Internet and digital economics principles, methods and applications

His research agenda focuses on the economics of institutionsand the economics of contracts, with two main applied fields: theeconomics of intellectual property rights and the economics o

Internet and Digital Economics

Principles, Methods and Applications

How are our societies being transformed by Internet and digital nomics? This book provides an accessible introduction to the econom-ics of the Internet and a comprehensive account of the mechanisms ofthe digital economy Leading scholars examine the original economicand business models being developed as a result of the Internet system,and explore their impact on our economies and societies Key issuesare analyzed, including the development of open source software andonline communities, peer-to-peer and online sharing of cultural goods,electronic markets and the rise of new information intermediaries,e-retailing and e-banking The volume examines how Internet anddigital economics have transformed the organization of firms, indus-tries, markets, commerce, modes of distribution, money, finance, andinnovation processes, and provides the analytical tools to understandboth these recent transformations and the likely future directions of the

Internet and Digital Economics

edited by

Eric Brousseau and Nicolas Curien

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK

First published in print format

Information on this title: www.cambridge.org/9780521855914

This publication is in copyright Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press

Published in the United States of America by Cambridge University Press, New Yorkwww.cambridge.org

hardback

paperbackpaperback

eBook (EBL)eBook (EBL)hardback

To Delphine

List of figure s page xList of table s xiiiNotes on contributo rs xivAcknowl edgement s xxv

E R I C B R O U S S E A U A N D N I C O L A S C U R I E N 1

Part I Towa rd a new econom y? 57

Part II On-line communi ties 171

M I C H E L G E N S O L L E N 173

of the digital commons

Pa rt III Ne twork exte rna lities and ma rket

Pa rt IV Produ cing, distrib uting and sharin g

inform ation goods 311

Pa rt V How e-marke ts perform 423

PAT R I C K B A J A R I A N D A L I H O R TA C ¸ S U 425

M A A R T E N C W J A N S S E N, J O S E ´ L U I S M O R A G A

-G O N Z A ´ L E Z A N D M AT T H I J S R W I L D E N B E E S T 460

and coalition strategy in Internet retailing

J A C Q U E S L AY E A N D H E R V E ´ TA N G U Y 484

M A R C B O U R R E A U A N D C H R I S T I A N L I C O P P E 510

Part VI Ev olving instituti onal infr astruc tures 537

B R U N O D E F F A I N S, YA N N I C K G A B U T H Y A N D

P H I L I P P E F E N O G L I O 539

perspectives in economics of banking

D AV I D B O U N I E A N D P I E R R E G A Z E´ 569

D E L P H I N E S A B O U R I N A N D T H O M A S S E R VA L 588

“second rank” institutional framework exist?

E R I C B R O U S S E A U 617

Part VII The impact s of the Internet at the macr o level 649

consumer welfare

G A RY M A D D E N, M I C H A E L S C H I P P A N D J O A C H I M TA N 651

or divergence in cross-country trends?

K E N N E T H L K R A E M E R A N D J A S O N D E D R I C K 663

A L A I N R A L L E T A N D F A B R I C E R O C H E L A N D E T 693

Refer ences 718Index 777

2.1 Cisco’s stock options, 1990–2004 page 91

option programs, ICT companies operating in the

semiconductors, Silicon Valley, Route 128, Dallas,

software publishers, Silicon Valley, Route 128, Dallas,

2.10 Stock price movements, Cisco, Lucent, AT&T, and IBM

x

8.7 Choice between compatibility and incompatibility 254

with i.i.d valuations uniformly distributed in [0,1]

non-monotonic function of competencies – the most

16.4 The impact of lower search cost c 477

and the pure ELOB for different values of c and M 606

24.10 Performance impacts of e-business in different

2.1 Employment, 1996–2003, at the top 20

“Old Economy” companies by 2003 sales page 672.2 Employment, 1996–2003, at the top 20

2.3 Top five companies by worldwide sales in computer

2.4 Cisco Systems acquisitions, by value, employees, and

6.2 Contribution and trust issues in experience-sharing

11.1 Equilibrium quantities, prices, and revenues in different

xiii

PAT R I C K B A J A R I is Professor at the University of Minnesota He ceived his PhD from the University of Minnesota in 1997 He teaches

re-in the areas of re-industrial organization and applied econometrics Hiscurrent research includes the econometrics of strategic interactions,demand estimation in differentiated product markets, and the empiri-cal analysis of asymmetric information He is Managing Editor of the

the Journal of Business and Economic Statistics

YA N N I S B A K O S is Associate Professor of Management at the Leonard

N Stern School of Business at New York University where he teachescourses on the economic and business implications of informationtechnology, the Internet, and online media Professor Bakos pioneeredresearch on the impact of information technology on markets, and inparticular on how Internet-based electronic marketplaces will affect pri-cing and competition He holds PhD, MBA, Masters and Bachelorsdegrees from MIT

P I E R R E-J E A N B E N G H O Z Iis presently Reseach Director at the NationalCenter for Scientific Research (CNRS) and is directing the Pole forResearch in Economics and Management at Ecole Polytechnique(Paris) He developed a research group on information technology,telecomunications, media, and culture His current projects drawattention to adoption and uses of ITC in large organizations, structur-ing of e-commerce and ITC-supported markets and supply chains.Pierre-Jean Benghozi publishes on these topics in French and English

He teaches regularly at Paris University

D AV I D B O U N I E is Assistant Professor of Economics at Te´le´com Paris

He completed his graduate work at the University of Paris 1Panthe´on-Sorbonne in public economics and received his PhD ineconomics from ENST His research interests include economics ofpayment systems and Internet economics He is currently working on

xiv

economics of payment and econometrics of consumer payment ior Prior to this affiliation with Te´le´com Paris, he served for three years

behav-as an economist with the Groupement des Cartes Bancaires (CB), theleading interbank payment and cash withdrawal system in France

