1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44111
Co...
30 downloads
349 Views
2MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44111
Competitiveness of New Industries
Information and communication technology (ICT) is considered to be one of the ‘drivers of growth’ by the OECD, while others regard it as a new type of general-purpose technology. Countries not able to establish themselves successfully in the ICT-related industries run the risk of losing important growth opportunities. Competitiveness of New Industries examines the extent to which the different institutional frameworks in Japan, the United States and Germany have encouraged and will continue to support the potential of ICTs. This is essentially a question of how rules should be designed in order to reach desired results – a traditional problem in economic policy. It has recently gained new impetus with regard to ICT industries due to the existence of network externalities and the public good character of some ICT subsectors. The volume also discusses the degree to which economic actors are capable of designing ICTs in the institutional framework within which they operate – a discussion that is strongly connected to the long-standing issue of path dependency. Cornelia Storz is Professor for Japanese Economy in the Department of Economics and Business Administration and Centre for Interdisciplinary Centre for East Asian Studies at the Goethe University of Frankfurt, Germany. Andreas Moerke is Head of the Business and Economics Section at the DIJ, the German Institute for Japanese Studies in Tokyo, and Partner at the consultancy iJEB Ltd (Tokyo, Berlin and Hamburg).
Routledge Studies in Global Competition Edited by John Cantwell, University of Reading, UK, and David Mowery, University of California, Berkeley, USA
Volume 1 Japanese Firms in Europe Edited by Frédérique Sachwald Volume 2 Technological Innovation, Multinational Corporations and New International Competitiveness The case of intermediate countries Edited by José Molero Volume 3 Global Competition and the Labour Market Nigel Driffield Volume 4 The Source of Capital Goods Innovation The role of user firms in Japan and Korea Kong-Rae Lee Volume 5 Climates of Global Competition Maria Bengtsson Volume 6 Multinational Enterprises and Technological Spillovers Tommaso Perez Volume 7 Governance of International Strategic Alliances Technology and transaction costs Joanne E. Oxley Volume 8 Strategy in Emerging Markets Telecommunications Establishments in Europe Anders Pehrsson
Volume 9 Going Multinational The Korean experience of direct investment Edited by Frédérique Sachwald Volume 10 Multinational Firms and Impacts on Employment, Trade and Technology New perspectives for a new century Edited by Robert E. Lipsey and Jean-Louis Mucchielli Volume 11 Multinational Firms The global–local dilemma Edited by John H. Dunning and Jean-Louis Mucchielli Volume 12 MIT and the Rise of Entrepreneurial Science Henry Etzkowitz Volume 13 Technological Resources and the Logic of Corporate Diversification Brian Silverman Volume 14 The Economics of Innovation, New Technologies and Structural Change Cristiano Antonelli Volume 15 European Union Direct Investment in China Characteristics, challenges and perspectives Daniel Van Den Bulcke, Haiyan Zhang and Maria do Céu Esteves
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Volume 16 Biotechnology in Comparative Perspective Edited by Gerhard Fuchs
Volume 24 Location and Competition Edited by Steven Brakman and Harry Garretsen
Volume 17 Technological Change and Economic Performance Albert L. Link and Donald S. Siegel
Volume 25 Entrepreneurship and Dynamics in the Knowledge Economy Edited by Charlie Karlsson, Börje Johansson and Roger R. Stough
Volume 18 Multinational Corporations and European Regional Systems of Innovation John Cantwell and Simona Iammarino
Volume 26 Evolution and Design of Institutions Edited by Christian Schubert and Georg von Wangenheim
Volume 19 Knowledge and Innovation in Regional Industry An entrepreneurial coalition Roel Rutten
Volume 27 The Changing Economic Geography of Globalization Reinventing space Edited by Giovanna Vertova
Volume 20 Local Industrial Clusters Existence, emergence and evolution Thomas Brenner
Volume 28 Economics of the Firm Analysis, evolution, history Edited by Michael Dietrich
Volume 21 The Emerging Industrial Structure of the Wider Europe Edited by Francis McGowen, Slavo Radosevic and Nick Von Tunzelmann
Volume 29 Innovation, Technology and Hypercompetition Hans Gottinger
Volume 22 Entrepreneurship A new perspective Thomas Grebel Volume 23 Evaluating Public Research Institutions The US advanced technology program’s intramural research initiative Albert N. Link and John T. Scott
Volume 30 Mergers and Acquisitions in Asia A global perspective Roger Y. W. Tang and Ali M. Metwalli Volume 31 Competitiveness of New Industries Institutional framework and learning in information technology in Japan, the US and Germany Edited by Cornelia Storz and Andreas Moerke
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Competitiveness of New Industries Institutional framework and learning in information technology in Japan, the US and Germany Edited by Cornelia Storz and Andreas Moerke
First published 2007 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group, an informa business
This edition published in the Taylor & Francis e-Library, 2007. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2007 for selection and editorial matter, Cornelia Storz and Andreas Moerke; individual chapters, the contributors. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. 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 Competitiveness of new industries: institutional framework and learning in information technology in Japan, the U.S., and Germany/edited by Cornelia Storz and Andreas Moerke. p. cm. Includes bibliographical references and index. 1. High technology industries. 2. New business enterprises. 3. Information technology – Economic aspects. 4. Telecommunication – Economic aspects. 5. Electronic commerce. 6. Competition, International. I. Storz, Cornelia. II. Moerke, Andreas. HC79.H53C6565 2006 338.6′048–dc22 2006031342
ISBN 0-203-96360-1 Master e-book ISBN
ISBN10: 0–415–41624–8 (hbk) ISBN10: 0–203–96360–1 (ebk) ISBN13: 978–0–415–41624–5 (hbk) ISBN13: 975–0–203–96360–9 (ebk)
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Contents
List of figures List of tables List of contributors Preface Abbreviations
ix x xii xviii xix
PART I
Introduction 1
Institutions and learning in new industries: an introduction
1 3
CORNELIA STORZ AND ANDREAS MOERKE
PART II
Institutional framework for ICT and options for political governance: Japan, the United States and Germany in comparison
17
Subsection A: Institutional conditions for introducing ICT
18
2
Legacies of the developmental state for Japan’s information and communications industries
19
MARK TILTON AND HYEONJUNG CHOI
3
Institutional framework and competitiveness of the US telecommunications market
41
MICHAEL SCHEFCZYK
4
Information and communication technologies in Germany: is there a remaining role for sector-specific regulations?
57
GÜNTER KNIEPS
Subsection B: The increasing role of self-regulation 5
Private solutions to uncertainty in Japanese electronic commerce CORNELIA STORZ
74 75
viii 6
Contents Institutional conditions for achieving effective implementation of ICT
103
ROB FRIEDEN
7
B2C e-commerce dynamics in Germany: do we need a new regulatory framework?
124
BERNHARD LAGEMAN, MICHAEL ROTHGANG AND MARKUS SCHEUER
PART III
Industrial organization, enterprise structure and ICT: Japan, the United States and Germany in comparison
153
Subsection A: Effects of ICT on industrial organization and on firm structures
154
8
ICT and corporate structure: the diffusion of e-commerce across Japanese companies
155
DENNIS S. TACHIKI
9
The rise and fall of ‘Wintelism’: manufacturing strategies and transnational production networks of US information electronics firms in the Pacific Rim
180
BOY LÜTHJE
10 Open innovation: novel deployment of ICT in new product development
210
RALF REICHWALD, FRANK PILLER, SASCHA SEIFERT AND CHRISTOPH IHL
Subsection B: The social construction of institutions and technology
232
11 Next generation information and communication technologies deployment in Japan
233
MOHAMMED AKHTAR, YOSHIYA TERAMOTO AND CAROLINE BENTON
12 Competitive advantage through co-evolution of technology and organization
256
JANET FULK
13 Shaping organizational technology: ICT as a learning process
270
GEORG SCHREYÖGG, SAMI KHIARI AND LEO SCHMIDT
Index
292
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Figures
2.1 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 4.1 7.1 8.1 10.1 11.1 11.2 11.3 11.4 11.5 13.1 13.2 13.3 13.4 13.5 13.6
International comparison of superiority in information and communications technologies Basic indicators, USA, Japan and Germany, 2003 Mobile indicators, USA, Japan and Germany, 1995 and 2003 Internet indicators, USA, Japan and Germany, 2003 Five phases of US telecommunications regulation Fixed-access competition, US Market share of local exchange carriers, US Market share of long-distance carriers, US High-speed Internet, US Internet periphery versus Internet service provision EU citizens who purchased on the Internet, 2003 Clustering of Internet companies in Tokyo Screenshot of the Adidas virtual customer lab Comparison of KDDI’s 1x, NTT DoCoMo’s FOMA and Vodafone’s W-CDMA subscriptions Cumulative growth of FOMA subscribers 3G licence fees in major countries in 2002 Monthly ARPU trend and total revenue data in major countries Asia’s iron triangle with complementary attributes Temporal modification patterns All adaptive activity differentiated between standard customizing and workbench changes with ABC All adaptive activity differentiated between standard customizing and workbench changes with GERO Level of adaptive activity at ABC on a monthly basis, 1996–98 Level of adaptive activity at GERO on a monthly basis, 1999–2001 Self-organized technology creation as a process
37 42 43 44 46 50 51 52 53 59 129 160 224 243 245 248 249 252 274 280 281 282 283 285
Tables
2.1 2.2 2.3 2.4 2.5 2.6 2.7 4.1 4.2 7.1 7.2 7.3 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10
OECD composite basket of residential and business telephone charges, August 2002, 2004 OECD Internet access basket at daytime discounted PSTN rates Japan’s software exports and imports, 1995–2000 Percentage of adults who went online at least once in last 30 days (1999–2003) Fixed Internet subscribers/100 population E-readiness rankings, 2005 E-readiness rankings, by category, 2005 The localization of monopolistic bottleneck facilities Local telecommunications networks as monopolistic bottleneck facilities Development of e-commerce in Western Europe and in selected European countries Products most frequently bought via the Internet in Germany, 2003 Selected rules and regulations relevant to e-commerce and central institutions Market size of e-commerce, 1998–2005 Users of e-commerce Organizational characteristics of Bit Valley Internet companies Major newly established dotcom companies (partial list) Leading users of e-commerce, 2000 Online services Online sales How establishments use the Internet to sell products and services Online procurement Impacts of conducting business online
24 25 28 32 33 33 34 62 64 128 130 142 156 158 162 164 165 168 170 172 175 177
Tables xi 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
9.1 9.2 9.3 9.4 9.5 11.1 11.2 12.1
Top ten electronics contract manufacturers, 2003 Flextronics International: one contract manufacturing network in Asia, 2003 Global supply chain consolidation, HP/Compaq Typology of outsourced production in the electronics industry EMS/ODM margins: industry average, 1999 to Q1 2004 Market landscape of Japan, August 2003 Mobile shipment in Japan Types of coupling in co-evolving systems
185 190 193 200 200 240 253 263
Contributors
Mohammad Akhtar is Vice President at Motorola and is responsible for telecommunication, infrastructure technology and marketing for Asia Pacific. He is an expert in wireless PCS/Cellular technology and has extensive experience in global wireless markets. He works closely with major global carriers to help them understand future communications requirements. He has a BSc degree in Electrical Engineering from the University of Texas and an MBA from the Kellogg School of Business, Northwestern University. Currently he is a PhD candidate at Waseda University. Caroline Benton is Professor at the University of Wales’s validated MBA programme in Japan. She earned her PhD degree in Management Engineering from the Tokyo Institute of Technology. She has held positions as a Director of a Japanese subsidiary of a European manufacturer and as Chief Consultant of a marketing consulting firm for foreign-affiliated firms in Japan. Hyeonjung Choi is currently Senior Secretary to a member of the Korean National Assembly. He received his PhD in Political Economy from Purdue University with a dissertation on Japanese IT policy and the software industry. His specialties are national ICT strategy, industrial policy and government/business relations. He worked at Korea Air Force Academy as a faculty member and was invited to the Institute of Social Science at Tokyo University. Before accepting his current position of the Legislative Public Officer at the Korean National Assembly, he recently was a Senior Fellow at the IT thinknet (Korean IT consulting institute). Rob Frieden serves as Pioneers Chair and Professor of Telecommunications at Penn State University where he teaches courses in management, law and economics. He also provides legal, management and market forecasting consultancy services. Before accepting an academic appointment, Professor Frieden served as Deputy Director of International Relations for Motorola Satellite Communications, Inc. He has held senior policymaking positions in international telecommunications at the Federal
Contributors 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
xiii
Communications Commission and the National Telecommunications and Information Administration. He holds a BA degree with distinction from the University of Pennsylvania and a JD from the University of Virginia. Janet Fulk is Professor of Communications in the Annenberg School for Communication and Professor of Management and Organization at the Marshall School of Business at the University of Southern California. She holds an MBA and PhD in Administrative Sciences from the Ohio State University. Her research interests include communication and knowledge management, information technology for strategic alliances, and social aspects of knowledge and distributed intelligence. A series of recent projects sponsored by three grants from the National Science Foundation examines how communication and information systems are employed to foster collaboration and knowledge distribution within and between organizations. Christoph Ihl is a Senior Researcher at the Institute for Information, Organization and Management (IOM), TUM Business School, Technische Universität, München. His research focus is the empirical analysis of consumer choice behaviour for technology-intensive products. Sami Khiari works as a Management Consultant at one of the world’s biggest auditing and consulting firms. As a manager for the Financial Services Industry, his work mainly focuses on the improvement of the organizational performance in the Finance & Accounting department. In this context he specializes in organizational analysis and optimization with a strong emphasis on the interplay of technology and processes. As part of his practical and scientific work, he has also conducted case studies on informal and evolving structures on the edge of intra-organizational technologies. Sami Khiari holds a degree of Diplomkaufmann in Business Administration from the Freie Universität, Berlin, where he also received his doctorate degree and where he has co-authored publications with Professor Georg Schreyögg. Günter Knieps is Professor of Economics at the University of Freiburg and Director of the Institute for Transport Networks and Regional Policy, Albert-Ludwigs-Universität Freiburg, Germany. Before that he held a position as Professor of Microeconomics at Groningen, the Netherlands. He has held visiting research positions at Princeton and the University of Pennsylvania, and has contributed numerous publications on network economics, (de)regulation, competition policy, industrial economics and sector studies on network industries. He is a member of the Scientific Advisory Councils of the German Federal Ministry of Economics and Technology, and the German Federal Ministry of Transport, Construction, and Housing. Bernhard Lageman is Chief of the Division ‘Empirical Industrial Organization’ in the RWI-Essen. His research interests lie in the fields
xiv
Contributors
of industrial and institutional economics. Recent research works include the survival of start-ups, the structural change of banking systems, entrepreneurship and enterprise performance, the influence of regulation policies on construction and German craft sectors, firm R&D and innovation, and technology policy. Boy Lüthje is a Research Fellow at the Institute of Social Research (Institut für Sozialforschung) and a Senior Lecturer at the Department of Social Sciences, University of Frankfurt. His research areas are the political economy of production and innovation, the international division of labour, and industrial relations with a focus on the United States and Greater China. Dr Lüthje holds a PhD from the University of Frankfurt where he worked as an Assistant Professor from 1990 to 1996. He has been a Visting Scholar at the University of California, Berkeley in 1996–97; the East–West Center, Honolulu, Hawaii, on several occasions since 2002 and the People’s University of China, Beijing in 2006. He also taught as a Visiting Professor at the University of Kassel, Germany. Andreas Moerke is Partner at the German–Japanese consultance JEB interlogue. He is responsible for market entry strategies to Japan and is consulting with companies in the automobile, supplier, machine tool and software industries. Besides this, Moerke serves as auditor on the board of Landscape Inc., a data solution and marketing company in Tokyo. He has held positions as Head of Business and Economics Section at the German Institute for Japanese Studies (DIJ) and Fellow at the Science Center Berlin doing research on industrial organization, corporate governance, and international management. Moerke holds a PhD in Management and an MA in Japanese Studies. Frank Piller is an Associate Professor of Management (‘Privatdozent’) at TUM Business School, Technische Universität, München, and a Faculty Member of the Smart Customization Group at the Massachusetts Institute of Technology (MIT) in Cambridge, MA. His recent research focuses on value co-creation between businesses and customers/users, and the interface between innovation management, operations management, and marketing. Ralf Reichwald is a Professor of Management at TUM Business School, Technische Universität München, and Head of the Institute for Information, Organization and Management (IOM) also at TUM Business School. His main areas of research are leadership and organization, information and communication, and service management. Michael Rothgang works as a researcher in the RWI-Essen. His work focuses on the application of new technologies on the firm level, the macroeconomic and sector-specific analysis of innovation processes, as well as the effects of technology policy measures. He has published
Contributors 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
xv
research results both on the national and international level and is currently working on several projects in this field. Michael Schefczyk is Professor (SAP Chair of Entrepreneurship and Innovation) at the Technical University of Dresden. He submitted his doctorate in Business Administration at the Rhineland-Palatinate Technical University of Aachen based on an empirical dissertation on key success factors in declining industries. He received the Bifego entrepreneurship research prize in 1999 and the entrepreneurship award from the Heinz-Assmann Foundation. Previous employment includes work in the Communications, Media and Technology practice at Booz Allen Hamilton as Principal of the German Management Team. Professor Schefczyk has contributed nearly 30 German and English language publications on topics of business administration. Markus Scheuer works as a researcher at the RWI-Essen. He is a member of the Council of the international research association on services RESER. Other than services, his research interests focus mainly on the labour market. Leo Schmidt works as a Technology and Business Process Consultant with a multinational consulting firm. His work focuses on designing, implementing and changing technological infrastructures where SAP’s ERP package plays a leading role. He combines his consulting work with involvement in research and publications on technology in organizations on a regular basis. The main contributions from his research in this field support the idea of ongoing processes of interaction between organizations and evolving internal technological settings. He therefore encourages researchers and practitioners alike to accept and manage new roles of technology, allowing it to change shapes and tasks over time as a result of ongoing adaptive activity conducted by the organization. He holds a Degree of Diplomkaufmann in Business Administration from the Freie Universität, Berlin, where he also received his doctorate degree, and where he has co-authored publications with Professor Georg Schreyögg. Georg Schreyögg is Professor of Business Administration, Freie Universität, Berlin. Since 1994 he has held the Chair of Management in the Department of Business at the Freie Universität Berlin where he currently teaches courses in management at undergraduate, graduate and postgraduate levels. Before accepting his current position, Dr Schreyögg was on the faculty of the Fern Universität Hagen, the University of Bamberg and the European School of Management Studies (EAP: Paris–Oxford– Berlin). He received his doctorate degrees from the University of Erlangen-Nürnberg. He is the (co)author of around a hundred articles and seven books on management, job design and strategic management and has conducted research on knowledge management, strategic decision making, organizational change and corporate culture. His research results
xvi
Contributors
have been published in national and international journals; including the Academy of Management Review, Organization Studies, Journal of Business Ethics, Die Betriebswirtschaft, and the Zeitschrift für Betriebswirtschaft. Sascha Seifert is a Consultant in an International Management Advisory Group. Previously, he was a Research Associate at the Institute for Information, Organization and Management (IOM), TUM Business School. His research is in the area of user-driven innovation, with a focus on the identification of lead users. Cornelia Storz is currently Professor of Japanese Economy at the Faculty of Economics and at the Interdisciplinary Centre for Asian Studies at the Johann Wolfgang Goethe University of Frankfurt, Germany. She received her PhD in Economics from the University of Duisburg with a thesis on the Japanese entrepreneur. Since 1993 she has served intermittently as researcher at the Research Institute of Economy, Trade and Industry, METI (RIETI). Further invitations include those from the Kansai University as a Guest Professor, the Institute of Social Science at the Tokyo University and from the Japanese Institute for Labour Policy and Training. She is a council member of the German Asia Pacific Society, a Board of Trustees member at the Jakob-Kaiser-Foundation and Treasurer of the European Association for Japanese Studies. Her research focuses on the comparison of economic systems, genesis and change of institutions, and comparative institutional analysis. Dennis S. Tachiki is currently Professor on the Faculty of Business Administration at Tamagawa University, Tokyo, Japan. He has held teaching and research positions at the University of Minnesota, the University of Michigan, Sophia University, the Sakura Research Institute and Fujitsu Research Institute. He has published widely in newspapers and journals and given lectures around the world on the Japanese economy and society and on Asian regional affairs. The main focus of his current research surrounds the diffusion of information technology in East Asia (World Bank) and human resource management and capacity building (Pacific Economic Cooperation Council). His most recent publications concern the role of multinational corporations in the process of regionalism in East Asia (Cornell University Press) and the diffusion of information technology in Japan (UC Irvine Center for Research on Information Technology and Organization). Yoshiya Teramoto is Professor at the Graduate School of Asia Pacific Studies, Waseda University. His specialities are corporate strategy, organization theory and strategic knowledge management. After some years of business experience in Fujitsu, he has worked for several universities and research institutes, including the Graduate School of Systems Management, the University of Tsukuba, the National Institute
Contributors 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
xvii
of Science and Technology Policy (Science and Technology Agency, Japan), and the School of Management at Cranfield University, UK. Mark Tilton is Associate Professor of Political Science at Purdue University and holds a PhD from UC Berkeley. He is the author of Restrained Trade: Cartels in Japan’s Basic Materials Industries (Cornell University Press 1996), ‘Ideas, Institutions and Interests in the Shaping of Telecommunications Reform’ in Linda Weiss (ed.) States in the Global Economy (Cambridge University Press 2003), and editor of Regulation and Regulatory Reform in Japan: Is Japan Really Changing Its Ways? (The Brookings Institution Press 1998). He recently was a George Washington University, Woodrow Wilson Center Asian Policy Studies Faculty Fellow.
Preface
This volume is the outcome of an international conference entitled ‘Information and Communication Technologies in Germany, Japan and the US: Institutional Frameworks, Competitiveness and Learning Processes’, which took place in Tokyo in October 2003. The conference was coordinated by the Centre of Japanese Studies of the Philipps University Marburg, Germany, and the German Institute for Japanese Studies in Tokyo, Japan. Most contributions in this volume were presented at this conference. As the organizers and editors, we would like to express our gratitude to the contributors to both the conference and this volume. All contributions are original contributions and are not yet published in other places. Moreover, we would like to thank in particular the Japan Foundation for its substantial support, which made the conference – and its international approach to the subject – possible, and the TÜV Rheinland Group for its additional and important donation towards the publication of the results of the conference. Thanks also go to the Goethe Institute in Tokyo whose rooms and infrastructure gave us a good and stimulating working atmosphere, and to the German Embassy in Tokyo for financial support and patronage of the conference. The venue and the publication would not have been possible without the help of numerous individuals: therefore, thanks go especially to Keiko Asano, Kristin Koop, Alexander Müller, Patricia Sughrue and Eiko Sugimoto. Last, but not least, we would like to express our thanks to two anonymous referees who gave us important stimuli for the finishing of this volume. The encouragement of Rob Langham and his colleagues at Routledge in taking this project through to publication is also much appreciated. We hope that this volume contributes in some way to increasing understanding of how information and communication technologies in three major economies are influenced by their institutional frameworks and now they, at the same time, influence the specific framework in which they are embedded, leading to often surprisingly heterogeneous solutions. Cornelia Storz and Andreas Moerke April 2005
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Abbreviations
1G 2G 3G 4G ABAP ADSL AHS AI AMPS APEC API ARC ARPU ASP AT&T ATIP B2B B2C B2G BOCs BSI BVA CAP CATV CDM CDMA CEN CEO CLUE CM COM CPNs
first generation second generation third generation fourth generation Advanced Business Application Programming asymmetric DSL American Hospital Supply Artificial Intelligence Advanced Mobile Phone System Asia–Pacific Economic Cooperation application programming interfaces Act Against Restraints on Competition (Germany) average revenue per user active serve pages American Telegraph & Telephone Asian Technology Information Program business-to-business business-to-consumer business-to-government Bell Operating Companies Bundesamt für Sicherheit in der Informationstechnik (Federal Office for Information Security) Bit Valley Association competitive access provider Community Antenna Television Contract Design Manufacturing Code Division Multiple Access Comité Européen de Normalisation chief executive officer Comprehensive Loss Underwriting Exchange Contract Manufacturing Commission of the European Communities cross-national production networks
xx
Abbreviations
CPU CRITO CSLR CTO CUST CVS DIN DSL DTI EC ECOM EDI EEIG EITO EMS ERP ETACS ETRI ETSI EU EWR FCC FOMA GATS GATT GPRS GSM GUI GWB Hanes DSD HDE HP HR HTML IBJ IBP ICT IDC IEC IIE ILEC IMG IMT IOC
central processing unit Center for Research on Information Technology and Organization Center for Social and Legal Research (US) customization-to-order customizing convenience stores Deutsches Institut für Normung digital subscriber line Department of Trade and Industry (UK) electronic commerce Electronic Commerce Promotion Council (Japan) electronic data interchange European Economic Interest Grouping European Information Technology Observatory Electronics Manufacturing Services Enterprise Resource Planning Extended Total Access Communication System Electronics and Telecommunication Research Institute European Technology Standard Institute European Union Early Watch Report Federal Communications Commission (US) freedom of mobile access General Agreement on Trade in Services General Agreement on Tariffs and Trade General Packet Radio System Standard Global System for Mobile Communication Graphical User Interface Gesetz gegen Wettbewerbsbeschränkungen Hanes Direct Store Distribution German Retailers Association Hewlett Packard human resource Hypertext Markup Language Industrial Bank of Japan Internet backbone provider information and communication technology/technologies International Data Corporation International Electronic Commission Institute for Information Economics incumbent local exchange carriers Implementation Management Guide International Mobile Telecommunications International Organization for Standardization
Abbreviations xxi 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
IP IPA IPR ISDN ISP IT ITU JAB JADMA JCCI JECC JEITA JETRO JFTC JIPDEC JIS JISA JISC JIT JJA JMRA JuSchG LAN LDP MBA METI MIC MITI MM MOA MOTHER MPHPT MPT NAP NCCs NFO NMT NPD NSF NTT OBM ODM
Internet Protocol Information-Processing Promotion Association (Japan) intellectual property rights integrated services digital network Internet service provider information technologies International Telecommunication Union Japanese Accreditation Board of Conformity Assessment Japan Direct Marketing Association Japanese Chamber of Commerce and Industry Japan Electronic Computer Company Japan Electronics and Information Technology Industries Association Japan External Trade Organization Japan Fair Trade Commission Japanese Information Processing Development Center Japanese Industrial Standard Japan Information Technology Service Industry Association Japanese Industrial Standards Committee just-in-time Japan Juku Association Japan Marketing Research Association Jugendschutzgesetz Local Area Network Liberal Democratic Party (Japan) Master of Business Administration Ministry of Economy, Trade and Industry (Japan) Ministry of Internal Affairs and Communications (Japan) Ministry of International Trade and Industry (Japan) materials management memorandum of agreement Market for the High-Growth and Emerging Stocks Ministry of Public Management, Home Affairs, Posts and Telecommunications Ministry of Posts and Telecommunications (Japan) network access point new common carriers National Family Opinion Nordic Mobile Telephony new product development National Science Foundation (US) Nippon Telephone and Telegraph Original Brand-name Manufacturing Original Design Manufacturing
xxii Abbreviations OECD OEM OSS P3P PC PDA PDC PDCA PHS PP PRD QoS R&D RBOCs RBV RegTP RF SAP SD SIC SII SIIA SME SoC SSCR SYST TACS TCA
Organisation for Economic Co-operation and Development Original Equipment Manufacturing Online Shopping Survey Platform for Privacy Preference Project personal computer personal digital assistant Personal Digital Communication plan-do-check-action personal mobile phones production planning Pearl River Delta quality-of-service research and development Regional Bell Operating Companies resource-based view Regulierungsbehörde für Telekommunikation und Post radio frequency Systeme Anwendungen, Produkte in der Datenverarbeitung sales and distributions standard industrial classification Structural Impediments Initiative (Japan) Software & Information Industry Association (US) small and medium-sized enterprise systems-on-a-chip SAP Software Change Registration workbench activities/system changes Total Access Communication System Information Technology and Telecommunications Association TCP transfer control protocol TDMA Time Division Multiple Access TGA German Association for Accreditation TKG Telekommunikationsgesetz TMS Transport Management System TMSC Taiwan Semiconductor TRIPS Agreement on Trade-related Aspects of Intellectual Agreement Property Rights TRON TRON (Real Time Operating-System Nucleus) Project TSE Tokyo Stock Exchange TÜV Technischer Überwach ungsverein UMTS Universal Mobile Telecommunications System UNCTAD United Nations Conference on Trade and Development VOIP Voice over Internet Protocol VPN Virtual private network WARC World Administrative Radio Conference
Abbreviations xxiii 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
WBWT WTO xDSL
Wissenschaftlicher Beirat beim Bundesministerium für Wirtschaft und Technologie World Trade Organization digital subscriber line
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Part I
Introduction
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
1
Institutions and learning in new industries An introduction Cornelia Storz and Andreas Moerke
ICT (information and communication technologies) are considered to be one of the new technologies with high growth rates predicted for the future. The OECD (Organisation for Economic Co-operation and Development) assesses the ICT industries as ‘drivers of growth’ (OECD 2001a), while others regard them as a new type of general-purpose technology. The insight that countries that are not able to successfully establish themselves in the ICT-related industries may lose important growth opportunities has led to a high awareness of international organizations, suggesting well-suited ICT policies.1 The leading question of this volume is thus addressed to the ways and the extent to which the different institutional frameworks in Japan, the United States and Germany have encouraged and will continue to support the potential of ICT. It starts from the general question of how different institutional settings influence the competitiveness of ICT, and then moves to discuss the degree to which economic actors are capable of designing ICT in the institutional framework within which they operate. The first part of this question – how rules should be designed in order to reach desired results – is a traditional problem in economic policy. It has recently gained new impetus with regard to ICT industries due to the existence of network externalities and the public good character of some ICT sub-sectors. The second part of the question – how institutions and technologies can be designed – is strongly connected to the long-standing issue of path dependence. The volume is thus connected to the general question of how institutions, organizations and new technologies interact. This volume starts with a discussion of what an appropriate institutional framework for exploiting ICT opportunities should look like. An institutional framework is understood as a set of rules that steer behaviour in a certain direction. This definition includes formal rules, such as governmental rules (also called regulation) or private rules (private constitution) as well as informal rules (e.g. customs, values, attitudes). Economic theory views formal institutions not generated via the market but via a top-down process as generally inappropriate since they tend to reduce the genesis of and the competition between solutions. Moreover, centralized institutions lead to a reduction of freedom and choice. Institutions should thus only be fixed in
4
Cornelia Storz and Andreas Moerke
advance when a common interest in a particular institution exists. In other words, the advantages of the common institution must be greater than those found in a comparable situation without the institution. For example, harmonized standards in ICT can reduce the freedom of choice, since products with a lower quality are no longer allowed on the market. On the other hand, this restriction would be appropriate if market failures can be avoided, thus creating a greater economic advantage than a situation without the common standard. In the ICT industry it may be even more necessary to formulate rules ex ante since common interests could be defined as the necessity to overcome network externalities by a common technical infrastructure or to gain credibility by the protection of consumers. In this context, private rules, as a specific form of ex ante formulated institutions, have recently gained attention as an alternative to counteract the problem of the failure of governmental rules. By examining the different economic, political and cultural backgrounds in Japan, the United States and Germany we can better understand not only the reasons for their differences but also the reasons behind different economic policy outcomes. The differences between the Japanese, American and European institutional set-up in ICT are particularly striking. The differences can in part be explained by the fact that government intervention in political processes has a deep-rooted history in Japan. This leads to path dependence in the policy-making process and to some extent reduces the general range of options for market solutions. It may not be astonishing that Japan’s economic outcome in the ICT sector is poor – especially in the area of standardized software, which has induced a broad discussion trying to identify the reasons for this (presumed) weakness (Fransman 1999; Fukutomo 2001; Noguchi 2003). Japanese policy makers and private entrepreneurs seem to be locked into paths that potentially reduce economic welfare. This volume intends to contribute to the understanding of why such lock-ins on the political and entrepreneurial levels take place. In addition it intends to complement previous research focusing on the appropriateness of rules by analysing how different institutional backgrounds influence the implementation of rules. As the informal institutional setting plays a large role in influencing a particular type of implementation, the different informal institutional conditions and their interaction with formal rules are analysed in this volume too. As noted above, the institutional framework steers behaviour in a certain direction. We thus start from the viewpoint that institutions are given, and that they influence the behaviour of economic actors. Indeed, this is a central point in the economic literature. At the same time, however, more research into the genesis of institutions should be conducted. Analyses of how institutions are created lead to a discussion of the mechanisms through which institutional change is achieved. One important presupposition for change is, obviously, a positive attitude towards change. Another important factor is the way economic actors reflect on possible outcomes of institutional
Institutions and learning in new industries 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
5
change. Reforms geared towards changing institutions, doing away with institutions and creating new institutions thus necessitate an institutional setting that is conducive to learning processes. Learning processes are based on comparisons, competition, trial and error, and on the simple insight that a change is more appropriate than holding on to a given institution. Thus, this volume not only contributes to the understanding of why lock-ins take place but also to our understanding of how institutions can be ‘unlocked’, i.e. how institutional change in a more desirable direction can be triggered. The concept of path dependence, which refers to the reasons behind and consequences of lock-ins, is important since it stresses specific historical conditions and acknowledges restrictions for change. Nevertheless, institutions are not simply ‘given’. They are also constructed and altered in order to remain viable. Actors can thus be regarded not so much as ‘mere’ adaptors but as creators – despite the fact that under certain conditions factors such as power or uncertainty may reduce viable options for lock-outs. Institutions themselves, then, can be considered as options, and not simply as restrictions. Such an approach is relatively new in the economic discipline. While it is true that learning theories and constructivist approaches are becoming better known and more integrated, especially in evolutionary and organization theory, institutional economics still tends to overemphasize the impact of institutions and the danger of lock-ins due to a somewhat simplified modelling of the world, still neglecting the creative potential of economic actors. In reference to a new technology such as ICT, however, the fact that entrepreneurial behaviour moves between adaption and creation becomes more evident. At this point a few general remarks about the central ideas of this volume are in order. These include institutions, lock-ins and learning.
Institutions and competitiveness While institutional economics employs an array of approaches to analyse economic problems, all assume that institutions are relevant for growth and development. One point of consensus is that the more transaction costs are reduced, the more transactions take place, and that this leads to a higher level of economic welfare. The quality of institutions thus is said to contribute to the explanation of growth rates and by doing so to a nation’s competitiveness. Competitiveness in relation to firms is understood as the monetary result of operating economically, e.g. market shares, earnings, or other performance indicators. For nations the definition is more difficult. Here, it is understood as a contribution to an increase in welfare, which might be, in the case of ICT, induced by foreign direct investment in the ICTrelated sector or an adequate economic policy. Since not all indicators are equally relevant in the sub-sectors of ICT, the authors of this volume themselves choose the appropriate criteria for each specific case.
6
Cornelia Storz and Andreas Moerke
Institutions include formal as well as informal institutions (e.g. attitudes towards entrepreneurship, innovation and/or risk) so the quality of these different institutions needs to be illuminated. For formal institutions, quality centres on credibility, which guarantees that the functions of institutions, such as the reduction of uncertainty, options for specialization and longterm orientation, can be fulfilled. In the case of informal institutions, empirical research has shown that innovative behaviour in a society, secondary virtues or generalized trust are beneficial for economic development (Knack and Keefer 1997). The quality of institutions can be measured by the collection of both objective as well as subjective data. Both forms are exemplified in this volume.2
Institutions and lock-ins Often, institutional change does not take place to the desired extent, even if certain reforms are seen to increase the economic welfare of all actors involved. One important reason behind holding on to inappropriate institutions is path dependence. Positive externalities lock the actors into a situation that may be beneficial in a particular case but may not be in the interest of the common good. This concept of path dependence is inspired by standardization economics, which tries to explain certain technical developments with reference to positive externalities – the fact that the value of a good often depends on the number of users. The more users acquire the good, the more the value of the good increases. On one side, the value of a good thus results not so much out of its technical qualities but rather out of the size of its network; on the other side, path dependence may lead to market failure, lock-in and a rigidity of less optimal standards. The concept of path dependence, developed in the world of standards, soon became the core concept in analyses of institutions in order to explain why a desired institutional change does not take place even if it would lead to an increase of welfare for all participants. It is well known that North (1990) suggested the transfer of the concept out of the world of technology and into the world of institutions with only minor modifications. Since then the discussion has centred on whether and to what degree such a transfer is appropriate. The discussion shall not be reiterated here, but one important result of this move has been that positive externalities important in the world of standards have indeed been found to play an important role in the stabilization of institutions as well. The value of institutions depends on the framework in which they are embedded. Every reform, therefore, not only requires changing particular rules but also implies changing the embedding institutions as well. Other factors that foster path dependence are uncertainties and the distribution of power. The problem of institutional path dependence is especially endemic to homogenous groups within which the implementation of political reforms is even more difficult than in heterogeneous groups (Eisenberg 1999). As a consequence, the
Institutions and learning in new industries 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
7
competitiveness of nations may be hindered because new developments need new institutional conditions that are not being produced due to pre-existing structures. Research has shown that the concept of path dependence is productive in explaining real economic processes. Obviously, there are ‘limits to institutional reforms’ (Eggertson 1998: 335). Nevertheless, there seems to be a tendency, especially in applied research, to overemphasize path dependence to the extent of becoming deterministic (cf. Crouch and Farrell 2004). The discussion about the decline of Japan’s competitiveness is, for example, often led by the idea that established formal and informal institutions that were helpful in the phase of catching up after the Second World War have now turned into competitive disadvantages, since new technologies such as ICT need different institutions. In other words, the institutional complementarity of the existing and the necessary new institutional conditions is not in place. The development of new industries is hindered because the needed structure is not there. The somewhat simplified argument here necessarily results in a pessimistic evaluation of reforms. It also implies that specific Japanese institutions are the reason behind the failure of certain industries. While cautiously incorporating a question mark, Anchordoguy (2000) explains the Japanese failure in the standardized software industry with a ‘failure of institutions’, including the political set of rules in competition policy, underdeveloped markets and industrial organization. The concept of institutions as a restriction can be found in the management literature about technology as well. As in the case of institutions, the emphasis given to structures lies in the focus of interest. Most contributions focus on the problem of how technology changes organizational structures and point out that organizational design needs to adapt to technological needs. Aoki’s argument (1996) – the ‘unintended fit’ of the Japanese economy as the reason for its success – clearly finds its exact replication here in the discourse of management literature. ‘Forces of production’ or ‘technology as an occasion of structuring’ (Noble 1984; Barley 1986) express a certain perspective on technology, namely that technology shapes the organizational design and that the organizational design has to fit technological needs. Without a doubt, institutions and technological conditions influence economic development, but the editors of this volume want to go a step further and stress that institutions and technologies do not simply exist and should not be interpreted primarily as an impact, but that due to creative actors implementing institutions and technologies, change is permanently going on. Technology thus receives a more plasticide character, since it is continuously adapted and created by the members of an organization. Therefore, the authors of this volume stress not so much the impact but the interaction of technologies (such as ICT) and institutions’ respective organizations.3 To summarize, we view the problematic issue as a strong tendency to perceive institutions and technologies as only marginally designable. The
8
Cornelia Storz and Andreas Moerke
background for this scepticism towards change stems not only from sobering experiences of holding on to familiar routines in the political and economic realms but it is also influenced by a relatively simple model of the world in which path dependence operates independently of actors. However, path dependence is not past dependence. We can expect actors to attempt to calculate the long-lasting effects of their behaviour, and we can expect that they have an interest in finding better, more effective solutions. In this sense, they are inevitably engaged in learning processes.
Institutions and learning In contrast to the interpretation of path dependence outlined above, the editors of this volume stress that institutional path dependence is different from path dependence of technical standards as we learnt it from the economics of standardization. Institutions and technologies do steer behaviour in certain directions, but this is not simply a process of a ‘passive’, costless or uncreative adaptation. Certainly, it cannot be denied that path dependence in institutional development is also a result of complementarities. Nevertheless, one simple insight seems to carry the most weight: actors reflect on formal and informal institutions with reference to their contribution to the expected outcome. We can assume that actors are ready to change an institution when the return from the change is expected to be larger than the cost of the change. This insight leads to a different approach to change and to the potential of developing new industries. Learning takes place primarily in two forms: as a routine-based and as a more active process of problem solving.4 Learning as a routine-based process means that, in the simplest version, learning follows certain stimuli, such as in Pavlov’s well-known experiments. In a broader sense, instrumental learning, understood as learning according to expected rewards and punishments, belongs to this type as well. A further learning type that becomes important in the case of best-practice learning is model-based learning as developed by Bandura (1976). Model-based learning means that learning takes place by observing other groups and then assessing the degree to which their behaviour is more successful than one’s own. Model-based learning is similar to routine-based learning in that the learning situation and the signals underlying how and in which direction behaviour should be changed are relatively clear. From this perspective, learning takes place in a routine manner. Non-routine situations that are not clearly structured and are characterized by a high degree of uncertainty need a higher degree of individual reflection. It is not sufficient to reflect whether another solution might be more appropriate than one’s own. Rather, individuals have to develop their own logic and their own problem-solving strategies. As cognitive science has shown, actors are indeed able to reflect and change their behaviour to a relatively far-reaching degree. This view presupposes that actors are
Institutions and learning in new industries 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
9
convinced that reforms are necessary and that an intrinsic motivation to learn and an open discourse exist. Learning processes in highly uncertain situations thus can no longer follow a specific logic inherent in the system. Instead, they need much more internal control. The most important strategies to handle uncertain situations in this context are trial and error, restructuring and creativity (i.e. the combination of two elements that are far remote from each other). With this approach to path dependence, the lack of complementarities itself may not be as problematic as it seems to be due to the fact that economic actors themselves are able to create new complementarities and are innovative in searching for new, previously ‘unthought of’ institutional solutions (on the condition that changing costs are reasonable). In this context, Vanberg (1994) and Buchanan (1994) have discussed how informal institutions, which are seen in economics for methodological reasons as stable and thus almost resistant to change, can be integrated in this approach as well. They propose dividing preferences – which are here equated with informal institutions – into theory-related and subject-related components. The theoryled component can be tested and is therefore objective from a scientific point of view, while the subject-related component is only subjectively correct. In order to change learned behaviour it is necessary to individually assess whether and to what degree a given preference has led to a desired aim. If an individual concludes that a certain aim was not reached, then the individual may change its preferences. With this understanding, a change of preferences can also be introduced into economic theory. Individuals thus cannot leave existing paths, but they can develop new paths by learning new preferences and new behaviour. Trial and error, restructuring and creativity are important tools by which one can change one’s behaviour and develop new paths (Holland et al. 1986). To summarize: since rules are learnt, not only do they offer orientation by their stability but they can also in principle be relearnt as well. Indeed, several examples in actual political and economic life demonstrate that institutional development is not as rigid as it is often presumed, and that there is plenty of leeway for action. Selected examples discussed in this volume are the success of i-mode in Japan, specific comparative advantages generated out of specific interpretations of ICT in German SMEs (small and mediumsized enterprises), or the genesis of new ventures in ICT in the United States and in Japan despite their completely different institutional conditions (see Schreyögg, Khiari and Schmidt, Chapter 13; Tachiki, Chapter 8; and Akhtar, Teramoto and Benton, Chapter 11). In order to discuss the competitive potential of ICT against the background of appropriate institutional and technological frameworks, path dependence and learning, the volume is structured as two main sections. Part II focuses on the institutional framework and Part III on the entrepreneurial level. Each section takes the specific backgrounds in the United States, Japan and Germany into account.
10
Cornelia Storz and Andreas Moerke
Part II starts from the finding that the institutional framework influences the development of ICT and carefully examines the ways regulation in the political and economic environment takes place in the United States, Japan and Germany. From a dynamic and comparative perspective, we look at the degree to which new rules are seen as necessary and thus influence positively the development of ICT. We find distinct, different and often unexpected patterns between the three systems and, moreover, that selfregulation in all three countries is rapidly beginning to replace governmental regulations. In the first sub-section of Part II, general institutional conditions for introducing ICT are analysed. The leading question is: which set of institutions is appropriate for the development of ICT technologies? As the contributions show, this question has to be answered differently depending on the sub-sector. In contrast to the United States and Germany where the idea of competition policy and a general rule orientation of economic policy is relatively well established despite divergent developments in selected areas, the Japanese case differs since the needs of the ‘developmental state’ have continued up to the present time to lead to comprehensive government intervention. Nevertheless, a more differentiated analysis shows that such tendencies can be found in the United States and in Germany as well. Mark Tilton and Choi Hyeonjung (Chapter 2) discuss the legacies of the developmental state for Japan’s ICT industries. They focus on Japan’s failures and successes in industrial policy for ICT. Japanese policies to promote indigenous technology have limited competition and are focused on funding technologies such as the development of large mainframe computers and a fibre-optic network. While this approach has produced some successes, overall Japan has found itself surpassed by information technology industries in the United States, where policies have emphasized market flexibility and lower prices. Recently, Japan has attempted to catch up by pursuing an ‘e-Japan Priority Policy Program’. While this has produced rapid growth in broadband internet access, regulatory protection of NTT, Japan’s leading telecommunication firm, continues to saddle Japan with expensive basic telecommunication services. The institutional framework and competitiveness of the US telecommunications market is analysed by Michael Schefczyk (Chapter 3). He first briefly puts into perspective the telecommunications markets of the United States, Japan and Germany. He then provides an overview of the development of the regulatory framework in the United States with an emphasis on the Telecommunication Act of 1996. Finally, Schefczyk analyses aggregate statistics to verify broad consequences of regulation on competition in the United States. The highlighted issues include, first of all, the different telecom market characteristics in the three countries in terms of their population density, cellular penetration, digitization and Internet/PC penetration. Second, he finds that the regulatory framework in the United States until 1996 was characterized by a reduction of market entry barriers
Institutions and learning in new industries 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
11
while keeping operators from serving multiple segments and using price cap regulation for incumbent carriers. Third, this framework changed in 1996 to allow operators multiple segments, enforce strong access and interconnection rules, and shift towards intervention-regulation of prices and universal service obligations. Günter Knieps (Chapter 4) discusses with regard to German ICT whether there is still a role for sector specific regulations. The traditionally different roles of government interventions and regulations in the media and in the IT and telecommunication sector may be challenged due to the convergence of all three sectors. On the one hand, convergence may outpace existing sector-specific regimes. On the other hand, sector-specific regulation may even be extended in the future to include markets not yet regulated. The question arises of how to achieve the proper role of government intervention. The focus of this paper is on Internet access and backbone providers. As far as Internet access is concerned, the recent introduction of regulatory unbundling of the European Parliament and the German Regulatory Commission is discussed as problematic due to its lack of justification. However, Internet backbone providers are not subject to sector-specific regulation, which is due to the absence of network specific market power. This results in a competitive market. In the second sub-section of Part II, the state’s decreasing ability to act and the increasing role of self-regulation in the context of ICT is discussed. This sub-section explores the conditions under which private rules may be more appropriate than governmental intervention. It examines the appropriate range of government involvement and the interaction between these two areas. First of all, the contributions draw attention to the fact that the question concerning ‘varieties of capitalism’ has to be considered carefully. In the United States, for example, the country regarded as the ‘model’ for private rules and market solutions, governmental interventions in the ICT sector are playing an increasingly prominent role. Second, most research has so far focused more on the appropriateness of rules and not enough on their implementation. Different social, political and economic contexts influence the implementation of rules, and thus become an important factor in the discussion about the appropriate design of the institutional framework. Cornelia Storz (Chapter 5) asks whether self-regulation is an adequate framework for electronic commerce in Japan. The contribution starts from the point that e-commerce creates new problems of uncertainty, which can lead to low growth rates in e-commerce as the Japanese case shows. Japan views this situation as a politically relevant problem. Although this new form of uncertainty has resulted in several new legal rules for e-commerce, the most adequate solution for this problem is expected to be found in the instrument of self-regulation, i.e. self-binding rules of private associations that define content, enforcement and adjudication. Japan is currently following this approach, as exemplified by both political actors and e-commerce associations. While self-regulation can be a general, adequate
12
Cornelia Storz and Andreas Moerke
political solution for reducing uncertainty and building trust, it seems questionable to choose it as a general, adequate institutional framework for Japan. Nevertheless, other innovative, entrepreneurial solutions have stepped in to solve this problem. Robert Frieden (Chapter 6) analyses the institutional framework and competitiveness of the US telecommunications market. Globalization, free trade and the constant pressure to improve productivity and generate growth create incentives to leverage ICT industries domestically and abroad. Regardless of political orientation, national governments have significant functions in ICT development – despite debates concerning the appropriate range of governmental involvement. Even the US government, which traditionally favours entrepreneurialism and private enterprise in ICT, has a lengthy record of incubation and stewardship that helps to develop and exploit competitive advantages. Since ICT is predicted to have a strong global economic impact, its incubation is a matter of great importance to the United States, Japan and Germany alike with regard to competitiveness. These nations cannot simply assume developing nations serve as untapped markets, since technology transfer may expedite development of a low-cost provider. On a macro level, these nations have to consider the impact of monetary, trade, immigration and industrial policy on ICT. On a micro level, they have to formulate nation-specific ICT policies. Bernhard Lageman, Michael Rothgang and Markus Scheuer (Chapter 7) ask whether e-commerce needs an independent e-commerce policy. From a German perspective, they argue that the debate surrounding e-commerce regulation has centred on single economic and public economics issues such as its impacts on taxation or securing property rights, as well as on social issues such as the protection of youth and individual rights. However, in the light of the important changes associated with technological innovation, emerging new markets and the globalization of business processes, the question as to whether there is a need for a new, ordo-political framework and how such an ordo-political framework for e-commerce could be shaped has not been addressed. Part III turns to the development and the use of ICT at the micro level. This approach illuminates the tension between the adaptive and the creative functions of firms. ICT has an impact on the structure and development of institutions and organizations, but the organization of industries and firms also influence the way in which ICT is generated and implemented; both ‘interact’, as described above. Thus, the first sub-section of Part III further examines the effects of ICT on industrial organization and firm structures in Japanese, the US and German contexts. Dennis Tachiki (Chapter 8) discusses whether Japan indeed lags behind in e-commerce, as is often stated. The analytical concern is in the extent to which e-commerce diffuses across industries and within establishments, and the consequent impacts on firm performance. The main data sources for this study are a telephone survey conducted by the International Data
Institutions and learning in new industries 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
13
Corporation in 2002 and primary data from the Fujitsu Research Institute. Overall, the author finds that Japanese groups (keiretsu) play an important role in adopting business-to-business technologies, but that the small and medium-sized enterprises are more active in adopting business-to-consumer technologies. These differences reflect the ability of companies to overcome the barriers and inefficiencies in the existing political economy. The relational context in Japan partially explains the variations in performance of adopting firms. Boy Lüthje (Chapter 9) discusses in the first subsection the American case: how are organizations influenced by ICT? Lüthje examines the profound economic and social changes in the IT industry that are shaping the conditions for electronic contract manufacturing and the Internet as an infrastructure for transnational production networks. The analysis explores three interrelated sets of questions: (1) the patterns of vertical specialization as they relate to new industry structures and changing architectures of IT networks; (2) the shift from traditional, multinational corporations to global network flagships that develop regional production networks with hierarchical layers of participants; and (3) the possible impact of contract manufacturing on international knowledge diffusion and local capability formation. The discussion is based on recent empirical research that compares the development of the contract manufacturing industry in the southern and western United States. Ralf Reichwald, Frank Piller, Sascha Seifert and Christoph Ihl (Chapter 10) focus on change and reorganization in firms, discussing the deployment of ICT in new product development and thus the central question of innovation management, how new knowledge in firms can be created. The contribution follows an institutional approach, e.g. referring to transaction cost theories. It starts from the concept of ‘open innovation’, which is relatively new in the literature and a fascinating concept since it includes a new perspective of the consumer, whose role is interpreted more as a proactive one. Referring especially to lead users, its role changes from a rather passive to a more creative one are analysed. The concrete example that is discussed in this contribution is based on a research project with adidas-Salomon AG. In the second sub-section of Part III, the other side of the coin is examined through analyses of the social construction of institutions and technology. As the contributions outlined above indicate, the development of new entrepreneurial concepts is an indicator of the alertness of entrepreneurs and may lead to lock-outs of inefficient arrangements. This sub-section discusses the conditions under which learning takes place. Subsequently, we look at how ICT gives rise to learning processes in terms of inducing knowledge creation, which in turn can lead to competitive advantages. Mohammad Akhtar, Yoshiya Teramoto and Caroline Benton (Chapter 11) discuss the next generation communication technologies in Japan and its leading economic position. They analyse competing deployments of mobile communications technology by evaluating the benefits that have been
14
Cornelia Storz and Andreas Moerke
delivered at the end-user level, operator level, country level and global level. They argue that operators who have successfully captured these benefits not only become formidable players in the Japanese marketplace but are also able to create a win–win relationship for all the stakeholders involved. The empirical part of the analysis examines the development of 3G in Japan, where both W-CDMA and CDMA 1X networks co-exist. Furthermore, the chapter gives insight into the reasons that lie behind the competitiveness of Japan and other Asian countries (mainly Korea and China) in technology development and commercialization. Janet Fulk (Chapter 12) analyses the competitive advantage of US firms by their use of ICT. This chapter reviews theory, research and case examples of the strategic application of communication and information technology to improve productivity and enhance competitiveness of firms headquartered in the United States. When implementation of such technologies is accompanied by changes to corporate strategy, culture and organizational design, these innovations can have a longer period of freedom from imitation by competitors. Trends in such applications often involve partnerships and alliances in which the different firms not only coordinate with each other but also learn from their partners as well. Georg Schreyögg, Sami Khiari and Leo Schmidt (Chapter 13) understand the implementation of ICT as a learning process and technology as a process that is shaped. The importance of technology for designing and managing organizations has long been a standing concern of organizational theory. The traditional approach focuses on the technological impact on organizations and how they inform organizational design to effectively match inherent technological demands. Technology is conceived as an external force that demands adaptation along prescribed patterns. This chapter focuses on these changes and advocates the view that the technological process has become an interactive one whereby organizations increasingly shape technologies. Even more importantly, the relationship between technology and organizational design transforms itself into a learning process. The empirical part of this chapter presents results from recently completed research on the introduction of SAP, one of the most-used customized business software, in a medium-sized enterprise. The volume thus is reasoned in the fact that the advanced OECD countries enter into a new era of knowledge-based industries, to which ICT belong. The question arises, how well Japan, the US and Germany are doing in developing these new technologies. Obviously, Japan and Germany possess strengths in industrial core industries but still have to adjust their policies and strategies in response to the new challenge. In contrast, it is often argued that the US system has been more flexible in its adaption process. One indicator is its strong position in software and most ICT technologies. In this volume, we have chosen an institutional approach to understand weaknesses and strengths. This unifying approach implicates a certain tension – namely, the tension between change, adaptability and
Institutions and learning in new industries 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
15
rigidity – that makes, in our view, the volume with its concrete focus on institutional issues and their respective changes (or non-changes) an exciting one. Moreover, we discuss ICT-related institutional, organizational and technological issues in a comparative view between the US, Japan and Germany. This again gives interesting insights in still different, and often unexpected, patterns of institutional settings, but makes the discussion necessarily less focused. Nonetheless, we hope that the reader will enjoy the volume, and that the implicit tensions will stimulate further discussions.
Notes 1
2 3 4
Compare, for example, Custodio (2004) and its analysis of the e-commerce work programme of the World Trade Organization, Hanna, Guy and Arnold (1995) in a discussion paper for the World Bank and publications of the OECD (2000, 2001) referring to selected issues such as privacy and security. On a critical note we should mention that causality flow from institutions to competitiveness has not gone unquestioned. The causality may also flow the other way around: economic growth induces high-quality institutions. In a pointed way, Crouch and Farrell (2004) did this in an interesting contribution about path dependence. For learning theories, see Brenner (2000) and Edelmann (1996).
Bibliography Ackermann, Rolf (2001) Pfadabhängigkeit, Institutionen und Regelreform, Tübingen: Mohr-Siebeck. Anchordoguy, M. (2000) ‘Japan’s software industry: a failure of institutions’, Research Policy, 29: 391–408. Aoki, Masahiko (1996) ‘Unintended fit’, in Aoki, Masahiko, Kim, Hyung-Ki and Okuno-Fujiwara, Masahiro (eds) The Role of Government in East Asian Economic Development, Oxford: Clarendon Press, pp. 233–53. Aoki, M. and Hayami, Y. (2001) ‘Introduction: communities and markets in economic development’, in Aoki, M. and Y. Hayami (eds) Communities and Markets in Economic Development, Oxford: Oxford University Press, pp. xv–xxiv. Bandura, A. (1976) Lernen am Modell: Ansätze zu einer sozialkognitiven Lerntheorie, Stuttgart: Klett. Barley, S. R. (1986) ‘Technology as an occasion for structuring: evidence from observations of CT scanners and the social order of radiology departments’, Administrative Science Quarterly, 31: 78–108. Brenner, Thomas (2000) ‘Diffusion and waves of innovations: a learning perspective’, in Computation in Economics, Finance and Engineering: economic systems – a proceedings volume from the IFAC symposium, Cambridge, UK, 29 June–1 July 1998, Oxford: Pergamon/Elsevier Science, pp. 87–92. Buchanan, James M. (1994) Ethics and Economic Progress, Norman OK and London: University of Oklahoma. Crouch, Colin and Farrell, Henry (2004) ‘Breaking the path of institutional development? Alternatives to the new determinism’, Rationality and Society, 16(1): 5–43. Custodio, Edsel T. (2004) ‘E-commerce: the work program of the World Trade Organization’, in Drysdale, Peter (ed.) The New Economy in East Asia and
16
Cornelia Storz and Andreas Moerke
the Pacific (Pacific Trade and Development Conference Series), London: RoutledgeCurzon, pp. 298–312. Denzau, Arthur T. and North, Douglass C. (1994) ‘Shared mental models: ideologies and institutions’, Kyklos, 47: 3–31. Edelmann, Walter (1996) Lernpsychologie, 5, Vollständig Überarbeitete Auflage, Weinheim: Psychologie Verlags Union. Eggertson, Thráinn (1998) ‘Limits to institutional reforms’, Scandinavian Journal of Economics, 100 (1): 335–57. Eisenberg, A. (1999) Die Lösungen Sozialer Dilemmata und der Wandel Informeller Institutionen, Jena, Max-Planck-Institut zur Erforschung von Wirtschaftssystemen. Fransman, Martin (1999) ‘Where are the Japanese? Japanese information and communications firms in an internet worked world’, Telecommunications Policy 23: 317–33. Fukutomo, Kazuo (2001) IT ka no shinkô to kigyôkan denshi shôtorihiki no kanôsei (Development of IT and the options for business to business e-commerce), Sangyô to Keizai (Nara Sangyô Daigaku), 16(1), June: 1–14. Hanna, Nagy, Guy, Ken and Arnold, Erik (1995) The Diffusion of Information Technology: experience of industrial countries and lessons for developing countries (World Bank Discussion Papers 281). Holland, John H., Holyoak, K. J., Nisbett, R. E. and Thagard, P. R. (1986) Innovation, Processes of Inference, Learning, and Discovery, Cambridge MA: MIT Press. Knack, Stephen and Keefer, Philip (1997) ‘Does social capital have an economic payoff? A cross-country investigation’, Quarterly Journal of Economics, 112 (4): 1251–88. Leipold, Helmut (1997) ‘Der Zusammenhang zwischen gewachsener und gesetzter Ordnung: Einige Lehren aus den postsozialistischen Reformerfahrungen’, in Cassel, D. (ed.) Institutionelle Probleme der Systemtransformation, Berlin: Duncker & Humblot, pp. 43–68. Meyer, John W. and Rowan, Brian (1977) ‘Institutional organizations: formal structure as myth and ceremony’, American Journal of Sociology, 83: 340–63. Noble, D. (1984) Forces of Production: a social history of industrial automation, New York: Knopf. Noguchi, Yukio (2003) ‘Is the IT revolution possible in Japan?’, in Barfield, Claude E., Heiduk, Günter and Welfens, Paul J. J. (eds) Internet, Economic Growth and Globalisation, Berlin, Heidelberg and New York: Springer, pp. 67–83. North, D. C. (1990) Institutions, Institutional Change and Economic Performance, Cambridge: Cambridge University Press. OECD (2000) Transborder Data Flow Contracts in the Wider Framework of Mechanisms for Privacy Protection on Global Networks (Working Party on Information Security and Privacy) (DSTI/ICCP/REG (99)15/FINAL) (www.oecd.org/Long Abstract/0,2546,en_2649_34223_2496189_1_1_1_1,00.html). OECD (2001a) Science, Technology and Industry Outlook. Drivers of Growth: information technology, innovation and entrepreneurship, special edition, Paris: OECD. OECD (2001b) Building Trust in the Online Environment: business to consumer dispute resolution, Joint Conference of the OECD, HCOPIL, ICC (Working Party on Information Security and Privacy) (DSTI/ICCP/REG/CP(2001)2) (www.oecd. org/document/22/0,2340,en_2649_34267_1864982_1_1_1_1,00.html). Vanberg, Viktor (1994) ‘Kulturelle Evolution und die Gestaltung von Regeln’, Walter Eucken Institut, Vorträge und Aufsätze; 144, Tübingen: J. C. B. Mohr (Paul Siebeck).
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Part II
Institutional framework for ICT and options for political governance Japan, the United States and Germany in comparison
Subsection A Institutional conditions for introducing ICT
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
2
Legacies of the developmental state for Japan’s information and communications industries Mark Tilton and Hyeonjung Choi
Introduction The information technology sector tells us much about how Japan’s political institutions shape its economy, and how change and learning takes place. Japanese leaders have observed American success in information technologies and have stated a strong interest in adopting policies that would enable Japan to catch up with the United States. Change and learning have in fact taken place, but in ways that are constrained by established institutions. Beginning in the late nineteenth century, the Japanese state adopted policies to catch up with Western industrial power. These policies evolved and changed over time, but in the period of rapid, post-Second World War growth they emphasized (1) close ties between the bureaucracy and industry and (2) a bias towards helping industries establish private market controls, such as cartels, in order to produce revenues for investment or to maintain jobs. Pro-consumer competition policy has been weak. This does not mean that there is little competition in Japan. However, industrial policy has been based on cooperation between the state and business to coordinate competition and channel resources to targeted industries. Undermining powerful companies by forcing tough, competitive conditions on firms has tended to be viewed with suspicion. The problems of Japan’s industrial policy approach become obvious in the computer and telecommunications sector. Strategies of targeting and managed competition translated into Japan’s late development of widespread access to PCs (personal computers), and this has been one factor that has held the country back in developing access to the Internet. Japan’s cautious approach to subjecting its old telecommunications monopoly, Nippon Telephone and Telegraph (NTT), to competition has meant high telecommunications prices for consumers, which also slowed the spread of Internet use. Recognizing this problem, the Japanese government has again employed a policy of industrial targeting by mandating low prices for digital subscriber line (DSL) interconnection1, while at the same time keeping interconnection rates for standard phone services high.
20
Mark Tilton and Hyeonjung Choi
The resulting low prices for broadband have fuelled a rapid growth in broadband services. Japan has leapfrogged ahead of the United States (not to mention most of Western Europe) in providing very fast broadband service at low prices to such an extent that by 2005, broadband penetration levels were higher than in the United States (Point-topic.com 2005). In addition to Japan’s success in targeting broadband access, the policy of sheltering the national champion has also not been without its advantages. NTT is in better financial shape than many global telecommunications firms, and NTT’s deep pockets have funded considerable innovation, most notably in mobile services and the creation of a fibre-optic network. Yet even without a tough competition policy, NTT is inevitably faced with new competition due to new technologies that have relaxed the advantages of an incumbent monopoly. Recent changes, such as the development of Voice over Internet Protocol (VOIP), the spread of mobile phones and the lowering of obstacles to building a rival, fixed-line network, as J-Phone is doing, all present direct challenges to NTT. Nevertheless, at this point network regulation still matters, and Japan still lags behind Germany and the United States in pushing a tough competition policy in telecommunications. The United States was successful to a great degree in information technologies due to its powerful competition policy. Japan adopted measures along the lines of its own industrial policy to try to catch up with the technology that was the fruit of American pro-competitive policies. Industrial policy enabled Japan to catch up with and surpass the United States in an area where there was a clear goal – the diffusion of high-speed Internet access. But Japan has not adopted the tough competition policy that enabled the United States to be the early innovator in mass use of the Internet. This chapter examines the effects of a pro-competitive versus a pro-industrial approach to information and communications technology (ICT) industries.
Telecommunications The telecommunications and computer industries are closely connected since the same companies produce hardware for both industries. NTT procurement has long been an important source of sales for computer manufacturers, and both industries are now crucial to the development of the Internet. Let us first take a look at telecommunications. As in most countries, Japan’s telephone company was a state-owned monopoly, and in Japan’s case it was self-regulating. It developed technology through what Martin Fransman (1995: 354–8) calls ‘controlled competition’, in contrast to the vertical or in-house procurement of American Telegraph & Telephone (AT&T) or the pure competition for procurement contracts that British Telecom used. This model of controlled competition was a source of trade friction with the United States, and, beginning in 1981, a series of agreements was signed to open up procurement to foreign firms.
Legacies of the developmental state, Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
21
Telecommunications regulation changed in important ways in response to the 1984 US break up of AT&T. In 1985, Japan privatized NTT, although the Japanese government kept most of the shares in NTT and competitors complained that it was difficult to compete in the areas, such as long distance, that were opened up to them because of NTT’s monopoly in local telephone service. A more significant change came with the split up of NTT in 1999. Local service was split between NTT East and West, mobile phone service was given to NTT DoCoMo, and NTT has ceded market share to a number of competitors although it still dominates most branches of the telecommunications market (Vogel 2000; Seifu Kisei Nado to Kyo¯so¯ Seisaku ni Kansuru Kenkyu¯kai 2002). When NTT was broken up, it was only with the important concession that Japan’s 52-year-old ban on holding companies be lifted. The ban on holding companies had been imposed by the US occupation because of suspicion that the zaibatsu, all held together by holding companies, had been responsible for triggering Japan’s military expansion in Asia (Hadley 1970; Haley 2001). Rapid change in the telecommunications market has been accompanied by controversy and political struggle. The policy-making process is deeply politicized, with NTT exerting significant pressure on regulators. Important liberalization and price drops have occurred, but at crucial times regulators lean in favour of protecting NTT. NTT is a large, powerful firm with over 200,000 employees. The NTT labour union is politically active and a number of Diet members are former NTT employees. Moreover, NTT is effective at mobilizing both the ruling Liberal Democratic Party (LDP) and the Democratic Party to support its efforts against regulations disadvantageous to NTT interests. The regulatory process has, nevertheless, created considerable opportunities for competitors. The challengers to NTT – the new common carriers (NCCs) – have taken a large share of the market in all areas of telecommunications service, with the exception of local service. But their workforces are small. The largest, KDDI, has only 10,000 workers. Their pockets are also shallower and their political clout is correspondingly less than that of NTT. The new common carriers have no significant supporters in the Diet (Tilton 2003b). At the time of privatization in 1985, NTT came under the formal jurisdiction of the Ministry of Posts and Telecommunications (MPT, which became part of the Ministry of Public Management, Home Affairs, Posts and Telecommunications in 2000, and was renamed the Ministry of Internal Affairs and Communications, or MIC, in 2004). The MIC is weak relative to the large firm it regulates. It is unable to match NTT’s expertise, budget or political clout with Diet members. The MIC has supported increased competition in telecommunications, but is sensitive to political pressure from the LDP and occasionally shrinks back from regulatory sallies it had intended. In devising its policy, the MIC consults with the Telecommunications Council, which includes representatives from telecommunications
22
Mark Tilton and Hyeonjung Choi
firms, consumers and academics. Several important changes have come about through the Council, although the power of NTT and the dominant, conservative mainstream of the LDP act as a brake on pro-competitive reforms (Kawabata 2001). The Ministry of Economy, Trade and Industry (METI, which before 2000 was called MITI, the Ministry of International Trade and Industry) expresses its opinions about telecommunications policy but does not play much of a role in shaping it. It largely confines itself to complaining from the sidelines that there should be more competition in the telecommunications industry and that high telecommunications prices are a drag on the Japanese economy. In contrast to the United States and Germany, Japan’s competition policy authority has had little influence on the development of the telecommunications sector until recently. In the United States, the Department of Justice played a key role in breaking up AT&T, and the courts have regularly intervened to promote competition. Although Germany came late to promoting competition in telecommunications, a variety of competition policy authorities, from the European Commission, the German courts and the Monopoly Commission, have backed the Regulierungsbehörde für Telekommunikation und Post (Regulatory Authority for Telecommunications and Posts, or RegTP, which on 13 July 2005 was folded into the Bundesnetzagentur, or Federal Network Agency) in its efforts to promote competition. But the corresponding antitrust authority in Japan, the Japan Fair Trade Commission (JFTC), has had little to do with telecommunications. In principle, there is a process in place for the JFTC to advise the MIC on telecommunications policy, but the MIC keeps the JFTC’s role quite limited (Tilton 2003a). The JFTC lacks the staff, expertise and, most importantly, political clout to influence telecommunications policy. Foreign pressure (gaiatsu) has played an important role in shaping Japanese telecommunications policy. US pressure has been especially important, and in recent years the European Union (EU) has also made suggestions for changes in telecommunications policy. Currently, however, the United States is less active in trying to influence Japan’s telecommunications policy. In part, this reflects considerable progress that has already been achieved in reforming telecommunications policy as well as the present existence of major telecommunications firms within Japan that can exert some pressure on their own (Tilton 2003c). It undoubtedly also reflects the American emphasis on securing Japan’s cooperation with American foreign policy rather than pushing for further economic reforms in Japan. In all countries, telecommunications is an industry in which states must intervene deeply and persistently in order to create competition. Most states owned telecommunications companies, and those that did not, such as the United States, had nationally regulated monopolies. States have dispensed with the old telephone monopolies, but have found that in order to create competition they must intervene aggressively. The problem is that
Legacies of the developmental state, Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
23
incumbents own a network of wires and switches, and if they are to be challenged in the marketplace, they must be forced by the state to share the network with competitors. At the same time, incumbents must be allowed to make profits from their network, or they will not be able to or have the incentive to continue ongoing investment in the network. Finding this balance between required competition and allowing incumbents to enjoy the profits of their established positions is difficult, and different states have found different solutions. All telecommunications industries have emerged from the status of a monopoly, and all regulatory authorities wrestle with how much competition to introduce. But in Japan there is an affinity between patterns of industrial policy and a tendency to coddle the telecommunications incumbent. Japan’s industrial policy approach to economic development included the following elements: • •
pursuing sectoral goals (such as the promotion of a particular industry or technology) rather than simply promoting economic growth in general; limiting (but not eliminating) competition in order to produce higher prices and more stable markets as well as to provide funds for investment.
Japan’s telecommunications policy still fits this pattern in important ways. Regulators have targeted the development of certain technologies rather than emphasizing low prices for consumers. Prices have tended to be high by international standards, and regulatory bodies have not pushed competition as much as their American counterparts have. The weakness of Japan’s telecommunications regulators in the face of NTT pressure was illustrated by the MIC’s difficulty in regulating interconnection rates in 2003 and 2004. In April 2003, the Telecommunications Council decided to raise interconnection rates, which are the rates charged to NTT competitors to hook up to its network. This rate increase invited protest from competing firms, from the US government and from the EU because Japanese rates were already twice as expensive as European or American rates. The reason the Telecommunications Council raised the rates was because of pressure from NTT-affiliated Diet members (Zaikai 2003). The rate increase produced a 5 per cent increase in NTT’s interconnection fee revenue in 2003, up to 621 billion yen (Jiji Press Ticker Service 2004). Telecommunications policymaking took a surprising twist in July 2003 when five telecommunications carriers sued the MIC for raising interconnection rates (Japan Times 2003). Observers suggested it was unlikely the suit would succeed, but it was a striking way for the new competitors to voice their objections to Japan’s conservative, anti-competitive policies. The generally gloomy prognoses of the litigation highlight a contrast between Japan’s telecommunications regulatory process and that of the United
24
Mark Tilton and Hyeonjung Choi
States or Germany. In the two latter countries, the courts have played an important role. A year later, in July 2004, the Telecommunications Council tacitly responded to the suit by floating a proposal to gradually reduce interconnection fees over five years. NTT East and West put out a statement the next month that was interpreted as ‘an indication of the vehemence of NTT’s reaction’ to that proposal. NTT called for the outright scrapping of the requirement that it should share the fixed phone network with companies (Jiji Tsûshin Kigyo¯ Nyûsu, 2004). METI has expressed the concern that the high telecommunications prices resulting from limiting competition stunt the growth of downstream uses of telecommunications technology, such as the use of the Internet. Weak regulation has kept prices in Japan higher than in the United States and Germany. OECD data for 2002 and 2004 show Japanese telecommunications prices were considerably higher than in both Germany and the United States, although the decline in the dollar’s value relative to the euro from 2002 to 2004 meant that American prices fell relative to Germany’s (see Table 2.1). High telecommunications prices have meant that dial-up Internet access in Japan has been expensive. Forty hours of Internet usage in 2000 cost $78 in Japan, $49 in Germany and $23 in the United States. Twenty hours cost $59 in Japan, $33 in Germany and $22 in the United States. By September 2002, these figures had dropped, although Japan’s rates were still higher than those in the United States and in Germany (see Table 2.2). The fact that Japanese telecommunications regulation was slower than that in the US or even Germany to push down telephone prices meant that Japanese consumers lacked the cheap phone rates that enabled Americans to experiment with widespread use of the Internet. While broadband was becoming more important in Japan by 2004, the initial move to the Internet had been slowed by high telephone prices.
Table 2.1 OECD composite basket of residential and business telephone charges, August 2002, 2004 Residential (US$)
Business (US$)
Ratio of residential charges to German residential charges
Ratio of business charges to German business charges
2002
2004
2002
2004
2002
2004
2002
2004
722 589 638
1,736 1,128 1,149
1,772 1,384 1,157
1.46 1.00 1.18
1.23 1.00 1.08
1.54 1.00 1.02
1.28 1.00 0.83
Japan 703 Germany 482 United States 567
Source: OECD 2003: 179, 181; OECD 2005: 185, 187. Note: Includes mobile and international calls, excludes VAT.
Legacies of the developmental state, Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
25
Table 2.2 OECD Internet access basket at daytime discounted PSTN rates (US$)
Japan Germany US
2000, 40 hours
2000, 20 hours
2002, 40 hours
2002, 20 hours
78 49 23
59 33 22
54 23 23
34 14 21
Source: OECD 2000a, 2000b, 2003: 170, 172. Note: the public switched telephone network (PSTN) ties together the world’s circuit-switched telephone networks.
The computer industry The computer industry in Japan differs from the telecommunications industry in that it has been under the jurisdiction of METI and thus close to the heart of full-fledged industrial policy. Nevertheless, as mentioned above, it has also been closely tied to the NTT procurement system. In the early 1960s, METI defined computers as a strategic industry. It protected the industry, selected players in it, and provided large subsidies, tax benefits and ‘Buy-Japanese’ policies as well. The computer industry had some of the ideal characteristics for industrial targeting. It was envisioned to produce high quality and high value-added machines, to require few natural raw materials and little energy, to use skilled labour power, and to have expanding domestic and international markets. METI obtained IBM patents and distributed them to select domestic companies with track records – Fujitsu, Hitachi, NEC, Toshiba, Mitsubishi Electric, Oki and Matsushita. All these firms, apart from Matsushita which withdrew in 1964, have remained the major Japanese computer companies and have been strategically protected and subsidized by the government. METI regulated the number of firms in each market segment to allow each firm to achieve economies of scale. After selecting the players, the government needed to create a market environment in which the computer companies would be willing to invest heavily to develop and manufacture sophisticated computers and Japanese users would be willing to purchase them. METI and the computer companies created the Japan Electronic Computer Company (JECC) in the early 1960s, which purchased and then rented out large numbers of Japanese-produced computers (Anchordoguy 1989: 59). In addition to stimulating the demand for and supply of the domestic computer, the JECC played various key roles in developing the industry. By giving the computer makers up-front cash, the JECC made it possible for them to invest heavily in improving their computers. The JECC also acted as a central agent for price and production coordination. In consultation with METI and the computer makers, the JECC set computer rental prices to prevent domestic price competition. In addition, although in general market demand and production costs were the primary determinants of the computers, the JECC used its
26
Mark Tilton and Hyeonjung Choi
discretion at times to pressure the computer makers to produce certain types of machines or to develop more advanced computers before market demand and production costs warranted such a strategy. For example, as the domestic computer makers became competitive in small computers, the JECC shifted its support from the rental of small computers to mediumscale and large-scale ones, and encouraged the makers to increase their efforts to develop and produce larger and more sophisticated computers. The computer industry was supported by subsidies, low-interest loans and loan guarantees, which gave the industry access to capital for research and development (R&D) and production.2 NTT played an important role in providing invisible financial support for the ‘NTT Family’ firms to create a computer industry. Financial support by NTT, which was just partly privatized in 1985, provided an additional means of putting public investment into the computer sector. Japan’s developmentalist strategy paid off in the late 1970s. Not only did Japan become a top exporter of semiconductors, but in 1979 Fujitsu became the first maker to beat IBM in any domestic market. Also, Japan became a mainframe computer exporter (Shimoda 1984: 178). But the advent of a new information technology environment in the 1990s proved that METI’s ‘vision’ for the computer industry and its development strategy was ultimately misguided. While Japan succeeded in catching up with existing technologies, it now faced an IT revolution based on personal computers, software-centred computer business and PC networks. METI failed to foresee the new computing world that was about to dawn in the early 1990s. The 5th-Generation Computer Project (1982–91), which was planned and implemented at a time when Japan had begun to have confidence in its computer industry, was an exemplary case. It was an ambitious attempt to develop a ‘thinking’ computer, the so-called Artificial Intelligence (AI) computer, which would infer new knowledge from its database, compute through its self-developed program and produce its own output without any input. It turned out that the massively funded national project had rather vague goals that were not realized.
Challenges to Japan’s PC and software industries In the United States, where the PC revolution originated, IBM deliberately adopted an ‘open-but-owned’ architecture strategy in order to attract thirdparty support in complementary products – components, expansion boards, peripherals and software. It published critical technical specifications that allowed outside independent vendors to develop peripheral devices and application software. Moreover, Microsoft’s operating system software controls the ‘application programming interfaces’ (APIs), which determine which software is compatible with which platform with the same ‘openbut-owned’ strategy. The open-but-owned strategy of the two major US computer firms not only included continued R&D but also drew on
Legacies of the developmental state, Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
27
outsourced peripherals and applications based on their architectures. Thus, with their core technologies in microprocessors and operating system software, Intel and Microsoft could increase the network externalities of their products, and their own profits as well (Egan 1996; Gevel 1997). The major Japanese mainframe computer makers also produced PCs based on spin-off technology from mainframe computers. They developed their own models of PCs based on ‘closed-and-owned’ architectures as they had done for their mainframe computers, which are incompatible with one another. By the early 1990s, NEC created a comfortable lead of more than 50 per cent with its PC-98 models in the domestic PC market. NEC’s ‘closedand-owned’ standards strategy enabled it to keep prices high and earn fat profit margins from locked-in Japanese users. However, NEC suffered from a problem endemic to successful pioneers, which West and Dedrick (2000) call ‘incumbent inertia’, because it failed to continue to develop advanced technology for users and network externalities. In other words, it failed to stimulate a network of software built on NEC’s own technology. Another serious problem was that, since funding was so dependent on government and keiretsu-based sources, there was no technological or financial basis for independent PC makers – such as Compaq and Dell of the United States – to enter the market. METI and the major computer companies never allowed their cooperatively developed computer engineering technologies to spill over to outsiders. Furthermore, it was difficult for outsiders to get financial support from the government or keiretsu-centred banks to start up their businesses. For software developers in Japan, it was harder to obtain funding from private investors because they were not producing tangible goods. Limited competition in the Japanese market has meant that the prices of PCs have been high and thus the PC penetration rate has been low even up to recent years. Japan’s weaknesses were exposed in 1992 when Compaq began selling PCs equipped with globalized Wintel architecture and with the availability of a larger software library at about half the level of prevailing Japanese prices. When Compaq began shipping cut-price computers to Japan, it captured only 4 per cent of the market, but computer prices in Japan fell abruptly by one-third (Edwards 1999: 5–7).3 Microsoft’s PC software, DOS/V (introduced in 1990 by IBM Japan) and Windows (1992), forced Japanese PC makers to abandon their ‘closed-and-owned’ architectures. This was a revolutionary development as DOS/V took a ‘software’ approach to displaying Japanese kanji, in contrast to Japanese PCs’ use of a ROMbased ‘hardware’ approach. Moreover, Windows (version 3.1) introduced the user-friendly Graphical User Interface (GUI) to the Japanese and the world market, which vastly simplified PC use. Once Wintel arrived on the scene, Japanese PC makers’ high prices cost them market shares. NEC’s market share slipped from over 70 per cent during the 1980s to 43 per cent in 1994 and again to 32 per cent in 1997. Moreover, Japan’s software makers have been overwhelmingly outpaced
28
Mark Tilton and Hyeonjung Choi
Table 2.3 Japan’s software exports and imports, 1995–2000 1995
1996
1997
1998
1999
2000
Exports 3,931 To the US 813 (Japan’s share of (20.7%) total world exports to the US)
5,679 1,394 (24.5%)
2,812 913 (32.5%)
8,752 1,251 (14.3%)
9,292 3,255 (35.0%)
8,981 3,934 (43.8%)
Imports 392,576 From the US 351,348 (US share of total (89.5%) imports into Japan)
393,540 344,978 (87.7%)
474,913 391,302 (82.4%)
595,165 544,051 (91.4%)
720,104 659,142 (91.5%)
918,860 821,653 (89.4%)
Source: JISA (Japan Information Technology Service Industry Association) 2001.
by American software makers, as seen in the trade imbalance between the two countries in this sector. From 1995 to 2000 Japan exported only 1 per cent of the value of software it imported, and some 90 per cent of its imports were from the US (see Table 2.3).
Industrial policy approach to the challenges METI did not pay much attention to software engineering when it established the Japanese computer industry. The first R&D project for the computer industry in 1962, the FONTAC Project, did not include software development but consigned it to an American software company. In 1966, in order to promote software development for the Super High-Performance Computer Project (1966–71), METI established the Japan Software Co., reluctantly joined by NEC, Fujitsu, Hitachi and the Industrial Bank of Japan (IBJ). As soon as the project ended in 1972, the company went bankrupt. METI next tried a different approach with the Software Module Project (1973–75), which concentrated 40 independent software vendors into five groups to develop various application programs. However, without any central body coordinating the participants, each of the five groups developed its own programs based on totally different standards so that the software programs developed by the participants could not communicate with others. Also, because the software programs did not address users’ demands, they were ignored in the marketplace. Inspired by the JECC’s success, the METI designed another national policy institution to play a similar role in the software business. The Information-Processing Promotion Association (IPA) was established in 1970 with a large government budget to promote the efforts of independent software vendors to develop standardized packaged software. The IPA rented out the software packages developed by the domestic software companies
Legacies of the developmental state, Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
29
at low rental fees, just as the JECC did for hardware, and issued loan guarantees to the software companies that needed venture capital. However, despite the fact that the IPA paid independent software vendors more than 10 billion yen to develop some 70 software programs from 1970 to 1978, very few of these programs were ever rented or sold on the open market (Kompyutopia 1978: 72). In the mid-1980s, METI recognized Japan’s dependence on IBM’s architecture for operating mainframe computers and tried to reduce it through another national project, the Software Industrialized Generator and Maintenance Aids Project (the Sigma Project, 1985–89). The programme was funded by the government as well as by 164 private firms, including foreign computer makers such as IBM, Hewlett Packard and AT&T. The Sigma Project also tried to reduce redundant software development due to incompatible hardware and to increase the supply of software engineers. Today, however, there is nothing much to show for the Sigma Project: there are no longer any Sigma Workstations or Sigma operating systems in existence, and software developers are still in short supply (Fujiwara 1995). While METI gave priority to the development of mainframe computers to catch up with IBM, it did not learn valuable lessons to be applied in the subsequent PC era. It did not recognize the global trend in the computer industry – the ‘open-but-owned’ strategy and the importance of network externalities from the standpoint of standardization. As a result, Japan developed a PC industry without standardized architecture for either hardware or software. It was Japan’s private sector that noticed the significance of standardization. Initiated in 1984 by Sakamura Ken, a professor at the University of Tokyo, in collaboration with all the leading Japanese computer firms without direct government intervention, the TRON (Real Time OperatingSystem Nucleus) Project aimed at developing its own hardware and software independent of Western standards and even used processing based on Japanese language commands. Despite the much boasted goal of producing a hi no maru (‘rising sun’) computer, the TRON Project failed. Because TRON PCs were incompatible with the architecture of both IBM-compatible PCs and NEC’s PC-98s, they could run none of the software for these platforms, which had become standard in the Japanese market. Although METI and the Ministry of Education favoured the TRON PCs, the Japanesedesigned PC failed to become a standard for educational purposes and never caught on in the commercial marketplace (Callon 1995). No matter how superior the TRON PC was in technological terms, it could not succeed because it did not consider network externalities. None of the government’s attempts to develop Japan’s own architecture and to reduce its dependence on the American-made standard succeeded. Due to the repeated failures of national projects and an economic slump in the 1990s, METI and Japanese computer firms practically gave up on trying to compete with the Wintel architecture and instead decided to use it themselves.
