstwhile ruled the law. more fundamentally. procedures or institutions. Not only does the new economy make apparent the limits of traditional. it allows for the emergence of a new. This chapter discusses the legal nature of these emerging.

which in many situations reveal themselves as largely unsuited to the task of governing such a borderless. Competing governance systems for the Internet Shortcomings of traditional forms of lawmaking in the context of the new economy The Internet continually raises a range of new and complex legal problems which. and in this sense. Interventions of a national authority in an international context could only have a partial impact. the methodology of economic analysis brings into the debates of legal theory a new perspective which could be particularly enriching in fields which are often described as legal voids. I will try to discern the implications of economic theory. major issues such as domain names and at the same time countless minor. Yet. territorially based legal rules in this context. and in particular the inter- disciplinary movement of law and economics. To go a step further. the idea that a centralized institutional structure could span the impressive range of issues raised by the worldwide web seems rather dubious and even undesirable. it is difficult to apply current. but there are also far-reaching questions such as the balancing of powers within the new organization. because they can hardly control activities for which physical location cannot be established with certainty. questions of content of policies and regulations. changeable and rapidly growing entity. invoke the role economics might play in pushing legal change. The problem of relationships between different national legal regimes. Because of the complex nature of the Internet. Such an organization would have to cope not only with problems of jurisdiction. for legal theory. the legitimacy and the enforcement of its decisions. if they are not solved rapidly and efficiently. unless the authority seriously transgresses her legitimate power. risk impeding the development of the enormous potential that the new economy is likely to offer our societies. territorially based legal institutions. My purpose is to prospect some of the ways in which economic analysis can contribute to an understanding of the relationship between the economic order and the evolution of legal systems. as well as the . Strengthening the role of international law. As the Internet is not a conventional territorial entity. Nevertheless. concrete issues. or even establishing a supra- territorial. yet they would pose considerable problems (Gould 1997). ( Johnson and Post 1997). the protection of unpopular minorities. the role of economics has been largely underestimated. etc. National legal systems are not really suited for the task either. such as the seemingly ungovernable cyberspace.290 Elisabeth Krecké to which these changes may engender legal change. the non-geographical and decentralized character of the Internet impairs a lot of traditional. the risk of capture of the organization by particular governments or industrial rent-seeking interests. Considerable informational as well as motivational obstacles cast serious doubts on the efficiency of such a model of governance for the Internet. internationally based model of governance are repeatedly discussed options. Legitimating national policies which are likely to directly or indirectly affect the jurisdictions of other countries would be a highly problematic enterprise.

governments were generally reluctant to open this market to competition.2 The Internet is also throwing up technological developments traditional lawmaking was not prepared to deal with for political reasons. whereby most of the time national and international concerns overlap. Legislators and regulators are often hopelessly overtaken by events in cyberspace and have serious difficulties not only to reach the fixed objectives. are important obstacles standing in the way of traditional. public and political characteristics. Private lawmaking on the Internet 291 conflicts that may appear among governments and already existing network authorities. under the strong pressure of the technological evolution within cyberspace. sometimes unreadable and thus unworkable pieces of legislation. considered as a mili- tary weapon. unknown and changeable context. So far. private forms of encryption. Complexity is enhanced by the fact that the Internet concerns various. have been monopolized by public authorities. sysops. how- ever. infringement of privacy. such as. as well as the consequences of the interaction of legislation with existing economic or legal policies. encryption technologies. and in particular international policies. Permanent and rapid changes in technology put an unprecedented pressure on the law. More importantly. or violation of property rights). it is not obvious to identify without ambiguity the legal liability of multiple actors such as domain name registries. con- glomerate trusts which largely dominate the fields of media and communication. a complex movement of liberalization of cryptographic policies appeared over the past few years in most industrial countries..1 The difficulties of governing the Internet in a centralized way furthermore engender a vagueness which inevitably leads to a dilution of responsibility among involved actors. as well as private law features. Yet. The governance of the Internet indeed has a national and at the same time an international dimension. Furthermore it involves constitutional. for which history offers no reference model. The biggest challenge of any centralized model of Internet governance is.3 The encryption example illustrates particularly well the fact that legislators have to deal with the often conflicting demands of politics. organizations and the like. highly different aspects of law. because market conditions have changed between the conception of the laws and their application. leading to the emer- gence of a private market for encryption products in cyberspace. etc. the speed of technological evolution. Legislators further have to face the strategic behavior of giant. centralized forms of Internet governance. unforeseeable side-effects risk transforming well-intended efforts to regulate cyberspace into a subtle threat to the development of the new. for instance. but also to define adequate objectives. Attempts to anticipate future changes in technology have in many cases led to the creation of highly complex. which sometimes renders laws obsolete as soon as they emerge. Regulating the new econ- omy would involve long and difficult processes of learning and adaptation in an entirely new. but certainly also because encryption procured them a significant instrument of power. nor is it easy to define in a satisfactory manner offenses with respect to online activities (for instance in the fields of financial criminality. For reasons of national security. international treaties. Indeed. users. . emerging economy within cyberspace.