M A R C B O U R R E A U is Assistant Professor at Ecole Nationale Supe´rieuredes Te´le´communications (ENST, Paris) and member of the Laboratory

of Industrial Economics (LEI) of the Center for Research in Economicsand Statistics (CREST) His main research interests are economic andpolicy issues relating to broadcasting, telecommunications, and theInternet

E R I C B R O U S S E A U is Professor of Economics at the University ofParis X, and Director of EconomiX, a joint research center betweenthe CNRS (French National Science Foundation) and the University

of Paris X He is also Co-Director of the GDR TICS (ResearchConsortium “Information Technologies and the Society”) of theCNRS His research agenda focuses on the economics of institutionsand the economics of contracts, with two main applied fields: theeconomics of intellectual property rights and the economics of theInternet and digital activities On this last issue he works both ondigital business models and on the governance of the Internet and ofthe information society

E R I K B R Y N J O L F S S O N is Director of the MIT Center for DigitalBusiness, the Schussel Professor of Management at the MIT SloanSchool, and director or advisor of several technology-intensive firms.His research focuses on the economics of information and informationtechnology, including productivity, organizational change, and thepricing of digital goods Professor Brynjolfsson previously taught atStanford Business School and Harvard Business School He holdsBachelors and Masters degrees from Harvard University and a PhDfrom MIT

N I C O L A S C U R I E N serves as Commissioner for the French RegulationCommission for Electronic Communications and Postal Services(ARCEP) He held several positions as an economist in theAdministration (France Telecom, Ministry of Defence, Institute ofStatistics and Economic Studies) and as an academic (Professor ofEconomics at the Conservatoire National des Arts et Me´tiers and atthe Ecole Polytechnique) He is also a member of the French Academy

of Technology and a member of the International TelecommunicationsSociety (ITS) He has published several books and many scientificarticles in the field of telecommunications and Internet economics

J E A N-M I C H E L D A L L E is Professor with University Pierre et MarieCurie and an associate researcher with IMRI-Dauphine (Paris,France) He works on the economics of innovation and has notablyfocused since 1998 on the economics of software and open sourcesoftware He is an alumnus from Ecole Polytechnique and ENSAE,and holds a PhD in economics from Ecole Polytechnique He is alsothe Managing Director of Agoranov, a major non-profit science-basedincubator located in Paris.

G O D E F R O Y D A N G N G U Y E N is Deputy Scientific Director at ENSTBretagne, an engineer’s school in France, and Professor at the College

of Europe in Bruges His research focuses on Internet and munications economics, on which he has written two books, and oninstitutional issues related to ICT He is the scientific director of anetwork of seven universities and institutions located in Brittany Thisnetwork carries out statistical and case studies on adoption and usage

telecom-of ICT Prtelecom-ofessor Dang Nguyen has been an expert with the EuropeanCommission since 1983, and has been consultant for many institu-tions including the World Bank and ITU

PA U L A D AV I D is known internationally for contributions to economichistory, economic and historical demography, and the economics ofscience and technology He divides his working life equally betweenStanford University in California, where he is Professor of Economicsand Senior Fellow of the Stanford Institute for Economic PolicyResearch, and the University of Oxford, where he is Senior Fellow

of the Oxford Internet Institute and Emeritus Fellow of All SoulsCollege In 2003 he edited (with M Thomas) The Economic Future

in Historical Perspective

J A S O N D E D R I C K is Co-Director of the Personal Computing IndustryCenter and Senior Research Fellow at the Center for Research onInformation Technology and Organizations (CRITO), at the University

of California, Irvine His research is focused on the globalization ofinformation technology production and use, and the economicimpacts of IT at the firm, industry, and national levels He is co-author

of Global E-Commerce: Impacts of National Environment and Policy,Cambridge University Press, 2006

B R U N O D E F F A I N S is Professor of Economics at University of Nancy,France For the past few years, he has been Vice-President forresearch activities there and he developed a new program in law andeconomics which is now associated to the CNRS He is the author ofarticles and books in law and economics His research has focused on

econom ics of accident s law, confl icts res olution, int ernet regu lation,comparis on of legal systems , as well as the questi on of the extent towhich law cons trains econom ic growth.

N I C H O L A S E C O N O M I D E S is Professor of Econom ics at the SternSchoo l of Busine ss of New York Univ ersity an d Exec utive Di rector

of the NET Institute His fields of specia lization an d research includethe eco nomics of networ ks, especial ly of te lecommun ications , com -puters , and informa tion, the eco nomics of te chnical com patibility andstanda rdization, ind ustrial org anizatio n, the stru cture and organiza -tion of financia l market s, app lication of publ ic policy to netwo rkindustrie s, and strategi c an alysis of markets He h as publish ed wide ly

in the areas of netw orks, telecomm unicatio ns, oligop oly, antitrus t,product positioning, and on liquidity and the organization of financialmarkets and exchanges His website on the economics of networks athttp:// www.stern nyu.edu/ netwo rks/ has been ranked by The Economi st

as one of the top four economics sites worldwide

E M M A N U E L L E F A U C H A R T is Assistant Professor of Economics at theLaboratory of Econometrics at CNAM, France Her interest in re-search is the field of industrial organization, and more particularlyinformation-sharing patterns, industrial dynamics, firm demography,and the economics of online communities

P H I L I P P E F E N O G L I Ois Assistant Professor of Economics at University

of Nancy (France) His research interests include industrial zation (oligopolistic competition, product differentiation) and eco-nomics of innovation and new technology He is also working on theeconomics of alcohol, tobacco, and illegal drugs and is the co-author

organi-of several reports on this topic for public organizations

PAT R I C E F L I C H Yis Professor of Sociology at Marne la Valle´e University,LATTS His main research topics include online computerizationmovements and technological utopianism

D O M I N I Q U E F O R AY holds the Chair of Economics and Management ofInnovation at the Swiss Federal Institute of Technology in Lausanne(EPFL) He is also Director of the Colle`ge du Management at EPFL.His research interests include all topics and issues related to economicpolicy in the context of the new knowledge-based economy This broadfield covers the economics of science, technology, and innovation.Intellectual property and competition policy, information technologyand the new economy, capital market and entrepreneurships, nationalsystems of innovation, education, and training policy are fields of highrelevance in his research In 2004 he published The Economics of Knowledge

YA N N I C K G A B U T H Y is Assistant Professor of Economics at theUniversity of Nancy and member of the BETA research laboratory(CNRS) since 2005 He received his PhD in economics from theUniversity of Lyon in 2003 and was a visitor at the University ofEssex in 2001 His main area of research is law and economics, withspecial interest in dispute resolution (bargaining and arbitration) Heconducts experimental research on these topics.