30
Mark Tilton and Hyeonjung Choi
Outcome: weak PC penetration The outcome of Japan’s industrial policy approach to computers is that Japan has been slow to expand PC penetration and thus slow to give consumers and businesses a platform for access to the Internet. In 2000, only about 28 per cent of Japanese households had at least one PC, about half the level of the United States. Most Japanese businesses had PCs that year but only 20 per cent had one for each employee (Reuters 2000). In mid-2000, one report stated, ‘In Japan, PC prices are still not considered cheap, and wired infrastructure is expensive to install’ (Scuka 2000). By 2001, over 40 per cent of Japanese households were estimated to have at least one PC, but this was still lower than the PC penetration rate of South Korea, Australia and Singapore at 64–68 per cent and Taiwan, Hong Kong and New Zealand at 58–59 per cent (Nielsen-NetRatings Co. 2001).
Targeting broadband In July 2000, the Japanese cabinet attempted to get around NTT’s lack of initiative in promoting Internet access and MPT ineffectiveness by establishing an IT Strategy Headquarters in order to promote comprehensive measures for the creation of an internationally competitive ‘IT nation’ (IT Strategy Headquarters 2001a: Preface). The government also established an IT Strategy Council, which put forth a ‘Basic IT Strategy’. On 6 January 2001, a new law came into force, the Basic Law on the Formation of an Advanced Information and Telecommunications Network Society (The IT Basic Law) (Advanced Information and Telecommunication Society Promotion Headquarters 2001). In certain respects, the e-Japan Strategy is reminiscent of traditional industrial policy. For one thing, the policy is clearly oriented towards industrial catch-up: as for penetration rate of the Internet, which other nations identify as the core of IT policies, Japan is still at a low level among major industrialized nations. . . . [T]he swift action of a nation in creating an environment necessary for realizing an advanced information and telecommunications network society determines the nation’s world competitive leadership in the 21st century. As nations in the Americas, Europe and Asia are intensively promoting efforts to create such environments, the delay in the world arena for socioeconomic structural changes at a high speed will result in irrecoverable competitiveness gap [sic] in the future. (IT Strategy Headquarters 2001b: Sec. I) The report echoes the concern of the Meiji period (1868–1912) with catching up with the West and, in fact, refers to that earlier time: ‘Just as a nation’s
Legacies of the developmental state, Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
31
response to the Industrial Revolution later determined its economic prosperity, the same will hold true with the IT revolution’ (IT Strategy Headquarters 2001b: Sec. I). The report states that the key problem is the high prices created by NTT’s monopoly, and puts forth the goal of providing high-speed Internet access. The new policies were quickly put into practice by the MIC. Japan’s stodgy regulatory policy had allowed little room for competitors to break into the market and promote Internet access. And NTT itself was doing little to promote broadband. It had decided early on a strategy for expanding Internet access by focussing on ISDN (integrated services digital network) and on creating a fibre-optic network. These technologies ended up being obsolete, and NTT had also been standing in the way of other companies introducing DSL. Based on this new initiative from the Cabinet, policy changed quickly. On 31 August 2000, the Telecommunication Council recommended introducing local loop unbundling for copper cables. New carriers were able to connect their networks to NTT East and West networks and to put their equipment in NTT buildings (collocation) as of December 2000. Several new carriers, as well as NTT East and West, immediately began offering DSL services (Ministry of Public Management, Home Affairs, Posts and Telecommunications 2001). In March 2000, regulators forced NTT to allow access to local lines (local loop unbundling). NTT East and West tried to prevent competitors from entering the market by opening only eleven offices for collocation. It also limited equipment space and the number of lines provided for unbundlers, and it dragged its feet in processing applications. In July 2000, the MPT forced NTT to open all its offices to collocation, but by December 2000 Japan still had fewer than 10,000 DSL subscribers (Point-topic.com 2003). Then, in December 2000, the MIC forced NTT to unbundle its lines to enable competitors to use them to provide DSL service. The charge for using the lines was dropped from 800 yen to 187 yen per month, which stimulated new entrants to provide DSL service, notably Yahoo Japan (Pointtopic.com 2003). The selection of DSL service as strategic and the regulatory decision to force NTT to cut charges for this service is in the tradition of earlier industrial policy. The policy was similar to previous policies in that it led to differential pricing. Since broadband Internet access was identified as the key problem, regulators required NTT to sharply cut its interconnection charges to encourage DSL access, making it possible for Yahoo to enter the broadband market and sell broadband access at some of the world’s lowest prices. Not long afterwards, interconnection rates for fixed line phone services not identified as strategically important were actually raised, even though they were already twice or more the rates of other advanced countries. In this way, Japan managed to catapult itself to the front ranks of the major industrialized countries in broadband, while at the same time keeping
32
Mark Tilton and Hyeonjung Choi
NTT revenues up (OECD 2004: 4). The United States and the EU complained that Japan’s high interconnection rates were unjustified because NTT was inefficient and its costs needlessly high. High interconnection charges kept phone rates high, kept NTT employees on the payroll and funded an unusual national project of fibre-optic network. Some American critics have argued that the United States has been wrong to press Japan to make NTT unbundle access to its network (Rohlfs and Sidak 2002). And indeed, in 2003, the United States Federal Communications Commission (FCC) voted to phase out the low prices for unbundled access to the local loop that had been required. The purpose was to encourage local providers to invest in infrastructure, especially fibre to the home, much as the MIC had been requiring. The ruling is that the incumbents will be allowed to close off access to their fibre networks, although they must continue to discount access to their copper wire networks. Surveys by Ipsos-Reid found that Japan was slow to develop widespread use of the Internet. In 1999 twice as many Americans and Canadians used the Internet as Japanese, Britons or Germans. By 2005 the Japanese, as well as Europeans, had closed most of the gap although Internet use continued to be most widespread in the US (see Tables 2.4 and 2.5). While Japanese industrial policy succeeded in creating widespread access to broadband, the industrial policy approach has left Japan behind Germany and the US with respect to broader successes in information technologies. The 2005 Economist Intelligence Report sums up Japan’s weakness. The report evaluates all of the advanced industrialized nations and many developing countries on their capacity to promote information and communications technology services. The US ranks second among the
Table 2.4 Percentage of adults who went online at least once in last 30 days (1999–2003); percentage of population who are Internet users, July 2005
US Canada South Korea UK Japan Germany France Urban Mexico Urban China Urban Brazil Urban India Urban Russia
1999
2000
2002
2003
2005
59 56 31 33 33 29 22 27 12 21 21 5
59 56 45 35 33 29 30 33 21 22 10 6
72 62 53 50 47 43 37 37 30 24 19 8
68 71 70 64 65 60 43 37 41 21 19 10
69 64 63 60 61 57 42 14 8 12 4 16
Source: Ipsos-Reid, 2002, 2004. Note: Data for 2005 are for the entire populations of Mexico, China, Brazil, India and Russia.
Legacies of the developmental state, Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
33
Table 2.5 Fixed Internet subscribers/100 population
Canada Japan US Germany
Dial-up 2003
Broadband 2003
Total 2003
Broadband, Aug. 2005
2 8 16 24 22
33 14 10 9 6
34 22 26 33 28
73.0 42.7 38.6 29.9 14.8
Source: OECD 2005: 126; European Travel Commission 2005.
65 nations, after Denmark, with a score of 8.73 (See Table 2.6). Germany ranks twelfth, with a score of 8.03, tied with Canada, and behind the UK and Nordic countries, but ahead of France and Italy. Japan ranks twentyfirst, with a score of 7.42. Japan lags behind all the north-western European countries and the wealthy East Asian economies (with the exception of Taiwan). The scores for individual categories of readiness generally show the same rank order, with the US ahead of Germany, which in turn is ahead of Japan (see Table 2.7). This rank ordering reflects Japan’s earlier stubbornness on insisting on a mercantilist, go-it-alone policy in developing computers, which put Japan behind Western countries in the development of computers, software and broad computer skills among the population. As witnessed in Japan’s continued high telecommunications prices, Japan lags behind the US and Germany in adopting policies that emphasize competition and consumer benefit.
Table 2.6 E-readiness rankings, 2005 Rank
Country
Score
Rank
Country
Score
1 2 3 4 5 6 (tie) 6 (tie) 8 9 10 11 12 (tie) 12 (tie) 14
Denmark US Sweden Switzerland UK Hong Kong Finland Netherlands Norway Australia Singapore Canada Germany Austria
8.74 8.73 8.64 8.62 8.54 8.32 8.32 8.28 8.27 8.22 8.18 8.03 8.03 8.01
15 16 17 18 19 20 21 22 23 24 25 26 27 28
Ireland New Zealand Belgium South Korea France Israel Japan Taiwan Spain Italy Portugal Estonia Slovenia Greece
7.98 7.82 7.71 7.66 7.61 7.45 7.42 7.13 7.08 6.95 6.90 6.32 6.22 6.19
Source: Economist Intelligence Unit 2005: 6–7.
34
Mark Tilton and Hyeonjung Choi
Table 2.7 E-readiness rankings, by category, 2005 Category weight Connectivity Business environment Consumer and business adoption Legal and policy environment Social and cultural environment Supporting e-services Overall score Overall rank among 65 countries surveyed
0.25 0.20 0.20 0.15 0.15 0.05 1.00
US 7.65 8.57 9.80 8.41 9.20 10.00 8.73 2
Germany Japan 6.40 8.23 9.10 8.09 8.60 9.25 8.03 12
6.90 7.34 8.00 7.27 7.60 8.00 7.42 21
Source: Economist Intelligence Unit 2005: 5, 22–3.
The political reasons for weak pro-competitive regulation In comparing Japan to the United States and Germany, the most striking difference is Japan’s lack of organizations that routinely help the telecommunications regulator promote competition. The courts play no role and the JFTC intervenes very little. NTT and its labour union have been much more effective at keeping regulators at bay than either AT&T or Deutsche Telekom. Why? Are things likely to change? It is often said that the reason procompetitive policy in Japan does not go forward is because of the opposition of powerful interest groups. We certainly can see this effect in the case of Japanese telecommunications. But German and American reformers also faced determined opposition from incumbent firms and unions. In the case of the United States, this was overcome due to a new consensus that developed among economists in the 1970s against regulation (Derthick and Quirk 1985). What made it possible was the fact that antitrust law was well established and the court, agreeing with the economists’ new consensus, was able to intervene and order the break up of AT&T in 1982. The political explanation for why the court was able to do this goes back to the populist antitrust movement that led to the 1890 Sherman Act (Freyer 1992). In no other country did a grass-roots political movement lead to the creation of a national antitrust policy. The political thinking in the United States that served as the foundation of the antitrust movement was the opposition to the concentration of power from the time of the revolution against the British Crown. The antitrust movement was based on political opposition to the concentration of power, rather than on ideas about the efficiency of competitive markets. In the case of American telecommunications regulation, existing ideas and institutions provide a repertoire from which contemporary political actors can draw inspiration and political support for new policies. The same populism that drove the antitrust movement also supported a regulatory
Legacies of the developmental state, Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
35
movement that created regulated prices and monopolies in the 1920s for a wide range of public utilities, such as telecommunications, electric power and transportation. The antitrust legal doctrines that developed from the populist antitrust movement then gave economists and politicians sources to take the opposite approach to breaking up AT&T’s monopoly in the 1970s and 1980s. Thus, US telecommunications policy is path dependent, as can be seen in the different paths taken by US policy compared to that of Japan and Germany, although the same path has led to different outcomes at different times. Germany was slow to break up its telecommunications monopoly. In that country there were two sources of change. The first was learning from abroad, as the thinking and experience of the United States and the United Kingdom changed minds in Germany about telecommunications regulation. This, however, did not directly lead to change. The second source of change was the drive from within the EU to create a unified market. This drive itself did not quickly result in tough, pro-competitive regulatory policy, nor was it by any means solely responsible for the change. But the framework of the EU created an arena within which European nations could negotiate the ‘mutual disarmament’ of their national telecommunications monopolies. Each state could agree to break up its monopoly, knowing that while they would open their own national markets to competition they would gain access to other national markets (Thatcher 2002). EU regulatory institutions gave assurance that market access would be real. Looking back further, the roots of the EU, similar to the American antitrust movement, come from non-economic political motives – an anti-nationalist drive to prevent war from reoccurring. While it is true that the push to integrate the European economy in the 1980s and 1990s had strong economic motivation, the political framework within which it was possible was created for noneconomic reasons. In industries with large economies of scale, of which network-based industries such as telecommunications are good examples, two powerful forces can block pro-competitive regulatory reform: (1) the political power of established firms with market power, such as AT&T, Deutsche Telekom or NTT and (2) nationalism, which supports defending national champions against foreign competitors. Against these anti-competitive forces, ideas favouring competition for the sake of efficiency are in themselves not very powerful. In the case of the United States and Germany, as well as the EU more generally, liberal economic ideas alone do not explain the choice to move to pro-competitive regulation. Instead, non-economic ideas about political power provide the deep roots of institutions that promote competition: in the United States the populist roots of antitrust and in Germany (and Europe more broadly) the anti-nationalist, anti-war roots of the EU. In Japan there is no comparable history of powerful, non-economic motives for creating competition policy. While some economists and some political leaders espouse liberal economic ideas, there are no powerful political
36
Mark Tilton and Hyeonjung Choi
movements or enforcement organizations to make these a reality. Antitrust in Japan was imposed from outside and has been met with suspicion (Tilton 1996). When, during the 1989–91 Structural Impediments Initiative (SII), the United States put pressure on Japan to adopt a stronger antitrust policy, there was powerful opposition to change, principally from the main Japanese big business organization, the Keidanren (Schoppa 1997). Nevertheless, the Japanese Diet did increase penalties to a modest extent under the Antimonopoly Law at that time and approved a gradual increase in the staff and budget at the JFTC. In 2003, the JFTC floated a new proposal to strengthen the Antimonopoly Law, but again was met with stiff opposition from NTT, the MIC, METI and especially from the Keidanren and the construction industry. The principal forces favouring the reform are the JFTC itself and Prime Minister Koizumi. Remarkably, some rank-and-file LDP members, especially junior members, also supported the revisions and opposed both the Keidanren and the construction industry (Nihon Keizai Shimbun 2004). In October 2004, the LDP agreed to increase penalty levels and to carry out further reforms of the Antimonopoly Law at a later time, although the increased penalties were not actually enacted. In April 2005 the Diet did pass a bill to increase penalties under the Antimonopoly Law, and perhaps the JFTC may begin to act more aggressively in the area of telecommunications and communications technology. In the past, the JFTC was constrained by LDP threats to take away JFTC funding and powers if it aggressively enforced the Antimonopoly Law (Beeman 2002). Yet tougher penalties combined with enthusiastic support from a large block of LDP diet members could translate into a significant change in Japanese competition policy. Again, learning and change have happened in Japan, but they have been constrained and shaped by established ideas and institutions. Radical liberal ideas about challenging national champions with competition are not easily introduced anywhere. In the United States and in Germany they required the support of non-economic political campaigns. This new, nearly successful initiative by the JFTC represents a political change in Japan, but this political change is weak and gradual compared to the tremendous increase in antitrust enforcement in Europe. The major challenge confronting Japan is how to raise the prestige, resources and political power of the JFTC to a level sufficient for it to intervene effectively in the ICT sector.
The Future of e-Japan The result of Japan’s use of high prices for telecommunications is that NTT has had a reliable source of funding for developing mobile telephony, and Japan is widely regarded as a leader in this technology, as can be seen in Figure 2.1. But high telecommunications prices held back Japan’s early development of Internet technologies, and industrial policy has not produced much success in computers or computer software.
Legacies of the developmental state, Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
37
Figure 2.1 International comparison of superiority in information and communications technologies Source: MIC 2003:10. Note: The figure shows the results of a survey of researchers in Japan and other countries as to which country or region was superior in each of the above information and communications technologies.
Japan still lags behind in many IT sectors, especially software, Internetand content-related businesses. In the early 1960s, Japan’s industrial policy for developing the computer industry began with regulating players and selecting the keiretsu-based, large electronics companies to be players in the industry. Because of successful results in the mainframe hardware sector, institutional inertia has held back further development of the computer industry and the national effort to become an IT superpower in the PC era. Japan’s developmentalist policies, while good at catching up, are holding it back in new fields. While Japan’s deft, differential pricing for access to NTT’s network has made rapid expansion of Internet access possible while simultaneously keeping NTT employees on the payroll, Japan may be missing out on other opportunities due to high telecommunications prices. Much of the success of Japan’s economy is based on the quickness of Japanese firms to learn from other countries. Japanese firms in electronics and other fields were able to ‘reverse engineer’ new technologies created abroad, make successful incremental improvements and produce goods efficiently. Marie Anchordoguy (2001) argues that Japan’s economy has slumped because it is no longer possible to maintain high growth rates based merely on incremental improvements on new technologies from abroad. On the other hand, Kozo Yamamura (2003) argues that as information technology passes its period of rapid innovation, Japan’s (and Germany’s) ability to learn radically new technologies from other countries and then improve on them will stand them in good stead. The jury is still out as to whether this will work for Japan in the computer and telecommunications sector.
38
Mark Tilton and Hyeonjung Choi
Notes 1 2 3
A digital subscriber line connects a subscriber to the Internet over copper telephone wires without interfering with voice telephone calls. Calculations are based on Japan Electronic Company Ltd (various dates) and appendixes in Anchordoguy 1989. Even a modest volume of foreign goods sold into the Japanese market can do a lot to encourage price competition. When Daiei, a supermarket retailer, decided in 1994 to import Dutch beer, domestic beer prices also dropped by one-third. The liberalization of petrol imports in 1996 reduced prices by a quarter.
Bibliography Advanced Information and Telecommunication Society Promotion Headquarters (2001) Basic Law on the Formation of an Advanced Information and Telecommunications Network Society, The Japanese Law no.144 of 2000, accessed at www. kantei.go.jp/foreign/policy/it/index_e.html. Anchordoguy, M. (1989) Computers Inc.: Japan’s challenge to IBM, Cambridge MA: Council on East Asian Studies, distributed by Harvard University Press. Anchordoguy, M. (2001) Whatever Happened to the Japanese Miracle?, Japan Policy Research Institute Working Paper 80: 1–6. Beeman, M. L. (2002) Public Policy and Economic Competition in Japan: change and continuity in antimonopoly policy, 1973–1995, Japanese studies series, The Nissan Institute/Routledge, London and New York: Routledge. Callon, S. (1995) Divided Sun: MITI and the breakdown of Japanese high-tech industrial policy, 1975–1993, Stanford CA: Stanford University Press. Derthick, M. and Quirk, P. J. (1985) The Politics of Deregulation, Washington DC: The Brookings Institution. Economist Intelligence Unit (2005) The 2005 E-readiness Rankings, White Paper, Economist Intelligence Unit, accessed at http://graphics.eiu.com/files/ad_pdfs/ 2005Ereadiness_Ranking_WP.pdf on 12 October 2005, pp. 6–7. Edwards, B. (1999) ‘Brace yourselves’, The Economist 353 (8147), November 27: 5–7. Egan, Edmund A. (1996) The Era of Microsoft? Technological Innovation, Network Externalities, and the Seattle Factor in the US Software Industry, BRIE Working Paper 87, Berkeley CA: Berkeley Roundtable on the International Economy. European Travel Commission (2005) New Media Review, accessed at www.etcnewmedia.com/review/default.asp?SectionID=10&OverviewID=8 on 8 October 2005. Fransman, M. (1995) Japan’s Computer and Communications Industry: the evolution of industrial giants and global competitiveness, Oxford and New York: Oxford University Press. Freyer, T. (1992) Regulating Big Business: antitrust in Great Britain and America, 1880–1990, Cambridge: Cambridge University Press. Fujiwara, H. (1995) Kono Gyo¯kai no Okite! (The Way the [Computer] Industry Really Works!), Tokyo: Gijutsu Hyo¯ronsha Co. Gevel, A. J. W. van de (1997) Wintelism and Production Networks in the Electronics Industry, Research Memorandum 750, Tilburg, Netherlands: Economics and Business Administration, Tilburg University. Hadley, E. M. (1970) Antitrust in Japan, Princeton NJ: Princeton University Press.
Legacies of the developmental state, Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
39
Haley, J. O. (2001) Antitrust in Germany and Japan: the first fifty years, 1947–1998, Asian law series, vol. 16, Seattle WA: University of Washington Press. Ipsos-Reid (2002) Internet Use Continues to Climb in Most Markets, 10 December, accessed at http://stat.nca.or.kr/news/files/ipsos-raid2.pdf on 20 October 2003. Ipsos-Reid (2004) Ipsos News Center Report, accessed at www.etcnewmedia.com/ review/default.asp?SectionID=10 on 8 October 2005. IT Strategy Headquarters (2001a) ‘e-Japan Strategy’, available at www.kantei. go.jp/foreign/policy/it/index_e.html. IT Strategy Headquarters (2001b) ‘e-Japan Priority Policy Program’, available at www.kantei.go.jp/foreign/policy/it/index_e.html. Japan Electronic Computer Company (various issues) JECC Kompyûta No¯to (Computer Notes), Tokyo: Japan Electronic Computer Company. Japan Times (2003) ‘Telecoms to sue government over NTT’s fee hike’, Japan Times, 11 July. Jiji Press Ticker Service (2004) ‘NTT group’s interconnection fee revenue rises in FY “03”’, Jiji Press Ticker Service, 30 July. Jiji Tsûshin Kigyo¯ Nyûsu (2004) ‘Shindenden e no kashidashi gimu teppai o, kotei denwa no shinai tsûshin mo¯ NTT Higashi Nishi ikensho’ (NTT East and West call for scrapping mandatory leasing of NTT’s lines for local fixed service to new common carriers), Jiji Tsûshin Kigyo¯ Nyûsu, 27 August. JISA (Japan Information Technology Service Industry Association) (2001) Report on Software Export & Import Statistics: FY2000 Result, 2001, Tokyo: Japan Information Technology Service Industry Association. Kawabata, E. (2001) ‘Sanction power, jurisdiction, and economic policy-making: explaining contemporary telecommunications policy in Japan’, Governance 14 (4): 399–427. MIC (Ministry of Internal Affairs and Communications) (2003) Information and Communication in Japan, White Paper, Tokyo: Ministry of Internal Affairs and Communications, p. 10. Ministry of Public Management, Home Affairs, Posts and Telecommunications (2001) Recent Regulatory and Policy Development in Japan, presented to the APEC Telecommunications and Information Working Group. Tokyo: Ministry of Public Management. Nielsen-NetRatings Co. (2001) Quarterly Global Internet Trends Report (Q2 2001), August, accessed at www.nielsen-netratings.com on 24 March 2002. Nihon Keizai Shimbun (2004) ‘Skirmish over amending AMA rekindles’, Nihon Keizai Shimbun, 11 April: 5. Nikkei BP Asia (2002) ‘Nikkei Market Access’, Nikkei Electronic Asia, 8 April, accessed at http://neasia.nikkeibp.com. OECD (Organisation of Economic Co-operation and Development) (2000a) Internet Access Price Comparison 20 hour basket, accessed at www.oecd.org/dsti/sti/it/ cm/stats/isp-40hrs.htm on 17 October 2000. OECD (2000b) Internet Access Price Comparison 2000: 40 hour basket, accessed at www.oecd.org/dsti/sti/it/cm/stats/isp-40hrs.htm on 17 October. OECD (2003) Communications Outlook, Paris: OECD. OECD (2004) Benchmarking Broadband Prices in the OECD, Directorate for Science, Technology and Industry, Committee for Information, Computer and Communications Policy, Working Party on Telecommunication and Information Services Policies, Paris: OECD.
40
Mark Tilton and Hyeonjung Choi
OECD (2005) Communications Outlook, Paris: OECD. Point-topic.com (2003) ‘Japan: broadband overview’, 8 April, accessed at www.pointtopic.com. Point-topic.com (2005) ‘World DSL lines close in on 100m in 2004’, 10 March, accessed at www.point-topic.com/content/dslanalysis/Q404DSLana050310.htm. Reuters (2000) ‘Internet fever spurs PC sales in Japan’, 9 February, accessed at http://today.reuters.com/news. Rohlfs, J. H. and Sidak, J. G. (2002) ‘Exporting telecommunications regulation: the US–Japan negotiations on interconnection pricing’, Harvard International Law Journal, 43 (2): 317–60. Schoppa, L. J. (1997) Bargaining with Japan: what American pressure can and cannot do, New York: Columbia University Press. Scuka, D. (2000) ‘Unwired: Japan has the future in its pocket’, J@pan, Inc., accessed at www/japaninc.net on 25 June 2002. Seifu Kisei Nado to Kyo¯so¯ Seisaku ni Kansuru Kenkyûkai (2002) Denkitsûshin Bunya no Seido Kaikaku Oyobi Kyo¯so¯ Seisaku no Arikata (Competition Policy and Systemic Reform in the Field of Telecommunications), Tokyo: Seifu Kisei Nado to Kyo¯so¯ Seisaku ni Kansuru Kenkyûkai (Study Group on Government Regulations and Competition Policy). Shimoda, H. (1984) IBM to no 10-Nen Senso¯ (The Ten-Year War against IBM), Kyoto: PHP Kenkyûjo. Thatcher, M. (2002) ‘The relationship between national and European regulation of telecommunications’, in Jordana, Jacint (ed.) Governing Telecommunications and the New Information Society in Europe, Cheltenham and Northampton MA: Edward Elgar, pp. 66–85. Tilton, M. (1996) Restrained Trade: cartels in Japan’s basic materials industries, Ithaca NY: Cornell University Press. Tilton, M. (2003a) Interview with JFTC official, July. Tilton, M. (2003b) Interview with official at new common carrier firm, July. Tilton, M. (2003c) Interview with US State Department official in Tokyo, July. Vogel, S. K. (2000) Creating Competition in Japan’s Telecommunications Market, Japan Information Access Project. Working Paper. Washington DC: Japan Information Access Project. West, J. and Dedrick, J. (2000) ‘Innovation and control in standards architectures: the rise and falls of Japan’s PC-98’, Information Systems Research, 11 (2): 197–216. Yamamura, K. (2003) ‘Germany and Japan in a New Phase of Capitalism: confronting the past and the future’, in Yamamura, K. and Streeck, W. (eds) The End of Diversity? Prospects for German and Japanese Capitalism, Cornell Studies in Political Economy, Ithaca NY: Cornell University Press. Zaikai (2003) ‘KDDI nado shin denden go sha ga “NTT no setsuzoku ryo¯ o sagero” to so¯musho o teisho¯’ (KDDI and four other new common carriers sue MPHPT to get it to lower NTT’s interconnection rates), Zaikai, August: 66–70.
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
3
Institutional framework and competitiveness of the US telecommunications market Michael Schefczyk
Introduction While information and communications technology (ICT) has been characterized as having the potential to foster economic development, it has also been portrayed as having the ability to advance political freedom and democracy, human rights, a more acceptable distribution of wealth and social mobility.1 In order to better address the parameters of an institutional framework on the nation level that can support and encourage ICT development, the purpose of this paper is to: • • •
briefly put the US, Japanese and German telecommunications markets in perspective; provide an overview of the development of the regulatory framework in the United States with an emphasis on the Telecommunications Act of 1996; and analyse aggregate statistics to verify broad consequences of regulation on competition in the United States.
Thus, in the next three sections I address the three purposes listed above in a stepwise fashion, and in the final section I summarize a few conclusions that may be drawn from the analysis.
The telecommunications markets in the United States, Japan and Germany in perspective The telecommunications markets in the United States, Japan and Germany are characterized by quite different circumstances. One important factor is population size and density as a proxy for number of customers and customer density, which ideally should also reflect business customers. Operating in a large and densely populated country is advantageous for network operators as well as equipment manufacturers and content providers, as fixed costs (e.g. regulation, standardization, branding and corporate management) can be better distributed across a larger number of customers. Density, in
42
Michael Schefczyk
particular, can contribute to the economics of an access network, as larger switches and shorter lines, including shorter long-distance lines, can be employed. Figure 3.1 compares the United States, Japan and Germany with respect to such basic indicators. We see that the US market is two to three times larger than the telecommunications markets in Japan and Germany respectively. On the other hand, the population density in the United States is less than a tenth of the density in Japan and less than a seventh of the density prevailing in Germany. Thus, metropolitan areas provide a better basis for comparison when it comes to telecommunications market issues across countries. Across these three markets, the number of telephone subscriptions per 100 inhabitants ranges between 115 and 145, indicating that many residents hold multiple lines (e.g. ISDN in Germany). These figures are by far above the global average of 40 lines per 100 inhabitants. The picture for mobile telecommunications in these three countries is very similar, as Figure 3.2 indicates. Differences in market size are to some degree compensated by the number of subscribers per 100 inhabitants, which is lower in the larger US market at 54 than in Japan and Germany (68 and 79 respectively), cumulating in 158, 86 and 65 million subscribers respectively. The smaller markets experienced a later but also a substantially steeper period of growth. Between 1998 and 2003, average annual growth 338 292
231
127 83 31
USA
Japan
Germany
116
119
144
Population (millions)
Population Density (Inhabitants per km?)
Telephone Subscribers per 100 Inhabitants (World average: 40)
Figure 3.1 Basic indicators, USA, Japan and Germany, 2003 Source: ITU ‘Basic Indicators’.
Institutional framework and competitiveness, US 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
43
158
+18 % p.a. 86 +13 % p.a.
69
65
47
+36 % p.a. 14
1995
USA
2003
1995 Japan 2003
1995 Germany 2003
68
79
54
Cellular Cellular Mobile Mobile Subscribers Subscribers (millions) (millions) 1995
2003
Cellular Mobile Subscribers per 100 Inhabitants (World average: 22)
Figure 3.2 Mobile indicators, USA, Japan and Germany, 1995 and 2003 Source: ITU ‘Cellular Subscribers’.
rates ranged from 13 per cent in Japan to 18 per cent in the United States up to 36 per cent in Germany. Subscriber density indicates, however, that future growth potential remains the highest in the United States.2 The relatively slow development of the telecommunications market in the United States can mostly be attributed to the late introduction of digital cellular telephony and to the larger number of regional network operators, which limited the use of cellular telephones and maximized less customerfriendly roaming charges.3 The situation in Internet and personal computer penetration shown in Figure 3.3 differs significantly from fixed and mobile telephony. At 66 PCs per 100 inhabitants, the United States is far ahead of Japan and Germany, both at around 40 PCs per 100 inhabitants. Internet connections vary less, but the United States still leads with 55 Internet users per 100 inhabitants over 45 to 50 in Japan and Germany.4 In each market, only 10 to 15 per cent of residential Internet users employ a broadband connection instead of dial-up access up to ISDN speed. In terms of broadband Internet technology, access based on cable TV leads in the United States with a share of 57 per cent of all connections, while the telephone network based digital subscriber line is the leading technology in Japan (72 per cent) and the dominating technology in Germany (96 per cent). Slower development
44
Michael Schefczyk 66 55 47
45
43 38
7
6
4
USA
Japan
Germany
Cable (57 %)
DSL (72 %)
DSL (96 %)
Internet Users per 100 Inhabitants Narrowband Broadband PCs per 100 Inhabitants
Leading Broadband Technology
Figure 3.3 Internet indicators, USA, Japan and Germany, 2003 Source: ITU ‘Internet Indicators’, FCC.
in Germany is usually attributed to Deutsche Telekom’s grip on both (local) telephony and cable TV, at least until some of the cable TV infrastructure was divested in the late 1990s, as well as on its dominance in the wholesale broadband Internet access market.
Regulatory framework in the United States The development of telecommunications regulation in the United States can be divided into five phases, as illustrated in Figure 3.4.5 The first phase, between 1877 and 1956, was characterized by a shift from an initially unregulated situation towards regulation and then a movement back in the opposite direction to partial liberalization and deregulation. The award of the Bell patent on voice telephony in 1877 marked the beginning of this phase. Prior to the Bell patent, telegraphy established by Western Union was in a quasi-monopolistic situation. Voice telephony, promoted by a system originating with Bell Operating Companies (BOCs), became a competitor to telegraphy and particularly to Western Union, which itself was seeking to enter the telephone market. Substitution was managed via a settlement between Western Union and the Bell System. Subsequently, American Telegraph & Telephone (AT&T) was founded as a long-distance carrier
Institutional framework and competitiveness, US 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
45
in 1889 to connect regional BOCs. Its monopolistic structures were far reaching. For example, proprietary telephone equipment developed by Western Electric, a predecessor to Bell Labs and acquired from Western Union, grew to dominate the entire telephony market, including local telephony and customer premise equipment. Competition in local telephony had increased after the expiration of Bell’s patents in 1893–94. Shortly thereafter, however, beginning approximately in 1907, competition in local telephone markets was reduced due to the growing importance of longdistance connections provided by AT&T. As a consequence, many competitors, except for less attractive rural carriers, were absorbed by the Bell System. Excesses during the Great Depression led to regulation of the telephone industry from 1929 onwards. Public utilities commissions served as regulators of the telephony industry on a state level parallel to regulating energy and water providers. The US Department of Justice first aimed to reduce the power of AT&T via a lawsuit in 1949. Seven years later the litigation was settled based on the agreement that AT&T would not enter into the computer market but would merely continue to develop and manufacture telephone equipment. The second phase from 1956 to 1971 and designated as ‘toward managed competition’ led to the first step-wise deregulation of some of the segments of the telecommunications market. First, customer premise equipment was deregulated in two steps in 1956 and 1968 based on conformity to published standards. In the meantime, microwave operators were licensed from 1959 onwards, which initiated some competition in long-distance and regional transport. The second phase created the ground for further deregulation in the following phase. First, the data transmission market was gradually deregulated between 1971 and 1980, beginning with corporate networks and eventually allowing multiple competitors to provide public network access. Second, between 1971 and 1977, long-distance carriers were licensed to compete with AT&T on a call-by-call basis when customers dialled a prefix number, which eventually led to the founding of firms such as MCI and US Sprint. Pre-selection was introduced later in this process, and until the early 1990s, even large corporate customers were able to reach longdistance carriers only via regional BOCs. The third phase, between 1971 and 1982, can be characterized as ‘managed competition’, as some competition did exist in the long-distance market while the telephony market at large remained dominated by AT&T across all segments. Parallel to long-distance telephony, satellite telecommunications operators were licensed in 1971, inducing international consortia of publicsector operators, in particular, to pool their satellite operations. In 1973, value-added services were deregulated to allow, for example, third-party directory providers to begin operations. Resale and infrastructure sharing was legalized in 1976 to enable firms to buy larger volumes of telecommunications services in order to resell them in smaller volumes without becoming a network operator. Finally, AT&T long-distance and seven Regional Bell
• Data transmission market deregulated 1971 – 1980
• Long distance carriers licensed 1971 – 1977
• AT&T founded as a long distance carrier 1889
• Great Depression and regulation via public utilities commissions 1929
Figure 3.4 Five phases of US telecommunications regulation
Major deregulation in five areas
1996-…
• Separation of AT&T long distance and Regional Bell Operating Companies 1982
• Liberalization of telephony vs. cable TV separation 1992
• Competitive Access Providers licensed 1991
• Price cap regulation 1989
• Value-added services deregulated 1973 • Resale and infrastructure sharing legalized 1976
• Open network architecture liberalization 1986
1982 – 1996 Toward Local Competition
• Satellite telecommunications licensed 1971
1971 – 1982 Managed Competition
Bitte in dieser Abbildung wie folgt korrigieren:
Source:***
• Microwave operators licensed 1959
• Customer premise equipment monopoly deregulated 1956, 1968
1956 –1971 Toward ManagedCompetition
• System of Bell Operating Companies established against Western Union (Telegraphy)
• Bell Patent 1877
1877 – 1956 Before Liberalization
BC
TEL
CATV
EQV
Telephony Equipment Cable TV Broadcasting Internet
WWW
– – – – –
• Major deregulation in five areas:
1996 – … Post 1996 Telecommunications Act
Institutional framework and competitiveness, US 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
47
Operating Companies (RBOCs) were separated in a settlement in 1982 (the ‘modified final judgement’). This was the result of an antitrust lawsuit by the United States versus AT&T filed in 1974. The RBOCs were confined to local and regional telephone service within their territory. This landmark event ended the phase of managed competition. The fourth phase, between 1982 and 1996, can be called ‘toward local competition’. Local competition was enabled via the creation of the RBOCs as noted above. As a second step, open network architectures were liberalized in 1986 to lay the ground for interconnection among competing local carriers. While the previous ‘management’ of competition had been based on setting a maximum return that licensed operators were allowed to achieve, price cap regulation was introduced in 1989 to link the price of various bundles of telecommunications services to a predefined development of prices in order to enforce efficiency gains. Since RBOCs took a gatekeeper position between end customers and long-distance carriers, competitive access providers (CAPs) were licensed from 1991 onwards to provide links to long-distance carriers for business customers and in such a way were able to bypass traditional RBOCs. Competitive access was important and successful because the actual cost of providing access to a long-distance carrier was approximately 0.2 cents per minute, while regulated access charges averaged at about 2 cents per minute for both the origination and the termination of a call. In 1992, the requirement to keep telephony and cable TV separate was also relaxed, enabling telephone and cable TV network operators to branch out into each other’s respective market. The fifth phase began with the Telecommunications Act of 1996. The Telecommunications Act advanced deregulation in five areas – telephony, equipment, cable TV, broadcasting and the Internet – and led to the key rules described below.6 The Telecommunications Act of 1996 aimed to reduce barriers to entry and increase competition. To achieve these goals, the Act relied on non-discriminatory access via handing over traffic among interconnected networks and via access to unbundled parts of telecommunications networks on a long-term, incremental cost basis. In many ways, the Telecommunications Act aims to abolish cross-subsidies among different products and different types of customers. Nevertheless, ‘universal service’ obligations were imposed at least upon larger carriers to ensure broad availability of basic services.7 To meet universal service obligation, a transparent and competitively neutral system of cross-subsidization was created via universal service funds. Rules on telephone service Based on the Telecommunications Act of 1996, RBOCs were finally allowed to offer long-distance service outside their own region and – after meeting certain requirements – also within their own region. The requirements focus on permitting local competition and abstaining from
48
Michael Schefczyk
cross-subsidizing different services in order to provide a level playing field for competitors in the region. Furthermore, access and interconnection became mandatory for all local carriers far beyond the standards of the open network architecture liberalization enacted ten years before. Requirements were created to negotiate agreements for interconnection, resale, number portability, dialling parity, access to rights-of-way and reciprocal compensation. In addition, the incumbent local carrier is required to provide non-discriminatory, unbundled access as well as resale at ‘wholesale’ rates to competitors, and disclose certain information. Exceptions are allowed for small and rural carriers because the US government perceived a need to shield such carriers from competition. Universal service obligations are maintained to guarantee availability of standard services at set prices in rural and high-cost areas. Within the universal service regime, reduced prices for low-income subscribers, schools, libraries and rural health-care providers, as well as infrastructure sharing among certain carriers are required. Eligible carriers fulfilling universal service obligations are compensated via a universal service fund. This fund effectively cross-subsidizes customers benefiting from universal service by drawing on mainstream customers. Rules on telecommunications equipment RBOCs were allowed to manufacture telephone equipment once they met the prerequisite of providing out-of-region, long-distance service provision, i.e. once they enabled an effective framework for competition. Crosssubsidization with telephone operations is prohibited and monitored by the Federal Communications Commission (FCC). As a measure of preventing cross-subsidization and breaking other non-discrimination safeguards, equipment manufacturing needed to be performed in ‘separate affiliates’ for three years. Furthermore, technical information and standards for customer premise had to be opened to all competitors as well as the research and development community. Bell Core, the traditional developer of AT&T telecommunications equipment, was prohibited from manufacturing equipment. Similar ‘separate affiliate’ requirements were introduced to enforce restrictions against joint marketing of local telephone and electronic publishing services. Rules on cable TV Rules capping cable TV rates were removed in a stepwise process. In 1996, price caps were removed for small cable operators in the case of effective competition by a telephone company that had entered the cable TV market. By 1999, remaining price caps were further reduced to the ‘basic tier’ of cable programming.
Institutional framework and competitiveness, US 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
49
In addition to competition across market segments traditionally kept separate, the possibility of cross-ownership of telephone and cable TV network operation was also further liberalized. A telephone company may now own more than 10 per cent of the shares of a cable TV company – or vice versa – under the following circumstances: • •
in cable TV markets of up to 35,000 inhabitants if the cable TV company serves less than 10 per cent of the households in the telephone company’s overall service area; and in cable TV markets of more than 35,000 inhabitants but one that is not among the top 25 markets nationwide served by multiple cable TV companies. The latter are subject to further restrictions.
The aim of this regulation is to prevent a telephone operator from acquiring a cable TV operator predominantly for the purpose of reducing competition. To compete directly with cable TV operators, telephone operators were permitted to run ‘open video systems’ where at least two-thirds of the content is provided by unaffiliated programmers. This is to guarantee that no single owner can control the entire value chain from programming down to distribution, which includes both cable TV and telephony. Furthermore, cable TV set top boxes were unbundled from cable TV operations, both in terms of technology used as well as in terms of retail availability. This rule was created to reduce switching costs for consumers and thus to increase competition. Rules on radio and TV broadcasting As far as the media was concerned, concentration limits were relaxed but not eliminated. Basically, any one company may own TV stations that reach up to 35 per cent of the national TV audience. This figure is up from a limit of 25 per cent prior to the Telecommunications Act. TV broadcast networks were allowed to own cable TV systems, but no network is allowed to acquire another network. Broadcasters were also allowed to temporarily use new spectrum for advanced TV services (such as High Definition TV) on the condition that these frequencies are either purchased or returned once the transition to digital TV broadcasting is completed. All nationwide limits on radio station ownership were repealed, and local limitations on radio station concentration were amended. Rules on Internet, online computer services and miscellaneous regulations Rules on Internet and online services in the United States focus on the prevention of use deemed to be inappropriate. In particular, unlawful use
50
Michael Schefczyk
of a computer or any other telecommunications device to transmit obscene material with the intention to annoy, abuse, threaten or harass another person or to transmit indecent material to a minor is prohibited. Such regulation is directed against the user or the sender of the information, and not against the network operator unless the operator knowingly or intentionally permits a facility under its control to be used for prohibited purposes. RBOCs (except for Ameritech) were prohibited from entering the alarm monitoring business until 2001, as this was felt to create an uneven playing field. As public pay telephones are frequently used for long-distance calls, calling card access to long-distance carriers provided no compensation to operators of pay telephones. Thus FCC was required to develop a plan to compensate pay telephone owners for long-distance calls placed. Regulation of direct broadcast satellite services was concentrated at the FCC. Furthermore, the FCC was authorized to relax tariff and rate regulation on competitive carriers who have little market power.
Development of competition in the United States When we consider the fact that telecommunications network access is still fixed, we find a remarkable level of competition after the passage of the Telecommunications Act. Competition was largely enabled by the effective regulation of access to unbundled network elements. Figure 3.5 illustrates the development between 1999 and 2002. The overall share of fixed-access lines provided by competitors (i.e. not by incumbent operators such as RBOCs or traditional rural or small % of lines offered by Alternative carriers
13.2
10.3 9.8
7.7
7.5
5.0 4.3 2.8
2.1
26 % Competitor’s own infrastructure • declining due to inefficiencies if infrastructure already exits
2.2
2.5
1.8
55 % Offers based on unbundled access • increasing due to efficient regulatory regime • relatively attractive for competitor as well as incumbent
19 % Resale • declining due to lack of control by competitive carrier
1999
2000
2001
2002
41 %
45 %
48 %
58 %
Residential and small business customers
Figure 3.5 Fixed-access competition, US Source: FCC ‘Trends in Telephone Services’.
Institutional framework and competitiveness, US 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
51
operators), grew from 4 per cent in 1999 to 13 per cent by the end of 2002. While approximately 19 per cent of such access is provided by resale, the overall percentage of lines attributable to resellers remained relatively low at 2 to 3 per cent. This is largely due to the fact that competing carriers lack control over the lines they are reselling. Some 26 per cent of competitive access is provided based on the competitors’ own infrastructure. While this segment grew, it did not grow proportionately, since inefficiencies generally result from the fact that already existing infrastructure is no longer being used. The largest segment and also the segment experiencing the strongest growth rates is the one that offers services based on unbundled access, such as the leasing of existing access lines from incumbent carriers. Since its regulatory regime is quite efficient and since it is relatively attractive for the competitor as well as for the incumbent, 55 per cent of the competition takes place in this manner. The competitor retains relatively strong control over customer relationships while at the same time minimizing investments. The incumbent can at least partly avoid idle infrastructure and can – depending on the price level permissible under the regulatory regime – collect relatively attractive revenues via infrastructure leasing contracts with competitors. The situation in local telephony is shown in Figure 3.6. The local exchange market has been declining since 1988 due to price cap regulation, which puts pressure on achieving efficiency gains and causing them to at least partly be passed on to customers. Traditional ‘rate of return’ regulations in monopolistic times enabled carriers to maintain inefficiencies and overheads since, to the extent these were accepted by regulators, they would tend to harm customers but not shareholders. One of the major consequences of the 1996 Telecommunications Act was to increase competitive pressure on RBOCs, and even more so on other Revenues in billion US$ 16 14 12
Competitive Local Exchange Carriers 1984 0% 1996 0% 2001 33 %
10 8
Other incumbent carriers 1984 27 % 1996 29 % 2001 8%
6 4 2
1984
1986
1988
1990
1992
1994
1996
1998
2000 2001
Figure 3.6 Market share of local exchange carriers, US Source: FCC ‘Trends in Telephone Service’.
Regional Bell Operating Companies 1984 73 % 1996 71 % 2001 59 %
52
Michael Schefczyk
incumbent carriers, such as rural and small operators. These other carriers have been necessary in order to ensure telecommunications access to otherwise underserved areas but they have nevertheless been frequently less customer friendly because they have been able to demand high prices. While the Telecommunications Act of 1996 included a few measures to shield these carriers from full competition (e.g. via exceptions on interconnection, resale and number portability requirements), innovative competitors frequently took advantage of opportunities to enter even these rural market segments. By 2001, new, competitive local exchange carriers had secured one-third of the local exchange market. Initially focusing on the business customer market similar to the strategy employed by the archetypal competitive access providers, they began to turn towards smaller customers over time. As the percentage figures at the bottom of Figure 3.5 show, the fraction of lines supplied by these new carriers to residential and small business customers grew from 41 per cent in 1999 to 58 per cent in 2002. Moreover, research has demonstrated that reasonably low interconnection prices offered by incumbent local exchange carriers (ILEC) are usually beneficial for both the competitor and the incumbent. In particular, relatively low interconnection prices contributed to an increase in fibre-optic and digital network deployment.8 The picture differs substantially in the long-distance telephony market, as shown in Figure 3.7. While AT&T’s market share declined from 90 per cent in 1984, two years after the break-up of AT&T and the RBOCs, to 37 per cent in 2001, AT&T revenues from long-distance services remained constant for the most part. Traditional competitors such as MCI and US Sprint increased their sales and market share until they reached a peak in 1999, thereafter experiencing a slight decline along with AT&T. The winners Revenues in billion US$ 100 Other Carriers 1984 3% 1996 17 % 2001 30 %
80
US Sprint 1984 3% 1996 10 % 2001 9%
60
40
MCI 1984 1996 2001
4% 25 % 24 %
AT&T 1984 1996 2001
90 % 48 % 37 %
20
1984
1986
1988
1990
1992
1994
1996
1998
Figure 3.7 Market share of long-distance carriers, US Source: FCC ‘Trends in Telephone Service’.
2000 2001
Institutional framework and competitiveness, US 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
53
in the long-distance market from 1996 onwards were other carriers. Their collective share grew to 30 per cent, and while they also experienced decline, their decline in market share came in 2001, i.e. slightly later than that of AT&T, MCI and US Sprint. Due to the Telecommunications Act, the composition of the ‘other carriers’ segment shifted from a large number of smaller providers towards: (1) a group of infrastructure-based new competitors focusing in the business customer market, e.g. Qwest, Level 3 and Williams; (2) RBOCs offering long-distance services to customers in their own region as well as elsewhere; and (3) a large number of generally smaller resellers. Overall, deregulation has produced remarkable results since 1996, as up to one-third of the mainstream telephony service market is occupied by providers that were not active in the respective segment in 1996. This can be taken as an indication that choices have increased from a customer’s perspective, and that quality and prices will have improved accordingly. The fact that in 2001 only slightly more than 10 per cent of fixed access lines were operated by major competitors while competitive local exchange carriers collected 33 per cent of the revenue further indicates that competition for the best customers must have been particularly strong. The situation in high-speed Internet access is shown in Figure 3.8. The number of high-speed lines grew tremendously from 1.4 million in 1999 to 11.4 million in 2002. Lines with >200 kbit/s in at least one direction are classified as high-speed, while >200 kbit/s in both directions classifies a line as advanced. Unlike the situation in many other countries, the largest proportion of high-speed Internet access lines is implemented via cable TV networks. Cable TV-based Internet access offers a strong technical basis (73 per cent of lines are advanced), but is limited to residential and small business customers. Approximately 4.5 million lines were implemented along Lines (millions)
Advanced2)
11.4 Cable
73 %
100 %
DSL
43 %
85 %
Other3)
90 %
24 %
7.1 6.5
3.6 2.0
3.9
1.8
2.0
1.4 1.0 0.4 1999
1.5
2000
2001
2002
1) > 200 kbit/s in at least one direction 2) > 200 kbit/s in both directions
Figure 3.8 High-speed Internet, US Source: FCC ‘Trends in Telephone Service’.
3) traditional wireline (60 %), fiber (27 %), satellite/fixed wireless (13 %)
54
Michael Schefczyk
the telephone network via the digital subscriber line (DSL) technology. The number of advanced lines are fewer but business customers do receive useful capabilities. Other technologies were used for two million high-speed lines, including traditional wireline leased lines, leased fibre lines, and satellite and fixed wireless access.9 Most of these lines qualify as advanced and the majority are implemented for business customers, which also reflects the higher costs associated with these technologies. In addition to increasing competition, then, we may say that deregulation after 1996 also fostered some innovation, albeit in a different direction from that taken in Japan and Germany, where high-speed residential Internet access tends to be based on the DSL technology.
Findings and conclusions The facts considered in this chapter support four conclusions. First, the US, Japanese and German telecommunications markets may be distinguished by: • • •
population density, with the US density a seventh and a tenth of Germany’s and Japan’s population density respectively; cellular penetration and digitization levels, with the United States at 30 per cent lower in penetration and 10 per cent lower in digitization; and Internet and PC penetration, with the United States 20 to 50 per cent higher in Internet and PC penetration than in Germany or Japan.
Second, until 1996, the regulatory framework in the US was characterized by a focus on: • • •
reducing barriers to market entry; keeping operators from serving multiple segments simultaneously, e.g. local and long-distance telephone; and using price cap regulation for incumbent carriers following traditional cost and rate-of-return-based regulation.
Third, in 1996 the regulatory framework in the United States was restructured to: • • •
allow operators to serve multiple segments simultaneously, e.g. local and long-distance telephony, customer premise equipment, cable TV and content; enforce strong access and interconnection rules; and move towards intervention regulation of prices and universal service obligations.
Fourth, results in terms of competition, price/productivity improvement and performance/quality have been strong:
Institutional framework and competitiveness, US 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
• • •
55
in long-distance and local telephony; in high-speed Internet access; and in mobile telephony, but less obviously so.
Beyond conclusions based on descriptive statistics, a couple of key economic arguments should be discussed briefly. One of the most interesting debates surrounds the reappearance of cost-based regulation for unbundled access and interconnection after the cost and rate-of-return-based regulation of end customer prices was abolished in favour of price cap regulation. Cost and rate-of-return-based regulation was seen to lead to high investment and operating expenditures and thus ultimately to high end customer prices. According to some observers, however, the new regulation based on longrun, incremental costs for services provided among competing network operators is seen to cement high cost and price levels, while others maintain that it provides free options to competitors and leads to underinvestment.10 I note, however, that the analysis provided here takes a mid- to longterm perspective with a focus on telecommunications. From this viewpoint, innovation and deregulation are likely to appear in a more positive light than from a viewpoint that focuses on a broader perspective of ICT infrastructure in general or from the more common perspective circulating at the moment that focuses on short-term comparisons across nations.11 One approach from where fruitful research contributions can be is comparative sector-specific analyses that focus more on selected segments of customers across countries. However, this type of approach requires more detailed data than those currently provided by the FCC and similar national entities.
Notes 1 See Sein and Harindranath (2004: 19–20) for a more detailed conceptual framework. 2 See Banerjee and Ros (2004: 128–31) for the situation in different types of countries. Low and middle income countries are not covered here. The United States, Japan and Germany fall in a cluster with high telephone and cellular density but relatively low growth rates in fixed and mobile telephony. Growth rates are found to be higher in lower income countries. 3 For an analysis of the impact of the European GSM standard on the development of mobile telecommunications, see Mina (2003: 444–50). 4 See Chapter 2 by Tilton and Choi, and Chapter 5 by Storz in this volume. See also Davis (2003: 296–8) for an analysis of computer and Internet penetration and its association to websites, electronic purchases and electronic sales in Canada. 5 For a more detailed analysis, see Sarkar (2001: 39–114) and Huurdeman (2003: 85–600). For an overview of the regulatory framework in Germany, see Knieps (2003: 383–97). 6 See for example Pitz (1999: 45–80), Laffont and Tirole (2000: 18–29) and Economides (2003: 196–204). 7 See Schement and Forbes (2003: 234–47) for details on universal service.
56
Michael Schefczyk
8 For a statistical analysis on the relationship between prices on the one hand and fibre-optic and digital network development on the other, see Chang et al. (2003: 679–88). 9 For a comparison of broadband access technologies with an emphasis on powerline see Tongia (2004: 569–76). 10 See Hausman (2001: 64) for one position. 11 See Chapter 6 by Frieden in this volume.
References Banerjee, A. and Ros, A. J. (2004) ‘Patterns in global fixed and mobile telecommunications development: a cluster analysis’, Telecommunications Policy, 28: 107–32. Chang, H., Koski, H. and Majumdar, S. K. (2003) ‘Regulation and investment behaviour in the telecommunications sector: policies and patterns in US and Europe’, Telecommunications Policy, 27: 677–99. Davis, T. (2003) ‘E-commerce measures and analysis’, Statistical Journal of the United Nations, 20: 289–301. Economides, N. (2003) ‘US Telecommunications Today’, in Brown, C. and Topi, H. (eds) I.S. Management Handbook, Boca Raton: Auerbach, pp. 191–212. Hausman, J. A. (2001) ‘Regulation by TSLRIC: economic effects on investment and innovation’, in Sidak, J. G., Engel, C. and Knieps, G. (eds) Competition and Regulation in Telecommunications: examining Germany and America. Boston MA: Kluwer, pp. 51–67. Huurdeman, A. A. (2003) The Worldwide History of Telecommunications, Hoboken: Wiley. Knieps, G. (2003) ‘Sector-specific regulation of German telecommunications’, in Madden, G. (ed.) World Telecommunications Markets: the international handbook of telecommunications economics, volume III, Cheltenham: Elgar, pp. 383–99. Laffont, J.-J. and Tirole, J. (2000) Competition in Telecommunications, Cambridge MA: MIT. Mina, A. (2003) ‘The creation of the European market for mobile telephony: overview of an instituted process’, International Review of Sociology, 13: 435–54. Pitz, D. (1999) Wettbewerb auf dem US-amerikanischen Telekommunikationsmarkt: Anbieterstrategien und Regulierungsphilosophie nach dem Telecommunications Act, Göttingen: Vandenhoeck & Ruprecht. Sarkar, R. S. (2001) Akteure, Interessen und Technologien in der Telekommunikation: USA und Deutschland im Vergleich, Frankfurt am Main: Campus. Schement, J. R. and Forbes, S. C. (2003) ‘Universal service in the information age’, in Madden, G. (ed.) Emerging Telecommunications Networks: the international handbook of telecommunications economics, volume II, Cheltenham: Elgar, 234–49. Sein, M. K. and Harindranath, G. (2004) ‘Conceptualizing the ICT artefact: toward understanding the role of ICT in national development’, Information Society, 20: 15–24. Tongia, R. (2004) ‘Can broadband over powerline carrier (PLC) compete? A technoeconomic analysis’, Telecommunications Policy, 28: 559–78.
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
4
Information and communication technologies in Germany Is there a remaining role for sector-specific regulations? Günter Knieps
Introduction The role of government interventions and regulations has different traditions in the media, information technologies (IT) and telecommunications sectors. The media industry, for example, has traditionally been viewed as a bearer of social, cultural and ethical values within society. As such, the media has historically been subjected to federal regulation to some extent, e.g. public broadcasting. Federal intervention in private communication in most western, industrialized countries, on the other hand, has been largely absent, as it has generally been in the computer and IT industry as well. In Europe, the computer and IT industry has been allowed to develop in an unregulated manner for the most part under the general competition law. Moreover, despite the fact that the telecommunications sector is fully liberalized in Germany – as well as in all other European countries – it is still characterized by a complex set of sector-specific regulations today. These various degrees of government intervention may be challenged in the future, however, due to the convergence of the telecommunications, media and IT sectors.1 The increasing convergence of these three sectors gives rise to two different scenarios: (1) convergence may outpace existing sector-specific regimes; and (2) sector-specific regulation may be expanded to cover presently unregulated markets, such as mobile telephones and Internet services. Convergence raises the question as to how to achieve the proper balance of government intervention within a comprehensive, institutional framework that will allow markets as much freedom as possible. The same question can also be applied to a related topic. Recent developments towards deregulation and subsequent vertical disintegration of telecommunication networks have illuminated a number of problems associated with network access and network interconnection. In this context, the question revolves around the extent to which interconnection and access problems can be solved voluntarily by private market contracts between the parties involved, or whether government intervention, such as
58
Günter Knieps
harmonization and integration policies set forth by the European Union (EU), becomes necessary. The basic idea behind a disaggregated approach to network regulation put forth in this chapter is to identify those parts of networks where market power remains and may be abused in the interconnection process. This chapter outlines the increasing vertical and horizontal interconnection problems within information and communication technologies (ICT) networks and the corresponding role of government intervention in this sector in a conceptual way. Several authors have raised questions along these lines with regard to the Internet. Mestmäcker (2001), for example, asks whether content will still be subject to regulation in the future considering the enormous scope of content production and distribution in the converging markets. Fisher (2000) and others wonder whether there is still a serious applications barrier to entry in the microprocessor market, given the enormous potential for middleware threats due to innovations on the browser market (see also Economides 2000; Sidak 2001). Kesan and Shah (2001) make an attempt to identify the potential and the limits to selfregulation in the organization of access to Internet Protocol (IP) number assignments and domain name systems (see also Hillebrand and Büllingen 2001), while Müller and Rannenberg (1999) question the safety of the Internet, and the enforcement of property rights within its domain has been considered, for example, by Engel (1999). These highly relevant and pertinent questions, however, will not be addressed here. Instead, the focus of this chapter is on those elements of the Internet periphery and Internet service provision heavily based on telecommunications, in particular Internet access and Internet backbone. Access to the Internet requires a connection between the Internet user and the interface to the Internet service provider (ISP). In the past, the local loops of the established telecommunications carriers have been considered as areas with network-specific market power and thus have been subject to sector-specific regulation. Gradual phasing out of this sector-specific regulation is now under debate due to increasing access alternatives. By contrast, transit and peering arrangements among Internet backbone providers (IBPs) are not subject to sector-specific regulation. The agreements that cover interconnection between IBPs are characterized by private negotiations and fall under non-disclosure rules. According to the economic theory of regulation, there is no need for sector-specific regulation due to the absence of network-specific market power. The input market of communications bandwidth is competitive, and each IBP can develop its own logistic concept to optimize its own backbone and set of transit and peering arrangements. This chapter is organized as follows. The first section deals with the role of telecommunications for the Internet, differentiating between local network access and long-distance network capacity. The next section discusses the new regulatory arrangements for communications services within Europe,
Information and Communication Technologies, Germany
with a particular emphasis on Germany. In order to analyse the future role of sector-specific regulation from a normative point of view, the following section introduces the network economic concept of a disaggregated regulatory approach. Subsequent sections then turn to the phasing-out potential for sector-specific regulation due to increasing competition within the local loop and to considering the role of ‘technology-neutral’ regulation – a term that implies that in an environment of competing network infrastructures, sector-specific regulation should not be expanded but rather removed completely. The final section explains the role of competition in the markets for backbone interconnections (interconnectivity).
The role of telecommunications for the Internet The main elements of Internet service provision are an inherent part of the Internet and would not exist without the Internet. However, Internet service provision also requires several complementary elements, designated here as the Internet periphery, which are viable on their own, even in the absence of the Internet (see Figure 4.1). ISPs offer their customers a wide spectrum of different services by combining both peripheral and main elements (Elixmann and Metzler 2001: 14–17). O’Donnell (2000: 17–21) has classified these services as fundamental networking and inter-networking, application services and customer relations.
Terminal Equipment
Content
ty
Internet Service Provision
B In ack te b rc on on e ne ct iv i
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
59
Local Network Access
Long Distance Network-Capacity
Figure 4.1 Internet periphery versus Internet service provision Source: Knieps 2003: 219.
60
Günter Knieps
Terminal equipment (e.g. PCs and cellular phones) can be used either with or without access to the Internet, although obviously the use of the Internet is not possible without any terminal equipment. Content (including broadband applications) may be provided via the Internet (e.g. video on demand, customized music and video libraries), but it is also available through other distribution channels, such as cinemas, traditional video libraries and traditional broadcasting. Internet service provision would be possible without any sort of content provision. For example, ISPs could specialize in interactive services such as e-mail. In order to provide Internet services, capacity of long-distance telecommunications networks (bandwidth) is required. Today, for the most part, investments in long-distance telecommunications infrastructure are strongly motivated by Internet demand, despite the fact that telecommunication transmission capacity has many different purposes.
New regulatory arrangements for communication services The basic goal of the 1999 Review of the European Commission (European Commission 1998) was to consider the extent to which phasing out sectorspecific market power regulation should take place. The key objectives stated at the beginning of the review process were: (1) to maximize the application of the general European competition law; (2) to minimize sectorspecific regulation; and (3) to rigorously phase out unnecessary regulation (European Commission 1998: 3). On 12 July 2000, the European Commission presented its ‘1999 Review Package’, which included five proposals for Directives of the European Parliament and the Council, and one proposal for a Regulation. Since then, the following directives have been enacted: an ONP Framework Directive,2 an Access and Interconnection Directive,3 a Licensing Directive,4 a Universal Service Directive5 and a Personal Data/Protection of Privacy Directive.6 In addition, the proposal for the regulation of unbundled access to the local loop was passed by the European Parliament and the Council, and enacted in January 2001.7 This regulation marks a first for the EU, as the legal instrument of a regulation has never been used before in telecommunications policy. In contrast to a Directive, which requires national legislative enactment measures, a Regulation is the most powerful legislative tool made available by the EC Treaty. It is binding in its entirety and directly applicable in all Member States, meaning that it automatically becomes law in all Member States as soon as it is enacted.8 According to this regulation, the incumbent operator with significant market power is obliged to provide full, unbundled access, as well as shared access, to the copper local loop under transparent, fair and non-discriminatory conditions. The implementation of price regulation is left to the national
Information and Communication Technologies, Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
61
regulatory authorities. As long as the level of competition for local access is insufficient to prevent excessive pricing, national regulatory authorities are required to ensure that the principle of cost orientation is applied. Both the Framework Directive and the Access Directive provide no clearcut definition of future sector-specific regulation. The Framework Directive provides a new interpretation of the criterion ‘considerable market power’, moving in the direction of establishing the criterion of dominance in a given market as a prerequisite for sector-specific market power regulation. It gives the Commission discretionary power to identify a variety of markets for which the introduction of sector-specific regulatory measures should at least be considered. The Access Directive already indicates that sector-specific regulation may be extended to competitive markets (e.g. mobile telephony) as well as newly developing innovative markets (e.g. the Internet). This would be a definite step backwards from the Access Notice of August 1998, which extended the role of competition policy by emphasizing the importance of ensuring non-discriminatory access to essential facilities. In Germany, the telecommunications law of 25 July 1996 has allowed global market entry since January 1998.9 The EU review process spurred on a review of the national communications law in Germany, which was finally revised in June 2004.10 The question thus arises as to whether additional sector-specific regulations now in place will be removed, or whether there is still a role for sector-specific regulations.
The disaggregated regulatory approach Criteria such as relative market share, financial strength and access to input and service markets can only serve as a starting point for evaluating the existence of market power. ‘Criteria for conjecturing a dominant position’ (‘Vermutungskriterien’) on the basis of market shares, for example, can lead to economically unjustified criteria for government intervention in network industries. From a competition economics point of view, the use of ex ante, sector-specific regulatory intervention constitutes massive interference with the market process and thus always requires a particularly well-founded justification based on modern network economics.11 Obviously, the development of an ex ante regulatory criterion creates a need for a more clear-cut definition of market power. Stigler’s concept of entry barriers, which focuses on long-term cost asymmetries between incumbent and potential entrants is essential here to identify a proper regulatory basis: A barrier to entry may be defined as a cost of producing (at some or every rate of output) which must be borne by a firm which seeks to enter an industry but is not borne by firms already in the industry. (Stigler 1968: 67)
62
Günter Knieps
The sector-specific characteristics of network structures (economies of bundling) do not constitute a sufficient basis on which to conclude that market power does actually exist. It is necessary to differentiate between those areas in which active and potential competition can work and other areas, so-called monopolistic bottleneck areas, where a natural monopoly situation (due to economies of bundling) in combination with irreversible costs exists. It can be demonstrated that the regulation of network-specific market power is only justified in monopolistic bottleneck areas. In all other cases, the existence of active and potential competition will lead to efficient market results as in the other sectors of an economy. The pressure of potential competition can be sufficient to discipline the behaviour of the active supplier, even in those cases where the supplier is the owner of a natural monopoly. Such networks are called ‘contestable’ (see, for example, Baumol et al. 1982). An essential condition for the functioning of potential competition in order to discipline a natural monopoly already providing network services is that the incumbent firm does not have asymmetric cost advantages compared to potential entrants. If sunk costs are relevant, consumers who would intrinsically be willing to switch immediately to less costly firms are unable to do so. Sunk costs are no longer relevant for the incumbent monopoly, whereas the potential entrant is confronted with the decision of whether to build network infrastructure and thus spend the irreversible costs or not. Thus, the incumbent firm has lower decision relevant costs than potential entrants, and this creates an outlet for strategic behaviour on the part of the incumbent firm to the extent that monopoly profits (or inefficient production) will not necessarily result in market entry (see Knieps and Vogelsang 1982). It follows, then, that sector-specific, ex ante regulatory intervention in order to discipline market power can only be justified in monopolistic bottleneck areas, i.e. where a natural monopoly in combination with irreversible costs is present (see Knieps 1997a; 1997b). This basic concept of disaggregated identification of network-specific market power is illustrated in Table 4.1. Table 4.1 The localization of monopolistic bottleneck facilities Network area
With sunk costs
Without sunk costs
Natural monopoly (bundling advantages)
(1) Monopolistic bottlenecks
(2) Potential competition (contestable networks)
No natural monopoly (bundling advantages exhausted)
(3) Competition among active providers
(4) Competition among active providers
Source: Knieps 2000: 96.
Information and Communication Technologies, Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
63
The network economic concept of monopolistic bottlenecks suggests a connection with the essential facilities doctrine resulting from US antitrust law, which is now also being used increasingly in European competition law (see Knieps 2000: 104).12 In accordance with this doctrine, a facility can only be regarded as essential if the following two conditions are fulfilled: (1) entry into the complementary market is not actually possible without access to this facility; and (2) providers on the complementary market cannot, using reasonable effort, duplicate the facility. There are no substitutes either (see Areeda and Hovenkamp 1988).13 The application of the essential facilities doctrine means that a traditional instrument of competition law can be used as a regulatory instrument. A facility is regarded as essential when it fulfils the criteria for classification as a monopolistic bottleneck facility in the context of the disaggregated regulatory approach. The starting point for this approach is to differentiate between those network areas in which active and/or potential competition is possible, and those network areas in which stable, network-specific market power can be localized. The disaggregated regulatory approach involves applying the essential facilities doctrine in a dynamic way. This means that it should be applied not only on a case-by-case basis, but also to a category of cases, namely to monopolistic bottleneck facilities. Moreover, non-discriminatory conditions of access to the essential facilities must also be set out in more detail. The aim is to design the conditions of access so as not to hinder infrastructure competition but rather to create an incentive for research and development as well as innovations and investments at the facility level. This is the only way to establish a balanced relationship between services and infrastructure competition.
Localization of monopolistic bottlenecks within telecommunications networks Competitive long-distance networks Although some markets for long-distance telecommunications services may be characterized by economies of scale as well as bundling advantages, there is nevertheless competition. Inefficient suppliers are replaced by less expensive ones because there is free market entry. Even when the incumbent’s market share is high, inefficient production or services not geared to market requirements will soon lead to a considerable loss in market share, as customers are not tied to a specific supplier and can react without delay to price cuts on the market. There is thus no regulatory need for disciplining the market power of alternative network providers. Since overall free entry has become possible, the performance of the German long-distance telecommunications market has markedly improved. Germany exhibits a large number of service providers who have increased their scope
64
Günter Knieps
of services, and the country has seen the entry of several network carriers, which has notably lowered prices for all services (Stumpf and SchwarzSchilling 1999; Gabelmann and Gross 2003: 113; Knieps 2006: 205). The market for long-distance transmission capacity is also very competitive (Laffont and Tirole 2000: 98). A large number of newcomers have built transnational network infrastructure as input for Internet backbone capacity (Elixmann 2000: 7), which represents another possibility for leasing transmission capacity from several alternative network infrastructure providers. Moreover, a larger number of carriers possess their own fibreoptic networks (Immenga et al. 2001: 14, Table 1). Presently, the telecommunications transport capacity is readily available from a variety of providers (Kende 2000: 25). The remaining regulatory problem in the local loop It has been traditionally assumed that local networks constitute monopolistic bottlenecks for which neither active nor potential substitutes are available (see Table 4.2). The EU regulation on unbundled access to the local loop proceeded from this assumption and concluded that there is still a need to regulate an incumbent operator’s local access network. To the extent that local networks constitute monopolistic bottlenecks, ex ante regulation appears justified. Non-discriminatory access to essential facilities has to be guaranteed (see Knieps 1997a: 328). Since unregulated tariffs would enable owners of monopolistic bottlenecks to generate excessive profits, the instrument of price-cap regulation should be introduced (see Beesley and Littlechild 1989). Its major purpose is to regulate the level of prices, and it takes into account the inflation rate (consumer price index) minus a percentage for expected productivity increase. It is important to restrict such price-cap regulation to those areas of telecommunications networks where market power due to monopolistic bottlenecks is a regulatory problem. In all other subparts of telecommunications networks, price setting should be left to competitive market forces. Table 4.2 Local telecommunications networks as monopolistic bottleneck facilities
Terminal equipment Telecommunications services (including voice telephone services) Satellite/mobile networks Long-distance, cable-based networks Local cable-based networks Source: Knieps 2000: 99.
Natural monopoly (economies of bundling)
Irreversible costs
–
–
X X – X
– – X X
Information and Communication Technologies, Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
65
Concentrating on regulating the local loop does indeed constitute the one remaining task of tailored, sector-specific market power regulation. Nondiscriminatory access to this bottleneck facility must be guaranteed for all competitors. The EU regulation on unbundled access to the local loop contains an obligation for full unbundling as well as line sharing. In order to guarantee competition in long-distance telecommunications markets, global access to local networks already seems to be sufficient (Engel and Knieps 1998). In any case, one variant of non-discriminatory access to the local loop should be considered sufficient to overcome the monopolistic bottleneck problem. Increasing competition within the local loop in Germany Local network competition started with business customers in urban centres where the preferred access technology was optical fibre (Distelkamp 1999: 94). However, after licences for point-to-multipoint microwave systems were granted by the state, the wireless local loop has also been gaining in importance in Germany (Regulation Office for Post and Telecommunication 1999: 24). Platform competition, where alternative ISPs have complete control of all aspects of their networks and their corresponding services, has effectively claimed a stake in the market. Consequently, ever since the comprehensive opening of the telecommunications market, the pressure of innovation has increased in local networks as well. The competition has led to considerable variety in technological platforms, e.g. optical fibre, wireless networks, Community Antenna Television (CATV) networks and satellite technology, as well as to an increase in product variety. In addition, because of these rapid developments, the local loop facilities in larger cities and agglomerations in Germany are increasingly losing their status as monopolistic bottlenecks. Although it is not possible at this point to predict exactly how long it will take for the monopolistic bottlenecks in the local loop to disappear completely, it is clear that the building of access networks parallel to monopolistic bottlenecks, which was until only recently considered a futuristic dream, is now in full swing. The development of alternative access networks indicates that the potential for phasing out sectorspecific regulation in telecommunications should be fully exhausted, particularly by not extending the essential facilities doctrine to facilities that have yet to be created, as this would tend to inhibit the necessary investments in new technologies in the first place.
Broadband versus narrowband access to the Internet: is there still a role for sector-specific regulation? Access to the Internet requires a connection between the Internet user and the interface to the Internet, which in turn requires access to a local telecommunications network. In addition, a (long-distance) link between the originating (local) network and the ISP is required.14
66
Günter Knieps
Several access technologies exist: copper, fibre optics, two-way cable TV infrastructure (CATV network), power line communication and radio in the loop. One may also differentiate between narrowband and broadband Internet access. Narrowband Internet access takes place on two-paired, copper cables via analogue modem and ISDN (integrated services digital network). Broadband Internet access can be provided either by upgrading two-paired copper cables by means of xDSL (digital subscriber line) technologies (the most popular is ADSL (asymmetric DSL) technology), CATV-based broadband Internet access, as well as broadband wireless technology (e.g. Universal Mobile Telecommunications System (UMTS)). Convergence and platform independence, however, does not mean that these broadband access technologies have the same cost characteristics. They also have different access quality attributes (e.g. mobility, reliability, startup speed, etc.). There are particularly strong quality differences between low-speed access (narrowband) and high-speed access (broadband). For example, the transmission of 100 text pages takes 120 seconds via modem, 25 seconds via ISDN and 0.4 seconds via ADSL. Five colour photos take 22 minutes via modem, five minutes via ISDN and four to five seconds via ADSL. A 30-minute video takes 38.8 hours via modem, 8.7 hours via ISDN and 8 minutes via ADSL (Fesenmeier 2001: 17). The difference in speed already indicates that narrowband Internet access does not provide an economically sensible way to consume data-intensive Internet services such as streaming video and interactive entertainment. On the other hand, dial-up access is sufficient for managing an e-mail account and surfing the Internet for a few hours a week. To the extent that the local loops of the established carriers are still monopolistic bottlenecks, there is still a need for sector-specific regulations, such as price caps, accounting separation and discriminatory-free entry. Alternative providers of broadband access (e.g. CATV networks) are not yet able to discipline the market power of the established local loop provider. Line-sharing obligations, however, which are geared towards stimulating broadband access, are superfluous from the perspective of this low-speed access market. Line-sharing regulation does also not appear to be justified from the perspective of broadband Internet access. From the dynamic perspective of convergence, the division of the Internet into one large narrowband section on the one hand and a rather marginal broadband section on the other is artificial. In order to develop innovation potential for data-intensive Internet services, broadband access is indispensable. Whereas the local loop of copper pairs via xDSL can provide one broadband access possibility, there are also a number of economically feasible access alternatives. Mobile Internet access based on GPRS (General Packet Radio System Standard), as well as UMTS in particular, demonstrates the large innovation potential and evolution of
Information and Communication Technologies, Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
67
mobile technologies for the Internet (see, for example, Büllingen and Wörter 2000; Büllingen and Stamm 2001). Looking ahead, the Europe 2005 Action Plan promotes a multi-platform approach to broadband development driven by strong competition between networks and services. In the meantime, the focus of the EU Commission has shifted towards considering the importance of technology-neutral regulation, which is designed to avoid favouring one technology over another. Technology-neutral regulation is considered to allow provision of new services to lead to competition between different network-access methods (facility based competition). The Commission concluded: ‘When there is effective facilities-based competition, the new framework will require ex-ante regulatory obligations to be lifted’ (Commission of the European Communities 2003: 6). Effective, facilities-based competition should include high-speed access. From the perspective of high-speed broadband access, the local loops of the established telecommunication carriers will thus lose the characteristics of a monopolistic bottleneck. Alternative, broadband-access technologies (such as cable modems, UMTS and mobile access) create economically sensible alternatives to xDSL. Due to the increasing role of product differentiation based on the different network characteristics of these access technologies, the long-run convergence towards a single, globally dominating access technology seems unrealistic. As a consequence, sector-specific regulation of broadband access – especially line-sharing obligations – seems superfluous. The aim of technology-neutral regulation is also stated in the revised German telecommunications law of 2004 (§ 1 TKG). It suggests that in an environment of competing network infrastructures, ex ante regulation should not be extended but removed entirely. Moreover, the traditional regulation of narrowband access should not be continued for historical reasons, but rather abolished as soon as narrowband access loses its bottleneck characteristics. Only then will the adequate incentives for investments in new network infrastructures be provided and an unbiased infrastructure and service competition be guaranteed.
Backbone interconnection: is there still a role for sector-specific regulation? Competitive markets for communications bandwidth Access to the IP-based backbone network is impossible without access to telecommunications transport capacity, which can be delivered by a variety of means, such as high-speed, fibre-optic networks, coaxial cables and satellite. The performance-price ratio for cutting-edge, optical communications technology has been improving rapidly. New transmission technologies work
68
Günter Knieps
most effectively over new fibre strands that have enhanced optical properties, and developments in optical technology have unquestionably made massive increases in bandwidth possible. Consequently, the growth of bandwidth for Internet traffic use has been dramatic since 1995. Nevertheless, expectations of a bandwidth revolution similar to Moore’s Law on the performanceprice ratio for computers have not yet been fulfilled. Costs and benefits of additional investment into bandwidth have to be counterbalanced, also by exploiting the benefits of substitution among bandwidth, storage and CPU (central processing unit) cycles (Galbi 2000). Communications bandwidth is readily available today from a variety of providers. IBPs own or lease communications bandwidth that is connected by routers which the backbones use to deliver traffic to and from their customers. The underlying network logistics is the TCP/IP protocol. Whereas the IP (Internet Protocol) is responsible for shifting the data packets from router to router, the TCP (transfer control protocol) is responsible for the reliability of transmission, including error correction. Internet backbone providers (IBPs) are also responsible for the quality of service and network management, including the capacity control of the backbone network. An additional dimension of Internet backbone services is the organization of interconnections with other IBPs by means of peering and transit arrangements. Unregulated interconnections: transit and peering Each IBP forms its own network that enables all end users and content providers connected to it to communicate with each other. End users, however, often want to be able to communicate with a wide variety of end users and content providers, regardless of the IBPs involved. In order to provide end users with such universal connections, IBPs must interconnect with one another to exchange traffic destined for each other’s end users. It is this interconnection that makes the Internet the ‘network of networks’. Peering and transit arrangements can be differentiated. Peering partners exchange traffic on a settlement-free basis (bill and keep type), that is, each peer terminates the traffic originating with other peers without charge. In contrast, in transit arrangements one IBP pays another IBP to deliver traffic between its customers and the customers of other IBPs (see Kende 2000: 5). Peering used to take place in the United States at public peering points, so-called NAPs (network access points),15 where different backbones could exchange traffic. Due to increased congestion at the NAPs, IBPs turned to bilateral peering arrangements (private peering). As each bilateral peering arrangement only allows backbones to exchange traffic destined for each other’s customers, backbones need a significant number of peering arrangements in order to gain access to the full Internet. Many IBPs have adopted a hybrid approach to interconnection, peering with a number of
Information and Communication Technologies, Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
69
backbones and paying for transit from one or more IBPs in order to have access to the backbone of the transit supplier as well as the peering partners of the transit supplier. Transit and peering arrangements among IBPs are not subject to sectorspecific regulation, neither by the Federal Communications Commission (FCC) nor by the regulatory agencies in Europe. The agreements that cover interconnection between IBPs are characterized by private negotiations and are subject to non-disclosure rules. The input market of communications bandwidth is competitive and each IBP can develop its own logistic concept to optimize its own backbone and set of transit and peering arrangements.16 In this context, there is indeed no role for exante regulation due to the absence of network-specific market power.
Summary In order to analyse the remaining role for sector-specific regulation the focus of this chapter has been on those elements of the Internet periphery and Internet service provision that are strongly based on telecommunications, in particular Internet access and Internet backbone. The first section dealt with the role of telecommunications for the Internet, differentiating between local network access and long-distance network capacity. In the next section the new regulatory arrangements for communications services within Europe, with particular emphasis on Germany, were explained. In order to analyse the future role of sector-specific regulation from a normative point of view, in the following section the network economic concept of a disaggregated regulatory approach was provided. In subsequent sections the phasing-out potentials for sector-specific regulation due to increasing competition within the local loop was dealt with and the role of technologyneutral regulation was considered, which implied that in an environment of competing network infrastructures sector-specific regulation should not be extended, but removed. The final section explained the role of competition in the markets for backbone interconnection.
Notes 1 For the important role of markets and committee solutions in order to deal with network externalities, see Blankart and Knieps (1993, 1995) and Holler et al. (1997). 2 Directive 2002/21/EC of the European Parliament and of the Council of 7 March 2002 on a common regulatory framework for electronic communications networks and services (Framework Directive), Official Journal of the European Communities L108/33, 24 April 2002. 3 Directive 2002/19/EC of the European Parliament and of the Council on access to, and interconnection of, electronic communications networks and associated facilities (Access Directive), Official Journal of the European Communities L108/7, 24 April 2002.