the Internet community had the chance to develop and continually test its own devices to cope with new. users have started to elaborate and customize sets of commercial norms and rules. Domain name and IP address registries. Based on the protection of tangible goods. for instance. create among economic actors an unprecedented interest for encryption technologies. rendering obsolete most of the strict legislation that has prevailed so far in this field.”4 As the constitution of reputation plays a crucial role in the context of Internet transactions in which anonymity undeniably creates a climate of mistrust and uncertainty. These seals progressively get accepted within cyber-communities as useful signals of reliability. for instance.292 Elisabeth Krecké and which use their extensive market powers to lobby for laws and policies that support their industrial interests. Users are developing genuine standards of behavior to be applied. in order to adjust them to their needs. adapted to the specificity of dematerialized business. Diverse Internet charters and codes of ethics recapitulate consensual. in news groups or chat rooms. and for which the established legal system is largely disarmed. which local rules to adopt. In cyberspace.” have been summed up under the label “lex electronica. more and more sites. they might allow parties to modify the legal rules that apply to their transactions. new contractual arrangements. Sysops are adopting rules determining which users to allow to sign on. in particular with respect to copyright. As mentioned earlier. Contracts can to a large extent avoid conflict by specifying the national laws that regulate the deal. etc. where consequential new needs appear. such as. for instance. which filters to install. the traditional legal system is not really fitted to deal with forms of property “that can appear with the speed of light” . which. As it could largely escape from authoritative interventions. Encryption is an additional method for protecting privacy and property in the context of digital business. contract has naturally imposed itself as a primary element of self- regulation. privacy or copyright. More importantly. Concerning electronic transactions. the necessity to protect the integrity and confidentiality of messages. As a consequence. The legal problems raised by the Internet. (Johnson and Post 1997: 67–8). with reference to the medieval “lex mercatoria. providers or organizations grant seals to traders which have proved their integrity and trustworthiness over longer periods of time. a new. with which other systems to connect. private market for encryption products is emerging within cyberspace. emerging problems. They are designed to cope with the particular problems posed by com- merce in a virtual world. have been setting up rules regarding the conditions to impose on the possession of an online address (Johnson and Post 1997: 67). top-down schemes of governance that may explain why the new economy could flourish the way it did. more or less informal and constraining rules of conduct prevailing within cyber- communities with regard to issues like politeness. The Internet permanently gives rise to new issues of protecting intellectual property rights that are unique to the technology. The comparative virtues of Internet self-governance It is precisely the shortcomings of centralized. are constantly developed and spread throughout the Internet.

for instance. as much as possible. or it tries to adapt the commercialization of music to the new conditions of the market. It is also developing devices which help cybernauts to cope with disputes once they have occurred. regulatory or judicial solutions. the music industry. or tying arrangements like online assistance for legitimate customers of software. thus contributing to a large-scale process of discovery and experimentation of new fencing techniques in the Internet. as extensions of property rights to new objects. Keys and passwords not only permit traders to select those they want to deal with. disputes in electronic transactions. In this perspective. digital objects (new software for instance). In the absence of an efficient traditional copyright protection in the field of digital property. such as regular updates. encryption technologies developed on the Internet appeared very soon as indispensable factors for the development of the new economy. the efforts of protecting legitimate property emanate from interested persons themselves (Mackaay 1997 2000). it either promotes the development of protection technologies itself. 2000). could not afford to rely on their governments to identify and sanction the numerous users without licenses.” “Online Ombuds Office. seriously concerned about the impacts of the increasingly effortless technological possibilities to download music from the Internet illegally. Assuring innovators the control over new. in as far as it provides the means to create a high degree of assurance of authenticity (Benson 2000: 26).” “CyberSettle” or “ClickNSettle” are examples of online arbitration and mediation which have had considerable success within cyber- communities. specific marketing techniques. Similarly. Solutions. who have started to develop new forms of encryption. property owners indeed had no alternative but to elaborate themselves new. can be viewed as fencing techniques adapted to cyberspace. but more fundamentally. Similarly. encryption produces trust and confidentiality among contractors. but from those who commercialize the new objects (Mackaay 1997. Instead. Private lawmaking on the Internet 293 (Schwartzstein 1994: 1068). however. The Internet not only provides private means to foresee and prevent. settlement systems and fencing techniques created by individuals or small groups seems to be increasingly approved by large communities of cybernauts as alternative means of dispute prevention and resolution. decided not to wait for legislative. are coming from the concerned persons themselves. encryption technologies furthermore function as what Mackaay calls virtual fences. As violators can be expelled instantaneously. services for settling disputes online have been multiplied over the past few years. The development of private means of property protection becomes an important variable in economic choice when protection from the traditional legal system largely fails. Thus. Here again. Allowing for a considerable reduction of transaction costs. For a long time they have been exploring their own technical solutions allowing for a relatively efficient protection of their property interests. exten- sions which do not emanate from legislation or judicial decision. Private and semi-private initiatives such as “e-resolution. it also allows easy enforcement (Benson 2000: 26). workable ways to protect their online activities. Software producers. Mechanisms that reveal themselves as profitable for many users . a growing number of successful rules.” “Virtual Magistrate.

in a process of private. developing and exchanging new technological and contractual devices allowing them to enhance the security and profitability of their transactions on the Net. progressively expand to larger communities. As Mackaay points out.294 Elisabeth Krecké (allowing them to efficiently protect their interest in online transactions by reducing costs linked to problems of trust. credibility and uncertainty). and finally become generally accepted. They work to a large extent without lawyers. nor do they require approval by the government. court decisions. Just like any other good. Entrepreneurial decisions thus initiate most evolutions that take place in cyberspace. entrepreneurial innovation. in particular fencing techniques. The shortcomings of traditional law and the existence of profit opportunities linked to the production of fencing techniques create incentives for entrepreneurs to become builders and inventors of fences. are in no way tied to the law of any particular