A L E X G A U D E U L is a lecturer at the School of Economics, University ofEast Anglia and a faculty member at the Centre for CompetitionPolicy He holds a PhD from the University of Toulouse, France and

is an industrial economist with an interest in the Internet, open sourcesoftware and media industries Current work examines the opensource software production model in order to evaluate its impact ontraditional methods of software production He is also working on howintermediaries regulate competition on the Internet between the firmsthey intermediate

P I E R R E G A Z E ´ is Assistant Professor of Economics at the University ofOrle´ans since 1999 He received his PhD in economics from theUniversity of Orle´ans His research interests are in economics ofbanking and industrial organization His recent works include bund-ling and tying practices in the banking industry and retail paymentsinnovations

M I C H E L G E N S O L L E N was trained as an economist and an engineer intelecommunications From 1990 to 2000 he was Chief Economist atFrance Telecom, in charge of the Economic and Strategic Studiesdepartment He is currently working at the SES (Economics andSocial Sciences) department at Te´le´com Paris His recent publicationsfocus on electronic commerce, network-based firms, informationeconomy, and the new business models triggered by the development

of the Internet and ICT

U L R I C H H E G E is Associate Professor in Finance at HEC School

of Management in Paris He has previously taught at TilburgUniversity and ESSEC and holds a PhD from Princeton University.His research interests are in corporate finance, on questions related toventure capital, corporate governance, joint ventures, bankruptcy,credit risk and credit structure, and internal capital markets

A L I H O R T A C ¸ S U is Assistant Professor of Economics at the University ofChicago He is interested in the theory and econometrics of auctionand matching markets, product markets with search frictions and

information asymmetries, and the vertical arrangement of industries.

He conducts empirical research on Internet auctions and matchmakingsites, government securities markets, electricity markets, the mutualfund industry, and the cement/concrete industry

M A A R T E N J A N S S E N is Professor of Microeconomics at ErasmusUniversity Rotterdam and Director of the Tinbergen Institute He haspublished numerous articles in internationally refereed journals such asthe Review of Economic Studies, International Economic Review, Journal ofEconomic Theory and Games, and Economic Behavior His current re-search interests are in incomplete information in markets and auctions

B R U N O J U L L I E Nis Director of Research at CNRS and a senior member

at IDEI He graduated from Ecole Polytechnique and ENSAE, andholds a PhD in economics from Harvard University His contributionscover a wide range of topics, with a particular strength in industrialorganization, contract theory, network economics, competition policyand regulation, and the economics of risk

K E N N E T H L K R A E M E R is Professor of Management and ComputerScience and Director of the Center for Research on IT and Organiza-tions, University of California, Irvine His research includes the socio-economic implications of IT, national policies for IT production anduse, and the contribution of IT to productivity and economic develop-ment His most recent book, co-authored with Jason Dedrick, is Asia’s

five-year study of the global diffusion of the Internet and e-commerce(Cambridge University Press, 2006) and is starting a new study of theoffshoring of knowledge work

G I L B E R T L A F F O N D is Professor of Economics and Statistics at theConservatoire National des Arts et Me´tiers (Paris) and Head of theLaboratory of Econometrics His main research concerns are in gametheory, public choice, and evolutionary economics, fields where he haspublished several major theoretical contributions He is also interested

in industrial organization and has worked more specifically on cations to the telecommunications industry and to the Internet

appli-J A C Q U E S L AY E has a PhD in economics from the Laboratory ofEconometrics (Ecole Polytechnique) and is a researcher at theLaboratory of Forestal Economics (LEF, Inra Nancy, France) where

he works on industrial organization and computational economics(oligopolistic competition, vertical relationships, product differenti-ation, coalition theory) applied to sectors like the wine industry, the

wood industry, or Internet retailing Other fields of research concernoperations research (stochastic control) applied to the management ofrenewable ressources, and computational economics.

WILLIAM LAZONICKis University Professor at the University of MassachusettsLowell and Distinguished Research Professor at INSEAD Previously

he was Assistant and Associate Professor of Economics at HarvardUniversity (1975–1984) and Professor of Economics at BarnardCollege of Columbia University (1985–1993) He specializes in theanalysis of industrial development and international competition

V I R G I N I E L E T H I A I Sis a lecturer in economics at the ENST-Bretagne, aFrench graduate engineering school in telecommunications She is amember of the CREM, the CNRS Research Laboratory on Economicsand Management of the Universities of Rennes and Caen She usesindustrial economics and applied econometrics to study pricing on theInternet and to analyze diffusion and uses of information and com-munication technologies

C H R I S T I A N L I C O P P E, PhD, trained in the history of science andtechnology, is developing a research program on the uses of infor-mation and communication technologies and the ways they are em-bedded in various forms of activity, such as commerce and servicerelationships, mobility and displacements, and forms of interpersonalcommunication He is currently Professor of Sociology of Technology

at the Social Science department of the Ecole Nationale Supe´rieure desTe´le´communications in Paris

G A R Y M A D D E N is Director of the Communication Economics andElectronic Markets Research Centre and Professor at the Department

of Economics of the Curtin University of Technology, Australia He iseditor of the International Handbook of Telecommunications Economicsseries, and serves on the board of management of the InternationalTelecommunications Society He is also a member of the editorialboard of the Journal of Media Economics

S E ´ B A S T I E N M I C H E N A U Dis a PhD candidate in finance at HEC School

of Management, in Paris, and is affiliated with GREGHEC He haspreviously worked at Bankers Trust International in mergers andacquisitions and as a consultant at Arkwright Enterprises where hewas an active observer of the Internet development His researchinterests are in corporate finance and behavioral finance