70
Günter Knieps
4 Directive 2002/20/EC of the European Parliament and of the Council of 7 March 2002 on the authorization of electronic communications networks and services (Authorization Directive), Official Journal of the European Communities L108/21, 24 April 2002. 5 Directive 2002/22/EC of the European Parliament and of the Council of 7 March 2002 on universal service and users’ rights relating to electronic communications networks and services (Universal Service Directive), Official Journal of the European Communities L108/51, 24 April 2002. 6 Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector (Directive on privacy and electronic communications), Official Journal of the European Communities L201/37, 31 July 2002. 7 Regulation (EC) No 2887/2000 of the European Parliament and of the Council of 18 December 2000 on unbundled access to the local loop, Official Journal of the European Communities L336/4, 30 December 2000. 8 This does not rule out the fact that Directives may have direct effect in member states, provided that the provisions of the Directive are sufficiently precise and unconditional. 9 Telekommunikationsgesetz (TKG) vom 25 Juli 1996 (BGBl. I S.1120). 10 Telekommunikationsgesetz (TKG) vom 22 Juni 2004 (BGBl. I S. 1190). 11 The traditional methods and approaches in general competition law, both with respect to merger control and the control of abusive practices, are fundamentally different from those of sector-specific, regulatory economics. Any mingling of these two different approaches is misleading. The chapter examines the question of a sector-specific need for regulation and thus does not comment on merger cases. Market shares and turnover are easily measurable and are therefore usually taken up as criteria in competition law. However, they must not in any way be confused with a sound economic analysis of the effectiveness of active and potential competition. When examining a merger case, the competition authorities use a large number of criteria to which they attach, by their own discretion, a varying degree of significance on a case-by-case basis. For the general control of abusive practices, as well, competition law envisions a correction of market processes on a case-by-case basis. 12 This means that access to ports, airports or railway networks can neither be refused nor granted under conditions that penalize competitors without factual justification. 13 The fact that use of this facility is essential for competition on the complementary market is also occasionally expressed as a third criterion, as it reduces prices or increases the volumes offered. This third criterion, however, only describes the effects of access. 14 Oftel (2001: 41) differentiates between ‘wholesale call origination’ and ‘wholesale Internet call termination market’. 15 In 1993, the US National Science Foundation (NSF) designed a system of geographically dispersed NAPs (Kende 2000: 5). 16 General competition law also applies to transit and peering arrangements. However, antitrust proceedings are geared towards dealing with concrete conflicts case by case and not towards designing a new, exante regulatory policy.
Information and Communication Technologies, Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
71
References Areeda, P. and Hovenkamp, H. (1988), ‘“Essential facility” doctrine? Applications’, in Areeda, P. and Hovenkamp, H. Antitrust Law, 202.3 (Suppl. 1988), pp. 675–701. Baumol, W. J., Panzar, J. C. and Willig, R. D. (1982) Contestable Markets and the Theory of Industry Structure, San Diego CA: Harcourt Brace Jovanovich. Beesley, M. E. and Littlechild, S. C. (1989) ‘The regulation of privatized monopolies in the United Kingdom’, Rand Journal of Economics, 20: 454–72. Blankart, Ch. B. and Knieps, G. (1993) ‘State and standards’, Public Choice, 77: 39–52. Blankart, Ch. B. and Knieps, G. (1995) ‘Market-oriented open network provision’, Information Economics and Policy, 7: 283–96. Büllingen, F. and Stamm, P. (2001) Mobiles Internet – Konvergenz von Mobilfunk und Multimedia, Discussion Paper no. 222, Bad Honnef: Wissenschaftliches Institut für Kommunikationsdienste (WIK). Büllingen, F. and Wörter, M. (2000) Entwicklungsperspektiven, Unternehmensstrategien und Anwendungsfelder im Mobile Commerce, Discussion Paper no. 208, Bad Honnef: Wissenschaftliches Institut für Kommunikationsdienste (WIK). Commission of the European Communities (2003) Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions, Electronic Communications: the road to the knowledge economy, Brussels: COM, 11 February, COM(2003) 65 final. Distelkamp, M. (1999) Möglichkeiten des Wettbewerbs im Orts- und Anschlußbereich des Telekommunikationsnetzes, Discussion Paper no. 196, Bad Honnef: Wissenschaftliches Institut für Kommunikationsdienste (WIK). Economides, N. (2000) The Microsoft Antitrust Case, Working Paper no. 2000–09, New York: Stern School of Business, New York University. Elixmann, D. (2000) ‘Wettbewerbsintensität auf dem Markt für Übertragungskapazität’, WIK Newsletter no. 41 (December): 6–9, Wissenschaftliches Institut für Kommunikationsdienste (WIK), Bad Honnef. Elixmann, D. and Metzler, A. (2001) Marktstruktur und Wettbewerb auf dem Markt für Internet-Zugangsdienste, Discussion Paper no. 221, Bad Honnef: Wissenschaftliches Institut für Kommunikationsdienste (WIK). Engel, C. (1999) Das Internet und der Nationalstaat; Gemeinschaftsgüter: Recht, Politik und Ökonomie, preprints of the Max-Planck – project group ‘Recht der Gemeinschaftgüter’, Bonn. Engel, C. and Knieps, G. (1998) Die Vorschriften des Telekommunikationsgesetzes über den Zugang zu wesentlichen Leistungen: Eine juristisch-ökonomische Untersuchung, Baden-Baden: Nomos. European Commission (1998) The 1999 Review of the Telecommunications Regulatory Framework, ONP COM 98–42, September, Brussels: Directorate General XIII ONP Committee. Fesenmeier K.-H. (2001) ‘Plötzlich wollen alle ADSL’, Badische Zeitung, 14 July: 17. Fisher, F. M. (2000) ‘The IBM and Microsoft cases: what’s the difference?’, American Economic Review, Papers & Proceedings, 90: 180–3. Gabelmann, A. and Gross, W. (2003) ‘Telekommunikation: Wettbewerb in einem dynamischen Markt’, in Knieps, G. and Brunekreeft, G. (eds) Zwischen Regulierung und Wettbewerb – Netzsektoren in Deutschland, 2nd edition, Heidelberg: Physica.
72
Günter Knieps
Galbi, D. A. (2000) Growth in the ‘New Economy’: US bandwidth use and pricing across the 1990s, FCC Working Paper, 9 July, Washington DC: FCC. Hillebrand, A. and Büllingen, F. (2001) Internet-Governance – Politiken und Folgen der institutionellen Neuordnung der Domainverwaltung durch ICANN, Discussion Paper no. 218, Bad Honnef: Wissenschaftliches Institut für Kommunikationsdienste (WIK). Holler, M. J., Knieps, G. and Niskanen E. (1997) ‘Standardization in transportation markets: a European perspective’, EURAS Yearbook of Standardization, 1: 371–90. Immenga, U., Kirchner, C., Knieps, G. and Kruse, J. (2001) Telekommunikation im Wettbewerb: Eine ordnungspolitische Konzeption nach drei Jahren Marktöffnung, Munich: C. H. Beck. Kende, M. (2000) The Digital Handshake: connecting internet backbones, Federal Communications Commission, OPP Working Paper no. 32, Washington DC: FCC. Kesan, J. P. and Shah, R. C. (2001) Fool Us Once Shame On You – Fool Us Twice Shame On Us: what we can learn from the privatizations of the internet backbone network and the domain name system, Law and Economics Working Paper no. 00–18, Urbana Champaign IL: University of Illinois College of Law. Knieps, G. (1997a) ‘Phasing out sector-specific regulation in competitive telecommunications, Kyklos, 50: 325–39. Knieps, G. (1997b) ‘The concept of open network provision in large technical systems’, EURAS Yearbook of Standardization, 1: 357–69. Knieps, G. (2000) ‘Interconnection and network access’, Fordham International Law Journal, 23: 90–115. Knieps, G. (2003) ‘Competition in telecommunications and internet services: a dynamic perspective’, in Barfield, C. E., Heiduk, G. and Welfens, P. J. J. (eds) Internet, Economic Growth and Globalization: perspectives on the new economy in Europe, Japan and the US, Berlin: Springer, 217–27. Knieps, G. (2006) ‘Privatisation of network industries in Germany: a disaggregated approach’, in Köthenbürger, M., Sinn, W.-H., Whalley, J. (eds) Privatization Experiences in the European Union, Cambridge MA; London: MIT Press, pp. 199–224. Knieps, G. and Vogelsang, I. (1982) ‘The sustainability concept under alternative behavioural assumptions’, Bell Journal of Economics, 13: 234–41. Laffont, J.-J. and Tirole, J. (2000) Competition in Telecommunications, Cambridge MA: MIT Press. Mestmäcker, E.-J. (2001)’ Unternehmenskonzentrationen und Urheberrechte in der alten und “neuen” Musikwirtschaft’, Zeitschrift für Urheber- und Medienrecht (ZUM), 3: 185–94. Müller, G. and Rannenberg, K. (1999) Multilateral Security in CommunicationsTechnology, Infrastructure, Economy, Munich: Addison-Wesley-Longman. O’Donnell, S. (2000) ‘Broadband architectures, ISP business plans, and open access’, paper presented at the 28th Annual Telecommunications Policy Research Conference, 23–25 September, Alexandria VA. Oftel (2001) ‘Oftel’s 2000/01 effective competition review of dial-up Internet access’, issued by the Director General of Telecommunications, London. Regulation Office for Post and Telecommunication (Regulierungsbehörde für Telekommunikation und Post) (1999) Jahresbericht, Bonn. Sidak, J.G. (2001) ‘An antitrust rule for software integration’, Yale Journal on Regulation, 18: 1–83.
Information and Communication Technologies, Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
73
Stigler, G. J. (1968) ‘Barriers to entry, economies of scale, and firm size’, in Stigler, G. J. The Organization of Industry, reprint, 67–70, Chicago: University of Chicago Press. Stumpf, U. and Schwarz-Schilling, C. (1999) Wettbewerb auf Telekommunikationsmärkten, Discussion Paper no. 197, Bad Honnef: Wissenschaftliches Institut für Kommunikationsdienste (WIK).
Subsection B The increasing role of self-regulation
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
5
Private solutions to uncertainty in Japanese electronic commerce1 Cornelia Storz
Introduction Electronic commerce is a central element in the OECD’s vision of the tremendous potential that our networked world now holds.
(OECD 2001) There is no doubt that the development of electronic commerce has become an important political issue. Its political and economic potential is seen to reside in major gains in efficiency, in the creation of new markets and in the worldwide integration of trade. Several guidelines and recommendations made by the OECD (Organisation for Economic Co-operation and Development) and its member states suggest the creation of best practice institutions to reduce many of the uncertainties that accompany electronic commerce, such as concerns about the fairness of exchange and the protection of privacy (OECD 1999).2 Institutions – defined here as a sum of rules that govern human behaviour and are secured by a credible sanctioning mechanism – aim at facilitating interaction by reducing uncertainty. They can be generated by the government (public policies) or by private actors (private regulation).3 Especially in the United States and in Japan, private regulation has been given the priority in fostering the development of electronic commerce. Japan, for example, has only just recently amended laws directed at electronic commerce after having relied on the creation of private rules, such as minimum quality standards aimed at the protection of consumer privacy, for many years.4 Nevertheless, the interesting point is that despite comprehensive and high-quality private regulations, electronic commerce in Japan today is still far less developed. Research in the field of information economics has shown that the probability of markets to fail is high without protective institutions. The design of appropriate institutions to prevent market failure has thus become a major theoretical focus. While it is indeed important to consider what kind of institution induces what kind of behaviour, a mere focus on the design of institutions overlooks the fact that institutions by themselves – whether privately or publicly designed – do not ‘automatically’ reduce
76
Cornelia Storz
uncertainty. In order for institutions to function well, they need to be regarded as legitimate by those who are encountering them. In other words, informal institutions become important. Since informal institutions vary across countries, formal, more structural institutions perceived in one country as legitimate and trustworthy may be perceived in another country as untrustworthy precisely due to the different values, norms and attitudes in which they are embedded. It thus becomes necessary to take a closer look at informal institutions and the ways in which they influence the acceptance of formal institutions. This reasoning becomes clearer when we examine the concept of path dependency. In economic theory, path dependency describes a process by which events of the past have long-lasting effects on events that follow, and actors can only influence this process to a certain degree. The development of electronic commerce, for example, is said to be constrained in societies such as in Japan where the normative framework of personal trust predominates. Newly created institutions set up to raise the credibility and legitimacy of electronic commerce need a more general type of normative framework, namely institutional trust. Path dependency further implies that it is extremely difficult to reform institutions due to the complementarities between formal and informal institutions. According to this approach, then, certain kinds of informal institutions may hinder the acceptance of new formal institutions, even in the long run. Electronic commerce in Japan and in other Asian countries is thus ‘locked in’ a specific path of development that renders these countries unable to take advantage of its potential.5 While I acknowledge the existence of and the implications associated with specific informal institutions such as trust, I want to go a step further by arguing that there is much more leeway for change in a society than institutional economics has often identified. I develop an alternative framework by incorporating an emphasis on the learning capacity of individuals and their ability to generate creative solutions to economic problems based on a case study of Japanese electronic commerce. The chapter is thus organized as follows: this paper starts with the general question of which institutions are appropriate to stimulate electronic commerce. Based on these reflections, the next section provides a description of the institutional solutions that have been implemented in Japan. The following section presents the argument that an important precondition for these appropriate institutions is that they are generally accepted and regarded as being trustworthy, and that this type of legitimacy necessitates a special form of trust, namely institutional trust. In a society in which personal trust predominates, even perfectly well-designed institutions will not be perceived as trustworthy and therefore will not be able to channel behaviour into the desired direction. Without widespread institutional trust, even best practice institutions would not lead to an increase in electronic commerce in Japan. However, I argue in the next two sections
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
77
that paths can be channelled into new directions by entrepreneurial alertness. Creative solutions are discussed that have been generated to solve the specific problems inherent in Japanese electronic commerce. In the final section I summarize my results, discuss the future perspectives for electronic commerce in Japan, and offer a few general considerations on a revised concept of path dependency.
Consumer uncertainty and institutional solutions Electronically traded goods as experience and credence goods: the problem of new uncertainties Electronic commerce takes place in electronic markets in which personal communication is replaced by electronic communication. Electronic commerce can therefore be described as a virtual room in which actors realize exchange. In the 1990s, the new and open HTML standard (Hypertext Markup Language) ushered in the starting point for worldwide business-toconsumer commerce (electronic commerce between business and consumers) by opening the electronic marketplace to all participants in contrast to the previous infrastructure based on proprietary standards.6 With the new standard, the use of the Internet became possible even for those with relatively little knowledge of information and communication technology. However, by replacing real with virtual rooms, electronic commerce leads to new uncertainties that are not present in conventional, stationary trade. Consumer uncertainty results out of a lack of information, which cannot be supplied even at high costs due to information asymmetries. Information asymmetries describe the fact that economic actors possess information to varying degrees and with varying quality. The problems associated with consumer uncertainty are nothing new in economic history and are generally solved by institutions providing information to consumers.7 In the wellknown example of the used car market as ‘lemon’ markets, Akerlof (1970) demonstrated that without uncertainty-reducing institutions, information problems can lead to less than optimal market outcomes and, in extreme cases, to market failure (see also Beales 1998). The wide range of institutions working to counteract market failure can be divided into public policies and private regulations. Such institutions place restrictions on both consumer and business choices. For example, lower-quality goods at lower prices are not allowed on the market. Such restrictions are appropriate, however, if the advantages resulting from them are on average superior to a situation without them. Institutions are, in this view, a response to the demand for better information about products and services and encourage the development of new markets.8 The information problem has a lot to do with differences in the characteristics of goods. Inspection goods are products that consumers can evaluate before their purchase at no major cost. Experience goods, on the
78
Cornelia Storz
other hand, can only be assessed after they have been bought and used. Consumer assessment of credence good quality is even more difficult since the product cannot be evaluated even after the purchase. A physician’s treatment is the classic example of a credence good. Assessing product quality is also difficult when product quality is expanded to mean not only the quality of the actual product itself but also the quality of the production and distribution processes that go into it, such as how well the consumer’s personal data are protected or how secure the transaction conditions for its purchase, return and payment are. When we define product quality in broad terms such as these, we find that the new uncertainties and information problems associated with business-to-consumer commerce are quite daunting: the quality of all goods traded in business-to-consumer commerce can only be assessed after the purchase, and sometimes, as with payment or data security, not at all, so that electronic commerce goods belong mostly to the class of experience or credence goods. At the same time, established signals such as the location of stores, the conduct of sales assistants, nonverbal communication in the ‘real world’ and the hands-on experience of inspection goods – all of which reduce uncertainty – no longer have any meaning in virtual trade. The reduction of consumer uncertainty, especially on the Internet, is thus crucial for the success of business-to-consumer commerce. In order to stimulate electronic commerce, the first and necessary step is to develop appropriate institutions to combat consumer uncertainty. A wide range of literature has discussed the positive and normative aspects of these rules, to which the following section will now turn. Public policies and private regulation In institutional economic theory there is a general consensus that government regulation should not interfere with exchange and should be provided in a way that leaves sufficient leeway for private actors to solve the problems that arise in the private market. The expression ‘sufficient leeway’ indicates that there are different ideas about the extent to which private regulation should take place, as the different modes of regulation in Japan, in the United States and in Germany demonstrate (see Knieps, Chapter 4; Lageman, Rothgang and Scheuer, Chapter 7). Over the past few decades and partly due to the new characteristics of computer technologies, most countries have preferred to rely on private regulation in business-to-consumer commerce for three reasons. First of all, private regulation is believed to be more efficient since the costs for governments are lower if formulation, decision making and sanctioning lie in the hands of private actors. According to transaction cost theory, even if control in the form of accreditation agencies is incumbent on the state, most of the costs stemming from the prior phases of the political process are not applicable (Campbell 1998–99). One problem with this viewpoint is that the extent to which transaction costs can be reduced varies depending
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
79
on the structural characteristics in a given society (e.g. distribution of information or the degree of consumer organization). Nevertheless, according to this argument, private solutions may be superior to public policies. A second argument for favouring private regulation over public policy centres on the market’s specialized knowledge. It is assumed that those actors involved with the day-to-day, actual market processes know better where major problems in the marketplace lie as well as the major difficulties associated with implementing solutions to those problems. In contrast to governmental regulation – an approach that presumes that the state as a central institution can correctly recognize market deficits and choose suitable instruments – private regulation avoids the pitfalls of centrally created knowledge since the creation of rules devolves on a variety of individual actors. More knowledge is created among decentralized, acting individuals and among them better rules (Michael 1995). Finally, economists influenced by insights from social psychology argue that the motivation to adhere to a rule is greater when it is privately agreed upon than when it is dictated by the government. Motivation to comply with a private contract is intrinsic to the contractual process itself (Frey 1997). The problem of information asymmetries can be solved by private actors by signalling or screening.9 The discussion will be restricted to signalling. The establishment of signals requires private regulations such as standards determined by independent parties as a basis for quality assessment. This takes place in private standardization committees on national, regional and international levels, such as in the German DIN (Deutsches Institut für Normung), the Japanese JISC (Japanese Industrial Standards Committee), the European CEN (Comité Européen de Normalisation), the ISO (International Organization for Standardization), or the IEC (International Electronic Commission). They all set minimum quality standards that regulate, for example, the use of private data and transaction conditions.10 In those cases in which such standards do not exist or in which they are considered to be insufficient, they may be supplemented by standards set by industrial associations or private companies. Firms introduce such labels and certificates in order to signal credibility. But the introduction of a signal is not equivalent to compliance with the signal.11 In order to ensure compliance with private regulation, then, it is crucial that credible sanctions accompany it. Such sanctioning normally takes place through monitoring systems that control an economic actor’s conformity to a standard. Private certification agencies formally accredited by a government agency, such as the German Association for Accreditation (TGA) or the Japanese Accreditation Board of Conformity Assessment (JAB), usually perform this function. Most OECD countries have such a certification and accreditation system for electronic commerce.12 Assuming that a minimum standard of quality and a working monitoring system are in place, private regulation can play a reliable and, in some cases, an even better informational role for consumers than mandatory,
80
Cornelia Storz
governmental rules. For this reason I will focus in the following on private regulation in Japanese electronic commerce and compare it to the United States.
Private regulation in Japanese business-to-consumer commerce13 The Japanese government has stressed the importance of private regulation in ensuring reliable transaction processes, data protection and consumer privacy. Private regulation has been encouraged by several guidelines put in place by Japanese central ministries. Nevertheless, in some cases the need for governmental regulation has been stressed. After a brief discussion of government policies regarding electronic commerce in Japan, I will turn to the area of private regulation. In the course of the growth of the Internet and its accompanying problems of privacy protection (CSLR (Center for Social and Legal Research) 2000b: 17–19), there ensued a growing demand in Japan to amend laws and to formulate new ones to apply to the Internet. In addition to consumer associations, even the leading business federation Keidanren called for governmental regulation of electronic commerce (Keidanren 1999). Most Japanese laws applicable to electronic commerce rely on general, non-specialized laws. This is somewhat unique in comparison to other countries. One of the most important laws is the Specific Commercial Transaction Act enacted in 1976 and amended in 2000. It requires direct marketers to inform consumers about important transaction issues such as their merchandise return policy, or to provide consumers with the name of a contact person (Doi 2001: 5–9; Mitsubishi Research Institute 2002). Other important new laws enacted are the Consumer Contract Act, the Privacy Bill and the Digital Signature Law: •
•
•
The Consumer Contract Act of 2001 is designed to protect consumers from unfair business practices due to insufficient information by allowing them to nullify contracts under certain conditions, e.g. if the supplier fails to disclose disadvantageous material facts (Doi 2001: 6). The so-called Privacy Bill of 2001, which concerns the protection of data and consumer privacy, aims to protect consumers by ensuring that their data is not misused and that it is handled appropriately. Violators may be subjected to punishment of up to six months in jail or to a fine of up to 300,000 yen – about $2,750 (Horibe 2000). The Digital Signature Law of 2000 grants electronic signatures the same legal status as handwritten signatures (Mitsubishi Research Institute 2002; see also Horie 2002).14
Not only the relatively late amendment and formulation of laws but also the early formulation of guidelines for private regulations make it obvious that the Japanese government was reluctant to enact strict regulations.
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
81
Already in 1989, the Japanese Ministry of Economy, Trade and Industry (METI) encouraged the private sector to take the initiative in formulating appropriate private rules for fostering Internet transactions. Over the following years, ministries and agencies developed guidelines for their particular constituencies.15 Private regulation of electronic commerce is promoted in international foreign trade relations as well, as the US–Japanese Joint Statement on Electronic Commerce announced in 1998 by the President of the United States and the Prime Minister of Japan demonstrates. This statement pinpoints privacy issues in particular as a very important area for the development of business-to-consumer commerce and suggests that governments should encourage private regulation through guidelines or model contracts (OECD 1999: 33).16 Guidelines play an important role in Japanese economic policy and are often implemented by business associations. The METI guidelines of 199717 were based on the earlier version of 1989 and derived from recommendations geared towards protecting personal data laid out by the OECD in 1980. The revised form of 1997 specifically addressed issues raised in the European Union Personal Data and Privacy Directive of 1995. Thus, the METI guidelines of 1997 are ultimately based on so-called ‘fair information principles’ very similar to those formulated by the OECD or the EU. They regulate data collection and usage, the rights of data subjects, the duties of data controllers and requirements in formulating standards for data security. The METI guidelines also explicitly call for industry guidelines and put pressure on Japanese associations to support their member companies in implementing privacy policies. Shortly after the revision of the guidelines in 1997, METI published a Personal Data Protection Handbook, which served as a model for standards such as the Japanese Industrial Standard JIS Q 15001 by virtue of its detailed explanation of guidelines and its incorporation of case studies (OECD 1999: 34; CSLR 2000a: 3, 19–20; Mitsubishi Research Institute 2002). This JIS standard provides minimum privacy requirements and applies to both online and offline activities. A certified company is required to prove that it carries out appropriate measures for consumer protection by ensuring that it adheres to the following technical and organizational procedures: • • • •
Consumers’ consent is required before a company acquires personal information. Consumers must be provided with appropriate information, such as how their personal data will be used. Consumers must be informed that the company is required to supply them with certain information. Organizational strategies, such as appointing an in-house data protection manager or implementing education and training programmes for employees, have to be undertaken.18
82
Cornelia Storz
In order to encourage further the development of private standards, the Japanese Information Processing Development Center (JIPDEC), a nonprofit organization founded in 1967 with the support of the METI to promote the use of information processing in Japan,19 developed a privacy mark programme in 1998. To do so it relied heavily on the newly created JIS Q 15001 standard, which thus served as a basis for measuring the quality of transaction terms. Such privacy mark programmes constitute a core element of the development of business-to-consumer commerce in the United States as well. Certification with the privacy mark of JIPDEC follows the general pattern of certification and accreditation systems: JIPDEC is approved by METI and works as the certifying and auditing organization for the privacy mark awarded to single firms and associations. Associations approved to carry out certifications themselves include the Japan Information Technology Service Industry Association (JISA), the Japan Marketing Research Association (JMRA) and the Japan Juku Association (cram schools) (JJA). The concrete activities of granting the mark, auditing and monitoring are carried out by the Privacy Mark System Committee within JIPDEC. The auditing fee is around 80,000 yen (about $730), while the mark use fee comes to about 50,000 yen ($460). The mark is awarded for a two-year period and can be renewed. JIPDEC may sanction certified enterprises for non-compliance. The severest sanction is the withdrawal of the mark, and this has been carried out in one case. In addition, there is a consulting centre for consumers within JIPDEC, which processes complaints regarding the illegal processing of personal data. According to the assessment of experts, the requirements and the certification system are similar to the US privacy seals programme developed by the two leading American certification organizations – BBBOnLine and TRUSTe.20 The signing of a mutual recognition agreement between JIPDEC and BBBOnline is a further indicator that the quality level of the American and the Japanese certification systems should be comparable (Matsumoto 1998; CSLR 2000a: 30, 32 and 2000b: 33; Privacymark 2003a, 2000b, 2000d). The number of companies certified with the JIPDEC privacy mark is 136, still low as of 2002 (Nagato 2003: 61), yet the promotion of the privacy mark has been successful in the sense that businesses and consumers are now more sensitive to privacy and security issues. This is evidenced by the fact that those associations active in electronic commerce, such as the Japan Chain Store Association, the Japan Direct Marketing Association, the Japan Department Stores Association, the Cyber Business Association and the Japan Federation of Consumer Credit Companies, have all adopted guidelines to protect personal data (for a list of organizations see the CSLR 2000a: 36). Another indicator is that 58 per cent of the single member companies of the private Electronic Commerce Promotion Council (ECOM) have established rules on personal data protection. The rest either have formulation under review (21 per cent) or their creation depends on the business division (15 per cent).21
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
83
The privacy mark described above reduces uncertainty when it comes to a firm’s privacy policy. However, another important issue is the reduction of uncertainty within the whole transaction process itself. With this aim in mind, ECOM was founded in 1996. Like JIPDEC, it is an approved association as well, but the majority of its members and staff belong to the private sector.22 One work group in ECOM established in cooperation with consumer associations launched the ‘Consumer Transaction Guideline’ in 1998 in order to reduce uncertainties in business-to-consumer commerce. According to this guideline, Internet suppliers are required to disclose information about their trading conditions and to clearly display the firm’s privacy policy on their homepage. The collection of personal click-stream data without informing the consumer is forbidden. Moreover, firms are obliged to fully disclose their transaction terms, introduce a system for cancellations of orders and return and implement measures for quickly settling disputes (CSLR 2000a: 31 and 2000b: 31–32; Plate 2000: 66). This guideline has been taken up by several members of ECOM, which then grants them a trust mark. The most important member association is the Japan Direct Marketing Association (JADMA), which adapted the ECOM guideline in its ‘Guidelines for Electronic Direct Marketing’ in 1999 and which grants the so-called ‘Online Trust Mark’ in cooperation with the Japanese Chamber of Commerce and Industry (JCCI). The Mark has been awarded to about 640 businesses, which constitute about half of the JADMA membership as of 2003 (ECOM 2001a; Mitsubishi Research Institute 2002; ECOM 2003). Moreover, ECOM has also started an international cooperation with leading American institutions (BBBOnline 2000). An ‘Online Trust Alliance’ between JADMA, JCCI, BBBOnLine and the Korea Institute for Electronic Commerce has been established, and European trustmark organizations such as the Association of European Chambers of Commerce and Industry and the Federation of European Direct Marketing have announced their commitment to the Alliance (ECOM 2001b). As a first result, an Asia Trustmark Alliance was established in 2003, also designed as a mutual recognition system. In summary, the rules reducing uncertainties in business-to-consumer commerce in Japan seem to be relatively comprehensive. The quality of the Japanese system is recognized by bilateral US–Japanese alliances and mutual recognition agreements. The JIS Q 15001 standard constitutes a minimum quality standard for privacy concerns that echoes the requirements for the protection of privacy set out by the OECD. Thus, the quality of certification should be as high as it is in the United States, where private regulation has also been enforced as the dominant strategy to reduce uncertainty in electronic commerce. Nevertheless, an astonishing difference between the volume of businessto-consumer commerce in Japan and that of the United States still persists. Even if the existing statistics are difficult to compare, they show at least an interesting trend. In Japan, the volume of business-to-consumer commerce
84
Cornelia Storz
lies at about $10.5 billion. In contrast, the volume of business-to-consumer commerce in North America is $135.3 billion – more than tenfold that of the Japanese.23 Japan is not even under the top ten OECD members with the highest share of online-shopping by Internet users (NFO 2003: 234, 310).24 The share of electronic commerce in retailing is about 1 per cent in Japan while it is about 1.9 per cent in the United States. In addition, business-toconsumer commerce growth rates in Japan are slow. Business-to-consumer commerce grew by 0.1 per cent in 1999 and 0.25 per cent in 2000, whereas the growth rates in the United States were 0.69 and 1.37 per cent respectively (METI 2001). The low volume and low growth rates for Japan are especially surprising when two additional facts are taken into account: Internet compatible mobiles such as i-mode are highly successful in Japan, so that mobile commerce could potentially stimulate more electronic commerce than the amount currently taking place in the United States. Moreover, Japanese consumers report the same advantages to business-to-consumer commerce as do their American consumer counterparts. From a customer’s point of view, the economic advantage of electronic commerce lies in its lower transaction costs. In mobile, highly industrialized, time-pressed and increasingly price-sensitive societies, we would expect enormous incentives to purchase via the Internet. But trade via the Internet in Japan has fallen far short of expectations. Even if one takes underestimated factors that favour traditional commerce into account, such as the Japanese retail system with long shopping hours, the numbers are perplexing. Institutional conditions, such as old-fashioned payment systems or high access costs for the Web, also appear to be insufficient as explanations in and of themselves.25 A more convincing explanation seems to be the role of trust. The problem of producing trust in electronic commerce is generally a widespread problem, as experimental economics has shown (Bolton et al. 2003). However, it gains special importance in countries with informal institutions that do not seem to fit with the necessity of electronic commerce. In the following section I discuss why trust and exactly what kind of trust is a sufficient precondition for business-to-consumer commerce and how the Japanese form of trust should be understood in this context.
Trust as a precondition for institutional solutions Trust and path dependency As briefly described above, reducing information asymmetries through private regulation encourages the development of trust. Since regulations are always incomplete – be it the type of sanctioning system or the rules themselves – a certain level of uncertainty always exists. It is in this sense that trust becomes important, since only the pre-existence of a certain level of trust makes the institutions work. If established economic rules are not
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
85
perceived to be trustworthy, they will not contribute to the aim of increasing the number of economic transactions. Trust is defined here as a special type of an informal institution that allows actor A to reasonably expect that actor B will comply to the arrangement agreed upon between A and B, even in a situation of uncertainty. Trust reduces complexity and risk and enables individuals to build up mutual expectations about the future. Individual expectations and perceived obligations thus constitute important elements of trust. As a consequence, trust becomes a functional substitute for control and monitoring, especially in those cases where the costs of searching for information are extremely high. Furthermore, trust also needs to be distinguished from similar terms that express expectations such as confidence or hope. Confidence designates a more passive mode of behaviour in the sense that an actor relies on a system functioning well, while hope is associated with chance or fortune. In contrast to both confidence and hope, trust implies an intention or motivation to carry out a concrete activity, in this case a transaction via the Internet (see Khalil 1994; Yamagishi and Yamagishi 1994). Arrow (1974) once stated that trust is an efficient lubricant in economic exchange processes. Nevertheless, extensive research on trust has only just begun. A well-known research project carried out by Knack and Keefer (1997) presents empirical evidence that different degrees of trust and civic norms – the two core elements of social capital – have an effect on a nation’s competitiveness, thereby supporting earlier, more exploratory research that focused on the interconnections between culture and economic growth (Casson 1993; Khalil 1994). On the micro level, Lane and Bachmann (1996: 365) come to the conclusion that ‘the new goals [such as high quality, versatility and innovativeness of products] cannot be achieved by isolated firms but only through the development of close and integrated relationships between manufacturing firms and their suppliers’, and that ‘trust plays an important role in the constitution of supplier relations’. Other findings confirm the important economic role of trust in functional areas such as marketing partnerships (Aulakh et al. 1996). In other words, trust is becoming more and more recognized as an important economic resource and increasingly relevant to economic analysis.26 In fact, trust becomes a factor of production, which explains its close association with the term ‘social capital’. Not only does it increase efficiency and make better coordination of economic exchange possible but, even more importantly, it becomes a precondition for exchange. Without trust, exchange would be impossible in many instances due to the characteristics associated with experience and credence goods noted above. In electronic commerce, trust gains even more importance. The tremendous importance of trust and the problems of specific cultural dispositions become obvious if the term trust is differentiated further according to its objects and its genesis. The first differentiation deals with the object of trust, which corresponds to the distinction between personal
86
Cornelia Storz
and institutionally based (or institutional) trust (Giddens 1990; Loose and Sydow 1994). This differentiation is so important that social psychologists do not even call personal trust ‘trust’ but rather ‘assurance’ or ‘commitment’ (Yamagishi and Yamagishi 1994). Institutional trust provides a ‘springboard in the attempt to break psychological inertia that has been mobilized to maintain committed relations’ (Yamagishi and Yamagishi 1994: 129), whereas personal trust is described as a commitment that becomes a liability rather than an asset as opportunity costs increase (Yamagishi 1988). The second important differentiation refers to the different ways in which trust is generated. Zucker’s work (1986) on the genesis of trust examines the role of trust in economic development in the United States from 1840 to 1920. She starts from the premise that there are three different types of trust: (1) process based; (2) characteristic based; and (3) institutionally based. Process-based trust develops out of the concrete experience of long-term and stable transactions by which expectations towards further transactions arise. Characteristic-based trust centres on the characteristics of the trustees, e.g. his or her family background or ethnicity. With characteristicbased trust, the concrete transaction experience is not necessary. Finally, institutionally based trust abstracts from the concrete transaction situation and from the concrete transaction partner. This form of trust is tied to institutional structures in a society, e.g. standards, certificates or labels. Zucker’s central point is that the replacement of process-based and characteristicbased trust – both of which were destroyed by exogenous instabilities in the nineteenth century – by institutionally based trust was critical for the successful industrialization of the United States. Putting these differentiations together, we may say that trust tied to a person can be generated by longterm transaction experiences or by his or her personal characteristics, while the trust tied to a system is generated by rules. Since institutional trust is independent of spontaneous individual preferences, it has also been termed ‘generalized trust’. Most works referring implicitly to Zucker’s concept assume that (only) institutional trust is the source of successful economic development. According to Khalil (1994: 340), for example, capitalist society needs institutional trust as ‘economic exchange becomes less intermingled with kinship and more based on formal contractual relationships [so that] the monitoring conducted by the kin members and the threat of ostracism almost vanish’. The same stance towards institutional trust can be found in the influential study by Knack and Kneefer (1997). They contrast high-trust environments (cultures with institutional trust) to low-trust environments (cultures that rely on personal trust), and describe the latter as environments in which transactions take place more among close friends and relatives than among strangers. They argue that high-trust cultures carry one of the major preconditions for economic success (see especially Knack and Kneefer 1997: 1256). Their conclusion that dense, horizontal networks do not reinforce trust fits into this line of argumentation (Knack and Kneefer
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
87
1997: 1284). Another common theory that draws on modernization literature is that personal trust loses its importance in advanced societies due to ever expanding anonymous markets. In the words of Tönnies, Gesellschaft gains in importance at the expense of Gemeinschaft (cited in Khalil 1994: 341). A lack of institutional trust induces economic stagnation and underdeveloped markets. Especially in the era of information and communication technologies, a lack of institutional trust may thus become a competitive disadvantage. Phrased more bluntly, in economic theory there is a favourable form of trust and a less favourable form of trust for economic development.27 In order to understand more clearly the implications of this research, a short recourse to institutional economics and its concept of path dependency is necessary. The concept is employed as a core concept in understanding institutional change, particularly in understanding why desirable reforms do not take place. The problem of institutional change becomes especially relevant when institutional arrangements are identified as harmful to economic development. Due to their embeddedness in societal structures and their complementarity to other rules, institutional change is difficult and slower than often deemed necessary. The adherence to established institutions and the resulting hesitancy towards reforms may be rational from an individual’s point of view, but as a collective process it may result in institutional lock-in. One of the first who accentuated the difficulty of institutional change is North (1993), who acknowledges at most the possibility of marginal reforms on the fringe.28 North concludes that for all economies that are not heirs to market-oriented belief systems, the transformation into a market economy is expected to be difficult: It takes much longer to evolve norms of behaviour than it does to create formal rules and for those economies without a heritage of such norms the reconstruction process is necessarily going to be long and the outcome very uncertain. (North 1993: 21) Other authors follow this argument, such as Eggertsson (1998) in his paper on ‘Limits to institutional reforms’. In a publication related to Japan and other Asian nations, Aoki and Hayami (2001) see formal and informal institutions as complementary to the degree that community norms such as personal trust become a prerequisite for economic exchange, but that such community norms can induce dysfunctions as well, since community trade tends to be small in volume and competition in the group becomes club-like and restricted. In other words, certain informal institutions can lead to institutional lock-in and hinder the development of markets. Their drive to analyse ‘what types of community relationships promote or deter market development’ (Aoki and Hayami 2001: xix) resembles very much the idea prominent in institutional economics discussed above – to differentiate between favourable and less favourable institutions. Finally, one can find, particularly in the
88
Cornelia Storz
German literature, strong scepticism towards change, which might partly stem from the negative experiences of the transformation processes in Eastern Europe.29 According to Eisenberg (1999), for example, the more homogenous the group is, the slower institutional change may occur. If one classifies Japan as a relatively homogenous culture,30 institutional change in its society is deemed to be extremely difficult. Earlier research on Japan, such as works concerning public policy or standardization policy, tends to support institutional path dependency theory (Pascha 2001; Storz 2003, 2005b). The negative experiences in developing and transformation countries have helped to sustain the path dependency tradition in institutional literature as a relatively deterministic concept. Groups without favourable informal institutions are restricted in their options to create better institutions. With the rise of new information and communication technologies making markets even more anonymous, institutional trust becomes an even more critical resource for a nation’s competitiveness and the lack of complementary informal institutions even more problematic.31 The lack of institutional trust in Japan and electronic commerce Institutional trust in a society is a precondition for the success of electronic commerce since societies in which it is not widespread will not readily accept a new market with newly established rules. Even market rules that have been successful in other countries need not be successful in the case of their transfer to another society if this important precondition is not given. Since informal institutions are fairly stable, we expect electronic markets to remain underdeveloped in the long run as well. Comparative empirical research has shown that there are indeed stark differences in the amount of institutional trust in Japan and the United States (Markus and Kitayama 1991; Yamagishi 1998). American respondents are more trusting in other people in general. They consider reputation to be more important than their Japanese counterparts. In contrast, the amount of institutional trust in Japan is low, while the amount of personal trust is high. Japanese respondents see more utility in dealing with others directly – a phenomenon that Befu (1989) has explained with the fact that interpersonal relationships are long-term oriented so that instrumental and personal aspects become more mixed. These insights correlate with Hofstede’s hypothesis that Japanese actors also tend to avoid uncertainties in the sense that they are more prone to risk aversion (see for its reception Kanbayashi 1999). These results may be somewhat astonishing since Japan has often been designated as a high-trust culture (see, for example, Casson 1993: 431), but this characterization is grounded in an insufficient differentiation of different forms of trust. Japan is a society in which there is a high level of personal trust and less institutional trust.
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
89
If these empirical results, on the one hand, and the thesis of incremental change discussed above, on the other hand, are correct, then it is not surprising that the Japanese electronic markets are underdeveloped. Public policies and private regulation are almost without effect since they require widespread institutional trust not found in Japan. The low success rates of business-to-consumer commerce in Japan can, in this view, be explained by the preference for personal trust. And personal trust cannot be transformed by appropriate rules into institutional trust in a mechanical way. Japanese actors simply will not put their trust into them. Yamagishi’s conclusions (1988) might even be correct in emphasizing that sanctions are needed more in Japan than they are in the United States in order to enforce commitment to rules. Indeed, the argument that the predominance of personal trust hinders the development of anonymous market transactions such as business-toconsumer commerce was grasped by leading Japanese think tanks in order to explain the slow growth of electronic commerce (Kitamura et al. 2000; Tagawa 2001).32
Learning processes and path dependency Entrepreneurial alertness and learning The previous sections have discussed how institutions may reduce uncertainty and that institutional trust is a necessary precondition for institutions to function well. This argument is based on the assumption that there are two different kinds of trust – institutional and personal. If one follows the argument that these two different forms of trust exist and if the concept of path dependency is correct, then institutional change is obviously restricted. Generating trust by introducing institutions, as suggested by Zucker (1986), becomes almost a tautology. At this point I would like to suggest a different approach. I depart from the view that informal institutions favouring market development, i.e. institutional trust, and those that favour communities, i.e. personal trust, are two different, alternative ways of organizing economic activities, and that economic development necessarily entails moving from personal to institutional trust. I see the common problem of many of the institutional approaches as residing in this viewpoint: that informal institutions are seen as given and that they perpetuate certain preferences that then lead to specific behavioural restrictions as well as specific rules for decision making. In this understanding, informal institutions are treated as exogenous determinants of individual behaviour. Nevertheless, a number of authors are increasingly acknowledging that this approach contains a number of deficiencies.33 In conjunction with neuro-biological theories and learning psychology it may perhaps be more useful to view informal institutions as internalized rules. Such internalization takes place by learning. The learning process is subjected both to exogenous control, such as sanctions
90
Cornelia Storz
or incentives, and endogenous control, which is initiated by the individual and carried out systematically. If internalization takes place by learning it seems questionable that rules persist practically unchanged after they are internalized, and such learning processes can challenge previously learnt rules. In order to understand the conditions under which and the ways in which re-learning takes place, it is helpful to follow the suggestion made by Vanberg and Buchanan (1994). They break informal rules down into their theory and interest components. The theory component is comprised of a validated finding, and the interest component entails a subjective evaluation. Since the theory component can be reflected upon in relation to its contribution to an expected outcome, there is an incentive for the individual to determine when and under which conditions established informal institutions are helpful in order to achieve this outcome. We can expect that individuals analyse the long-term outcomes of informal institutions with respect to their efficiency, i.e. whether and to what degree these institutions contribute to a long-term goal. A change in informal institutions can be expected if their effectiveness is perceived to be improved by the change over time.34 Since learning takes place in paths, re-learning also does not normally take place in radical bursts. It is more often the case that one learns in unspectacular, incremental steps in everyday life. Informal institutions, such as personal or institutional trust, can evolve in a slow, almost imperceptible process of change. In such a way, informal institutions undergo processes of change and may become locked out of given paths. But even if one is more sceptical about learning processes and tends to interpret informal institutions as stable, there is still another way of interpreting the existence of paths. If one understands paths as plasticine and considers the range of varieties in existing paths, then one may say that the existence of a path itself is less important. More important is the entrepreneur’s ability to find creative answers to (supposed) path restrictions. Such an approach renders the determinism that shines through in at least applied economic research questionable (Garud and Karnoe 2001; Storz 2005a, b). The role I attach to learning capacity and to entrepreneurial creativity does not deny that learning takes place on the basis of accumulated storages of experiences, and that certain informal institutions may restrict economic development if incentives for learning are not present. Nevertheless, the thesis of institutional rigidity loses some of its strength when we consider that internalized rules can be modified or even rejected through creativity and learning. The issue is then not so much the fact that certain informal institutions hinder a certain type of development as it is more the underlying learning processes in a society and the fact that they might be hindered. The availability of information and its credibility, public discourse about the ‘truth’ of rules and the subjective conviction that new rules would be
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
91
more desirable are central elements for successful reforms. The precondition for change, therefore, are not favourable institutions but rather the creation of an institutional framework that enables learning processes and that accommodates well-functioning feedback processes. Within this approach, options for reforms and change emerge. New solutions in Japanese electronic commerce Electronic commerce is a fairly recent form of individual economic transactions, so that it may be too early for an assessment of learning processes; even general statistics on business-to-consumer commerce began to be collected only a few years ago. Nevertheless, the following section will attempt to interpret the recent developments on the Japanese business-toconsumer market within the conceptual framework developed above. As mentioned earlier, the volume of business-to-consumer commerce in Japan is low. However, one turning point appears to be in the making by way of connecting the ‘old’ to the ‘new’ market. A unique form of business-toconsumer commerce has developed in Japan by which convenience stores have been converted into ‘windows of electronic commerce’ (Matsumoto 2000: 100). Japanese convenience stores, originally an invention from the United States, are small supermarkets with long business hours (90 per cent of them are open 24 hours a day) and a large number of small outlets at convenient, consumer-dense intersections. Since their introduction in the 1970s, they have become the most successful form of retailing in Japan, boasting more than 30,000 in number. In Japan’s urban areas, at least one, and usually two or three, are within walking distance. The largest companies are 7-Eleven, Lawson and Family Mart. Together they comprise more than 50 per cent of all outlets. Over the past few years they have developed two very dynamic electronic commerce strategies: (1) their own portals in electronic commerce and (2) service for other virtual shops. 1
In 2000, several convenience stores began to go online. 7dream.com, founded by 7-Eleven and seven partners including NEC, Sony, Mitsui and Nomura Research Institute, is one of the largest virtual stores in Japan and offers about 100,000 items. It also offers services in the areas of travel, digital photo printout, books and mobile phones. Another start-up of the 7-Eleven group entered into the meal delivery market (7meal.services). Lawson offers retail electronic commerce as well, also in cooperation with partners who are top leaders in their industry, such as Mitsubishi and Toyo Information System. Consumers can use easyto-select items in terminals installed in Lawson shops and pay at the shop. Smaller products are picked up at the shops while bulkier items are delivered to the customer’s home (Whipple 2000; Joffe 2001).
92 2
Cornelia Storz A second strategy is to offer a delivery and payment service to other virtual stores. Since their distribution network, their IT-based logistics and their demanding payment system are the best in Japanese retailing, convenience stores possess a competitive advantage in the logistical distribution of goods. Today, about 75 per cent of Internet shoppers reportedly pick up their purchases from stationary retailing (Ebusinessforum 2001). The idea of combining virtual stores with stationary retailing is so successful that even companies with high-priced items and no financial investments in convenience stores, e.g. Toyota, plan to incorporate business-to-consumer commerce via convenience stores. One plan is to sell more than half of the automobiles originally to be sold by Toyota’s own homepage Gazoo through convenience stores (Ebusinessforum 2001: 27).
Within the theoretical framework presented here, the second strategy is particularly interesting. Its uniqueness is grounded in the cooperation between business-to-consumer commerce and conventional retailing. One way to interpret this strategy would be to say that this is simply another case of strict path dependency. Business-to-consumer commerce organizers found a way to generate these new business models by relying on consumers’ personal relationships with well-known retailers, thereby circumventing restrictions associated with certain informal institutions. The trustworthiness of business-to-consumer commerce is strengthened as a high percentage of convenience stores are successors to traditional neighbourhood stores, and often the franchises are operated by the former. One could argue that personal trust has come to electronic commerce through the backdoor.35 The problem with this interpretation is that it continues to employ a static approach to economic development. Other approaches allow more room for creative potential and stress the importance of personal trust as a precondition from which institutional trust may derive. According to organization theory, for example, the sales staff in convenience stores can be interpreted as actors with special organization roles known as ‘boundary spanning roles’ (Loose and Sydow 1994). Even if they fulfil more the function of an institution than a person, they can help through their reliable behaviour to build trust into the system. Trust is thereby stabilized since consumers can fall back on established norms and schemes in which previous trustworthy relationships have not changed. As consumers conduct businessto-consumer commerce via stationary retailing, they learn the advantages of electronic commerce while positive feedback loops concerning the quality of products prove to them that a change in their behaviour may be in their interest. Such a dynamic perspective allows the possibility that institutional trust may result out of personal trust and may be built up incrementally to the benefit of new market opportunities. Personal and institutional trust are in this perspective neither contradictory nor merely complementary but dynamic and interactive.36 Even if a lack of institutional trust somewhat
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
93
restricted the development of electronic commerce in Japan in the beginning, the above-average growth rates of mobile commerce in business-to-consumer commerce may be an indicator that institutional trust is growing.37 Another indicator of the potential of this new retailing strategy lies in the fact that it has been successfully exported to other Asian countries which exhibit to some degree similar informal institutions. But even if one takes informal institutions as given, the case of Internetbased retailing provides a good example of the way in which presumed restrictions may trigger entrepreneurial creativity, which in turn can overcome the presumed restrictions. The case of Internet retailing via convenience stores thus closely resembles the development of the Japanese enterprise groups, which was a strategic response to formal institutional restrictions of underdeveloped markets of labour, capital and knowledge in the 1950s and 1960s. Here again, because of presumed restrictions, this new strategy could be generated and expanded upon. The coupling of business-to-customer commerce to convenience stores is not only interesting as a case of clever entrepreneurial alertness. It also strongly reminds us that economic development is always open. The fact that paths exist is not a problem per se if an adequate solution to the economic problem at hand can be found. It is too early to evaluate whether the solution of coupling business-to-customer commerce to stationary trade is indeed an adequate one, but it has clearly opened up new and innovative business possibilities.
Conclusion This chapter started with the assertion that electronic commerce – here the business-to-customer commerce form – is characterized by new consumer uncertainties that result from new information asymmetries. According to information economics, uncertainties can hinder the development of markets. Institutional answers to uncertainty are generally supplied by public policies and private regulation. Since private regulation may be more flexible and superior with regard to knowledge creation, it has gained much attention in recent years. In order to solve the problem of uncertainty, Japan has chosen to rely on private regulation, as has the United States. Mutual recognition agreements between American and Japanese certification organizations indicate that Japan’s mode of regulation should have the same quality as the mode in the United States. But, despite an elaborate and comprehensive institutional infrastructure, Japanese electronic markets are still underdeveloped compared to the United States. Why is this the case? I find one important explanatory factor in the different kinds of trust that exist in each of the two countries. Empirical research has shown that institutional trust is weak in Japan (as well as in other Asian countries), while personal trust is widespread. Since private regulation requires institutional trust in the sense that consumers must be able to assume that certified parties will comply with the rules and that defection is sanctioned,
94
Cornelia Storz
and because of the extent to which personal trust dominates in Japan, it is not surprising that private regulation did not lead to the expected results in Japan. Nevertheless, uncertainty and information asymmetries have been tackled in Japan in a different way. Japan as well as other Asian countries rely on personal solutions by incorporating anonymous Internet shopping into stationary retailing. Western countries have tended to accept the more anonymous shopping on the Internet without much ado. One could stop at this point and remain sceptical about the future development of electronic commerce in Japan. Such scepticism would be underscored by a certain stream in institutional economics in which restrictions to institutional change are stressed. But the conceptual framework presented here departs from this view since informal institutions are learnt and relearnt: the sales staff in convenience stores fulfil special organizational functions in boundary-spanning roles, thereby contributing to the build-up of institutional trust – specifically but not only in electronic commerce – through their trustworthy behaviour. Although it is too early to state that these business models are functional equivalents of the type of electronic commerce we find in the United States, Japanese convenience stores have clearly developed into focal points of business-to-consumer commerce. They function as distribution and payment centres, and have even exported the model into south-east Asian markets. The above-average increase of mobile commerce may be an indicator that the needed institutional trust is slowly beginning to develop. Moreover, entrepreneurs have an incentive to develop their own, creative solutions. Indeed, as I showed for Japan, innovative business models have been created exactly by drawing on pre-existent informal institutions. In directing attention to the option of unlocking, I do not deny that learning takes place in paths and that institutional lock-ins are possible. Pathdependency is a given fact. I merely emphasize that we should not underestimate an actor’s capacity to learn, nor should we disregard the leeway of options within paths. The selection of a certain analytical perspective tends to portray certain institutional developments in a more favourable light or – as the case may be with business-to-consumer commerce – in a more unfavourable light than reality might suggest. This contribution is meant as a possible starting point to discuss further other possible modes of economic development.
Notes 1 This contribution was supported by an invitation to conduct research at the Kansai University from September to October 2003, and I would like to thank the University for its generous hospitality, especially Professors Mitsuru Tanaka and Hirohiko Yasuki. During this research stay I was able to conduct several interviews with staff in institutions active in self-regulation, and the results have been incorporated into this chapter. Special thanks go to Kazuhiko Wakaizumi and Shogo Asanuma from ECOM and to Hideo Shindo from METI.
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
95
2 The validity of electronic signatures, a new problem caused by the disintegration of identity and authenticity, will not be touched upon here. Special laws regarding electronic signatures and private certification systems have been formulated in almost all OECD countries. 3 Private regulation is defined here as a set of rules formulated by the private sector, also called market solutions or private constitution. Some scholars prefer the term ‘self-regulation’. ‘Self’ refers to the fact that the rules are formulated by the actors themselves. ‘Regulation’ refers to the activity of the subjects, namely the formulation of suitable rules, their enforcement, and the control over their enforcement. 4 The appropriateness of private rules depends very much on the respective sector. In the case of broadband penetration, for example, private rules lead to very low penetration, which may be explained by the fact that federal regulation is often necessary in order to guarantee a common standard (see the contributions by Frieden; Knieps; Lageman, Rothgang and Scheuer; Schefcyzk in this volume). 5 For South Korea and Singapore, see Tagawa 2001. 6 Electronic commerce includes every type of electronic commerce that is conducted between different actors, independent of the medium (e.g. it includes mobile commerce as well). It can be carried out between firms and consumers, between firms and other firms, and between public institutions and firms and/or consumers. Since the economic problems arising with each of these types differ, this chapter focuses only on electronic commerce conducted between firms and consumers, or business-to-consumer commerce. 7 The systematic analysis of information problems started in the 1960s with the idea that both sides of the market do not have perfect information. Stigler (1961), who can be seen as the founder of the economics of information, discusses the consequences of information problems and demonstrates that information asymmetries may be in the interest of sellers. The theories of adverse selection and signalling (Akerlof 1970; Spence 1974) are formulated in this neoclassic tradition. 8 Since institutions restrict individual behaviour, a common interest in the solution of a problem is necessary before the institution can work. Only when a common interest can be established, in the sense that ‘it can be assumed that all individuals might agree to it’ (Kerber and Vanberg 2001: 67), may limits on individual freedoms be legitimized. 9 Screening is the process by which the uninformed consumer searches for information about products and services, while signalling refers to the information about product or service quality by the supplying actor. 10 Brand names can also be interpreted as signals. It can be argued that sanctioning here occurs internally since economic actors tend to stick to one particular brand on the basis of its reputation. 11 The dilution of rules is usually subsumed under the term ‘defection’. 12 Other problems associated with collectively formulated standards are the creation of less demanding rules and the restriction of competition. While neither can be further discussed here, both are highly relevant for the problem of technology clubs. For a general theory of clubs, see Buchanan 1965 and Olson 1971. For information on technology clubs, see Gilroy 1992 and Kuno 1998. 13 B2C stands for business-to-consumer commerce. 14 The problem of electronic signatures will not be discussed here, but the Japanese government relies on private regulation in this area as well. Private certification authorities provide so-called public key infrastructures, which require the government to recognize the formal equivalence of electronic and handwritten signatures (Keidanren 1999; Nikkei Communications 2000). A list of federal laws and corresponding administrative guides relating to online trade can be found in Horibe 2000. 15 For an overview, see CSLR 2000a: 3–6, 21–3 and 2000b: 21–3.
96
Cornelia Storz
16 Further international cooperation takes place in the APEC (Asia–Pacific Economic Cooperation), where an Electronic Commerce Task Force has been established (CSLR 2000b: 27–8). 17 The full name is ‘Guidelines Concerning the Protection of Computer Processed Personal Information in the Private Sector’. 18 In Japanese management literature, JIS Q 15001 is often seen as an entity within the international standards ISO 9001 (quality management standard) and ISO 14001 (environmental management standard), since together they outline a firm’s social responsibility. As in the case of ISO 9001 and ISO 14001, the single requirements of JIS Q 15001 are embedded in a strategy called PDCA (plan-docheck-action), which aims to permanently improve the company’s policy. Both internal and external audits are prescribed (see Horie 2002; Nagato 2003: 55, 136). For further details on this standard see CSLR 2000a: 30 and 2000b: 25–6 and Nagato 2003: 25, 138. 19 There are several different English translations for the JIPDEC certification system, including the System of Granting Privacy Marks, the Privacy Mark Award System, or the Privacy Seal Granting Program. JIPDEC receives tax abatements as an incorporated foundation (zaidan hôjin). This status requires annual approval by the corresponding ministry (Privacymark 2003a, 2003b, 2003c, 2003d; for Japanese associations see Schaede 2000). 20 The licensing fees at BBBOnline are graduated depending on the size of the enterprise. Total company sales of one million or less pay $50, one to five million pay $250 and more than five million pay $500 per year. 21 Information relayed through interviews. 22 Nevertheless, ECOM is also relatively closely associated with the METI since about 30 per cent of its staff is sent by JIPDEC to ECOM on a temporary basis. 23 In Europe, the volume is about $82 billion. The absolute volume for electronic commerce for 2003 lies at $674.6 billion in the United States, $170 billion in Germany, and $87.4 billion in Japan (NFO 2003: 235, 237, 241). Statistics present different absolute levels of business-to-consumer and business-to-commerce, but the trend is the same in all statistics: North America leads followed by Europe, while the volume of Japanese electronic commerce is much lower. 24 The United States is again in first place, South Korea in second and Germany in third. 25 Most of the procured goods in Japan are paid by bank drafts or postal giro. This form of payment is often cited as a barrier to electronic commerce. But the volume of electronic commerce in Germany, a country in which it is equally not very common to pay by credit card, is much higher than in Japan. Access costs for the Internet in Japan are indeed relatively high – an issue that has been discussed as Japan’s ‘backwardness in IT’ at length (Nihon okureta IT kakumei). It was hoped that this could be overcome by the so-called ‘e-Japan Strategy of 2001’. Nevertheless, the share of mobile commerce, which could be an important pillar of business-to-consumer commerce since its access costs are cheap, was – at 7.2 per cent of the overall business-to-consumer volume in 2000 – still low (METI 2001; MPHTP (Ministry of Public Management, Home Affairs, Posts and Telecommunications) 2003). 26 On this point, see for further details Mayer et al. 1995; Uslaner 1999; Alesina and La Ferrara 2002. 27 For further analysis, see Hardin 1998 and Rousseau et al. 1998. 28 For further case studies of transformation countries, see Sztompka 1995; for trust building in medieval Italy, see Putnam 1993. 29 Leipold (1997: 64) states that because of the necessity of reform policy to set even further into deeper social layers, there are no prospects for quick reform successes in Russia (see also Panther 1998). One central reason for this assumption lies in the reception of a certain strand of social psychology that stresses that
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
30 31
32
33
34 35
36
37
97
certain moral rules are learnt primarily in youth and are then more or less fixed (Nunner-Winkler 1992; Lind 1993). This statement does not negate important foreign cultural influences in Japan. In institutional economics the concept of path dependency is not necessarily static, and many scholars examine ways in which institutional change can be triggered. Nevertheless, most of the research tends to address the problems of path dependency and the impact of institutions on human behaviour while neglecting the questions of institutional lock-out and the genesis of new institutions. The case of credit cards is another example in point. One qualifying remark here is necessary. The history of private regulation enacted to reduce information asymmetries is quite young in Japan. Private regulation was often closed to corporatist structures both in the domestic as well as in the export market. In the domestic market agreements in the banking sector, within the steel industry in export markets and in the household appliances industry have all become wellknown examples. See Schaede 2000 for a general overview. For critical voices towards self-regulation, see Fujita 1999 and Nihon Keizai Seisaku Gakkai 2001. Institutional economics focuses more on the effect of institutions and less on their genesis. Eggertsson (1993) and Gäfgen (1983: 25) have already voiced this criticism. Both state that there is a growing consensus that a theory of institutional change requires a theory of the formation of value systems. If one adds this concept of effectiveness to the analytic framework, the change of preferences can be integrated into the economic model of behaviour as well. The volume of TV shopping is about tenfold that of business-to-consumer commerce, and TV shopping has shown growth rates of about 4.7 per cent (Plate 2000: 64; Nikkei Weekly 2003a). The case of TV shopping is interesting, since important personal interaction still exists, albeit at a physical distance (e.g. by personal presentation of goods even by company presidents). The idea of combining electronic commerce with personal relationships is not new and has been applied successfully in other areas as well: about one-third of catalogue shopping payments takes place in neighbourhood shops. The utilization of personal trust in order to create institutional trust has been observed in other industries as well, for example, in the insurance sector where access to the customer takes place via insurance brokers (see Loose and Sydow 1994: 183–4). The increasing sums of purchase in business-to-consumer commerce can also be interpreted as learning curves. People who have been using shopping sites for less than six months spend about 6,900 yen in Japan, whereas the sum for people with experience in this area for more than five years is much higher at 18,600 yen (Nikkei Weekly 2003b).
Bibliography Akerlof, G. A. (1970) ‘The market for “lemons”: quality uncertainty and the market mechanism’, The Quarterly Journal of Economics, 84 (3): 488–500. Alesina, A. and La Ferrara, E. (2002) ‘Who trusts others?’, Journal of Public Economics, 85: 207–34. Aoki, M. and Hayami, Y. (2001) ‘Introduction: communities and markets in economic development’, in Aoki, M. and Hayami, Y. (eds) Communities and Markets in Economic Development, Oxford: Oxford University Press, pp. xv–xxiv. Arrow, K. J. (1974) The Limits of Organisation, New York: Norton. Aulakh, P. S., Kotabe, M. and Sahay, A. (1996) ‘Trust and performance in crossborder marketing partnerships: a behavioral approach’, Journal of International Business Studies, special issue: 1005–32.
98
Cornelia Storz
BBBOnline (2000) New Online Privacy Protection Tool to Transcend Borders: BBBOnline and privacy seal program (JIPDEC) announces plans for joint online privacy seal, accessed from http://www.bbbonline.org./about/press/2000/051800.asp on 5 May 2003. Beales III, J. H. (1998) ‘Licensing and certification systems’, in Newman, P. (ed.) The New Palgrave Dictionary of Economics and the Law, vol. II, London: Macmillan pp. 578–81. Befu, H. (1989) ‘A theory of social exchange as applied to Japan’, in Sugimoto, Y. and Mouer, R. E. (eds) Constructs for Understanding Japan, London: Kegan Paul International, pp. 39–66. Bolton, G. E., Katok, E. and Ockenfels, A. (2003) Bridging the Trust Gab in Electronic Markets, accessed http://ockenfels.uni-koeln.de/pubs.php?w=all on 20 August 2004. Buchanan, J. M. (1965) ‘An economic theory of clubs’, Economica, 32: 1–14. Campbell, A. J. (1998–1999) ‘Self-regulation and the media’, Federal Communications Law Journal, 51 (3): 711–71. Casson, M. (1993) ‘Cultural determinants of economic performance’, Journal of Comparative Economics, 17: 418–42. CSLR (Center for Social and Legal Research) (2000a) Guide to Privacy and Data Protection Program in Japan, Washington DC: CSLR. CSLR (2000b) Guide to E-Commerce and Privacy Developments in Japan, Washington DC: CSLR. Doi, E. (2001) Recent Developments on Internet Related Laws and Regulations in Japan, accessed at www.nzls.org.nz/conference/pdfpercent20files/DoiF12.pdf on 19 September 2003. Ebusinessforum (2001) Seven-Eleven Japan: blending e-commerce with traditional retailing, 24 May, accessed at www.ebusinessforum.com/index.asp?layout=rich_ story&doc_id=3544&categoryid=&channelid=&search=Sevenpercent2DEleven+ Japan on 1 April 2004. ECOM (Electronic Commerce Promotion Council) (2001a) Electronic Commerce: onrain mâku seido ni tsuite (Electronic Commerce: about the online mark system), accessed at www.ecom.jp/onlinemark/index.html on 15 April 2003. ECOM (2001b) ECOM Newsletter No. 16, 12 October, accessed at www.ecom.or.jp/ ecom_e/latest/newsletter_no16.htm on 15 April 2003. ECOM (2003) ECOM Newsletter No. 31, 29 January, accessed at www.ecom.or.jp/ ecom_e/latest/newsletter_no31.htm on 15 April 2003. Eggertsson, T. (1993) ‘Mental models and social values: North’s institutions and credible commitment’, Journal of Institutional and Theoretical Economics (JITE), 149 (1): 24–8. Eggertsson, T. (1998) ‘Limits to institutional reforms’, Scandinavian Journal of Economics, 100 (1): 335–57. Eisenberg, A. (1999) Die Lösungen sozialer Dilemmata und der Wandel informeller Institutionen, Discussion Paper no. 04–99, Jena: Max-Planck-Institut zur Erforschung von Wirtschaftssystemen. Frey, B. S. (1997) Markt und Motivation. Wie ökonomische Anreize die (Arbeits) Moral verdrängen, Munich: Vahlen. Fujita, S. (1999) ‘ISO kankyô kokusai kikaku ninshô shutoku kigyo ni taisuru ankêto chôsa hôkoku. Kikyô no kankyô taisaku to kankyô risuku manejimento’ (A Survey of Companies Certified under the International Environmental ISO Standard. Corporate Management of Environment and Risk), Hokengaku Zasshi, 567: 94–115.
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
99
Gäfgen, G. (1983) ‘Institutioneller Wandel und ökonomische Erklärung’, in Boettcher, E., Herder-Dorneich, P. and Schenk, K.-E. (eds) Jahrbuch für Neue Politische Ökonomie, vol. 2, Tübingen: J. C. B. Mohr (Paul Siebeck), pp. 19–49. Garud, Raghu and Peter Karnoe (2001) Path Dependence and Creation, London: Lawrence Erlbaum Associates. Giddens, A. (1990) The Consequences of Modernity, Oxford: Polity Press. Gilroy, B. M. (1992) International Production, Technology Clubs and Governments, Discussion Paper No. 65, Department of Economics, University of St Gallen: St Gallen. Hardin, R. (1998) ‘Trust’, in Newman, P. (ed.) The New Palgrave Dictionary of Economics and the Law, vol. III, London: Macmillan, pp. 623–8. Horibe, M. (2000) ‘Denshi-shô torihiki to puraibashii’ (E-Commerce business and Privacy), Jurisuto, 1183 (8): 77–85. Horibe, M. (2002) ‘E-komaasu ni okeru hoshô to kansa no gainen wakugumi’ (Security in E-commerce and the framework of a concept for inspections), Hitotsubashi Ronso, 128 (4): 470–85. Joffe, H. (2001) ‘Japanese business models for electronic commerce – laying the foundation of a ubiquitous networking infrastructure with mobile phones and convenience stores’, Vierteljahresheft zur Wirtschaftsforschung, 70 (4/2001): 546–70. Kanbayashi, N. (1999) ‘Jôhô gijutsu shisutemu no sekkei to bunka kôzô’ (The design of the system of information technology and the construction of culture), Kokumin Keizai Zasshi, 177 (5): 39–51. Keidanren (1999) Denshi Torihiki no Suishin ni Kansuru Teigen. III. Kakuron (Suggestions for the promotion of E-Commerce. III. The arguments), accessed at www.keidanren.or.jp/japanese/policy/pol240/part3.html on 15 April 2003. Kerber, W. and Vanberg, V. (2001) ‘Constitutional aspects of party autonomy and its limits – the perspective of constitutional economics’, in Grundmann, S., Kerber, W. and Weatherill, S. (eds) Party Autonomy and the Role of Information in the Internal Market, Berlin/New York: Walter de Gruyter, pp. 49–79. Khalil, E. L. (1994) ‘Trust’, in Hodgson, G. M., Samuels, W. J. and Tool, M. R. (eds) The Elgar Companion to Institutional and Evolutionary Economics. L-Z, Aldershot: Edward Elgar, 339–45. Kitamura, Y., Ôtani, A. and Kawamoto, T. (2000) Denshi-shô Torihiki no Genjô to Kadai: atarashî chûkai-gyô no shussan to shinrai keisei (The present condition and the tasks of e-commerce business: birth of new intermediation-industries and the development of trust), Discussion Paper No. 2000-J-13, IMES (Institute for Monetary and Economic Studies, Bank of Japan), Tokyo. Knack, S. and Keefer, P. (1997) ‘Does social capital have an economic payoff? A cross-country investigation’, The Quarterly Journal of Economics, November: 1251–88. Kuno, A. (1998) ‘Kokusai hyôjun no keizai bunseki’ (Economic analysis of international standards), SRIC Report, 3: 49–64. Lane, C. and Bachmann, R. (1996) ‘The social construction of trust: supplier relations in Britain and Germany’, Organization Studies, 17 (3): 365–95. Leipold, H. (1997) ‘Der Zusammenhang zwischen gewachsener und gesetzter Ordnung: einige Lehren aus den postsozialistischen Reformerfahrungen’, in Cassel, D. (ed.) Institutionelle Probleme der Systemtransformation, Berlin: Duncker & Humblot, pp. 43–68.
100
Cornelia Storz
Lind, G. (1993) Moral und Bildung. Zur Kritik von Kohlbergs Theorie der moralischkognitiven Entwicklung, Heidelberg: Asanger. Loose, A. and Sydow, J. (1994) ‘Vertrauen und Ökonomie in Netzwerkbeziehungen – strukturationstheoretische Betrachtungen’, in Sydow, J. and Windeler, A. (eds) Management interorganisationaler Beziehungen. Vertrauen, Kontrolle und Informationstechnik, Opladen: Westdeutscher Verlag, pp. 161–93. Markus, H. R. and Kitayama, S. (1991) ‘Culture and the self: implications for cognition, emotion and motivation’, Psychological Review, 98 (2): 224–53. Matsumoto, To. (2000) ‘Konbini daikin shûnô – ukewatashi sâbisu’ (‘Convenience stores’ service: cash pay and handing over of goods’), Nikkei NetBusiness, January: 98–103. Matsumoto, Ts. (1998) The Development and Future Challenges of Self-Regulation in Japan, with Special Regard to Electronic Commerce, accessed at www.law. kyushu-u.ac.jp/luke/ecommerce.html on 19 September 2003. Mayer, R. C., Davis, J. H. and Schoorman, F. D. (1995) ‘An integrative model of organizational trust’, Academy of Management Review, 20: 709–34. Michael, D. C. (1995) ‘Federal agency use of audited self-regulation as a regulatory technique’, Administrative Law Review, 47 (2): 171–253. METI (Ministry of Economy, Trade and Industry) (2001) Heisei Jûni-Nendo Denshishô torihiki ni Kansuru Ichiba Kibo, Jittai Chôsa ni tsuite, accessed at www. meti.go.jp/kohosys/press/0001317/0/0201ecom1–1-html on 5 September 2003. MPHPT (Ministry of Public Management, Home Affairs, Posts and Telecommunications) (2003): Jôhô Tsûshin Hakushô (White paper on Information and Communications), accessed at www.johotsusintokei.soumu.go.jp/whitepaper/ja/ h15/pdf/index.html on 06 April 2004 (short English version available at www. johotsusintokei.soumu.go.jp/whitepaper/eng/WP2003/2003-index.html.) Mitsubishi Research Institute (2002) ‘Japan’, in JETRO (ed.) Report on ‘Asia-Pacific B2C E-Commerce Legal Framework’, January: 299–330, accessed at www.jetro.go. jp/ec/e/stat/surveys/b2c/index.html on 20 April 2004. Nagato, N. (2003) Yoku Wakaru Puraibashii Mâku (Understanding privacy mark), Tokyo: Nihon Jitsugyô. NFO (2003) Monitoring Informationswirtschaft. 6. Faktenbericht 2003, accessed at www.nfoeurope.com/ib/Newsitem.cfm?lan=en&ObjectId=E2A82F11-E282–4202BBA422E19776630C on 6 April 2004. Nihon Keizai Seisaku Gakkai (Japan Economic Policy Association) (2001) ‘Nijûisseiki no Nihon no saisei to seido tenkan – kankyô seisaku’ (Reforms and institutional change in Japan in the 21st century – Environmental Policy), Nihon Keizai Seisaku Gakkai Nenpô (The Annual of Japan Economic Policy Association), 49: 21–6. Nikkei Communications (2000) ‘Denshi-shô torihiki wo sasaeru PKI no dônyû-hô’ (The methods of introducing an e-commerce supporting PKI), Nikkei Communications, 19 June 2000: 202–7. Nikkei Weekly (2003a) ‘Consumers take to TV, online shopping’, The Nikkei Weekly, 10 November: 20. Nikkei Weekly (2003b) ‘A world of content at their fingertips’, The Nikkei Weekly, 15 December: 22. North, D. C. (1993) ‘Institutions and credible commitment’, Journal of Institutional and Theoretical Economics (JITE), 149 (1): 11–23. Nunner-Winkler, G. (1992) ‘Zur moralischen Sozialisation’, Kölner Zeitschrift für Soziologie und Sozialpsychologie, 44: 252–72.
Private solutions to uncertainty 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
101
Olson, M. (1971) Logic of Collective Action: Public Goods and the Theory of Groups, Cambridge MA: Harvard University Press. OECD (Organisation for Economic Co-operation and Development) (1999) Inventory of Instruments and Mechanisms Contributing to the Implementation and Enforcement of the OECD Privacy Guidelines on Global Networks, accessed at www.oecd.org/ dataoecd/12/54/2092454.pdf on 14 April 2004. OECD (2001) Electronic Commerce, accessed at www.oecd.org/document/29/ 0,2340,en_2649_201185_2346205_1_1_1_1,00.html on 6 April 2004. Panther, S. (1998) ‘Historisches Erbe und Transformation’, in Wegner, G. and Wieland, J. (eds) Formelle und Informelle Institutionen: Genese, Interaktion und Wandel, Marburg: Metropolis, pp. 211–52. Pascha, W. (2001) ‘Ordnungspolitik in Japan? – Zur möglichen Rolle von Regelbindung und unabhängigen Agenturen’, in Bosse, F. and Köllner, P. (eds) Reformen in Japan, Hamburg: Institut für Asienkunde, 337, pp.167–192. Plate, P. A. (2000) ‘Elektronischer Handel in Japan’, Japan Aktuell. Wirtschaft, Politik, Gesellschaft, 1/00 (February): 61–71. Privacymark (2003a) Nintei Jigyô-Sha no Kojin Jôhô ni Kakaru Jiko nado ni tsuite (Certification companies and violation of private data use), accessed at www. privacymark.jp/pr/20021210.html on 11 September 2003. Privacymark (2003b) Diagram of accreditation and certification processes in ecommerce, accessed at www.privacymark.jp/pr/19980325a.gif on 18 September 2003. Privacymark (2003c) Yôgo-Shû (Glossary), accessed at http://privacymark.jp/misc/ glossary.html on 18 September 2003. Privacymark (2003d) Puraibashî Mâku Fuyo Nintei Shitei Kikan (Certification institutes für privacy), accessed at www.privacymark.jp/list/dlist on 18 September 2003. Putnam, R.D. (1993) Making Democracy Work: civic traditions in modern Italy, Princeton NJ: Princeton University Press. Rousseau, D. M., Sitkin, S. B., Burt, R. S. and Camerer, C. (1998) ‘Not so different after all: a cross-discipline view of trust’, Academy of Management Review 23 (3): 393–404. Schaede, U. (2000) Cooperative Capitalism: self-regulation, trade association, and the antimonopoly law in Japan, Oxford: Oxford University Press. Spence, A. M. (1974) Market Signalling: information transfer in hiring and related screening processes, Cambridge MA: Harvard University Press. Stigler, G. J. (1961) ‘The economics of information’, Journal of Political Economy, 64 (3): 213–25. Storz, C. (2003) ‘Globalisierung, Technik, Normen – Warum weichen japanische Unternehmen von internationalen Normen ab?’, in Deutsches Institut für Japanstudien (ed.) Japanstudien 15, Munich: iudicium, 219–46. Storz, C. (2005a) ‘Private Regulierung aus institutionenökonomischer Sicht. Das Beispiel Japan’, in Pascha, W. and Storz, C. (eds) Wirkung und Wandel von Institutionen: Das Beispiel Ostasien, Stuttgart: Lucius & Lucius, pp. 199–228. Storz, C. (2005b) ‘Standardization and the convergence of production systems’, in Pascha, W. (ed.) Systemic Change in the Japanese and German Economies: convergence and differentiation as a dual challenge, London: Routledge, pp. 203–30. Sztompka, P. (1995) ‘Vertrauen: Die fehlende Ressource in der postkommunistischen Gesellschaft’, Kölner Zeitschrift für Soziologie und Sozialpsychologie, special issue 35 (Politische Institutionen im Wandel): 254–76.
102
Cornelia Storz
Tagawa, Y. (2001) ‘E-komaasu hatten no kongen-teki mondai wo kangaeru: Ichiba ni okeru shinrai’ (The critical issue for e-commerce: trust in the market), InfoCom Review, 25: 4–16. Uslaner, E. M. (1999) ‘Trust but verify: social capital and moral behaviour’, Social Science Information, 38 (March): 29–56. Quoted version accessed at www.bsos. umd.edu/gvpt/uslaner/verify.pdf on 5 April 2004. Vanberg, V. and Buchanan, J. N. (1994) ‘Interests and theories in constitutional choice’, in Vanberg, V. (ed.) Rules and Choice in Economics, London: Routledge, pp. 167–77. Whipple, C. T. (2000) Japan Rushes to Catch Up on the Internet, accessed at www.iht.com./articles/3720.html on 31 March 2004. Yamagishi, T. (1988) ‘The Provision of a sanctioning system in the United States and Japan’, Social Psychology Quarterly, 51 (3): 265–71. Yamagishi, T. (1998) Shinrai no Kôzô: kokoro to shakai no shinka gêmu (The structure of trust: the evolutionary games of mind and society), Tokyo: Tokyo Daigaku. Yamagishi, T. and Yamagishi, M. (1994) ‘Trust and commitment in the United States and Japan’, Motivation and Emotion, 18 (2): 129–66. Zucker, L. G. (1986) ‘Production of trust: institutional sources of economic structure, 1840–1920’, Research in Organizational Behaviour, 8: 53–111.
Glossary Certification Agency Certifying and Auditing Organisation for the Privacy Mark Consulting Centre for Consumers Incorporated Foundation In-house Data Protection Manager Japan Accreditation Board of Conformity Assessment Japan’s ‘Backwardness in IT’ Japan Information Processing Development Center Mutual Recognition Agreement Privacy Mark System Committee Specific Commercial Transaction Act
Puraibashî Mâku Fuyo Shiteikikan Puraibashî Mâku Fuyo Kikan Shôhisha Sôdan Madoguchi Zaidan Hôjin Kojin Jôhô Hogo Kanri Sekininsha Nihon Tekigôsei Nintei Kyôkai Nihon Okureta IT Kakumei Nihon Jôhô Shori Kaihatsu Kyôkai Sôgô Shônin Puraibashî Mâku Seido Iinkai Tokuteishô Torihikihô
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
6
Institutional conditions for achieving effective implementation of ICT Rob Frieden
Introduction Information and communications technologies (ICT) can effectively stimulate a nation’s economy and the welfare of its citizens (OECD (Organisation for Economic Co-operation and Development) 2002).1 Efficient information age infrastructures enhance productivity (Grace et al. 2004) by providing intelligent networks that can efficiently handle converging voice, data and electronic commerce applications (Hukill et al. 2000). These infrastructures provide a comparative advantage in ‘knowledge-based’ (OECD 1995: 3)2 industries, which include such diverse fields as data processing, banking, insurance, management and technical consulting, travel planning, customer relations management, business logistics and others. With an increasingly global economy enhanced by fewer trade barriers and the quest by companies to find new growth opportunities, substantial incentives exist for public and private players to leverage comparatively greater competency in information and communication markets domestically and abroad (Zhen-Wei Qiang et al. 2003).3 Curiously, the track record for ICT implementation achieved by individual companies and nations does not always correlate with other indicators of success in trade, development and quality of life. The United States, for example, has excelled in a number of information industries, including public sector leadership in developing the Internet and private sector success in electronic commerce, and other ICT markets such as computers, software and integrated circuits (Leiner et al. n.d.). Nevertheless, observers across the political and social spectrum have roundly criticized the state of broadband network development (Grant and Latour 2003)4 in the United States (Ferguson 2002; Morgan 2002).5 Comparatively poor deployment of broadband network access juxtaposes with far greater success achieved by other nations, including ones with no comparatively greater prior success in ICT development and with fewer financial resources than the United States (Hopkins 2004). The International Telecommunication Union (ITU) reported that as of 1 January 2006, the five top nations for broadband network market penetration
104
Rob Frieden
were Iceland, Korea, the Netherlands, Denmark and Hong Kong. (ITU 2006). The ITU ranked the United States sixteenth in broadband penetration. The OECD reported that at year end 2004 the top market penetration for member nations occurred in Korea, the Netherlands, Denmark, Iceland and Canada, with the United States ranked twelfth (OECD 2005). One might infer that comparatively poor, new, telecommunications infrastructure development (FCC (Federal Communication Commission) 2004)6 in the United States and in other developed economies would adversely affect overall ICT development and global marketplace success. Indeed, some stakeholders seeking more aggressive governmental support and regulatory relief claim that the United States has forgone billions of dollars in lost business revenues and productivity gains (Pociask n.d.).7 These assertions make sense as national and private investments in ICT have a multiplier effect that accrues individual and societal benefits well in excess of the amount invested (New Zealand Trade and Enterprise n.d.; Digital Opportunity Initiative 2001).8 Aside from the obvious geographical and demographic advantages accruing to small nations with large urban populations, broadband telecommunications network development has become a national priority for many nations. Governments in many developed and developing nations have organized a cohesive and comprehensive strategy for stimulating capital investment in ICT infrastructure, and for expediting the deployment of ICT services in ways US public and private sector stakeholders have not embraced. Such efforts have accrued ample benefits, including lower broadband access, higher broadband market penetration and enhanced opportunities to use the Internet for individual and corporate benefits. For example, the ITU reports that in 2002, Japanese consumers paid $0.09 per 100 kilobits per second of broadband access compared to $3.53 in the United States (ITU 2003a). This wide disparity exists in part because Japanese broadband service providers have made the investment necessary to accommodate robust demand and to achieve economies of scale. This chapter will identify the institutional conditions in legislative, regulatory and business forums that help achieve success in developing faster, better, smarter, cheaper and more convenient broadband services. The chapter will concentrate on strategies that have worked well in Canada, Japan and Korea, as well as the institutional constraints that have handicapped broadband development in the United States. The chapter uses broadband development as a proxy for considering how institutions can achieve greater efficiency and effectiveness in ICT development, despite their vastly different geographical, political and economic conditions.
The importance of ICT incubation Both developed and developing nations recognize that ICT provides an effective opportunity to improve national living standards through enhance-
Institutional conditions for implementation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
105
ment of productivity and accrual of other efficiency gains (Hanna 1994; Zhen-Wei Qiang et al. 2003).9 Few would dispute that telecommunications and information processing technologies serve as powerful agents for economic and social development by improving access to information, enhancing trade in commodities and services, reducing costs and improving efficiency: ICT can help enhance the working of markets and reduce transaction and coordination costs within and across organizations. This is of particular relevance to developing countries where transaction costs are very high because of logistical problems. ICT applications can enable improvements in productivity and quality in a number of sectors . . . such as agriculture, manufacturing, infrastructure, public administration, and services such as finance, trade, distribution, marketing, education and health. (Sein and Harindranath 2004, p. 18) Nations must continually improve ICT innovation, incubation and exploitation because an increasingly integrated global economy can quickly erode a nation’s comparative advantages, particularly ones prone to volatility resulting from technological innovation. For example, ICT provides developing nations with greater opportunities to accelerate their rise along a technology development learning curve through technology transfer and foreign direct investment. While ICT first might generate threats to employment in developed nations through outsourcing, it might subsequently jeopardize wealth generation in knowledge industries as emerging nations establish their own research and product development prowess. When developing nations wean themselves of dependency on developed nations’ patents and innovations, the upside revenue generating opportunities become more contestable among all nations. China, for example, has quickly evolved from providing cheap labour for component assembly, e.g. cellular radiotelephones, to a nation that can challenge developed nations on the intellectual property and standards that next generation equipment will use. In the span of a few years, Chinese companies have increased the amount of value they contribute to a product and, in turn, the financial returns accruing for such an effort. Chinese mobile telephone manufacturers initially assembled handsets for sale within the country. Soon these companies provided world-class quality assurance in addition to cheap labour, so that their assembled handsets rivalled anything offered in the global marketplace. Not content with low-margin assembly work, some manufacturers have collaborated with the Chinese government, technology parks and universities to develop the intellectual property needed for next generation mobile telephones. Over a short period of time, Chinese companies have migrated from ICT original equipment
106
Rob Frieden
manufacturers for other companies, to ICT royalty paying ventures marketing their own equipment, to ICT innovators possible soon to seek royalty payments from other manufacturers (Yan n.d.).10 ICT development presents both challenges and opportunities to all nations. Developing nations no longer face the inevitability of having to organize their economies primarily for the benefit of developed countries that provide the market demand for cheaply produced products. Developed nations, for their part, can no longer consider technology transfer as largely a one-sided transaction that expands market penetration without risk of lost markets in the future.