J O S E ´ L U I S M O R A G A-G O N Z A ´ L E Zis Professor of Industrial Organization

at the University of Groningen He obtained a PhD in Economics

from University Carlos III Madrid His research focuses mainly oninformation economics and R&D His work has been published

in journals including the Review of Economic Studies, Rand Journal

of Economics, European Economic Review and International Journal ofIndustrial Organization

F R A N C ¸ O I S M O R E A U is Associate Professor in Economics at theConservatoire National des Arts et Me´tiers in Paris, France His maincurrent fields of research are the economics of the Internet and mediaeconomics

T H O M A S PA R I S is Affiliated Professor at HEC School of Management,Paris, and Associate Researcher at CRG Ecole Polytechnique Hisresearch focuses on the management, the economics, and the regula-tion of creative and immaterial industries He has written a book onthe system of author’s rights (Le Droit d’auteur – L’ide´ologie et le syste`me,PUF, Paris, 2002) and has edited one about the challenges of thetelevision industry (La Libe´ration audiovisuelle – Enjeux technologiques,e´conomiques et re´glementaires, Dalloz, Paris, 2004)

T H I E R R Y P E ´ N A R D is Professor of Economics at the University ofRennes 1 and is Director of the Industrial Organization department

at the Center for Research in Economics and Management (CREM).His main fields of interest are the economics of networks (Internet,telecommunication), the economics of contract (franchising), gametheory, and antitrust policy

A L A I N R A L L E Tis Full Professor of Economics at the University of ParisSouth and researcher at ADIS, research center in economics there.His fields of interest are Internet economics (e-commerce, emergence

of online multimedia markets on the Internet, the impact of ICTs onorganizations) and regional and urban economics (ICTs and localiza-tion of economic activities, network infrastructure, and local develop-ment)

F A B R I C E R O C H E L A N D E T holds a PhD from the University of ParisPanthe´on-Sorbonne His dissertation addressed the question of copy-right in the digital economy He is currently Assistant Professor ofEconomics at the University of Paris 11 His research focuses on theeconomic impact of digital technologies through topics such as elec-tronic delivery of cultural good, broadcasting, intellectual property,and digital divide He has published several papers on the digitalprotection of contents, the efficiency of copyright collecting societies,and the European copyright harmonisation

D E L P H I N E S A B O U R I N was a PhD student at CEREMADE, ParisDauphine University, Research Fellow at CREST, and teaching as-sistant at Paris Dauphine University Her research interests includedthe microstructure of securities markets, with emphasis on the compe-tition between alternative trading mechanisms and corporate finance.

M I C H A E L S C H I P P is Research Associate at the CommunicationEconomics and Electronic Markets research centre (CEEM) at theCurtin University of Technology His research focus is networkeconomics This interest led Michael to empirically measure the mag-nitude of the Internet network effect Michael is currently developinghis expertise in e-market competition modelling An early outcome ofthis research program is a model of network investment contained in acompetitive game between an incumbent firm and a virtual entrant

T H O M A S S E R VA Lis a graduate from Ecole Normale Supe´rieure in Parisand Ecole Nationale de la Statistique et de l’Administration Economique

He focused his PhD work on Internet economics with a special est in transformation of electronic capital market (ECN), law andeconomics (software patent), and value creation in cyberspace Thomas

inter-is the President and CEO of Baracoda and Vice President of the Frenchchapter of the Internet Society

J O A C H I M T A N is Research Associate at the Communication Economicsand Electronic Markets Research Center (CEEM) at the CurtinUniversity of Technology His research orientation is in the empiricalmodelling of aspects of electronic, information, and communicationsmarkets in the search for dynamic regularities Data examined includecircuit- and packet-switched traffic, product adoption, and telecom-munications company share price data He is currently working ondeveloping methods to enable reliable long-horizon forecasts based onshort time series

H E R V E ´ T A N G U Y is Director of Research at INRA/LORIA and at theLaboratory of Econometry of Ecole Polytechnique He performs re-search in industrial economy applied to problems of strategy, finance,and business organization He has participated in the conception andrealization of new business software systems in several organizations(multinational firms, public companies, wine industry, etc.)

S Y LV I E T H O R O Nis a maıˆtre de confe´rences at the University of Toulonand a researcher at the GREQAM, a CNRS research group in econo-mics in Marseille Her research fields are public economics and theformation of coalitions and agreements in economics In the framework

of cooperative and non-cooperative game theory, she studies theincentives and behavior of individuals and groups of individuals whendeciding whether to contribute to public goods.

M AT T H I J S W I L D E N B E E S Tis a PhD student at the Tinbergen Instituteand the Erasmus School of Economics His major field of interest isindustrial organization His current research interests include con-sumer search and the structural estimation of equilibrium models

J E A N-B E N O Iˆ T Z I M M E R M A N Nis CNRS Research Director at GREQAM(Groupement de Recherche en Economie Quantitative d’Aix-Marseille)and IDEP (Institut d’Economie Publique) in Marseille His maintopics of interest are economics of interactions and networks, proxim-ity dynamics, innovation, and intellectual property economics (opensource software)

This work draws from a process of interactions among the contributors,which led to a consistent set of chapters that have been written inter-actively This book is therefore really a collective one, and the editorswould like to warmly thank the authors who accepted the disciplineneeded to produce such an outcome These interactions were initiated

in a seminar held in 1999–2000 which led to the edition of a special issue

of the Revue Economique entitled “L’Economie de l’Internet” in 2001.This publication set the basis of the contributions to this entirely newbook that was elaborated in the frame of a working group of the CNRSResearch Consortium “Information Technologies and the Society” (GDRTICS) in 2003–2004 This working group organized in particular twoworkshops held in 2004 at GREQAM (a joint research center from theFrench CNRS, the Universities of Marseille and EHESS) and at theLaboratory of Econometrics of the CNAM (Paris)