ICT incubation by government Regardless of political and economic philosophy, national governments have significant positive roles in successful ICT development. Strategies have included an expansive governmental role in several areas, including: • • • • • • • •
developing a vision and strategy; promoting digital literacy, i.e. the ability to use digital technologies to pursue information, communications and entertainment interests; investing in infrastructure, aggregating demand and serving as an anchor tenant; fostering facilities-based competition of telecommunications and information processing infrastructures; creating incentives for private investment and disincentives for litigation and other delay tactics; offering electronic government services, including healthcare, education, access to information job training and licensing; promoting universal service through subsidies and grants; and revising and reforming governmental safeguards to promote a high level of trust, security, privacy and consumer protection in ICT services, including electronic commerce.
Nations pursuing successful ICT development strategies do not appear to quibble about whether government should meddle in areas that the private sector possibly could manage exclusively. However, one person’s view of government stewardship might come across to another as ‘industrial policy’, centralized management by the public sector and pre-emption of private initiatives. Successful ICT incubation appears to require government involvement, albeit with a light hand that stimulates and rewards investment, reduces unneeded regulatory scrutiny, and promotes global marketplace attractiveness without ‘tilting the competitive playing field’ in favour of one technology or company and without foreclosing market-driven industries.
Institutional conditions for implementation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
107
ICT incubation in the United States Curiously, even as the United States severely lags in broadband market penetration, this nation has achieved global supremacy in other ICT markets in part through the successful partnership of the public and private sectors. The United States model for ICT development favours entrepreneurism and private enterprise coupled with direct and indirect financial support by government primarily through early stage incubation. The Internet, for example, originated as a collaboration of government agencies and universities under the auspices of the Defense Advanced Research Projects Agency, a branch of the United States Defense Department. The Internet later evolved under loose management and financial support from the National Science Foundation with the Department of Commerce administering domain name registration. While the US government later eliminated direct financial underwriting when it privatized the Internet backbone, few would argue that early underwriting and anchor tenancy exemplified effective and successful government incubation of ICT. In the United States, governmental financial underwriting of Internet development had a short time span due to an institutional predisposition against government management of commercial markets and the perception by entrepreneurs that high monetary rewards justified risk-taking. In stark contrast to the absence of broadband incubation, the US government got involved early, but calibrated a timely exit strategy when a critical mass of private resources had developed. The government was able to make a quick exit because venture capital could readily replace taxpayer financed investment, research and development. Additionally, a well-developed marketplace for lawyers, accountants, consultants and entrepreneurs made it possible for private risk taking. A favourable tax climate ensured that ample rewards provided incentives for entrepreneurship. High technology hotbeds, such as Silicon Valley, California, demonstrate a largely private orientation to ICT development in the United States. In an assessment of what makes Silicon Valley a high technology development success, the authors of The Silicon Valley Edge (Lee et al. 2000) suggest ten factors: 1
2
3
favourable rules of the game – laws, regulations and conventions for securities, research and development, taxes, accounting, corporate governance, bankruptcy, immigration and development designed to support entrepreneurship and risk taking; knowledge intensity – the region has achieved a critical mass of ideas for new products, services, markets and business models. Silicon Valley serves as a magnet for entrepreneurs, educators, venture capitalists and people with vision; a high quality and mobile workforce – talented, educated and motivated people seek to make a home and a fortune in the region;
108
Rob Frieden
4
a results-oriented meritocracy – talent and ability accrue rewards in Silicon Valley without regard to race, ethnicity and age; 5 a climate that rewards risk-taking and tolerates failure – the region supports a high risk/high reward calculus, but also makes it possible for entrepreneurs who have experienced failure to regroup and try again; 6 an open business environment – the region supports robust competition but also knowledge sharing. This win/win environment results from the frequent formal and informal interactions among people with similar interests and objectives. Networking and relationships matter as much as technological innovations; 7 universities and research institutes that interact with industry – major universities such as Stanford foster exchanges among academics and entrepreneurs; 8 collaborations among business, government and non-profit organizations – the region houses universities, trade associations, labour councils, service organizations and companies, all of which collaborate and network with an eye on a successful future; 9 a high quality of life – despite traffic congestion, soaring housing prices, a relentless work pace and recent power outages, Silicon Valley offers proximity to open spaces and urban amenities; 10 a specialized business infrastructure – the region provides access to specialists needed for economic development, including consultants, lawyers, venture capitalists and executive recruiters. ICT incubation in the United States has achieved great success, in part thanks to governmental involvement. One should not discount the effect of early government financial involvement coupled with ongoing financial benefits accrued through favourable tax treatment and other financial incentives, e.g. tax holidays, revenue repatriation, infrastructure improvements and favourable immigration policies.
ICT development failures in the United States The fact that the United States lags significantly in broadband infrastructure development provides a stark contrast to the success story outlined above. Several legislative and regulatory initiatives have failed to achieve the intended results, or have backfired. While the United States may lead in technology incubation at technology parks and in regions such as Silicon Valley, the nation substantially lags in the increasingly essential, first and last kilometre access to broadband services and the Internet. This failure juxtaposes with this nation’s leadership in development of the Internet and Internet-mediated services, including electronic commerce. The failure of the United States to develop best-in-class broadband infrastructure results in part from lack of investment in new ICT technology by incumbent ventures such as telephone and cable television companies.
Institutional conditions for implementation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
109
Most market entrants in telecommunications markets concentrated on serving long-haul transmission markets while opting to rely on incumbents to provide first and last kilometre access at promotional prices as required by law. The Telecommunications Act of 1996 (Frieden 1997; United States of America, 1996) provided a legislative quid pro quo for incumbent Bell Operating Companies: the authority to provide long-distance toll telephone services in exchange for providing local exchange access based not on historical technology deployment costs, but on forward-looking best practices and new technology costs, i.e. theoretical costs of installing the infrastructure needed to provide access to local exchange facilities using the latest and cheapest technology (Frieden 2003a). While the incumbent Bell Companies welcomed the opportunity to generate new, long-distance telephone service profits, they objected as ‘confiscatory’ the duty to offer access to the local exchange network at rates well below what they considered cost and what they would demand in commercial negotiations (Baumol and Merrill 1998; Spulber and Yoo 2003). The Telecommunications Act of 1996 did not stimulate the development of viable, local, exchange service competition and the upgrading of local networks to provide broadband, high-speed data services (Dibadj 2003). The Bell Operating Companies refused to make the necessary investments based on the view that they should not have to continue subsidizing competition, particularly in the light of the fact that many new competitors did not appear inclined ever to migrate from reselling Bell network capacity to building and operating their own local facilities. Moreover, both incumbents and newcomers suffered heavy financial losses as a result of a severe reversal in the markets for anything relating to the Internet. The bursting of the dotcom bubble shifted investor sentiment from irrational exuberance to extraordinary pessimism. Investment bankers quickly moved from supporting acquisition of market share to requiring evidence of near-term profitability, thereby making capital investment contingent on largely unachievable short-term results. In addition, the proliferation of operating standards, particularly in wireless services, had fragmented telecommunications equipment markets making it difficult for any one company or technology to reach a critical mass. Consensus on operating standards can occur as a result of government stewardship, market forces or stakeholder collaboration. In the United States none of the three options has occurred, thereby causing a slow down in the commercial rollout of some ICT technologies and adoption by consumers. The combination of market downturn, legislation failure and lack of consensus on operating standards has removed much of the incentive for risk taking and investment, even as the need for network upgrades proved essential for the evolution of high-speed broadband ICT services. Stakeholders appeared more intent on competing in the courtroom than in the marketplace. The incumbent Bell Operating Companies made infrastructure investment contingent on securing massive, regulatory liberalization that,
110
Rob Frieden
if implemented, might result in the establishment of a shared monopoly among telephone and cable television companies without significant government oversight. Broadband network development in the United States already has begun to accelerate as the overall economy improves and as the Bell Operating Companies succeed in securing regulatory relief, including forbearance from having to comply with requirements specified in the Telecommunications Act of 1996. However, the potential for much faster and earlier rollout of new ICT technologies existed in the United States without government legislators and regulators capitulating to threats of stalled investment by dismantling still essential regulatory safeguards.
Best practices in ICT development Nations as diverse as Canada, Japan and Korea can provide insights on how to achieve maximum success in ICT development and what roles governments can effectively pursue. These and other nations offer insights on how government-led integration of technology incubation and development can generate favourable results. While these governments readily encourage private enterprises and direct foreign investment in technology ventures, they do not shy away from pursuing an active and vital role. In vivid contrast to the most effective strategy pursued by the United States, namely government incubation followed by a quick departure and reliance on marketplace forces, best-practices ICT development in many nations demonstrates the benefits of long-term involvement by honest, technologically sophisticated government officials. Public officials may better understand the macro-level, societal stakes involved and work conscientiously to establish a transparent, efficient, flexible and positive business environment for the long run. In many nations, for example, governments sponsor science and technology parks where the government or a government-appointed manager integrates all the necessary elements for ‘the production and commercialization of advanced technologies by forging synergies among research centres, educational institutions and technology-based companies’ (Petree et al. 1999). Governments can achieve this synergy primarily through investments, preferential policies and focused leadership under the auspices of an economic development board that underwrites programmes designed to finance research and development projects and to promote commercialization of applied research. Put another way, nations have expedited ICT development by mastering the ability to foster an efficient and favourable business environment without abdicating ongoing oversight responsibilities. This environment results in part from the ability of technology parks and other development vehicles to foster:
Institutional conditions for implementation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
• • • • •
111
cooperation in both pure and applied research and development with scientific research institutes and laboratories; ease of access to venture capital; the availability of professional, technical, administrative and legal assistance; state-of-the-art information and telecommunications services; and a fair and transparent business infrastructure.
Indigenous comparative advantages Before considering the types of public and private actions that can expedite ICT development, one should appreciate that a significant set of indigenous factors contribute to or deter progress, regardless of what affirmative steps are taken. A number of localized characteristics favour ICT development independent of concerted actions. For example, geography and demographics can make ICT development tasks easier or harder as a function of nation size, population density, per capita income, percentage of high-rise housing and size of households. Nations and administrative regions such as Korea, Singapore (Aizu 2002; Wong 2003)11 and Hong Kong (ITU 2003b) should have an easier ICT development task simply because telecommunications carriers have fewer lines to install and greater population density served by these lines. Geographically small nations with little rugged terrain and high incomes can achieve ubiquitous digital network access on a timely and efficient basis, perhaps even without having to create a sizeable fund for subsidizing services to rural and low-income residents. Similarly, with a population skewed to youthful, urban apartment dwellers, telecommunications carriers can more readily introduce new broadband services and achieve comparatively higher penetration rates than carriers in other nations would achieve. A nation such as Korea enjoys a larger percentage of technology ‘early adopters’ – people who are keen on accessing services that provide faster, better, smarter, cheaper and more convenient solutions to existing requirements coupled with a willingness to use technologies to serve new wants, needs and desires. Well-educated Korean youth with sufficient discretionary funds supported ICT development, first by frequenting personal computer gaming rooms, known as PC bangs, and later by embracing new markets, including streaming music, Internet and wireless messaging and online photography. Furthermore, one cannot underestimate the impact of attitudes towards ICT and the extent to which entrepreneurs will take risks to provide services offering clearly better consumer benefits. A culture favouring education, speedy resolution of problems and risk-taking favours ICT development, simply because consumers will more readily embrace technologies that provide tangible improvements. The push of new technologies meets an equally aggressive demand pull.
112
Rob Frieden
Acquired comparative advantages Indigenous comparative advantages cannot reliably propel a nation into ICT development supremacy, nor do the identified factors fully explain why some nations excel while others falter. Acquired comparative advantages result from concerted efforts by the public and private sector to achieve ICT development with an eye towards fostering improvements in the quality of life, individual wealth and national economic development (OECD 2003). The best advantages result when governments effectively calibrate the scope of intervention (1) to the degree of market stimulation required, and (2) to the extent to which ICT development would not occur but for government subsidization, demand aggregation and sponsored pilot projects.
Government vision, strategy and stewardship The acquisition of comparative advantages in ICT development appears impossible without some degree of ongoing government involvement. No matter how attractive ‘blue sky’ technologies appear on the horizon, governments may need to jump-start new technology adoption and thereby accelerate the accrual of a critical mass needed to achieve scale economies and the ability to offer services at rates a mass market will support. Before private enterprises can operate largely free of government involvement and/or support, a technology incubation phase typically must occur, as was the case for Internet development in the United States. Governments willing to undertake an active role need to reach closure on a vision of what constitutes ICT development success and what steps they should take in order to achieve those outcomes.
Canadian government efforts The Canadian government also launched a series of early ICT development initiatives articulated in the 1990s (Industry Canada 1994; Government of Canada 1996; Connecting Canadians website n.d.). The Ministry of Industry articulated a strategy to make Canada the most connected country in the world.12 It set out to achieve ICT development primarily through the promotion of online access, developing ICT-intensive, ‘smart communities’, creating incentives for the development of indigenous content for transmission via the Internet, expediting electronic commerce and delivering electronic government services (ITU 2003c). In 2001, a National Broadband Taskforce specified a strategy for achieving ubiquitous access to broadband networks and services by 2005 (Government of Canada 2004). Specifically, the Task Force established several access priorities, including the view that all communities, including small businesses and residential users, should have Internet access at throughput speeds in excess of 1.5 megabits per second, rural access rates should not exceed urban rates, and the local broadband infrastructure should extend to schools, public libraries and other public access points.
Institutional conditions for implementation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
113
The Task Force identified two funding vehicles for achieving these goals: 1 2
a top-down, infrastructure government support model that creates broadband network and service investment incentives; and a bottom-up, ‘community aggregator model’, where government-funded pilot programs and the delivery of electronic government services helps stimulate the generation of sufficient demand to use existing network capacity and stimulate the construction of new facilities.
Korean government efforts The Korean government articulated an action plan in 1997 entitled Cyber Korea 21 (Government of Korea 1999, 2002). The Ministry of Information and Communications articulated a vision of a ‘knowledge-based economy’ where every citizen would have access to a personal computer, the government would expedite development of an information infrastructure and all stakeholders in ICT would work together (ITU 2003d) to increase productivity, employment and exports (Dahlman and Andersson 2001; Government of Korea 2003). The Korean government recognized that the scale and ambitiousness of such a vision would require several types of initiatives and financial inducements (Lee n.d.), including: • • •
• • • •
efforts by regulatory authorities to encourage infrastructure investment by incumbents and market entrants (OECD 2004); regulatory parity among operators with an eye towards promoting facilities-based competition, but also market entry by operators who might need to access some facilities of the incumbent; direct financial underwriting as well as offering loans and loan guarantees, favourable tax treatment, and other types of financial support for the construction of new, high-capacity backbone digital broadband networks; financial support for research, development and technology demonstration projects; subsidies for purchase of personal computers by low-income citizens; the promotion of digital literacy, including the ability to use information technologies for interacting with government and for acquiring information, communications and entertainment services (Han 2003); the support of electronic government, education, e-commerce (Lee et al. 2003), healthcare and other types of ICT-mediated services.13
Japanese government efforts Japan developed a high-level, national, information ‘e-Japan’ strategy in 2001 with ambitious goals addressing infrastructure, human resources, e-commerce, e-government and network security (see Prime Minister of Japan
114
Rob Frieden
and his Cabinet 2001 for the e-Japan strategy).14 Perhaps smarting from less robust development than nearby Korea, Japan expedited the development of the world’s most advanced telecommunications and information networks, blending private and public sector initiatives (ITU 2003e; Ishii 2003). The e-Japan strategy triggered the development of 220 separate projects in its first year, as well as achieving the goal of linking 30 million households to high-speed Internet access options (Takada 2003; Taniwaka 2003, 2004; Yamada 2004). Today, Japanese residential consumers have the highest throughput speeds and the lowest per megabit cost.
Regulatory initiatives Perhaps the key regulatory initiative pursued by nations such as Canada, Korea and Japan lies in effectively changing the regulatory climate without triggering the kind of costly and protracted litigation that has thwarted progress in the United States. Nations can use regulatory change to promote facilities-based and resale competition through incremental deregulation of the sector, liberalization of rules affecting incumbent carriers and mandating cost-based and compulsory access to the incumbent carrier’s switches and transmission capacity at fair and compensatory rates. Progressive tax policies, including investment tax credit, further stimulate incentives to invest in ICT infrastructure. National regulatory authorities have to find a way to create incentives for incumbents and newcomers alike to invest in infrastructure needed to provide high-speed broadband data services. Only with sustainable facilitiesbased competition with a multiplicity of operators in each of the technologies will broadband services thrive. Incumbent wireline telephone companies initially will provide broadband services using existing copper and later using fibre-optic facilities. Additionally, competition will come from wireless operators and cable television ventures. Whether robust competition will evolve depends primarily on decisions by incumbent carriers. Both telephone and cable companies will need to consider broadband service market share as essential to ongoing commercial viability in the light of declining margins in, and the maturation of existing services. Faced with such competitive necessity, it follows that an incumbent would have to diversify services and pursue new profit centres, including value-added network, wireless and broadband services (Frieden 2003a). The need to respond to declining revenues in core business lines and new deregulatory opportunities has begun to stimulate interest in expediting delivery of broadband services by US carriers. However, years earlier, carriers in Canada, Japan and Korea made such investments as a result of governmental encouragement, the real or perceived competitive necessity and internal market forecasts. Meanwhile, in the United States, incumbent telephone companies complained about the unfairness of having to unbundle their networks and offer access to individual elements at below market
Institutional conditions for implementation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
115
rates. Cable television ventures succeeded in thwarting efforts to force them to provide open access to any Internet service provider (ISP) in lieu of dedicated access to a corporate ISP affiliate or joint venture partner. Carriers in best-practice nations accept regulatory mandates and emphasize capitalizing on new market opportunities. Carriers in the United States resorted to litigation and delay with an eye towards conserving capital until such time as the demand for broadband services became unassailable. Such stalling tactics resulted from conditions of heightened fear, uncertainty and doubt resulting from an economic downturn largely triggered by the decline in confidence that the Internet and demand for Internet services would trigger perpetual growth. Carriers more willing to embrace change and to accept the onset of a ‘new world order’ predominated by data services (Kiser and Collins 2003; Frieden 2003b) appear better equipped to capitalize on new market opportunities. Carriers keen on conserving capital and reducing risk exposure appear less able to migrate from a business plan predominated by voice services, despite the fact that this once core market has deteriorated and will continue to decline as consumers migrate to wireless and Internetbased services.
Supply-side stimulation: underwriting research, funding pilot programmes and community champions Nations offering best practices in supply-side stimulation recognize the importance of triggering an expedited migration from narrowband to broadband services and promoting widespread availability of new services at attractive prices (United Kingdom Department of Trade and Industry and Brunel University 2002).15 While preferring private carriers to make the transition to broadband on the basis of competitive necessity and declining margins in basic voice telephony markets, governments at the local, provincial and federal level volunteered to provide financial support under conditions of market failure, i.e. the unwillingness of private firms to make the investment based on the view that they lacked certainty as to whether or not they could accrue a sustainable and adequate profit. Such self-help programmes brought broadband digital services to hinterland locations north of the Arctic Circle in Canada, primarily through the assessment of business plans created by community groups, also known as ‘community champions’, and the grant of up to 50 per cent of the costs to develop a broadband network. Ironically, the use of metered pricing for narrowband services made it financially more attractive to migrate to unmetered, always on (‘all you can eat’) broadband services. Unlike in the United States, telecommunications consumers in many nations have to pay per minute rates for access to voice telephone and Internet services. With the onset of broadband services charged on a flat-rated, monthly basis, even moderate World Wide Web surfers could substitute unlimited access for metered service at only a slightly higher cost.
116
Rob Frieden
Demand-side stimulation: promoting digital literacy, aggregating demand and delivering e-government services Best practices also include efforts by national governments to stimulate and aggregate demand primarily by offering citizens better ways to acquire government, education and health services. While youthful video gamers needed no inducement to appreciate the benefits of high-speed online access, others grew to appreciate the time-saving and productivity enhancements available from broadband services. For example, high-speed data networks make it possible for remote communities to secure medical consultations between local nurses and doctors based at urban teaching hospitals, as well as quick transmission of X-ray images. E-learning possibilities include highspeed access to databases, multi-media learning tools and video conferencing with teachers in a virtual classroom environment.
New challenges to developed nations Developed nations such as Germany and the United States increasingly have to rely on ICT markets to accrue competitive advantages that generate wealth and sustain current high standards of living. These nations can no longer simply assume that developing nations will serve as largely untapped markets or as low-cost assemblers and manufacturers of goods using intellectual property created in developed nations. For developed nations, ICT development generates new risks and insecurity, not only in the light of employment losses due to outsourcing, but also because ICT incubation in developing nations helps them become innovators as well as low-cost assemblers. Developing nations such as China have already established themselves as least-cost manufacturers by using the intellectual property created elsewhere. ICT incubation for some developing nations provides the opportunity to become more than technology licensees and copiers. For developed nations to maintain their comparative advantage in ICT, they must continually prime the pump through research and entrepreneurship. As never before, ICT incubation provides upside opportunities for all nations.
Conclusions ICT development, including investment in a robust broadband infrastructure, requires extensive coordination and cooperation among private and public sector players. Successful ICT development typically occurs if, and only if, both types of participants stick to roles proven to maximize benefits. For government, the empirically proven optimal role involves neither a laissezfaire abdication of responsibility nor intrusive, heavy-handed, commandand-control regulation that predominated when private or government monopolies largely controlled how the public received telecommunications
Institutional conditions for implementation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
117
services. Governments can enhance ICT development by articulating from the top a broad vision of what telecommunications and information processing services can do for a nation and its citizens, while at the same time leaving to community champions the flexibility to propose specific, ‘bottom-up’ projects that aggregate the supply of services needed to support the build-up of needed infrastructure. For the private sector, the proven role does not involve extensive litigation and delayed investment, nor does it constitute the leveraging of ICT investment in exchange for even greater deregulatory relief. The private sector needs to make the necessary investments in ICT incubation, but government can create incentives for such investment by underwriting and guaranteeing loans, providing favourable tax treatment and financially supporting a portion of the necessary research, development and technology demonstration projects. Governments do not serve as a catalyst simply by throwing money at the ‘problem’ of insufficient ICT development. Wasted investment in ICT development can occur if government relies on one category of private sector participant, e.g. incumbent local exchange telephone companies, to administer the major programmes designed to promote universal access to basic telecommunication services. The incumbent develops a reliance on and expectation for this funding source. It thus has little incentive to achieve a universal service goal, as opposed to justifying an ongoing source of subsidies for preferred beneficiaries, which includes the carrier itself. Developing a recurring subsidy and funding mechanism, as opposed to relying primarily on ad hoc project funding, typically justifies the need for an extensive bureaucracy similarly keen on pursuing an ongoing mission or expanding broad development goals. Ironically, the universal service funding mechanism in the United States, which promotes subsidized access to often unmetered basic telephony, has created disincentives for consumers to migrate to available, yet unsubsidized, broadband services. Nations achieving comparatively greater success in ICT development demonstrate the value in having a specific mission, achievable goals and policies designed to achieve success. The governments of Canada, Japan and Korea articulated a vision of what ICT could do for both public and private sector beneficiaries. At the macro-level, these nations designed laws that created incentives for risk taking and innovation, and penalized litigation and strategies which delayed making the necessary investment in capitalintensive projects. At the micro-level, these nations linked public funding with private initiatives that aggregated demand, generated matching funds and justified the installation of ICT, even in comparatively unattractive locales. The United States has largely failed to match its comparative advantage in private ICT incubation, such as Silicon Valley, with similar, world-class governmental incubation, despite having achieved success in developing and then privatizing the Internet. The lack of success in recent governmental incubation efforts, e.g. in broadband market penetration, stems largely from
118
Rob Frieden
the failure to appreciate the need to blend and integrate both private sector entrepreneurialism and public sector stewardship. Such stewardship involves active governmental involvement, as cheerleader, referee, loan guarantor, grant funder and anchor tenant in a sector that many in the United States believe warrants little, if any, government involvement. Nations exhibiting best practices in ICT development clearly show the benefit in a combination of public and private initiatives.
Notes 1 ‘The capacity of countries and firms to develop and manage knowledge assets has become a major determinant of economic growth and competitiveness. . . [I]nvestment in and exploitation of knowledge remains a key driver of innovation, economic performance and social well being. Over the last decade, investments in knowledge – as measured by expenditures on research and development (R&D), higher education, and information and communication technologies (ICTs) – grew more rapidly than gross fixed capital formation’ (OECD 2002). 2 ‘Knowledge is now recognized as the driver of productivity and economic growth, leading to a new focus on the role of information, technology and learning in economic performance. The term “knowledge-based economy” stems from this fuller recognition of the place of knowledge and technology in modern . . . economies’ (OECD 1995: 3). 3 See also World Summit on the Information Society n.d.; ICT for Development website n.d.; ITU n.d. 4 ICT development covers many diverse segments of a national economy. Accordingly, broadband development, by itself, may not serve as a complete measure of national success or failure in ICT development. On the other hand, one cannot overemphasize the importance of broadband network access for a variety of ICT services, including high speed Internet access and an increasing percentage of Internet-delivered services, such as Voiceover Internet Protocol and telephone services. ‘Currently VOIP [Voiceover Internet Protocol] accounts for less than 3% of global voice phone calls, according to an AT&T estimate. But a number of trends are working in its favor, say industry executives: the boom in demand; the evolution of the technology, which permits companies to offer services beyond the reach of conventional phones; and the spread of broadband connections, which make VOIP much easier to use’ (Grant and Latour 2003). 5 ‘The pace of deployment and technological progress in broadband, or high-speed, services remains seriously inadequate, a problem that results from the monopolistic structure, entrenched management, and political power of incumbent local exchange carriers (ILECs) such as BellSouth and Verizon and the cable television industry. It is worsened by major deficiencies in the policy and regulatory systems covering these industries. Failure to improve broadband performance could reduce U.S. productivity growth by 1 percent per year or more, as well as weaken public safety, military preparedness, and energy security’ (Ferguson 2002). ‘[Bill] Gates said US broadband would lag behind European and Far Eastern countries by “five to six years”. He slammed US telecom providers and cable networks for recently increasing prices’ (Morgan 2002). 6 Current broadband infrastructure enhancements primarily involve upgrading existing telephone and cable television networks. Digital subscriber line (DSL) service from local exchange carriers involves an upgrade to local loop, copper wires. They offer more bandwidth capable of providing both legacy voice
Institutional conditions for implementation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
7 8
9
10
11 12 13 14
15
119
telephone service and Internet access. Cable modem service from cable television companies involves the partitioning of an existing broadband wire into separate video delivery and Internet access links. Unlike most nations, cable modem access had predominated in the United States. Cable modems provide 75.3 per cent of all broadband services in the United States, while DSL serves 14.9 per cent as of December 2003. ‘Waiting for computer screens to fill has resulted in $25 billion a year in lost e-commerce and countless billions of dollars in lost time for consumers’ (Pociask n.d.). ‘ICT is an integral component of every sector in the New Zealand economy – working behind the scenes as an “enabler” – and is also a major sector in its own right. . . . The information and communications technology sector: makes a significant contribution to export growth; can have a multiplier effect across other sectors; can transform business and operational processes; lifts productivity; improves health and education outcomes; has the potential to add exceptional value to New Zealand’s traditional industries’ (New Zealand Trade and Enterprise n.d.). ‘Well-targeted ICT interventions in five key interrelated areas can play a crucial role in igniting and sustaining this development dynamic by creating the necessary conditions to achieve critical mass and to reach the thresholds required for significant multiplier effects and increasing returns to scale’ (Digital Opportunity Initiative 2001). ‘Information technology dramatically increases the amount and timeliness of information available to economic agents – and the productivity of processes to organize, process, communicate, store, and retrieve information . . . [thereby impacting] countries, as producers and users of this technology’ (Hanna 1994: 1). In a joint venture with Siemens, the China Academy of Telecommunications Technology has developed the TD-SCDMA mobile radio standard for third generation mobile radiotelephones. This is the first telecommunications standard proposed by the Chinese industry and accepted as one of several standards by international forums. Izumi Aizu compared successful deployment in Korea versus mixed results in Singapore. For background on Canada’s broadband initiatives, see www.broadband.gc.ca/ pub/media/index.html. For extensive research and reports on ICT issues in Korea and elsewhere, see Korea Information website n.d. and Korea Information Strategy Development Institute website n.d. See Prime Minister of Japan and his Cabinet n.d. ‘We will strive to establish an environment where the private sector, based on market forces, can exert its full potential and make Japan the world’s most advanced IT nation within five years by: (1) building an ultra high-speed Internet network and providing constant Internet access at the earliest date possible, (2) establishing rules on electronic commerce, (3) realizing an electronic government and (4) nurturing high-quality human resources for the new era’ (Prime Minister of Japan and his Cabinet 2001). Stewardship by the Korean government offers several vehicles for expediting broadband deployment and use. See United Kingdom Department of Trade and Industry and Brunel University 2002.
References Aizu, I. (2002) A Comparative Study of Broadband in Asia: deployment and policy, accessed at www.anr.org/web/html/index_e.htm on 14 November 2004.
120
Rob Frieden
Baumol, W. and Merrill, T. (1998) ‘Deregulatory takings, breach of the regulatory contract and the Telecommunications Act of 1996’, New York University Law Review, 72 (5): 1037–67 (1998). Connecting Canadians website (n.d.), accessed at http://cap.ic.gc.ca/english/5000.shtml on 9 November 2004. Dahlman, C. and Andersson, T. (2001) Korea and the Knowledge-Based Economy: making the transition, accessed at http://oecdpublications.gfi-nb.com/cgi-bin/ OECDBookShop.storefront/EN/product/922000061P1 on 18 November 2004. Dibadj, R. (2003) ‘Competitive debacle in local telephony: is the 1996 Telecommunications Act to blame?’, Washington University Law Quarterly, 81 (1): 1–46. Digital Opportunity Initiative (2001) Creating a Development Dynamic: final report of the digital opportunity initiative (Section 4), accessed at www.opt-init.org/ framework/pages/title.html on 8 September 2005. FCC (Federal Communications Commission) (2004) Availability of Advanced Telecommunications Capability in the United States: fourth report to congress. (FCC 04–208), Washington, DC: FCC, 9 September, accessed at http://hraunfoss. fcc.gov/edocs_public/attachmatch/FCC-04–208A1.pdf on 18 November 2004. Ferguson, C. H. (2002) The U.S. Broadband Problem, accessed at www.brookings. edu/comm/policybriefs/pb105.htm on 1 October 2002. Frieden, R. (1997) ‘The Telecommunications Act of 1996: predicting the winners and losers’, Hastings Communications & Entertainment Law Journal, 20 (1): 11–57. Frieden, R. (2003a) ‘Fear and loathing in information and telecommunications industries: reasons for and solutions to the current financial meltdown and regulatory quagmire’, The International Journal on Media Management, 5 (1): 25–38. Frieden, R. (2003b) ‘Adjusting the horizontal and vertical in telecommunications regulation: a comparison of the traditional and a new layered approach’, Federal Communications Law Journal, 55 (2): 207–50. Government of Canada (1996) Building the Information Society: moving Canada into the 21st century, accessed at www.ifla.org/documents/infopol/canada/ihac9601.pdf on 18 November 2004. Government of Canada (2004) The New National Dream: networking the nation for broadband access, Report of the National Broadband Taskforce, accessed at http://broadband.ic.gc.ca/pub/program/NBTF/table_content.html on 18 November 2004. Government of Korea (1999) Cyber Korea 21, accessed at the Ministry of Information and Communication website www.innovazione.gov.it/ita/intervento/banda_larga/ corea_cyber.htm on 8 September 2005. Government of Korea (2002) E-KOREA VISION 2006, accessed at the Ministry of Information and Communication website www.mic.go.kr/index.jsp on 8 September 2005. Government of Korea (2003) Information and Communication White Paper 2003, accessed at the Ministry of Information and Communication website www.mic.go.kr/ index.jsp. on 8 September 2005. Grace, J., Kenney, C., and Zhen-Wei Qiang, C. (2004) Information and Communications Technologies and Broad-based Development: a partial review of the evidence, World Bank Working Paper No.12, accessed at www-wds.worldbank.org/servlet/ WDS_IBank_Servlet?pcont=details&eid=000090341_20040302090454 on 8 September 2005.
Institutional conditions for implementation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
121
Grant, P. and Latour, A. (2003) ‘Battered telecoms face new challenge: internet calling’, The Wall Street Journal, 9 October: 1. Han, G. (2003) ‘Broadband adoption in the United States and Korea: business driven rational model versus culture sensitive policy model’, Trends in Communication, 11 (1): 3–25. Hanna, N. (1994) Exploiting Information Technology for Development: a case study for India, World Bank Discussion Paper, No. WDP 264: 1, accessed at www1. worldbank.org/wbiep/decentralization/saslib/hanna.htm on 8 September 2005. Hopkins, J. (2004) ‘Other nations zip by the USA in high-speed net race’, USA Today Online Edition, 18 January, accessed at www.usatoday.com/tech/techinvestor/ 2004–01–19-broadband_x.htm. Hukill, M., Ono, R. and Vallath, C. (2000) Electronic Communication Convergence: Policy Challenges in Asia London, Thousand Oaks CA: Sage. ICT for Development (n.d.) Website accessed at http://topics.developmentgateway.org/ ict on 8 September 2005. Industry Canada (1994) Spectrum, Information Technologies and Telecommunications Sector – the Canadian information highway – building Canada’s information and communications infrastructure, accessed at www.ifla.org/documents/infopol/ canada/cihac001.txt on 18 November 2004. Ishii, K. (2003) ‘Diffusion, policy and use of broadband in Japan’, Trends in Communication 11 (1): 45–61. ITU (International Telecommunication Union) (2003a) Birth of Broadband Executive Summary (Sec. 8), September, accessed at www.itu.int/osg/spu/publications/sales/ birthofbroadband/exec_summary.html on 8 September 2005. ITU (2003b) Broadband as a Commodity: Hong Kong, China internet case study, workshop on promoting broadband, May, accessed at www.itu.int/ITU-D/ict/cs/ hongkong/material/CS_HKG.pdf on 8 September 2005. ITU (2003c) Promoting Broadband: the case of Canada, workshop on promoting broadband, 9 April, accessed at www.itu.int/osg/spu/ni/promotebroadband/ casestudies/canada.pdf on 8 September 2005. ITU (2003d) Promoting Broadband: the case of Korea, March, workshop on promoting broadband, accessed at www.itu.int/ITU-D/ict/cs/korea/ on 8 September 2005. ITU (2003e) Promoting Broadband: the case of Japan, 7 April, workshop on promoting broadband, accessed at www.itu.int/osg/spu/ni/promotebroadband/casestudies/japan. pdf on 8 September 2005. ITU (2006) Top 15 Broadband Economies, 1 January, accessed at www.itu.int/ osg/spu/newslog/ITU+Broadband+Statistics+For+1+January+2006.aspx. ITU (n.d) World Summit on the Information Society website (n.d.) accessed at International Telecommunication Union web page www.itu.int/wsis/ on 8 September 2005. Kiser, C. R. and Collins, A. F. (2003) ‘Regulation on the horizon: are regulators poised to address the status of IP telephony?’, CommLaw Conspectus, 11: 19–44. Korea Information Strategy Development Institute website (n.d.), accessed at www.kisdi.re.kr/ on 8 September 2005. Korea information website (n.d.), accessed at Ministry of Information and Communications website www.mic.go.kr/index.jsp on 8 September 2005. Lee, C. M., Miller, W. F., Hancock, M. G. and Rowen, H. S. (2000) ‘The Silicon Valley habitat’, in Lee, C. M., William F. Miller, Marguerite Gong Hancock and
122
Rob Frieden
Henry S. Rowen (eds) The Silicon Valley Edge: a habitat for innovation and entrepreneurship, Stanford CA: Stanford University Press, pp. 1–15. Lee, H., O’Keefe, R. M. and Yun, K. (2003) ‘The growth of broadband and electronic commerce in South Korea: contributing factors’, The Information Society, 19: 81–93. Lee, J. H. (n.d.) Korea Information Strategy InstituteVision of E-Korea and Policy Agenda, accessed at www.kisdi.re.kr/img/eng/pdf/oecd_kisdi_ppt.pdf on 8 September 2005. Leiner, B. M., Cerf, V. G., Clark, D. D., Kahn, R. E., Kleinrock, L. et al. (n.d.) A Brief History of the Internet, accessed at www.isoc.org/internet/history/brief. shtml#wolff on 8 September 2005. Morgan, G. (2002) Gates Warns US Lags Behind in Broadband, 4 February, accessed at Vnunet website www.vnunet.com/vnunet/news/2117463/gates-warns-lagsbehind-broadband on 8 September 2005. New Zealand Trade and Enterprise (n.d.) ICT: growing our high-tech advantage, accessed at www.nzte.govt.nz/section/11757.aspx on 8 September 2005. OECD (Organisation for Economic Co-operation and Development) (1995) The Knowledge-Based Economy (OCDE/GD(96)102), accessed at www.oecd.org/ dataoecd/51/8/1913021.pdf on 8 September 2005. OECD (2002) Towards a Knowledge-Based Economy – recent trends and policy directions from the OECD, Directorate for Science Technology and Industry background paper for the OECD-IPS workshop on promoting knowledge-based economies in Asia, accessed at www.oecd.org/dataoecd/32/15/2510502.pdf on 8 September 2005. OECD (2003) Broadband Driving Growth: policy responses (DSTI/ICCP(2003)13), 9 October, accessed at www.oecd.org/dataoecd/18/3/16234106.pdf on 8 September 2005. OECD (2004) ICT Diffusion to Business: peer review, country report: Korea (DSTI/ICCP/IE(2003)9), working party on the information economy, 7 May, accessed at www.oecd.org/dataoecd/8/6/31787529.pdf on 8 September 2005. OECD (2005) Broadband subscribers per 100 inhabitants, by technology, December 2004, accessed at www.oecd.org/dataoecd/40/16/34919335.xls on 8 September 2005. Petree, R., Petkov, R. and Spiro, E. (1999) Technology Parks – concept and organization, Columbia International Affairs Online Working Paper, accessed at Columbia International Affairs Online website www.ciaonet.org/wps/per01/ on 8 September 2005. Pociask, S. (n.d.) Putting Broadband on High Speed – new public policies to encourage rapid deployment, accessed at www.epinet.org/studies/broadband_pociask.pdf on 8 September 2005. Prime Minister of Japan and his Cabinet (2001) E-Japan Strategy, accessed at www.kantei.go.jp/foreign/it/network/0122full_e.html on 8 September 2005. Prime Minister of Japan and his Cabinet (n.d.) Information Technology, accessed at http://www.kantei.go.jp/foreign/it_e.html on 8 September 2005. Sein, M. K. and Harindranath, G. (2004) ‘Conceptualizing the ICT artifact: toward understanding the role of ICT in national development’, The Information Society, 20 (1): 15–24. Spulber, D. F. and Yoo, C. S. (2003) ‘Access to networks: economic and constitutional connections’, Cornell Law Review, 88 (4): 814–54. Takada, Y. (2003) Promoting Broadband: the case of Japan, Powerpoint presentation at workshop on promoting broadband, Geneva, Switzerland, 9 April, accessed
Institutional conditions for implementation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
123
at www.itu.int/osg/spu/ni/promotebroadband/presentations/06-Takada.pdf on 8 September 2005. Taniwaka, Y. (2003) ‘Emerging broadband market and the relevant policy agenda in Japan’, Journal of Interactive Advertising, 4 (1), accessed at www.jiad.org/vol4/ no1/taniwaki/ on 8 September 2005. Taniwaka, Y. (2004) Outline of the Japanese Broadband Market, The Alliance for Public Technology: 2004 Broadband Forum, 5 March, accessed at www.apt.org/ confer/yasu-taniwaki-presentation.ppt on 8 September 2005. United Kingdom Department of Trade and Industry and Brunel University (2002) Investigating Broadband Deployment in South Korea – broadband mission to South Korea, accessed at www.broadbanduk.org/reports/SKorea_report.pdf on 8 September 2005. United States of America (1996) Telecommunications Act of 1996, Pub. L. No. 104–104, 110 Stat. 56 (codified in scattered sections of 47 USC), accessed at www.fcc.gov/telecom.html. Wong, P. K. (2003) ‘Global and national factors affecting e-commerce diffusion in Singapore’, The Information Society, 19: 190–213. World Summit on the Information Society (n.d) World Summit on the Information Society and the Role of ICT in Achieving the Millennium Development Goals, accessed at The World Bank, InfoDev, Development Gateway website http:// topics.developmentgateway.org/ict/sdm/previewDocument.do~activeDocumentId=8 15843 on 8 September 2005. Yamada, M. (2004) Broadband Development and Security Policies in Japan, 10 January, accessed at the Communications and Information Network Association of Japan website www.ciaj.or.jp/content/topics/Yamada.pdf on 8 September 2005. Yan, X. (n.d.) The Economic Context of 3G Development in China, accessed at TD-SCDMA Forum website www.tdscdma-forum.org/en/pdfword/2004614124 00489465.pdf on 8 September 2005. Zhen-Wei Qiang, C., Pitt, A. and Ayers, S. (2003) The Contribution of Information Communication Technologies to Growth, accessed at http://info.worldbank.org/ict/ WSIS/docs/comp_ICTGrowth.pdf on 8 September 2005.
7
B2C e-commerce dynamics in Germany Do we need a new regulatory framework?1 Bernhard Lageman, Michael Rothgang and Markus Scheuer
Why does consumer oriented e-commerce develop ‘so slowly’? Is there a simple answer to a simple question? Since the burst of the new economy bubble it became clear that the highflying expectations that had accompanied the growth of the Internet economy during the 1990s would not be fulfilled. This holds true for the United States, where e-commerce (electronic commerce) began and where it reached its highest growth rates. But it also holds true, perhaps even more so, for most other developed market economies such as Germany, where e-commerce developed with a time lag of about five to eight years compared to the United States. The immature Internet dreams of the 1990s notwithstanding, the growing importance of the new electronic chains of distribution should not be underestimated. ICT (information and communications technology) equipment and processes are an indisputable element of business life today. It remains difficult, if not impossible, to judge at present the future effects of ICT on the economy and society (Carr 2004: 149). From a long-term point of view, the Internet and the electronic firm networks expanding within it are still very young mediums. Against the background of the novelty of these electronic communication channels, the expansion of electronic transactions, which could be observed within a fairly short period of time, is remarkable. Nevertheless, the diffusion of new technologies often takes a long time. This general assertion applies to all major inventions in the industrial era, so why should it be different in the case of e-commerce? In this sense, then, the question of whether e-commerce has been developing at too slow a pace since its initiation is ill posed. For one thing, we lack a suitable benchmark for the right pace of its diffusion. The evolution of e-commerce depends on many factors, such as relevant technology and hardware, software applications, number and quality of Internet firms and e-commerce business models on the supply side, and consumers’ technical equipment, preferences and behaviour on the demand side. Regulation, as well, plays an important role in the development of
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
125
markets. Placed within this context the question arises as to whether or not existing institutional conditions are responsible for the seemingly low number of business transactions in cyberspace. Recent literature has provided ample evidence that the regulatory framework is crucial in determining the output of market activities. Many authors discuss the effects of regulation and the necessity of regulation in general (Delong and Fromkin 1997, 1999; Klodt 2001), the protection of intellectual property rights (Quah 2000), the role of standards (Varian 2001) and the functioning of the market mechanism within the Internet (Donges and Mai 2001). Other contributions have raised the question as to whether or not and to what extent an e-commerce policy is necessary (Rothgang and Scheuer 2001) and have investigated experiences gained with e-commerce governance (Christiansen 2003). If e-commerce governance and regulation were a serious problem impeding the evolution of e-commerce, then the setters of rules and regulations on different levels should do everything possible to establish a more suitable institutional setting with which to create a more favourable environment for the Internet economy (Zerdick et al. 1999: 257). Of course, that does not mean that the state itself should define the rules for e-commerce. Co-operative arrangements between firms would be an interesting alternative. Thus, before calling for the state to deregulate or re-regulate e-commerce we should examine the actual development of e-commerce both nationally and internationally in order to determine where the barriers to the further development of e-commerce are. In the next two sections we look at the development of e-commerce in Germany and then focus on the development of business-toconsumer (B2C) e-commerce. In the subsequent section we discuss existent regulations for e-commerce in Germany, including the rules defined by the European Union and international organizations, and ask whether Germany is in need of changes in its institutional setting for e-commerce. We summarize the results of our investigation in the final section.
Business-to-consumer e-commerce in Germany Definitions of B2C There is no generally agreed definition of electronic commerce. The term is used very differently in various settings depending on whether one is talking about different kinds of activities (e.g. trade and/or services), different data transmission channels (e.g. the Internet, firm intranets, extranets and other networks) and/or the different types of technical media involved (e.g. computers, phones and television sets). Furthermore, there is no consensus concerning which parts of a complex purchase or sales transaction extending from the electronic order to after-sales services must be carried out by electronic transmission to be counted as ‘e-commerce’ (Preissl 2003: 1). Most observers will agree that the use of the Internet as a source
126
Lageman, Rothgang and Scheuer
of information for comparing offers does not justify its classification as ecommerce. But what about the form of payment chosen? Is electronic payment (e.g. by credit card) a prerequisite for classifying a transaction as e-commerce? The sale of goods on the Internet normally includes the physical transmission of the sold articles by post services. In this case, the electronic dimension is restricted to certain parts of the transaction. Circumstances are different when tickets or music files are sold, articles are published, or weather forecasts are supplied through the Internet. The delimitation of e-commerce is important, particularly if we want to assess the volume and development of e-commerce. Following the definition used by the US Census Bureau, e-commerce is ‘any transaction completed over a computer-mediated network that involves the transfer of ownership or rights to use goods or services’ (Mesenbourg 2000: 3; Atrostic et al. 2000: 2). According to this definition, electronic payment is not a precondition for a transaction to be classified as e-commerce. Payment may be carried out in conventional ways, yet the transaction may still be defined as e-commerce. Electronic transactions between enterprises are defined as business-tobusiness (B2B) e-commerce. The electronic sale of goods and services by firms to consumers, on which we concentrate in this article, is denoted as B2C. We use term ‘e-commerce’ interchangeably with ‘Internet commerce’, because the Internet is the dominating medium in electronic sales to consumers and will remain so in the near future. However, e-commerce may also be based on different communication channels such as electronic data interchange (EDI) networks that are not connected with the Internet.2 Phone order transactions in which no computer is involved are also usually not included in e-commerce. The buying and selling processes classified as e-commerce are completed as soon as agreement is reached between seller and buyer upon the transfer of ownership or rights to use goods or services. The evolution of B2C and overall e-commerce sales E-commerce as the act of purchasing goods and services performed by a customer via the Internet is still a field of the economy where reliable statistical data are scarce (UNCTAD (United Nations Conference on Trade and Development) 2003: 16–17). Most relevant statistics to date are produced by consulting firms and private research organizations specialized in information technologies and e-commerce, such as Forrester Research and Jupiter Research in the United States or Bitkom in Germany. Statistical offices have just begun to pay attention to e-commerce. In most cases they do not collect data on e-commerce in any systematic way, but some offices such as the Federal Statistical Office of Germany have carried out pilot surveys (Schnorr-Becker 2004). In 2000, the US Census Bureau initiated a programme that aims to fill this data gap. It now collects and publishes relatively detailed data on e-commerce in the United States.
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
127
The discrepancies between the e-commerce statistics produced by the US Census Bureau on the one hand and those produced by market research companies on the other indicate the difficulties associated with collecting the ‘correct’ data. Forrester Research estimated the US online retail e-commerce sales at $110.7 billion in 2002. The corresponding value for online retail spending published by Jupiter Research for 2002 was $41.2 billion. The latter value was not far from the figure for e-commerce sales in retail trade published by the Census Bureau, which was $44.3 billion (US Census Bureau 2004: 664–5, 790). One partial explanation for the difference is that the retail e-commerce projection by Forrester Research included service sales such as travel arrangements and event tickets. The Census Bureau estimated the value of services sold by electronic transmission channels in 2002 to be $41.5 billion. Total B2C would then add up to $82.7 billion, thus more closely approximating the Forrester Research estimate. For a long time, public discussion on e-commerce concentrated on B2C. The perspective that millions of consumers would order their purchases via the Internet proved to be a very fascinating idea. However, B2B electronic transactions over the Internet already surpassed B2C in the early development stage. All relevant data sources agree that in all developed market economies only 10 to 15 per cent of all e-commerce sales can be attributed to B2C, whereas B2B accounts for 85 to 90 per cent. This relation is not expected to change in the near future (TNS Infratest/IIE (Institute for Information Economics) 2004: 220–3). Table 7.1 shows the European Information Technology Observatory (EITO)3 estimates of the present volume and development trends of ecommerce in selected European countries. Western European e-commerce, which added up to € 172 billion in 2000, was still on a relatively modest level. The increases over the last few years have been very impressive in all countries compared to sectoral growth records, even those of new branches. These data should also be interpreted with care. They are based on rather rough estimates and the forecasts for 2008 may be too optimistic. Nevertheless, these data clearly show that the Internet economy and e-business are thriving. Internet commerce in Germany has developed remarkably well compared to the larger member states of the European Union. That appears to be the case both for B2B and B2C. Total e-commerce sales per capita reached € 2,460 in 2004 in Germany, €2,030 in the United Kingdom, and only € 1,650 in France, €1,290 in Italy and €950 in Spain. The German position in e-commerce is especially strong in the field of B2B transactions. The sales in Germany (€ 180.3 billion) approximated to the sum of B2B sales in France, Italy and Spain (€184.3 billion). The Scandinavian countries, which are not included in Table 7.1, are clearly ahead of the Central and Southern European countries in e-commerce. Thus, Germany assumes a relatively stable, middle position in the development of e-commerce in Western Europe.
128
Lageman, Rothgang and Scheuer
Table 7.1 Development of e-commerce in Western Europe and in selected European countries 2001 (€ bn)
2002 (€ bn)
2003 (€ bn)
2004 (€ bn)
2008 (€ bn)p
Average growth 2001–08 %
Total Internet commerce France 21.1 Germany 44.9 Italy 17.3 Spain 6.9 United Kingdom 38.3 Western Europe 171.6
38.7 87.8 32.3 14.2 58.5 308.9
65.9 138.1 52.1 24.9 84.9 476.7
97.7 202.6 73.9 38.3 120.3 680.0
331.9 670.0 239.9 127.2 399.8 2,217.2
48.2 47.1 45.6 51.6 39.8 44.1
Business-to-business France 18.6 Germany 39.7 Italy 15.4 Spain 5.7 United Kingdom 32.4 Western Europe 149.1
34.1 78.3 28.9 12.3 48.8 270.3
57.8 122.7 46.2 22.0 68.0 413.6
85.7 180.3 65.3 33.3 97.7 591.7
272.9 580.6 199.4 111.0 329.9 1,866.1
46.8 46.7 44.1 52.9 39.3 43.5
Business-to-consumer France 2.5 Germany 5.3 Italy 1.8 Spain 1.2 United Kingdom 5.9 Western Europe 22.5
4.6 9.5 3.4 1.8 9.7 38.6
8.0 15.4 5.9 2.9 16.8 63.1
11.9 22.3 8.6 4.0 22.6 88.3
59.0 89.4 39.6 16.2 69.8 351.1
57.1 49.7 55.5 45.0 42.3 48.1
Source: EITO 2002: 28–9; EITO 2003: 26, 28; EITO 2004: 21; EITO 2005: 21–2. Note: p Projected volume.
Business-to-consumer e-commerce in Germany: dimension, dynamics and structure The volume of German B2C transactions in 2004 was estimated by EITO to be €22.3 billion (Table 7.1). This figure includes electronic retail trade and services. A more conservative source, the Association of Retail Traders, estimated the volume of B2C sales (retail trade and services) to be €13 billion for the same year (HDE 2004: 15). Data published in September 2004 from the GfK[pc332] WebScope study4 (GfK 2004) show that in the first half-year of 2004 €5.3 billion were spent on goods purchased through the Internet. This would equal a total turnover of online sales of about €11 billion in 2004. The Federal German Statistical Office (Statistisches Bundesamt) estimated the B2C turnover of goods in Germany to be about €6 billion in the year 2002 as a minimum level (Petrauschke and Kaumanns 2003). However, the Association of German Direct Marketers (Bundesverband des Deutschen Versandhandels e.V. 2005) published an estimation of the volume of turnover in e-commerce in Germany (excluding digital
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
129
services and travel bookings) of only €4.9 billion. This would mean that in comparison to the year 2000, e-commerce turnover would have more than quintupled, and compared to 2003 would have increased by 36 per cent. The statistical evidence on e-commerce is, to say the least, not very clear. It is reasonable to assume that both consumer goods and services in B2C make up about 50 per cent of the total volume of sales. Estimated retail trade sales in Germany amounted to about €470 billion in 2004. Thus, according to EITO figures, the percentage of e-retailing with respect to total retail trade sales would have been 2.4 per cent. This figure seems to be rather surprising. In 2002, the US Census Bureau estimated that e-retailing in the United States made up 1.4 per cent of total retail trade sales ($44.3 billion to $3,230.1 billion; US Census Bureau 2004: 665). Even if we considered that this percentage increased in the United States to about 1.9 per cent in 2004 (estimation in US Census Bureau 2005: 1), the share of e-commerce in Germany would still be higher. This seems highly implausible. Most available qualitative information on e-commerce in Germany and in the United States, indicates that B2C not only developed earlier in the United States but that it also has been accepted to a higher degree by American consumers. Perhaps EITO figures overestimate B2C in Germany and thus also in other European countries. In that case, the number given by the Association of German Direct Marketers (€4.9 billion in 2004) might be more realistic. Their estimate implies that the share of electronic sales in total retail sales would have been a little over 1 per cent in 2004. In general, a number of studies conclude that B2C will continue to grow rapidly over the next few years, although the estimates have become more Greece Portugal Italy Spain Belgium France EU15 Ireland Austria Germany Finland United Kingdom Luxembourg Netherlands Denmark Sweden 0
5
10
15
20
25
30
Figure 7.1 EU citizens who purchased on the Internet, 2003 (%)
35
40
130
Lageman, Rothgang and Scheuer
reserved than before (European Commission 2001a: 25; NFO Infratest 2002: 343; TNS Infratest/IIE 2004: 264, Table). The number of online shoppers in particular is rising with the increasing use of the Internet by customers (Enigma GfK/TNS Infratest 2004: 8) and is expected to continue. At the same time, the relations in the degree of Internet e-commerce between these countries do not seem to have changed essentially in the last few years. As the Eurobarometer published in March 2004 shows, the percentage of citizens who have bought articles on the Internet is higher in Germany than in the EU-15 (Figure 7.1) and even more so compared to the EU25 because Internet penetration rates are relatively low in the post-accession countries. Yet Internet penetration in Germany is still lower than in the Scandinavian countries, the Netherlands and the United Kingdom. At the same time, the United Kingdom is behind Germany in B2B e-commerce, as we can see in the EITO figures (Table 7.1). Once again, one has to be very careful in interpreting these data. They result from population surveys that might not be very precise in gauging e-commerce activities. These numbers show the percentages of people who bought anything via the Internet (and related media), which does not say anything about the amounts spent nor the frequency of purchases and sales made. At the same time, however, the relevance of single groups of products in B2C can be traced, at least for Germany, fairly well on the basis of ongoing surveys. The relevance of groups of products is determined by a series of different factors, such as the target group, the need for consultation, the ability of the product to be individualized and transaction costs (Table 7.2). These criteria have been met for many years and have remained almost unchanged for the same group of products. The target group of this survey contained 17.59 million buyers who had made purchases of at least one product in the last 12 months via the Internet. If the breakdown of turnover by products rather than the frequency of purchases is considered, the result changes due to the average prices of products traded. In this category, clothing and shoes had the highest share Table 7.2 Products most frequently bought via the Internet in Germany, 2003 Item sold
Percentage of respondents
Books Music/CDs Clothing/shoes Gifts Admission tickets Computer hardware/accessories CD-ROMs/DVDs Train tickets User software
44.1 34.9 31.4 28.0 22.2 22.2 20.6 19.8 19.7
Source: AGIREV 2003.
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
131
in 2003 at 11.5 per cent, followed by PC accessories (10.8 per cent) and consumer electronics/photo/video (8.7 per cent). Books, including cards and journals, at 6.1 per cent and CDs/CD-ROM at only 4.1 per cent are lagging behind (TNS Infratest/IIE 2004: 275). The diversity of the successful products in e-commerce shows that there seems to be no general and simple rule when it comes to whether a product can be marketed via the Internet or not. Nevertheless, it seems to be beyond dispute that the reasons behind success can be found to a certain extent in the nature of the products themselves: the sooner the potential buyer can determine the use and the quality of the product, the easier it is to be sold over the net. Only on that basis can potential price advantages of direct marketing be made clear to customers. On the seller’s part, there are also logistic factors which are important for success in direct selling via the net.5 One important aspect of e-commerce is that it potentially makes crossborder sales between different countries easier. Thus, national distribution systems in retail trade and consumer services could theoretically be undermined by cross-border electronic sales. In fact, considerable transactions costs, particularly cross-border transportation costs of goods, make orders in other countries less attractive than could be expected at first glance. In order to assess the cross-border relevance of products traded via the Internet it is not possible to revert to exact statistics. According to one consumer survey conducted by EOS Gallup Europe, almost half of the German households that had bought online declared that they had done this at least once through a foreign supplier (EOS Gallup Europe 2002: 62). In relation to total turnover, however, the share of this kind of purchasing from private customers in another country is still very low. Based on a survey carried out with general managers and marketing directors, one EU study estimates that in the 12 months between mid-2001 and 2002 only 3 per cent of the purchases made by consumers via the Internet were made outside their own country (Press and Communication Directorate-General, Public Opinion Analysis Unit 2002: 25). In evaluating such data, one has to bear in mind that it is sometimes difficult for the household to know where the Internet-supplier is actually located. The country of the location of the enterprise and of the website are not necessarily identical (OECD 2002: 68). Price differences play a decisive role in shaping private household decisions to purchase in another country. Goods suited for sale on the Internet have to be much cheaper in a foreign country in order to be considered as an attractive alternative. As is evident from the current and relatively low volume of Internet-based, cross-border distance trade, price differences that would be able to overcome consumer uncertainties do not appear to be widespread at the moment. This view is backed by a study commissioned by the European Commission for the period between March 1999 and March 2000 on price differences in the European Single Market (European Commission 2001b:
132
Lageman, Rothgang and Scheuer
18–19). So far, e-commerce has not yet become a very well-developed, and surely not, ‘borderless market’ (GFA Management 2002: 48).
Factors influencing the structure and volume of B2C e-commerce in Germany Differentiated picture As discussed above, international comparisons appear to be fairly complicated due to the novelty of the Internet and the general problem of data collection. Consumer and supplier surveys provide only a rudimentary, neither an authentic nor a reliable, picture of the actual development of electronic commerce. Solid market studies that incorporate all relevant aspects, such as total or partial number of electronic transactions, number of customers using the Internet as a complementary medium of information, supplier networks on the Internet, and substitution relations between electronic and conventional mail-order business, are missing. Moreover, when looking at international differences in the volume of e-commerce, several additional factors should be taken into account. First of all, Internet use originated in the United States and was diffused throughout the industrial world, eventually also reaching industrializing and developing countries. Different patterns of diffusion within Western European nations could thus simply be a result of time lags. Second, the English-speaking industrial countries and countries where the use of English as second language is widespread – e.g. Scandinavian countries – certainly show an advantage over Roman countries. Finally, cost factors, such as lower telephone fees (i.e. flat rates), might also have contributed to an earlier diffusion of the private use of the Internet in some countries. A broader acceptance of the Internet in the United States could likewise be a result of earlier technological distribution. The effect of the time lag qualifies the importance of other factors, including institutional factors. After all, it seems that the difference between the United States and other industrial countries has not increased to a noticeable degree. The development of B2C e-commerce has unfolded in a remarkably manifold way. Strong growth patterns in the electronic banking sector, electronic ticket sales and Internet-based hotel reservations contrast with very sluggish growth rates in other areas. Not at all astonishing is the fact that electronic products such as PCs, printers and digital cameras are often purchased online. Electronic book trade has become a serious competitor of the traditional book trade within a fairly short period of time. Nevertheless, limits to the growth of e-commerce have also become increasingly apparent as the development of e-commerce has been relatively sluggish in other consumer product markets. The main causes for these differentials can be found in three areas: supply-side factors, demand-side factors and institutional factors. These are analysed in more detail below.
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
133
Technology The development of e-commerce is completely based on ICT. Rapid technological change in ICT automatically leads to improvements in the technical environment of e-commerce, and ICT’s remarkable progress over the last 30 years has often been described in the literature. The average hardware equipment of firms and private PC users in 2005 is incomparably better than the equipment sold in the ICT markets 10 years ago. For instance, the speed of PC calculation as well as PC memory capacity has multiplied. Software development has followed the increasing availability of advanced hardware equipment. There are acceptable software programs for nearly all standard business problems, even for SMEs for which the software solutions developed in the context of large enterprises had often not been attractive. Today, consumers can use highly sophisticated and reliable software programs on their personal computers. At the same time, software applications for the Internet are far more developed and reliable than ten years ago. The capacity of e-commerce enterprises to make attractive offers to consumers via the Internet is far better today than some years ago for purely technical reasons. Better technologies on the hardware and software side also make it possible to find convincing solutions for security problems, which are critical for the efficient use of the Internet for commercial purposes. Major developments in ICT that can decisively influence B2C in the next few years are the dissemination of broadband communications channels, which has just started on a wider basis, and the convergence of IT and telecommunication (EITO 2005: 25). Broadband communication not only allows speedier electronic communication processes but also the establishment of far better Internet portals for electronically trading firms as well. The convergence of IT and telecommunication will lead to the stronger penetration of Internet applications into the daily lives of consumers. This might foster the readiness of many consumers to make the Internet an integral part of their buying decisions in the future. Internet firms: supply-side factors The Internet and other data transmission channels existed before any firms could offer their products and services over these channels. The use of the Internet for commercial purposes started in the United States in the late 1980s and in Europe in the early 1990s. The presence of a critical number of consumers with access to the Internet was a precondition for the commercial use of the Internet. Among the pioneers of e-commerce in Germany were many young idealists without any experience in commercial activities, retail traders who discovered an interesting new business field, and who also established firms such as Amazon that tried to extend the geographic area of their activity to make use of economies of scale and scope. B2C via the Internet gained
134
Lageman, Rothgang and Scheuer
a decisive impetus in the second half of the 1990s as a ‘gold-digging euphoria’ developed around the Internet economy. At the same time, the market values of Internet firms exploded in the American and European stock exchanges, including the German ‘new market’. During these years of seemingly unlimited development, thousands of firms attempting to initiate ‘e-commerce’ in one way or another were founded in Germany. Many of these start-ups surely never traded with anything, while other hopeful firms came to an end already in the start-up phase. Unfortunately, there are no reliable figures on Internet start-ups that could reconstruct the development of the e-commerce firm population. The only thing that seems clear from the fragmented information available is that there was a very tough market selection process among Internet start-ups so that the chances of survival were very low. Only 5 to 10 per cent of the firms established between 1990 and 1995 are estimated to have survived ten years later. This picture of industry population dynamics is well known from economic history. A newly constituting branch at first attracts a large number of start-ups attempting to make commercial use of new technologies or product families. In the next phase of market consolidation, a hard selection process takes place among these newly founded firms. Competition leads to a strict selection among the market participants where the most efficient and innovative firms have the best chance of survival. This process occurred in the automobile industry during the first decades of the twentieth century, as well as in other durable consumer goods industries. B2C follows this same evolutionary pattern. Although the general pattern of early market dynamics in e-commerce is similar to the pattern found in industrial consumer and investment goods sectors, it would be wrong to assume that e-commerce should automatically evolve into a highly concentrated branch as, for instance, the automobile industry did. It is true that building up efficient Internet marketing organizations demands considerable investments, as Amazon’s example shows. In this case, it may take years to reach the break-even point in sales turnover, and high losses are unavoidable in the first phase of business operation. But it is also possible to enter the market on a far more modest scale with relatively low risks. Establishing a website and offering a product to customers is not necessarily associated with high set-up costs. The severe selection process among Internet suppliers was associated with a basic change in the patterns of ICT use in the surviving firms. The first generation of Internet firms concentrated on establishing a more or less impressive homepage and offered goods and services in an often arbitrary way. Limited sales and deceptive exchange results have led the focus of attention away from the presentation of the firm to the outside world via an Internet homepage to a better organization of ICT-based business structure within the firm (OECD 2004: 106). Thus, e-commerce and internal value creating processes are better integrated today.
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
135
At present, e-commerce activities constitute only a small part of far more complex marketing activities by the firms that use the Internet as marketing channel. Pure Internet sellers have been supplanted by retailers who are conducting e-commerce along with other diverse marketing activities. Multi-channel firms that originally were not involved in the B2C market play a strong role in the field today. According to a recent survey, multichannel firms that offer online sales achieve only 3.8 per cent of their total sales through online transactions (Hudetz and Tanaskovic 2004: 11). E-commerce is a very young economic branch still in its infancy. It is thus no surprise that this young market has experienced major structural changes in recent years. At first, specialized Internet traders dominated a very small market. As soon as conventional retail traders realized the great potential of the Internet marketing channel, they eagerly took advantage of the opportunity and established their own Internet marketing departments. Hence, the combination between conventional channels of distribution and the possibilities opened up by the Internet, the so-called ‘bricks and clicks models’, is one of the current most promising avenues for further proliferation of online trade (European Commission 2001c: 12). This view is shared by the German Retailers Association (HDE), which expects total online turnover in Germany to increase from € 1.25 billion in 1999 to €14.5 billion in 2005, with multi-channel shops as the main winners (HDE 2005). EITO expectations for 2007 are even higher at €48 billion in online shopping turnover and an additional €40 billion offline turnover by 16 million crosschannel shoppers who have been influenced by information and price comparisons via the Internet before they actually buy in a traditional shop (EITO 2004: 181, Figure 14).
Demand-side fctors: Internet access, learning processes and fading fears of misuse To date, B2C has played a minor role in retail trade and in the marketing of services in general. At present, less than 2 per cent of all consumer goods are sold via electronic channels, even in the United States as the market forerunner. The fact that Americans spend more money on online purchases than Germans looks less impressive if we take into consideration the small share of their online transactions in total retail expenses (1.9 per cent in 2003). Nevertheless, the growth rates of this young market segment in the 1990s have been striking, and e-commerce is just beginning to consolidate its marketing position in relation to traditional marketing channels. The most important questions regarding B2C development are: What is the percentage of consumers who are effectively able to use online services because they have Internet access? Who buys online and why? Which goods and services are bought online and what are the interrelations between Internet purchases and traditional marketing channels? Finally, what factors impede or foster Internet purchases? These are the questions discussed in the following two sections.
136
Lageman, Rothgang and Scheuer
Adequate computer equipment and Internet access – ‘e-readiness’ – is an elementary precondition for participation in Internet-based e-commerce. The number of people who own a personal computer and have direct access to the Internet has grown steadily since the early 1990s. According to a BITCOM estimate (2005: 9), the percentage of the German population using the Internet was 54 per cent in 2004 (44.6 million). This percentage was lower than the percentage of Internet users in the United States (65 per cent), but Germany outranks Western Europe (45 per cent) and Eastern Europe (11 per cent). The Scandinavian countries are clearly ahead of the rest of Europe with their extremely high Internet user rates (73 per cent in Denmark and 69 per cent in Sweden). Internet users do not automatically resort to the Internet for shopping purposes. Not all consumers who have private Internet access order and buy goods or services via the Internet. According to the OECD ICT database, 44.9 per cent of German Internet users searched the Web for information about goods and services in 2002, but only 19.2 per cent used it for purchasing and ordering goods and services (OECD 2003b, Tables 4.3.3 and 4.3.4). This percentage was far higher in the United States (39.1 per cent), in Denmark (37.6 per cent), in Sweden (36.9 per cent) and in the United Kingdom (32.77 per cent), and all figures appear to be compatible with the data on Internet access and user practices cited above. The numbers of buyers on the Internet has also been growing over the past few years. At the same time, the frequency of purchasing transactions and the amount of goods and services bought via the Internet is also on the rise. Apart from the steady improvement in user equipment, other factors have contributed decisively to the growing amount of Internet sales to consumers. Trust is undoubtedly a decisive factor for the success of e-commerce (Kollmann 2004: 30). The clients have to be convinced that information they send via the Internet to the sellers, e.g. credit card information, is honestly used and that security standards against criminal misuse are reasonably high. As diverse analyses show, German B2C firms have succeeded in establishing relatively stable trust relationships and loyalty with their clients (see, for example: Mentzel 2003; Hudetz et al. 2004). The observation that a steadily increasing number of German consumers use the Internet for their daily banking transactions might indicate that the psychological barriers of German consumers towards perceived electronic transaction risks are not as high as to impede more use of this medium. The growing awareness of consumers of the shopping potential of the Internet and the increasing readiness to buy via the Internet are essentially due to learning processes. The Internet is still a relatively new medium. Consumers and suppliers alike have yet to learn to make the best use of it. The predominance of younger people among Internet shoppers shows that learning processes are very important for the acceptance of the Internet with regard to demand. In general, the prospects of B2C development seem to be better than many observers believed after the burst of the ‘dotcom
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
137
bubble’. Nevertheless, while it does seem clear now that the Internet creates an additional marketing channel for retail trade in goods and services, it will also probably not replace stationary commerce on a broad base. Institutional factors: Regular settings and the development of B2C e-commerce E-commerce in Germany – as in other highly developed market economies – has not been subject to special restrictions that could have prevented its development. At the same time – and this appears to be the central aspect of the legal question – German law has in no way been prepared for the legal problems caused by the Internet medium. The same appears to be true for other countries. The German economy has gained a bad reputation for being highly regulated. With respect to consumer market regulations, such a judgement does not stand the test of unemotional assessments. The consumer goods markets in Germany are – in general – accessible both for domestic and foreign suppliers from other EU countries. In addition, as a consequence of the process of European integration, goods market regulations are decreasing in importance. A somewhat different situation prevails with respect to service markets. As most European countries regulate access to these markets, German regulations do not deviate from the neighbour countries in any essential way. Regulations that are specific for Germany are mainly concentrated in branches of the economy that are irrelevant to B2C, e.g. the recently (and only partially) liberalized German ‘Handwerksordnung’ for craft trades. This regulation makes market access dependent on an examination (Meisterprüfung). For the development of the Internet economy, however, this regulation is of no relevance. Internet suppliers in Germany are not subject to any more special restrictions than they are in the United States. Other regulations that restrict the free trade of goods in Germany – the German law for drug sales – are in danger of being eluded by electronic commerce. Since 2004, the law permits German pharmacies to sell nonprescription drugs via the Internet. It cannot prevent consumers from buying prescription drugs without a prescription from suppliers abroad. However, this practice is still unlawful in Germany.6 A striking example of the replacement of private contractual regulation by a law represents the book retail price maintenance (Buchpreisbindung) in Germany. The Uniform Fixed Retail Book Price Bill was enacted on 1 October 2002. This bill obligates publishers to set uniform, fixed retail prices for all their new publications regardless of the channel through which the book is distributed, e.g. small book store, large book chains or the Internet. This exemption from the otherwise reigning principle of price competition is culturally motivated ‘in order to safeguard the necessary width and variety of the offer’ because ‘by the book as a medium the fundamentals of our culture are communicated’ (Bundesregierung 2002).
138
Lageman, Rothgang and Scheuer
A system of fixed retail prices for books had already existed in the form of a contractual agreement between publishers and booksellers, but this agreement came increasingly under pressure from cartel concerns voiced by the European Commission. Eventually, however, the EC officially acknowledged the application of nationally regulated book-price systems (European Commission 2001c). By regulating the issue in a national bill, anti-cartel arguments became pointless, as these only address private parties agreeing on restrictions to competition. Moreover, the new German bill now makes it impossible for any publisher not to fix the prices for new books. Before the enactment of the bill, it had been possible to avoid the fixation of the book prices by not joining the voluntary contractual agreement. Even private persons selling new books in larger quantities on eBay are now subject to the prices fixed by the publishers, as a recent court ruling in Germany has decided (Börsenverein des deutschen Buchhandels 2004a). E-commerce has played an important role in stipulating the re-regulation of the book market as a reaction to Austrian direct marketer (LIBRO)7 attempts to undermine the German contractual fixing of book prices by cross-border trade (European Parliament 2002). As the German Association of Booksellers states, the German form of regulation has become accepted as the new EU standard (Börsenverein des deutschen Buchhandels 2004b). With the exception of Finland and the English-speaking countries exposed to the competition of the world market, all other European Community member states already have fixed-price systems on the market for books or are considered to be heading in that direction.
Regulation of B2C e-commerce in Germany How to regulate B2C e-commerce In Germany, as in all other industrialized countries, the development of B2C took place within a highly structured regulatory setting for commercial trade in general. The new channel of distribution for goods and services is subject to the same legal conditions for commercial, contract and corporate law as the older ones. This fact has partly been neglected by the discussion surrounding Internet regulation. The perspective that nothing would remain the same was part and parcel of the ‘stock market hysteria’. Initial scientific discussion was also inclined to stress the revolutionary changes that would accompany the dissemination of the Internet. Today, the question regarding ‘appropriate’ regulatory settings for B2C e-commerce can be discussed in a different light as some time has passed since the Internet boom and bust. Two central fields of scientific discourse, which also underlie the contributions of this volume, have emerged: (1) a discussion of whether existing regulatory frameworks on the national level foster the development of ICT in general and e-commerce in particular, which also concerns the question of possible path dependency inducing some countries
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
139
to fall behind in the diffusion of e-commerce; and (2) a discussion about the extent to which the state should intervene by regulating the Internet, which also often points to private alternatives to state intervention (such as with regard to security in the Internet). In Germany, the discourse of similar questions – particularly to what extent the state should regulate economic activity – has a long-rooted tradition that goes back to the historical school and the Freiburg school in the 1920s. Some economists, such as Walter Eucken (1990), stressed the importance of instituting a strong state guided by strong principles of economic policy to prohibit the creation of cartels. This idea was complemented by the view that the complexity of economic processes poses a central obstacle for an economic policy oriented towards specific goals. Hence, every intervention into the economic process potentially impairs an economic system’s capacity to adapt to changes and makes the window of evolution smaller (Hayek 1972; Streit 2001). This problem has motivated some economists to postulate an economic policy in accordance with regulatory settings that do not affect the economic process in a negative way. Both aspects of state regulation – checking the individual market power of firms while not interfering with market activities – have influenced the emergence of the framework conditions for economic activity in Germany since the Second World War. Hence, many issues that are important for economic trade are not at all subject to any restrictive government regulation. The observation that there are no specific restrictions that could hamper economic trade also applies to B2C. Thus, we could ask whether we are in need of specific e-commerce regulations at all. However, when it comes to e-commerce regulations and possible hindrances for the development of B2C, this general question boils down to very concrete questions about specific deficiencies in the regulatory setting. What are the prerequisites for the diffusion of e-commerce? This topic somewhat overlaps with the regulatory agenda for the diffusion of the Internet in general (DTI (Department of Trade and Industry) 1998: 13–18) and could possibly make an adaptation of the regulatory framework to changing conditions necessary in the following respects: • • • •
to ensure inexpensive and nationwide access to Internet services; to keep markets open on the supply side in order to ensure a degree of consumer choice between alternative service providers; to enhance safety standards which raise customer acceptance of B2C (e.g. data transfer and payment security); and finally to eliminate Internet-based criminal activities.
Deficits with respect to these conditions for the development of e-commerce could, in fact, lead to a possible future backlog. It is therefore worthwhile to look at both the existing and the evolving regulatory framework for e-commerce in Germany and to ask what effect it is going to have on future development.
140
Lageman, Rothgang and Scheuer
E-commerce policy in Germany There are fundamental differences with respect to ‘regulation philosophies’ between the United States and the European Union. The United States seems to prefer a laissez-faire approach by which government only defines and enforces property rights for e-commerce, while delegating further regulation if necessary to the market. However, policy cannot always abstain from regulatory action, especially when ‘national interests’ are at stake. Apart from the United States, the European Union proactively promotes a political programme that very clearly focuses on federal regulation of e-commerce as a basis for electronic market activities. In general, the US approach is based on the assumption that the positive effects of the new technologies that are apparent in lower transaction costs will best flourish in markets that are to a large extent free of interference from authorities. One example of this philosophy may be found in the area of data protection as stated in the ‘Platform for Privacy Preference Project’ (P3P), which has set a private standard for the use of personal data. Another example of a private response to the challenges accompanying modern technology is represented by the creation of private escrow firms for source codes of computer programs. In this context, software piracy is the most striking example of the limits of private regulation. The US Software & Information Industry Association (SIIA 2000) has continually fought for internationally binding and enforced laws. Like all other forms of crime, fighting against ‘cyber crime’ sets narrow limits for private solutions because the use of a state’s monopoly of force might be necessary when it comes to interfering with individual rights. In Germany, the dispersion of e-commerce took place within the legal framework of commercial law, which also applies to electronically based trade transactions. The German commercial code (Handelsgesetzbuch) constitutes the basis for all business transactions in any form. Moreover, a large number of already existing institutions, laws and regulations are in one way or another way relevant to electronic transactions. Table 7.3 highlights only a few of the most important ones. Traditionally, national regulations determined the framework for commercial trade in general and B2C in particular. These have increasingly been amended and replaced, first by international regulations and second by private initiatives. Thus, many problems in competition policy, the protection of property rights and the protection of data and young people or consumers are already regulated on the national as well as on the European level. This co-existence of European and German regulations is also institutionalized in the form of two antitrust and two patent organizations. German federal and state regulation covers mainly the fields of radio/media and publications, both of which are potentially harmful to young people. Over the last few years, the EU has issued several important directives concerning e-commerce. These include the applicable law in B2C e-commerce (country-of-origin rule), transparency in contractual terms
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
141
and conditions, the conclusion of contracts, data protection, liability for content, the use of electronic signatures and copyrights. The EU also aims at regulating data protection in international exchanges, which concerns mainly the protection of consumers and young people who, in the United States, are generally left to voluntary self-commitment. In 2000, the European Union and the United States joined together on the so-called ‘Safe Harbor Agreement’ concerning the protection of personal data that are transferred from the EU to recipients in the United States. In this instance, different European and US approaches constitute a general need for transnational regulation on the one hand, which led to the EU Data Protection Directive (Directive 95/46/EC) of 1998, and a patchwork of laws on state and federal levels complemented by self-regulation in the US economy on the other (Arkell 2001). On the supranational level, the protection of property rights has been pursued via an OECD directive on consumer protection. The rules of consumer protection set up there appear to be reasonable. However, they are not binding because the matters addressed fall under national jurisdiction. The World Trade Organization (WTO), especially through the General Agreement on Trade in Services (GATS), intends to deal with the new challenges arising from the emergence of e-commerce in subsequent rounds of negotiations. So far, the main outcome of the WTO meetings has been the TRIPS Agreement (Agreement on Trade-related Aspects of Intellectual Property Rights), which sets minimum standards for the protection and enforcement of intellectual property rights. The so-called ‘Ministerial Declaration on Global Economic Commerce’ took up a comprehensive work programme on e-commerce in 1998 to initiate the further evolution of GATT, GATS and TRIPS.8
How should e-commerce be regulated in the future? When examining the influence of regulations on the development of B2C e-commerce, we arrive at the somewhat paradoxical result that the problem is not over-regulation. In fact, adequate regulations are required in certain fields due to the nature of the new medium. Existing regulations – in Germany as well as in other countries such as the United States – do not take two problems adequately into account. First of all, many consumers mistrust the Internet because in their view the security standards are too low. Second, providers have to be sure that they receive payment for their products and services. E-commerce thus leads to specific legal problems that the existing framework does not deal with in a sufficient manner. The agenda for regulating different fields of e-commerce should mainly concentrate on three important aspects (Zerdick et al. 1999: 257): • • •
priorities and content of regulations; the regional level of regulation; and state or self-regulation.
Data protection
IT security
Telecommunications Act, Telecommunication Services Act (The Regulatory Authority for Telecommunications and Post) BSI Act – Gesetz über die Errichtung des Bundesamtes für Sicherheit in der Informationscechnik (Federal Office for Information Security – BSI) Data Protection Regulations (Federal Data Protection Act, Information and Communication
Media rights
Media Services Inter-State Agreement
Patent Law (Deutsches Patent- und Markenamt)
Property rights (patents, licences)
National German Commercial Code (Handelsgesetzbuch) Act Against Restraints on Competition (ARC) (Gesetz gegen Wettbewerbsbeschränkungen, GWB) (Bundeskartellamt)
Local–regional
Commercial transactions Competition policy
Policy fields
Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 on
EU Television Directive
Copy Right Directive, 22 May 2001; Convention on the Granting of European Patents (European Patent Office)
European Directives of GD Competition
EU
Table 7.3 Selected rules and regulations relevant to e-commerce and central institutions
US–EU Safe Harbour Agreement
GATT (General Agreement on Tariffs and Trade), GATS (General Agreement on Trade in Services), WTO (World Trade Organization) Treaties under the Umbrella of the World Intellectual Property Organization; TRIPS Agreement (Traderelated Aspects of Intellectual Property Rights)
Global
Concerted efforts on federal and state levels concerning publications that are morally harmful to young people
Source: author’s compliation.
Cultural and educational policy
Youth protection
Consumer protection
the Protection of Individuals; European Directive on Privacy and Electronic Communications of 12 July 2002; European Commission’s Directive on Data Protection of October 1998 Gesetz zum elektronischen E-commerce Directive of 8 June OECD Guidelines for Geschäftsverkehr (2001; enacts the 2000; Electronic Signatures Consumer Protection 2000 European E-commerce Directive); Directive of 13 December 1999; Telecommunications Customer Council Directive of 20 Protection Ordinance; Act on December 1985 to Protect the Digital Signatures; Act against Consumer with Respect to Unfair Competition enacted on Contracts Negotiated Away 3 July 2004 (therein: § 1 Against from Business Premises Spam), Distant Trade Act (85/577/EEC); Youth Protection Law EU Green Paper on the (Jugendschutzgesetz – JuSchG). Protection of Minors and Human Law Against the Diffusion of Dignity Media Harmful to Young Persons (Federal Department for Media Harmful to Young Persons; FSK (Freiwillige Selbstkontrolle der Filmwirtschaft); Board of Film Control) Uniform Fixed Retail Book Price Bill (Buchpreisbindungsgesetz)
Services Act), Voluntary IT Security Audit and Update Check, BSI-registered Protection Profiles
144
Lageman, Rothgang and Scheuer
Important issues with respect to the future development of e-commerce lie in creating reliable ‘framework conditions’, the protection of legal property rights and the possible emergence of monopolies.9 Obviously, the abovementioned fields require different regulatory responses depending on the level of regulation – state, self, national or international. The most important challenge is to establish ‘framework conditions’ that ensure frictionless transactions over the Internet (Schaaf 2002: 2). The new Internet medium brings supply and demand together. As a consequence, several aspects have to be settled, such as ensuring legal certainty, consumer data protection, electronic signatures and consumer protection. The open questions with respect to regulatory environment become more pressing in international transactions as different legal frameworks are involved. Legal certitude in B2C e-commerce poses no special problem as long as both buyers and sellers are in the same country, which is still the case in most B2C transactions by far. Here, national legislation applies, and for Germany there appears to be no reason to change the legal framework. For e-commerce on the European level, the EU directive on e-commerce enacted into national law in 2001 specifies the country-of-origin principle. Thus, in general the sellers do not have to orient themselves towards the country where they sell the goods. However, exceptions such as national consumer protection laws in a particular buyer’s country may still apply. As far as the fundamental legal framework is concerned, private solutions play only a minor role. However, markets for technical solutions that are closely connected to the legal framework, such as the provision of electronic signatures and encryption techniques, have emerged. Both services are provided by private firms and at the same time influenced by national legislation. Digital signatures are needed in order to be sure that a declaration of intent over the Internet can be attributed to one person without doubt (Grunwald 2001). With the signature directive, the EU has set the minimum standards for the use of signatures, thereby strengthening the diffusion of this service. For B2C e-commerce, however, digital signatures remain insignificant. The reason is that the expenses necessary for technical equipment are not worthwhile for most consumers who buy on the Internet. Along the same lines as signatures, encryption techniques aim to increase consumer trust in the Internet as a medium of transaction. They allow the sending of information through the Internet to proceed safely. In this case, the lack of state regulation in Europe entails a competitive advantage for producers of encryption software to the United States, where the sale of safe encryption methods is limited, so that the investigation of crimes is not obstructed (Zerdick 1999: 263). Consumer protection is an important arena of B2C e-commerce framework conditions as it is central to reducing reservations against the new medium for business transactions. The catalogue of important, general guidelines for consumer protection set by the OECD in 1999 (OECD 2003a) includes:
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
145
(1) fair business, advertising, and marketing practices; (2) clear information about the identity of the businesses engaged and the goods and services offered; (3) transparent procedure for confirming transactions; (4) secure payment mechanisms; and (5) protection of privacy. Here, again, the existing legal rules offer an appropriate framework for national transactions. For international transactions, however, at least minimum requirements should be set by regulations on the international level. In some respects, consumer protection and overcoming consumer reservations can also be accomplished by private security labels. In Germany, the privately organized association for product certifications (TÜV (Technischer Überwachungsverein)) assigns a certificate (called ‘s@fer shopping’) for online shops and virtual tourist agencies. Criteria for the certificate are, among others, data security and data protection. For governmental promotion of B2C e-commerce there are also alternatives to regulations such as providing information about the possibilities of B2C e-commerce. The protection of legal property rights is, of course, also an important issue in B2C. Computer technologies render it easy and costless to copy software programs, music, electronic books and movies. The shrinking possibilities of capturing the monetary benefits from these products may in fact reduce incentives to produce and supply them. The Internet offers opportunities to circumvent property rights and can possibly make this problem worse. An additional necessity to regulate this issue, however, does not automatically follow from this concern. The above-mentioned TRIPS Agreement offers one international minimum standard. Thus, the crucial question here is less how to regulate property rights protection and more how to enforce the already existing rules. Technical solutions for copyright protection should play an important role. The consequences of the ICT technologies for competition policy have already been studied at length (WBWT 2001; Picot and Heger 2003). Of course, ICT opens up new opportunities to gain competitive advantages. At the same time, commentators suspect that competitive advantages might also be more prone to erode as innovation cycles become shorter. Meanwhile, a major share of the B2C turnover is concentrated on big business firms, which raises the question of whether possible economies of scale in the storage and delivery of goods and services make a difference. However, one important competitive advantage is that large firms are able to build on consumer trust associated with well-known brand names. Their existing creditability reduces consumer reservations to buy goods over the Internet. In discussing possible consequences for competition policy, one has to keep in mind that a well-established regulation regime for monopoly control exists on the national, as well as to a lesser extent on the international level. No additional state regulations are thus necessary for B2C. Do we
146
Lageman, Rothgang and Scheuer
need more private regulatory mechanisms? Such developments have to be seen in the context of the total retail trade sector. B2C will play an important role in only some branches. Smaller firms with no established brands do indeed have a competitive disadvantage due to the lack of trust. Instruments that increase trust in transactions with these firms such as certificates should be able to partly overcome these disadvantages. In summary, we can say that there are several important questions regarding international e-commerce trade that should be settled in the future. Some of these regulatory issues are important only for the further development of e-commerce trade (such as consumer protection), some are important only for B2C transactions, and others (such as cryptography) apply equally to B2B and B2C e-commerce. Most of these regulations are required on the international level, while national administrations should transform the international regulations into federal law and make both compatible. In general, the ongoing tendency to erode federal regulation by international agreements should continue in the future. B2C e-commerce represents a field of regulation where national law will be losing importance over the long term. This observation applies to all industrialized countries. Thus, national differences in the regulatory environment will become less important for differences in B2C trade. Self-regulation ought to receive attention as a more flexible alternative to state intervention. There is no evidence that state regulation of details is needed in addition to very general framework conditions and minimum standards. Thus, one should be careful in setting new rules. Too much regulation may well impede further development.
Conclusions The regulatory framework is important for the development of new markets. This is also the case with respect to e-commerce. However, one should not invest too much hope in the regulation of economic activity as a means stimulating B2C development. As our considerations show, the existing regulatory environment is in many respects suitable to ensure market functions. The fact that e-commerce has, as yet, not developed in a more dynamic manner is mainly due to other factors, such as the technical weaknesses of Internet solutions, costs, the sluggish development of attractive web shops, the customer learning processes enabling full use of the Internet supply and existing and changing consumer preferences. It also seems quite possible that savings in transaction costs and economic welfare effects of B2C have occasionally been overestimated. At the same time, there are clearly fields of action where framework conditions for the Internet economy could be improved. Yet doubts remain as to whether this has to be done by the government. From our analysis it becomes obvious that most of the remaining problems could be solved in a more efficient manner by private initiatives and co-operative action between
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
147
market actors. Private firms and cooperative arrangements already provide security of Internet transactions and trust-creating devices, and this line of action should be pursued in the future. The task of the government would then be to establish general rules for electronic commerce without getting involved in detailed solutions. While these observations refer to e-commerce in general, national differences in e-commerce appear to be fairly persistent. The development of ICT and Internet sectors in general has been associated with varying levels of employment and productivity effects. The first-mover countries such as the United States have so far profited more than others, such as Germany. However, there are no indications that these developments have established a lock-in with respect to the use of e-commerce. It appears that the existing ‘backlog’ with respect to the United States can largely be explained by a time lag in the diffusion of the Internet.
Notes 1 2 3
4
5
6 7 8
The authors are grateful to Rainer Graskamp and Corina Zoch for their valuable help. Electronic transactions between firms are generally conducted via EDI networks, which often make use of a web-enabled gateway. However, they can also be based on other data exchange networks. The annual EITO report is published by the European Economic Interest Grouping (EEIG), a private research organization, with the support of the European Commission, DG Enterprise and DG Information Society and Media in Brussels and the OECD Directorate for Science, Technology and Industry in Paris. According to the online information of the company, the GfK Group (Gesellschaft für Konsum-, Markt- und Absatzforschung, www.gfk.com) is the No. 4 market research organization worldwide. The GfK Webscope survey is based on representative written surveys of 10,000 Internet users which are carried out both online and offline. There are also products that have played only a subordinate role in direct marketing up to now but that meet, in principle, the requirements of a successfully marketed product via the Internet and thus possess considerable sales potential. Currently, the best example for this group of products is motor vehicle parts. This is the result of a survey conducted in the context of the Online Shopping Survey (OSS), a joint survey by NFO Infratest and Enigma GfK (GfK 2003). Another prominent example is the highly regulated drug market in the United States, which is under increasing pressure from the less regulated and hence much cheaper market in Canada. LIBRO is one of the leading suppliers of office supply and entertainment (books, music, DVDs) in Austria. Its subsidiary, LIBRO online, tried to circumvent the contractual fixing of book prices. However, there is hardly much progress to be seen. There seems to be a political blockade in which, among others, efforts are made to find ways to maintain the difference between goods and services for digital transmissions. In its current form, the GATS Agreement is neutral concerning the technology by which the service is performed. Following this interpretation, e-commerce would already be subject to existing regulations. This is not beyond dispute and ongoing negotiations will show the extent to which the agreement’s existing regulations
148
9
Lageman, Rothgang and Scheuer
are appropriate for digital commerce (Arkell 2001). The process described here is a very slow one. There has been almost no progress over the last few years, as one can see by reading the respective e-commerce gate website of WTO accessed at www.wto.org/english/ tratop_e/ecom_e/ecom_e.htm on 17 February 2005. Other issues, such as tax policy and varying national regulations concerning consumer protection or national take-back obligations for producers, should become more important for B2C e-commerce as the share of international e-commerce trade increases.
Bibliography AGIREV (2003) Online Reichweiten Monitor ORM 2003 II, accessed at www. agirev.de/download/AGIREV_ORM2003_II.pdf on 14 November 2003. Arkell, J. (2001) Regulators Issues for Global Services Activities: how suppliers using new technologies may be impacted by the Rules of the General Agreement on Trade in Services, Paper presented to the 11th Annual International Conference of RESER, Grenoble, 25–26 October. Atrostic, B. K., Gates, J. and Jarmin, R. (2000) Measuring the Electronic Economy: current status and next steps, draft, Washington DC: US Census Bureau. BITCOM (eds) (2005) Daten zur Informationsgesellschaft. Status quo und Perspektiven Deutschlands im internationalen Vergleich, accessed at www.bitkom.org/files/ documents/BITKOM_Daten_zur_Informationsgesellschaft_2005.pdf on 15 November 2005. Börsenverein des deutschen Buchhandels (2004a) Buchpreisbindungsgesetz hat sich bewährt. Pressemitteilung 15.06.2004, accessed at www.boersenverein.de/de/69814 ?template=news_print on 15 March 2005. Börsenverein des deutschen Buchhandels (2004b) Buchpreisbindung, accessed at www.boersenverein.de/de/64603 on 17 March 2004. Bundesregierung (2002) Buchpreisbindung sichert notwendige Breite und Vielfalt des Angebots in Deutschland. Presse- und Informationsamt der Bundesregierung, Pressemitteilung Nr.141, 20.03.2002, accessed at www.bundesregierung.de/ dokumente/413.73184/Pressemitteilung/dokument.htm on 17 March 2004. Bundesverband des Deutschen Versandhandels e.V. (2005) Bestellen per Internet beliebt wie nie. Pressemitteilung des Bundesverbandes des Deutschen Versandhandels vom 08. März 2005, accessed at www.versandhandel.org/index.php?id= 96&tx_ttnews[tt_news]=52&tx_ttnews[backPid]=95&cHash=0143a6469a on 15 March 2005. Carr, N. G. (2004) Does It Matter? Information technology and the corrosion of competitive advantage, Boston MA: Harvard Business School Press. Christiansen, P. (2003) ‘Internet governance’, in Oberender, P. (ed.) Wettbewerbspolitik in der Internetökonomie (Schriften des Vereins für Socialpolitik, 292), Berlin: Duncker und Humblot, pp. 39–64. DeLong, B. and Froomkin, M. (1997) The Next Economy? Draft 1997, accessed at http://personal.law.miami.edu/~froomkin/articles/newecon.htm on 28 August 2001. DeLong, B. and Froomkin, M. (1999) Speculative Microeconomics for Tomorrow’s Economy, Draft 1999, accessed at http://personal.law.miami.edu/~froomkin/articles/ spec.htm on 28 August 2001. Donges, J. and Mai, S. (2001) ‘Elektronischer Handel im Kontext der Wirtschaftspolitik’, in Donges, J. (ed.) E-commerce und Wirtschaftspolitik, Stuttgart: Lucius & Lucius, pp. 1–8.
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
149
DTI (Department of Trade and Industry) (ed.) (1998) Regulating Communications: approaching convergence in the information age, London: DTI. EITO (European Information Technology Observatory) (ed) (2002) European Information Technology Observatory 2002, Frankfurt: EITO. EITO (ed.) (2003) European Information Technology Observatory 2003, Frankfurt: EITO. EITO (ed.) (2004) European Information Technology Observatory 2004, Frankfurt: EITO. EITO (ed.) (2005) European Information Technology Observatory 2005, Frankfurt: EITO. Enigma GfK/TNS Infratest (ed.) (2004) Online Shopping Survey, Welle 2, März 2004. accessed at www.enigma-gfk.de on 1 July 2004. EOS Gallup Europe (ed.) (2002) Internet and the Public at large: flash Eurobarometer 135, Brussels, accessed at http://europa.eu.int/comm/public_opinion/flash/fl135_en. pdf on 15 May 2003. Eucken, W. (1990) Grundsätze der Wirtschaftspolitik, 6, Auflage, Tübingen: Mohr. European Commission (2001a) The Impact of the E-economy on European Enterprises: economic analysis and policy implications, Brussels: COM(2001) 711 final. European Commission (2001b) Internal Market Scoreboard 8, accessed at http:// europa.eu.int/comm/internal_market/en/update/score/score8en.pdf on 15 May 2003. European Commission (2001c) ‘Council Resolution of 12 February 2001 on the application of national fixed book-price systems’, Official Journal of the European Communities, (2001/C 73/03), 6.3.2001, accessed at http://europa.eu.int/comm /avpolicy/legis/key_doc/legispdffiles/ book_en. pdf on 17 March 2004. European Commission (2004) Issues Relating to Business and Consumer E-commerce, 2003 – 3. Executive summary, Brussels, accessed at http://europa.eu.int/comm/ public_opinion/archives/ebs/ebs_201_executive_summary.pdf on 2 July 2004. European Parliament (2002) Committee on Legal Affairs and the Internal Market: report with recommendations to the Commission on the drawing-up of a directive of the European Parliament and of the Council on the fixing of book prices, rapporteur: Willy Rothley, 21 February, FINAL A5–0039/2002 (2001/2061(INI), accessed at http://www2.europarl.eu.int/omk/OM-Europarl?PROG=REPORT&L= EN&PUBREF=-//EP//TEXT+REPORT+A5–2002–0039+0+DOC+ SGML+V0//EN&LEVEL=3&NAV=S#Content45f930 on 15 March 2005. European Union (ed.) (1996) Green Paper on the Protection of Minors and Human Dignity in Audiovisual and Information Services, COM/96/0483FINAL. GFA Management (2002) Ex-ante Impact Assessment of the options outlined in the Green Paper on EU Consumer Protection, final report, Brussels: EU Commission, accessed at http://europa.eu.int/comm/consumers/cons_int/safe_shop/fair_bus_pract/ green_pap_comm/studies/gfa_report_en.pdf on 15 May 2003. GfK (2003) e-commerce mit Autozubehör an der Schwelle zum Milliardengeschäft, Pressemeldung 9. July 2003, accessed at www.enigma-institut.de/download/pd.oss-03-07-09-fin.pdf on 20 December 2006. GfK (2004) Online-Einkauf gewinnt an Marktbedeutung: Ergebnisse der Studie GfK Web*Scope zum Online-Kaufverhalten von Internet-Nutzern, Pressemeldung 06.09.2004, accessed at www.gfk.de/index.php?lang=de&contentpath=http%3A// www.gfk.de/presse/pressemeldung/contentdetail.php%3Fid%3D616 on 16 March 2005.
150
Lageman, Rothgang and Scheuer
Grunwald, A. (2001) ‘Rechtliche Rahmenbedingungen für den e-commerce: Viel Lärm um nichts?’, Economics, Deutsche Bank Research (ed.), No. 14, Frankfurt am Main. Hayek, F. W. (1972) Die Theorie komplexer Phänomene, Tübingen: Walter Eucken Institut, Vorträge und Aufsätze 36. HDE (2005) E-Business: Mehrheit der Branche erwartet steigende Umsätze. HDEPressemitteilung 9. März 2005, accessed at www.einzelhandel.de/servlet/PB /menu/ 1044106/index.html on 16 March 2005. Hudetz, K., Duscha, A. and Wilke, K. (2004) Kundenbindung über das Internet: Ergebnisse einer empirischen Studie, Ausgewählte Studien des EEC Handel 9, Köln: Institut für Handelsforschung an der Universität zu Köln. Hudetz, K. and Tanaskovic, N. (2004) Internet im Handel 20004: Status Quo und Entwicklungen. Ausgewählte Studien des EEC Handel 12, Köln: Institut für Handelsforschung an der Universität zu Köln. Klodt, H. (2001) ‘Und sie fliegen doch: Wettbewerbsstrategien für die Neue Ökonomie’, in Donges, J. (ed.) E-commerce und Wirtschaftspolitik, Stuttgart: Lucius & Lucius, pp. 31–48. Kollmann, T. (2004) ‘Vertrauensmanagement im E-Business’, in Wegweiser GmbH (ed.) eBusiness-Jahrbuch der deutschen Wirtschaft, Berlin: Wegweiser GmbH, pp. 30–3. Mentzel, I. (2003) Kaufverhalten und Kundenloyalität im E-Commerce – zwei empirische Untersuchungen, inauguraldissertation, Philosophische Fakultär der Ernst-Moritz-Arndt-Universität Greifswald, Greifswald, URN: urn:nbn:de:gbv: 9–200331–2. Mesenbourg, T. L. (2000) Measuring Electronic Business: definitions, underlying concepts, and measurement plans, Washington DC: US Census Bureau. NFO Infratest (ed.) (2002) Monitoring Informationswirtschaft, 5, Faktenbericht, München: NFO-Infratest. OECD (2002) Measuring the Information Economy, Paris: OECD, accessed at http://www.oecd.org/pdf/M00036000/M00036089.pdf on 16 May 2003. OECD (2003a) Consumers in the Online Marketplace: the OECD guidelines three years later, report by the Committee on Consumer Policy on the Guidelines for Consumer Protection in the Context of Electronic Commerce, JT001386–46, 3 February, accessed at www.olis.oecd.org/olis/2002doc.nsf/43bb6130e5e86e5fc 12569fa005d004c/af6ec39d8631ca3ac1256cc2005c0edf/$FILE/JT00138646.PDF on 16 May 2003. OECD (2003b) OECD Science, Technology and Industry Scoreboard 2003: towards a knowledge-based economy, accessed at www1.oecd.irg/publications.e-book/ 02–2003–04–1-7294/GB-03–2-2.htm on 2 July 2004. OECD (2004) OECD Information Technology Outlook: information and communications technologies, Paris: OECD. Petrauschke, B. and Kaumanns, S. C. (2003) ‘Informationstechnologie in Unternehmen’, WISTA 2/2003, Wiesbaden: 106–14. Picot, A. and Heger, D. (2003) ‘Braucht das Internet eine neue Wettbewerbspolitik?’, in Oberender, P. (ed.) Wettbewerbspolitik in der Internetökonomie (Schriften des Vereins für Socialpolitik, 292), Berlin: Duncker und Humblot, pp. 9–38. Preissl, B. (2003) ‘E-business dreams and facts: Ergebnisse einer Unternehmensbefragung’, in Fuchs, G., Purschke, I. and Teutsch, B., Fuchs, G., Purschke, I. and Teutsch, B. (2003) E-Commerce revisited. Workshop Dokumentation. Arbeitsbericht 239, Stuttgart: Akademie für Technologiefolgenabschätzung in Baden-Württemberg,
B2C e-commerce dynamics in Germany 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
151
accessed at http:///elib.uni-stuttgart.de/opus//volltexte/20004/1732 on 2 July 2004, pp. 1–19. Press and Communication Directorate-General, Public Opinion Analysis Unit (ed.) (2002) Public Opinion in Europe: views on business-to-consumer cross-border trade, Standard Eurobarometer 57.2, Flash Eurobarometer 128, Brussels, accessed at http://europa.eu.int/comm/consumers/cons_int/safe_shop/fair_bus _pract/green_ pap_comm/studies/eb57-fb128_final_report_en.pdf on 15 May 2003. Quah, D. (2000) The Weightless New Economy, paper presented at the EVA Seminar on 8 September, accessed at www.eva.fi/english/artikkelit/evakije01_2000.htm on 10 October 2001. Rothgang, M. and Scheuer, M. (2001) ‘Braucht e-commerce eine eigen ständige e-commerce-politik? Überlegungen aus ordnungspolitscher sicht’, TA-DatenbankNachrichten, Heft 4: 79–89. Schaaf, J. (2002) ‘Framework conditions for e-commerce: all in good order?’, E-conomics, Deutsche Bank Research, 24. Schnorr-Becker, S. (2004) ‘Moderne Informations- und Kommunikationstechnologien in Deutschland 1995 bis 2003’, Wirtschaft und Statistik, 7, 736–49. SIIA (Software & Information Industry Association) (ed.) (2000) SIIA’s Report on Global Software Piracy 2000, Washington DC: SIIA. Streit, M. (2001) ‘Wirtschaftspolitik in einem komplexen System’, Zeitschrift für Wirtschaftspolitik, Bd. 50 (1): 68–76. TNS Infratest (Taylor Nelson Sofres Infraset) IIE (Institute for Information Economics) (2004) Monitoring Informationswirtschaft, 7. Faktenbericht 2004: E-Commerce. Eine Sekundärstudie und Ergebnisse einer Expertenumfrage im Auftrag des BMWA München, April, accessed at http://193.202.26.196/bmwi/Faktenbericht_7/pdf/FB7_ EK_E-Commerce.pdf on 7 July 2004. UNCTAD (United Nations Conference on Trade and Development) (ed.) (2003) ECommerce and Development Report, accessed at www.unctad.org/en/docs// ecdr2003_en.pdf on 7 July 2004. US Census Bureau (ed.) (2004) Statistical Abstract of the United States: 2004–2005, the national data book, 124th edition, Washington DC: US Census Bureau. US Census Bureau (ed.) (2005) Quarterly Retail E-Commerce Sales, 4th Quarter 2004, United States Department of Commerce News CB-05–16, Washington DC: US Census Bureau. Varian, H. (2001) High-Technology Industries and Market Structure, working paper, accessed at www.sims.berkeley.edu/~hal/Papers/structure.pdf on 19 July 2004. WBWT (Wissenschaftlicher Beirat beim Bundesministerium für Wirtschaft und Technologie) (2001) ‘Wettbewerbspolitik für den Cyberspace’, Bertelsmann Briefe, Heft 145, Sommer: 4–8. Zerdick, A., Picot, A. and Schrape, K. (1999) Die Internet-Ökonomie. Strategien für die digitale Wirtschaft, European Communication Council Report. Berlin: Springer.
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Part III
Industrial organization, enterprise structure and ICT Japan, the United States and Germany in comparison
Subsection A Effects of ICT on industrial organization and on firm structures
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
8
ICT and corporate structure The diffusion of e-commerce across Japanese companies1 Dennis S. Tachiki
Introduction The 1990s is often referred to as the ‘lost decade’ in Japan, where the bursting of the asset bubble economy left it mired in a prolonged recession. In contrast, during the same period the ‘new economy’ in the United States and Europe is said to have taken off, spearheaded by highly innovative dotcom companies and a strong bull market. An International Data Corporation report (IDC 2000), however, paints a more mixed picture. Japan’s overall information society index score still ranks among the developed countries, while its Internet infrastructure and informatization scores are closer to those found in developing countries. The main research problem driving our analysis, then, is whether or not Japan lags behind other countries in the diffusion of e-commerce, and if so, what implications this lag has for the flagging Japanese economy. In answering this question it is easy to compare Japan with other countries, especially with the United States and the European countries. This approach, however, biases us towards a convergence model of economic development. If the United States is the exemplary model, for example, how can we account for the rapid increase in mobile commerce in Japan? Rather than assuming a strictly exogenous imperative, a recurrent question we ask is to what extent are endogenous factors also affecting the diffusion of e-commerce? Our analytical concern here is the extent to which e-commerce does and does not diffuse across industries and within establishments and the consequent impacts on firm performance. Our measure of e-commerce diffusion is based on revenues generated online over the Internet (CRITO (Center for Research on Information Technology and Organizations) 2002). The comparative merits and demerits of e-commerce against the existing Japanese style of management suggest it will diffuse unevenly across industries and within companies. For the past three years, the ECOM (Electronic Commerce Promotion Council of Japan) has been improving the definition of and data collection on e-commerce to create a more reliable database (ECOM 2002). Although the annual ECOM survey is getting better at capturing current trends, its
156
Dennis S. Tachiki
Table 8.1 Market size of e-commerce, 1998–2005a 1998b 1999
2000
2001c 2002c 2003c 2004c 2005c
Total (million yen) 8,685 12,656 22,414 35,511 46,781 66,304 90,925 141,727 B2Bd 8,620 12,320 21,590 34,027 43,950 61,270 78,430 125,430 B2B ratio (%)e 99.3 97.3 96.3 95.8 93.9 92.4 86.3 88.5 EC rate (%)f – – 4.1 5.0 6.6 9.2 11.5 14.1 B2Cg 65 336 824 1,484 2,831 5,034 12,495 16,297 B2C ratio (%) 0.7 2.7 3.7 4.2 6.1 7.6 13.7 11.5 EC Rate (%) – – 0.3 0.6 1.1 1.9 3.1 4.5 Source: ECOM 2002. Notes: a The definition of e-commerce used for the ECOM study is ‘the conduct of commerce’ (e.g. exchange of goods, services, information and money between suppliers and buyers, associated with the commercial transfer of assets between economic units) ‘through computer network systems using the Internet technology’ (i.e. using the TCP/IP protocol. Network lines include the Internet, extranet, Internet VPN (virtual private network) and dedicated IP lines), ‘the transactional values of which can be identified’ (i.e. giving quotations, providing information and other pre-order conduct are included as long as it is clearly identified that the conduct has led to purchase and/or sales order). b First year of survey. c Projected figure. d B2B e-commerce is where businesses/government bodies pay businesses in exchange for commodities (goods, services, information, etc.). It includes both B2G (Business-toGovernment), where government bodies pay businesses in exchange for commodities (goods, services, information, etc.), as well as the e-marketplace, where B2B e-commerce takes place on platforms used by multiple selling and buying enterprises. e B2B or B2C per cent of total e-commerce. f The proportion of e-commerce against the total interim demands and final demands for the applicable segment. g B2C e-commerce is where households pay businesses in exchange for commodities (goods, services, information, etc.), such as mobile e-commerce using mobile terminals and e-commerce involving pre-order stages (quotes, commissioning, etc.) for automobile, real estate, etc.
future forecasts often fail to foresee emerging ones. Moreover, the future forecasts look suspiciously similar to the government policy goals of the e-Japan Strategy. Despite these caveats, we find that the ECOM survey is still slightly more reliable than many of the forecasts generated by consultants. According to the ECOM annual survey, Table 8.1 shows that the projected size of the e-commerce market in Japan in FY2002 (fiscal year) should amount to nearly 47 trillion yen (ECOM 2002). This is a fivefold increase over the amount of e-commerce transactions in 1998, when this survey first began. By FY2005, the ECOM predicts that the e-commerce market should triple in size to 142 trillion yen. During this same period of time, the B2C market should account for 12 per cent of total revenues, up from 6 per cent in FY2002. In the following sections we disaggregate the B2B and B2C data to examine (1) the diffusion of e-commerce across industries and (2) the nature of Internet use within companies.
ICT and corporate structure 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
157
Diffusion across industries We seek evidence that the keiretsu (corporate groupings) fault lines across the industry sectors ease and channel the flow of B2B and B2C e-commerce. In addition, since the GEC Japan Database does not include ‘small-scale businesses’ (1–19 employees), we elaborate on the rise of Internet companies to flesh out this side of the diffusion story. Users of e-commerce According to a Ministry of Economy Trade and Industry survey, the diffusion of the Internet in Japanese companies has increased from less than 10 per cent to 96 per cent for ‘enterprises’ (>300 employees) and from 6 to 45 per cent for ‘establishments’ (<300 employees) since the early 1990s. Table 8.2 orders these individual companies into their industry subsectors and types of e-commerce (B2B or B2C) to examine the spread of e-commerce across industries. Business-to-business In Table 8.2 we see that the manufacturing sector accounted for most of the e-commerce transactions across the B2B market in FY2001. The top three manufacturing sub-sectors were electronic and information products (44 per cent), automobile (40 per cent) and industrial and precision machinery (3 per cent). Although the wholesale/retail sector is difficult to separate from the manufacturing-related component in the ECOM data, at face value the textile and sundry goods (2 per cent), food (2 per cent) and paper and office products (<1 per cent) are the most obvious candidates to be characterized as wholesalers rather than retailers. In the bank and finance sector, the financial and insurance services accounted for a nominal 0.03 per cent. Since the GEC10 Survey uses the standard industrial classification (SIC) to define the range of industries for inclusion in this study, we note that wood products, furniture, leather and ceramics industries are missing from Table 8.2. These sub-sectors correspond roughly to companies falling outside the inter-firm boundaries of the horizontal and vertical keiretsu. In short, the diffusion of B2B e-commerce roughly follows the contours of the horizontal and vertical keiretsu relationships and spills over into recently liberalized sectors. These findings combined with our previous discussion suggest that we find nominal e-commerce activities where establishments have weak ties to a keiretsu group and/or where market entry is still regulated. The ECOM data becomes more problematic in its future forecasts of e-commerce trends. By FY2006, ECOM projects that electronic and information products (25 per cent) and automotive sub-sectors (19 per cent) should continue to top the B2B market. Moreover, industrial and precision machinery should grow to 7 per cent, joined by chemical products
158
Dennis S. Tachiki
Table 8.2 Users of e-commerce Sector
2000
2001
2006
B2B Electronic/information products Automotive Industrial/precision machinery Iron/non-ferrous/raw materials Textile/sundry goods Food Transportation/travel services Chemical products Info processing/software-related services Construction Paper/office products Utility-related services Communications/broadcasting services Financial/insurance services Total
55.5 33.6 0.5 1.8 2.7 3.2 1.3 0.1 NA 1.3 <0.1 NA NA NA 21,590
44.3 39.7 2.8 2.6 2.4 2.4 1.6 1.3 1.1 1.1 0.4 NA <0.1 <0.1 34,027
24.7 18.7 5.6 6.6 8.8 6.5 5.1 5.7 1.8 11.5 4.1 0.7 0.1 <0.1 125,430
24.5 21.4 11.0 7.4 7.2 6.6 3.8 5.3 3.3 4.0 2.7 2.4 0.5 824 50
23.4 22.0 10.0 8.0 7.4 6.6 4.7 4.3 3.9 3.8 3.3 2.3 0.5 1,484 93
14.2 8.7 3.5 14.6 6.9 6.5 15.6 3.8 8.2 7.3 6.5 3.3 1.0 16,297 –
B2C Automotive Real estate PC and related goods Travel Entertainment Other products sales Other services Financial Clothing and accessories Food Hobbies/misc./furniture Books and music Gifts Total (Digital contents) Source: ECOM 2002.
Notes: Total percentages do not add up to 100 per cent due to rounding errors. NA = data not available.
(7 per cent) and iron and non-ferrous metals (8 per cent). This scenario implies greater integration of the supporting industries – i.e. the cluster of companies in the materials industries, material processing industries and associated industries (dyes and moulds, machine tools, founding and forging machinery, industrial furnaces, etc.) – at the foundations of the vertical keiretsu production networks. However, these sub-sectors are relevant to an export-oriented economy, and it is already clear that China and the other Asian economies will come to dominate the low and middle technologies in these sub-sectors, pushing Japan to move up the technological ladder to more knowledge-intensive technologies. This shift suggests that the ECOM
ICT and corporate structure 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
159
forecast underestimates the potential growth in the service sector as an important future B2B player. Our quasi-measures for the wholesale and retail sector also suggest greater growth in the textile and sundry goods (9 per cent), food (7 per cent), and paper and office products (4 per cent). These are the sub-sectors where intense competition and consumer preferences require companies to pursue greater rationalization of their distribution channels. Although bank and finance activities should increase as the economy improves, ECOM projects only a 0.03 per cent growth in its online business. Business-to-consumers As shown in Table 8.2, the retail and wholesale sector accounted for significantly more of the e-commerce revenues across the B2C market in FY2001 than the other two sectors in this study. The top wholesale and retail sub-sectors were automotive (23 per cent), PC and related goods (10 per cent), clothing and accessories (4 per cent), food (4 per cent), hobbies (3 per cent), and books and music (2 per cent). This list of sub-sectors essentially covers all the segments in the wholesale and retail sector, but the depth of this diffusion is shallow. In contrast, financial services (bank and finance) accounted for 4 per cent, while the manufacturing sector registered a very small percentage of B2C total revenues. By FY2006, ECOM projects that travel (from 8 to 15 per cent) will move near the top of the B2C list and replace real estate (from 22 to 9 per cent) as the top services sub-sector in B2C revenues. The automotive (23 to 14 per cent), PC and related goods (10 to 4 per cent) and finance services (stable at 4 per cent) sub-sectors will contract, but clothing and accessories (4 to 8 per cent), food (4 to 7 per cent), hobbies (3 to 7 per cent), and books and music (2 to 3 per cent) should continue to grow. While the ECOM forecasts are becoming more accurate, they tend to miss nascent trends, such as youthoriented markets. At the same time, the rapid aging of Japanese society suggests that other types of B2C services will emerge in the near future. This fundamental demographic trend is virtually ignored in the ECOM forecast. Thus, while we expect further growth in the above industries, we acknowledge the fact that the major players five years down the road may look very different. Locations and structure of e-commerce companies In this section we turn to the dotcom companies currently challenging the established brick-and-mortar business models of Japanese firms. Dotcom companies appeared out of nowhere in the 1990s to become an important part of the industrial landscape over the past decade. If one asks a knowledgeable person in Japan the question ‘What is a prototypical e-commerce company?’, most close observers would mention one associated with the Bit Valley.
160
Dennis S. Tachiki
Clusters of innovation Since 1999, one hot topic among Internet business people in Japan has been the Bit Valley scene, a community of the Internet entrepreneurial companies situated in the Tokyo ward of Shibuya. One reading of the Chinese characters for Shibuya is ‘bitter valley’, leading to the nickname Bit Valley. The word ‘bit’ is meant to convey the obstacles that these entrepreneurs have had to overcome (bitter), while ‘valley’ symbolizes their aspirations to create a Japanese version of Silicon Valley. Bit Valley led to the birth of the Bit Valley Association (BVA), a non-profit organization aimed at promoting personal contacts between people engaged in Internet businesses. A Fujitsu Research Institute survey reveals that there were 1,541 Internet companies in the 23 wards of Tokyo at the end of February 2001 (Yukawa 2003). Figure 8.1 depicts a detailed breakdown of the wards with the highest density of Internet companies. Out of the 1,061 companies responding to the survey, 426 were located in the five central wards of Tokyo, especially Shibuya-ku, Shinjuku-ku, Chiyoda-ku, Chuo-ku and Minato-ku. The increasing concentration of Internet companies in the AkasakaAoyama-Shibuya corridor of Bit Valley fame since 1994 is mainly explained by the unique social amenities catering to the lifestyle of young entrepreneurs. Yukawa (2001) identified several attributes of the locations where Internet companies cluster, including: (1) the fulfilling social amenities for the young
Ochiai/ Nakai 4
Shinjuku Ward
Hatsudai/ Hatagaya 26
Takadanobaba/ Okubo Waseda 6 14 Ichigaya/ Okub Kagurazaka 22
Shinjuku 77
Yotsuya 25
Chiyoda Ward Iidabashi/ Kudan 27 Imperial Palace
Bancho 47
Kanda 96
Otemachi/ Marunouchi 17 Nihombashi
Harajuku/ 60 Kasumigaseki/ Yoyogi Nagatacho 62 Ginza Tomigaya Aoyama Akasaka/ Roppongi 3 17 7 Shibuya 69 95 Ward Shinkawa/ Shibuya Shimbashi/ Tsukiji 121 Toranomon Ebisu/ 40 Azabu 32 43 Hiro-o 79 Shiba/ Shirokane / Hamamatsucho Mita Minato 51 21 Ward
Figure 8.1 Clustering of Internet companies in Tokyo Source: Yukawa 2003
Chuo Ward
ICT and corporate structure 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
161
generation; (2) the existence of affordable spaces; (3) the existence of artists; (4) the existence of related educational institutions; and (5) existing industries as the clients and the sources of human resources for Internet companies. These characteristics have developed in other parts of Tokyo as well, which has led to a revival of declining urban areas. For example, the Kanda area, with the third largest concentration of Internet companies in Tokyo, has become identified with the animation (animé) industry through which Japanese entrepreneurs have leaped to international fame in game software and animated pictures. Other clusters of Internet companies are located in Sapporo in northern Japan and in Fukuoka in the south. The increasing numbers of clustering cases bring into relief the classical importance of private–public partnerships that involve the government, universities and businesses as another contributing factor to these clustering of Internet companies (Yukawa 2001). In addition, this geographical dispersion of industry clustering outside the Tokyo and Osaka corridors is a healthy example of the ways in which economic activity is stimulated in the more rural areas of Japan. Organizational structure Approximately 24 per cent of the Internet companies were founded before 1990. The remaining 76 per cent came afterwards. Table 8.3 shows that the average Internet companies are capitalized at less than 30 million yen. Most have fewer than ten full-time and fewer than four part-time employees. According to capitalization and establishment size, these Internet companies would be classified as small-scale businesses. Since this type of business is not covered by the GEC Japan Database, we include them in our study in order to flesh out the contours of this emerging segment of the e-commerce economy. The main business activity of these companies is concentrated in the area of Internet application (55 per cent), such as consulting, website development, search engines, multimedia, software and databases. Intermediary companies that provide products and services are the next largest category, accounting for 22 per cent of the cases. For B2C-related business activities, marketmakers providing services such as online brokerage, travel and advertising characterize this category. Companies specializing in infrastructure (10 per cent) and e-commerce (8 per cent), which constitute the last two categories, represent the supply side of the story – the local capacity to implement e-commerce. The low level of business activities in the last category is due to the low level of enterprise integration activities among Japanese companies. Established e-commerce companies It is difficult to enumerate the number of domestic Internet companies established since 1994 – the year the commercial use of the Internet is claimed
162
Dennis S. Tachiki
Table 8.3 Organizational characteristics of Bit Valley Internet companies Establishments
Business activities
Founding (year) <1990 1990–1993 1994–1997 >1998
Per cent 23.7 12.8 30.4 33.0
Business
Capital (million yen) <10 10–29.9 30–49.9 50–99.9 >100
Per cent 8.9 39.5 8.7 11.0 31.9
Employees (#)
Per cent
Software and databases for the Internet
Full-time <4 5–9 10–24 25–49 >50
20.8 20.4 28.4 13.6 15.1
Market-maker Online broker, travel agency Portal Content provider Internet advertising
Part-time <4 5–9 10–19 >20
58.3 19.5 10.4 11.7
Infrastructure 9.5 %
Examples Telecommunications ISP Security provider Hardware for networking Data centre, ASP (active serve pages), payment system, etc.
Application 54.5%
Intermediary 22.3%
Consulting Creation of websites Search engines EC (electronic commerce) and multimedia applications
Content aggregator Research, etc. E-Commerce 7.8%
Online shopping Auctions, etc.
Source: Yukawa 2001.
to have started full scale in Japan. There are many websites that connect buyers and sellers, such as online recruitment. These intermediary businesses often run their websites from community sites, and in many cases their source of revenue cannot be easily identified. In the absence of data on all Internet companies in Japan, we limit our discussion to a survey of Internet companies in the Kanto region of Japan (Yukawa 2003). Local vs. global companies According to a survey by Yukawa (2003), most new Internet companies are actually hybrid domestic companies. There are only a few, purely domestic dotcom companies that dominate the local market as Amazon.com or eBay do in the United States. For example, the auction site operated by Yahoo Japan and the online bookstore Amazon.co.jp are the most popular websites in Japan. But Yahoo Japan is not regarded as a purely domestic
ICT and corporate structure 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
163
company because of its large equity stakes in Yahoo (US), and Amazon.co.jp is a Japanese offshoot website from its American parent company. In addition, Japanese dotcom companies tend to base their business models on those found in successful American Internet companies. Many of the Internet companies established by the pioneer Softbank Company, for example, are American versions adapted to the Japanese market. As for the types of domestic Internet companies established since 1994, Yukawa (2003) finds that most of them conduct B2C transactions. Table 8.4 shows that these companies are mainly in the wholesale and retail and the bank and finance sectors, but almost none of them are in the manufacturing sector. In the travel industry, market reorganization accompanies the diffusion of e-commerce. For example, Kokunaisen.com, which was jointly established by three large airline companies in 2000, deals with domestic air ticket bookings for all three companies. Given the rigid divisions separating the major keiretsu, this type of intra-industry cooperation was virtually inconceivable before the diffusion of e-commerce. Moreover, the bank and finance sub-sectors are fairly well represented. In the field of finance, most of the major finance companies have acquired large equity stakes in online brokers such as E*TRADE Securities and MONEX rather than attempting to establish their own subsidiaries. The major brick-and-mortar companies are now conducting e-commerce in such sub-sectors as online shopping and securities brokerage. Most of the established convenience stores (CVS) started building their own websites in the late 1990s to encourage online shopping. The keiretsu companies, especially the sogo shosha (general trading companies) and financial services companies, have been particularly active in this regard. For example, the largest CVS, 7-Eleven Japan, established 7dream.com in 2000 through joint capital investment from six large companies, such as NEC and the Sony Corporation. In the same year, Family Mart established Famima.com, a joint venture with other large companies such as Toyota Motor and NTT Data. The other CVSs are now following Seven-Eleven and Family Mart’s click-and-brick business model by building their own websites for ecommerce. This business model is now spreading throughout the retail sector, especially among department stores and various other types of retailers trying to fend off online competitors. Independent vs. subsidiary By fiscal year 2000, Sotec Company became the first Japanese company to surpass 10 billion yen in online sales. Notably absent from Table 8.5 are dotcom companies such as Rakuten and others that have attracted media attention. Instead, the list consists of well-known companies such as Sofmap, a PC (personal computer) manufacturer, and peripheral retailers in the Kanto region (Tokyo/Yokohama). One pull to their website is that consumers can earn points that can be used for future purchases. Rounding
164
Dennis S. Tachiki
Table 8.4 Major newly established dotcom companies (partial list) Year of establishment
B2B
<1998
1999
2000
Info Mart
EC-COM
MetalSite Japan Kouzai.com Smart Online Corp. Food Net
Oisix (97)
DeNa Price-down Engine e-Shopping! Books cbook24.com Book1.Inc Autobytel MPEC Flowerfarm Winetsu.com
B2C Manufacturing Wholesale and retail Auction Books Cars Flowers Food Games Cyber mall Photo printing Pharmaceuticals
GameOn Rakuten (97) ARSeed (97) Digipri (96) Online Store (95)
Bizseek Netprice eLady
e-seikatsu CURIOCITY
INDI
Publishing
Book-ing Digi-Book, Japan
Toys Travel
2001
e-Shopping! Toys ComNet Enterprise (97)
Tavigator
tabini
Skygate Kokunaisen.com Banking and finance banks Securities Commodity Futures Source: Yukawa 2003.
E*TRADE
Web lease E-LOAN Japan MONEX kabu.com JET Securities
SonyBank
e-Commodity
ICT and corporate structure 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
165
Table 8.5 Leading users of e-commerce, 2000 Company name
1 Sotec Company 2 Sofmap Company 3 Xing, Inc. (JOYSOUND)
Sales URL (million yen)
Main items
10,279 www.sotec.co.jp 9,536 www.sofmap.com 5,000 www.xing.co.jp
PCs, peripherals PCs, peripherals Online music distribution, cell phone call melody 4 Cecile Company 4,283 www.cecile.co.jp Clothing, sundries, etc. 5 Japan Air Systems Company 3,654 www.jas.co.jp Airline tickets 6 Entertainment Plus, Inc. 3,600 eee.eplus.co.jp Various tickets 7 Prince Hotels, Inc. 3,511 www.princehotels.co.jp Hotel reservations 8 Yodobashi Camera Company 3,421 www.yodobasi.com Consumer electronics 9 Kinokuniya Company 3,000 www.kinokuniya.co.jp Books 10 Nissen Company 2,427 www.nissen.co.jp Clothing, sundries, etc. 11 Sega Corporation 2,153 www.d-direct.ne.jp Game software, toys, amusement goods 12 Giga Networks, Inc. 1,837 www.giga.co.jp Online music distribution, cell phone call melody 13 Freeway Company 1,800 www.freeway.co.jp PCs, peripherals 14 Book Services Company 1,694 market.bookservice. Books co.jp 15 Murauchi Company 1,619 www.murauchi.co.jp PCs, peripherals 16 Nikkei Business Publications, 1,479 store.nikkeibp.co.jp Books Inc 17 Mytrip Net Company 1,087 www.mytrip.net Hotel reservations 18 Fancl Corp. 1,000 www.fancl.co.jp Pharmaceuticals, cosmetics 19 Laox Company 1,000 www.laox.xo.jp PCs, peripherals 20 Nihon Ryokou Kurabu 1,000 www.jtam.co.jp Travel Tomonokai Co. Source: Nihon Keizai Shimbun 2002.
out the list are well-known companies such as Cecile, Japan Air Systems, Prince Hotels, Yodobashi Camera and Kinokuniya. In short, the top 20 companies deriving revenue through the Internet all share one common characteristic: they are relatively well-known, brick-and-mortar companies that have established click-and-brick business models. Most of the new Internet companies are not independent because the entrepreneurial environment in Japan is not well developed (Tachiki et al.
166
Dennis S. Tachiki
2002). In the absence of bank loans and alternative financial options, entrepreneurs finance their new companies from personal resources. For cash-strapped entrepreneurs, this leads to large cash drains on parents and relatives or an excessive financial dependence on large companies. This is a major reason why the Bit Valley area has not become a fully fledged cluster like Silicon Valley, where venture capitalists play active roles in coordinating innovation by matching people and new technologies. Moreover, the large companies themselves tend to establish their own subsidiaries to start new Internet businesses, a process that squeezes out some of the smaller players. The company listings on the MOTHER (Market for the High-Growth and Emerging Stocks) board should provide one indicator of the maturity of Internet companies in Japan. The MOTHER board was created in November 1999. Compared to the rules for listing on the first (large companies) and second (SMEs) sections of the Tokyo Stock Exchange (TSE), the basic idea motivating the MOTHER board is to ease the minimum capitalization and number of profitable year requirements before an entrepreneur can become listed on the stock exchange. Conventional wisdom suggests that a new stock exchange member should have a critical mass of around 50 listed companies. Yet by 2001, only 26 companies were listed on the MOTHER board (MOTHER 2001). Moreover, the NASDAQ Japan has withdrawn from its collaboration with the Osaka Stock Exchange because the market has not grown as fast as originally anticipated. These developments will hinder the creation of Internet companies that are independent from the pull of the keiretsu companies.
Diffusion of e-commerce The diffusion of e-commerce across the three industry sectors essentially follows two paths. First, the most global-oriented keiretsu companies are the major carriers of B2B e-commerce across the manufacturing sub-sectors and then into the vertical keiretsu in the wholesale and retail, and the bank and finance sectors. Second, the domestic-oriented wholesale and retail sector emerges unexpectedly as the most active in B2C e-commerce. We found that the liberalization and deregulation of specific sub-sectors creates economic space for SMEs, especially Internet companies, to flourish online. After an initial euphoric take-off period, access to capital has been a constraining factor in their further growth, forcing them into the orbits of the keiretsu companies. Adoption within companies In order to uncover evidence about the ways in which e-commerce technologies are replacing or complementing existing business practices to meet online B2B and B2C customer demands we use the CRITO GEC
ICT and corporate structure 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
167
Japan Database to develop this analysis one step further. If customer demand is the main driver of Internet use in Japan, then the ‘voice of the customer’ is the point of departure for understanding the adoption of Internet use within Japanese companies. In other words, Internet activity starts with the sales and after-sales segments of the value chain closest to the customer, in contrast to the conventional wisdom of new product development beginning with R&D (Akao 1990). Since the CRITO survey does not cover all segments of the value chain, we focus only on customer services, sales, distribution channels and procurement segments. Customer services One common definition of a customer in Japan is ‘the person(s) or organizational unit that is the next step in your process for taking a product or service from concept to market’ (Tachiki 2000: 3). Table 8.6 shows that at 15 per cent, Japanese companies are half as likely to provide both B2B and B2C online services to customers as the global average of 33 per cent. Moreover, they are slightly more likely to provide only B2B (14 per cent) or only B2C (19 per cent) services to customers than the global averages (11 and 13 per cent respectively). For those companies that conduct B2B or B2C, the percentage of online transactions for business services (15 per cent) and customer services (6 per cent) is roughly similar to the global averages of 11 and 8 per cent. These rather unremarkable findings become more insightful when we consider that in the manufacturing sector, the customer is the next business unit in moving from the forging of raw materials into parts and then assembling the components into a final product (i.e. supplier-assembler relationship), and the wholesale/retail and the bank/financial sectors are essentially directly involved with the individual consumer. Consistent with these sector definitions of a customer, the adoption of online services across the industries follows along the globalization fault lines in the Japanese industrial landscape: B2B online services are more likely to develop in the manufacturing sector (54 per cent) followed by the wholesale and retail sector (22 per cent), and finally in the bank and finance sectors (19 per cent), whereas B2C online services follows the reverse pattern beginning with the sectors with the most direct consumer contact (bank and finance at 35 per cent and wholesale and retail at 23 per cent), and then manufacturing (4 per cent). Moreover, large companies are more likely to provide online services than SMEs. This overall pattern of online services underscores the fact that the B2B and B2C stories are different and re-confirms our earlier general finding that Japanese companies use the Internet for special purposes according to business functions rather than systematically integrating it across business units. Table 8.6 distinguishes types of online service use by industry sector. This allows us to understand why online service is more segmented than
45.4 22.5 21.2 14.3 33.4 83.7 76.1 54.4 12.3 14.0 60.1 31.3 53.4 38.0 11.1 87.7 87.7 36.0
29.3 14.1 18.6 5.6 14.5 94.0 74.6 47.3 24.6 27.4 99.1 49.3 1.2 0.3 0.3 65.3 65.3 39.8
93.3 74.7 47.7 23.8 26.6
53.6 37.6 3.8 1.1 24.0
98.1 48.8 2.6 1.3 0.6
22.3 0.7 23.0 10.6 11.7
67.8 39.4
67.8
18.5 2.0 35.4 7.0 27.0
Wholesale, Bank and retail and financial distribution services
Manufacturing
SME
Large
Industry sector
Establishment size
67.8 39.4
67.8
98.1 48.8 2.6 1.3 0.6
93.3 74.7 47.7 23.8 26.6
29.8 14.5 18.7 6.0 15.2
Japan
Total
57.3 52.0
53.9
69.8 20.6 48.6 21.7 21.3
79.9 54.7 54.4 17.0 21.5
23.1 11.0 12.9 7.6 33.3
Global
Source: CRITO Global E-Commerce Survey 2002. Notes: See notes a–d for Table 8.1. e Percentages are based on the full sample (all establishments). Exact wording of question: ‘Are these online services to other businesses or to consumers or to both?’ f Percentages are based on the full sample (all establishments). Exact wording of question: ‘What per cent of your establishment’s total services to businesses are conducted online?’ g Percentages are based on the full sample (all establishments). Exact wording of question: ‘What per cent of your establishment’s total services to consumers are conducted online?’ h Percentages are based on only those establishments that have a website and conduct business within the specified sector.
Type of online servicee % B2B only Mean % of online business servicesf % B2C only Mean % of online consumer servicesg % both B2B and B2C % of mfg websites which supporth Product specification Product configuration Service and technical support Account information Order tracking % of WRD websites supportingh Product catalogue Gift certificates and/or registry Product reviews Account information Individual customization % of B/F websites supportingh Online services (e.g. filing applications, claims, paying bills, transferring funds) Access to account information Online tools (research and planning tools, etc.)
% indicating a significant factor
Table 8.6 Online services
ICT and corporate structure 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
169
integrated in Japanese companies compared to global averages. The manufacturing sector uses its website more for B2B procurement transactions (release information on product specifications at 93 per cent) and product configuration (75 per cent) than for back-office functions such as service and technical support (48 per cent), account information (24 per cent) and order tracking (27 per cent). The wholesale and retail distribution sector focuses primarily on the marketing side of B2C transactions (product catalogue at 98 per cent and gift certificates and/or registry at 49 per cent), but noticeably less so for sales functions such as product review (3 per cent), account information (1 per cent) and individual customization (1 per cent). The bank and finance sector packages online services (68 per cent) with access to account information (68 per cent) for sales functions, but less so for marketing functions such as providing online tools (40 per cent). The story that emerges from Table 8.6, regardless of industry sector, is that companies provide online services to meet their immediate customer demands, but they are less likely to provide online services where privacy is paramount. This finding elaborates on our earlier insight concerning the defining factors that divide the drivers of and obstacles to e-commerce. The story changes somewhat according to the size of the establishment. SMEs in the manufacturing sector clearly conduct more of their customer services online, both for procurement and back-office functions, than larger companies. For the wholesale and retail sector, the SMEs are more active in conducting sales related to product catalogue and gift certificates, but the large (wholesale) companies provide greater access to product reviews, account information and individual customization. In the bank and finance sector, the SMEs (regional banks, trust banks, securities) are less likely than large keiretsu banks to conduct customer service over their websites, but they are slightly more likely to provide access to online tools. SMEs in the manufacturing sector and to some extent in the bank and finance sector are moving online to allow the voice of the customer to reverberate electronically further back into their value chain, whereas the wholesale and retail sectors use their websites to reach out to new customers. Anecdotal evidence suggests that both of these cases represent SMEs that fall outside a keiretsu nexus and/or ones that fall within a segment of the economy undergoing liberalization or deregulation. Nevertheless, further research beyond the GEC Japan survey is necessary to development this storyline. Sales With regard to moving the voice of the customer electronically back into the sales segment of the value chain, Table 8.7 shows that Japanese respondents report less online sales to both businesses and customers (13 per cent), to businesses only (7 per cent) or to customers only (1 per cent) compared to the global averages (15, 13 and 7 per cent respectively). Nevertheless, when we limit the sample to only those Japanese companies actually
13.0 14.1 5.1 1.4 8.9
2.7 11.4
71.1
12.8 7.0 1.3 5.0 36.6
3.0 15.2
96.6
100.0
2.6 8.5
16.1
0.6
2.1 28.5 1.7
64.2
3.3 19.8
39.6
6.6
16.3 0.5 0.5
31.6
0.1 1.2
3.2
0.7
16.0 0.2 13.3
Wholesale, Bank and retail and financial distribution services
Manufacturing
SME
Large
Industry sector
Establishment size
94.1
3.0 15.1
35.6
4.9
12.8 7.2 1.4
Japan
Total
33.6
4.0 15.1
18.6
3.8
15.0 12.9 7.1
Global
Source: CRITO Global E-Commerce Survey 2002. Notes: see notes a-d for Table 8.1. e Percentages are based on the full sample (all establishments). Exact wording of question: ‘Are these online sales to other businesses or to consumers or to both?’ f Exact wording of question: ‘What per cent of your establishment’s total consumer sales are conducted online?’ g Exact wording of question: ‘What per cent of your establishment’s total business-to-business sales are conducted online?’
Type of online salese % both B2B and B2C % B2B only % B2C only B2C Mean % of total consumer sales conducted online (all establishments)f Mean % of those only doing B2C sales onlinef B2B Mean % of total business sales conducted online (all establishments)g Mean % of those only doing B2B sales onlineg Web payment % websites that support online payment (only those doing online sales)
% indicating a significant factor
Table 8.7 Online sales
ICT and corporate structure 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
171
conducting B2C, we find that they are almost twice as likely to conduct sales online than the global average (36 versus 19 per cent). Turning to the case of B2B, Japanese companies (15 per cent) are just as likely as the global average (15 per cent) to conduct business sales online. In addition, 94 per cent of the Japanese companies support online payments through their websites compared to 34 per cent for the global average. This suggests a bimodal split in the use of e-commerce among Japanese companies: on the one hand, we find a large majority of companies conducting very little B2C and B2B sales online, while on the other hand the data reveals that a distinct minority of companies intensively conduct on average a third of their B2C sales online with a strong link to their accounts receivable system. The data on industry sector and establishment size provide some insight into where online sales activities are most advanced. Table 8.7 shows that when the sample is limited to only those conducting B2C or B2B sales online, the wholesale and retail sector conducts a greater percentage of B2C (40 per cent) and B2B (20 per cent) than the other two sectors. The wholesale and retail sector (64 per cent) also reports a high percentage of website support for online payment. In contrast, the manufacturing sector conducts only 16 per cent of its B2C and 9 per cent of its B2B sales online, yet it backs up these sales 100 per cent with an online payment system. These figures suggest that these two sectors are reorganizing their sales and payment activities to an online system: the wholesale and retail sector for both sales and payment and the manufacturing sector with hybrid EDI (electronic data interchange) and Internet networks and payment systems. In particular, the SMEs are more likely to engage in such activities than large companies. It is thus the SMEs in the wholesale and retail sector that are the most active minor players using Internet-based networks for B2C online sales. Distribution channels In Table 8.8 Japanese companies report ‘competing directly with traditional distribution channels’ (38 per cent) as the main reason for using the Internet to sell products and services, followed by another 29 per cent indicating that they plan to use the Internet to replace traditional distribution channels. In other countries this is known as ‘channel conflict’, but in the Japanese context it is a way of getting around channel bottlenecks. The distribution system in Japan is quite hierarchical, consisting of more than three intermediaries between producer and customer. Indeed, foreign companies often cite the complex, multi-layered distribution system as a major structural impediment to doing business in Japan. The remaining companies strive only to enhance their traditional distribution channels (22 per cent) or expand their distribution channels using the Internet (12 per cent). In this segment of the value chain, then, the voice of the customer becomes a function of improving quality, reducing cost and decreasing delivery time (Tachiki 1990).
30.0
Replace traditional distribution channels
11.4
Address new markets only
14.1
29.5
18.5
37.8
24.6
48.8
3.0
23.6
0.9
3.7
47.5
47.9
32.7
31.5
16.6
19.2
11.5
21.8
29.4
37.2
Japan
Total
15.3
44.1
13.2
27.4
Global
Source: CRITO Global E-Commerce Survey 2002. Notes: See notes a-d for Table 8.1. e Exact wording of question: ‘Which of the following statements best characterizes how you are using the Internet to sell products and services?’
21.4
Address traditional distribution channels only
Enhance or expand channels
37.2
Compete with traditional distribution channels
Channel conflict
% indicating Internet used to . . .e
Wholesale, Bank and retail and financial distribution services
Manufacturing
SME
Large
Industry sector
Establishment size
Table 8.8 How establishments use the Internet to sell products and services
ICT and corporate structure 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
173
Among the three sectors, the manufacturing sector has the greatest number of distribution layers between a company and its customers. This sector primarily uses the Internet to enhance traditional distribution channels (49 per cent), but it is less likely to replace traditional distribution channels (3 per cent). Consequently, the EDI supplier-manufacturer networks remain relatively intact. Nevertheless, they are moving towards the use of the Internet in the downstream segments of their value chain to distribute products and services. When we look downstream at the retail side of the story, we find that supermarkets and department stores dominate the sector, followed by speciality stores, convenience stores and finally cooperatives. The supermarkets and department stores are using the Internet to procure fresh and/or reasonably priced products directly from producers for consumers. For example, Aeon (formerly Jusco Company), a leading retailer, bypasses wholesalers and orders goods directly from domestic and overseas producers. Speciality stores, once a vibrant sector, finds that younger consumers are turning to discount stores or to the Internet for computers, music, books, etc., forcing them to adopt a click-and-brick business model. Squeezed between manufacturers and retailers, then, is the wholesale sector that is responding the strongest to the channel conflict questions. The biggest threat to wholesalers is ‘disintermediation’ from the distribution process. In this connection, not only do online purchases pose a threat to their intermediary role but the liberalization of this sector (e.g. Large and Small Store Law) has also led to the emergence of competitive challenges from direct marketing (telephone call centres, catalogue orders, etc.) and large mega-stores (e.g. Carrefour, Costco, etc.). In response, wholesalers are increasingly purchasing private brands by importing from China and other overseas vendors to bypass high-cost domestic producers (JETRO (Japan External Trade Organization) 2003). The bank and finance sector reports less channel conflict than the other two sectors. However, the ‘big bang’ financial liberalization of the sector in the mid-1990s opened the door to market entrance from non-bank bank competitors. The Japanese government has historically favoured debt financing over equity markets and has thereby restricted market entry through its monetary policies. Subsequent to the liberalization of this sector and the rise of e-commerce after 1994, non-bank banks, businesses in sectors other than financial services providing banking services such as IY Bank and Sony Bank, have been making headway in the area of retail banking and securities, which in turn has required traditional banks and financial services to protect and expand their market share. Consequently, companies in this sector are more likely to use the Internet to enhance traditional branch distribution channels (32 per cent) or to expand into new markets (33 per cent). Overall, the initial impacts of both the Internet and liberalization have led to a chain reaction that has spread across the three sectors and has gradually flattened out and internationalized the previously hierarchical domestic distribution channels in Japan.
174
Dennis S. Tachiki
Procurement When we move further up the value chain, we see in Table 8.9 that customer demand for Internet-based transactions is weaker in Japan than the global average: only 33 per cent of Japanese companies purchase online compared to the global average of 51 per cent. Much of this lag lies downstream in B2C purchases and in the integration of suppliers in B2B purchases. Whereas the manufacturing sector is the most active in upstream online purchasing, nearly half of its procurement online is parts for production. At the centre of a manufacturing company’s procurement segment of the value chain is some derivative of Toyota Motor’s just-in-time (JIT) and the kamban delivery system (Monden 1983). Under this procurement system, companies decide whether to use an open or closed procurement system depending on the product architecture. For products with a modular design, i.e. products with standardized, mass-produced components, online procurement is an option. But for integrated product designs, such as products with high-tech core components, a closed EDI system is the most secure way to protect intellectual property (Fujimoto 2002). Japanese companies tend to use closed EDI networks for integrated product designs but are more catholic about modular product designs. The subdued pattern of online procurement activity is thereby due to the existence of hybrid EDI and Internet-based networks for procuring parts for production. The wholesale and retail sector (26 per cent), and bank and finance sector (26 per cent) are half as likely as the manufacturing sector to purchase online. This outcome for the wholesale and retail sector is consistent with our earlier finding that it is a heavy user of EDI networks. Nevertheless, the GEC Japan Database unaccountably shows no online purchasing for resale goods. For the bank and finance sector, in contrast to its active use of the Internet to reach external customers, its nominal amount of online orders for supplies and equipment suggests fewer intra-firm, online business activities. Adoption of e-commerce E-commerce has spread within companies in segments of the value chain closest to the customer. However, companies that spend the most on ICT and are prepared to conduct e-commerce are not necessarily the largest adopters of the Internet. Harnessed to their EDI legacies, the large keiretsu companies have adopted a hybrid, open and closed network. Moreover, in the relative absence of privacy and security for Internet transactions, the large keiretsu companies have not integrated the Internet across their business functions. The organizational boundaries of Japanese companies may still be mapped according to the keiretsu intra- and inter-firm relationships, except at the customer interface and distribution segments. However, the rise of Internet support services and company interest in outsourcing business processes could possibly induce change in keiretsu relationships in the future.
21.1 0.0 0.1
Mean % spent on parts for productione
Mean % spent on goods for resalef
Mean % spent on supplies and equipment for business ordered onlineg
0.5
1.4
7.5
45.4
0.0
–-
20.0
54.5
0.0
0.0
–-
25.9
1.6
–-
–-
26.4
0.01
0.0
20.0
32.8
Japan
Total
8.3
6.8
8.3
50.8
Global
Source: CRITO Global E-Commerce Survey 2002. Notes: See notes a–d for Table 8.1. e Question asked only to those in the manufacturing sector; percentage based on all manufacturing establishments. Exact wording of question: ‘What per cent of the money your establishment spends on direct goods for production, such as parts and components, is ordered online?’ f Question asked only to those in the wholesale/retail distribution sector; percentage based on all wholesale/retail establishments. Exact wording of question: ‘What per cent of the money your establishment spends on goods for resale is ordered online?’ g Percentage based on all establishments. Exact wording of question: ‘What per cent of the money your establishment spends on supplies and equipment for doing business is ordered online?’
32.4
Wholesale, Bank and retail and financial distribution services
Manufacturing
SME
Large
Industry sector
Establishment size
% doing online purchasing
Table 8.9 Online procurement
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
176
Dennis S. Tachiki
Impacts of the Internet and e-commerce The globalization and liberalization of the Japanese economy provides a good picture of the fault lines within the industrial landscape. In this section, we conclude our discussion by examining the impacts of the Internet and e-commerce on the performance of Japanese companies that fall on either side of these fault lines, especially in the areas of efficiency, coordination and commerce. Efficiency In Table 8.10 we see that Japanese companies are less likely to experience efficiency in internal processes (29 per cent) and staff productivity (24 per cent) than the global averages of 34 and 27 per cent respectively. When analysed according to industry sector, the manufacturing sector beats the global average on internal process efficiency (41 versus 34 per cent). In our discussion of the GEC Japan Database, the manufacturing sector uses the Internet to rationalize operations management functions. We would have expected a higher degree of improvement in internal efficiencies of the distribution channels for the wholesale and retail sector, and for the backroom operations in the bank and finance sector. The bank and finance sector is low on both internal process efficiency (21 per cent) and staff productivity (12 per cent). The large companies (32 per cent) are more likely to achieve internal process efficiency than the SMEs (29 per cent), but the SMEs (24 per cent) are slightly more likely to report increases in staff productivity than the large companies (23 per cent). Coordination With regard to coordination measures, Japanese companies are less likely than the global average to report decreases in procurement costs (4 versus 18 per cent) and decreases in inventory costs (5 versus 14 per cent). Nevertheless, they do report more improvement in coordination with suppliers (34 versus 30 per cent). This is an area where the manufacturing sector has made improvements, and these improvements show up when we examine the data by industry. The manufacturing sector equals the level of the global average on procurement costs and clearly exceeds it on inventory costs and coordination with suppliers. We attribute this result to the hybrid open and closed networks. In other words, Japanese companies have only adopted open e-commerce technologies to the extent that they improve on existing business practices. Where this is not the case, they still rely on closed EDI networks. The wholesale and retail sector only outperforms the global average with regard to the ‘coordination with suppliers’ measure, a segment of the value chain in which we documented significant changes with the introduction of e-commerce technologies. The bank and finance sector reports
3.1 1.1 5.0 10.1 10.9
3.9 5.4 34.0
28.6 24.3
12.3 6.9 5.8 9.1 17.9
12.3 4.3 27.9
31.5 22.7
9.3 1.4 20.6 14.6 42.3
16.2 20.3 40.4
40.5 25.7
0.8 0.4 0.0 8.8 0.8
0.3 0.3 33.2
25.2 24.6
12.3 13.5 0.0 6.1 6.9
0.8 5.0 10.5
20.7 11.5
3.4 1.2 5.0 10.1 11.2
4.2 5.3 33.8
28.7 24.3
Japan
Total
31.4 20.5 19.5 29.8 34.8
17.7 14.0 29.8
33.9 27.2
Global
Source: CRITO Global E-Commerce Survey 2002. Notes: See notes a–d for Table 8.1. e Exact wording of question: ‘Using a 5-point scale where 5 is “a great deal” and 1 is “not at all”, please rate the degree to which your establishment has experienced the following impacts since it began using the Internet for business’. A score of 4 or 5 was classified as ‘high impact’.
Commerce Sales area widened Sales increased International sales increased Competitive position improved Customer service improved
Coordination Procurement costs decreased Inventory costs decreased Coordination with suppliers improved
Efficiency Internal processes more efficient Staff productivity increased
% indicating a significant factore
Wholesale, Bank and retail and financial distribution services
Manufacturing
SME
Large
Industry sector
Establishment size
Table 8.10 Impacts of conducting business online
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
178
Dennis S. Tachiki
low results on all three measures. Nevertheless, internal coordination is not a significant issue in this sector. Instead, its more pressing issue is meeting new competitive challenges from non-bank banks (e.g. Sony Bank and IY Bank, etc.). The SMEs are more likely to report decreases in inventory costs and coordination with suppliers than the large companies. Commerce Japanese companies have not benefited as much as the average global company on the measures of commerce, i.e. widening of the sales area, increased sales, increased international sales, improved competitive position and improved customer service. Only the bank and finance sector shows improvement in widening sales area and increasing sales – a key reason companies give for adopting e-commerce. The wholesale and retail sector reports an improvement in its competitive position and an improvement in its customer service. Since the wholesale and retail, and bank and finance sectors focus their e-commerce in downstream activities, we would have expected more improvement in these sectors. Perhaps a combination of their domestically oriented market focus and the poor state of the Japanese economy has muted the potential positive impact of the Internet. The manufacturing sector exceeds the global average on improved customer service and increased international sales. In terms of establishment size, large companies have benefited more than the SMEs. Nevertheless, none of the measures exceed the global averages. Because the diffusion of ecommerce tends to enhance the efficiency of each trade, a large company’s full-scale entry into e-commerce promotes the market reorganization of each trade.
Notes 1
This research is part of the Globalization and E-Commerce Project of the Center for Research on Information Technology and Organizations (CRITO) at the University of California, Irvine. The material is based upon work supported by a grant to CRITO from the National Science Foundation (CISE/IIS/DST, Grant No. 0085852). Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.
Bibliography Akao, Y. (1990) Quality Function Deployment: integrating customer requirements into product design, Portland OR: Productivity Press (originally published as Hinshitsu tenkai katsuyo no Jissai by the Japanese Standards Association in 1988). CRITO (Center for Research on Information Technology and Organizations) (2002) Global E-Commerce 10-National Survey Database, Irvine CA: CRITO, University of California at Irvine.
ICT and corporate structure
179
ECOM (Electronic Commerce Promotion Council of Japan) (2002) Market Survey of E-Commerce 2001 in Japan: current situation in 2001 and future outlook through to 2006, Tokyo: ECOM, Ministry of Economy, Trade and Industry, and NTT DATA Institute of Management Consulting. Fujimoto, T. (2002) ‘Relocation and reorganization of Japanese industries: automobiles’, in Tachiki, D. S. and Kurihara, J. (eds) Can Japan Be a Global Player? Industry and national perspectives, Tokyo: Fujitsu Research Institute. International Data Corporation (IDC) (2000) The 2000 IDC/World Times: Information Society Index, www.idc.com: 8080/Data/Global/isi/isiMain.htm JETRO (Japan External Trade Organization) (2003) White Paper on Trade and Investment, Tokyo: JETRO. Monden, Y. (1983) Toyota Production System, Norcross GA: Institute of Industrial Engineers. MOTHER (2001) ‘Mothers Monthly Report’, Tokyo: Tokyo Stock Exchange. Nihon Keizai Shimbun (2002) Japan Economic Almanac 2002, Tokyo: Nihon Keizai Shimbun. Tachiki, D. S. (1990) Total Quality Control: the Japanese approach to continuous improvement, Tokyo: Sakura Institute of Research. Tachiki, D. S. (2000) ‘What is TQM: a Japanese approach to continuous improvement’, unpublished paper for the Japan International Cooperation Agency (Tokyo). Tachiki, D. S., Hamaya, S. and Yukawa, K. (2002) E-Commerce National Environment and Policy Environment in Japan, CRITO Working Paper, Irvine CA: Center for Research on Information Technology and Organizations, University of California at Irvine. Yukawa, K. (2001) Survey of Internet Companies in Tokyo, Tokyo: Fujitsu Research Institute in collaboration with the Bit Valley Association, Tokyo Metropolitan Government, and Ministry of Economy, Trade and Industry. Yukawa, K. (2003) ‘A cluster of Internet companies in Tokyo – review of Bit Valley’, Journal of Korean Regional Science, 18 (3).
9
The rise and fall of ‘Wintelism’1 Manufacturing strategies and transnational production networks of US information electronics firms in the Pacific Rim Boy Lüthje
Introduction The extraordinary dynamics of the information and communications technology (ICT) industry during the decade of what was called the ‘New Economy’ have produced many myths. One of the most persistent surrounds the decentralization of technologies, corporate organization and production often associated with the demise of the Chandlerian model of the vertically integrated corporation (Chandler 1962). Neo-liberals like George Gilder (1988) have hailed the US microelectronics industry and its once revolutionary epicentre Silicon Valley as the ultimate example of the decentralization of innovative powers. Through its ever-exploding production of new technologies, it resembles a nineteenth-century world of small, innovative entrepreneurs at the frontiers of industrial development. Theorists of flexible specialization (Saxenian 1994) and network capitalism (Castells 1996) have analysed this scenario in terms of ‘technological communities’, in which cooperation in trust-based network relationships has become the ultimate power of innovation. This type of cooperation has also been a key factor for the recovery of US dominance in the global ICT industry against the massive challenge from vertically integrated multinationals from Japan in the 1980s. In spite of their very different theoretical and political leanings, these theories converge on the view that specialization in technologies and products calls for decentralization in business organization. The approach follows a simple equation of technological and institutional path dependence, which the editors of this volume call into question (for a theoretical discussion see Lüthje 2001a). Only few have understood that the meteoric rise of the key players in the New Economy, such as Intel, Microsoft, Cisco or Sun, was intrinsically linked to a highly complex recombination of the patterns of technological innovation and industrial reorganization. Further-
The rise and fall of ‘Wintelism’ 181 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
more, throughout this process of reorganization, both of these elements often developed along different and sometimes contradictory paths. From this perspective, then, the resurgence of Silicon Valley and the survival of the US computer industry in a ‘post IBM-world’ (Ferguson and Morris 1993) involved a distinct synthesis of vertical specialization in technology development on the one hand, and the development of new forms of mass production in global production networks on the other (Lüthje 2001a). The analysis of decentralization has generally been informed by three related concepts: ‘Wintelism’, global production networks and modular manufacturing. They all relate to the complex processes of vertical specialization and global consolidation that have reshaped the ICT industry since the 1980s (Ernst and O’Connor 1992). The term ‘Wintelism’ was introduced as an analytical category for the form of innovation and market control established by the key players of the New Economy. It refers to the defining of new markets by breakthrough innovations in product design, which has been supported by ‘open-but-owned’ standards that disseminate high-end commodity technologies (Borrus and Zysman 1997). In this context, manufacturing is no longer regarded as a core competency of innovative firms but rather as a ‘service’ performed by cross-national production networks made up of sub-assemblers and specialized component producers (Borrus 2000). In the key sectors of the Wintelist ICT industry, a new form of outsourced manufacturing emerged, called electronics Contract Manufacturing (CM) or Electronics Manufacturing Services (EMS). This type of production has also been described as a ‘new American model of manufacturing’ and an alternative to the Japanese keiretsu organization (Sturgeon 1997; 1999). However, this analysis and its underlying concepts have to be reassessed in the light of the massive crisis in the ICT sector since 2001, aptly described as the most severe recession in the history of the industry. The downturn opened a new round of accelerated restructuring characterized by a massive shift of manufacturing jobs from the United States to low-cost locations, increased outsourcing of product development and design (which has made many well-paid engineering jobs vulnerable to relocation as well), and the emergence of China as both a manufacturing powerhouse and a key growth market. Much of this change was driven by leading actors of the Wintelist segments of the ICT industry. They followed the trajectory of organizational innovation and learning processes created during the heyday of the New Economy. While this new round of restructuring has engendered a reshaping of manufacturing strategies, it nevertheless seems to involve an at least partial resurgence of vertically integrated models of innovation, market control and production. This trend, in turn, may be primarily attributed to the shift of innovation to Asia (Ernst 2003; Ernst and Lüthje 2003), in which Asian multinationals with distinctive strategies of vertical integration have persisted or emerged as global flagship companies in new growth markets in communications and consumer electronics.
182
Boy Lüthje
We examine these developments in five steps by concentrating on the production systems of major US ICT companies and their subcontractors in the Pacific Rim. We first review the basics of the Wintelist model of production and the role of electronics contract manufacturing within it. Second, we look at the causes of the current recession and the restructuring of global production networks in major US ICT companies driven by the crisis. Next we trace the massive, vertical reintegration of US-led production networks in the course of their relocation to Asia and China. Against this background we then discuss sectoral restructuring trends and the emerging variety of outsourced production models from the perspective of brandname companies, here designated as Original Equipment Manufacturing firms (OEMs), and focus on the current battle between US-based contract manufacturers and their competitors from Taiwan, who generally operate on a more sophisticated model of production services called Original Design Manufacturing (ODM). In a concluding section, we point to the analytical implications of the current discussions on modular manufacturing, and touch upon some consequences for innovation in the ICT industry and the broader balance of industrial power in the Pacific Rim.
Wintelism and the rise of transnational production networks Since the 1980s, the days of the ‘Japanese challenge’ to the US microelectronics industry (Borrus 1988), the US ICT industry has undergone permanent restructuring. In the course of its massive attempt to catch up with vertically integrated Japanese keiretsus and their followers from Korea, the industry was reshaped by a new brand of companies that specialized in hardware and software products for the PC industry, including computer servers, Internet equipment and various kinds of mobile computing devices such as laptops or handheld digital assistants. This type of specialized company was initially created during the 1970s in California’s Silicon Valley – particularly among the newly emerging ‘merchant’ chip companies such as Intel, National Semiconductor or AMD, and later in the PC industry under the leadership of Apple, Microsoft and Compaq. Contrary to the first generation of computer and electronics companies, such as IBM or Digital Equipment in the United States, Fujitsu in Japan or Siemens in Germany, many Silicon Valley companies did not produce entire computer systems. They concentrated on producing only a few key components, such as a microprocessor or a software operating system (Ferguson and Morris 1993). With the ‘PC revolution,’ which took off during the 1980s, Intel, Microsoft, Sun, and Cisco among others acquired global dominance. This break up of the traditional model of innovation and market control engendered a massive shift in industrial structure and was accompanied by a crisis in many older computer companies (most visibly IBM and Digital Equipment) around 1990. Most of these older companies designed and produced the
The rise and fall of ‘Wintelism’ 183 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
key components of their computer systems under their own roof, including computer chips, operating software and the hardware – following the Fordist model of the vertically integrated corporation (Chandler 1962). With the emergence of these new, specialised, technology companies such as Intel or Microsoft, the production system of the computer and later the telecommunications industry became more and more ‘modular.’ In the new world of vertically disintegrated production, items such as computers, servers and Internet-routers are assembled from standard components such as chips, operating software, disk-drives, modems and displays. They all can be bought on the open market and assembled and configured in various ways in products for different competitors (Lüthje 2001a). The emerging production systems of the ‘horizontal computer industry’ (Grove 1996) engendered a full-scale reversal of the existing logic of Fordist mass production, as well as of the lean production model pioneered by Japanese automobile manufacturers, notably Toyota (Womack et al. 1990). In reference to the brand names of Microsoft and Intel, Borrus and Zysman coined the term ‘Wintelism’ as an analytical concept to describe the forms of technological innovation and market control in vertically disintegrated industries and their inherent logic of ‘fragmentation and centralisation’ (Ernst and O’Connor 1992): In the auto industry, competition remains centrally a battle among the assemblers such as Toyota, GM, and Renault who design and integrate the final product. That competition has been dominated by production innovation and marketing (. . .). In electronics over the last decade, by contrast to the auto industry, competition has shifted away from final assemblers and the strategy of hierarchical (i.e. vertical) control of technologies and manufacturing. (. . .) In this new epoch, firms located anywhere in the dis-integrated value chain can, potentially, control the evolution of key standards and in that way define the terms of competition not just in their particular segment but, critically, in the final product market as well. (. . .) Such Wintelist strategies effectively attenuate the link between market power and the ownership of the assets of production that characterized the prior era of competition, and at the extremes, a company like Cisco Systems, can completely decouple control of final markets from ownership of manufacturing assets. . . . But production and scale do not vanish from the story; they are still significant . . . (Borrus and Zysman 1997: 4 ) It should be noted that this definition does not merely refer to a new model of manufacturing or of corporate organization – a frequent misunderstanding in recent debates on the topic. Rather, ‘Wintelism’ refers to the specific mode of innovation in technologies and markets pursued by the lead firms of specific industry segments (e.g. the PC or the data networking equipment
184
Boy Lüthje
industry), and is essentially related to industry-wide processes of restructuring (for a theoretical discussion see Lüthje 2001a). In this context, Wintelist innovation systems constitute the restructuring of production within crossnational production networks (CPNs), a term that is used ‘as a label (. . .) to the consequent disintegration of the industry’s value chain into constituent functions that can be contracted out to independent producers wherever those companies are located in the global economy’ (Borrus and Zysman 1997: 2). The concrete forms of such globalized production systems can vary widely. Indeed, there is no simple equation between patterns of innovation and institutional configurations of production networks. The changes in the division of labour in the ICT industry propelled by ‘Wintelism’ can be seen in three basic facts: •
•
•
Most ICT products have become complex commodities, assembled from traded parts and components supplied by various industry segments. The control of the time-cycle of new technologies and products has become the chief problem for manufacturing organization in the industry. As market control has shifted from assemblers to ‘product definition companies’, product innovation is increasingly separated from manufacturing. One important implication is that brand-name companies increasingly lose interest in keeping manufacturing close to their headquarters in industrialized countries. In contrast to older industries such as automobile manufacturing, the Wintelist ICT industry has no ‘focal corporations’ (Sauer and Döhl 1994) that coordinate the value chain through their own manufacturing operations. The ‘supplier pyramid’ governed by large-scale final assemblers is replaced by networks of interacting industry segments. Hierarchy is defined by the flagships’ ability to control technological development in key market segments (Borrus et al. 2000; Ernst 2002).
The growth of the Wintelist industry model sparked a proliferation of subcontracting networks in manufacturing. Traditionally, subcontractors in the electronics industry were relatively small firms that assembled printed circuit boards and standard electronic components such as resistors, coils or cables. During the 1990s, a new type of sub-assembly firm emerged, called Contract Manufacturing. These companies, which tend to be very large and global in scope, provide integrated manufacturing services for brand-name companies called Electronics Manufacturing Services. In contrast to traditional sub-assemblers, CM companies provide every element of systems manufacturing, including product engineering, highly automated assembly of printed circuit boards, final assembly, and configuration of computers and other ICT devices (called box-build), as well as components purchasing, distribution logistics and repair services (Sturgeon 1997, 1999).
The rise and fall of ‘Wintelism’ 185 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Contract manufacturers have become important players in the production chain, currently accounting for about 15–20 per cent of global value added in ICT manufacturing. As listed in Table 9.1, these companies exhibit multi-billion dollar revenues. However, their names are hardly known since they do not post their brand names ‘inside’ the manufactured products. Their toughest competitors are now found among another brand of subcontractors, called original design manufacturers. Most of them are based in Taiwan and assemble products for brand-name companies, but they still own the product design. For example, these companies manufacture most of the world’s notebook computers that are sold under brand names such as Dell and Compaq/Hewlett-Packard. Nevertheless, the small-scale subcontractor has not disappeared. Most companies of this kind today are concentrated in low-wage manufacturing areas, such as China’s Guangdong province, where they work as suppliers of cheap, standard components assembled at low wages for both CMs and brand-name companies (Shameen 2003). The relationships between contract manufacturers and older, brand-name firms in the United States and Europe (OEMs) developed rapidly during the second half of the 1990s. They began mainly through the acquisition of entire plants from established companies such as IBM, Texas Instruments or Lucent. In 1997, the Swedish telecommunications manufacturer Ericsson was the first European brand-name firm to sell off entire production units. It was followed by Europe’s largest electronics producer Siemens who sold an important server manufacturing facility in Germany in 1999 and several other PC and mobile-phone plants in 2000. By this expansion into industry segments with long-standing traditions of Fordist manufacturing operations in the United States and Europe, the flagship companies of the emerging EMS industry rapidly gained high-end manufacturing experience, often at the price of difficult processes of integration into their respective Table 9.1 Top ten electronics contract manufacturers, 2003 Company
Country
Revenue (US$ bn)
Type of business
Flextronics International Solectron Foxconn (Hon Hai) Sanmina-SCI Quanta Celestica Asustek Jabil Circuit Compal Mitac
US US TW US TW CDN TW US TW TW
13,822 11,144 10,899 10,795 8,576 6,735 5,747 5,170 4,760 4,564
EMS EMS EMS/OEM EMS ODM EMS ODM/OBM EMS ODM ODM
Source: Electronic Business Magazine, 1 August 2004. Note: TW: Taiwan; CDN: Canada.