The editors are grateful to the organizers of these events, and inparticular to Jean-Benoıˆt Zimmermann (GREQAM) and to Franc¸oisMoreau (CNAM), as well as to the CNRS for its financial support.They would also like to thank the President of the Revue Economique,Jean-Paul Polin (University of Orle´ans), for his support to the project.The editors are also indebted to Marie-Line Priot (EconomiX, CNRS-University of Paris X) for secretarial support

While this book was being published, Delphine Sabourin passed away.Delphine was the youngest contributor to this book, but she was far frombeing the least brilliant scientist on the team All of us will miss herbrightness and kindness, and we are all so sad that she has been deprived

of such a promising future

Since this book aims at providing the reader with an up-to-datesynthesis of what is essential in digital and Internet economics, some of

already published in journals The editors and the publishers would like

xxv

(Chapter8), and the Institute for Operation Research and Management

originally published respectively, in the Journal of Economic Literature(vol 42, pp 457–486, 2004), in Marketing Science (vol 19, pp 63–82,2000), and in the International Journal of Industrial Organization (vol 14,

no 6, pp 673–699, 1996)

1 Internet economics, digital economics

Eric Brousseau and Nicolas Curien

Since the “privatization” of the Internet in the United States in the 1990s, the network of networks has developed rapidly This has beenmatched by a wave of innovation in information technology, as well as inmany areas of its application, giving rise to a multitude of on-lineservices and new “business models” In the same period, the UnitedStates experienced unprecedented non-inflationist growth As a result ofthis conjunction, certain commentators considered the Internet as theheart of a new growth regime, qualified as the “new economy” Thiscontributed to the creation and then amplification of a speculativebubble around businesses involved in the Internet

mid-As these unfounded hopes necessarily met with disappointment, theeuphoria disappeared at the turn of the 21st century At the same time,the forecasts of a certain number of economists were confirmed a poster-iori These forecasts had highlighted, firstly, that the use of informationand communication technologies (ICTs) does not lead ipso facto to animprovement in microeconomic performances (Brousseau and Rallet

and thirdly that American growth in the 1990s was not necessarilyfounded on the innovations linked to the use of ICTs exclusively

these analyses do not claim that nothing changes with the large-scaledissemination of these digital networks and their associated practices By

that the changes are slower and more complex than is generally admitted,

We would like to thank Michel Gensollen, Thierry Pe´nard and Alain Rallet for their helpful remarks on this text However, should any inaccuracies or errors subsist, they are

to be entirely attributed to the signatories.

1

precisely because they have a fundamental character David (1990,

2000b) refers to digital technology as general purpose technology whoseimpact on economic performance is linked to transformations in prac-tices in all dimensions of economic and social life: norms of consump-tion, modes of production, organizational forms, and so on In the sameway as the great inventions at the end of the 19th century, this tech-nology will radically change economics, spark growth and transform theface of society, but only in the long run

In this long-term perspective, the Internet is situated at the confluence

of two older economic evolutions: on the one hand, that of nications networks, created in the 19th century and becoming electronic

telecommu-in the second half of the 20th century; on the other, that of computers,beginning during the Second World War The transformation of soci-eties, under the influence of these technologies, had been set in motionwell before the sudden emergence of the network of networks thattherefore does not mark the beginning of ICT-related transformations.Conversely, it is premature to consider it as the culmination of thistechnology Given that the majority of the trajectories of change havebarely begun, it is difficult to pronounce judgment on the final result

of movements which are still emerging and unstable However, theInternet, and more generally speaking digital networks, possess specificproperties which leave their mark on a number of phenomena for whichthey become the basis: information processing and circulation, commer-cial transactions, organizational coordination, network management,and so on

The ambition of this book and of this chapter is to highlight theaspects of the Internet and digital technologies that appear to be trulyinnovative, in terms of both economic practices and analytical concepts.Since the 1980s, telecommunications networks have constituted amelting pot producing, on the one hand, new practices in the manage-ment of “public facilities”, the regulation of competition, the design ofnetwork services, etc., and, on the other, new analytical concepts such asthe notions of “contestable” market, “yardstick” competition, “incen-tive” pricing, etc., which were then applied to all industries Similarly,the Internet is today giving rise to innovative practices that call forrenewed conceptualization

There are three principal reasons explaining the Internet’s double role

as a catalyst of practices and theories First of all, the Internet is aplanetary federation of digital networks, whose technical potential,notably the ability to act as a medium for very differentiated modalities

of information management, induces a growing “digitization” of ities: access to these interconnected and flexible networks incites

activ-economic actors to increase the informational intensity of their servicesand to multiply their informational exchanges Secondly, the logic of thismodular and decentralized network, serving as a platform for the provi-sion of services founded on information and innovation, deployed in aglobal space, makes it an archetype of contemporary economies, whereindustry tends to be organized according to a flexible assembly modelthanks to standardized interfaces; where competitiveness is stronglyassociated with the ability to innovate; where products and services areundergoing an increase in informational intensity; where the economicspace is more and more transnational, etc Finally, the organizationalinnovations induced by the digital networks federated by the Internet aregradually spreading to the entire economy.

We shall highlight the fact that one of the Internet’s central tics is the ability it grants economic agents to very finely control the infor-mation they exchange in accordance with the individual preferences of theissuing and receiving parties Moreover, this control can be totally decen-tralized through the use of standardized interfaces This double charac-teristic founds both the specificity of the Internet as a network and that ofthe digital economy it serves Other frequently mentioned characteristicssuch as the global and multimedia nature of the Internet or its impact oninformation costs are certainly important factors, but they probably do notalone justify giving such marked attention to Internet economics.This chapter is divided into three parts In the first part, we will recallsome of the factual principles and elements clarifying the nature of theInternet network as well as the issues it raises for economic activities Inthe second part, we will seek to specify the link between the “internal”morphology and economics of the Internet network and the new types ofrelations and exchanges which accompany the development of this net-work, which we will refer to using the term “digital economics” We willshow that this link has more to do with engendering than with causality

characteris-To put it in other terms, rather than simply considering the Internet as

a technological tool, a determining factor in a new type of economicdevelopment, it is more productive to analyze Internet economics stricto

ability to manage information in a decentralized and customized way ismassively exploited In the final part, we will seek to draw up a researchprogram covering the issues raised by the multifaceted development

of the Internet and the concomitant emergence of digital economics

We will explain how the different chapters of this book, as a contribution

to the flourishing corpus of publications devoted to Internet economics,constitute the pieces of a puzzle which is yet to be completed, and whosefinal form is just beginning to emerge