186
Boy Lüthje
organizations. On the part of the older, vertically integrated ICT corporations, their selling off of key manufacturing assets was part of the broader scenario of restructuring in the 1990s driven by the finance market. Vertically integrated ICT firms aggressively sold off entire units and restructured along market lines as defined by the key players of the Wintelist industry segments (Lüthje et al. 2002; for a discussion of the role of financial markets for Wintelist industry models, see Jürgens and Sablowski 2004). The rapid expansion brought about a highly diverse spectrum of outsourcing relationships. For ‘fabless’ technology definition companies such as Cisco, 3Com or Microsoft (for its X-Box game console), CMs perform full-scale system manufacturing. Older brand-name companies maintain similar production relationships through their outsourced plants, often in competition with their own remaining facilities. Specialized mass producers in the computer industry (such as Dell) who still consider final assembly an important interface with the customer use CMs for the large-scale manufacturing of printed circuit boards or pre-assembled product kits. In addition, such companies outsource systems assembly in key foreign markets, mostly to medium-sized, local CMs. The only group of major brand-name firms hesitant to use CMs has been Japanese and Korean electronics firms. They prefer to rely on subcontractors within their own corporate ‘families’ (Lüthje et al. 2002). The globalization of manufacturing networks has emerged as a key feature of the EMS industry. As Borrus (2000) has analysed in detail, developing and effectively using such networks is a specific strength of the vertically specialized US ICT firms. The relatively open structures of these networks allowed for a wider choice of skilled, low-cost suppliers than the comparatively hierarchical and closed subcontracting arrangements in Japanese and Korean firms. US-based CMs emerged as a key element of these global production infrastructures. By their rapid acquisition of factories in industrialized countries and the simultaneous build-up of advanced manufacturing resources in Asia (primarily Malaysia and China), in Latin America (Mexico), and in Eastern Europe (Hungary, also the Czech Republic, Poland, Romania and Estonia), they acted as transnational network builders in this field and created low-cost, mass-production capabilities around the triad of industrialized world regions available to multinational brand-name companies on a ‘one-stop-shopping’ basis (Lüthje et al. 2002).
The recession and the global restructuring of production networks The massive recession in the ICT industry, which began late in 2000 and led to the subsequent bursting of the stock market bubble surrounding the New Economy (Brenner 2002), produced wide-sweeping changes in this scenario. The downturn also demonstrated the specific potential for crisis
The rise and fall of ‘Wintelism’ 187 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
and instability inherent in the Wintelist model of innovation and production. The potential for crisis can be described as follows: •
•
•
Rapid build-up of over-capacities in key industry segments that reflect extremely short cycles of innovation and amortisation. This has been particularly visible in the PC and server industry, where the recession coincided with the fading of the ‘PC-revolution’ as a driver for the expansion of Wintelist models of innovation (Lüthje 2001b). Overcapacities have been even more dramatic in the other key growth markets of the 1990s, such as data and telecommunications networking equipment (including optical fibre) and mobile communications. In the latter case, the capacity glut on the manufacturing side has been directly linked to the highly visible battles about future UMTS networks, culminating in the spectacular auctions of UMTS licences in European countries such as Germany, and the subsequent realization that carriers had overpaid to such an extent that investment costs had become almost impossible to recover. In the telecommunications industry, this crisis has also been linked to the enormous expansion of fibre-optics communications networks driven by the speculative pressures of financial markets. This crisis became visible in the massive accumulation of unused network capacity and in the crash of telecommunications providers Worldcom and Global Crossing in the United States. The high degree of vertical disintegration of production systems achieved during the 1990s, coupled with ‘ultra-lean’ inventory and manufacturing strategies, reinforced important domino effects throughout the supply chain and resulted in a massive shift of the impact of the crisis to the middle and lower tiers of the manufacturing system. Such effects were particularly felt in the vast networks of key component suppliers to Wintelist industry segments, such as disk-drive makers, and in contract manufacturing. In the disk-drive industry, the reduction of demand led to the near collapse of leading companies, such as Seagate, causing the loss of several tens of thousands of jobs, mostly in huge production sites in Malaysia and Thailand (SJMN 2003). A mounting structural problem behind the more cyclical aspects of the downturn is the rising cost of innovation in vertically disintegrated production systems, especially in chip design and manufacturing. This problem has been looming in the core sectors of the chip-making industry for quite some time. Particularly in the microprocessor and PC industries, the surging costs for investment in new fabrication facilities incited a proliferation of ‘fabless’ chip manufacturers (Angel 1994) and an increased level of outsourcing of downstream components and systems assembly (Lüthje 2001a). Productivity problems in chip design are now most visible in the key growth markets of consumer electronics, where the shift of growth to this particular area coincides with the emergence of ultra-complex methodologies of chip design, such as
188
Boy Lüthje systems-on-a-chip (SoC; Chang et al. 1999). As Ernst (2003) has explained in detail, the chip design industry has been suffering from a notorious decline in productivity in recent years, and this has initiated a process of vertical specialization and an accompanying shift of development work to low-cost locations in Asia.
Contract manufacturing is at the centre of the restructuring of key industry segments spurned by structural processes of overproduction in computer, data networking and telecommunications equipment. The unexpected stagnation of growth rates in 2001 indicated a rupture in the ultra-rapid growth pattern of the 1990s. However, the slowdown seems to be only temporary in nature, as the recession is also accelerating the outsourcing of manufacturing on the part of brand-name firms. Developments such as IBM’s sale of most of its PC manufacturing capacities to Sanmina-SCI (WSJ 2002), the acquisition of Hewlett-Packard’s PC factory in France by the same CM (FT 2002a) or the transfer of large-scale manufacturing assets from Lucent to CMs Celestica and Solectron seem to justify the expectation of further growth. Against this background, contract manufacturing has assumed a critical role in the regulation of industrial overproduction. Latent over-capacities have been built up in almost every segment of contract manufacturing, and since 2001, the major CMs have suffered heavy losses and declining revenues (Lüthje 2001b). The leading companies had to take massive restructuring charges for plant closures, over-priced acquisitions and inventory excess (FT 2002b). At the same time, they carried out large-scale, job-reduction programmes and plant closures. Lay-offs mostly affected volume production sites in low-cost areas in the United States, such as Texas, the Carolinas or Georgia, and in similar areas in Europe (Scotland and Ireland in particular). Cut-backs were even heavier in the newly established manufacturing complexes in low-wage countries such as Mexico, Malaysia or Hungary (FT 2002c). Loans of some very large CMs were downgraded to junk bond status. Particular sources of financial losses were the huge inventories of electronic components that CMs held on behalf of their customers. These components were either purchased directly by the manufacturers and built into the assembled products or owned by brand-name firms or parts distributors and assigned for use and inventory management to the manufacturing service companies. The complicated arrangements in this field produced what a wellknown consulting firm, iSuppli, aptly called a ‘supply chain disaster’. In many cases, the ownership of excess inventory was indeterminable. Both sides, CMs and OEM customers, tried to leverage their respective customers or manufacturing partners to take financial responsibility for significant amounts of excess parts and components (EN 2001). Few details became public, but market data indicate that the CMs were forced to bear the major brunt of the excess. According to iSuppli, at the peak of the crisis during the third quarter of 2001, CMs held about 49 per cent of the global inventory excess in semiconductors – an estimated $5.9 billion globally (EB Asia 2002; EN 2002).
The rise and fall of ‘Wintelism’ 189 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Vertical reintegration: the expansion of EMS production networks in Asia A key element of restructuring in the wake of the crisis was the accelerated relocation of transnational production networks to Asia (primarily China) by major CMs and a rapidly increasing number of new Taiwanese CMs. These Taiwanese firms were able to encroach into the turf of US CMs by way of cost advantages and superior capabilities in product design. They rapidly established very large plants in Chinese high-tech manufacturing regions, such as Guangdong, the Greater Shanghai area and the BeijingTianjin corridor. Within the networks of the global EMS industry, Asia has clearly emerged as the central region for advanced manufacturing (UNCTAD 2002). The production systems of major CMs in Asia are massive and complex, with large clusters located in Malaysia, Thailand and China, and regional headquarters and a few smaller, specialized plants in Singapore, Hong Kong, Taiwan and Japan. The shift to China is following the lead of major brandname ICT multinationals. Since China has emerged as the fastest growing and the world’s largest market for advanced computer, telecommunications and consumer electronics products and network infrastructure, it offers the economic potential for growth and continued innovation in the ICT that has been lacking in the developed world. At the same time, the country is becoming increasingly important for technological development, particularly in future standards for mobile communications. Furthermore, the very low level of wages and salaries based on a seemingly incessant supply of migrant labour from rural provinces remains a major pull factor for manufacturing in China. Flextronics, the largest CM in the world, is a perfect example of the complexity of CM networks in Asia (see Table 9.2). The company has its operational headquarters for Asia in Singapore (also nominally the world headquarters), and maintains a total of 25 locations throughout Asia. The Asian facilities employ 50,000 workers, more than a half of Flextronics’ global headcount of 95,000, and have 14 million square feet of manufacturing space, more than three-quarters of the company’s total 17.9 million (company information as of July 2003). The two main manufacturing bases are located around Singapore in southern Malaysia and in the Pearl River Delta (PRD) in South China. They are complemented by additional factories in other high-tech locations in China (Shanghai, Nanjing, Beijing) and in Malaysia (Kuala Lumpur, Penang). They also own a number of facilities that produce important supplies, such as plastic parts and metal enclosures, and a few design centres. Strategic manufacturing capabilities are concentrated in certain plants, which also function as know-how centres for other plants in the particular product category. Southern Malaysia, for instance, specializes in consumer printer products, PRD facilities in consumer electronics and mobile communications. Flextronics’ major US-based competitors,
190
Boy Lüthje
Table 9.2 Flextronics International: one contract manufacturing network in Asia, 2003 Country
Location
Type of facility
Japan
Tokyo Nagoya Beijing Qingdao Changzhou Nanjing Shanghai
Sales office Assembly Assembly Enclosures Enclosures Assembly Assembly plastics
Dongguan Gongming (Shenzhen) Xixiang (Shenzhen) Shenzhen Guanlan/Shajing Doumen (Zhuhai) Guangzhou Hong Kong
Plastics enclosures ass. Plastics Assembly plastics Enclosures Plastics Industrial park R&D Asia HQ
Bangalore Samutprakara Penang Shah Alam Melaka Senai Tampoi Woodland Changi
Assembly Enclosures Assembly Assembly Assembly Assembly plastics Assembly plastics Plastics HQ, R&D
China
Industrial park planned
Plastics HQ North Asia India Thailand Malaysia
Singapore
Source: Flextronics company information. Notes: Assembly: PCB- and systems assembly; enclosures: metal enclosures; plastics: injection molding for plastic enclosures and parts
Solectron, Sanmina-SCI, Celestica and Jabil, have production networks with a similar geographical division of labour, albeit smaller in size. European and Taiwanese CMs follow different integration strategies. Elcoteq, from Finland and the smallest of the global CMs, maintains a production network exclusively centred on China, with three plants in Beijing, Dongguan and Shenzhen. Taiwanese CMs, on the other hand, have their own world order. Their headquarters and major design and development operations are in Taiwan, their new product introduction operations are in US high-tech regions such as Silicon Valley, and their volume manufacturing is concentrated in China. As Flextronics demonstrates, CMs follow a distinctive strategy of vertical integration by combining the various stages of the production chain for ICT hardware systems within a global corporation. In Asia and in other low-cost locations, this structure is basically achieved in two ways: either
The rise and fall of ‘Wintelism’ 191 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
by constructing very large plants, often called industrial parks, that combine a broad spectrum of manufacturing resources in one place, or by integrating various mid-sized plants together in one region. The Flextronics industrial park in Doumen, South China, is a well-advertised example of the first strategy. It encompasses two major facilities for the assembly of printed circuit boards, a fabrication cluster for raw printed circuit boards featuring highly advanced manufacturing technology, a large plant for plastic enclosures, another one for metal enclosures, and fulfilment and distribution capabilities. The park boasted a workforce of more than 12,000 in 2003. Its major products include mobile phones, game consoles and ink-jet printers (Flextronics 2003). The operations of CMs from Taiwan are even larger. The factories of leading ODM producers Quanta, Compal or Arima in the greater Shanghai region equal the Flextronics park in size. Foxconn, the largest Taiwanese player (a subsidiary of the Hon Hai industrial conglomerate), practises the philosophy of vertical integration in one place in an even more pronounced way. It maintains a huge manufacturing facility with about 80,000 workers in Shenzhen. Encompassing about fifteen manufacturing buildings dedicated to top, international OEMs in the computer and communications field, as well as design and manufacturing facilities for components, it is probably the largest electronics plant in the world (interview data). Foxconn has seen revenues grow by up to 60 per cent in the last few years. According to figures from the Chinese government (RR 2003), the company has become the mainland’s second largest exporter. Flextronics in South China also provides a good illustration of the second strategy of vertical integration. Its smaller facilities in Guangdong form another production cluster of considerable size, including a regional manufacturing centre for mid-volume consumer products (2,300 employees), two further plastics production plants with a joint workforce of 1,500, a smaller, specialized enclosures facility in Shenzhen with 300 employees, and an engineering centre in Guangzhou (Flextronics 2003). In the course of 2003, this network was expanded by the acquisition of a number of plastics and enclosures facilities from NatSteel Broadway, formerly a leading CM from Singapore, of which two large plastics plants with 2,300 and 2,900 workers respectively are located in the PRD area. Elcoteq, with two complementary facilities for high-volume and low-volume products, and Sanmina-SCI, with a printed circuit board (PCB) assembly and an enclosures plant on one single campus, are following similar integration strategies. Celestica in Dongguan and Jabil in Guangzhou have stand-alone plants in the area that are integrated into the larger manufacturing networks of their respective companies in China and Asia. Many of these mid-sized factories were acquired from entrepreneurs based in Hong Kong or Taiwan in the early 1990s, or from major multinationals, such as two Elcoteq facilities bought from Nokia and from a joint venture between IBM and a local company. For many of these plants, a CM acquisition entailed significant
192
Boy Lüthje
upgrading of product portfolios and manufacturing technologies. As a result of this very rapid and very recent build-up (since about 1998), Guangdong has become host to the largest CM complex in the world. All in all, about fifteen facilities belonging to major CMs from the United States, Europe and Taiwan could be counted on the mainland side of the PRD area by mid-2003. Based on interview data and company information, the overall employment in these facilities can be estimated to stand at a minimum of 100,000. As can also be observed in other low-cost regions (Lüthje and Sproll 2003), the current process of vertical reintegration of manufacturing resources in newly industrializing countries engenders a gradual but accelerating shift of higher-end segments of the value chain to their lowcost respective locations. In China, most CM plants were started under overseas Chinese ownership in the early 1990s as assembly sites for lowend products, mostly computer peripherals such as mice or keyboards. Today, there are still large volumes of such low-end mass products in most CM plants. Nevertheless, a definite upgrading in the product portfolio has been taking place and is increasingly centred on mass products of higher complexity with very strict quality requirements in the assembly process, such as mobile phones. At the same time, there is a trend towards locating more products with lower volumes and larger variety in product specifications, such as telecom network equipment or higher-end medical products in Chinese plants. Furthermore, an increasing number of product introduction processes is being located in PRD factories, often on the request of major brand-name customers. This upgrading trend is supported by the accelerated adaptation of design capabilities into the service portfolios of major CMs, following the ODM model of their Taiwanese competitors. A very important element in this picture is the growing significance of the emerging Chinese electronics multinationals in the region as brand-name CM customers. Companies such as Huawei, ZTE or TCL all conduct substantial amounts of business with foreign CM firms, thereby generating important, local, upstream linkages in the manufacturing chain. Recent press reports indicate that TCL is planning to use Flextronics’ manufacturing facilities in Hungary in its bid to become a major player on the market for advanced television sets in Europe (NYT 2004).
Diverging paths of restructuring in key segments of the ICT sector At every level of the production system, original equipment manufacturing companies and contract manufacturing firms have reacted to the crisis with massive restructuring in manufacturing, supply and purchasing operations. The process has been put forth within the context of a ‘classical’ reconcentration and centralization of capital. Highly publicized mega-mergers and takeovers such as the one between Compaq and Hewlett-Packard only
The rise and fall of ‘Wintelism’ 193 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
marked the tip of the iceberg. Further downstream, major OEM companies reorganized their relationships with CMs across various corporate departments and product lines. Cisco, for instance, plagued by a huge inventory excess of chips, circuit boards and components of $2.5 billion in early 2001 (SJMN 2001), reduced the number of its CMs from 9 to 4, and the number of its CM facilities around the world from 30 to 16. This restructuring entailed a major shift of production between established US-based CMs to low-cost EMS companies from Taiwan (interview data). Lucent, the leading US manufacturer of telecommunications equipment and in possession of a 100-year manufacturing tradition within the former US telephone monopoly AT&T, shifted almost its entire manufacturing operation from 29 production plants to CMs Celestica, Solectron, Sanmina-SCI and Jabil. They now perform 95 per cent of Lucent’s manufacturing. Together with the spin-off of its chip-making and network services arms, Lucent’s revenue shrunk from $38.5 billion in 1999 to $8.5 billion in 2003 and its workforce from 157,000 to 32,500. This shift coincided with a massive reconcentration of plants and logistics and produced a similar push on CMs to downsize and concentrate manufacturing operations as Cisco did (Takahashi 2004). The merger between Hewlett-Packard (HP) and Compaq illustrates the magnitude of such restructuring. Their huge supply chain organizations were folded into one. As Table 9.3 shows in detail, this reorganization resulted in the reduction of the numbers of manufacturing facilities from 63 worldwide to 32, of distribution hubs from 142 to 88, and of first-tier suppliers from about 1,500 to 700. Also involved was a reorganization of the company-wide ICT system, with substantial reductions in the number of ICT applications, data servers and data centres. According to HP company information, this resulted in huge cost savings of $3 billion over the first two years after the merger. One key element in the restructuring of production systems has been a concentration of relationships with CMs. Whereas Compaq traditionally
Table 9.3 Global supply chain consolidation, HP/Compaq Premerger HP/Compaq 64 142 1,500 385 7,000 22,000 300 approx.
HP 2004 Manufacturing facilities 32 Distribution hubs 88 Suppliers 700 approx. Logistics partners 119 IT applications 4,000 approx. Data servers 19,000 Data centres 88 $3 bn cost savings over first 2 years
Source: HP company information.
194
Boy Lüthje
had used EMS only to a relatively small extent (see Lüthje et al. 2002), HP was one of the largest customers of the EMS industry and a key OEM that had nurtured the emerging industry during the 1990s (Lüthje 2001a). However, relationships between HP and EMS firms were established under the responsibility of individual HP business units, leading to a patchwork of diverging models and substantial diseconomies on the purchasing end. In 2001, still before the merger, HP began to centralize its CM networks by designating three preferred company-wide EMS partners controlled through key accounts at the top corporate level (2002 interview data). These policies are explicitly designed to counteract the growing bargaining power of large CMs. Part of the recentralization was a substantial change in purchasing operations. HP consolidated 80 per cent of its $40 billion parts purchases into 35 suppliers (ESM 2004). The purchasing of key electronics components, such as chips, motherboards or disk-drives, became strictly concentrated in the hands of HP, which restricted components purchase by CMs on behalf of HP to non-electronic or low-end components. For components purchased by HP, a new policy called ‘price masking’ was introduced. HP keeps the prices of components hidden from the CM in order to withhold strategic information on purchasing conditions and reduce its bargaining power (2004 interview data). In the wake of the recession, price masking has emerged as a widespread practice among major OEMs (such as Motorola, Siemens, Sony-Ericsson, etc.; EN 2004; interview data). According to recent figures from iSuppli, about 40 per cent of the electronics industry’s top OEMs use price masking schemes (EMSNow 2004). This practice has profound implications for CMs, since it severely limits their ability to capitalize on profits from their role as intermediary purchasers and vendors of electronic components. During the growth phase of the industry in the 1990s, component sourcing was key to the success of the EMS model, since profits from purchasing and resale of materials to OEM customers under general conditions of high demand were able to compensate for the CMs’ low profit margins in the assembly process (Lüthje et al. 2002). Today’s price-masking policies also shift competencies in materials management back to the OEMs, thereby reducing CMs’ roles in the organization of this important element of global production networks (Sullivan 2003). EMS companies, for their part, have reacted to the loss of control over the purchasing chain with a massive centralization of supply chain management. Solectron, for example, announced a sweeping reorganization of its global purchasing corporation, which was designed to overcome inefficiencies caused by the rapid growth in recent years. According to company sources, responsibility for purchasing decisions is shifted from individual plants or customer specific teams to a company-wide organization – a practice also seen in other major EMS firms. The goal of this policy is a radical cut in the number of key suppliers, as well as concentration at the global and regional level. About 250 suppliers, down from 550 in recent years, are
The rise and fall of ‘Wintelism’ 195 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
targeted to make up about 80 per cent of Solectron’s purchasing power – roughly $14 billion annually. The supply chain organization is developing company-wide schemes for supplier evaluation, bidding and purchasing procedures. This standardization is considered crucial for the broad implementation of web-based supply chain tools. It is intended to lift the proportion of web-based purchasing procedures from 35 to 75 per cent of the overall purchasing volume (Carbone 2002). Behind this transformation of the role of contract manufacturers and the revival of the traditional potential of vertical integration on the part of OEMs, we can see a diverse and often contradictory scenario when it comes to restructuring of sectoral systems of innovation. We also see remarkable differences between industry segments. Personal computers The PC industry, the role model for Wintelist industry structure, has seen an almost complete move to ‘fabless’ manufacturing on the part of all major US OEMs, accompanied by another shift to low-cost locations and more transition to ODM-based manufacturing models. IBM exited the PC industry almost completely, with the exception of business desktop PCs sold as a part of contracts with business customers. Manufacturing activities in this area were transferred to Sanmina-SCI (including three plants in North Carolina, Scotland and Mexico), and relocated to several plants in Malaysia, China and Eastern Europe (with heavy competitive bidding among the respective locations and massive job losses, particularly in the facilities in Scotland). Desktop (along with notebook) computer manufacturing for the Asian market was concentrated in a rapidly growing facility in Shenzhen, China, owned by a joint venture between IBM and its Chinese partner, Great Wall Computers. In the wake of this restructuring, IBM had its production of commercial desktops to an overwhelming degree performed by CMs or in its own joint venture, while still maintaining definition or product architecture and design as core competencies. However, this configuration was only transitional in nature, since at the end of 2004 IBM announced the sale of its entire PC business to Chinese PC maker Lenovo (Digitimes 2004b). The shift to ODM models was driven by HP and Dell. Dell, the winner of the restructuring of the PC industry in recent years, basically retained its production organization. It continues to rely on US CMs and on EMS and ODM companies from Taiwan as manufacturers of key components (especially motherboards and enclosures). It also retains its own plants for customized, final assembly and delivery in the United States, Asia and Europe. One major geographical relocation for its desktop PCs was the shift of final assembly for China and north-east Asia from Penang, Malaysia, to a new Dell facility in Xiamen, China. This shift was accompanied by an increased transfer of component prefabrication from US CMs to
196
Boy Lüthje
low-cost providers from Taiwan, notably Foxconn. At HP, PC manufacturing had been based on an EMS model. After the merger with Compaq, HP transitioned consumer and business PC manufacturing from EMS partnerships to an entirely ODM model with three ODMs as key partners (Tatung, Trigem and FIC). Hewlett-Packard sold its manufacturing plant in France (also to Sanmina-SCI), as well as a facility formerly owned by Compaq in Scotland to Jabil. As a result, almost all manufacturing and product design capabilities are outsourced. Following Dell’s lead, only industrial design (primarily casings, user interfaces and logos) remains a core OEM competency. Most electronics-related elements of product design are performed on a commodity basis by third parties and are no longer considered to be an innovative capacity. As US-based CMs had to bear the brunt of the recession, ODMs have emerged in a more favourable position. Mobile phones In mobile telecommunications handsets, an industry in which Wintelist models of innovation and ‘fabless’ manufacturing rapidly gained ground during the second half of the 1990s, restructuring paths have remained diverse and often contradictory. As mentioned earlier, mobile phone handset operations of vertically integrated manufacturers, notably Ericsson, Siemens, Motorola and, to a lesser extent, Nokia, had been transferred to a massive extent to CMs. Ericsson went even further when it announced the outsourcing of all of its handset manufacturing in 2001. Nokia, on the other hand, has maintained a relatively integrated manufacturing organization, including inhouse final assembly. It uses CMs mainly as component suppliers (Bengtsson and Berggren 2002). In the wake of the recession, the pendulum swung back to a certain extent: Motorola and Siemens have limited their outsourcing targets to roughly 30–40 per cent of their manufacturing volume, while Ericsson’s alliance with Sony has brought substantial in-house manufacturing assets back into their joint production system. The joint venture has split outsourced production between Flextronics, traditionally the preferred EMS, and Taiwanese ODM Arima. Together they now perform about 60 per cent of the production volume. In addition, Ericsson has invested heavily into a joint venture with Chinese state-controlled Putian group, which serves as the main production base for the Chinese market (Digitimes 2004a). The levelling-off of EMS-based outsourcing in handset production has been essentially due to the fact that all major mobile phone manufacturers have built up massive manufacturing and design facilities in China that integrate large, low-cost, in-house manufacturing operations with extensive modular supplier networks, including small and large component manufacturers that relocated from Europe (such as Philips audio systems as the world largest supplier of handset speakers and microphones). The strongest factor indicating a revival of vertical reintegration has been the emergence of Korean electronics manufacturers Samsung and LG as leading vendors
The rise and fall of ‘Wintelism’ 197 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
of mobile handsets. Both maintain vertically integrated manufacturing models with little outsourcing but huge, low-cost manufacturing and development complexes in China. Samsung’s strength rests particularly in the vertical integration of mobile communications with its chip manufacturing and development arm. It has also secured strong technological capabilities for the development of mobile communications chipsets sold to third parties. A key element of restructuring in mobile phone handset manufacturing has been the emergence of ODM companies as manufacturing partners and providers of intellectual property to US and European OEMs. The commodification of product know-how was mainly driven by the standardization of radio frequency (RF) technologies, which undermined the traditional link between handset and proprietary network technologies. Motorola has been a driving force in this development. By 2004 it had built an ODM supply infrastructure with four, top-tier ODMs from Taiwan (BenQ, Chi Mei, Compal and HTC). Siemens, Ericsson and, most recently, Nokia followed suit. Nokia established an ODM relationship with BenQ for the manufacturing of clamshell phones in reaction to analysts’ reports about weak showings in this field (Taipei Times 2004). In this sense it diverged from its traditional path of ‘component-oriented outsourcing’ (Bengtsson and Berggren 2002). Similar to the computer industry, major producers with deep-rooted traditions of in-house manufacturing and research such as Siemens have today narrowed down their definition of core competencies in mobile handset production to industrial design and the display software (the major technology components of brand-name building; interview data). From the perspective of manufacturing, the particular attraction of ODMs seems to lie in their reintegration of product development and manufacturing, which eases the ramping-up of new volume products. In addition, ODMs offer low manufacturing costs, since almost all handset manufacturing is done on mainland China (interview data from ODM and OEM companies; industry consultants). Computer game consoles As in telecommunications, the new markets of consumer electronics have also not been restructured predominantly along Wintelist lines. This seems to be true for most new growth segments, especially in the field of advanced audio and video systems. The case of game consoles is illustrative. Microsoft entered the market with its newly created X-Box in 2001 on a completely EMS-based manufacturing model. The X-Box was designed to break the dominance of the large Japanese players in computer games – particularly Sony and Nintendo. As a software company, Microsoft did not dispose of any substantial manufacturing assets of its own. Instead, it made CMs the natural partners for a quick build-up of hardware production capacities designed to establish control over the rapidly emerging computer game
198
Boy Lüthje
software markets. In the beginning, the manufacturing organization for the X-Box was developed with Flextronics, which built on the company’s capacity to offer integrated global manufacturing and logistics solutions with initially two manufacturing sites – one in Guadalajara, Mexico, and the other in Sarvar, Hungary. Microsoft’s production network for the XBox entailed a complex web of commodity component suppliers, including Intel and Samsung as producers of microprocessor and Dynamic Random Access Memory (DRAM) chips, ‘fabless’ chipmaker Nvidia as the designer for the key graphics processor, Taiwan Semiconductor (TMSC) as foundry manufacturer for Nvidia, Western Digital as designer and manufacturer for the high performance disk-drive, and a host of other specialized suppliers for commodity components. Flextronics was involved in the design of most of the key, non-electronics components, especially plastics parts, and coordinated the complex supply chain, which included the channelling of a huge number of parts and the challenging task of beefing up production to volumes of 100,000 units per week within a little more than a month (Carbone 2002). However, this genuinely ‘fabless’ model of manufacturing encountered massive competition from the Japanese incumbents in the global market. In 2002, about one year after launching the product, Microsoft had to substantially reduce the selling price for the X-Box, even though the initial price of $299 was already estimated to be about $150 below manufacturing costs. The increased cost pressure triggered a major restructuring of its global manufacturing network, resulting in the closure of the production lines in Hungary and Mexico (with the loss of several hundreds of jobs in each location), and in a shift of production to Flextronics’ industrial park in South China. Following some undisclosed problems in Flextronics’ supply chain, Wistron of Taiwan (also in possession of a manufacturing facility in South China) was brought in as a second EMS partner, and Solectron took charge of repair services in autumn of 2002. From a long-term perspective, however, Microsoft has not been able to challenge its Japanese competitors, especially the global market leader Sony. As of the beginning of 2004, Microsoft had sold 13.7 million X-Boxes (never with a profit), while Sony had sold more than 70 million of its Playstations (SJMN 2004). Sony also uses CMs to reduce manufacturing costs. The preferred manufacturing partners, however, are Foxconn and Asustek from Taiwan, mostly on an OEM basis. Foxconn also assembles large volumes of the Nintendo Game Cube (Custer Global SMT 2003). Sony keeps full control of the product design for the hardware and software, including the microprocessor chip. For future product generations, Sony has formed a large-scale alliance with two additional, vertically integrated players, IBM and Toshiba. The three companies lead a consortium for the development of a powerful new processor code named Cell, designed to provide a system–on-chip platform for future generation consumer products with modular architecture for various types of hardware consoles (EN 2004).
The rise and fall of ‘Wintelism’ 199 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
The recombination of production networks: EMS vs. ODM The ongoing restructuring of production models among brand-name firms and CMs indicates that there is no simple way into vertical disintegration, as has been suggested by theorists of market-based ‘modularity’. The changes on the part of the OEMs indicate a partial resurgence of vertically integrated models of product development. Nevertheless, they are different in nature from ‘Fordist’ models of mass production. The restructuring of production models on the part of CMs is an essential piece to this trend towards vertical reintegration in an environment of ‘global flagship networks’ (Ernst 2002). As in most other cases, the competition involves a redefinition of models of production, which leads to a reshaping of the particular elements of the manufacturing infrastructure as new industrial segments. The competition is essentially about defining manufacturing models, somewhat similar to the redefinition of product markets by new architectures of technology as it is practised under the Wintelist system of innovation. Among manufacturers, however, competition is essentially not about technology but about the ‘packaging’ and marketing (albeit without a brand name) of certain portfolios in manufacturing, design and logistics services. These elements involve a higher or lower degree of participation on the part of the manufacturer in product design, and a corresponding definition of ‘core competencies’ on the part of the brand-name customer. As explained above (see Table 9.2), all major types of outsourced manufacturing can be found in production networks formed by the top ten CMs today (see Table 9.2; for a definition of manufacturing models, see Table 9.4). OEM, as the oldest form of subcontracted manufacturing (and historically the domain of the traditional sub-assembly sweatshop), is used today in large-scale contracts between major global brands and their big, no-name suppliers in Asia. For instance, mobile phone bodies are produced for Nokia by Foxconn. The move from pure OEM subcontracting to EMS seemed to be the order of the day during the explosion of ‘fabless’ manufacturing in the heyday of the New Economy. EMS extended the role of assemblers from traditional subcontractors to ‘manufacturing partners’ – which gave rise to the expectation that the new American model of subcontracted ‘full-service’ manufacturing would outperform Asian subcontractors and put pressure on countries with strong OEM and ODM manufacturing models (Lüthje 2002; Sturgeon and Lester 2002: 30–3). Obviously, the overexpansion driven by speculation on EMS stock titles and the subsequent disaster of inventory pile-ups and plant closures marked the limits of the EMS model. It also added another chapter to the renewed crisis of US manufacturing industries in the post 9/11 era. In this period, the ODM model, as practised in the notebook industry and then transported into the PC, mobile handsets and consumer electronics industries, proved to be more sustainable, especially in terms of profit margins (see Table 9.5).
200
Boy Lüthje
Table 9.4 Typology of outsourced production in the electronics industry Acronym
Abbreviation
Main characteristics
Original Brand-name Manufacturing
OBM
Original Equipment Manufacturing
OEM
Electronics Manufacturing Services (also Contract Manufacturing)
EMS (CM)
Original Design Manufacturing
ODM
Contract Design Manufacturing
CDM
‘Classical’ in-house production. Core functions: product design, components purchasing, manufacturing, and logistics controlled by brand-name company. ‘Classical’ subcontracting: manufacturing of brand-name products by contract assembler, product and manufacturing process designed and controlled by brand-name firm, parts and components purchased by brand owner. Manufacturing of brand-name products by contract manufacturer, product designed by brand-name company, manufacturing process and logistics chain designed by contractor, including manufacturing-related product design, purchasing of major volumes of components, and after-sales services (e.g. installation and repair). Manufacturing and design of brand-name products by contract manufacturer; hard- and software IP of brand-name firm reduced to key elements of brand-name building (e.g. industrial design, logos, and software). Logistics and aftersales services by ODM. Hybrid between EMS and ODM. EMS, including shared IP between brand-name firm and contract manufacturer on certain components and manufacturing tools.
ODM companies accumulated manufacturing experience in the production networks of major US PC manufacturers, notably Dell and Compaq. Firms such as Mitac, FIC, Lite-On or Tatung emerged as key manufacturing partners for some of the leading US players in the PC industry during the Table 9.5 EMS/ODM margins: industry average, 1999 to Q1 2004 (%) EMS/ODM margins rolled-up averages
1999
2000
2001
2002
2003
Q1, 2004
EMS gross margins ODM gross margins EMS net income margins ODM net income margins
10.7 15.8 3.7
8.9 12.6 2.0
8.0 13.6 0.4
6.7 12.2 –11.0
6.0 10.0 –6.0
5.3 7.9 0.4
11.5
9.6
9.1
7.0
6.2
6.5
Source: Digitimes.
The rise and fall of ‘Wintelism’ 201 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
early 1990s. Most PC production of this kind emerged as OEM production for US brand names in the late 1980s and early 1990s, primarily motherboards and key components such as graphics cards, as well as final assembly (Dedrick and Kraemer 1998: 146–52). The move into ODM was propelled by the increasing role of Taiwanese manufacturing partners as suppliers of notebook computers in the late 1990s and supported by continuing efforts on the part of the Taiwanese government to facilitate the adoption of cuttingedge technologies. During this period, manufacturers from Taiwan added sophisticated capabilities in customization-to-order (CTO) and logistics, supported by the far-flung transportation networks of the country’s shipping and cargo firms (Chen 2002). The latest expansion of ODM into areas such as computer servers, mobile phones and consumer electronics is accompanied by a massive build-up of vertically integrated manufacturing facilities, most of them located on the Chinese mainland in Shanghai, Kunshan and Suzhou (2004 interview and site visit data). Today, the combination of world-class design and manufacturing capabilities puts Taiwanese CMs into the position of full-service ‘system suppliers’ with extensive capacities in sub-architectural technology development in key product markets. Their role and structure increasingly resemble those of top, independent system suppliers in older manufacturing industries such as automobiles (in some respect comparable to European companies such as Bosch, Siemens-VDO or Valeo). The chief attraction of such a manufacturing model to brand-name firms and one of their important strategic advantages over CMs is their integration of product development and manufacturing under one roof. This combination facilitates the transfer of new product designs into production and reduces ramp-up problems. It includes the possibility of buying full-scale product designs off the shelf – sometimes a welcome quick fix for product portfolios in highly cyclical markets (2004 OEM interview data). In addition, Taiwanese ODM and OEM firms seem to be superior in acquiring large outsourcing deals from Japanese first-tier electronics companies, long hoped for by US EMS firms (an illustrative case is Sony’s Play Station game console, which in early 2003 was shifted to the Chinese plants of two major Taiwanese assemblers, Foxconn and Asustek; SCMP 2003). Several vertically integrated electronics producers from Korea seem to pursue similar strategies of partial outsourcing, as in the case of LG’s choosing of Lite-On as an ODM for its production of mobile phones (EMSNow 2004). It should be noted that there has been hardly any significant cooperation between Korean OEMs and US-based EMS firms. As it seems, one of the greatest advantages of Taiwanese ODMs lies in their technological, organizational and cultural flexibility, which allows them to cooperate with a broad spectrum of very diverse brand-name firms ranging from ‘fabless’ US companies to vertically integrated corporations of European, US or Asian origin. The expansion of ODMs into the global production chains and the increasingly important role of China are driving rapid changes in the division
202
Boy Lüthje
of labour across the Taiwan Strait. Taiwanese top computer manufacturers had already shifted substantial manufacturing capacities to the mainland, Guangdong in particular, during the 1990s (around 50 per cent of manufacturing capacities in major product segments by 1995; Dedrick and Kraemer 1998: 152). Since these companies have become an integrated part of first-tier global production systems, most manufacturing capacities are located on the mainland, which has transformed top computer manufacturers into ‘hybrid companies’ with two-pronged corporate citizenship on the mainland and in Taiwan (Leng 2005). The ‘made in China by Taiwan’ model has involved the relocation of procurement and technology development capacities of major OEMs to Taiwan, resulting in a series of recent openings and expansions of technology centres by companies such as HP, IBM (especially for computer servers), Motorola and Siemens (EMSNow 2003, 2004). Apple recently expanded its staff in Taipei from 20 to 100 and announced that it was going to increase procurement from Taiwan to a volume of $5 billion per year (Digitimes 2004).2 However, as Taipei has become the world’s ODM marketplace (Chen 2004), the increasing extent to which Taiwanese ODMs are dependent on the mainland as a manufacturing base and their extensive political networking with local and national government authorities in China makes ODM manufacturing particularly vulnerable to political conflicts arising from the complicated relationships between China and Taiwan, including struggles over the relocation of jobs and labour practices in Taiwanese electronics firms (Leng 2005). Apart from such political risks, the ODM model has its own inherent problems. As Sturgeon and Lester have pointed out, ODM firms command a broader spectrum of capabilities than EMS companies, but they concentrate on fewer products (2002: 63–6). The recent and rapid expansion of their product spectrum would seem to overcome this limitation. However, with increasing technological capabilities (and thus bargaining power), the danger of know-how transfer from brand-name firms to potential competitors has become one of the key limitations of ODM partnerships. This situation is particularly dramatic in mobile communications. Here, many Taiwanese ODMs, such as BenQ or DBTel, appear on the market with their own brandname products, so far primarily in Taiwan and China or developing Asian markets such as India. In addition, the advent of a strong ODM infrastructure is producing a major shake-up in marketing relationships between brandname telecom vendors and telecom operators. For major network operators such as Vodafone or Deutsche Telekom, ODMs have become attractive partners for the manufacturing of mobile phone handsets to be sold under the operators’ own brand name. This practice is widespread in Asia, but it is also increasingly used in Europe and in the United States. Not only does it put pressure on established manufacturers as far as volumes and margins are concerned, but it also undermines the established order in technology definition, since operators can form alliances with powerful technology partners independent of their existing brand-name vendors, such as Deutsche
The rise and fall of ‘Wintelism’ 203 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Telekom’s alliance with Microsoft on 3G display software architecture (Economist 2004). In their attempts to counteract the challenge from their Taiwanese competitors, EMS firms are trying to exploit the weaknesses of the ODM model. Apart from the restructuring of manufacturing operations and supply chain organization, the leading EMS companies have put forward a variety of responses. One set of strategies is to concentrate on their own core competencies and emphasize the value of the traditional service model. Solectron’s restructuring is following this pattern. Ongoing financial troubles forced the former world market leader to sell substantial assets in the design field (notably subsidiaries Force Computers and Zhone). The company is now advertising an approach exclusively focused on production, which is in turn accompanied by company-wide efforts to implement lean manufacturing practices from top Japanese auto manufacturers in order to boost manufacturing quality. In the design field, Solectron is developing contract design manufacturing, which expands contract design activities at the component level but still leaves the intellectual property for the product architecture with the brand-name firm. This approach is being advertised as particularly appealing to OEM customers interested in ‘protecting their brand by not developing competitive IP’ (Purvis 2004). Celestica, who in the component field sold an important subsidiary for power supplies, as well as Jabil, seems to be following similar strategies. They do not, however, seem to be as driven by financial restructuring needs as Solectron does (Jorgenson 2004). In contrast to these more defensive strategies, industry leader Flextronics and, to a lesser extent, Sanmina-SCI, are trying to expand vertical integration by building on existing strengths in non-electronic component manufacturing (plastic and metal enclosures in particular), while simultaneously adding new product design capabilities. In the autumn of 2004, Flextronics announced a variety of important steps in this direction, including offering expanded ODM services and enhanced capabilities in chip design. They formed a partnership with ARM, the top IP vendor in system-on-chip designs, and they acquired software design capabilities from Hughes and a digital camera module business from Agilent (2004 interview data). However, in spite of such recent competitive gains by the ODM model in the broader field of contracted manufacturing services in the electronics industry, Taiwanese ODM companies are not exempt from the overall structural problems in the CM industry. These problems stem from the industry’s primary role as a low-cost manufacturing infrastructure that encourages rapid profit maximization among global brand-name firms. As Table 9.5 shows, ODM firms enjoyed higher profit margins than CMs during the recession years. However, as a result of the massive restructuring and price-cutting on the part of brand-name firms, ODM margins are also beginning to decline rapidly. Overall, the shift of CM infrastructures to Asia and the restructuring of manufacturing models have not been able
204
Boy Lüthje
to overcome the long-term decline of profit margins in the ICT industry. The decline is due to the structural costs of vertically disintegrated innovation and declining productivity rates in the generation of new technologies, particularly in the field of chip design. Whether the ongoing restructuring of the production infrastructure will result in a scenario of EMS and ODM convergence into a hybrid model (Pick 2004) or in a protracted weeding out of EMS pioneers from the 1990s is a matter of speculation. The net effect of this restructuring remains the ongoing relocation of massive, hightech manufacturing potential to Asia and its vertical integration, particularly within the ‘China circle’ (Naughton 1997).
Conclusion: Wintelism and beyond In summary, the recent crisis in the ICT industry has not ended the complex processes of sector-wide ‘fragmentation and centralization’ (Ernst and O’Connor 1992) as a way of restructuring the post-Fordist ICT industry (Lüthje 2001a). Rather, the recession demonstrated the limits of Wintelism as a model of network-based mass production, thereby opening the way to more diverse pathways of restructuring. Our analysis illustrates the thesis that there is no such thing as path dependence, defined as a ‘relatively simple model of the world in which complementarities of rules (. . .) result in rigidity and path dependence’ (Storz and Moerke, Chapter 1). This, of course, implies that the structural problems of Wintelism are not simply encouraging a ‘return’ to older forms of vertical integration, nor are they stimulating a revival of small-scale networks of flexible specialization. Instead, we see an accelerated globalization of production networks, which is leading to greater diversity and complexity, but is also reinforcing the hierarchical character of ‘global flagship networks’ (Ernst 2002). ‘Modularity’ is not developing along a classical, liberal market model of free exchange in contract-based manufacturing and design services. It is also apparently not evolving along trust-based networks of intra-firm cooperation, as theories of flexible specialization or informational capitalism have suggested. The study of global production networks in the electronics industry cannot fall back on yesterday’s debates but must examine the concrete dynamics of globalization and the ways in which older and newer forms of mass production are being recombined. One lesson from the more recent developments is that the interface between product development and manufacturing – traditionally the Achilles heel of all forms of outsourced manufacturing – remains the key to restructuring manufacturing and innovation in the ICT industry. The processes of vertical reintegration outlined in this chapter – i.e. the partial revival of vertically integrated production models on the part of OEMs, the massive reintegration of manufacturing resources by US and Taiwanese CMs in China, and the increased role of ODMs in integrating development and manufacturing along the more classical lines of ‘system suppliers’ – all point in this direction.
The rise and fall of ‘Wintelism’ 205 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
However, the reintegration of design and manufacturing is increasingly occurring under the auspices of third parties, particularly first-tier ODMs with product development capabilities, and often with a totally different set of business and locational strategies (see Ernst and Lüthje 2003). As a result, we find an increasingly diversified landscape of innovation, which so far has limited any further ‘Wintelization’ of innovation and market control in key, ICT industrial growth markets beyond PCs, servers, routers and some consumer products such as Apple’s iPod music player. In the context of the increasing shift of innovation to Asia, this constellation has enhanced the role of first-tier Asian ICT companies in the global arena and generated long-term pressures on both the American and the European ICT industry widely subscribing to the Wintelist logic of innovation driven by short-term, financial market interests. A recent article in Business Week on Samsung summarizes this development: Samsung is challenging basic New Economy dogma. In high tech, the assumption is that developing proprietary software and content gives you higher margins and a long lead time over rivals. Yet Samsung defiantly refuses to enter the software business. It’s wedded to hardware and betting it can thrive in a period of relentless deflation for the industry. Rather than outsource manufacturing, the company sinks billions into huge new factories. Instead of bearing down on a few ‘core competencies,’ Samsung remains diversified and vertically integrated – Samsung chips and displays go into its own digital products. ‘If we get out of manufacturing,’ says CEO and Vice Chairman Yun Jong Yong, ‘we will lose’. (Business Week 2003) In the arena of outsourced manufacturing, the success of Taiwanese ODM companies makes explicit the long-term, hidden problems of know-how transfer and brand-name control as the key questions for the future growth of contract-based manufacturing. This conflict may leave ample space for ‘pure’ EMS-based manufacturing models, yet it reiterates the fact that there remains a positive trade-off between large-scale, vertical integration of state-of-the-art know-how in product development and manufacturing and the ability to build brand-name control in mass-production industries. The new generation of rapidly growing OEMs from China, such as TCL, Huawei or Konka, seems to be learning this lesson. They can benefit from a huge domestic market backed by a developmental national state that exercises strong control over markets and industrial policies. At the same, the availability of first-class resources in third-party manufacturing, design and product development pushed by the shift of CMs to China facilitates learning processes in the respective fields. This story has implications for debates on industrial upgrading. The key to linking up with global flagship networks does not seem to lie merely in
206
Boy Lüthje
the upgrading of industrial districts by developing a skilled base of local suppliers and perhaps specialized technological product design to serve certain segments of global production networks (the now classical pathway of economies such as Singapore or Taiwan). In today’s ICT industry, learning effects are increasingly occurring within globalized production networks, and technological knowledge – including the very top end of product design and development – is increasingly losing its ‘local stickiness’ (Ernst 2003). The current dynamics in East Asia as a high-tech region is based on a largescale recombination of advanced production infrastructures that work with new potential in technological development created by vertically integrated players from Asia and the simultaneous global outsourcing of hardware, component, software and chip design. This opens up new possibilities for recombining innovative forces in East Asia, often across borders. Wintelism has acted as a key driver in the globalization of production networks and in the transfer of knowledge. However, it has not resulted in an ‘American model of manufacturing’ becoming a globally dominant way of doing things and a new pillar of US hegemony.
Notes 1
2
This paper emerged from the Institute of Social Research project ‘New Models of Production and International Division of Labor in the Electronics Industry,’ sponsored by the German Research Foundation (DFG). My thanks for continued cooperation to the members of this group: Wilhelm Schumm, Martina Sproll, Stefanie Hürtgen, and Peter Pawlicki. I am also grateful for ideas and suggestions from many discussions with our partner, Dieter Ernst, at the East-West Center, Honolulu, Hawaii. This paper benefited greatly from Peter Pawlicki’s excellent work as research assistant. Apple declined to be interviewed for this research project.
Bibliography Angel, D. P. (1994) Restructuring for Innovation: the remaking of the US semiconductor industry, New York: Guilford Press. Bengtsson, L. and Berggren, C. (2002) Horizontally Integrated or Vertically Divided? A comparison of outsourcing strategies at Ericsson and Nokia, working paper, University of Linköping/University of Gävle. Borrus, M. (1988) Competing for Control: America’s stake in microelectronics, Cambridge MA: Ballinger. Borrus, M. (2000) ‘The resurgence of US electronics: Asian production networks and the rise of Wintelism’, in Borrus, M., Ernst D. and Haggard S. (eds) International Production Networks in Asia: rivalry or riches?, London: Routledge, pp. 57–79. Borrus, M. and Zysman, J. (1997) Wintelism and the Changing Terms of Global Competition: prototype of the future?, BRIE Working Paper 96B, Berkeley CA: Berkeley Roundtable on the International Economy. Borrus, M., Ernst D. and Haggard S. (eds) (2000) International Production Networks in Asia: rivalry or riches?, London: Routledge.
The rise and fall of ‘Wintelism’ 207 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Brenner, R. (2002) The Boom and the Bubble: the US in the world economy, London/ New York: Verso. Carbone, J. (2002) ‘Solectron prepares for the upturn’, Purchasing, 24 October. Castells, M. (1996) The Rise of the Network Society, Oxford/Cambridge: Blackwell. Chandler, A. D. (1962) Strategy and Structure: chapters in the history of the American industrial enterprise, Cambridge MA/London: Harvard University Press. Chang, H., Cooke L., Todd, L., McNelly, A. et al. (1999) Surviving the SOC Revolution: a guide to platform-based design, Boston MA: Kluwer Academic Publishers. Chen, S. (2002) ‘Global production networks and information technology: the case of Taiwan’, Industry and Innovation, 9 (3): 249–65. Chen, S. (2004) ‘Taiwanese IT firms’ offshore R&D in China’, unpublished manuscript, Taipei: Chung-Hua Institution for Economic Research. Dedrick, J. and Kraemer, K. L. (1998) Asia’s Computer Challenge, New York/Oxford: Oxford University Press. Ernst, D. (2002) ‘The economics of electronics industry: competitive dynamics and industrial organization’, in Lazonick, W. (ed.) The International Encyclopedia of Business and Management (IEBM), Handbook of Economics, London: International Thomson Business Press. Ernst, D. (2003) ‘The new mobility of knowledge: digital information systems and global flagship networks’ in Latham, R. and Sassen, S. (eds) Digital Connections in a Connected World, published for the Social Science Research Council, Princeton NJ: Princeton University Press. Ernst, D. (2004) ‘Limits to modularity – reflections on recent developments in chip design’, Industry and Innovation, September, 12(3): 205–25. Ernst, D. and Lüthje, B. (2003) Global Production Networks, Innovation, and Work: why chip and system design in the IT-industry move to Asia, issues and topics for future research, East–West Center Economics Research Papers, No. 64. Honolulu, Hawaii. Ernst, D. and O’Connor, D. (1992) Competing in the Electronics Industry: the experience of newly industrializing economies, Paris: OECD. Ferguson, C. H. and Morris C. R. (1993) Computer Wars: how the West can win in a post-IBM world, New York: Times Books. Flextronics (2003) Corporate Fact Sheet Asia, CD-ROM files, Singapore. Gilder, G. (1988) ‘The revitalization of everything: the law of microcosm’, Harvard Business Review, 66, March–April. Grove, A. S. (1996) Only the Paranoid Survive: how to exploit the crisis point that challenge every company and career, New York/London: Currency Doubleday. Jorgenson, B. (2004) ‘EMS prepares for its next phase’, Electronic Business, December. Jürgens, U. and Sablowski, T. (2004) ‘A new model of industrial governance? Wintelism in the InfoCom industry’, in Faust, M., Voskamp, U. and Wittke, V. (eds) European Industrial Restructuring in a Global Economy: fragmentation and relocation of value chains, Göttingen: SOFI, pp. 221–40. Leng, T. (2005) ‘State and business in the era of globalization: the case of crossstrait linkages in the computer industry’, The China Journal, 53, January: 63–79. Lüthje, B. (2001a) Standort Silicon Valley: Ökonomie und Politik der vernetzten Massenproduktion, Frankfurt/New York: Campus. Lüthje, B. (2001b) Arbeitnehmerinteressen in einerm transnationalen IT-Unternehmen. Eine Fallstudie am Beispiel des Compaq-Konzerns, Edition der Hans-BöcklerStiftung 48, Düsseldorf.
208
Boy Lüthje
Lüthje, B. (2002) ‘Electronics contract manufacturing: global production and the international division of labor in the age of the internet’, Industry and Innovation. Special Issue: Global Production Networks (eds Ernst and Linsu Kim), 9 (3): 227–470. Lüthje, B. and Sproll, M. (2003) ‘Electronics contract manufacturing: networks of transnational mass production in Eastern Europe’, in Faust, M., Voskamp, U. and Wittke, V. (eds) European Industrial Restructuring in a Global Economy: fragmentation and relocation of value chains, Göttingen: SOFI, pp. 249–66. Lüthje, B., Schumm, W. and Sproll, M. (2002) Contract Manufacturing: transnationale produktion und industriearbeit im IT-Sektor, Frankfurt/New York: CampusVerlag. Naughton, B. (1997) The China Circle: economics and technology in the PRC, Taiwan, and Hongkong, Washington DC: Brookings Institution Press. Pick, A. (2004) ‘Survival of the fittest?’, EMSNow, 19 October. Purvis, D. (2004) ‘New approach to design partnerships: Solectron’s design approach preserves customer’s valuable intellectual property’, Solectron World, IV (1):1–3. Sauer, D. and Döhl, V. (1994) ‘Arbeit an der Kette – Systemische Rationalisierung unternehmensübergreifender Produktion’, Soziale Welt, 45. Jahrgang (H. 2), pp. 197–215. Saxenian, A. (1994) Regional Advantage: culture and competition in Silicon Valley and Route 128, Cambridge MA/London: Harvard University Press. Shameen, A. (2003) ‘Outsourcing: you want it, we make it’, Far Eastern Economic Review, 20 March. Sturgeon, T. J. (1997) Turnkey Production Networks: a new American model of industrial organization? BRIE Working Paper 92A, Berkeley CA: Berkeley Roundtable on the International Economy. Sturgeon, T. J. (1999) Turn-Key Production Networks: industry organization, economic development, and the globalization of electronics contract manufacturing, Ph.D. dissertation, Berkeley CA: University of California. Sturgeon, T. J., Lester R. K. (2002) Upgrading East Asian Industries: new challenges for local suppliers, Working Paper, Cambridge MA: MIT Industrial Performance Center. Sullivan, L. (2003) ‘Motorola wirft Auftragsfertiger aus Preisverhandlungen’, EETimes.de, 11. November, www.eetimes.de. Takahashi, D. (2004) ‘Back from the brink’, My-ESM, October 10, www.techweb.com. UNCTAD (United Nations Conference on Trade and Development) (2002) World Investment Report 2002: transnational corporations and export competitiveness, New York/Geneva: UNCTAD. Womack, J. P., Jones, D. T., Roos, D. (1990) The Machine That Changed the World, New York: Rawson Associates.
Newspapers and periodicals Digitimes EB Asia Economist EMSNow EN ESM
DigiTimes, Taipei Electronic Business Asia The Economist, London EMSNow Electronics News, San Jose CA Electronics Supply & Manufacturing
The rise and fall of ‘Wintelism’ 209 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
FT NYT RR SJMN SCMP WSJ
Financial Times, London New York Times, New York Renmin Ribao (People’s Daily), Beijing San Jose Mercury News, San Jose CA. South China Morning Post, Hong Kong Wall Street Journal, New York
10 Open innovation Novel deployment of ICT in new product development Ralf Reichwald, Frank Piller, Sascha Seifert and Christoph Ihl
According to some authors, the development of new information and communication technologies (ICT) is slowly undermining the traditional economic reasons behind hierarchical forms of economic transactions. Even the importance and necessity of customer relationships as a mode of economic exchange between firms and customers has been put into question due to the rise of new ICT, particularly new media and the Internet. According to this view, which has become known as the ‘move-to-themarket’ hypothesis, electronic commerce is assumed to encourage market transparency and price competition and thus favour free market rather than hierarchical, corporate structures (Malone et al. 1987). At the same time, however, new ICT may also improve a firm’s absorptive capacity for customer knowledge by allowing virtual means of customer integration and interaction – a strategy we call open innovation. In this chapter, we focus on the micro level of the development and use of ICT. An individual actor can be regarded not only as an adaptor but also as a creator in the sense of shaping institutional settings. The first section of this paper provides a literature review of customer participation in new product development (NPD), which helps to conceptualize our understanding of open innovation. In the next section, the concept of open innovation is described as both an adaptive as well as a creative response of firms to new developments in ICT. The next section turns to the micro level of the development and the use of ICT at Adidas-Salomon. Here, the practical implementation of the concept of open innovation is explored in an action research project. We identify a set of strategic capabilities a firm has to develop in order to profit from open innovation. The evaluation of the case project uncovers a number of new management tasks and organizational structures that appear to be relevant for companies operating in heterogeneous, volatile markets. While open innovation undoubtedly requires an appropriate, institutional framework, we argue in the concluding discussion that on the micro level, corporate culture is the most important factor for success when a firm attempts to capitalize on the benefits of open innovation.
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
211
The concept of open innovation No one in this day and age needs to be convinced about the importance of innovation. Intense competition, ever shortening product life cycles, the growth of heterogeneous user needs and rapidly changing markets and technologies continually ensures its necessity (Booms et al. 1983). Yet innovation often entails enormous investments, and a failure in terms of accumulated flops could easily jeopardize the viability of the entire company. Indeed, the failure rate of innovations in recent years has been high, especially in consumer goods and services. Over the past few years, innovations in these markets have shown flop rates of often up to 50 per cent or more (Davidson 1976; Crawford 1977; Brockhoff 2003). Thus, the key question for firms today is not whether or not to innovate, but how to innovate successfully (Drucker 1998). In recent economic literature, a number of surveys discuss potential success factors of innovation (Easingwood and Storey 1991; Edgett 1994; de Brentani and Ragot 1996; de Brentani 2001). As of late there is a growing consensus about the importance of a customer–market fit. Successful innovations are those that are exactly tailored to the needs and demands of a well-defined, target market (Cooper 1993; Kogut and Meitu 2000; Katila and Ahuja 2002; Vojak and Suarez 2002; Brockhoff 2003; Chesbrough 2003; Lüthje 2003). From the point of view of a specific firm, this perspective requires two types of information: need-related and solution-based information. Needrelated information refers to customer needs based on past purchasing behaviour, usage and customer perceptions of a firm’s products, competitive behaviour and individual or group experiences with a firm. In contrast, solution-based information refers to the way a firm transforms customers’ needs into specific products or service designs in general, or to the process of applying a specific technology to transform needs into new products and services (Freeman and Soete 1997; Utterback 1994). For decades, a seemingly convincing innovation pattern was proposed: customers possess need information while firms possess solution information (Kaulio 1998; Brockhoff 2003). In a first stage of this traditional, innovation process (idea generation), companies expand their knowledge base and access information in order to increase the number of new product and service ideas. A variety of ways of doing so exist, ranging from mining research labs to soliciting creative input from manufacturers, marketers, customers and suppliers. After generating a bundle of ideas, the second challenge is to screen them in order to focus resources on the most attractive opportunities. This process is based on a set of screening criteria that accommodate the company’s technological capacities while at the same time making effective use of its development resources in meeting strategic and financial needs. In a second stage (concept), potential ideas are transformed and conceptualized within a product concept review by bundling all current technical, organizational
212
Ralf Reichwald et al.
and economic pieces of information. Based on this review, feasibility studies geared towards assessing possible risks and benefits are carried out. Early screens tend to be primarily technical in nature, whereas later screens shift to emphasize manufacturing feasibility and fundamental economics. In a third stage (prototype), a potential innovation is prototyped by building an actual and functional mock-up model. This mock-up should meet all the specifications designed in the concept phase and work exactly as it should under real-world conditions while remaining within the given budget. In a final step (market), successful prototypes are transferred into the mass production process. Test and pre-test markets are used to reduce uncertainty about entrance, sales forecast and additional marketing objectives. Within this traditional, innovation process, customers’ need-related information is iteratively transferred to the firm by means of standard market research techniques (Brockhoff 2003). The better the firm is at capturing need-related information, the better it can incorporate the information into problemsolving activities. In doing so, firms are seen as the locus of solution-based information. Von Hippel (1978) coined the term ‘manufacturing active paradigm’ to describe the conventional role of customers in the innovation process. Within this paradigm, customer input is meticulously gathered from customers in a chosen target market. The customer herself is regarded as a statistical (representative) unit in terms of reflecting the mean demand of the target market. Having identified an initial product idea, firms integrate iterative, need-related customer information at multiple points of the development process. Nevertheless, within this traditional process, the role of the customer remains a passive one. He or she speaks only after having been spoken to. Customer information accessed by the company remains restricted to needrelated information (Dahan and Hauser 2002). Chesbrough (2003) was among the first to coin the term ‘open innovation’ as a label for a novel approach to NPD. It is characterized by cooperation in innovation processes within large, horizontal and vertical networks of actors. Companies utilize external ideas as well as those from their own research and development departments in order to advance their technology, since internal departments often show a tendency of repeating old successes rather than producing radical innovation (Bower and Christensen 1995). In addition, Chesbrough argues that open innovators can commercialize external ideas by deploying outside as well as in-house pathways to the market. This process redefines the boundary between the firm and its surrounding environment. It makes the firm more porous and embedded in loosely coupled networks of different actors, each collectively and individually working towards commercializing new knowledge (Laursen and Salter 2004). Chesbrough’s understanding is shared by other authors who also demand the use of multiple sources of input in generating ideas for innovation (Kogut and Meitu 2000; Quinn 2000). Open innovation in this understanding draws
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
213
on technologies from networks of universities, start-ups, suppliers and even competitors. Yet while these authors address the demand for cooperation between professionals in the innovation process, the customer as a source of innovation plays no dominant role. The importance of customer input is acknowledged, but innovation is primarily regarded as resulting from a corporate network of professional players. Open innovation in this context thus also follows the traditional pattern of innovation outlined above: customers possess need information, while manufacturers develop solution information. Empirical research in a number of fields has shown, however, that customers rather than manufacturers are frequently the first to develop and use prototype versions of what later become commercially significant new products and processes (Cooper 1993; Kogut and Meitu 2000; Katila and Ahuja 2002; Vojak and Suarez 2002; Brockhoff 2003; Chesbrough 2003; Lüthje 2003). Anywhere from 10 to nearly 40 per cent of users in the fields sampled to date have engaged in developing or modifying a product for in-house industrial use or for personal use. These findings provide evidence that the role of the customer is changing from a passive to a proactive one – a shift in customer emancipation that von Hippel (1978) labels the ‘customer active paradigm’. In this paradigm, customers are the locus of most of the activities carried out in the innovation process. They discover and conceptualize new needs and demands, build and test prototypes, select a manufacturer capable of bundling the product or service and then request the product or service from this manufacturer. In other words, the customer freely reveals a ready-made prototype that can be exploited by a manufacturer in order to satisfy previously unmet consumer needs (von Hippel 1988). Such innovative potential is not equally distributed among a group of customers but rather tends to be concentrated among the ‘lead users’ in a user population. Lead users are members of a user population with two distinguishing characteristics. First of all, they are at the cutting edge of important trends, and as such are currently experiencing needs that will later be experienced by many users in that marketplace. Second, they anticipate relatively high benefits from obtaining a solution to their needs, and as such are willing to share their solution-based information with a company (von Hippel 1986). From the perspective of a firm, the first characteristic could be seen as a type of filter that identifies user innovations that manufacturing firms would likely find commercially attractive. The second characteristic, on the other hand, can be used to identify users with a higher likelihood of innovation potential, as there is a positive association between profit expectations and innovative activity. But what are the basic characteristics and attributes that distinguish potential lead users from the rest of the user population? One of the foremost discussed characteristics is consumption expertise, which is closely related to product knowledge and usage experience (Jacoby et al. 1986; Alba and Hutchinson 1987; Lüthje 2003). Lüthje (2003) draws the association between product expertise and
214
Ralf Reichwald et al.
customer co-creation first of all based on the customer’s experience with a product due to frequent use, and second on their knowledge about a product’s architecture, materials and technologies. This knowledge influences a customer’s ability to anticipate (future) problem needs before the rest of the target market. Furthermore, there is evidence that lead users are also active as opinion leaders (King and Summers 1970; Childers 1986). Opinion leadership reflects the extent to which individuals provide information about a topic to others and the extent to which information is sought by others from those individuals. In addition, empirical evidence shows a high level of innovativeness among lead users (Craig and Ginter 1975; Manning et al. 1995). On an individual basis, every consumer is, to some extent, an innovator. All of us, over the course of our lives, adopt some objectives or ideas that are new to our perception. Lead users, however, are people who adopt new products and innovations before others in a timely manner. They lead the market in trying out new innovations and thus are quicker to figure out new needs and solutions. Further, lead users can also be determined by personality traits. These include intelligence, creativity, novelty or varietyseeking behaviour, and/or the ability to handle cognitive complexity (see Hirschman 1980; Reichwald et al. 2004). Open innovation does not have to follow the radical extremes of either the manufacturing or the customer active paradigm. Depending on the type of innovation, customer involvement can also be limited to specific, welldefined stages or tasks of the innovation process. We thus use the term ‘open innovation’ to refer to the way that customers are integrated into the process of creating new capabilities and resources in cooperation with the firm. The creative process leads to innovation value in terms of a new solution space, characterized either by new products or new processes. From an economic point of view, value in this system is not simply accumulated but is mutually created and re-created among a company and its potential customers. Building on this understanding, open innovation is characterized by a shift of labour that moves value-creating activities from the manufacturer to the customer. The customer, in turn, moves in the direction of becoming a prosumer and co-designer (see Enos 1962; Freeman 1968; von Hippel 1978; Rice and Rogers 1980; Herstatt and von Hippel 1992; Anderson and Crocca 1993; Cooper 1993). Both the customer and the manufacturer supply and share important resources. Moreover, the customer is integrated in a process of knowledge creation, thereby expanding the manufacturer’s existing resources and capabilities. By generating a process in which manufacturers and customers work together to develop, create and produce a final product that satisfies needs, open innovation builds cooperation between both parties. Cooper (1993) suggests that customer input and feedback should be sought at every step of the way throughout the entire development phase as the product or service takes shape. In an open innovation system, companies solicit solution-based information from the customers instead of merely need-related information.
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
215
One problem with this open innovation system is that solution information is stickier and more difficult to transfer to firms compared to need information. According to von Hippel (1998), the stickiness of information is defined as the incremental expenditure required to transfer a specific unit of solution information from the customer to the manufacturer. When the expenditure is low, information stickiness is low; when it is high, stickiness is high. This is where new ICT gain in importance. ICT are seen as the main protagonist that can lower the costs of transferring sticky information, both from the firm’s and the customer’s perspective.1
The relevance of open innovation in the light of new ICT Product differentiation and customer interaction According to the move-to-the-market hypothesis, electronic commerce is assumed to induce more market transparency and price competition, as well as stimulate an overall shift towards more use of markets and more use of the price mechanism rather than hierarchical modes of economic relations. This shift towards perfect, transparent markets as described in neoclassical economics is caused by a reduction in switching costs due to a technologyinduced reduction in transaction costs for search and coordination. The theoretical existence of perfect competitive markets relies on several preconditions, such as homogeneous products and services and actors’ rationality and complete information about all possible future outcomes. As analysed in information economics (Akerlof 1970), agency theory (Bergen et al. 1992) or transaction cost theory (Williamson 1985), departure from these preconditions as well as the assumption of opportunism lead to uncertainty about a transaction partner’s behaviour, which in turn grounds the economic justification for relational or hierarchical modes of governance. These market imperfections are reflected in the existence of transaction costs. Transaction costs can be differentiated into at least two distinct cost dimensions: coordination and motivation costs (Milgrom and Roberts 1992). Coordination costs refer to the costs associated with obtaining the information necessary for matching buyers and sellers, while motivation costs are associated with safeguarding against a transaction partner’s potential opportunistic behaviour, which can occur in the wake of information incompleteness and asymmetries as well as imperfect commitment. These are important concepts behind the theoretical explanation of market failure, imperfect competition, and, most importantly, hierarchical business relationships as a mode of governance in economic transactions. Taking the impact of the digitalization of information due to the Internet into account, one stream of literature emphasizes the mitigation of certain market imperfections. According to transaction cost theory, digitalization decreases the impact of information asymmetry, perceived product complexity and asset specificity – all of which are factors that usually favour
216
Ralf Reichwald et al.
hierarchical and hybrid forms of economic cooperation (Malone et al. 1987). In other words, the widespread use of information technology decreases the ‘unit costs’ of coordination. Since searching for partners in free markets tends to raise coordination costs compared to predetermined partners in hybrid or hierarchical relations, the market mode of transaction benefits to a much greater extent from a decrease in coordination costs. By the same token, lower production costs due to scale economies now over the Internet, which has always been the relative strength of the market method over hybrid or hierarchical modes, are also achieved and become a significant factor. Another traditional reason for hierarchies – the perceived complexity of products – is also said to be mitigated through improved communication of product descriptions. In terms of transaction cost theory, decreasing the transaction costs stemming from information search, coordination and product evaluation reduces information asymmetries and product complexity and, in turn, behavioural uncertainty. To the extent that information technology promotes flexible manufacturing technology, asset specificity will be reduced as well. Therefore, Malone et al. (1987) plausibly propose that ICT development will lead to a shift away from hierarchies in the direction of freer, more competitive markets and increased competition based on price. In sum, new media makes communication and information gathering less expensive, thereby decreasing the transaction costs for coordination and increasing the transparency of markets. According to transaction cost theory (Williamson 1985), this will favour market governance. Empirical evidence seems to support the thesis for a shift towards electronic markets, at least for standardized consumer goods (Smith et al. 2000). However, the above described effects of increased market transparency are not in the interest of firms. Firms benefit from market non-transparency because it leads customers to engage in long-term relationships and allows firms to extract consumer surplus. To reintroduce market non-transparency and switching costs for customers as well as to enhance their possibilities for price discrimination, many firms have started to develop more differentiated market offerings, such as by producing more variants, bundling products and services or customizing. Yet this strategy of differentiation shifts the problem of transparency to the issue of effectively and efficiently developing, producing and marketing these highly differentiated offerings. Customizable product and service bundles with many variants entail significantly more uncertainty and complexity for both consumers and firms. From a customer’s perspective, this strategy can lead to market failure as described above. From a firm’s perspective, it can lead to high flop rates in product development and many unsold product variants on stock. To properly manage this new uncertainty and complexity at the customer interface, firms have to realize that customers themselves possess valuable information about product solutions that meet market demand. Firms thus
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
217
have to devise means to elicit such customer information and incorporate it into value creation. At the same time, the combination of ICT and flexible manufacturing in particular has meanwhile altered the traditional product development and production process. Customer knowledge in the form of ideas, preferences, tastes and opinions can now easily be incorporated throughout the entire value-creation process. New opportunities for customer interaction In terms of communication, incorporating customer knowledge implies a shift away from a one-way communication (usually from firm to customer) towards a two-way communication process. This trend is growing and is manifested among others in two prominent but apparently distinct approaches: (1) mass customization (Pine 1993; Piller 2003) and (2) userdriven innovation (von Hippel 1986). Common to these and other approaches is a new perspective on customers as active value creators rather than mere value recipients. Their competencies should be co-opted by firms. Intuitively, established customer relationships seem to be necessary when tacit customer knowledge is to be made explicit and transferred to firms in order to co-create more value in an interactive, transformative process. But the driving force behind this ‘new’ type of customer relations cannot be seen to lie in the forces behind the ‘traditional’ form of relationships with customers as value recipients, i.e. in information asymmetries and asset specificity as suggested by transaction cost theory and agency theory. The driving force behind the need to transform customer relations is due to customers’ twofold increase in empowerment brought about by advances in ICT (Lindbeck and Wikström 1999). The first source of empowerment has been described above. New ICT mitigates problems of information asymmetries, uncertainty and complexity in product evaluations to the advantage of the customers. Customers benefit from cheaper information searches and price comparisons, and searching for the cheapest supplier for prospective purchases becomes a worthwhile endeavour. Increased market transparency results in stiffer competition based on price, and customers’ loyalty and commitment to a single supplier firm as in traditional relationships is weakened. Second, new ICT and digitalization have facilitated a cost structure of information production that allows customers to generate and widely distribute content as information goods and services in the market. Along with a reduction of search and communication costs for matching customers with common interests, public, product-like communities and discussion groups (Kollock 1999) emerge, either independently of firms or as a replacement of services formerly produced in the domain of firms (e.g. consumer support and after-sales service). In analogy to these two sources of empowerment, two types of information can be distinguished: (1) information as a coordination good, which needs to be acquired by customers
218
Ralf Reichwald et al.
at certain costs to make informed purchasing decisions; and (2) information as a market good, which can be created by customers and serve as input in a firm’s production as well as in other customers’ purchasing and production processes. From a firm’s perspective, the overall goal in the open innovation model is to unleash and utilize this new, productive potential of customers to (partly) externalize formerly internal value-creation processes. In this context, new ICT have enhanced a firm’s ability to integrate customers into open innovation processes in several ways. Previously, communicating for knowledge transfer in the physical world required physical proximity or personal interaction with customers, and this is no longer necessary. On the other hand, interaction with a large audience as enabled by new communication technologies entails potential compromises in the quality of the dialogue. The new ICT break with this trade-off between reach and richness because it is interactive in nature (Evans and Wurster 1999). Alba et al. (1997) suggest that interactive communication is characterized by three factors: it is (1) immediate (responses can occur right away), (2) multi-way (it involves two or more actors) and (3) contingent (the response of one actor follows logically from another’s action): 1
2
3
Immediate communication allows firms to transform episodic and oneway customer interactions into a continuous, two-way dialogue with customers at a considerably higher speed and level of persistence. Given the cost, timely effort and cognitive burden of traditional market research techniques such as focus groups and surveys for both firms and customers, new ICT and corresponding forms of virtual customer interaction facilitate more frequent and less time-consuming mechanisms of open innovation. Multi-way communication extends a firm’s reach of customer interactions beyond individual customer knowledge. By creating virtual communities, firms can step into the social dimension of customer knowledge that is created and shared by groups of customers (Kozinets 1999). Virtual communities let the firm delve into the experiential context of customer consumption patterns. Finally, new ICT and the Internet facilitate contingent communication, i.e. size and scope of possible customer input become contingent on prior contributions as well as on an individual customer’s level of motivation, interest and knowledge. Firms in particular benefit from contingency, as value creation and innovation tasks are complex and thus sequential in nature (Bessen and Maskin 2000).
In this situation, individual customer contributions can accumulate in a complementary fashion to increase the likelihood of overall success. Furthermore, customers can choose the size and scope of the input they would like to contribute. The choice of various levels of involvement is driven by customers’ perceived benefit from the interaction as well as by
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
219
their interests, knowledge and skills. Since customers have been shown to be very heterogeneous in these areas (Franke and Piller 2003), it is important for the success of the open innovation model that firms ‘modularize’ complex innovation tasks before they are externalized to customers (Bessen and Maskin 2002). Modules of various sizes and scopes can be self-selected by customers according to their preferences. In sum, new ICT improve a firm’s absorptive capacity for customer knowledge by allowing virtual means of customer integration and interaction to be more continuous and frequent, to take place in the social sphere (i.e. customer-to-customer), and to be more flexible in terms of time, size and scope of individual customer contributions.
From closed to open innovation at Adidas-Salomon Action research methodology Given the early stage of firm-initiated, open innovation, we have chosen an exploratory, qualitative research design to study the phenomenon in question. Understanding the structure and complexity of a rather new phenomenon calls for case study research to generate empirical data (Daft and Lewin 1990; Gummesson 2000; Danneels 2003). Action research is one method that can be adopted when working with case studies. Here, a researcher takes on the role of an active consultant or a ‘participant’ in the process under study (Gummesson 2000). The concept was introduced by Argyris and Schön (1978) under the name of a ‘theory of action’, based on the work of the social psychologist Kurt Lewin (1946). Lewin wanted to develop an approach that connected theory and practice, or action, in one process. According to Rapoport (1970: 499), ‘action research aims to contribute both to the practical concerns of people in an immediate problematic situation and to the goals of social science by joint collaboration within a mutually acceptable ethical framework’. The basic argument in favour of action research is access. Access refers to the opportunities available to the researcher to find empirical data and information for analysis and conclusions. The action research approach regards the active participation of the researcher in the process to be investigated as a better opportunity to develop a ‘pre-understanding’ of the topic at hand compared to the pre-understanding gained by interviews or detached observers (Gummesson 2000). Action researchers should become agents of change by participating in and fostering a process within an organization. By doing so, they gain deeper access to more relevant information. Gummesson (2000) has proposed a methodology for action research that consists of repeated cycles of modelling, intervention and testing, which Ödmann (1985) called a ‘hermeneutic spiral’ between pre-understanding and understanding. Each stage of the research should provide a new level of knowledge and thus understanding.
220
Ralf Reichwald et al.
The object of the action research process discussed in this paper is the NPD process in Adidas’s ‘Performance Division’, which is responsible for about 80 per cent of Adidas’s worldwide sales. Since the 1990s, the market for sports shoes has consolidated and new market entrants from the fashion industry have claimed a growing market share. Adidas is thus continuously forced to find a way to outperform its competitors. Moreover, customers are increasingly demanding exceptional design and product performance according to their individual needs, and retailers are beginning to move into the new arena of experience shopping by adding a variety of services in addition to their core products. As a reaction to these market demands, Adidas developed a range of products and successfully introduced them to the market in the form of mass customization. The programme, called mi adidas, offers customized shoes to consumers in specialized sales units located in dedicated retail stores. Consumers are given the opportunity to create their own unique footwear to meet exact personal specifications in terms of fit, function and design (Berger and Piller 2003). All shoes are made-to-order at an Asian factory, and the delivery time is about three weeks. Mi adidas can be considered as Adidas’s first move towards more intensive, customer–firm interaction. Today, the Performance Division serves as an agent of change and as a piloting unit for new business models and initiatives within the whole corporation. Encouraged by the insight gained from the mi adidas experience – that individual interaction with single consumers can be both possible and profitable, even for a mass producer – Adidas decided to explore customer integration further. The firm now aims to provide not only a small segment of customization but also to integrate large numbers of consumers into an open innovation process. The current state of NPD at Adidas NPD at Adidas today is a perfect example of closed innovation. New ideas from the innovation department, one of the most respected internal functions of the firm, are transformed into a product concept. At the same time, financial details, brand information and product strategies are taken into account. The conceptualization ends at a senior management meeting where decisions according to a specific assortment of factors are made. Based on these decisions, rough prototypes are produced and internally approved concepts are transferred to the Asian suppliers, who are responsible for producing mock-ups for trade shows, sales training and promotions. If all of these steps are successfully completed, production starts. Over the years, increasing competition and shorter product life cycles have increased the pace of this NPD process: You are looking at a so much more mature industry. Running shoes, for example, have been in huge mass production for almost 30 years. We are driven by steadily shortening product life cycles. About six
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
221
years ago, the average product life cycle lasted three seasons. Today, our products degenerate at least after one season and that is not the end of the story.2 Given the emphasis of innovation management literature on integrating customer input in the innovation process, we expected to find at least some of these methods in use at Adidas – especially when one considers the company’s brand-name consciousness and aggressive marketing orientation. But customer input seems to be of almost no importance to NPD at Adidas: Idea generation and evaluation starts with a brand strategy meeting, coming up with ideas and requirements. Those ideas can come from different places. Usually we just do an evaluation of the last stuff we did. So, in 2004, we did an evaluation of the stuff we did in 2003. The biggest input comes from our subsidiaries. Their input comes from the retailer, trend magazines, the competition and their own analysis of what the consumer is buying. The next biggest factor is us. Our info comes from the same people. The last big input is from senior management. This comment is a striking example of a lack of any customer integration, not even customer orientation, in the innovation process. When asked explicitly about open innovation and customer contributions to the innovation process, one participant stated: We work with athletes who are used to prototypes. We take a lot of prototypes out and we show them. The level of feedback from those users is quite good. Getting it is pretty good. Further, we do focus groups. For example, we picked the six coolest kids from six schools in New York. We showed them our stuff, explained the technology and asked for feedback. Adidas thus partly relies on ‘lead users’, here the professional athletes who test new shoes and provide feedback. The focus groups made up of the ‘six coolest kids’, however, is a typical, market-driven approach. This kind of customer integration is neither representative nor particularly insightful. It is more likely an attempt by product management to signal to the management board that it is ‘truly customer orientated’. The Adidas Customer Project Driven by the positive feedback from the mi adidas business unit and certain interventions carried out by the research team, Adidas’s management decided to build a system for customer integration in NPD. This ‘Adidas Customer Project’ covers the entire innovation process. It integrates customers on the level of idea generation and evaluation, technology
222
Ralf Reichwald et al.
assessment and concept testing. In addition, feedback about purchased products and the service experience in retail is ascertained. The project is divided into five ‘campaigns’, each consisting of different surveys, rating procedures, a conjoint analysis tool and a virtual customer lab. This Internetbased lab allows customers to come up with their own ideas in a structured manner as part of an idea generation competition. It also allows them to evaluate both the ideas of other users and the ideas generated in-house. In the following, we focus our discussion on this customer lab, which was developed jointly by our research team and Adidas.3 The design of this lab follows the requirements of a ‘toolkit for user innovation and co-design’ (von Hippel and Katz 2002). Toolkits for user innovation differ from product configuration systems, which are used in the mi adidas sales system. Customers are not limited to restricted assembling rules but can easily express their ideas within a wide solution space. In order to integrate untrained users as well, the virtual customer lab aims at a high level of user friendliness. Users do not have to be prepared or trained in any special way, nor do they need to have any particular skills or prior knowledge of the system to participate in the project. Building on research on the design of such customer-firm interaction systems (Dahan and Hauser 2002; Franke and Schreier 2002; Nambisan 2002; Thomke and von Hippel 2002; von Hippel and Katz 2002; Thomke 2003; Füller et al. 2004; Franke and Piller 2004), the Adidas virtual customer lab was conceptualized, coded, tested and placed online in July 2004. It follows a number of design requirements, including: •
•
•
•
Creativity. Innovative problem solutions do not come ad hoc to the users’ minds. It is crucial to create an environment that stimulates creativity, and several techniques can be applied to enrich the user’s imagination. For instance, this lab can present different future scenarios or a few vaguely drafted solutions as catalysts that lower the barrier to brainstorming. In addition, customers are able to return to a previous design stage in order to modify their input. In such a way, the virtual customer lab incorporates a ‘trial and error’ process. Service innovation, so-called ‘experience innovation’. The idea behind the toolkit is not only to generate feedback on individual products or product categories, but also to serve as an interaction tool with regard to the total customer experience with Adidas by including a focus on the purchasing process and the usage phase. The inclusion of ‘service innovation’ is a radical step for a product-centred brand such as Adidas. Efficiency. The toolkit evaluates and clusters user input automatically. It also represents user feedback in such a way that less editing work is needed. Of course, user ideas have to be read and understood by other human beings, but the grouping and clustering of these ideas is performed by the system. Community functionality. The benefit of interacting in communities is that its members can accept, enhance or simply comment on new product
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
•
223
ideas and continue to work on them in a collaborative manner. The virtual customer lab thus offers community functionality, which enables users to work jointly on a problem or provide each other with feedback and inspiration. Clear and transparent rules. A clear disclaimer of legal aspects concerning the exploitation rights of generated ideas and protection of data privacy ensures a transparent and trustworthy mode of communication.
During the piloting phase, the virtual customer lab was only accessible to customers of the mi adidas mass customized products in selected markets for two reasons. First, the product and application knowledge as well as the involvement of these customers is very extensive. We thus originally hypothesized that mi adidas customers are more willing to share innovative ideas. Second, at present Adidas has direct access only to its mi adidas clients. As the direct relationships already existed, channel conflicts with retailers were less likely. For legal reasons, customers were asked at the sales unit to participate in the project. Customers willing to participate received a personal access code to the project’s website. On the welcome page, users were addressed personally, and a clickable picture of their last purchased shoe was provided. Users were also invited to participate in the virtual lab and idea competition, which was divided into two parts: creation and evaluation. The creation process spanned the total customer experience with the brand, including service. Structured into 12 zones, users were able to generate and evaluate ideas with regard to, for example, the pre-sales phase, the sales process itself, the usage phase (including product innovation), as well as with regard to their ideas of the types of additional services Adidas could offer in the future. The 12 stages of the service experience were clustered in four topic areas (see Figure 10.1): • • • •
appointment and registration (subtopics: appointment process, layout of sales unit and registration process); adjustment and design (subtopics: scanning, fitting and design); purchase and delivery (subtopics: purchase process, waiting period and delivery process); usage and reorder (subtopics: usage, after-sales services and advanced services).
To support and foster creativity, all stages were illustrated by explicative pictures. An easy-to-use interface increased the users’ flow experience. Any user input on one of the stages was stored in the database in such a way that a subsequent and systematic analysis was able to be performed. After an input, customers were asked if they intended their own contribution as a comment, a suggestion for improvement or as a new idea. They were also requested to evaluate the perceived innovativeness, benefit for other customers, market potential and practicability of their own idea on a seven-
224
Ralf Reichwald et al.
point Likert scale. During the piloting stage, customers were also surveyed in order to get further insight into their motives as to why they wanted to contribute in the first place. After a user had submitted a contribution, he or she returned to the start pages in a loop-function to either send in another idea or to try the ‘evaluate’ function. Cross-evaluation of user ideas was one of the central community features in the starting phase of the toolkit. In this case, as well, evaluation followed the twelve-staged structure, represented by twelve pictures. Each user could select a situation and add a comment to another customer’s input. This comment could be a pure annotation, but it could also be an innovative continuation of the existing idea. After contributing, customers were again surveyed as described above. Evaluation Since the tool was launched in July 2004, 570 customers have been invited to participate (after they had purchased a mi adidas product). Of these, 272 customers visited the virtual customer lab website, which corresponds to a response rate of 47.7 per cent. Of the 272, 127 customers actively participated in the idea competition (46.7 per cent of responding customers). The rest contributed solely to other campaigns. This high level of willingness to participate may be related to the high degree of involvement shown by mi adidas customers (Berger and Piller 2003). However, it may also be seen
Figure 10.1 Screenshot of the Adidas virtual customer lab Source: The authors.
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
225
as an indicator of the general applicability of the open innovation system. After three months, 93 full contributions and 85 evaluations were posted. In an intermediate evaluation of the ideas by an Adidas expert jury, a crossfunctional team of five members of product management, marketing and innovation management evaluated and ranked all contributions with regard to their innovativeness, customers’ benefit, market potential and practicability (using the same seven-point Likert scale as the customers). From all contributions, about 10 per cent were evaluated as new ideas, while about 60 per cent were suggestions for improvements and 30 per cent were pure comments without any innovative input whatsoever. Overall, the Adidas team was impressed by the number of contributions and the innovativeness of many of the ideas, a number of which also incorporated ideas on how to improve the virtual customer lab itself.
Discussion and conclusion Open innovation is characterized by a shift of labour which moves valuecreating activities from the manufacturer to the customer. The customer thus becomes a prosumer and co-designer. The Adidas case demonstrates that open innovation can be applied as a viable strategy to increase the success rate of new innovations. At the same time, the effectiveness of open innovation is closely related to the development and deployment of new ICT. While open innovation no doubt demands an appropriate, institutional framework (which generally implies governmental, private and informal rules), we show that a successful, open innovation system also consists of important internal structures, tools, organizational values, norms and rules. Entrepreneurial behaviour moves between adaptation and creation. Thus, we argue that in a given institutional setting, corporate culture is the most important factor for a successful application of open innovation.4 Specifically, open innovation consists of two closely connected and often inseparable parts: (1) initiating a system of leadership for customers and employees alike; and (2) building an organizational structure for open innovation. Both tasks are rather new in management research and practice. Leadership of open innovation Strategies for managing customer behaviour in open innovation systems could draw upon models of employee behaviour (Bowen 1986). According to recent literature on leadership in corporations, three factors influence employee behaviour: ability, role clarity and motivation (Daft 2002). The behaviour of customers as co-creators can be viewed as being shaped by these same three factors: •
Ability. Are customers able to perform as expected? Ability refers to selecting the right customers and recruiting them as participants of an open innovation system within the organization (Anselmsson 2001;
226
•
•
Ralf Reichwald et al. Lüthje 2003). The evaluation of customer input at Adidas revealed a broad scope of quality customer contributions. Given limited resources to evaluate and follow up on the ideas, a pre-screening of customers would be beneficial. Pre-screening, however, would require developing a corresponding and adequate instrument. Role clarity. Do customers understand how they are expected to perform? Customers who would be able to participate based on their personality characteristics but who lack object or application knowledge need to be trained and educated (Lengnick-Hall et al. 2000). Education in this sense also includes matching the role expectations of customers with the expectations of the firm. Contributing customers to the Adidas project showed strong expectations of a timely response and realization of their ideas, which in turn requires a new kind of communication strategy for the firm. Motivation. Are there valued rewards for customers performing as expected? In the Adidas project, customers freely revealed information without asking for compensation (see also Henkel and von Hippel 2003). But customers in general do not release information for free. For them, the decision to participate in an open innovation system is basically the result of a simple economic equation (Franke and Piller 2003): If the (expected) returns exceed the (expected) costs, then customer participation may increase. The connection between performance and appropriate rewards has to be made apparent for contributing customers (Gurgul et al. 2002).This would involve a preliminary investigation into the types of benefits that motivate customers or users in a given market segment and a selection process for an appropriate reward structure.5
Customers able and willing to contribute need to be able to find competent and motivated firm employees who understand open innovation. Leadership for open innovation thus includes the ability to manage change in an existing organization that is moving from closed to open innovation (Anderson and Crocca 1993; Prahalad and Ramaswamy 2004): ‘Only when managers are freed from the conventional wisdom of innovation as a preserve of the manufacturer, can they assess their own firm’s most efficient role in the innovation process’ (von Hippel 1988: 117). Otherwise, a new kind of ‘not-invented-here’ problem will prevent customer input from being exploited by the NPD team. In the Adidas project, internal change management and cross-functional acceptance that open innovation can supplement conventional, closed innovation processes was recognized as the dominating factor for the long-term success of this initiative and its expansion. The virtual customer lab itself can be seen as an early concept test, as the head of the Adidas team remarked: The Virtual Customer Lab is an innovation of innovations. We are experiencing in a space that seems to have no borders, but many constraints. Suddenly, people at the company, who have never cared
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
227
about consumer innovation before, want to be part of the project. The customer lab itself is an experiment in how to deal with innovation in our corporation. Organization of open innovation The leadership tasks have to be supplemented by adequate organizational structures. The firm’s internal organization and its customer interfaces have to be set up in such a way that the firm’s ability to access, evaluate and utilize these new, external resources can be accomplished. The objective is to build up a sufficient amount of absorptive capacity (Cohen and Levinthal 1990). Jacob (2003) pinpoints the idea of absorptive capacity for open innovation by the use of a construct called ‘customer integration competence’. It describes the capabilities of a firm to perform customer integration efficiently and effectively, and consists of three different dimensions: (1) communication, i.e. the ability to gain access to relevant customer information during the innovation process; (2) management, i.e. the ability to create an organizational environment within the firm that can combine external customer input with internal firm resources effectively; and (3) efficiency, i.e. the ability to perform the integration processes in a way that they generate profits. In this context, one important task is to set the optimal extent of customer integration by clearly delimiting the activities a firm is willing to engage its customers in. At times, customer participation is seen as beneficial per se, following the logic of ‘the more the better’ (see Prahalad and Ramaswamy 2004). Greater levels of customer participation are seen to be related to a greater likelihood of innovative performance. This view, however, is too simplistic and ignores financial and cognitive costs for both the firm and the customer – costs that may be directly related to the degree of openness within the innovation system (Gales and Mansour-Cole 1995). In our Adidas case study, this particular question was a topic of heated debate. Only experimentation and trial-and-error processes within an open innovation system itself can unveil the optimal extent of customer integration within a given firm structure. Open innovation carries crucial implications for the design of the entire value-creating system within a firm (Ramirez 1999). It carries implications for the design of product-service bundles, which are most effective if they fit the customers’ business (or use) logic; for the design of the organizational structures through which open innovation strategies are constructed and implemented; and for the design of the tools and competencies essential to the performance of these activities. All of these design tasks are, in turn, closely related to the development and deployment of ICT as well as to an appropriate, institutional setting. The fact that innovations have already been global phenomena for a long period of time6 suggests that the main challenge for firms today is to successfully orchestrate a balanced move between the adaptation of new, institutional frameworks and the creation of an effective corporate culture.
228
Ralf Reichwald et al.
Notes 1 2 3 4 5 6
For a general discussion of the institutional conditions that need to be taken into account when introducing ICT, see also Part II of this book. This and all following quotes are taken from interviews with the product design team of Adidas’s Performance Division. Coding and graphic design was outsourced to a professional multimedia company. See also Schreyoegg, Khiari and Schmidt, Chapter 13 for a detailed understanding of the implementation of ICT as a cultural learning process. Hars and Ou (2002) discuss a possible structure. For more on the competitive advantages of ICT, see Fulk, Chapter 12.
Bibliography Akerlof, G. (1970) ‘The market for “lemons”: quality uncertainty and the market mechanism’, Quarterly Journal of Economics, 84: 488–500. Alba, J. W. and Hutchinson, J. W. (1987) ‘Dimensions of consumer expertise’, Journal of Consumer Research, 13 (March): 411–54. Alba, J., Lynch, J., Weitz, B., Janiszewski, C., Lutz, R., et al. (1997) ‘Interactive home shopping: consumer, retailer, and manufacturer incentives to participate in electronic marketplaces’, Journal of Marketing, 61 (July): 38–53. Anderson, W. L. and Crocca, W. T. (1993) ‘Engineering practice and co-development of product prototypes’, Communications of the ACM, 36 (4): 49–56. Anselmsson, J. (2001) Customer Perceived Service Quality and Technology Based Self Service, Dissertation, Lund Institute of Economic Research, Lund: Lund Business Press. Argyris, C. and Schön, D. A. (1978) Organizational Learning; a theory of action perspective, Cambridge MA: Addison-Wesley. Bergen, M., Schantanu, D. and Walker, O. C. (1992) ‘Agency relationships in marketing: a review of the implications and applications of agency and related theories’, Journal of Marketing, 56, July: 1–24. Berger, C. and Piller, F. (2003) ‘The miAdidas mass customization strategy’, IEE Manufacturing Engineer, 82: 42–6. Bessen, J. and Maskin, E. (2000) Sequential Innovation, Patents and Imitation, Working Paper, Cambridge MA: MIT. Bessen, J. and Maskin, E. (2002) ‘Sequential innovation, patents and imitation’, research on Innovation Working Paper, Boston MA (revised version July 2002). Booms, B. H., Davis, D. and Guseman, D. (1983) ‘Participant perspectives on developing a climate for innovation of new services’, in George, W. R. V. and Marshall, C. E. (eds) Developing New Services, Chicago IL: American Marketing Association, pp. 23–6. Bowen, D. (1986) ‘Managing customers as human resources in service organizations’, Human Resource Management, 25: 371–83. Bower, J. L. and Christensen, C. M. (1995) ‘Disruptive technologies: catching the wave’, Harvard Business Review: 43–53. Brockhoff, K. (2003) ‘Customers’ perspectives of involvement in new product development’, International Journal of Technology Management: 464–81. Chesbrough, H. (2003) Open Innovation: the new imperative for creating and profiting from technology, Boston MA: Harvard Business School Press. Childers, T. L. (1986) ‘Assessment of the psychometric properties of an opinion leadership scale’, Journal of Marketing Research, 23: 184–8.
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
229
Cohen, W. M. and Levinthal, D. A. (1990) ‘Absorptive capacity: a new perspective on learning and innovation’, Administrative Science Quarterly, 35: 128–52. Cooper, R. G. (1993) Winning at New Products: accelerating the process from idea to launch, 2nd edition, Boston MA: Perseus Books. Craig, C. S. and Ginter, J. L. (1975) ‘An empirical test of a scale for innovativeness’, Advances in Consumer Research, 2 (1): 555–62. Crawford, C. M. (1977) ‘Marketing research and the new product failure rate’, Journal of Marketing, 41: 51–61. Daft, R. L. (2002) The Leadership Experience, Orlando FL: Harcourt College Publishers. Daft, R. L. and Lewin, A. Y. (1990) ‘Can organization studies begin to break out of their normal science straitjacket’, Organization Science, 1: 1–9. Dahan, E. and Hauser, J. R. (2002) ‘The virtual customer’, Journal of Product Innovation Management, 19 (5): 332–53. Danneels, E. (2003) ‘Tight-loose coupling with customers: the enactment of customer orientation’, Strategic Management Journal, 24: 559–76. Davidson, J. H., (1976) ‘Why most new consumer brands fail’, Harvard Business Review, 54: 117–22. de Brentani, U. (2001) ‘Innovative versus incremental new business services: different keys for achieving success’, Journal of Product Innovation Management, 18 (3): 169–87. de Brentani, U. and Ragot, E. (1996) ‘Developing new business-to-business professional services: what factors impact performance?’, Industrial Marketing Management, 21: 231–41. Drucker, P. F. (1998) ‘The discipline of innovation’, Harvard Business Review, 76 (6): 149–57. Easingwood, C. J. and Storey, C. (1991) ‘Success factors for new consumer financial services’, International Journal of Bank Marketing, 9 (1): 3–10. Edgett, S. (1994) ‘The traits of successful new service development’, Journal of Services Marketing, 8 (3): S. 40–9. Enos, J. L. (1962) Petroleum Progress and Profits: a history of process innovation, Cambridge MA: MIT Press. Evans, P. and Wurster, T. S. (1999) Blown to Bits: how the new economics of information transforms strategy, Boston MA: Harvard Business School Press. Franke, N. and Piller, F. (2003) ‘Key research issues in user interaction with configuration toolkits in a mass customization system’, International Journal of Technology Management, 26: 578–99. Franke, N. and Piller, F. (2004) ‘Toolkits for user innovation and design: an exploration of user interaction and value creation’, Journal of Product Innovation Management: 401–15. Franke, N. and Schreier, M. (2002) ‘Entrepreneurial opportunities with toolkits for user innovation and design’, International Journal on Media Management, 4: 225–34. Freeman, C. (1968) ‘Chemical process plant: innovation and the world market’, National Institute Economic Review, 45: 29–57. Freeman, C. and Soete, L. (1997) The Economics of Industrial Innovation, London: Pinter. Füller, J., Bartl, M., Ernst, H. and Mühlbacher, H. (2004) ‘Community based innovation: a method to utilize the innovative potential of online communities’, Proceedings of the 37th Hawaii International Conference on System Sciences 2004, Kona HA: IEEE.
230
Ralf Reichwald et al.
Gales, L. and Mansour-Cole, D. (1995) ‘User involvement in innovation projects: towards an information processing model’, Journal of Engineering and Technology Management, 12: 77–109. Gummesson, E. (2000) Qualitative Methods in Management Research, 2nd edition, Thousands Oaks, CA London: Sage. Gurgul, G., Rumyantseva, M. and Enkel, E. (2002): ‘Customer integration: establish a constant bilateral knowledge flow’, University of St Gallen Discussion Paper, 47, St. Gallen. Hars, A. and Ou, S. (2002) ‘Working for free? Motivations for participating in opensource projects’, International Journal of Electronic Commerce, 6: 25–39. Henkel, J. and von Hippel, E. (2003) Welfare Implications of User Innovation, Cambridge MA: MIT Sloan School of Management Working Paper. Herstatt, C. and von Hippel, E. (1992) ‘Developing new product concepts via the lead user method’, Journal of Product Innovation Management, 9: 213–21. Hirschman, E.C. (1980) ‘Innovativeness, novelty seeking, and consumer creativity’, Journal of Consumer Research, 7 (December): 283–95. Jacob, F. (2003) ‘Kundenintegrations-Kompetenz’, Marketing-Zeitschrift für Forschung und Praxis, 25 (2): 83–98. Jacoby, J., Troutman, T., Kuss, A. and Mazursky, D. (1986) ‘Experience and expertise in complex decision making’, Advances in Consumer Research, 13: 469–72. Katila, R. and Ahuja, G. (2002) ‘Something old, something new: a longitudinal study of search behavior and new product introduction’, Academy of Management Journal, 45: 1183–94. Kaulio, M. A. (1998) ‘Customer, consumer, and user involvement in product development’, Total Quality Management, 9: 141–9. King, C. W. and Summers, J. O. (1970) ‘Overlap of opinion leadership across product categories’, Journal of Marketing Research, 7 (1): 43–50. Kogut, B. and Meitu, A. (2000) ‘The emergence of e-innovation: insights from open source development’, Reginald H. Jomes Center working paper No. 00–11, Wharton School of Business, University of Pennsylvania. Kollock, P. (1999) ‘The economies of online cooperation: gifts and public goods in cyberspace’, in Smith, M. and Kollock, P. (eds) Communities in Cyberspace, London: Routledge. Kozinets, Robert V. (1999) ‘E-tribalized marketing? The strategic implications of virtual communities of consumption’, European Management Journal, 17 (June): 252–64. Laursen, K. and Salter, A. (2004) ‘Open for innovation: the role of openness in explaining innovation performance among UK manufacturing firms’, Tanaka Business School Working Paper, Imperial College London. Lengnick-Hall, C., Claycomb, V. and Inks, L. (2000) ‘From recipient to contributor: examining customer roles and experienced outcomes’, European Journal of Marketing, 34: 359–83. Lewin, K. (1946) ‘Action research and minority problems’, Journal of Social Issues, 2: 34–6. Lindbeck, A. and Wikström, S. (1999) ‘The ICT revolution in consumer product markets’. Seminar Paper No. 670, Institute for International Economic Studies, Stockholm University. Lüthje, C. (2003) ‘Characteristics of innovating users in a consumer goods field’, Technovation, 23: 245–67.
Open innovation 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
231
Malone, T., Yates, J. and Benjamin, R. (1987) ‘Electronic markets and electronic hierarchies’, Communications of the ACM, 1998, 30 (6): 484–97. Manning, K. C., Bearden, W. O. and Madden, T. J. (1995) ‘Consumer innovativeness and the adoption process’, Journal of Consumer Psychology, 4 (4): 329–45. Milgrom and Roberts (1992): Economics, Organization and Management, Englewood Cliffs NJ: Prentice Hall. Nambisan, S. (2002) ‘Designing virtual customer environments for new product development: toward a theory’, Academy of Management Review, 27 (3): 392–413. Ödman, P.-J. (1985) ‘Hermeneutics’, in Husen, T. and Postkethwaite, N. (eds) The International Encyclopedia of Education, Oxford: Pergamon, pp. 2162–9. Piller, Frank T. (2003) Mass Customization, 3rd edition, Wiesbaden: Gabler. Pine, B. J. (1993) Mass Customization, Boston MA: Harvard Business School Press. Prahalad, C. K. and Ramaswamy, V. (2004) The Future of Competition: co-creating unique value with customers, Boston MA: Harvard Business School Press. Quinn, J. B. (2000) ‘Outsourcing innovation’, Sloan Management Review, 41, Summer: 13–28. Ramirez, R. (1999) ‘Value co-production: intellectual origins and implications for practice and research’, Strategic Management Journal, 20: 49–65. Rapoport, R. N. (1970) ‘Three dilemmas in action research’, Human Relations, 23: 499–513. Reichwald, R., Ihl, C. and Seifert, S. (2004) ‘Innovation durch Kundenintegration’, in Frey, D., Rosenstiel, L. v. and Graf Hoyos, C. (eds) Wirtschaftspsychologie (Angewandte Psychologie, Band 2), Weinheim: Beltz/PVU. Rice, R. and Rogers, E. M. (1980) ‘Reinvention in the innovation process’, Knowledge: Creation, Diffusion, Utilization, 1: 499–514. Smith, M. D., Bailey, J. and Brynjolfsson, E. (2000) ‘Understanding digital markets: review and assessment’, in Brynjolfsson, E. and Kahin, F. (eds) Understanding the Digital Economy, Cambridge MA: MIT Press, pp. 99–136. Thomke, S. (2003): Experimentation Matters, Boston MA: Harvard Business School Press. Thomke, S. and von Hippel, E. (2002) ‘Customers as innovators: a new way to create value’, Harvard Business Review, 80 (4): 74–81. Utterback, J. (1994) Mastering the Dynamics of Innovation, Boston MA: Harvard Business School Press. von Hippel, E. (1978) ‘Successful industrial products from customer ideas: presentation of a new customer-active paradigm with evidence and implications’, Journal of Marketing, 42: 39–49. von Hippel, E. (1986) ‘Lead users: a source of novel product concepts’, Management Science, 32 (7): 791–805. von Hippel, E. (1988) The Sources of Innovation, Oxford: Oxford University Press. von Hippel, E. (1998) ‘Economics of product development by users: the impact of sticky local information’, Management Science, 44 (5): 629–44. von Hippel, E. and Katz, R. (2002) ‘Shifting innovation to users via toolkits’, Management Science, 48: 821–33. Vojak, B. and Suarez, C. (2002) ‘Sources of information used in new product and process technology planning within the electron device industry’, Proceedings of the IEEE International Engineering Management Conference IEMC-2002, edited by Tariq S. Durrani, pp. 623–8. Williamson, O. E. (1985) The Economic Institutions of Capitalism, New York: Free Press.
Subsection B The social construction of institutions and technology
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
11 Next generation information and communication technologies deployment in Japan Mohammed Akhtar, Yoshiya Teramoto and Caroline Benton Introduction ‘Ubiquitous’ computing and communication is a natural development in the realization of the future information-based society. In order to get to the point where one is able to communicate with others anywhere and anytime with voice, data and multimedia, existing wireless networks and corresponding handsets must be upgraded to support high-speed data communication. At present, a number of different digital networks and analog networks on this level of sophistication exist worldwide, yet most of them simply co-exist and are not interoperable. The next generation of mobile communications will have to be based on a standardized protocol that major companies and governments worldwide agree upon. Yet not all of the many operators in the different countries that have begun deploying such next generation systems are experiencing the same degree of success. In this chapter we review the deployment of third generation (3G) wireless communications technology in Japan with the goal of identifying the causes of the success and failure of different operators. We find that when a wireless network operator delivers benefits at the end-user level, the operator level, the national level and the global level, it becomes a formidable player in the market. An operator meeting these four-tier benefits creates a win–win relationship for all stakeholders involved. In the Japanese market where two of the main 3G specifications co-exist, the ‘W-CDMA’ and ‘CDMA 1x’, our analysis finds that selection of technology, strong marketing, unique service and product advantages were the main success factors. Being the first to enter the market did not guarantee instant success, as seen in the case of NTT DoCoMo’s FOMA. Such lessons are increasingly being fed back into Japanese policy-making, and their effects will keep Japan and other Asian countries well poised as leaders in next generation mobile technologies. This chapter is organized as follows. The next section briefly describes the developments leading to the current generation change and thus to the emergence of the third generation (3G). Subsequent sections describe a
234
Mohammed Akhtar et al.
four-tier benefit model to evaluate an operator deploying the 3G mobile networks and take a closer look at the Japanese market of mobile communications, analysing the performance of the three major companies – NTT DoCoMo, KDDI and Vodafone KK. The next section compares the Asian market with others to put the Japanese situation in perspective. Based on the lessons learned from 3G deployment in Japan, in the following section we look beyond 3G and predict Asia’s leading role in future technologies. The next section highlights a number of the factors behind Japan’s predicted success, and the final section summarizes our observations.
Generation change in mobile communication technologies Over the past decade, mobile communication has been one of the fastest growing areas in technology. In order to appreciate the impact of next generation mobile communication technology, it is necessary to briefly review the ongoing trends of different generations of communication technology. What becomes clear from an overview is that rapid changes between different generations and different applications took place, and these changes made the evolution of the present-day third generation wireless networks both possible and necessary. In the following paragraphs, we take a look at the previous generations and demonstrate how their limitations led to the development of the next. We then take a closer look at the features that characterize third generation systems. First generation (1G) – late 1980s The first generation mobile communication systems were analog systems, which are defined as circuit-switched, voice-telephone communications via cellular radio channels. They are also referred to as Advanced Mobile Phone System (AMPS) in the United States, Nordic Mobile Telephony (NMT) in Europe and Total Access Communication System (TACS)/Extended Total Access Communication System (ETACS) in Europe and much of rest of the world. Out of this first generation of mobile communication came the concepts of basic mobility, basic service and compatibility in communications (Holman and Toskala 2000). First generation provides analog voice service but no data service. It is convenient for communication, especially in sparsely populated regions, but its shortcomings – lower frequency availability, same frequency band interference and poor security – became very apparent in the urban areas. Second generation (2G) – early 1990s Innovations in digital technology allowed for the introduction of digital networks, which marked the beginning of a significant growth phase in
Next generation ICT deployment in Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
235
wireless communications. This evolution towards the second generation (2G) of cellular systems was driven by the rapid growth of subscribers and the limitations faced by the incompatible 1G system. The 2G mobile communications systems aimed at moving forward from basic mobility to advanced mobility (i.e. roaming), from basic service to more services (e.g. data), and from small, incompatible coverage towards global solutions. The 2G technologies include Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Global System for Mobile Communication (GSM) and Personal Digital Communication (PDC), the latter found exclusively in Japan. Second generation CDMA, TDMA and PDC offered data transmission but only one-way. The 2G mobile communications also provided enhanced calling features such as caller ID, but lacked an alwayson data connection. The Japanese 2G cellular data networks delivered data rates up to 9.6 kb/s (kilobits per second) for upload and up to 29.8 kb/s for download. Moreover, and with regard to technology, 2G provided more frequencies, better security, more capacity and better speech quality-ofservice (QoS) over the first generation communication systems. However, interoperability was still a problem as Japan, the United States, Europe and the rest of the world were on totally different protocols. Second generation with data enhancements (2.5G) – mid-1990s 2.5G is essentially a software overlay on an existing 2G network that utilizes existing base stations but requires new handset devices. It acts as a transition bridge from 2G to a future network, especially for the service provider. Most carriers migrate to this interim technology before upgrading to the next generation. With an increasing number of mobile users, a shortage of bandwidth made a transition to new technologies inevitable. Third generation (3G) – early 2000 The 3G technologies were developed in order to meet the demand for more capacity and support for innovative broadband multimedia services. They needed to support a wide range of vendors and operators and offer choice, competition and affordability in order to rise to the challenge of truly implementing communication with anyone in any place at any time. The 3G mobile communications systems met this challenge first of all by introducing revolutionary innovations; 3G networks can transmit data at 144 kb/s, or up to 2 mb/s (million byte per second), from fixed locations with features such as improved system capacity, backward compatibility with second generation systems, multimedia support and high-speed packet data services, thus enabling not only voice service but also data delivery. They are specifically designed for multimedia communication, whereby person-to-person communication can be enhanced with high quality images and video. Secondly, by converging GSM and CDMA into a single, official,
236
Mohammed Akhtar et al.
globally roamable standard, 3G provides seamless roaming and inserts a long-term, market-driven strategy of service and models for global radio access and global solutions. In such a way, 3G generates an innovative value chain and real user benefits that can drive genuine market demand. In sum, access to information and services on public and private networks are enhanced by the higher data rates and the new, flexible communication capabilities of 3G systems. Third generation CDMA modes are most commonly known as cdma2000, W-CDMA and TD-SCDMA. Their deployment is not only a technical problem, but also involves regulatory and economic issues. The manner in which 3G is rolled out depends on the business policy of the mobile operators and on the licence requirements imposed by the regulatory authority. Subscriber demand is also a crucial factor in the deployment of 3G. As the users’ expectations for more service increase, any successful 3G licence bidder has to take the time-to-market factor seriously in order to be successful. In recent years, standardization has been actively pursued by all countries in order to achieve high speed packet data transmission over wireless networks (3GPP 2001a). Key demands made on new generation systems are (1) a high degree of design commonality worldwide, (2) compatibility of services, (3) use of small pocket terminals with worldwide roaming capability, (4) Internet and other multimedia applications and (5) a wide array of services and terminals (3GPP, 2001b). The global cellular community introduced the term ‘3G’ mainly to refer to the next generation of mobile services (3GPP2 2000), and in 1992, the World Administrative Radio Conference (WARC) defined the frequency spectrum for 3G to support advanced services, particularly data transmission. International Mobile Telecommunications-2000 (IMT-2000) is the International Telecommunications Union’s (ITU) effort to develop a rigorous definition regarding 3G wireless communication (3GPP2 2001a, 2001b). Hence, 3G networks, which represent wireless access to global and metropolitan area data networks, are standardized in the IMT-2000 (Chan and Lin 2002). The original target of the third generation process was a single, common, global IMT-2000 air interface (Nikula et al. 1998), and several terms are used to describe networks that meet 3G specifications. According to IMT2000, minimum 3G data rate requirements fall into three categories: (1) 2 mb/s to stationary users (i.e. fixed location); (2) 384 kb/s to pedestrian users (3 m/hr (million/hour)); and (3) 144 kb/s to vehicular users (60 m/hr). Other requirements include support of symmetrical and asymmetrical data transmission, standardized interfaces, improved voice quality, greater capacity, multiple simultaneous services, global roaming ability across networks, improved security and flexibility of service (Parry 2002). Multiple radio technology options were included in the IMT-2000 standard to allow seamless service evolution from the various 2G mobile standards that have been extensively deployed around the world. For example, the wide-band CDMA (W-CDMA) was adopted in Europe and Japan. W-CDMA can support mobile voice, images, data and video communications at up to
Next generation ICT deployment in Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
237
2 mb/s for local area access and 384 kb/s for wide area access. In the United States, cdma2000 is employed for backward compatibility with CDMA (IS-95). For instance, the W-CDMA system allocates its upload and download frequency band at 1.92–1.98 GHz (GigaHertz) and 2.11–2.17 GHz, respectively. It allows a maximum data transmission rate of 2 mb/s (indoor) with a channel spacing of 5 MHz (MegaHertz). Operating at a new frequency band may require additional upgrading of a particular operator’s infrastructure and capital expenditure. In contrast, cdma2000 can be deployed in essentially all cellular systems, PCs or in the new IMT-2000 spectrum. Its high spectral efficiency enables high traffic deployment in any 1.25MHz channel of spectrum. The difference in modulation schemes between GSM and W-CDMA also result in different performance requirements. In Japan, W-CDMA and cdma2000 are the two principal versions of 3G with superior voice quality, and they continue to be the most popular. W-CDMA offers much greater data speed and data bandwidth (Feng et al. 2004). FOMA (freedom of mobile access) is a proprietary version of W-CDMA technology incorporated by NTT DoCoMo and thus will be referred to as W-CDMA here. On the other hand, cdma2000 offers complex scrambling to provide privacy protection. The 3G wireless networks in Japan deliver data rates from 64 kb/s for upload and in the order of 200 kb/s for download. The 3G networks in Japan are being upgraded at a fast pace in order to deliver data connection rates from 2 to 10 mb/s. This kind of higher speed allows the transmission of video and two-way telephony, rapid download of movie sketches, music and JAVA applications. The Java language is an object-oriented programming language created by James Gosling and other engineers at Sun Microsystems. Other data connections such as downloading information or JAVA applets are also several times faster on 3G networks than on older 2G networks. The above discussion emphasizes the fact that third generation systems meet basic bandwidth demand in combination with global mobility and interactivity. This unique combination of capabilities stimulates new applications and opens up new areas to business opportunities (Qiu et al. 2002).
Four-tier benefits of next generation wireless networks As discussed above, shortage of bandwidth itself is not driving the move to deploy 3G and beyond (Raivio 2001). The transition is taking place because over time a demand for new services that cannot be implemented with the existing systems has emerged. With the introduction of new systems, it is expected that competition, innovation and customer benefits will increase dramatically. But how does one know whether these goals are being met or not? We set our focus on the effects of the deployment of 3G mobile communications technology and attempt to evaluate the delivery of benefits at the end-user, operator, national and global levels. We call it a four-tier benefit model of evaluation and discuss it in more detail below on the basis
238
Mohammed Akhtar et al.
of trends displayed by equipment manufacturers and communications operators in mobile telephony. We maintain that operators who successfully capture these four-tier benefits generally become formidable players in the marketplace and are able to create a win–win relationship for all stakeholders involved. End-user level The end user or targeted users are, for the most part, the general consumers. User demand is a key factor for long-term, market-driven creativity, and this demand is increasing at a steady pace. The crucial issues for the user include price competitiveness, ease of use, and the usefulness and attractiveness of the running applications. Furthermore, mobile phone users are experiencing a need to interact with machines and to have machines interact with other machines over radio connections. Customer choice depends not only on the technology and implementation, but also on other issues, particularly those related to the local region, such as content. Content should be appropriate to the local culture and in the local language (Watson 2001). Thus, local content in the mobile communication environment has a major effect on the end-user level. Since end users usually choose services based on the services’ ability to meet their own personal needs and not purely on their technical merit, the deployment of 3G networks and devices must capture user needs in terms of mobility, increased choices, competitive and more affordable pricing, as well as personalized and tailored offerings (Ralph and Shephard 2001). Operator level Operators need to integrate the goals of positioning themselves as a leader in technology and an innovator of service differentiation in the marketplace with the goals of delivering value to shareholders and increasing both average revenue per unit (ARPU) as well as total revenue. In recent decades, mobile operators have converged on the idea that mobility will become the norm for any communication service. Operators around the globe are beginning to undergo a paradigm change from telephone company to value-added information provider. Many operators are moving away from the time-based approach to one that gives users the freedom to access certain services as much as they want. As a result of this paradigm change, a variety of different technologies are already in use throughout the world on the operator level. Due to the different technologies and frequency allocations, specific arrangements between operators, such as multi-mode and multi-band handset and roaming gateways between the different core networks, need to be in place in order to achieve global roaming capacity. Cost, attractiveness to future customers, service to existing customers and competitiveness with other technologies are the important parameters on which operators should base their strategy in order to run a business
Next generation ICT deployment in Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
239
successfully. At the same time, operator arrangements should not be visible to the end user, and global roaming terminals should merge for the users who are willing to pay for global service (Holman and Toskala 2002). In order to clearly differentiate services and market these services successfully to users, operators also need to integrate the role of the entertainment and media industries with shareholders as well. National level National level benefits in this four-tier benefit model for wireless network deployment consist of benefits or opportunities such as higher employment rates, growth in national income, positioning as a regional telecoms hub, positioning as a leader in 3G and beyond and export opportunity in mobile contents and next generation technology. At the same time, however, it also encompasses creating a unique chain of value for customers, business frontiers and shareholders through collaboration with equipment manufacturers, content providers and operators. Although every company has its own way of implementing technology and providing service that reflects its own approach to the market and subscribers, for harmonization at the national level it is necessary for a company to work in a highly collaborative manner. Global level Harmonization of standards and delivery of ubiquitous communications any time, anywhere and across any device are the main features of this level. From the viewpoint of standardization, level four enables the inter-working of different systems and products, which in turn allows both mass production and reasonable prices. From the global perspective, level four provides an incentive to work together with other prospective organizations in a way that would allow knowing each other and sharing experience for the maximum benefit of all the counterparts. At the same time, the market player cannot lose sight of the objectives of ubiquitous communications. A company must keep the compatibility of its products in mind. In order for the next generation communication systems to be marketed and deployed, smooth passage from the end-user level to the global level should be ensured. This ideal is reflected in the recent discussion of the ‘glocal’ idea, which is a combination of the terms ‘global’ and ‘local’, or end-user level.
The 3G market in Japan Snapshot of the Japanese market Japan’s telecommunications market consists of three main players – NTT DoCoMo, Vodafone and KDDI. In the beginning, these three Japanese companies offered quite different visions of new developments in the
240
Mohammed Akhtar et al.
direction of third-generation systems. NTT DoCoMo was the first to launch 3G services in mobile communication in Japan in 2001, followed by KDDI in April 2002 and Vodafone in December 2002. These three companies may be characterized as three parallel, independent and competing 3G networks (see Table 11.1). NTT DoCoMo’s 3G network brand name is FOMA and it uses the WCDMA technology standard. FOMA experienced slower growth due to limited coverage, short battery life and expensive handsets. To address these shortcomings, DoCoMo launched a series of new handsets with longer battery life in early 2003, and FOMA service coverage expanded from 93 to 96 per cent of the population by the end of September 2003, reaching one million subscribers. As of 31 May 2004, DoCoMo’s 3G subscribers numbered about four million, which indicates the momentum that they have finally been able to achieve. While Vodafone uses the same standard as NTT DoCoMo, it had a soft launch with limited coverage and still has a very small number of 3G subscribers – nearly 0.15 million as of May 2004. Table 11.1 Market landscape of Japan, August 2003 Topic
KDDI
NTT DoCoMo
Vodafone KK
Total subscribers
15.08 million
44.86 million
14.56 million
3G subscribers
9.73 million
0.79 million
0.08 million
Network standard
cdma2000–1x
W-CDMA
W-CDMA
Data transmission rate (maximum uplink)
64 kb/s
64 kb/s
64 kb/s
Maximum transmission speed
144 kb/s
384 kb/s
384 kb/s
Launch date
1 April 2002
October 2001
20 December 2002
3G handset market price (new model)
approx. 20,000 yen approx. 35,000 yen approx. 35,000 yen
3G major contents
GPS TV-Telephone TV-Telephone movie clip delivery, transmission of images and music
Different services
EZweb
i-mode
J-sky
International roaming CDMA countries Not yet (outbound, fully with special Global handset-based) Passport handset
GSM countries with special Vodafone Global Standard dual GSM/W-CDMA handset
Major technology partners
Japan / North America / Europe
Japan / North America
Source: Japan Telecom Monthly 1999.
Japan / North America
Next generation ICT deployment in Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
241
KDDI uses a different version of the cdma2000–1x standard. However, KDDI 1x achieved 7 million subscribers after its first year of operation and surpassed 10 million subscribers by September 2003. KDDI has moved almost all subscribers from 2G to 3G networks and reached about 14 million in May 2004 (Fasol 2003). This rapid increase has been attributed to nationwide, broad service coverage, backward compatibility with the existing IS-95B network and KDDI’s extensive CDMA operations experience. Key factors behind KDDI’S 1x service success KDDI was not first to launch 3G but it achieved a high level of success within a short period of time. Figure 11.1 depicts KDDI’s success relative to other wireless providers in Japan. The driving factors behind this success can be summed up by four main parameters: marketing; service and products; technology, and timing. Marketing Upon closely examining KDDI’s marketing practices, we found that their services were on the front line of advertisement and placed little emphasis on their technology platform. Only after having acquired a significant market share did they start to use technical terms like ‘3G’ in their ads. In other words, they targeted for the market push, not for a technology pool with an innovative value chain. The approach is one of gradual availability of increasingly higher speed service offerings as the market demands them. They provided rapid rollout, announced monthly status of coverage completion and incorporated a variety of other marketing strategies such as targeting young user groups by providing an up to 50 per cent discount for students and a 25 per cent discount for members of the same family. They aimed to be the first in the world to offer international compatibility for photo transmission with their South Korean counterparts. They were also conscious about the charges they were offering. Under the newly envisioned two-tiered levy system, KDDI planned to cut the uniform rate for users whose data transmission volume does not exceed the highest number of allowed packets, which is 400,000. Today, KDDI’s 3G network covers more than 90 per cent of the Japanese population. Service and products KDDI capitalized on the benefits of cdma2000 in order to build a significant competitive advantage. Its 3G cdma2000–1x provided uncompromised QoS and a much better service portfolio compared to those of 2G cdmaOne. It offered new services and improved end-user experience by deploying rapidly and utilizing long-term, market-driven creativity. There is no degradation in specification, and the increase in price was minimal compared to 2G
242
Mohammed Akhtar et al.
cdmaOne phones. They also provided a wide range of market-focused applications. In this regard, they offered photo-mail (camera-equipped phones), movie-mail (movie recorder-enabled phones) and content rich applications at a user-friendly cost. With cdma2000–1x, KDDI also significantly expanded their EZWeb Internet services. At present, the EZWeb portfolio includes eznavigation (location-based services), ezmovie (video distribution), EZweb@email (multimedia messaging) and ezplus (high-speed web access). KDDI charges a uniform amount a month with its EZ flat rate 3G data service that allows users to send an unlimited amount of data at the flat rate. Soon KDDI will offer a service that enables music to be downloaded to cellular phones so that they can be used as portable music players. This development will make KDDI the first Japanese mobile phone company to offer such a service. In the beginning, the handsets will be able to store five-minute song pieces, and fifty song models are under development. Once stored, the songs can be played repeatedly. KDDI paid attention to such kinds of handset improvements as they are crucial to sustain the company’s success. Technology The inherent backward compatibility to IS-95 guaranteed KDDI’s 3G continuous service availability within 2G cdmaOne’s service coverage areas. (Interim Standard 95 (15–95) is the first CDMA-based digital cellular standard pioneered by Qualcomm.) The maturity of the technology inherited from IS-95 operational records helped KDDI to ensure ‘No increase in handset physical dimensions’ and ‘No degradation in handset battery life’. KDDI decided to continue with the increasing numbers of cdmaOne users and to improve its data transmission speed rather than introduce the cdma2000 system hastily. They could therefore utilize with slight modifications its existing, nationwide cdmaOne facilities, which provided a high-speed data communication service across the country without having to invest a large amount in new installation, as NTT DoCoMo was forced to do. At the same time, they offered advanced, lightweight, easy-to-use terminals with intuitive interfaces that provided instant, real-time, multimedia communications. Finally, KDDI’s expertise in 2G cdmaOne network operations contributed to the stable service offering of cdma2000–1x. Timing Deciding when a technology or standard should be marketed is a crucial matter. When a technology is first discovered there is a burst of research activity. After a while this activity subsides, people become more aware of it and its implementation, and eventually customers begin to respond to the new technology. Arriving in the market too late may negatively impact a company’s success (Tanenbaum 2003), but so too can arriving too early. NTT’s launch in October 2001 enabled it to be the first to enter the market,
Next generation ICT deployment in Japan 12,000K 11,000K 10,000K 9,000K 8,000K 7,000K 6,000K 5,000K 4,000K 3,000K 2,000K 1,000K
3G Total
FOMA
Au 1x
D Ja ec n ’0 3 Fe b M ar Ap r M ay Ju ne Ju Au ly gu st
Ju n Ju l Au g Se pt O ct N ov
r
ay
Ap
M
Fe b ar M
ov D Ja ec n ’0 2
N
ct
’0 1
0K
O
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
243
Vodafone KK
Figure 11.1 Comparison of KDDI 1x, NTT DoCoMo’s FOMA and Vodafone’s W-CDMA subscriptions Source: Morgan Stanley 2003.
but it proved to be too early, as users were not as aware of its usefulness in Japan as they later became. At that time the technology was in its early stage of research and marketing. Second to launch 3G technology in April 2002, KDDI’s timing proved to be the most accurate of the three major competitors, and this proper timing contributed to KDDI’s success (see Figure 11.1). In short, cdma2000 came out ahead of W-CDMA as the most successful 3G network technology because it cost less to deploy than the W-CDMA equivalent. The cdma2000 networks are supported by a more extensive range of handsets and can be more easily upgraded to next generation technology. This advantage was utilized by KDDI. The company offered content-rich applications at a user-friendly cost, reliable handsets with high functionality and an equal or better coverage area than 2G service. KDDI was able to exploit their expertise in data communication on the basis of movie-mail enabled handsets that supplied their customers with a large amount of meaningful and entertaining data. FOMA gains momentum In 1999, NTT DoCoMo launched the i-mode Internet service for mobile phones on the back of its previous, national DoPa packet switched network. I-mode provided users one more step of freedom to view web pages and
244
Mohammed Akhtar et al.
send e-mail. Since its introduction, i-mode has become the driving force behind m-mode, e-motion and other i-mode derivatives throughout the world. Using i-mode one can access the Web instantly without a computer. The beauty of i-mode is that it allows one to view the contents of any Internet website through a small, liquid-crystal monitor on the phone terminal itself. Even people who lack computer skills can read web pages as easily as they can operate the buttons of a telephone. The recently developed handset P505i provides more easy-to-use features and is equipped with both an inner and outer CCD camera. This handset allows easy access to incoming calls, cameras and e-mail through the convenient location of the function button. Its one-push, call-answer button situated on the clamshell hinge blinks when there is an incoming call, and when pushed, automatically opens the handset and connects the call. Today in Japan there are over 40,000 sites that can be viewed on Internet-capable mobile phones featuring simple graphics and quicker download times and navigation. VodafoneLive also originated J-sky in Japan, and J-sky was developed in reaction to i-mode’s explosive success. Even today, i-mode is driving VodafoneLive through competitive pressure. There was no service like i-mode in Europe or the United States at the time it was marketed in Japan. It was characterized as the ‘Japanese IT revolution’. By February 2000, only one year after the service was introduced, the total number of mobile phone users reached 55.5 million, 4.2 million of whom had i-mode. In addition, even after that the first year the number of i-mode users increased at a rate of roughly seven million per month. Presently, 40 million subscribers use i-mode in Japan every day. I-mode has become a part of the essential infrastructure of communication in Japan as well as the driving force behind the development of the global wireless Internet revolution. The trial period of FOMA services began at the end of May 2001 for a limited number of users, and full commercial service started in October (NTT DoCoMo 2003). FOMA is a proprietary version of W-CDMA technology that operates in 3G networks utilized by DoCoMo. This system uses a totally different technology from that of 2G services, and therefore all the facilities required installation of new equipment at a cost of about one trillion yen. But the system supports an advanced version of NTT DoCoMo’s popular i-mode mobile Internet service, making it easier to access e-mail and the Internet while travelling with a few extra features such as attaching sound files as well as still images via e-mail and JPEG in display. FOMA offers a packet communication service with downlink speeds of up to 384 kb/s, and a circuit switched service offering of 64 kb/s uploading/ downloading for large-volume data (Ralph and Bonner 2002). Japan got its first taste of third-generation cellular phone services with the launch of NTT DoCoMo’s FOMA services, but at that time the technology failed to attract people’s attention and its deployment was slow in the early stage compared to KDDI’s later deployment. One of the reasons
Next generation ICT deployment in Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
245
behind this slow start may be that the service area was then limited to the area around Tokyo. A second reason may be the initially high billing rate most likely due to its high installation costs. Another reason is that while DoCoMo made little progress in starting services, wireless LAN (Local Area Network) services had been spreading rapidly in Japan. Data transmission speed by the wireless LAN reaches a maximum of 11 mb/s. This speed was far faster than that of 3G services then offered by DoCoMo. Moreover, the wireless LAN systems fully utilize the Internet technology in order to reduce investment in facilities and thus reduce its monthly usage fee compared to DoCoMo’s 3G service. In spite of its attractive features, FOMA became a victim to the monumental success of i-mode and 2G. The major drawbacks of that success can be outlined as follows: • • • • •
FOMA’s coverage area was smaller than DoCoMo’s i-mode/2G service area. FOMA’s in-building coverage was not as good as the 2G service area. FOMA’s mobile handsets were larger in size than those for 2G service, and initially only a few models were available. FOMA’s mobile handset battery life was not as long as that of the 2G service. FOMA’s mobile handsets were costlier than the standard mobile phone at that time.
However, NTT DoCoMo decided to collaborate more closely with user companies for application development and review its handset strategy. Soon, Cumulative Subscribers 1M 950K 900K 850K 800K 750K 700K 650K 600K 550K 500K 450K 400K 350K 300K 250K 200K 150K 100K 50K 0 O N D
2001
DoCoMo FOMA* reached 1 million subs in Sept. 2003
J
F
M
A
M
J
J
A
S
O
N
D
2002
J
F
M
A
M
J
J
A
S
2003
Figure 11.2 Cumulative growth of FOMA subscribers Source: TCA (Information Technology and Telecommunications Association), JEITA (Japan Electronics and Information Technology Industries Association), calculated by the author, October 2004.
246
Mohammed Akhtar et al.
NTT DoCoMo’s FOMA will be equipped with the ‘FeliCa’ contact-less IC (Integrated Card) technology and its handset will be the ‘F900iC’ model. A further handset model, the ‘N900iL’, will be the first Linux-based model of NTT DoCoMo’s mobile phone. We anticipate that this kind of handset variety will make its deployment faster and reach a higher number of subscribers than ever before. Facing phenomenal competition from cdma2000–1x, the Japanese W-CDMA carriers, DoCoMo and Vodafone, have decided to reinvent themselves. DoCoMo introduced attractive service pricing and a new array of handsets in the first half of 2003. Hence, FOMA subscriber growth is finally picking up due to enhanced coverage and increased handset choice. Figure 11.2 shows FOMA’s cumulative subscriber numbers, and we can see that the efforts of DoCoMo have begun to pay off.
Comparison of Asian and European/US mobile communication markets The role of Asia, more specifically Japan and Korea, as the most mature wireless markets is moving towards taking the lead in the 3G market and in mobile data deployment. It is conceivable that 3G battle lines will be drawn in Asia. Such an achievement can be predicted from three factors – technology, end users and business strategies – as well as from the significant changes that have taken place in the Japanese and Korean communications industries over the last decade. With regard to technology, these countries are front-runners in many ways. Japan was the first to launch W-CDMA and thus the first country to actually experience real competition between the CDMA 1x and W-CDMA 3G communication systems. Compared to North American and European countries, Japan also is ahead in the market in terms of IPv6-related technology. At the same time, Korea was the first to launch CDMA 1x and DO. The Korean Ministry of Information and Communications launched a programme under the supervision of the Electronics and Telecommunication Research Institute (ETRI) with the aim of completing installation by the year 2000 and deploying 3G no later than 2001. Selection of technology was crucial for their success, as they opted for the cost effectiveness of CDMA over migration through the GSM/UMTS system. In the meantime, CDMA-1x, which is the 3G technology of choice for both Japanese and Korea operators, has demonstrated its capabilities and has been introduced in various parts of Asia and in the United States. As far as the end user is concerned, favourable market conditions, a mature mobile phone industry and highly technology sophisticated Asian consumers help to make Japan the future leader in the 3G markets. Japan’s mobile phone industry expects to ring up a banner year for sales of 3G mobile phones in 2004, and wireless operators around the world will be watching closely as they prepare for similar rollouts. Japan has always been regarded as an example that has encouraged investors, operators, equipment
Next generation ICT deployment in Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
247
vendors and others who pursue this market. Without a doubt, Japan has one of the most advanced and most competitive telecommunication markets in the world. Japan is unique in the number of competing wireless communications networks that offer a wide range of options to subscribers, and its mobile data market is much more developed compared to that of the United States or Europe, both of which still in their infancy. We may also presume that as long as customers are satisfied with the quality of services and range of features offered by the 2G systems, the service provider would have little reason to consider anything else (Schwarz Da Silva et al. 2003). Lastly, from the business perspective, no spectrum purchase was required in Japan. The United States supposedly lags in the 3G markets due to its lack of 3G spectrum as well as its fragmented approach to the market. In addition, the successful deployment of 2G systems in the United States translated into huge investments that needed to bring significant returns before 3G replacement plans were even considered. In other words, operators in the United States were rethinking the implementation of third-generation systems while Japan rolled them out. Ever since the members of the International Telecommunication Union (ITU) voted to allocate frequencies for a third-generation mobile communication system in 1992, Japan has adopted a somewhat different philosophy from Europe or the United States in terms of how a third-generation system should be developed and chosen. While in Europe the need for a fully fledged third-generation system was anticipated for 2005, the situation in Japan was such that the Japanese Ministry of Posts and Telecommunication (MPT) felt compelled to accelerate the standardization process to have a new, high-capacity system developed by the year 2000. The Japanese Telecommunication Ministry proposed allowing high-speed wireless Internet service providers to use part of the 25 gigahertz bandwidth without having to seek ministry approval. The decision by the European Technology Standard Institute (ETSI) to adopt DoCoMo’s technology as a major part of its third-generation standard gives Japanese manufacturers the opportunity to significantly expand their share of the worldwide cellular phone business. From the chart comparing 3G licence fees in major countries (Figure 11.3), we can see that Japan pays no licence fees and at the same time they have the highest Data ARPU (Average Revenue Per User) in the world (Figure 11.4). Other Asian countries are also paying much less in 3G licence fees than their European counterparts. The deteriorating revenues of mobile phone operators in the United States and Europe have affected the deployment of 3G services in those two areas. On the other hand, Japanese operators have understood that it is the customer experiences that matter, which has led them to adopt an effective, segment-focused marketing approach. In order to maintain their leadership, the operators in Japan and Korea devised win–win, profit-sharing models with content providers. We can thus say that Asia’s leadership in 3G is a unique combination of success on the technology, the end-user and the business front.
248
Mohammed Akhtar et al.
US$Mil 50,000
45,850
45,000 40,000 35,460 35,000 30,000 25,000 20,000 15,000 10,000 5,000
1
0
26
150 51
3,300 0
1,398
Au
st
ra Au lia s C tria an D ad en a m a Fi rk nl an Fr d a G nc e e H rma on n g y Ko n Ja g p M a N ala n et ys he ia r Si lan n N ga ds ew p Z or So ea e ut lan h Ko d Sw rea ed Ta en iw an U K
0
2,508 352 601 931 472 0 1,102
Figure 11.3 3G licence fees in major countries in 2002 Source: 3G News Room, April 2003.
Enormous changes have occurred in the Japanese communication industry in the past decade. The wireless communication industry has been transformed from a monopoly into a competitive industry. Nippon Telegraph and Telephone (NTT), a public corporation, monopolized the Japanese telecommunications industry until the mid-1980s. They started their analog cellular phone service in 1979. After that, several significant changes took place in Japan: • • •
The telecommunications industry was privatized in the mid-1980s. By the mid-1990s, competition between ‘keitai denwa’ (mobile phones), personal mobile phones (PHS) and pagers was heated. In the late 1990s, a successful NTT DoCoMo corporation was established.
The year 1985 is considered to be the turning point in the Japanese telecommunications industry due to the privatization of NTT and the acceptance of new common carriers in the market. Moreover, Motorola began to supply handsets in 1986 through a trade agreement between the United States and Japan (ATIP (Asian Technology Information Program) 2002). Prior to 1985, the entire communications industry in Japan had been controlled by the NTT family of companies.
Next generation ICT deployment in Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
249
Monthly ARPU Trend
US$ 70
64 60 54
63
62
62
59
57
50 40 30 20 10 0 4Q'01
1Q'02
2Q'02
Germany
3Q'02
Italy
4Q'02
Japan
1Q'03
Korea
UK
2Q'03
US
Share of Data in Total Revenue 25% 22% 21% 20% 19% 18%
19%
18%
17% 15%
10%
5%
0% 4Q'01
1Q'02 Germany
2Q'02 Italy
3Q'02 Japan
4Q'02 Korea
1Q'03 UK
2Q'03 US
Figure 11.4 Monthly ARPU trend (top) and total revenue data as % (bottom) in major countries Source: Merrill Lynch 2003.
During the period when NTT was in control of the communications industry, wireless communications was considered by NTT to be an unusual case and a minor business segment. Wireless communication at that time was used primarily for the defence system, broadcast media, electric companies,
250
Mohammed Akhtar et al.
local governments and railways, And the government enforced the regulations relating to spectrum allocation much more directly than it enforced antimonopolistic measures against corporations such as NTT. Wireless communication was included merely as one part of the hybrid wireless/wired, long-distance relay infrastructure before the mobile phones and pagers were introduced. Fifty per cent of the connections for long-distance calling consisted of wireless communications that relied on a microwave relay system. This ratio was flexible, and both forms of communications were competing with each other over transmission capacity and cost. New companies entered the fixed-line communications market by using long-distance relays. Long-distance calls were much more expensive than local calls, and long-distance customers had to pay much more than the actual cost of the calls. Because the long-distance market held the promise of a good profit, new companies began to enter. The competition created from the entrance of new operators caused a drop in long-distance fees. Nevertheless, even after the approval of new common carriers, NTT still owned most of the subscription lines fifteen years later. NTT DoCoMo had the largest share of the Japanese market. It had access to NTT’s well-funded research programmes: it was planning to operate services overseas; and it was being pressured by Japan’s MPT either to create or adopt a worldwide standard. By mid-1996, few participants in the worldwide mobile communication field outside of Japan believed that NTT DoCoMo had a chance of creating worldwide standards. Despite the fact that NTT DoCoMo had brought assets over from NTT, which gave it an advantage over new operators, as soon as new operators began to enter the market, NTT DoCoMo’s market shares began to drop sharply. The strategy of the new common carriers in the mobile communications industry in the wireless market was different. The service that they provided encompassed the entire country. A mobile communication business’s main objective is to first build a base of subscribers. Therefore, the providers had to create an independent, nationwide network for cellular services. Japanese firms in the telecommunications field had recognized for a long time that although NTT’s unique wire-line standards and DoCoMo’s unique mobile communication standards reduced the amount of foreign competition in the Japanese market, these unique standards at the same time made it difficult for them to penetrate into foreign markets. It was at that point when NTT DoCoMo began the process of attracting support from other Asian carriers. The Japanese mobile telecommunications market is reaching a stage of maturity while the demand for data communication is steadily expanding. In this regard, NTT DoCoMo plans to implement sustainable strategies that will pertain to multimedia, ubiquity and globalization. They are also determined to enhance their internal business structure by comprehensive cost cutting and strengthening of management. Korea is also working for the advancement of wireless communications along the same lines as Japan. NTT DoCoMo’s efforts with Asian carriers
Next generation ICT deployment in Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
251
began to pay off when Korea Mobile Phone and China Telecom, and shortly thereafter Ericsson and Nokia, agreed to support DoCoMo’s standard. The South Korean telephone company, SK Telecom, built the first thirdgeneration mobile telecommunication service in that country. It is the second largest network in the world after Japan’s NTT DoCoMo, and like DoCoMo it employs CDMA, which is the method expected to become the global standard for 3G wireless communications. The cell phone will always be on and will always be online because 3G mobile telecommunications service transmits data in the form of packets that are compatible with the Internet. This assumption is spurring Japan and Korea’s plans to develop a fourth-generation (4G) mobile telecommunications network as well as their collaboration on developing new technical standards later in 2004. The promise of 3G and beyond networks envisioned by this collaboration is that the service they intend to provide to mobile phone handsets will be without fault and will include video conferencing and movies on demand. The two countries are also planning to cooperate with each other in e-commerce policies and high-speed Internet services.
Japan and Asia’s leadership in next generation technologies As discussed above, Asia is a significant industrial cluster for 3G technologies and beyond. The question is whether the area can maintain this leading position. We believe that Asia will continue to be the epicentre for the development of technology and the commercialization of next-generation communication systems beyond 3G due to government and industry initiatives, its industry expertise, its market size and potential, and its market maturity. Asian governments are encouraging huge investments in research and development in order to gain a leading edge in securing intellectual property rights (IPR) and worldwide leadership in developing standards in 3G and beyond. Increasingly sophisticated and low-cost production facilities and cooperative development of standards between Japan, Korea and China will play an important role in Asia becoming that epicentre. Moreover, mobile operators will have the opportunity to gain the vital business and market experience of providing high-speed mobile data services. Japan was the first to announce specific plans to introduce wideband radio networks based on W-CDMA technology (3G Newsroom 2003). Japan and Korea are home to half of the top-ten mobile phone manufacturers and half of the top-ten semiconductor manufacturers in the world. They have strong operators with a long history of marketing and implementing data services, and understanding of user needs in advanced generation services and devices. Hence, initial, beyond-3G markets will more than likely be concentrated in Korea and Japan. NTT DoCoMo and Hewlett-Packard have announced that the two companies are jointly developing technologies for beyond 3G
252
Mohammed Akhtar et al.
Government& Industry Synergy R&D Innovation Market Potential Manufacturing Scale & Expertise Mobile Applications Expertise Government Direction
Figure 11.5 Asia’s iron triangle with complementary attributes Source: Pyramid Research, July 2004.
wireless communications. They have named this new technology platform ‘MOTO-Media’. The world’s two most advanced data markets, Japan and Korea, closely situated next to China with its massive wireless subscriber market make these countries together a most fertile ground for growth in next generation mobile communication. Figure 11.5 depicts the complementary attributes of this iron triangle of Asia. Japan, Korea and China together hold 30 per cent of the total, global mobile phone market, representing a formidable force in leading technology development beyond 3G.
Summary of Japan’s strength in next-generation mobile communication The strength of Japan in its next-generation mobile communication deployment can be summed up as follows: • • • •
Japan’s potential market in the Asia Pacific has approximately 420 million subscribers, making it the largest in world. Japan may develop products to support Chinese consumers. Asia contributed 50 per cent of the world’s GSM growth in 2002, with China alone providing 60 million. Users of mobile-based Internet services exceeded 65 million in Japan as of August 2003, whereas its i-mode users number more than 39 million. Vodafone Sha-mail picture messaging customers exceeded 10 million as of August 2003.
Next generation ICT deployment in Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
• • • • • • • • • • • •
253
KDDI CDMA-1x subscribers surpassed 10 million in September 2003. NTT DoCoMo FOMA UMTS/3G exceeded 1 million subscribers in September 2003. Both KDDI and NTT DoCoMo have introduced flat rate fee plans for specific data services. Potentially 70 per cent or more of the Japanese population will have some form of wired or wireless Internet connection by 2005. Asia Pacific and Europe will dominate other geographic regions during the 2000–10 decade. UMTS Forum predicts annual 3G revenues in Asia Pacific will reach US$118 billion by 2010. Asia pursues very strong research and development programmes for wireless technologies as depicted in Figure 11.5. 3G services in Asia have gradually reached the same degree of coverage as 2G. Improved handsets that generally have one or two built-in digital cameras and one or two colour screens are in use today in Japan. After facing initial problems, NTT DoCoMo is also showing strong 3G growth along with KDDI. Table 11.2 depicts a strong 3G phone market. All carriers cover approximately 95 per cent of the population at this point. Finally, the selection of technology, strong marketing, unique service and product advantages have been the major success factors for the three major Japanese competitors.
Table 11.2 Mobile shipment in Japan FY 2003
Cellular phone 1k units
April 2003 May June July August September October November December January 2004 February March Total (12 months from April 2003 to March 2004) Source: JEITA 051304.
PHS (for public service) FY’03/ FY’02 (%)
1k units
FY’03/ FY’02 (%)
4,133 3,891 4,744 5,104 3,988 4,007 4,006 4,158 4,675 2,935 3,475 4,725
140.3 123.7 119.6 139.0 122.4 119.7 130.6 112.4 120.1 77.8 97.6 99.6
190 111 134 149 160 121 106 58 36 48 20 40
148.1 118.6 112.8 204.4 117.3 164.3 117.3 73.5 52.6 73.3 27.2 19.1
49,841
115.7
1,173
102.2
254
Mohammed Akhtar et al.
Conclusion We have analysed the market of next-generation mobile communication technology with an emphasis on the Japanese market as a case study to understand the dynamics of success and failure in launching new-generation wireless technologies. Among the key lessons learnt is that being the first to enter the market is not necessarily a guarantee of success. Rather, it is more important to get the formula right before launching a new service that replaces an existing service. The pillars of success remain the same along a technology migration curve. User contents and services that are rich in experience and personalized handsets that are attractive and user-friendly at the right prices, a tariff plan that attracts customer loyalty and the extent of network coverage are all very important factors in successfully launching a new technology. The business model needs to include greater visibility for the user and to satisfy end-to-end requirements. Many other interface options and service specifications should be determined in order to ensure smooth roaming across operators, access technology and end users. In addition to having recognized the importance of these issues, all the operators in Japan also are searching for new ways to create more variety in products and services. The crucial factors for success in the future will be international roaming capabilities, device availability and network deployment costs. It is therefore necessary to encourage the research and extensive testing of next generation communication technologies in order that they can be rolled out rapidly once the market is ready. On average, it takes some time before momentum gathers for a new technology because the users need time to become comfortable with it and its applications. Furthermore, we also conclude that technology selection has a significant impact on the fundamental success of a wireless business strategy.
Bibliography 3G Newsroom (2003) Introduction Into 3G, accessed at www.3gnewsroom.com/ html/about_3g/intro_3g.shtml, April 2003. 3GPP (2001a) Physical Layer Aspects of UTRA High Speed Downlink Packet Access (Release 2000), 3rd Generation Partnership Project Technical Report 25.848 V0.6.0 (2000–05), Technical Specification Group Radio Access Network Working Group1 meeting, Las Vegas NV, 27 February to 2 March. 3GPP (2001b) ‘UTRA high speed downlink packet access (release 4)’, 3rd Generation Partnership Project Technical Report 25.950 V1.1.0 (2001–02), Technical Specification Group Radio Access Network Working Group 2 meeting, Sophia Antipolis, France, 19–23 February. 3GPP2 (2000) ‘High-speed data enhancement for cdma2000 1x—integrated data and voice, Stage 1 requirements’, 3rd Generation Partnership Project 2 Special Report 0026, October 17. 3GPP2 (2001a) ‘Joint physical layer proposal for 1xEV-DV’, 3rd Generation Partnership Project 2 Technical Specification Group-C meeting, Phoenix, 15 February.
Next generation ICT deployment in Japan 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
255
3GPP2 (2001b) ‘1xTREME 1xEV-DV framework components chandler’, 3rd Generation Partnership Project 2 Technical Specification Group-C meeting, Phoenix, 15 February. ATIP (Asian Technology Information Program) (2002) ‘Japan’s third generation mobile communication handsets’, accessed at www.atip.or.jp on 4 December 2002. Chan, H. A. and Lin, S. (2002) ‘Standards and deployment issues in wireless data networks’, IEEE Industrial Electronics Society Annual Conference, 4: 3407–12. Fasol, G. (2003) KDDI’s Success Story, accessed at www.eurotechnology.com/index. html in November 2003. Feng, M., Shen, S. C., Caruth, D. C. and Huang, J. J. (2004) ‘Device technologies for RF front-end circuits in next-generation wireless communication’, Proceedings of the IEEE, 92 (2), Feb: 354–75. Holman, H. and Toskala, A. (2000) W-CDMA for UMTS Radio Access for Third Generation Mobile Communications, Chichester: John Wiley & Sons. Holman, H. and Toskala, A. (2002) W-CDMA for UMTS, Chichester: John Wiley & Sons. Japan Telecom Monthly (1999) World Telecommunication Development Report, ITU press release/99–22 IMT-2000 radio interface specifications approved in ITU meeting in Helsinki, 5 November. JEITA press releases, 13 May 2004. Merrill Lynch, Pierce, Fenner & Smith Incorporated (MLP F & S) (2003), Global Wireless Matrix, September. Nikula, E., Toskala, A., Dahlman, E., Girard, L., and Klein, A. (1998) ‘FRAMES multiple access for UMTS and IMT-2000’, IEEE Personal Communications Magazine, April. NTT DoCoMo (2003) 3G/FOMA, accessed at www.nttdocomo.com/corebiz/network/ 3g/index.html in October 2003. Parry, R. (2002) ‘Overlooking 3G’, IEEE Potentials, October/November: 6–9. Qiu, R. C., Zhu, W. and Zhang, Y. (2002) ‘Third-generation and beyond (3.5G) wireless networks and its application’, IEEE Intl. Symposium on Circuits and Systems: 141–4. Raivio, Y. (2001) ‘4-G hype or reality’, 3G Mobile Communication Technologies, IEE Conference Publication, 477, March: 26–28. Ralph, D. and Bonner, I. (2002) ‘3G and beyond-the applications generation’, 3G Mobile Communication Technologies, IEE Conference Publication no. 489, 8–10 May. Ralph D. and Shephard, C. (2001) ‘Services via mobility portals’, BT Technol J, 19 (1), January: 88–99. Schwarz Da Silva, J., Arroyo-Fernándes, B., Barani, B., Pereira, J. and Ikonomou, D. (2003) Evolution Towards UMTS, accessed at www.tbm.tudelft.nl/webstaf/rudiw/ TB9612/UMTS/Umts0.htm, on 10 October 2003. Tanenbaum, A. S. (2003) Computer Networks, Englewood Cliffs NJ: Prentice-Hall. Watson, A. (2001) ‘The UMTS third generation market – structuring the service revenues opportunities’, 3G Mobile Communication Technologies, IEE Conference Publication, 477, 26–28 March.
12 Competitive advantage through co-evolution of technology and organization Janet Fulk
Research has shown that information technology investment in the United States, in the aggregate, is positively related to productivity (Brynjolfsson 2003). Yet, there is great variation across organizations in their ability to harness the productivity benefits of such investments: some organizations show marked increases in performance, while other organizations with the same level of information technology investment reap only meager benefits (Brynjolfsson 2003). A key question is: why do some companies disproportionately benefit from information technology investments? This chapter begins to answer this question by bringing to bear theory, research and case examples. In combination, these threads build the argument that information technology is part of a dynamic and complex system that evolves with changes in its subsystems. Thus, in order to secure organization-wide performance improvements from information technology investments, changes in technology must be matched with complementary changes in related systems. Information technology is not a medication that, once injected into a system, targets problems and solves them. Successful information technology investment is more like an ongoing whole-system makeover in which many parts of the system change and continuously adjust in relation to each other. The theory that underlies this perspective is organizational evolutionary theory (Campbell 1960, 1965; Aldrich 1999). The theory specifies a variety of mechanisms by which interrelated systems co-adapt and adjust to each other as they evolve and change. This theory has proven useful in understanding a wide variety of organizational issues (see, for example, the compendium in Baum and McKelvey 1999). The argument draws on empirical research and case examples from a variety of industries and contexts, including computing, medical supplies, consumer products distribution, insurance underwriting and law enforcement. This evidence also suggests that not only can co-evolution of systems with information technology investments increase organizational performance, but it also can contribute to sustainable competitive advantage. The complexity and embeddedness of the co-evolutionary changes in the organizational subsystems can serve as powerful barriers to imitation by competitors.
Co-evolution of technology and organization 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
257
The presentation is organized as follows. First, I review several basic tenets of evolutionary theory of organizations, highlighting mechanisms most relevant to information technology investment. Second, I discuss empirical research and case evidence bearing upon a co-evolutionary explanation of variation in outcomes of information technology investment. Finally, I conclude with suggestions for future research and theory.
Evolutionary theory of organizations An organization can be seen as a nested hierarchy of systems in simultaneous activity. Systems may be specialized by function (e.g. human resources versus financial administration), project (wireless versus wireline service provision), client (government versus consumer market) or other groupings that cluster the activities necessary to organizational functioning. Systems can be broad-ranging (e.g. research and development (R&D)) or highly specific (engineering for a particular product, which is itself nested in the broader R&D system). Systems may overlap such that some elements are associated with more than one system (e.g. human resources delivers instruction in the MBA instructional system and in the undergraduate educational system). Campbell (1994) points out that evolution in any system is a pathdependent exploration of possible variations from its current state in an attempt to continually improve fitness. Variation, whether accidental or intentional (created), is experienced as a perturbation in a system. For example, peer-to-peer file-sharing technology is an important upheaval for organizations that deliver music and films through traditional channels. Intentional variation occurs when firms proactively pursue a change in their current situations. These variations are created by the organizations themselves. Examples include managerial inventions such as implementation of balanced scorecard processes or strategic redirection of the firm such as entering or exiting a market. Perturbations may be of great magnitude, such as peerto-peer file-sharing technology or strategic redirection, but many variations are incremental rather than discontinuous (Tushman and Romanelli 1985). Fitness typically means likelihood of survival. In organizational studies the concept of fitness has been variously measured as longevity, death, performance or productivity (Carroll and Hannan 1989; Miller 1994). When a system explores a variation that can improve its fitness, it will select that variation as one to be incorporated within the system. Selection may be a conscious, creative process or may occur more incidentally. Once the variation has been selected, the system must absorb the variation and through the absorption process achieve a new system state, a process labeled retention. Retention may be intentional or may be a product of imposing environmental forces. The new Internet-based music download sites maintained by major record companies (e.g. Sony’s Connect music store) are examples of intentional retention in response to unintentional variation.
258
Janet Fulk
Nested systems in organizations are linked to each other and to systems in other organizations. Because of such coupling, the retention process for a selected variation entails adjustments in multiple other systems in addition to the retaining system. For example, delivery of music via the Internet impacts the nature of music production, contracts with artists, relations with brick and mortar retailers in the distribution system, and marketing activities. Thus, multiple systems co-evolve as they experience cycles of variation, selection and retention. Because organizational systems are nested, coupled and co-evolving, evolution in one system is likely to affect the fitness of other systems in the organization. As an organization evolves it may find that the improvement in fitness achieved in one of its systems may not be advantageous for one or more of its other systems—that is, evolution may increase the fitness of some systems while simultaneously decreasing the fitness of others. Those systems experiencing a rebounding decrease in fitness will then be motivated to seek improved fitness through their own cycles of variation, selection and retention. This process, in turn, may perturb yet other systems to which it is linked. Thus, evolution is a process of constant change of systems in relation to each other as each seeks improved fitness. When evolution in any system has only positive consequences (sustained or improved fitness) for the other systems in the organization, it is possible to achieve optimal overall organizational fitness. When there are at least some negative consequences for other systems attendant to evolution in one system, such global optimization is not possible. The “best” outcome for the organization in this situation is Nash equilibrium (Baum 1999). In the Nash equilibrium state, the mix of levels of fitness for each of the systems is the best possible given the full set of needs of the other systems. Nash equilibrium often is less than the maximal fitness possible for any one system in isolation. Evolutionary theorists argue that the greater the complexity of the coupling (and, hence, co-evolution) of the systems in an organization, the less likelihood there is of achieving Nash equilibrium as systems compete for their own selfish maximal fitness. Several examples from the realm of team sports illustrate the tension between self-maximization and global optimization. In basketball in the United States, players who excel in number of baskets scored can secure a strong personal reputation (and a lucrative salary). Self-interest can induce a player to take a shot on basket himself, even when a better play is to pass the ball to a team member in a better shooting position. Optimization of team performance may require that stars pass the ball to teammates. Indeed, when stars play their own game and do not work with the team, the performance of the whole team can suffer. Drucker (1995) points out that doubles tennis is an example of how players must constantly change their strategy and game play in relation to each other in order to maximize team success. For example, a team’s game plan will differ depending on whether each player’s strengths are at the baseline or in net play, and on how strong
Co-evolution of technology and organization 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
259
each player’s serve is. During the match, each player adjusts to changing game dynamics. For example, when one partner is pulled wide by an opponent’s shot, the other partner moves toward the center to cover the open territory. The team cannot win on the efforts of just one player. Brown and Eisenhardt (1998) offer an analogy in which an individual cannot win on his or her own efforts, but a team can win by adjusting their own moves to support a win by an individual player. In the Tour de France, only one cyclist wins. The winning cyclist, however, does not have the stamina to finish such a grueling race without the help of teammates. Brown and Eisenhardt (1998: 62) state: Although the top rider on the team is the one whom everyone knows, other team members are essential. They draft one another, a key strategy in which one cyclist leads while the others conserve energy by following in his slipstream . . . They muscle and block to protect teammates from passing moves by competitors. Each member of the team has a different role, such as hill specialist or blocker. Sometimes, argue Brown and Eisenhardt (1998), survival of the organization as a whole rests on outstanding performance by a single business, requiring other businesses to play a supporting role, adjusting as needed to bring success to a crucial business. The tension between individual and team interests also illustrates another concept of co-evolution: part–whole competition (e.g. Baum 1999). A subset of the competition between hierarchically nested organizational systems occurs between a system and its parts, as in the star versus team in basketball. A particular challenge of such competition is that lower level systems adapt and evolve more quickly than do higher level systems. Adaptation is swifter at lower levels because there are fewer potential variations to explore, select and retain. This more rapid adaptation at lower levels perturbs higher levels, exerting pressure for them to learn and co-adapt, and contributes to global suboptimization. Pressure cascades upward toward the whole organizational system from changes and co-adaptations taking place in lower level systems.
Information technology investment as variation, selection and retention Information technology implementations have been studied extensively in the management and information systems literature. Traditional diffusion of innovations approaches highlight how technology adoption spreads throughout a system from initial users to a broader network (Rogers 2003), while institutional theory (DiMaggio and Powell 1991) offers explanations for mimetic adoption in other organizations. Other models highlight the stages of development and implementation of technological systems (Nolan 1992). Extensions to these approaches highlight how users reconfigure new technologies to better fit their own needs (Rice and Rogers 1980).
260
Janet Fulk
Several examples serve to illustrate cycles of variation, selection and retention in information technology investments. The experience of Hanes Direct Store Distribution (DSD) with implementation of a new information system based on hand-held computers in their distribution network in the early 1980s illustrates how intentional variation can create other unplanned opportunities for additional variation. This case is described in detail in Morrison and Vitale (1988). Sales merchandisers who managed the physical distribution and merchandising of hosiery products for the retail chain were using a cumbersome and lengthy paper form for tracking hosiery in the system. New technology was implemented (intentional variation) by which sales merchandisers entered sales and tracking information into a hand-held computer. The software provided information on orders ready for pick-up, the most efficient routes, and the day’s calls. The device was also linked to the organization’s main data center providing email and planning programs. The unanticipated variation was new business opportunities, including offering the service to manufacturers with non-competing products in the same retail chain. Brynjolffson and Hitt (2000) describe the challenges to retention of selected intentional variations at a large medical products manufacturer. The organization implemented computer integrated manufacturing. A variety of complementary changes were made in other related systems: (1) compensation was flat-rate rather than quantity based; (2) operators took over their own quality control and generally had flexible rather than fixed, narrow job responsibilities; (3) work rules changed to focus on keeping up quality rather than keeping production flowing regardless of problems; (4) input materials were outsourced rather than made in-house; (5) concurrent engineering replaced functional hand-offs; and (6) several layers of management were eliminated. Despite such attention to co-evolution in other organizational systems, the investment did not pay off as much as hoped. A key retention problem was that workers were reluctant to change established work practices and learn new ones. When the system was implemented in a “greenfield” site with workers who did not already have established practices, there were significant productivity improvements. This example illustrates the importance of attention to evolution in all related systems—including, in this case, helping workers to learn new work practices—for an information technology investment to be effectively retained. Because organizational systems are coupled with systems in other organizations, evolution in one organization’s information technology system is linked to changes in coupled systems in other organizations. The classic example is American Hospital Supply (AHS), who pioneered direct order entry in their industry. In the late 1970s AHS customers were offered the option to enter orders directly into AHS’s computer system remotely, making the hospital purchasing agent’s job much easier if the agent ordered from AHS rather than competitors. AHS furthered their hold on customers by offering a free service to reorganize the hospital’s inventory stores to
Co-evolution of technology and organization 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
261
improve efficiency (and better align with AHS’s system, furthering AHS’s lock-in of customers; Vitale, 1985). Intentional variation followed when AHS developed a ‘reverse’ system for linking directly to its own suppliers. Nature of coupling among systems To understand dynamics in related systems, it is important to understand the nature of the coupling among systems. Evolutionary theory describes couplings as either symbiotic or commensalist. Symbiotic relations occur among systems that draw on different resource bases but benefit from each other’s activities. A classic biological example is the relationship between giraffes and tickbirds. Tickbirds feed on insects, whereas giraffes feed on leaves from acacia and mimosa trees. Tickbirds ride on giraffes where tickbirds eat the ticks and other insects present on the skin of giraffes. Tickbirds get food, and giraffes get rid of pesky biting insects. An organizational example from Aldrich (1999) is the relationship between a venture capitalist and a high-technology firm, both of which benefit from the other’s activities. Commensalist relations occur between two systems that draw on the same resource base, and can take three forms: competition, mutualism or neutrality. Competitive relations can be balanced such that changes in either system negatively impact the other, or imbalanced such that only one system can negatively impact the other. Carrying through the biological theme, an example of a balanced competitive relationship occurs between animals that hunt the same prey, such as leopards and cheetahs. Growth in the leopard population can negatively affect the availability of game for cheetahs, and vice versa. An analogous organization situation is between competitors in a mature industry from which neither can exit. The mutual effects of escalating competitive behavior can potentially destroy the industry structure for all competitors. An imbalanced relationship is illustrated by that of the lion and the cheetah, which also hunt the same prey. If the cheetah population grows to the point of putting pressure on food supply for lions, lions can kill cheetahs. Cheetahs, however, cannot kill lions. Such is the situation faced by new entrants to industries where a dominant firm holds proprietary technology and a popular brand, and where customers face heavy switching costs. Mutualistic relations, in which neither system is harmed by its relationship to the other system, can also be balanced or imbalanced. In the balanced case, each system benefits from the other’s activities; in the imbalanced scenario, only one of the systems benefits from the other. A balanced relationship can be seen for the two organisms that make up the lichen plant: fungus and alga. The alga provides food and oxygen for the fungus, while the fungus provides protection, water and salts for the alga. An analogous situation is the functional design of an organization, where each function contributes a unique portion of the value added by a firm. An imbalanced relationship is
262
Janet Fulk
illustrated by the bush sponge and Antarctic scallop. The sponge attaches to the swimming scallop to create water flow for the sponge’s filter feeding. The scallop gets no particular benefit from its free-riding sponge, but is not harmed by it. A business example is offered by Aldrich (1999). In the mid-twentieth century sales of wine imports in the United States increased, generating new demand for wine, which benefited the domestic industry. Sales of domestic wines, however, did not influence sales of imports. Neutral coupling occurs when two systems do not directly influence each other, even though they may influence other systems. Blue shark populations do not directly influence white shark populations, although as predators both of them influence sea life in their food chain. In financial services, growth of commercial banks has been shown to be unrelated to growth in savings banks (Aldrich 1999). Table 12.1 summarizes the types of coupling. It is important to note that relations between systems are often multiplex; two systems may be coupled differently on different dimensions. For example, the elephant may have a neutral relationship with a giraffe in that they do not hunt each other, they feed on different foods, and population growth in one does not typically influence the other. They also have an imbalanced competitive relationship in that elephants often topple trees. If elephants destroy too many tall trees, there will not be enough high tree branches from which the giraffe can graze. This example also highlights the dynamics of population co-evolution. The growth and/or decline in the giraffe and elephant populations may not influence each other in the presence of sufficient water, but if conditions change and water becomes scarce, any population growth can influence the survival of other populations that depend on those scarce water sources. Industry evolution offers an organizational example. Two competitors in a very new industry may have a balanced competitive relationship in that each benefits from the industry growth and customer education that the other provides, and each is harmed by the other’s actions to the extent that they confuse potential customers or disappoint sources of capital. However, as the industry matures and no longer faces the hazards of newness, those same competitors may enter into an unbalanced relationship where growth in one company negatively impacts the market share of the other. Precompetitive alliances offer an additional example of the temporal dimension. Firms that cooperate in research and development that benefit both (e.g. Intel and Motorola in the SEMATECH alliance; Browning, Beyer and Shetler 1995) must subsequently compete against each other in product markets that employ the lessons from the research and development partnership (e.g. computer chips). To understand how information technology investments can change and potentially improve organizations, one can identity the relevant systems and nature of the couplings between them. Adjustments can be made in related systems as information technology is implemented and reconfigured. Of course, changes in related systems may also have implications for the fitness
Co-evolution of technology and organization 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
263
Table 12.1 Types of coupling in co-evolving systems Coupling
Description
Example
1 Symbiotic type
Relations among systems Venture capitalists and that draw on different high-technology firms resource bases but benefit from each other’s activities
2 Commensalist type
Relations among systems that draw on the same resource base
2a Competitivebalanced
Changes in either system Price wars that reduce the negatively impact the other profitability of all participants
2b Competitiveimbalanced
Only one system can negatively impact the other
Major airlines with their own reservation systems in relation to travel agents that book airline reservations
2c Mutualisticbalanced
Each system benefits from the other’s activities
Different functional groupings contribute to overall firm success
2d Mutualisticunbalanced
Only one of the systems Wine imports improved sales benefits from the other, of domestic wines through but the other is not harmed customer education and market expansion, but not vice versa
2e Neutral
Two systems do not directly influence each other
Growth of commercial banks is unrelated to growth of savings banks
Source: The table and labels are modified from Aldrich 1999: 302.
of the information technology systems; organizations need to be prepared to modify the information technology system itself as needed in response to changes in coupled systems. The following cases serve as examples. Competition and cannibalism: the instant camera business The battle between Polaroid and Kodak in the instant camera market illustrates how technological advancements can lead to a cascade of changes that modify the relationships within organizations as well as between them (Porter 1983; Berg and Merry 1984). Polaroid was the sole manufacturer of instant cameras in the mid-1970s. Kodak decided to enter the market, coupling with Polaroid in a balanced competitive relationship. These two manufacturers sold a variety of models, each model designed for a particular niche in the market. Over time, technology investments combined with the accumulation of production experience at Polaroid made it possible to sell high-end cameras at a lower price (to compete with Kodak’s new entries)
264
Janet Fulk
but still maintain profit margins on those models. The competitive coupling between organizations created downward pressure on prices on both brands. High-end cameras became available at mid-range prices, linking the two types of camera into a new and unbalanced competitive relationship for the same set of customers within each company. When prices on midrange cameras necessarily dropped in response, these cameras changed to an unbalanced competitive relationship with low-end cameras. As an inexperienced player in the instant camera portion of the camera market, Kodak was unable to imitate the technological capabilities that Polaroid had invested in, and thus could not keep costs as low. The cascade of price reductions between the two organizations eventually pushed Kodak out of the instant camera business altogether, after taking tremendous losses (Porter 1983). Had Kodak approached the instant camera business and Polaroid as a potentially mutualistic relationship, perhaps different and more favorable industry dynamics could have resulted (Porter 1983). From balanced mutualistic to balanced competitive: Comprehensive Loss Underwriting Exchange (CLUE) The insurance industry has employed information technology to support a joint database on claims and insureds across companies (pre-competitive alliance; Monge et al. 1998). Insurance companies send claims information to the CLUE database in much the same way as creditors send information to major credit bureaus. Any insurer can access CLUE information to obtain a five-year history on any person or property that represents possible new business or risk (http://www.choicepoint.com/industry/insurance/pc_ins_4. html). Thus, insurance companies cooperate in a set of balanced mutualistic relationships relative to writing new business. All benefit from accurate and accessible information related to risk assessment and risk reduction in the industry. At the same time, these same insurance companies compete in the markets for insurance products, in competitive relationships. These multiplex relationships serve the interests of each of the insurers. Building mutualism: El Centro El Centro (a pseudonym) is a technology-driven information-sharing system among state, federal, and local narcotics enforcement officers in one region of the United States (Monge et al. 2000). In this region, there was a history of conflict among these organizations, which were pitted in competitive relationships as a result of specific features of asset forfeiture laws. Such laws gave to the jurisdiction that made a narcotics arrest the drug trafficker’s assets that had been used in the illegal activity. Those jurisdictions that helped other jurisdictions in an arrest received nothing, while the jurisdiction that made the arrest took all. The result was hoarding of information and in some rare cases officers from one jurisdiction unknowingly firing
Co-evolution of technology and organization 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
265
weapons upon officers from other jurisdictions during undercover operations. A significant information technology investment was made to create a “deconfliction” database of planned operations so that each jurisdiction would be notified in advance of any possible conflicts. The deconfliction task was housed in a federation that represented all of the jurisdictions. Importantly, the technology investment was supported by evolution in other systems. The chief executive officers of the jurisdictions signed a memorandum of agreement (MOA) indicating that each would require their officers to submit information on planned operations to the database. This MOA was essential to secure the cooperation of undercover officers who, due to the risks involved, are highly secretive, even from their own bosses. The Chief Executive Officers (CEOs) also reached an agreement that any forfeited assets became the property of the federation that managed the deconfliction system, to be subdivided among all jurisdictions having any part in an operation that eventually led to asset forfeiture. The technology investment was substantial and important, but by itself was not able to achieve the goal of deconfliction. This co-evolution of other systems through the MOA and the agreements on sharing assets were critical to the success of the technological intervention and the transformation of relationships among jurisdictions to become more mutualistic. Systemwide co-evolution: Dell Computer In 1996 sales of personal computers had become a commodity business. Dell Computer, which had originally introduced the concept of build-toorder in the personal computer industry, revamped its organization from a mail-order business to an Internet business. Dell deployed Internet-based information technology to permit direct sales to customers, rather than selling through distributors or brick-and-mortar stores. The goal was to move at least 50 percent of sales to the Internet. This move altered the relationship between Dell and its distributors and retailers from a balanced mutualistic relationship to an imbalanced competitive relationship. As described by Rangan and Bell (1998) the new “Dell Direct” model permitted both corporate and home customers to purchase and track orders online, obtain technical support online, and to download software. Large corporate customers were offered customized web pages (“premier pages”) for technical support, access to their account executives, and for some customers, the ability to order online with special pricing on standard configurations. Large customers were treated as partners; Dell technical staff consulted them extensively for input on future changes to computer models. At the same time these customers were able to use premier pages as a management control tool for enforcing product standards and reducing workloads. Dell was able to offer these customers online diagnostics and the ability to load customer specific software at the factory. The new model
266
Janet Fulk
of Dell Direct, at least for large corporate customers, was a balanced mutualistic one. For home users, Dell benefited from the efficiencies of no longer having to invest in education and substantial technical support, since there were many fewer first-time buyers as personal computers became commoditized. Dell’s Direct model modified not only the relationships with customers, but also with suppliers. In combination, supply- and demand-side changes generated what Chief Executive Officer Michael Dell labeled “virtual integration” (Magretta 1998). Virtual integration involves a shorter distance between customer and supplier, without intermediaries, and a blurring of the boundaries between customer and supplier. For suppliers, Dell evolved a number of significant relational changes. Dell reduced the number of suppliers and made long-term commitments to them, invited supplier technical staff onsite at Dell, shared design databases with “partnersuppliers”, and delegated the task of combining components into a single shipment to the delivery companies (e.g. United Parcel Service would pick up monitors from Sony and computers from Dell and match them in transit to the customer; Rangan and Bell 1998). The evolving mutualistic relationships with suppliers meant better information on which to forecast demand, less inventory with associated costs and risks, and the fastest delivery time in the industry. Within Dell, all business units needed to adjust to these changes. Sales representatives needed to learn how to work online and telephone queues, rather than the previous onsite visits based on massive catalog distribution. New metrics were needed on which to forecast demand and measure service effectiveness. Online resources needed to be resident in all business units in addition to the major thrust of the Dell Online group that orchestrated the changes. The balanced mutualistic relationships among the business units and with the Dell Online group meant that all units needed to change and adjust in relation to each other’s needs. Putting online resources in the business units permitted Dell Online to adjust strategy better to business unit needs, and permitted business units to better understand the changes that the Dell Direct program would entail locally. The complex set of changes in strategy, structure, culture and boundaries that accompanied the technological innovation created a huge barrier to imitation for a period of years. Because evolution and change permeated all aspects of the value system, competitors faced significant challenges in identifying the exact configuration of changes that would need to be imitated. In addition, some of these changes were more monumental for competitors who had traditionally relied totally on distributors, since it required adopting the whole model, which eliminated the roles of distributors. Dell’s preemption of the corporate markets and complex integration of major customers with Dell systems substantially increased the switching costs for customers— and the challenge for competitors targeting those relatively locked-in customers.