1.2 Internet economics: some principles and facts

The Internet is not a network per se but a network of networks that relies

on common standards allowing machines that process information ally to “interoperate” More precisely, Internet standards enable thetotally decentralized interconnection of computerized networks Onthe Internet there is no technical discrimination between the resourcesdedicated to the administration of the network and the terminals thatprocess the information carried, as is usual in a traditional telecommuni-cation network such as the telephone network, where terminals processinformation while the administrative equipment of the network –switches – connects information flow and transmission (Curien and

(IPDs) connected to the network are simultaneously terminals androuters In addition, the use of a standard interface and a genericaddressing system creates a sort of “meta-network” presenting itself as

a homogenous and seamless system to the user

The devices interconnected by the Internet, essentially computers,process information digitally The network organizes communicationbetween these machines on the basis of the “client–server” model The

“client” sends requests to the “server”, which processes them and thensends a response Any device connected to the Internet can be both aclient and a server This is notably the case for the most commonapplications on the Internet, e-mail and the Web Sending an e-mailinvolves asking the server (recipient) whether he agrees to receive infor-mation If he does, the client (sender) sends the information In practice,e-mail servers carry out these operations Unlike user terminals, theservers are permanently connected to facilitate data flows Similarly,when consulting a website, the visitor sends a request to a computer inwhich information is stocked The information server sends backHyperText Markup Language (HTML) codes to the client that enablethe computer to re-build pages on the client’s screen Generally speak-ing, independently of the application being considered, requests andresponses are broken down into data packets, which their senders andrecipients identify and which circulate within the network where they arerelayed by routers After the packets are transmitted and received, thereceiving terminal reconstitutes the original programming lines contain-ing an informational content or instructions to the machine that pilots itfrom a distance The following elements must be in place for everything

to function correctly:

 All IPDs conn ected to the netwo rk must be cl early identified so thatthe data packe ts actu ally reac h the ad dressee ; this identifier is theInterne t protocol (IP) numbe r.

 An address ing language has to allow users to form ulate their requ estwith a suitable server: domain names (www.identi fier.com) m ake upthe visible part of this addressing system; servers known as domainname systems (DNS) transform these addresses close to “human”language into machine addresses (IP addresses)

 Common communication protocols are required for communicationand routing between IPDs; the Internet protocol is at the center of avast group of technical standards responsible for communication andinteroperability among network components

 The machines must use compatible programming languages in order

to code and decode requests and responses transmitted betweenclients and servers; in this respect, HTML language is the foundationstone of the Web, enabling different kinds of IPDs to exchange texts,images, data and sounds

Thus the very existence of the Internet stems from the use of a genericaddressing system (IP numbers and domain names) and standards(Internet protocol and HTML) These form the basis of interoperabilitybetween various sub-networks The technical regulation of the Internettherefore is essentially based on the management of these resources withthe goal of guaranteeing interoperability Three main organizations carryout this regulation:

 ICANN (Internet Corporation for Assigned Names and Numbers),set up in 1998, is a non-profit organization based in the US Under

a delegation contract with the US government (Department ofCommerce), ICANN is responsible for organizing the distribution of

IP numbers and domain names In both cases, the addressing system

is a hierarchical one in which a limited number of roots (e.g .com, org or.net) enables the creation of addresses This hierarchy makes it possible todelegate the practical distribution of addresses among entities thatmanage portfolios of addresses according to their own rules ICANNtherefore supervises two distinct functions: the distribution of IPnumbers that is ensured by the administrators of the subscriber networks(the Internet service providers – ISPs) and the distribution of domainnames, by setting the features of the available roots (first-order domains,

or suffixes, such as com, fr, etc.) and by selecting and supervising theorganizations in charge of collecting and registering users’ claims.ICANN’s importance derives from its power over the private companyVerisign (formerly Network Solution Inc – NSI), the entity responsible

for the technical management of the domain names’ root computer.This server contains the source file able to translate the domain namesinto IP addresses ICANN can therefore “erase” addresses or entitiesthat do not comply with the rules it sets out and thus exclude themfrom the Web This is why ICANN is one of the possible sources of anon-technical governance system of the Internet.

 The IETF (Internet Engineering Task Force) is de facto the entityresponsible for the standardization of the communication protocols:the Internet protocol system It has no legal status and is only aworking group of the Internet Society (ISOC) The latter is a not-for-profit organization, founded by some of the “inventors” of theInternet, that constitutes a forum of reflection and a tool of influenceaimed, notably, at promoting the development of an open and efficientnetwork, which would benefit the greatest number

 The W3C (World Wide Web Consortium) is responsible for thestandardization of the multimedia languages used on the Internet It

is a kind of club where access is reserved to those organizations thatcan afford the relatively high membership fee

These three organizations are, however, not really regulating bodies

 They do not combine the three faculties of setting rules, supervisingoperators and users, and imposing sanctions for lack of compliance orfor practices going against the principles which they stand for, such asfair competition, public freedom and secure operations Indeed, theIETF and the W3C constitute mechanisms for sharing technical devel-opments in a manner similar to that with open source software Theirstandards are not mandatory for Net users and operators; however, they

do facilitate interoperability ICANN certainly has formal powers but itdoes not have sufficient means to supervise the Web, which would makethe exercise of its prerogatives efficient and independent

 Furthermore, the legal status of each of these organizations is unclear,

or even non-existent in the case of the IETF In principle, they areresponsible for the traditional regulation of the World Wide Web.However, they are subject to American law and operate under con-tract with the United States government Their functioning andmembership principles do not guarantee their independence, whichweakens their authority and casts doubt on the legitimacy of theirdecisions The credibility of the norms they set and the decisions thatthey make is affected as a result because their enforceability is notguaranteed