Co-evolution of technology and organization 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
267
Technology co-evolution It is also important for information technology implementations to be sufficiently flexible to co-adapt to changes in other systems. As other systems change in order to secure or reacquire their best fitness, they are likely to have impacts that can decrease the fitness of the technology system itself. As Rice and Rogers (1980) note, reinvention is a key to success of any technology implementation. Dell’s expansion to the Chinese market (intentional variation) is an example. Dell in China: technology co-evolution The Dell Direct system relies on a number of features that posed a challenge to the China market upon its entry in the mid-1990s (Ng, Lovelock and Farhoomand 2000). Few people used the Internet or had credit cards, transportation networks were unreliable, and potential customers preferred to see and use a computer in a showroom before buying, since the cost was a substantial investment (equivalent of two months salary; Roderick 2002). Dell adjusted by having delivery persons carry wireless debit card machines that permitted customers to pay upon delivery—a significant change in its technology strategy. El Centro: failure to co-evolve technology El Centro’s case management system is one example of a failure to adjust technology implementation in a timely manner. In addition to the deconfliction system, El Centro spent millions of US dollars to design and implement a case management system that permitted officers from different jurisdictions to pool information on ongoing investigations. This system was designed to permit information sharing about persons under investigation, suspect properties, findings from interrogations and undercover contacts, and other information not generally available to anyone other than the individual narcotics agent. The technology was a “communal system” (Fulk et al., 1996) that had the feature of making such contributed information available to any officer of any jurisdiction at any time. In contrast to the deconfliction system, the case management system was not successful. Narcotics officers were unwilling to contribute their sensitive information to a common repository. Unlike deconfliction information, case information was private, known only to the individual officer. The chiefs could not enforce compliance because they could not know whether an officer was withholding information. By the time El Centro’s management realized the need to modify the system to permit exclusive sharing of information between known pairs or only persons trusted by an individual officer, the reputation of the case management system was established and no officers contributed truly valuable information. The technology that worked for deconfliction did not
268
Janet Fulk
work for case management. Failure to adapt the technology to the very real needs of officers who put their lives on the line in their undercover work undermined the acceptance of the new technology. Co-adaptation between technology and officer procedures was needed.
Conclusion I started this chapter with the question: At the same level of IT investment, why do some organizations have only a fraction of the productivity of their competitors? The answers from evolutionary theory, research and case examples are that successful organizations understand principles of coevolution and implement processes that recognize the need for interdependent systems to co-adapt with changes in information technology. An important future direction is to further articulate the theoretical and practical aspects of co-adaptation of the technology itself to other systems. Adaptive structuration theory (Poole and DeSanctis 1990) offers one valuable theoretical approach to supplement the important lessons from theories of organizational evolution. Insight from theory offers an important tool for helping to evolve our understanding of the complex ways in which information technology investment is implicated in firm productivity.
Bibliography Aldrich, H. (1999) Organizations Evolving, Thousand Oaks CA: Sage. Baum, J. A. C. (1999) ‘Whole-part coevolutionary competition in organizations’, in Baum, J. A. C. and McKelvey, B. (eds) Variations in Organization Science: in honor of Donald T. Campbell, Newbury Park CA: Sage, pp. 113–35. Baum, J. A. C. and McKelvey, B. (1999) Variations in Organization Science: in honor of Donald T. Campbell, Newbury Park, CA: Sage. Berg, N. A. and Merry, G. W. (1984) Polaroid–Kodak, Boston MA: Harvard Business School Press. Brown, S. and Eisenhardt, K. (1998) Competing on the Edge: strategy as structured chaos, Boston MA: Harvard Business School Press. Browning, L. D., Beyer, J. M. and Shetler, J. C. (1995) ‘Building cooperation in a competitive industry: SEMATECH and the semiconductor industry’, Academy of Management Journal, 38 (1): 113–51. Brynjolfsson, E. (2003) ‘The IT productivity gap’, Optimize, issue 21, accessed at www.optimizemag.com/showArticle.jhtml;jsessionid=DAH4O4FCXEXTUQSNDB GCKH0CJUMEKJVN?articleID=17700941> on 1 August 2003. Brynjolfsson, E. and Hitt, L. (2000) ‘Beyond computation: information technology, organizational transformation and business performance’, Journal of Economic Perspectives, 14 (4), 23–48. Campbell, D. T. (1960) ‘Blind variation and selective retention in creative thought as in other knowledge processes’, Psychological Review, 67, 380–400. Campbell, D. T. (1965) ‘Variation and selective retention in socio-cultural evolution’, in Barringer, H. R., Blanksten, G.I. and Mack, R. (eds) Social Change in Developing Areas: a reinterpretation of evolutionary theory, Cambridge MA: Schenkman, pp. 19–49.
Co-evolution of technology and organization 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
269
Campbell, D. T. (1994) ‘How individual and face-to-face-group selection undermine firm selection in organizational evolution’, in Baum, J. A. C. and Singh, J. V. (eds) Evolutionary Dynamics of Organizations, New York: Oxford, pp. 23–8. Carroll, G. R. and Hannan, M. T. (1989) ‘Density dependence in the evolution of newspaper organizations’, American Sociological Review, 54: 524–41. DiMaggio, P. and Powell, W. W. (1991) The New Institutionalism In Organizational Analysis, Chicago IL: University of Chicago. Drucker, P. (1995) Managing in a Time of Great Change, New York: Truman Talley Books. Fulk, J., Flanagin, A. J., Kalman, M. E. Mange, P. R. and Ryan, T. (1996) ‘Connective and commercial public goods in interactive communication systems’, Communication Theory, 6, 60–87. Magretta, J. (1998) ‘The power of virtual integration: an interview with Dell Computer’s Michael Dell’, Harvard Business Review, March–April, 73–84. Miller, D. (1994) ‘Selection processes inside organizations: the self-reinforcing consequences of success’, in Baum, J. A. C. and McKelvey, B. (eds) Variations in Organization Science, Newbury Park CA: Sage, pp. 93–109. Monge, P., Fulk, J., Kalman, M., Flanagin, A., Parnassa, C. and Rumsey, S. (1998) ‘Production of collective action in alliance-based interorganizational communication and information systems’, Organization Science, 9: 411–33. Monge, P., Fulk, J., Parnassa, C., Flanagin, A., Rumsey, S., and Kalman, M. (2000) ‘Cooperative interagency approaches to the illegal drug problem’, International Journal of Police Science and Management, 2 (3): 229–41. Morrison, J. E. P. and Vitale, M. R. (1988) Hanes DSD, Boston MA: Harvard Business School Press. Ng, P., Lovelock, P. and Farhoomand, A. (2000) Dell: selling directly, globally, Boston MA: Harvard Business School Press. Nolan, R. (1992) ‘The stages theory: a framework for IT adoption and organizational learning’, in Mechling, J. and Rosenberg, C. (eds) America’s Information Technology Agenda, Cambridge MA: John F. Kennedy School of Government. Poole, M. S. and Descanctis, G. (1990) ‘Understanding the use of group decision support systems: the theory of adaptive structoration’, in Fulk, J. and Steinfield, C. (eds), Organizations and Communication Technology, Newbury Park, CA: Sage, pp. 173–91. Porter, M. E. (1983) Polaroid Kodak B(1) through B(11), Boston MA: Harvard Business School Press. Rangan, V. K. and Bell, M. (1998) Dell online, Boston MA: Harvard Business School Press. Rice, R. E. and Rogers, E. M. (1980) ‘Reinvention in the innovation process’, Knowledge: Creation, Diffusion, Utilization, 1 (4): 499–514. Roderick, D. (2002) ‘For whom the Dell tolls: can legend computer save China from the world’s largest boxmaker?’, Time Asia, 15 (11), 8 April, accessed at www. time.com/time/asia/magazine/article/0,13673,501020325–218368,00.html on 8 April 2002. Rogers, E.M. (2003) Diffusion of Innovations, 5th edition, New York: Free Press. Tushman, M. and Romanelli, E. (1985) ‘Organizational evolution: a metamorphosis model of convergence and reorientation’, in Cummings, L. L. and Staw, B. M. (eds) Research in Organizational Behavior, 7: 171–2222. Vitale, M. (1985) American Hospital Supply Corp: the ASAP system (A), Cambridge MA: Harvard Business School Press.
13 Shaping organizational technology ICT as a learning process Georg Schreyögg, Sami Khiari and Leo Schmidt
Introduction Technology and technological constraints have been long-standing topics in both management theory and in practice. The capability of an organization to effectively handle technological demands is considered one of the most important factors affecting a firm’s competitive position. In recent technological studies, information and communication technology (ICT) and its specific impact in particular have come to the forefront (Fulk and DeSanctis 1995). Most studies of technologies in an organizational context have focused on the impact of technology and technological change on organizational structure and design (for an overview, see Scott 1990; Schreyögg 1995; Roberts and Grabowski 1996). Although the cumulative results of research studies have remained chronically inconsistent, it is widely agreed that technological conditions shape organizational forms or, to put it differently, that organizational design has to match technological demands. This research stream carries clear connotations of technological necessity and determinacy (Noble 1984). Some authors even consider technology an imperative that requires congruent organizational structures and management systems (Woodward 1965). Whatever the propositions are in detail, the bulk of this research has taken technology as a given external force constituting objective constraints for an organization and its design. Technology is conceived as a clear-cut material body that enters an organization as a finished self-contained entity. During the past decade this line of reasoning and conceptualizing technology has been called into question. Many authors have urged a major re-think and various new views have been developed (e.g. Barley 1986; Weick 1990; Orlikowski 1992; Fulk 1993). These studies emphasize the artificial character of technology developed and embedded in a specific historical context. To put it another way, technology is considered ‘hardened history, frozen fragments of human and social endeavour’ (Noble 1984: xi). Weick (1990), for instance, stresses the open and equivocal character of technology, which brings about stochastic, continuous and abstract events.
Shaping organizational technology 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
271
Instead of a clear-cut imperative, in this view technology needs interpretation, imagination and making sense of, thereby construing a workable technological solution for the organization. As a result, technology is conceived as a continuous process, which is basically interactive in nature. The outcome of a specific social process imprints future choices and behaviour, and by implication the process becomes self-referential (Luhmann 1984). The major conclusion of this research was that technology should no longer be conceived as an exogenous determinant. It is not deterministic in its application nor in its development. Technology is a socially construed artefact that leaves more or less broad scope for its usage. The application of technology is much more open than initially thought and actually comes close to an interactive process. Another result of this new thinking was dropping conceptions of technology that have materiality as a primary characteristic. As early as the late 1960s Charles Perrow (1967) pioneered the re-defining of technology in a radical way. In his knowledge-oriented view, technology consists of a ‘body of ideas’ – thereby pointing to rules and practices as the actual core of technology as opposed to material technology.1 This stream of thought has been strongly reinforced by the insight that the software has increasingly become the major component of technology, and, most importantly, the dramatic expansion of information and communication technology has changed the prevailing notion of technology anyway. By now we are used to viewing technology in a much more integrated way, and nobody seriously insists on treating technology as an exclusive hardware feature. The bulk of the constructionist literature has focused on the societal or market processes in defining and construing technologies, including the assumptions underlying technology development and implementation. Only recently has this constructionist perspective been applied to an intraorganizational perspective as well.
Organizational technology The constructionist view of technology has initiated a considerable amount of technology studies on the organizational level, addressing the whole organization and its subsystems (divisions, strategic business units, departments, etc.). The major focus is on the interactive nature of technology and the specific organizational enactment of a technology (Fulk and Steinfeld 1990; Scarbrough and Corbett 1992; DeSanctis and Poole 1994; Orlikowski et al. 1995; Yates et al. 1999; etc.). In this perspective organizations are no longer simple recipients of technology and subject to its causal effects but rather agents with individual scope and intentions (Ortmann et al. 1990). Following this line of reasoning technology eventually becomes ‘organizational technology’ – basically being imprinted by one specific organization. The traditional boundaries between technology development, implementation and usage are removed.
272
Georg Schreyögg et al.
This new perspective of technology not only broadens the scope for remodelling the relationship between management system and technological features, but also brings along a new strategic dimension (Burgelman et al. 2000). The traditional cause–effect–conception implied a standardized reaction pattern to a specific technology – at least in the medium term. All those organizations unable to find the appropriate prescribed design that answers to the demands of a specific technology are assumed to fail (because of inefficiency reasons) and subsequently become eliminated by market forces (Schreyögg 1995). As a result, organizational responses and usage of technology increasingly match the standard pattern. By implication, according to this view, firms cannot differ significantly along the technological dimension. In contrast to this external pressure model the new perspective stresses the internal processes of an organization and the scope to shape technology in a specific way. A CEO recently hit it on the nail by describing his firm’s technology approach: ‘We don’t much worry if the competition also has access to the technology, we think we can be smarter in how we use it’ (Hopper 1990: 120). From this perspective, technology amounts to a design variable that allows for strategic differentiation. Organizations can develop unique and superior capabilities in shaping and using technologies that may result in competitive advantage and enable the firm to generate greater than normal rents. Standard technologies are likely to become ‘organizational technologies’. This conception of organizational technology obviously has a clear affinity with the resource-based view (RBV) of competitive advantage (Barney 1991; Grant 1991; Wernerfelt 1989). The interactive construction of organizational technology adds up to a source of heterogeneity (Penrose 1959) and offers the chance to develop a ‘strategic resource’ that uniquely contributes to a firm’s competitive advantage and sustained profitability (Barney 1991). One of the primary preconditions to make a resource a strategic resource is its inimitability. In the RBV framework, major conditions for resource inimitability are causal ambiguity and uncertainty (Mosakowski 1997). Ambiguity is assumed to arise from social complexity, which means the way in which organizations combine resources and the underlying multifaceted interplay of human interaction and routines (Barney 1991). The bulk of this interaction is self-organized and emergent in nature, not deliberately built and therefore difficult to understand (‘complex’). This evolving unplanned nature prevents competitors from imitating it. In most cases even the focal firm does not fully understand the inherent logic of the resource building process. The new interactive conception of technology stresses this evolving nature of technology. In terms of RBV it is a system-level intangible resource, i.e. resources conceived as outcomes from internal interaction (Colbert 2004). Such intangible organizational resources are, however, not only complex in character, they must also be conceived as being in flux. The organizational
Shaping organizational technology 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
273
construction and definition process has no clear end, as conditions are changing and new interpretations emerge, etc. As a result, the technological process in organizations potentially offers scope for redefinition and change. Organizations always have the potential to change their modes of use and in this way initiate new construction processes (Orlikowski 2000; Boudreau and Robey 2005). Quite obviously, technological processes defined this way have a natural affinity to organizational learning. The effectiveness of technology-building processes is at least partly dependent on the system’s ability to learn and on the availability of efficient learning mechanisms. All of these conceptual ideas and frameworks simultaneously raise a number of interesting empirical questions: How are organizational technologies actually built? Are there typical interaction patterns and which ones are superior? How do technology-construction processes develop over time? What is the timing of successful technological initiatives? Do typical stages emerge? Throughout the studies investigating these questions three features stand out (for an overview see, for example, Orlikowski 2000): 1 2 3
temporal patterns of technology enactment (use of opportunity); learning processes in technology building; users as inventors.
In the following sections we report on two empirical studies that focus on these three features of organizational technology.
‘Windows of opportunity’ In a much quoted study Tyre and Orlikowski (1994) addressed issues of temporal patterns of technological enactment in organizations. Their major finding was that organizations actually re-work technology during and after installation. However, the firms observed did not do it on a continuous basis as expected. Instead, the researchers found a strict temporal pattern with only a few modification periods. The bulk of organizational definition and modification activities occurred right after the start. The initial phase opens a short ‘window of opportunity’ as Tyre and Orlikowski put it. Modification activities decreased as technologies became ‘more thoroughly embedded and routinized in the user environment’ (Tyre and Orlikowski 1994: 111). Later users only occasionally re-examined and redefined technology. The method of data collection was ex post. It mainly included interviews and questionnaires as well as the company’s annual report, plant and project documents. The researchers studied technical projects as part of the production process of three US and European organizations: a European precision metal component manufacturer; a multinational software consulting firm; and a research university in the US.
274
Georg Schreyögg et al.
Tyre and Orlikowski (1994: 105) defined adaptive activities as actions that were targeted at altering the installed technology, its operating procedures or its context of use. Respondents to their questionnaires and interviews were asked to report the number and the level of adaptive activities on a timeline starting with the installation of new technology. They calculated both the total level of adaptive activity reported for a given project and the percentage of the total completed during the first, second and up to the thirty-fifth month following installation. Results by month were then averaged across the (32) projects. This in-depth-analysis was only conducted at the European firm. As can easily be seen in Figure 13.1 nearly 30 per cent of all modifications implemented in all the projects at the European research site were conducted during the first month after installation. On average, 54 per cent of all modification activity was completed after 2.8 months. The first and largest window of opportunity directly followed the implementation of a new technology. The interviews revealed that subsequent changes to the initial burst didn’t occur until either disruptive events or new discoveries by users triggered subsequent adaptive activities (Tyre and Orlikowski 1994: 112). The study showed that users perceived new technology as being shapeable only during the implementation stage and shortly after. Once developed, the solutions seemed to become organizationally locked in – at least from the perception of the organizational members.
Percent of all adaptive activity per month
30
25
20
15
10
5
0 0
2
4
6
8
10
12 14 16
18 20 22
24 26
Time in months from installation
Figure 13.1 Temporal modification patterns Source: Tyre and Orlikowski 1994: 106.
28 30 32
34
Shaping organizational technology 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
275
The authors (Tyre and Orlikowski 1994: 106) identified four organizational forces that may explain this periodic shaping of the adaptive activities and the early closure of the windows of opportunity: 1 2 3 4
the the the the
pressure of production; constraining effect of habitual patterns of use; adjustment of expectations based on experience; and erosion of team membership and enthusiasm over time.
The results of this study have been surprising as they stand in contrast to many other studies, mentioned above, which stress the continuous character of technology enactment. The question arises as to whether the way these users dealt with the technological opportunities and the resulting periodic nature of modification activities found in this study can be considered as representing a typical pattern. Are beginnings actually so important in terms of enactment? The authors claim that their findings are remarkably consistent across different organizational and technological settings. In our view there are two major problems with the Tyre and Orlikowski study that call the generalizability of its results into question. One problem is the retrospective character of the data. Respondents were asked to report retrospectively on modifying activities.2 The data and the resulting pattern, therefore, very much depends on the perception of the participants and their ex post interpretation (rationalization?) of the events. Adjustments right at the beginning may make the most sense for them. Later adjustments may be seen as failures signalling that the organization is not able to handle the new technology. Second, we do not learn anything about the importance of the modifications observed. Are these major ones or only marginal in character? The first of the studies to be reported in this chapter addresses both of these concerns.
SAP R/3 – exploring adaptive activities Our research focuses on organizational reactions to the entry of R/3 (a product of the German IT corporation SAP AG), one of the most important standardized enterprise software packages that are often referred to as Enterprise Resource Planning (ERP) systems. R/3 promises the seamless integration of all data of an organization’s business processes throughout the value chain. All business tasks from the purchasing process of services and materials, the handling of sales and production orders, the complete inventory and materials management, shipping and billing to the customer should ideally run on a single R/3 system. Even complementary business processes such as customer service, project administration, human resource management and finally the complete range of financial services such as accounting, controlling and business intelligence are integrated into one software application.
276
Georg Schreyögg et al.
All relevant information is processed by a range of functionally differentiated modules, all using a single database. For example, the module for sales and distributions (SD) was designed for processes starting with a quotation or sales order entry and passing all resulting material requirements on to the neighbouring modules for materials management (MM) and production planning (PP). Related processes for either purchasing the required materials (MM) or making them (PP) are triggered in real time as soon as the corresponding sales order (SD) is posted onto the database. SAP’s R/3 system is one of the most prominent and successful among the standardized ERP packages. In order to align business and technology needs one can first choose from among the modules included in the R/3 software, and then begin configuring the system. Due to the broad range of functionalities provided, anyone responsible for implementing and/or running R/3 in his/her organization faces its legendary complexity. Critics increasingly raise questions about its deterministic properties. Davenport (1998), for instance, states that effectively using an R/3 system requires an organization to adopt entirely SAP’s explicit and implicit philosophy. The ubiquity of the SAP R/3 system within an organization’s processes is said to come close to a ‘technological imperative’. There is, however, a configuration tool included in the R/3 system that is commonly known as the ‘customizing’ feature. The tool is called ‘Implementation Management Guide’ (IMG). Customizing (individualized configuration) provides a range of predetermined adaptations to business process parameters, i.e. the user has a limited choice between given alternatives. This highly standardized and supposed deterministic structure makes SAP R/3 an interesting candidate for studying technology enactment in organizations. How do organizations react to this rather dominant software system and its ‘imperatives’? How do they articulate and implant their individual problems and expectations? These questions are all the more interesting as SAP, while highly standardized in nature, has technically opened up the source code of R/3 and thereby provided in principle the opportunity to make changes and to adapt the software’s functionality (in contrast to other prominent software producers such as Microsoft). As part of the service and maintenance functions SAP included programming features integrated into a combined tool known as the ‘ABAP development workbench’ (ABAP: Advanced Business Application Programming). Having acquired the software licence, skilled programmers can modify and enhance standard R/3 functions by using this programming language and the development workbench. Modifying standard programs may, however, seriously impair software quality and reduce system integration benefits. SAP is not willing to take responsibility for these modifications so the firm integrated compulsory control systems. Seeking to distinguish standard and modified (or added) functions and to document all changes made to an R/3 system, automatic monitoring features were introduced as follows:
Shaping organizational technology 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
•
•
277
SSCR (SAP Software Change Registration). The SSCR is a procedure that registers all manual changes to standard R/3 sources and other dictionary objects. Every user carrying out programming activities has to be registered as a development user via the SSCR. The modification of standard SAP sources or dictionary objects requires the registration of the respective object or code as altered and voids SAP’s warranty obligation for that function. Apart from modifying the R/3 standard system, registered developers can create custom code and objects within the R/3 system without limitation. TMS (Transport Management System). According to the official SAP guidelines, system changes should only be done on a separate customizing and development system and transported to the live R/3 system once they are properly documented, tested and approved of by a project or IT manager. The main purpose of the TMS is to move customizing (configuration tables) and dictionary objects (database tables, source code, etc.) from one R/3 system to another and monitor the transportation process (export, import and activation). In order to assure that all the affected customizing and dictionary objects are being transported, users are forced to assign a change order to any adaptation they make. The change order therefore enforces a complete list of all the objects modified or created as well as the source and target system of the changes.
The data automatically collected by these monitoring tools represent all modifications and changes made to R/3 and proved to provide an ideal mode of access to observe the enactment activities of the organization. Definition and measurement of adaptive activities We defined organizational technology adaptation as a deliberate alteration or enhancement of standard R/3 system functionality carried out through the organization by using the SAP development workbench (ABAP). Program corrections officially released by SAP have been ignored since they are not interactive from our framework point of view. The same is true in principle for changes implemented through the standard customizing tool. However, for the sake of completeness and to better expose the varying character of adaptation activities, we included customizing changes as a point of reference in parts of the data presented here. For measuring the number and the type of adaptive activities within the organizations, we used the automatically compiled memory file described above. This type of data collection not only freed us from carrying out (in many respects troubled) observations in the manner conducted by Tyre and Orlikowski, but also allowed for collecting data on adaptation activities from the past without the danger of bias or manipulation through respondents. The digital protocol provides unobtrusive data on all modifying and
278
Georg Schreyögg et al.
workbench activities. Every adaptation ever made to an R/3 system is listed in these central transport files. The main repository for the data to be analysed was the R/3 database table ‘E070’. It keeps an electronic real-time logbook of all the change orders ever registered by the TMS. In addition to the SSCR and TMS data mentioned above, we explored an additional valuable source of information: The Early Watch Report (EWR). This is a another standardized SAP service tool that summarizes the results of a working week’s digital monitoring of a live R/3 system providing information on a number of topics such as the system’s overall configuration and performance and also vital database parameters are presented along with information on the actual use of the software during the monitored timeframe. The EWR provides a clear picture of a company’s adapted R/3 system, its current condition and its actual usage. In particular, it allows for identifying the intensity of use of the modifications and the number of users working with these changes, so that the actual importance of adaptations made to the system can be assessed properly. Analysis of data We set out to investigate the technology adaptation processes following the implementation of R/3, the relevance of the modifications for business operations and the context of exploiting the opportunities to make changes to the R/3 system. We did not rely on distinct projects (as Tyre and Orlikowski had done) but tried to collect all modifications realized in the defined period at a given company. Following the Tyre and Orlikowski study, the focus of our study was on patterns of adaptations over time in order to check whether the ‘windows of opportunity’ thesis with its major emphasis on the beginning of technology use holds. As stated by the ‘windows of opportunity’ thesis, changes in an organization’s technology occur in a rather short period of time immediately after the implementation of new technology. Once that time window has been closed further alterations carried out by members of the organization become rare. Analysing the change orders logged by the R/3 transport management system (mainly database table E070), we plotted all of the measured system changes on a timeline showing the level of overall adaptive activity per month as the sum of all registered change orders excluding – as mentioned – SAP program fixes (and standardized customization). Identifying the users listed with each change order enabled us to distinguish properly between internally and externally produced changes. Changes were marked external when they were traced back to consultants or programmers from outside the organization. While being objective in nature, there are still clear limitations to this technical approach of data collection and analysis. The major problem is its lack of sensitivity for the surrounding context of a measured adaptive
Shaping organizational technology 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
279
activity. We obtain a complete picture of all the changes made, but we are blind to any possible connections between them. Second, this method focuses exclusively on the users listed in the E070 table as being responsible for transporting the changes. This individualistic perspective is likely to distract the interactive context of the changes. We do not learn anything about the initiatives, teams involved or the role of management in the changes made. To compensate for these shortcomings, additional sources were used to collect data. Interviews, the analysis of project documents and company reports played an important role in that part of the investigation. Sample R/3 is used in many industries and companies of all sizes worldwide. Investigating R/3 changes in an organization in the way defined above requires unlimited access to the companies’ confidential R/3 database. This requirement dramatically reduced the number of candidates. Granting this access is a matter of trust limiting the sample to those firms with which the research team had a long-standing relationship. A further requirement was that the sample firms had been using R/3 for at least 36 months (to match the time span of the Tyre and Orlikowski study). Matching the above criteria left a total of six mid-sized German organizations in different industries. Among these we identified two firms as suitable for an in-depth analysis. We selected ‘ABC’,3 a European manufacturer of electrical components such as instrument transformers and conductors with production sites in Berlin (Germany), Jicin (Czeck Republic), Slovenia and Brazil. ABC produces relatively few types of products; however, these products are available in a large number of variants, each with different technical specifications. At the time of our investigation ABC employed roughly 500 people, 100 of which used the company’s R/3 system on a regular basis. We decided to run all the pretests at ABC. This company’s SAP system was based on the previously used R/2 system. In the beginning only the Human Resource (HR) module and parts of the sales processes in the SD module were used. ABC implemented the engineering and production modules in 1996. First observations and interviews with R/3 users showed that the firm had actually modified the system. We had access to data starting with the PP implementation project at ABC in May 1996 through to September 2001. As part of ABC’s staff had previously worked with the SAP R/2 system, users were somewhat experienced. For our second research site, we selected ‘GERO’,4 a Berlin-based wholesale and retail company of building materials with some 250 employees, roughly 200 of whom used some part of the R/3 system in their daily work (SAP users). We had access to data starting with the implementation project of R/3 at GERO in 1998 through to March 2004.
280
Georg Schreyögg et al.
Adaptive activities For a first approach we compiled the annual ratio of overall adaptive activities. As mentioned above, for drawing comparisons we not only included ABAP workbench activities, but also changes stemming from the standard customizing settings. Figure 13.2 shows all registered changes to the R/3 system, including standard customizing (CUST) and workbench activities (SYST). The timeline starts with the implementation of the PP variant configuration in the freshly founded ABC spin-off in early 1996 up to September 2001. These results come as a surprise. They stand in sharp contrast to the results of Tyre and Orlikowski and contradict the ‘windows of opportunity’ thesis. The level of system changes (SYST) at the beginning of the use of the SAP system at ABC was remarkably lower than in two of the three years following. Even in the fifth year of effective usage a remarkably high level of adaptation activity was measured in the realm of the SAP system. The years of both 1996 and 2001 have to be viewed cautiously as we do not have data of the full year. Also, it is amazing that in contrast to the constantly high number of workbench changes the number of customizing activities has been decreasing significantly. What is striking is that workbench changes (SYST) clearly dominated system changes at all times. Workbench changes (interactive self-developed system changes) amounted in total to 74–83 per cent of all annual changes in five consecutive years. The results are very much at variance from the Tyre and Orlikowski results, so the question arises as to whether ABC represents a rare exception.
Adaptive Activity over Time Number of changes made
450 400 350 300 250
SYST CUST
200 150 100 50 0 1996
1997
1998
1999
2000
2001
Time from installation in years
Figure 13.2 All adaptive activity differentiated between standard customizing and workbench changes with ABC
Shaping organizational technology Adaptive Activity over Time 2500 Number of changes made
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
281
2000
1500 SYST CUST 1000
500
0 1998
1999
2000
2001
2002
2003
Time from installation in years
Figure 13.3 All adaptive activity differentiated between standard customizing and workbench changes with GERO
To test this assumption we conducted exactly the same analysis with GERO. The results are shown in Figure 13.3. Surprisingly enough, we found with GERO more or less the same pattern of modifications as with ABC. Again, in sharp contrast ot the results of the Tyre and Orlikowski study, the frequency of system changes (SYST) did not reach its peak right after the implementation. Rather, at the beginning the frequency was remarkably lower than in the years following. Within the first two years of system use, adaptive activity even increased to almost double the changes made during the implementation year. As was the case with ABC even in the fifth year a remarkably high level of modification activities could be observed – on average in 2003 more than two changes per working day. Once again the number of standardized customizing changes continuously decreased were opposed to this pattern as it had its peak level right at the beginning. For the purpose of a direct comparison with the Tyre and Orlikowski study we adopted their time frame of 34 months after installation of the new technology. As the point of ‘time of installation’ we picked the ‘going live’ of the R/3 system at ABC in May 1996, as that was the time when ABC first used the new SAP R/3 system in its own business environment. In accordance with Tyre and Orlikowski the number of all adaptive activities registered in the R/3 system has been calculated on a monthly basis. As Tyre and Orlikowski defined ‘adaptive activity’ as all actions that were targeted at altering the technology, Figure 13.4 includes both customizing and workbench changes.
282
Georg Schreyögg et al. Level of adaptive activity
Number of adaptive activities
120 100 80 60 40 20
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
0 1996
1997
1998
Time from installation in months
Figure 13.4 Level of adaptive activity at ABC on a monthly basis, 1996–98
In total, Figure 13.4 echoes the foregoing contradictions. It reveals a temporal pattern that deviates dramatically from the Tyre and Orlikowski study and rejects the ‘windows of opportunity thesis’. The results come much closer to the thesis of continuous re-definition and re-construction of technology than to the ‘windows’ thesis. Again, we replicated the analysis with GERO. As the point of ‘time of installation’ we picked the ‘going live’ of the R/3 system in January 1999, the time when the users were first exposed to the technology in a real life context. The number of all adaptive activity registered on the GERO R/3 system has been calculated on a monthly basis and it is no longer a surprise that we found the same pattern and supporting evidence for the ‘windows of opportunity’ thesis contradiction. The GERO results (see Figure 13.5) reveal more or less the same temporal pattern that we found at ABC, which comes much closer to the conception of continuous re-definition and reconstruction of technology than to the ‘windows’ thesis. From the results of the windows study we expected the level of adaptive activity to rapidly drop after the initial boost of changes made upon arrival of the new technology in the real world. The May 1996 (ABC) and January 1999 (GERO) results indeed actually showed the highest level of monthly adaptive activity across the 36-month time frame. However, the expected slump and halt of adaptive activity after that boost never occurred. On the contrary, in April 1997 at ABC exactly the same level was reached again – almost a year later. An in-depth analysis revealed that a release change
Shaping organizational technology 600 Number of adaptive activities
500 400 300 200 100 0 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 11 12
1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
283
1999
2000
2001
Time from installation in months
Figure 13.5 Level of adaptive activity at GERO on a monthly basis, 1999–2001
was responsible for that high level of adaptive activity. With GERO the same general pattern emerged. The level of adaptive activity remained constantly high during the years 2000 and 2001 and the sum of system changes even exceeded that of 1999 by approximately 150. Apart from some apparent amplitude due to the year 2000 change and the euro conversion, the level of adaptive activity remained constantly high at an average of more than 150 changes per month over the whole period. ABC and GERO never seem to close the window of opportunity. Maybe the changes that our studies are focusing on are much more important for the daily operations than those studied in the firms of the Tyre and Orlikowski study. The changes our study are focusing on enable the organization to continuously adapt their technological setting to better meet its customers’ needs. A more detailed analysis is needed. The acquisition of knowledge to modify the SAP R/3 system enabled ABC and GERO to make the desired system changes actually happen. Each firm was able to learn, i.e. to adopt and rework R/3 in cases where it seemed necessary to enhance business performance. Although only small or mediumsized, the complexity of the R/3 system and its powerful status did not intimidate the firms as they insisted on enacting ‘their’ technology. The constantly high level of adaptive activity and the adoption of modification skills can be understood as a continuous learning process of the organization. Both organizations were able to acquire the required capability to understand and to change the system. In total, the results are very much in line with the new conception of technology and the constructionist idea of enacting individual technology
284
Georg Schreyögg et al.
in organizations. Further attempts are needed, however, to unveil the nature of technological enactment. There are many questions left open. For instance, questions may be raised on the dynamics of the adaptation process itself and on the nature of technological initiatives in organizations. Where do they start? Who takes the initiative? How did this adoption process happen? In a planned way or informally embedded in organizational politics? Is there such a thing as an ‘emerging technology’? Furthermore, who invents technological changes in organizations? What are the reactions of the higher level management? What are the processes of authorization: official routine or incrementally legitimating? Is there a typical pattern for self-organized technology creation (e.g. logical, chronological)? And other questions. To examine these subsequent questions in more detail we conducted a second study at a different research site directed on the processes of technology inventions by organizational users.
Self-organized technology creation in organizations: the case of an IT-consultancy The study was conducted as in-depth descriptive field research in an IT consultancy with 40 offices and over 1,500 employees. The research, data synthesis and interpretation took approximately 30 months, from February 2001 to August 2003. The research focused on self-organized software developments and modifications such as spreadsheet calculations or the recombination of existing software and functions. Different research methods were applied: analysis of documents, technological screening, semi-structured interviews, direct and indirect observations as well as simulations of daily work routines. Stages of a self-organized technology creation process The case study focuses on the sequence of events around technological dissatisfaction and the subsequent substitution of a time based recording tool, which had been designed to document individual working hours of consultants and projects. The billing process of the consultancy was closely linked to this time recording tool, bills were automatically generated from the time records entered. The story briefly goes as follows. Consultants started complaining about the dysfunctional and time-consuming technology/tool; however, the management insisted on using this software tool, which was a part of the overall SAP system. Some consultants began to informally develop a new time recording tool and set out to work with both the informal and formal tool at the same time. Colleagues came to be familiar with the new informal tool and they adopted and amended it. Subsequently, project leaders became informally involved in the distribution process. Finally, top management
Shaping organizational technology 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
I
II
III
IV
Dysfunctional ,, exogenous Dysfunctiona lexogenous Dysfunctional exogenous organizational organizational technology technology
Solution problems Solution of of problems problems through informal, through informal, informal, endogenous technology endogenous technology technology creation creation
Functional and Functional deficits deficits and and solutions become public solutions become becomepublic public
Organizational Organizational reflex reflex
(technological bundle, (technologicalbundle, bundle, complexity and emergence) emergence complexityand and emergence)
(exploration, (exploration, adaption adaption and and recombination) recombination)
(organizational (organizational revelation revelation and announcement) and announcement) announcement)
(indifference, (indifference, plan plan adjustment adjustmentor or revolution) revolution)
285 ...
...
Figure 13.6 Self-organized technology creation as a process
found the new tool and rejected it initially. Later on, they looked for a compromise and initiated the official re-configuration of the initial tool. The software producer became involved in this process. The whole process can be differentiated into four stages (see Figure 13.6). Stage I: dysfunctional, exogenous organizational technology The consultancy expected their organizational members/consultants to use the standard SAP-based time recording tool to document their individual working hours. However, from the consultants’ point of view the prescribed time recording technology was lacking in many respects: basic needs such as offline recording, pre-selection of data (e.g. primarily used project accounts) and individualization of the work process of time recording could not be properly matched by the SAP system. Asked to describe the attitude towards the SAP time system, the following answer of one of the consultants aptly reflects the prevailing feeling: ‘The time recording system? First of all it is too complex by far, second it is user-unfriendly, confusing and rigid. . .’. Furthermore in their view the system had many additional flaws such as causing occasional system crashes and the quite complicated and annoying procedure for revising false time entries. Any revision had to be officially documented so that everybody could see it (and ask questions about the revisions). As a result, the employees became more and more frustrated with the given technology/software and started to complain about it, but the management decided to stick with the SAP time recording tool for the organization. They valued the automatic invoice creation and billing process and furthermore the management did not want to deviate substantially from the SAP system architecture as they were afraid any deviation might spoil its coherence. Besides that, they did not appreciate the high degree of actual dissatisfaction with the tool at that time.
286
Georg Schreyögg et al.
Stage II: steps towards an informal problem solution The unsatisfying situation with the time recording tool prompted employees to explore other technologies and tools. Some tried to change the SAPsystem itself but failed: I tested the possibilities of changing the code or system structure directly in the SAP-system. A colleague of mine – a real SAP-freak – helped me out, but it is a pretty closed system, there is no reasonable way of changing anything there except the color of your data mask. . . So the employees widened their exploring focus and looked for solutions from other office software and groupware systems. Available tools and platforms were recombined. Highly qualified programmers started to create new problem solutions through self-organized programming and recombination of existing functions of alternative software systems. One consultant described his personal exploration as follows: . . . well, I use this Excel-sheet on a daily basis, which we tacitly crafted here in my department and on my computer – it is open all day long. And since I have to document an internal project for each . . . I continuously enter my working hours in that sheet: When did I begin working on the project? When did I finish? What project did I work for? And so on – and on Fridays I usually copy all the recorded data to the SAP system via drag-and-drop . . . nobody can see that I am using another tool for time recording, everything keeps in order. New solutions were tacitly built and tested at various places in the organization, and the management did not notice all these hidden activities. The employees were acting on their own to meet their individual functional needs in recording working time. The most common modifications had two salient features. The first one was a comfortable interface to the official SAP system. It had been achieved through the combined use of Microsoft Excel, a personal digital assistant (PDA), a piece of time recording software on the PDA and the programming of interfaces between the components (e.g. Visual Basic for Applications). Self-created templates and Microsoft Access based solutions came to the fore. The preferred solution not only provided better functionality but had also one striking advantage: time data could be imported easily from the self-created time records into the official SAP system so that the official requirements could be fulfilled on time. Besides that, the tool enabled employees to handle their individual work process more flexibly, because they had no longer to go online when recording work hours as was necessary with the SAP system. The second basic feature was a much more comfortable pre-structuring of time related data and simplification of data input. The official SAP data
Shaping organizational technology 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
287
system was fairly complex. It contained more than 46,000 different projects, over 30 individual identification codes and many additional options from which employees had to choose. The new technology made data organization much easier. Consultants could check the plausibility of the recorded distribution of working hours on projects throughout the week before entering their time finally records into the SAP system. They no longer had to do it 20 or 30 times a week – just once a week. One consultant commented: . . . sometimes I just enter my time data every ten or twelve days, no one pressures me on it, but I tend to keep a record for myself with this excel-solution each day so I can . . . hmmm well let’s say ‘adjust’ the hours a little bit to better fit the customers’ or the company’s demands. After this personal adjustment I import the data into the SAP-system so that no one gets curious due to modifications – it appears as if the official process worked smoothly. . . The self-created solution did not, however, become a fixed tool – rather it was in a constant state of flux. The employees were continuously adding amendments and new functions. Given the hidden development of the tool, the employees had to find a smart way of handling this self-developed technology. The façade of using the official solution had to be maintained as it was quite clear that management definitively wanted them to stick with the SAP tool. So the self-created tools were used ‘under cover’ parallel to the SAP system. Timerecording technology amounted in this organizational setting to a ‘dual technology’. The everyday acting oscillated between formal and informal areas of the organization and employees found smart mechanisms to make their behaviour appear to be in line with the official requirements. Stage III: the solution becomes known The self-developed tools did not remain unknown in the organizational ‘catacombs’. Rather, they soon spread out informally to many parts of the organization. The tools were passed on to co-workers and colleagues. In most cases inventors or persons with high technological reputations (technological ‘Fire Fighters’) managed the distribution: That PDA-Excel-solution, well, after I finished the programming, I sent it to everyone from my team here in Hamburg as well to those guys from the functional department – you know, I worked with them on a large scale project last year . . . but whoever comes to me and shows interest; I will give it to them . . . They transferred the solutions via e-mail or disk, but they also limited the reach and spread by setting up special distribution lists. Initially, only fellow
288
Georg Schreyögg et al.
co-workers were handed the self-developed tools. The group of persons who became involved in this process were confided in the trust of the tool-builders. In a way, the sharing represented the tool-builders individual confidence-hierarchy. Confidence appears to be the most important factor in determining this distribution process. At a later stage, however, even many project leaders became part of the tacit circle of users. The individual use of the tool varied; it differed from one user to the next and individual amendments and new functions were again added. As a consequence the functionality of the tools rose with the degree to which they were organizationally distributed. On the other hand, the delicacy of the tool also rose with the number of users. Each distribution made the ‘illegal’ usage more likely to become generally known. However, once under way there was obviously no stopping it. Eventually, middle managers came to know more and more the informal technology and the actual dual practice. Increasingly, they found themselves in the position of deciding between ‘useless legality’ and ‘useful illegality’. The common use of ‘illegal’ technology increasingly challenged the authority of the hierarchy. It became obvious that the informal technology creation process was not only a question of functionality but also one of power and domination. Stage IV: the reaction of the hierarchy Finally, in an official meeting with upper management, some middle managers came up with the uneasy question of how to handle the dual technology. They unveiled the hidden technology. The organization (and the upper management) was more or less forced to act. It tried to find a compromise. Management provisionally accepted the informal technology but concurrently asked the software producer (SAP) to come up with a revised version of the initial (faulty) tool. The idea was that the self-developed tools should soon be substituted with the substantially revised ‘official’ software tool. The revision should adopt the bulk of improvements, which the internal and informal technology development had brought about. The modifications and recombinations also led to a change in the performance appraisal system. It was modified in a way that the quality and process of recording individual working hours became an essential element of the appraisal system – not only at the employee level but also at the project and team manager level. By adopting some amendments and by stressing the need to stick closely to the official time recording system the management hoped to bring the dual practice situation to an end. The future will reveal whether they succeeded in doing so. The stages of this case study not only explain how organizational technologies may emerge in a practical setting, it also highlights the multidimensional character of technology developments. It is a long way from
Shaping organizational technology 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
289
the classical technological imperative, explained at the beginning of this article, to internal technology creation as illustrated in this case study. Quite obviously, the whole process also has a close affinity with organizational learning cycles. In accordance with the March/Olsen learning theory (March and Olsen 1979) discrepancies between expectations and technological capability triggered an organizational learning process. Members of the organization started to search for solutions to overcome the perceived deficiencies. Gradually, a new solution spread throughout the organization and formed the basis of acting. Finally, the internal environment reacted to the acting based on the new technology. The feedback from the environment triggered a new learning cycle. The organization was able to learn by reacting to the perceived discrepancies. It developed new technological solutions to overcome deficiencies.
Conclusions and outlook This chapter has advocated an alternative perspective on technology that emphasizes three features: (1) technology as an organizational product; (2) technological heterogeneity as a strategic resource of firms; (3) technology as an ongoing process (i.e. as a dynamic construct). These insights have implications for both theory and practice. First of all and most importantly, organizations are encouraged to experiment with technologies, to take them as plastic tools rather than as deterministic forces that require congruent design patterns. Organizations are also encouraged to reflect on the use of their technologies. Once established in an organization, patterns of technological usage are more often than not perceived as iron laws. The new technological perspective calls into question all these seeming necessities and advocates a continuous audit. Furthermore, the conception of organizational technology encourages organizations to take care of the evolution of technological amendments, changes and redefinitions in their departments and subsystems. Our case studies, and others as well, have shown that many organizations enact adaptive technological processes to match unforeseen challenges and to seek new innovative solutions. Organizations should learn to make use of these processes and not suppress them because of their mostly informal character. Those firms that sensibly explore their technological scope and are able to create a climate encouraging technological learning are likely to develop unique technological solutions in terms of a strategic resource. In the future the strategic dimension of technology in terms of technological enactment and differentiation is likely to play a much more important role. The new non-deterministic view of technology offers an important avenue to design and re-design technologies and their use in organizations. Opposed to material technology, the organizational technology perspective opens the window for exploring the scope of unique resource development and opportunities of setting up socially complex solutions that
290
Georg Schreyögg et al.
may be difficult for competitors to imitate. Much more research is needed to understand and exploit the strategic options inherent in the features of organizational technology.
Notes 1 2 3 4
Actually, it was Max Weber who first put this conception on the agenda of academic discourse; see for an interesting discussion Joerges and Czarniawska 1998: 367f. The use of project documents is mentioned but not explained. The name of the firm has been modified. The name of the firm has been modified.
Bibliography Barley, S. R. (1986) ‘Technology as an occasion for structuring: evidence from observations of CT scanners and the social order of radiology departments’, Administrative Science Quarterly, 31: 78–108. Barney, J. B. (1991) ‘Firm resources and sustained competitive advantage’, Journal of Management, 17: 99–120. Boudreau, M. C. and Robey, D. (2005) ‘Enacting integrated information technology’, Organizational Science, 16: 3–18. Burgelman, R. A., Maidique, M. A. and Wheelwright, St C. (2000) Strategic Management of Technology and Innovation, 3rd edition, Chicago IL: Irwin. Colbert, B. A. (2004) ‘The complex resource-based view: implications for theory and practice in strategic human resource management’, Academy of Management Review, 29: 341–58. Davenport, Th. H. (1998) ‘Putting the enterprise into the enterprise system’, Harvard Business Review, 66 (4): 121–31. DeSanctis, G. and Poole, M. S. (1994) ‘Capturing the complexity in advanced technology use: adaptive structuration theory’, Organization Science, 5: 121–47. Fulk, J. (1993) ‘Social construction of communication technology’, Academy of Management Journal, 36: 921–50. Fulk, J. and DeSanctis, G. (1995) ‘Electronic communication and changing organizational forms’, Organization Science, 6: 337–49. Fulk, J. and Steinfeld, C. W. (eds) (1990), Organizations and Communications Technology, Newbury Park CA: Sage. Grant, R. M. (1991) ‘The resource-based theory of competitive advantage’, California Management Review, 33 (3): 114–35. Hopper, M. (1990) ‘Rattling SABRE – new ways to compete on information’, Harvard Business Review, 68 (3): 118–25. Joerges, B. and Cziarniawska, B. (1998) ‘The question of technology, or how organizations inscribe the world’, Organization Studies, 19: 363–85. Luhmann, N. (1984) Soziale Systeme, Frankfurt: Suhrkamp. March, J. G. and Olsen, J. P. (1979) Ambiguity and Choice in Organisations, 2nd edition, Universitetsforlaget, Bergen. Mosakowski, E. (1997) ‘Strategy making under causal ambiguity: conceptual issues and empirical evidence’, Organization Science, 8: 414–42. Noble, D. (1984) Forces of Production: a social history of industrial automation, New York: Knopf.
Shaping organizational technology 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
291
Orlikowski, W. J. (1992) ‘The duality of technology: rethinking the concept of technology in organizations’, Organization Science, 3: 398–427. Orlikowski, W. J. (2000) ‘Using technology and constituting structures: a practice lens for studying technology’, Organization Science, 11: 404–28. Orlikowski,W. J., Yates, J., Okamura, K. and Fujimoto, M. (1995) ‘Shaping electronic communication: the metastructuring of technology in the context of use’, Organization Science, 6: 423–44. Ortmann, G., Windeler, A., Becker, A. and Schulz, H. J. (1990) Computer und Macht in Organisationen, Opladen: Westdeutscher Verlag. Penrose, E. T. (1957) The Theory of the Growth of the Firm, New York: Wiley. Perrow, C. (1967) ‘A framework for the comparative analysis of organizations’, American Sociological Review, 32: 194–208. Roberts, K. H. and Grabowski, M. (1996) ‘Organizations, technology and structuring’, in Clegg, S., Hardy, C. and Nord, W. R. (eds) Handbook of Organization Studies, London: Sage, pp. 409–23. Scarbrough, H. and Corbett, J. M. (1992) Technology and Organization: power, meaning and design, London: Routledge. Schreyögg, G. (1995) Umwelt, Technologie und Organisationsstruktur, 3rd edition, Bern/Stuttgart: Haupt. Scott, W. R. (1990) ‘Technology and structure: an organizational level perspective’, in Goodman, P. S., Sproull, L. S. and Associates (eds) Technology and Organization, San Francisco CA: Jossey Bass, pp. 109–43. Tyre, M. J. and Orlikowski, W. J. (1994) ‘Windows of opportunity: temporal patterns of technological adaptation in organizations’, Organization Science, 5: 98–118. Weick, K. E. (1990) ‘Technology as equivoque: sensemaking in new technologies’, in Goodman, P. S., Sproull, L. S. and Associates (eds) Technology and Organizations, San Francisco CA: Jossey Bass, 1–44. Wernerfelt, B. (1989) ‘From critical resources to corporate strategy’, Journal of General Management, 14 (3): 4–12. Woodward, J. (1965) Industrial Organization: theory and practice, London: Oxford University Press. Yates, J. W., Orlikowski, W. J. and Okamura, K. (1999) ‘Explicit and implicit structuring of genres: electronic communication in a Japanese R&D organization’, Organization Science, 11: 83–103.
Index
Pages containing relevant figures and tables are shown in italic type. Unless a specific country is mentioned, topics are of a general or global nature. Differences between chapters in spelling and capitalisation have been standardised. ABC, Germany, SAP R/3 system study 279, 280–4, 280, 282 action research 219 acts/bills: Consumer Contract Act (2002), Japan 80; Privacy Bill (2001), Japan 80; Specific Commercial Transaction Act (1976), Japan 80; Telecommunications Act (1996), USA 47–50, 51–2, 109; Uniform Fixed Retail Book Price Bill (2002), Germany 137–8 Adidas-Salomon, innovation 220–1, 225–7; Adidas Customer Project 221–5, 224 AHS (American Hospital Supply) 260–1 Akerlof, G. A. 77 Alba, J. 218 American Hospital Supply (AHS) 260–1 American Telegraph & Telephone (AT&T) 44–5, 47, 52, 52 analog mobile communications systems 234 Anchordoguy, Marie 7, 37 Antimonopoly Law, Japan 36 antitrust law, USA 63 antitrust movement, USA 34–5 Aoki, M. 87 Argyris, C. 219 ARPU (average revenue per unit), mobile phones 247 Arrow, K. J. 85 Asia see China; Japan; Korea; Pacific Rim; Taiwan AT&T (American Telegraph & Telephone) 44–5, 47, 52, 52
average revenue per unit (ARPU), mobile phones 247 Bachmann, R. 85 backbone interconnection, Internet 58, 67–9 bandwidth markets, Internet 67–8 Bell Operating Companies (BOCs), USA 44–5, 109–10; Regional (RBOCs) 47, 47–8, 48, 50, 51 Bit Valley, Tokyo, Japan 160–1, 160, 162, 166 BOCs see Bell Operating Companies, USA book prices, Germany 137–8 Borrus, M. 183–4 bottlenecks, monopolistic 62–3, 62, 63–5, 64 broadband services 118n4, 118n6; best practices 115–16; Canada 112–13, 115; Japan 20, 30–2; vs. narrowband 65–7; penetration, worldwide 43–4, 44, 103–4; regulation 114–15; US development failures 108–10, 118nn5–6 Brown, S. 259 Brynjolffson, E. 260 B2B/B2C commerce see e-commerce Buchanan, J. N. 90 business-to-business/consumer (B2B/ B2C) commerce see e-commerce cable services, USA 48–9, 53, 53, 118n6 camera market, Polaroid/Kodak, USA 263–4
Index 293 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Campbell, D. T. 257 Canada: broadband services 112–13, 115; ICT development 117 capacity glut, electronics 187 CAPs (competitive access providers), USA 47 CDM (contract design manufacturing), electronics 200, 203 CDMA see Code Division Multiple Access cdma2000 236, 237; see also KDDI, Japan Celestica 203 cellular phones see mobile communications technology; mobile phone industry Chesbrough, H. 212 China: electronics manufacturing services (EMS) 189, 191–2; mobile phone industry 105–6, 119n10, 196–7, 252, 252; Taiwanese electronics manufacture/rs 201–2 chip manufacture, rising costs 187–8 Cisco, USA 193 ‘closed-and-owned’ strategy, Japan 27 CLUE (Comprehensive Loss Underwriting Exchange), USA 264 CMs see contract manufacturers, electronics Code Division Multiple Access (CDMA) 236–7, 246; see also KDDI, Japan; NTT DoCoMo, Japan; Vodafone, Japan commensalist relations, evolutionary theory of organizations 261–2 commerce see e-commerce communication, customer/firm interaction 217–19, 227 Compaq, USA 27; Hewlett-Packard (HP) merger 193–4, 193, 196 competition law 70n11 competitive access providers (CAPs), USA 47 competitive advantage, resource-based view (RBV) 272 competitiveness see individual companies, countries, industries and technologies Comprehensive Loss Underwriting Exchange (CLUE), USA 264 computer game industry, restructuring 197–8 computer industry see electronics manufacture/rs
Consumer Contract Act (2001), Japan 80 consumers/customers: Adidas Customer Project 221–5, 224, 225–7; consumer behaviour, e-commerce 135–7; consumer uncertainty/trust, ecommerce 77–80, 84–9, 93, 95nn9–10; e-commerce, Japan 157–9, 158, 167–9, 168; in innovation process 211, 212, 213–14, 216–19, 225–7; ‘lead users’ 213–14; mobile communications 238 Consumer Transaction Guideline, Japan 83 contract design manufacturing (CDM), electronics 200, 203 contract manufacturers (CMs), electronics 184–6, 185; EMS vs. ODM 199–204, 200; expansion in Asia 189–92, 190; price masking 194; restructuring 188, 192–8, 193; see also electronics manufacture/rs ‘controlled competition’ 20 convenience stores (CVSs), Japan 91–3, 163 Cooper, R. G. 214 coordination costs, imperfect markets 215, 216 coupling, evolutionary theory of organizations 261–6, 263 CPNs (cross-national production networks) 184 credence goods 78 CRITO (Center for Research on Information Technology and Organizations) see GEC Japan Database cross-national production networks (CPNs) 184 ‘customer active paradigm’, innovation 213 ‘customer integration competence’ 227 customers see consumers/customers customer services, e-commerce, Japan 167–9, 168 CVSs (convenience stores), Japan 91–3, 163 Cyber Korea 21 113 Dell Computer, USA 195–6, 265–6, 267 deregulation, telecommunications, USA 45, 46, 47, 53 Digital Signature Law (2000), Japan 80
294
Index
Direct Store Distribution (DSD), Hanes, USA 260 distribution channels, e-commerce, Japan 171–3, 172 domino effects, supply chain, electronics 187 DOS/V, Microsoft 27 dotcom companies, Japan 159–66, 160, 162, 164–5 Drucker, P. 258 DSD (Direct Store Distribution), Hanes, USA 260 DSL services: Japan 31; USA 53–4, 53 ECOM (Electronic Commerce Promotion Council), Japan 82–3, 155–6, 156, 157–9, 158 e-commerce 78, 95n6, 96nn23&25, 132, 146–7; consumer behaviour 135–7; consumer uncertainty/trust 77–80, 84–9, 93, 95nn9–10; definitions 125–6; evolution 124–5, 126–8, 128; Germany 127, 128, 128–41, 129–30, 142–3, 145; Japan 75, 76, 80–4, 88–9, 91–4, 97nn32&37, 155–78, 156, 158, 160, 162, 164–5, 168, 170, 172, 175, 177; regulation 75–6, 78–84, 93–4, 95n3, 97n32, 137–46, 142–3, 147n6; United States 81, 82, 83–4, 126–7, 129, 140, 141 EDI (electronic data interchange), Japan 174 Eggertsson, T. 87 Eisenhardt, K. 259 EITO (European Information Technology Observatory), ecommerce trends 127, 128, 129 e-Japan Strategy 30, 36–7, 37, 113–14 El Centro, USA 264–5, 267–8 electronic commerce see e-commerce Electronic Commerce Promotion Council (ECOM), Japan 82–3, 155–6, 156, 157–9, 158 electronic data interchange (EDI), Japan 174 electronic signatures 95n14 electronics manufacture/rs 180–2, 204–6; EMS vs. ODM 199–204, 200; global recession/restructuring 181, 186–8, 192–8, 193; Hewlett-Packard (HP)–Compaq merger 193–4, 193, 196; Japan 25–30, 28, 36–7, 37; transnational networks, rise of 182–6,
185; USA 26–7, 180–6, 185, 193–4, 193, 195–6, 197, 197–8; see also electronics manufacturing services (EMS); telecommunications industry electronics manufacturing services (EMS) 184–6, 185, 199–204, 200; and Hewlett-Packard (HP) 194; Asia, expansion in 181, 189–92, 190, 193, 205–6; supply chain centralization 194–5 encryption software, e-commerce 144 end-users see consumers/customers Enterprise Resource Planning (ERP), SAP R/3, Germany 275–89, 280–3, 285 entry barriers, markets 61–2 e-readiness, world rankings 32–3, 33–4, 136 Ericsson–Sony alliance, mobile phones 196 ERP (Enterprise Resource Planning), SAP R/3, Germany 275–89, 280–3, 285 essential facilities doctrine 63 EU see European Union Eucken, Walter 139 European Information Technology Observatory (EITO), e-commerce trends 127, 128, 129 European Union (EU): e-commerce 127, 128–9, 130, 131, 137, 140–1, 142–3, 144; mobile communications technology 247, 248–9; regulation 60–1, 64–5, 64, 67, 137, 140–1, 142–3, 144; telecommunications industry 35; see also Germany evaluation, mobile communications, four-tier benefit model 237–9 evolutionary theory of organizations 256, 257–61; coupling 261–6, 263 EZWeb Internet services, KDDI, Japan 242 ‘fabless’ electronics manufacture 186, 187, 195, 196, 198, 199 Family Mart, Japan 163 Federal Communications Commission (FCC), USA 48, 50 5th-Generation Computer Project (1982–91), Japan 26 Fisher, F. M. 58 fitness, evolutionary theory of organizations 257, 258
Index 295 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Flextronics International, Asia 189–91, 190, 191–2, 197–8, 203 FOMA see freedom of mobile access FONTAC Project, Japan 28 foreign pressure (gaiatsu), Japan 22 Foxconn, Taiwan 191 freedom of mobile access (FOMA) 240, 243–6, 245; see also NTT DoCoMo, Japan gaiatsu (foreign pressure), Japan 22 GATS (General Agreement on Trade in Services) 141, 147n6 GEC Japan Database 168, 170, 172, 175, 177 General Agreement on Trade in Services (GATS) 141, 147n6 Germany: e-commerce 127, 128, 128–41, 129–30, 142–3, 145; e-readiness 136; SAP R/3 system studies 275–89, 280–3, 285; telecommunications industry 22, 24, 24–5, 35, 41–4, 42–3, 44, 54, 61, 63–4, 65, 67; see also European Union GERO, Germany, SAP R/3 system study 279, 280–4, 281, 283 goods, credence 78 governments, role in ICT development 106, 107, 110–11, 112–14 Gummesson, E. 219 Hanes Direct Store Distribution (DSD), USA 260 Harindranath, G. 105 Hayami, Y. 87 Hewlett-Packard (HP): Compaq merger 193–4, 193, 196; mobile communications technology 251–2 Hitt, L. 260 HP see Hewlett-Packard IBM, USA 26, 195 IBPs (Internet backbone providers) 58, 67–9 ICT (information and communication technologies): investment/ organization co-evolution 256–7, 259–68, 263; national importance 103, 104–6, 118nn1–2, 119n9; New Economy theories/myths 180–1; worldwide comparisons 37; see also electronics manufacture/rs; ICT
development; telecommunications industry; individual services and technologies ICT development 103–4, 116–18; acquired/indigenous advantages 111–12; best practices, worldwide 110–11, 115–16; government role 106, 107, 110–11, 112–14; Japan 113–14, 117, 119n14; New Zealand 119n8; regulation 114–15; US success/failures 103, 104, 107–10, 117–18, 118nn5–6 i-mode Internet service, NTT DoCoMo, Japan 243–4 IMT-2000 (International Mobile Telecommunications-2000) 236 ‘incumbent inertia’, NEC, Japan 27 information: asymmetries 77, 79, 95n7; and customer empowerment 217–19; digitalization and costs 215–16; narcotics information sharing, USA 264–5, 267–8; need-related/solutionbased 211–12, 214–15 information and communication technologies see ICT; ICT development Information-Processing Promotion Association (IPA), Japan 28–9 innovation: at Adidas 220–5, 224, 225–7; closed/traditional 211–12, 220–1; open 210, 212–19, 225–7 institutional change, lack of see path dependence/lock-ins institutional trust see trust, institutional/personal institutions (rules/regulations) 5–6; ecommerce 75–6, 77; frameworks for ICT 3–5; and learning 8–9, 89–91; and path dependence/lock-ins 4, 5, 6–8, 8–9; see also policies, public; regulation insurance industry, CLUE, USA 264 intellectual property rights, e-commerce 141, 145 International Mobile Telecommunications-2000 (IMT-2000) 236 Internet: access 30–3, 32–3, 33–4, 53–4, 53; bandwidth markets 67–8; Germany/Japan/USA compared 24, 24–5, 43–4, 44; i-mode service, NTT DoCoMo, Japan 243–4; peering and transit 68–9; regulation 49–50, 58, 66–7, 69; service provision 59–60, 59; and transaction costs 215–16;
296
Index
see also broadband services; e-commerce; telecommunications industry; telephone industry Internet backbone providers (IBPs) 58, 67–9 Internet Protocol (IP) 67–8 Internet service providers (ISPs) 59–60, 59 investment/organization co-evolution 256–7, 259–68, 263 IP (Internet Protocol) 67–8 IPA (Information-Processing Promotion Association), Japan 28–9 ISPs (Internet service providers) 59–60, 59 IT Basic Law, Japan 30 Jacob, F. 227 JADMA (Japan Direct Marketing Association) 83 Japan: competition policies 19–20; ecommerce 75, 76, 79, 80–4, 88–9, 91–4, 96nn18–19, 97nn32&37, 155–78, 156, 158, 160, 162, 164–5, 168, 170, 172, 175, 177; electronics manufacture/rs 25–30, 28, 36–7, 37; ICT development 113–14, 117, 119n14; Internet access 30–3, 32–3, 33–4; telecommunications industry 20–4, 24–5, 34, 35–6, 41–4, 42–3, 44, 54; see also mobile communications technology, Japan Japan Direct Marketing Association (JADMA) 83 Japan Electronic Computer Company (JECC) 25–6 Japanese Industrial Standard (JIS) 81, 96n18 Japanese Information Processing Development Center (JIPDEC) 82, 96n19 Japan Fair Trade Commission (JFTC) 22, 36 Japan Software Co. 28 JAVA applications 237 JECC (Japan Electronic Computer Company) 25–6 JFTC (Japan Fair Trade Commission) 22, 36 JIPDEC (Japanese Information Processing Development Center) 82, 96n19 JIS (Japanese Industrial Standard) 81, 96n18
KDDI, Japan 239, 240, 241–3, 243, 253 Keefer, P. 85, 86 Keidanren, Japan 36 keiretsu system, Japan 27, 157, 158, 166, 174 Kesan, J. P. 58 Khalil, E. L. 86 Knack, S. 85, 86 Kodak, USA 263–4 Kokunaisen.com, Japan 163 Korea 111, 117, 119n13, 246, 248–9, 250–1, 251–2, 252; Cyber Korea 21 113 Lane, C. 85 LAN (Local Area Network) services, Japan 245 law/laws: Antimonopoly Law, Japan 36; antitrust law 63; competition law 70n11; Digital Signature Law (2000), Japan 80; IT Basic Law, Japan 30; see also litigation Lawson, Japan 91 LDP (Liberal Democratic Party), Japan 21–2, 36 leadership, open innovation 225–6 ‘lead users’ 213–14 learning processes: e-commerce, Japan 91–3; and institutions (rules/regulations) 8–9, 89–91; organizational technology 273, 289 Lee, C. M. 107–8 Lewin, Kurt 219 LG, mobile phones 196–7 Liberal Democratic Party (LDP), Japan 21–2, 36 litigation: AT&T, USA 45, 47; NTT competitors, Japan 23–4 Local Area Network (LAN) services, Japan 245 lock-ins see path dependence/lock-ins Lucent, USA 193 Lüthje, C. 213–14 manufacture/rs see electronics manufacture/rs; telecommunications industry ‘manufacturing active paradigm’, innovation 212 March, J. G. 289 Market for the High-Growth and Emerging Stocks (MOTHER) board, Japan 166
Index 297 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
market power 61–2; monopolistic bottlenecks 62–3, 62, 63–5, 64 markets: entry barriers 61–2; imperfect 215–17; move-to-the-market hypothesis 210, 215; telecommunications, Germany/Japan/ USA compared 41–4, 42–3, 44, 54 MCI, USA 52, 52 Mestmäcker, E.-J. 58 METI see Ministry of Economy, Trade and Industry, Japan mi adidas 220, 223–5, 224 MIC (Ministry of Internal Affairs and Communications), Japan 21, 23–4 Microsoft, USA 26–7, 27; X-Box game console 197–8 Ministry of Economy, Trade and Industry (METI), Japan 22, 25–6, 28–9; privacy guidelines (1997) 81 Ministry of Industry, Canada 112 Ministry of Information and Communications, Korea 113 Ministry of Internal Affairs and Communications (MIC), Japan 21, 23–4 Ministry of International Trade and Industry (MITI) see Ministry of Economy, Trade and Industry (METI), Japan Ministry of Posts and Telecommunications (MPT) see Ministry of Internal Affairs and Communications (MIC), Japan MITI see Ministry of Economy, Trade and Industry (METI), Japan mobile communications technology, Japan 233–4, 237, 239–46, 240, 243, 245, 252–3, 253, 254; Asia/Europe/ USA compared 246–51, 248–9; fourtier benefit model of evaluation 237–9; generation change 234–7; Japan and Asia’s leadership 251–2, 252 mobile phone industry: average revenue per unit (ARPU) 247; China 105–6, 119n10, 196–7, 252, 252; restructuring 196–7; subscribers, Germany/Japan/USA 42–3, 43 monopolies, telecommunications 34–6 monopolistic bottlenecks 62–3, 62, 63–5, 64 Morrison, J. E. P. 260
MOTHER (Market for the High-Growth and Emerging Stocks) board, Japan 166 motivation costs, imperfect markets 215 Motorola, mobile phones 197 move-to-the-market hypothesis 210, 215 MPT see Ministry of Internal Affairs and Communications (MIC), Japan Müller, G. 58 NAPs (network access points), USA 68 narcotics information sharing, USA 264–5, 267–8 narrowband vs. broadband services 65–7 Nash equilibrium, evolutionary theory of organizations 258 National Broadband Taskforce, Canada 112 NCCs (new common carriers), Japan 21 NEC, Japan 27 network access points (NAPs), USA 68 new common carriers (NCCs), Japan 21 New Economy, ICT theories/myths 180–1 new product development (NPD) see innovation New Zealand, ICT development 119n8 Nippon Telephone and Telegraph (NTT), Japan 20, 21, 23–4, 26, 31–2, 248, 249–50; see also NTT DoCoMo, Japan Nokia, mobile phones 196, 197 North, D. C. 6, 87 North America see Canada; United States NPD (new product development) see innovation NTT see Nippon Telephone and Telegraph, Japan NTT DoCoMo, Japan 237, 239–40, 240, 243, 243–6, 245, 250–1, 251–2, 253 OBM (original brand-name manufacture/rs) 200 ODM see original design manufacture/rs O’Donnell, S. 59 OECD (Organisation for Economic Co-operation and Development) 75, 81, 141 OEM see original equipment manufacture/rs Olsen, J. P. 289
298
Index
online sales see e-commerce online services see Internet Online Trust Alliance, e-commerce 83 ‘open-but-owned’ strategy, USA 26–7 open innovation 210, 212–19, 225–7 organizational technology 270–3, 289–90; SAP R/3 system studies, Germany 275–89, 280–3, 285; ‘windows of opportunity’ thesis 273–5, 274, 278, 280, 282 Organization for Economic Co-operation and Development (OECD) 75, 81, 141 organizations: evolutionary theory of 256, 257–66, 263; ICT investment/organization co-evolution 256–7, 259–68, 263 original brand-name manufacture/rs (OBM) 200 original design manufacture/rs (ODM) 185, 199–204, 200; mobile phone industry 197; PC industry restructuring 195–6; see also electronics manufacture/rs original equipment manufacture/rs (OEM) 185, 199–204, 200; price masking 194; restructuring 192–8, 193; see also electronics manufacture/rs Orlikowski, W. J. 273–5, 274, 280–2 Pacific Rim: electronics manufacturing services (EMS) 181, 189–92, 190, 193, 205–6; see also China; Japan; Korea; Taiwan path dependence/lock-ins 4, 5, 6–8, 8–9, 76, 87–8, 97n31, 204; and innovation 34–6; and learning processes 89–93 PCs see personal computers peering arrangements, Internet 68–9 Perrow, Charles 271 personal computers (PCs): industry restructuring 195–6; Japan 26–8, 28, 29–30; penetration, Germany/Japan/USA 43, 44 personal trust see trust, institutional/personal piracy, software, USA 140 Polaroid, USA 263–4 policies, public: e-commerce 78–9, 80; see also institutions (rules/ regulations)
population density 111; Germany/ Japan/USA compared 41–2, 42 price-masking, electronics 194 Privacy Bill (2001), Japan 80 privacy standards, e-commerce, Japan 80–3 procurement, e-commerce, Japan 174, 175 protocols: Internet Protocol (IP) 67–8; Voice over Internet Protocol (VOIP) 118n4 quality assessment, e-commerce 79, 95nn9–10 radio industry, USA, regulation 49 Rannenberg, K. 58 Rapoport, R. N. 219 RBV (resource-based view), competitive advantage 272 Real Time Operating-System Nucleus (TRON) Project, Japan 29 Regional Bell Operating Companies (RBOCs), USA 47, 47–8, 48, 50, 51 regulation 57–8; broadband implementation 114–15; e-commerce 75–6, 78–84, 93–4, 95n3, 97n32, 137–46, 142–3, 147n6; European Union 60–1, 64–5, 64, 67, 137, 140–1, 142–3, 144; Internet 49–50, 58, 66–7, 69; and market power 61–3, 62; technology-neutral 67; telecommunications, USA 44–50, 46, 54, 55; see also institutions (rules/ regulations) resource-based view (RBV), competitive advantage 272 retail price maintenance, books, Germany 137–8 retention, evolutionary theory of organizations 257–8, 259–68, 263 Review of the European Commission (1999) 60 ‘Safe Harbor Agreement’, EU/USA 141 sales, online see e-commerce Samsung, mobile phones 196–7 SAP R/3 system studies, Germany 275–89, 280–3, 285 satellite services, regulation, USA 50 Schön, D. A. 219 screening, quality assessment, e-commerce 95n9
Index 299 1111 2 3 4 5111 6 7 8 9 1011 1 2 3111 4 5 6 7 8 9 20111 1 2 3 4 5111 6 7 8 9 30111 1 2 3 4 5 6 7 8 9 40111 1 2 3 44 45111
Sein, M. K. 105 selection, evolutionary theory of organizations 257, 259–68, 263 7-Eleven, Japan 91, 163 Shah, R. C. 58 shoe industry see Adidas-Salomon Sigma (Software Industrialized Generator and Maintenance Aids) Project, Japan 29 signalling, quality assessment, e-commerce 79, 95nn9–10 Silicon Valley, USA 107–8, 180–1, 182 SK Telecom, Korea 251 software: and e-commerce 133, 140, 144; Japan 27–8, 28, 28–9; see also electronics manufacture/rs Software Industrialized Generator and Maintenance Aids (Sigma) Project, Japan 29 Software Module Project, Japan 28 Solectron 194–5, 203 Sony–Ericsson alliance, mobile phones 196 Sony game consoles 198 Specific Commercial Transaction Act (1976), Japan 80 sports analogy, evolutionary theory of organizations 258–9 standardization, electronics, Japan 29 standards, e-commerce, Japan 79, 80–3 ‘sticky’ information 215 Stigler, G. J. 61, 95n7 stock exchange listings, e-commerce, Japan 166 Super High-Performance Computer Project, Japan 28 supply chains: centralization 194–5; domino effects, electronics 187 symbiotic relations, evolutionary theory of organizations 261 Taiwan, contract manufacturers (CMs) 189, 190, 191, 201–2, 203–4 TCL, China 192 TD-SCDMA 119n10, 236; see also Code Division Multiple Access (CDMA) technology see ICT (information and communication technologies); mobile communications technology; organizational technology
Telecommunications Act (1996), USA 47–50, 51–2, 109 Telecommunications Council, Japan 21–2, 23–4 telecommunications industry: Germany 22, 24, 24–5, 35, 61, 63–4, 65, 67; Japan 20–4, 24–5, 34, 35–6, 41–4, 42–3, 44; markets, Germany/ Japan/USA compared 41–4, 42–3, 44, 54; monopolistic bottlenecks 62–3, 62, 63–5, 64; USA 20, 22, 24, 24–5, 32, 34–5, 41–54, 42–4, 46, 50–3, 54–5, 109; see also electronics manufacture/rs; Internet telegraphy, USA 44 telephone industry: Germany/Japan/ USA compared 24, 24–5, 42, 42; USA 44–5, 47–9, 50–4, 50–3, 109 television industry, cable services, USA 48–9, 53, 53, 118n6 television shopping, Japan 97n35 3G see mobile communications technology, Japan Toyota, Japan, e-commerce 91 Trade-related Aspects of Intellectual Property Rights (TRIPS) Agreement 141 transaction costs, imperfect markets 215–16 transaction cost theory 78–9 transit arrangements, Internet 68–9 transnational production networks, electronics: expansion in Asia 181, 189–92, 190, 193, 205–6; recession/restructuring 181, 186–8, 192–8, 193; rise of 182–6, 185; see also electronics manufacture/rs TRIPS (Trade-related Aspects of Intellectual Property Rights) Agreement 141 TRON (Real Time Operating-System Nucleus) Project, Japan 29 trust, institutional/personal 76, 84–8, 89, 97n36; e-commerce 88–9, 92–3, 94, 136 Tyre, M. J. 273–5, 274, 280–2 UMTS (Universal Mobile Telecommunications System) networks 187 unbundling, Japan 31–2 Uniform Fixed Retail Book Price Bill (2002), Germany 137–8
300
Index
United States (USA): broadband regulation 114–15; e-commerce 81, 82, 83–4, 126–7, 129, 140, 141; electronics manufacture/rs 26–7, 180–6, 185, 193–4, 193, 195–6, 197, 197–8; ICT development success/failures 103, 104, 107–10, 117–18, 118nn5–6; ICT investment/ organization co-evolution 256, 260–1, 263–8; institutional trust 86, 88; mobile communications technology 247, 249; narcotics information sharing 264–5, 267–8; New Economy theories/myths 180–1; peering, Internet backbones 68–9; telecommunications industry 20, 22, 24, 24–5, 32, 34–5, 41–54, 42–4, 46, 50–3, 54–5, 109 Universal Mobile Telecommunications System (UMTS) networks 187 users see consumers/customers US–Japanese Joint Statement on Electronic Commerce (1998) 81 US Sprint 52, 52 Vanberg, V. 90 variation, evolutionary theory of organizations 257, 259–68, 263 vertical integration/disintegration, electronics 181, 183, 186, 187, 190–2, 196–7, 199–204, 200 virtual integration, Dell Computer, USA 266 Vitale, M. R. 260 Vodafone, Japan 239–40, 240, 243
Voice over Internet Protocol (VOIP) 118n4 voice telephony, USA 44–5 VOIP (Voice over Internet Protocol) 118n4 von Hippel, E. 212, 213, 215 W-CDMA (wide-band Code Division Multiple Access) 236–7, 246; see also Code Division Multiple Access (CDMA); NTT DoCoMo, Japan; Vodafone, Japan Weick, K. E. 270–1 Western Union, USA 44 Windows, Microsoft 27 ‘windows of opportunity’ thesis, organizational technology 273–5, 274, 278, 280, 282 Wintelism 184–6, 195, 196, 199, 204; crisis potential 186–8; definition 181, 183–4 wireless technology see mobile communications technology World Trade Organization (WTO) 141 X-Box game console, Microsoft 197–8 Yahoo Japan 31 Yamamura, Kozo 37 Yukawa, K. 160 Zucker, L. G. 86 Zysman, J. 183–4