 Finally, the scope of these organizations’ jurisdiction is unclear Inprinciple, they are responsible only for the technical regulation of the

Internet, the IETF and the WC3 in particular However, for technicaland historical reasons, the regulation of the network and its modalities

of use are closely linked Indeed, the regulatory bodies decide for oragainst authorizing the development of particular categories ofservices depending on the way in which the interoperability standards,the security mechanisms and the mechanisms for managing priorities

strong tradition in the Internet community to set ethical, social andeconomic norms in addition to technical rules For example, until

1995, Netiquette, Net ethics, prohibited any commercial use of the Net

The regulatory bodies of the Net appear very different from the itional mechanisms for regulation and international standardization,such as, notably, the International Telecommunication Union (ITU),the International Standard Organization (ISO) or the InternationalElectrotechnical Commission (IEC) When the Internet was “invented”,

trad-it was decided not to go through these different authortrad-ities for fourprincipal reasons: firstly, they are known for being slow in elaboratingstandards, which seemed incompatible with the high rate of innovation

in technologies linked to the Internet; secondly, these bodies did notimmediately grasp the originality and the power of Internet standardsand, moreover, certain aspects of the Internet put it into competitionwith the areas these bodies govern; thirdly, the Internet was limited tothe North American continent until 1998 and even today the network isstill predominantly American, despite a clear trend of internationaliza-tion; finally, the technicians and the entrepreneurs of the Internet,marked by a libertarian or liberal ideology, from the very beginningexpressed an almost visceral suspicion with regard to State or inter-national bureaucracies

Despite their limits, the combined efforts of ICANN, IETF and W3Cplay an essential role in the technical governance of the Internet However,this governance conditions in part the socio-economic regulation ofactivities for which the network is the basis In fact, on a network ofthe Internet type, each participant has the possibility of influencing theway in which the flow of communication is administered This possibil-ity exists in relation to digital technologies’ ability to encrypt informa-tion: in a digital system, all information is coded in the form of a digitalsequence, which is then easy to encrypt The encryption thus constitutesthe key to a filtering of information use: according to the settings, onecan authorize access to all or to a part of the information potentially

available, depending on the identity of the user or other criteria Thecombination of encryption abilities and the decentralized administration

of communication therefore enables all Internet users to regulate the use

of information which they make available on the network, because theycan lay down “norms” of use for this information which the technologywill enforce This opens practically infinite margins for maneuver toeconomic agents who wish to define informational services, as well as

to social actors desiring to implement specific rules of informationalinteraction However, interoperability imposes one notable constraint:

in order for the actors to be able to implement such specific rules, thesemust be compatible with the interoperability standards defined by theIETF or the W3C This gives these bodies a notable influence on theuses that are likely or not to develop on the Internet Moreover, the way

in which the addressing system is managed also influences uses Themodalities for the management of domain names – that is to say therecognition, or non-recognition, of brand rights, the question of creatingcategories allowing service providers to be classified and therefore

“labeled”, as well as the definition of rules of inclusion for these categories –have an influence on the conditions of competition between the operators.Therefore, in the long term, they influence the nature of the services andthe uses made of them

Given these factors, and because the technical mechanisms for thegovernance and regulation of the Internet are neither totally legitimatenor perfectly complete, States have progressively become more involvedwith the socio-economic regulation of the network More exactly, as long

as the Internet affected only a coherent and closed community – that ofscientists – the American State had little interest (and other States evenless so) in intervening in a network which functioned according to rulesthat were specific to this community, and which in any case was ultim-ately monitored by the State Yet with the diversification of uses andactors, the need to complement the technical regulation in order toorganize competition, to enable the development of commercial activi-ties, to protect citizens, and so on, made itself felt strongly This resulted

in intense legislative activity in the American Congress from 1995 onwards

for Economic Cooperation and Development (OECD) and those of theEuropean Union, were soon to follow, from 1997/1998 The WorldSummit on the Information Society is the symbol of the globalization

of this process, and of the attempts to harmonize the various nationaldevelopments

The first reflex in most countries was to extend the field of applicationfor the existing regulations, as well as the jurisdiction of the authorities

responsible for enforcing them However, this movement ran into twomajor obstacles:

 The re-evaluation of borders between industries inherited from thepre-digital era needs time and cannot take place without conflict andinconsistencies The Internet is a platform integrating all communi-cation and information-processing technologies, which inevitablytends to blur the borders traditionally established between the respect-ive areas of voice, image and data; that is to say, in industrial terms,between the computer, telecommunications, audiovisual and publish-ing industries For this reason, Internet activities have been subject tomultiple regulations which are sometimes contradictory and sometimessimply costly and complex to combine This leads to a “patchwork” oflegislative and regulatory environments which is consequently verysensitive to the interpretation of administrative or judiciary authorities,which more often than not appear to be unprepared for the Internet’soriginality and technical complexity

 The Internet’s global and open character is not very favorable forestablishing national regulations The Yahoo! case in France in 2000showed that it is a complicated matter to enforce a judge’s rulingobliging a portal to deny its clients access to certain content.Furthermore, this case, arising from complaints lodged by anti-racistassociations, raises questions concerning legal disputes In the name ofthe “Gayssot” law condemning the justification of the Holocaust andrevisionism, the French judge obliged Yahoo!, the provider of a portaland search engine, to deny French surfers access through its site to anAmerican auction site selling Nazi objects, failing which Yahoo! would

be subject to periodic fines However, a decision of this kind imposestechnical problems: how is it possible to effectively recognize thenationality of the surfers and, above all, to deny access, since it issufficient merely to know the URL to access the “proscribed” content?There is also a legal problem: what standard should be establishedbetween the French conception of the regulation of certain contentand American ethical standards defending total freedom of expres-sion? The world legal system is organized on a territorial basis.However, the Internet is aterritorial, to the extent that its architec-ture ignores the geographical localization of information-processingoperations: the data carried by the network uses pathways thatcannot be regulated and the information bases consulted or usedcan be fragmented or duplicated in several places, in such a way that

it is totally invisible to the user, or even for the administrators of thenetwork Given that operations carried out on the network lack a

geographical base, to a great extent legal norms turn out to beinefficient.

The pre-existence of regulation mechanisms for the Internet and theirincompleteness, the limits of traditional State approaches, their lack oflegitimacy in the eyes of the liberal and libertarian ideology that presidedover the development of the Net, are all factors which explain why a so-called “co-regulation” approach imposed itself progressively, notably atthe instigation of the US, the OECD and the European Union

In fact, from the beginning of the 1990s, the United States tried toimpose a model of self-regulation, which the creators of the Internet andindustry had been lobbying for However, it rapidly turned out to beimpossible to maintain this logic, given the Internet’s effects on intellec-tual property, national security, public freedoms, and so on Moreover, itwas also necessary to adapt the existing legislative framework in crypt-ography, evidentiary law and so on in order to allow for the development

of economic activities on the Internet, notably electronic commerce.Thus, the idea of implementing cooperation between the State andnon-governmental organizations (NGOs) for the regulation of theInternet progressively emerged There were two aspects to this cooper-ation: on the one hand, to informally delimitate areas of responsibilitybetween the State and involved NGOs, notably by applying the principle

of subsidiarity, and on the other, to allow the parties involved in theInternet to fully participate in the elaboration of State standards, gener-ally using the means of the network The Europeans adopted an ap-proach of this type straight away However, they sought to use existingdemocratic institutions more than in the American approach

The self-organization of the Internet is therefore not synonymous with atotal lack of an institutional framework: it is “framed” self-organization.Moreover, even if, technically speaking, it would be theoretically pos-sible, the network does not function in a perfectly homogenous andundifferentiated mode in which all the actors play the same symmetricroles There are “network operators” which serve the purpose of supplying

a range of telecommunication services – essentially access to the Weband management of communication services – making them an interfacebetween users and the transport providers, that is to say the cableoperators, the telephone companies, the owners of transmission infra-structure, etc These Internet operators implement the standards andaddressing systems described above and they perform the interconnec-tion between networks For reasons of efficiency, the different access

networks are not systematically directly connected to each other The

the bottom, the ISPs connect the final users’ machines (or local works) to the Internet using various local distribution networks: tele-phone and cable television networks for private customers, specializednetworks of all kinds for professionals On top, backbone administratorsinterconnect the ISP through high-speed continental and intercontin-ental networks Although ISPs and backbones are competitors, they arenonetheless interconnected in order to ensure the general connectivity ofthe network

net-The conditions for interconnection between operators of Internetservices are complex, due to the convergence of two dimensions:

 Firstly, they are based on vertical agreements with elements of tition, to the extent that the operators are both complementary andrivals On the one hand, each provides a generalized connectivityservice, since the users wish to have access to correspondents or tocontent hosted on the other networks On the other hand, each seeks

compe-to attract the greatest possible number of users on its own networkbecause the production function is characterized by high fixed cost

 Secondly, strong externalities mark relations between operators, due

to the decentralization of the network administration Once two works are interconnected, their respective operators no longer directlycontrol the use made of them The logic of the Internet protocol issuch that the packets of information themselves choose the mostefficient path to reach their target by selecting the unsaturated parts

net-of the network In these conditions, the available capacity net-of a givenoperator is “automatically” used, which may degrade the quality of theservices offered to its own users

These elements combine to cause original problems where industrialorganization is concerned In particular, relations between operatorsmust overcome a cooperation–competition dilemma that takes on aneven more brutal form given the wide scope for the opportunistic ma-nipulation of interconnection conditions This results in a great variety

of modes of governance, going from the institution of modalities for freeinterconnection through peering agreements intended to minimize trans-action costs, to vertical integration internalizing the externalities, as well

as various pricing formulae and organizational arrangements These areall strategies that question competition policies, as they are situated in acontext where the logic of fixed costs encourages concentration.Finally, it should be noted that these characteristics are not unique torelationships between Internet operators: they can also be found in many

relationships between providers of informational services as well as inrelations between informational providers and providers of networkservices.

Originating in the American military and scientific community, thenetworks which became the Internet were progressively transferred toprivate initiatives It was these private initiatives that were responsible forthe universal success of the open standards and the addressing systemthat characterize the Internet However, the process leading to theemergence of the Internet was far from spontaneous

The Internet and its precursors were developed and administeredthanks to research fellowships and licensing contracts from theAmerican government Until 1985, it was essentially made up of diverse,closed and, to a great extent, experimental networks, elaborated within theframework of large American research programs: ARPANet in Defense,SPAN in space technology, CSNet in computers, etc ARPANet, theprecursor launched in 1969, constituted the principal laboratory for thefuture Internet This was the program in which the concepts and tech-niques structuring the network today were invented (Leiner et al [200 1])

In 1985, when the National Science Foundation (NSF) decided tofavor the development of a network open to the entire scientific commu-nity, the ARPANet network logically became the principal basis forNSFNet The latter was itself rapidly transformed into the Internetwhen, in 1988, the NSF adopted an interconnection policy with privatenetworks (however this was not to lead to an opening up to commercialapplications until much later)

The NSF’s will to open up was not limited to a simple interconnectionpolicy From 1985 the principal technical component of the Internetunderwent an active transfer to industry For successive administrations,opening up to private investment and to cooperation with industryeffectively constituted the best way of developing the network and itsassociated technologies This served a double purpose: on the one hand,

to give the United States a more efficient information infrastructure, and

on the other, to reinforce the advance of the American computer try in digital technologies This policy was finalized in 1998 whenresponsibility for the development of the Internet was transferred fromthe NSF to the Department of Commerce

indus-The US government therefore played a determining role in the gence and promotion of the Internet and the policy of industry transfer,initiated in 1985, gave rise to a specialized industrial base The